From 96b5d2e807f667320c66f41ddc1c473023a73ab2 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Fri, 20 Dec 2024 09:29:40 +0200 Subject: [PATCH 0001/1443] drm: xlnx: zynqmp_dpsub: Fix kernel doc Fix two kernel doc warnings introduced by the recent DP audio patch: - Add a doc line for the new "audio" field - Remove a reference to zynqmp_dpsub.c from zynqmp.rst, as the .c file no longer has structured comments Fixes: 3ec5c1579305 ("drm: xlnx: zynqmp_dpsub: Add DP audio support") Closes: https://lore.kernel.org/all/20241220154208.720d990b@canb.auug.org.au/ Reviewed-by: Vishal Sagar Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20241220-xilinx-dp-audio-doc-fix-v1-1-cc488996e463@ideasonboard.com --- Documentation/gpu/zynqmp.rst | 2 -- drivers/gpu/drm/xlnx/zynqmp_dpsub.h | 1 + 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/Documentation/gpu/zynqmp.rst b/Documentation/gpu/zynqmp.rst index f57bfa0ad6ec..1a6f9193de22 100644 --- a/Documentation/gpu/zynqmp.rst +++ b/Documentation/gpu/zynqmp.rst @@ -144,6 +144,4 @@ Internals .. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_dp.c -.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_dpsub.c - .. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_kms.c diff --git a/drivers/gpu/drm/xlnx/zynqmp_dpsub.h b/drivers/gpu/drm/xlnx/zynqmp_dpsub.h index 49875529c2a4..d771b8b199e0 100644 --- a/drivers/gpu/drm/xlnx/zynqmp_dpsub.h +++ b/drivers/gpu/drm/xlnx/zynqmp_dpsub.h @@ -60,6 +60,7 @@ struct zynqmp_dpsub_audio; * @layers: Video and graphics layers * @dp: The DisplayPort controller * @dma_align: DMA alignment constraint (must be a power of 2) + * @audio: DP audio data */ struct zynqmp_dpsub { struct device *dev; From 3f60dbd40d3f7aca312d1aa1131e204eb97a68b3 Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Tue, 31 Dec 2024 17:44:17 +0800 Subject: [PATCH 0002/1443] drm/rockchip: dw_hdmi_qp: Add platform ctrl callback There are some control bits for IO and interrupts status scattered across different GRF on differt SOC. Add platform callback for this IO setting and interrupts status handling. Signed-off-by: Andy Yan Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20241231094425.253398-2-andyshrk@163.com --- .../gpu/drm/rockchip/dw_hdmi_qp-rockchip.c | 79 ++++++++++++------- 1 file changed, 51 insertions(+), 28 deletions(-) diff --git a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c index e498767a0a66..4c82b8400b41 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c @@ -62,6 +62,12 @@ struct rockchip_hdmi_qp { int port_id; }; +struct rockchip_hdmi_qp_ctrl_ops { + void (*io_init)(struct rockchip_hdmi_qp *hdmi); + irqreturn_t (*irq_callback)(int irq, void *dev_id); + irqreturn_t (*hardirq_callback)(int irq, void *dev_id); +}; + static struct rockchip_hdmi_qp *to_rockchip_hdmi_qp(struct drm_encoder *encoder) { struct rockchip_encoder *rkencoder = to_rockchip_encoder(encoder); @@ -226,9 +232,44 @@ static irqreturn_t dw_hdmi_qp_rk3588_irq(int irq, void *dev_id) return IRQ_HANDLED; } +static void dw_hdmi_qp_rk3588_io_init(struct rockchip_hdmi_qp *hdmi) +{ + u32 val; + + val = HIWORD_UPDATE(RK3588_SCLIN_MASK, RK3588_SCLIN_MASK) | + HIWORD_UPDATE(RK3588_SDAIN_MASK, RK3588_SDAIN_MASK) | + HIWORD_UPDATE(RK3588_MODE_MASK, RK3588_MODE_MASK) | + HIWORD_UPDATE(RK3588_I2S_SEL_MASK, RK3588_I2S_SEL_MASK); + regmap_write(hdmi->vo_regmap, + hdmi->port_id ? RK3588_GRF_VO1_CON6 : RK3588_GRF_VO1_CON3, + val); + + val = HIWORD_UPDATE(RK3588_SET_HPD_PATH_MASK, RK3588_SET_HPD_PATH_MASK); + regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON7, val); + + if (hdmi->port_id) + val = HIWORD_UPDATE(RK3588_HDMI1_GRANT_SEL, RK3588_HDMI1_GRANT_SEL); + else + val = HIWORD_UPDATE(RK3588_HDMI0_GRANT_SEL, RK3588_HDMI0_GRANT_SEL); + regmap_write(hdmi->vo_regmap, RK3588_GRF_VO1_CON9, val); + + if (hdmi->port_id) + val = HIWORD_UPDATE(RK3588_HDMI1_HPD_INT_MSK, RK3588_HDMI1_HPD_INT_MSK); + else + val = HIWORD_UPDATE(RK3588_HDMI0_HPD_INT_MSK, RK3588_HDMI0_HPD_INT_MSK); + regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON2, val); +} + +static const struct rockchip_hdmi_qp_ctrl_ops rk3588_hdmi_ctrl_ops = { + .io_init = dw_hdmi_qp_rk3588_io_init, + .irq_callback = dw_hdmi_qp_rk3588_irq, + .hardirq_callback = dw_hdmi_qp_rk3588_hardirq, +}; + struct rockchip_hdmi_qp_cfg { unsigned int num_ports; unsigned int port_ids[MAX_HDMI_PORT_NUM]; + const struct rockchip_hdmi_qp_ctrl_ops *ctrl_ops; const struct dw_hdmi_qp_phy_ops *phy_ops; }; @@ -238,6 +279,7 @@ static const struct rockchip_hdmi_qp_cfg rk3588_hdmi_cfg = { 0xfde80000, 0xfdea0000, }, + .ctrl_ops = &rk3588_hdmi_ctrl_ops, .phy_ops = &rk3588_hdmi_phy_ops, }; @@ -261,7 +303,6 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master, struct resource *res; struct clk_bulk_data *clks; int ret, irq, i; - u32 val; if (!pdev->dev.of_node) return -ENODEV; @@ -278,6 +319,12 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master, if (!cfg) return -ENODEV; + if (!cfg->ctrl_ops || !cfg->ctrl_ops->io_init || + !cfg->ctrl_ops->irq_callback || !cfg->ctrl_ops->hardirq_callback) { + dev_err(dev, "Missing platform ctrl ops\n"); + return -ENODEV; + } + hdmi->dev = &pdev->dev; hdmi->port_id = -ENODEV; @@ -357,31 +404,7 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master, return ret; } - val = HIWORD_UPDATE(RK3588_SCLIN_MASK, RK3588_SCLIN_MASK) | - HIWORD_UPDATE(RK3588_SDAIN_MASK, RK3588_SDAIN_MASK) | - HIWORD_UPDATE(RK3588_MODE_MASK, RK3588_MODE_MASK) | - HIWORD_UPDATE(RK3588_I2S_SEL_MASK, RK3588_I2S_SEL_MASK); - regmap_write(hdmi->vo_regmap, - hdmi->port_id ? RK3588_GRF_VO1_CON6 : RK3588_GRF_VO1_CON3, - val); - - val = HIWORD_UPDATE(RK3588_SET_HPD_PATH_MASK, - RK3588_SET_HPD_PATH_MASK); - regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON7, val); - - if (hdmi->port_id) - val = HIWORD_UPDATE(RK3588_HDMI1_GRANT_SEL, - RK3588_HDMI1_GRANT_SEL); - else - val = HIWORD_UPDATE(RK3588_HDMI0_GRANT_SEL, - RK3588_HDMI0_GRANT_SEL); - regmap_write(hdmi->vo_regmap, RK3588_GRF_VO1_CON9, val); - - if (hdmi->port_id) - val = HIWORD_UPDATE(RK3588_HDMI1_HPD_INT_MSK, RK3588_HDMI1_HPD_INT_MSK); - else - val = HIWORD_UPDATE(RK3588_HDMI0_HPD_INT_MSK, RK3588_HDMI0_HPD_INT_MSK); - regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON2, val); + cfg->ctrl_ops->io_init(hdmi); INIT_DELAYED_WORK(&hdmi->hpd_work, dw_hdmi_qp_rk3588_hpd_work); @@ -394,8 +417,8 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master, return irq; ret = devm_request_threaded_irq(hdmi->dev, irq, - dw_hdmi_qp_rk3588_hardirq, - dw_hdmi_qp_rk3588_irq, + cfg->ctrl_ops->hardirq_callback, + cfg->ctrl_ops->irq_callback, IRQF_SHARED, "dw-hdmi-qp-hpd", hdmi); if (ret) From 3a5981b65f77bfb8abcc6f29c13f1b89ce9df3a3 Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Tue, 31 Dec 2024 17:44:18 +0800 Subject: [PATCH 0003/1443] dt-bindings: display: rockchip: Add rk3576 hdmi controller RK3576 HDMI TX Controller is very similar to that of RK3588, but with some control bits for IO and interrupts status scattered across different GRF. Signed-off-by: Andy Yan Acked-by: Krzysztof Kozlowski Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20241231094425.253398-3-andyshrk@163.com --- .../bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml index d8e761865f27..b838b5ec2dca 100644 --- a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml @@ -29,6 +29,7 @@ allOf: properties: compatible: enum: + - rockchip,rk3576-dw-hdmi-qp - rockchip,rk3588-dw-hdmi-qp reg: From 36439120efbdc62e2f47053e7ddfcc4e34364640 Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Tue, 31 Dec 2024 17:44:19 +0800 Subject: [PATCH 0004/1443] drm/rockchip: dw_hdmi_qp: Add basic RK3576 HDMI output support The HDMI on RK3576 shares the same IP block (PHY and Controller) with rk3588. However, there are some control bits scattered in different GRF. Signed-off-by: Andy Yan Tested-by: Detlev Casanova Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20241231094425.253398-4-andyshrk@163.com --- .../gpu/drm/rockchip/dw_hdmi_qp-rockchip.c | 142 +++++++++++++++++- 1 file changed, 140 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c index 4c82b8400b41..9916843a6ff3 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c @@ -25,6 +25,41 @@ #include "rockchip_drm_drv.h" +#define RK3576_IOC_MISC_CON0 0xa400 +#define RK3576_HDMI_HPD_INT_MSK BIT(2) +#define RK3576_HDMI_HPD_INT_CLR BIT(1) + +#define RK3576_IOC_HDMI_HPD_STATUS 0xa440 +#define RK3576_HDMI_LEVEL_INT BIT(3) + +#define RK3576_VO0_GRF_SOC_CON1 0x0004 +#define RK3576_HDMI_FRL_MOD BIT(0) +#define RK3576_HDMI_HDCP14_MEM_EN BIT(15) + +#define RK3576_VO0_GRF_SOC_CON8 0x0020 +#define RK3576_COLOR_FORMAT_MASK (0xf << 4) +#define RK3576_COLOR_DEPTH_MASK (0xf << 8) +#define RK3576_RGB (0 << 4) +#define RK3576_YUV422 (0x1 << 4) +#define RK3576_YUV444 (0x2 << 4) +#define RK3576_YUV420 (0x3 << 4) +#define RK3576_8BPC (0x0 << 8) +#define RK3576_10BPC (0x6 << 8) +#define RK3576_CECIN_MASK BIT(3) + +#define RK3576_VO0_GRF_SOC_CON12 0x0030 +#define RK3576_GRF_OSDA_DLYN (0xf << 12) +#define RK3576_GRF_OSDA_DIV (0x7f << 1) +#define RK3576_GRF_OSDA_DLY_EN BIT(0) + +#define RK3576_VO0_GRF_SOC_CON14 0x0038 +#define RK3576_I2S_SEL_MASK BIT(0) +#define RK3576_SPDIF_SEL_MASK BIT(1) +#define HDCP0_P1_GPIO_IN BIT(2) +#define RK3576_SCLIN_MASK BIT(4) +#define RK3576_SDAIN_MASK BIT(5) +#define RK3576_HDMI_GRANT_SEL BIT(6) + #define RK3588_GRF_SOC_CON2 0x0308 #define RK3588_HDMI0_HPD_INT_MSK BIT(13) #define RK3588_HDMI0_HPD_INT_CLR BIT(12) @@ -167,6 +202,37 @@ static const struct dw_hdmi_qp_phy_ops rk3588_hdmi_phy_ops = { .setup_hpd = dw_hdmi_qp_rk3588_setup_hpd, }; +static enum drm_connector_status +dw_hdmi_qp_rk3576_read_hpd(struct dw_hdmi_qp *dw_hdmi, void *data) +{ + struct rockchip_hdmi_qp *hdmi = (struct rockchip_hdmi_qp *)data; + u32 val; + + regmap_read(hdmi->regmap, RK3576_IOC_HDMI_HPD_STATUS, &val); + + return val & RK3576_HDMI_LEVEL_INT ? + connector_status_connected : connector_status_disconnected; +} + +static void dw_hdmi_qp_rk3576_setup_hpd(struct dw_hdmi_qp *dw_hdmi, void *data) +{ + struct rockchip_hdmi_qp *hdmi = (struct rockchip_hdmi_qp *)data; + u32 val; + + val = HIWORD_UPDATE(RK3576_HDMI_HPD_INT_CLR, + RK3576_HDMI_HPD_INT_CLR | RK3576_HDMI_HPD_INT_MSK); + + regmap_write(hdmi->regmap, RK3576_IOC_MISC_CON0, val); + regmap_write(hdmi->regmap, 0xa404, 0xffff0102); +} + +static const struct dw_hdmi_qp_phy_ops rk3576_hdmi_phy_ops = { + .init = dw_hdmi_qp_rk3588_phy_init, + .disable = dw_hdmi_qp_rk3588_phy_disable, + .read_hpd = dw_hdmi_qp_rk3576_read_hpd, + .setup_hpd = dw_hdmi_qp_rk3576_setup_hpd, +}; + static void dw_hdmi_qp_rk3588_hpd_work(struct work_struct *work) { struct rockchip_hdmi_qp *hdmi = container_of(work, @@ -182,6 +248,43 @@ static void dw_hdmi_qp_rk3588_hpd_work(struct work_struct *work) } } +static irqreturn_t dw_hdmi_qp_rk3576_hardirq(int irq, void *dev_id) +{ + struct rockchip_hdmi_qp *hdmi = dev_id; + u32 intr_stat, val; + + regmap_read(hdmi->regmap, RK3576_IOC_HDMI_HPD_STATUS, &intr_stat); + if (intr_stat) { + val = HIWORD_UPDATE(RK3576_HDMI_HPD_INT_MSK, RK3576_HDMI_HPD_INT_MSK); + + regmap_write(hdmi->regmap, RK3576_IOC_MISC_CON0, val); + return IRQ_WAKE_THREAD; + } + + return IRQ_NONE; +} + +static irqreturn_t dw_hdmi_qp_rk3576_irq(int irq, void *dev_id) +{ + struct rockchip_hdmi_qp *hdmi = dev_id; + u32 intr_stat, val; + + regmap_read(hdmi->regmap, RK3576_IOC_HDMI_HPD_STATUS, &intr_stat); + + if (!intr_stat) + return IRQ_NONE; + + val = HIWORD_UPDATE(RK3576_HDMI_HPD_INT_CLR, RK3576_HDMI_HPD_INT_CLR); + regmap_write(hdmi->regmap, RK3576_IOC_MISC_CON0, val); + mod_delayed_work(system_wq, &hdmi->hpd_work, + msecs_to_jiffies(HOTPLUG_DEBOUNCE_MS)); + + val = HIWORD_UPDATE(0, RK3576_HDMI_HPD_INT_MSK); + regmap_write(hdmi->regmap, RK3576_IOC_MISC_CON0, val); + + return IRQ_HANDLED; +} + static irqreturn_t dw_hdmi_qp_rk3588_hardirq(int irq, void *dev_id) { struct rockchip_hdmi_qp *hdmi = dev_id; @@ -232,6 +335,21 @@ static irqreturn_t dw_hdmi_qp_rk3588_irq(int irq, void *dev_id) return IRQ_HANDLED; } +static void dw_hdmi_qp_rk3576_io_init(struct rockchip_hdmi_qp *hdmi) +{ + u32 val; + + val = HIWORD_UPDATE(RK3576_SCLIN_MASK, RK3576_SCLIN_MASK) | + HIWORD_UPDATE(RK3576_SDAIN_MASK, RK3576_SDAIN_MASK) | + HIWORD_UPDATE(RK3576_HDMI_GRANT_SEL, RK3576_HDMI_GRANT_SEL) | + HIWORD_UPDATE(RK3576_I2S_SEL_MASK, RK3576_I2S_SEL_MASK); + + regmap_write(hdmi->vo_regmap, RK3576_VO0_GRF_SOC_CON14, val); + + val = HIWORD_UPDATE(0, RK3576_HDMI_HPD_INT_MSK); + regmap_write(hdmi->regmap, RK3576_IOC_MISC_CON0, val); +} + static void dw_hdmi_qp_rk3588_io_init(struct rockchip_hdmi_qp *hdmi) { u32 val; @@ -260,6 +378,12 @@ static void dw_hdmi_qp_rk3588_io_init(struct rockchip_hdmi_qp *hdmi) regmap_write(hdmi->regmap, RK3588_GRF_SOC_CON2, val); } +static const struct rockchip_hdmi_qp_ctrl_ops rk3576_hdmi_ctrl_ops = { + .io_init = dw_hdmi_qp_rk3576_io_init, + .irq_callback = dw_hdmi_qp_rk3576_irq, + .hardirq_callback = dw_hdmi_qp_rk3576_hardirq, +}; + static const struct rockchip_hdmi_qp_ctrl_ops rk3588_hdmi_ctrl_ops = { .io_init = dw_hdmi_qp_rk3588_io_init, .irq_callback = dw_hdmi_qp_rk3588_irq, @@ -273,6 +397,15 @@ struct rockchip_hdmi_qp_cfg { const struct dw_hdmi_qp_phy_ops *phy_ops; }; +static const struct rockchip_hdmi_qp_cfg rk3576_hdmi_cfg = { + .num_ports = 1, + .port_ids = { + 0x27da0000, + }, + .ctrl_ops = &rk3576_hdmi_ctrl_ops, + .phy_ops = &rk3576_hdmi_phy_ops, +}; + static const struct rockchip_hdmi_qp_cfg rk3588_hdmi_cfg = { .num_ports = 2, .port_ids = { @@ -284,8 +417,13 @@ static const struct rockchip_hdmi_qp_cfg rk3588_hdmi_cfg = { }; static const struct of_device_id dw_hdmi_qp_rockchip_dt_ids[] = { - { .compatible = "rockchip,rk3588-dw-hdmi-qp", - .data = &rk3588_hdmi_cfg }, + { + .compatible = "rockchip,rk3576-dw-hdmi-qp", + .data = &rk3576_hdmi_cfg + }, { + .compatible = "rockchip,rk3588-dw-hdmi-qp", + .data = &rk3588_hdmi_cfg + }, {}, }; MODULE_DEVICE_TABLE(of, dw_hdmi_qp_rockchip_dt_ids); From 69c0d83f0eeac6d0d573790c768fcbc902f0dfc5 Mon Sep 17 00:00:00 2001 From: Vignesh Raman Date: Tue, 17 Dec 2024 21:36:52 +0530 Subject: [PATCH 0005/1443] drm/ci: uprev IGT Uprev IGT to the latest version and update expectation files. Signed-off-by: Vignesh Raman Reviewed-by: Dmitry Baryshkov # msm Signed-off-by: Helen Koike Link: https://patchwork.freedesktop.org/patch/msgid/20241217160655.2371138-1-vignesh.raman@collabora.com --- drivers/gpu/drm/ci/gitlab-ci.yml | 2 +- .../gpu/drm/ci/xfails/amdgpu-stoney-fails.txt | 8 +- .../drm/ci/xfails/amdgpu-stoney-flakes.txt | 21 +- .../gpu/drm/ci/xfails/amdgpu-stoney-skips.txt | 5 + drivers/gpu/drm/ci/xfails/i915-amly-fails.txt | 20 +- .../gpu/drm/ci/xfails/i915-amly-flakes.txt | 14 + drivers/gpu/drm/ci/xfails/i915-amly-skips.txt | 1 + drivers/gpu/drm/ci/xfails/i915-apl-fails.txt | 8 +- drivers/gpu/drm/ci/xfails/i915-cml-fails.txt | 22 +- drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt | 7 + drivers/gpu/drm/ci/xfails/i915-cml-skips.txt | 1 + drivers/gpu/drm/ci/xfails/i915-glk-fails.txt | 31 +- drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt | 7 + drivers/gpu/drm/ci/xfails/i915-glk-skips.txt | 298 ++++++++++ drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt | 18 +- drivers/gpu/drm/ci/xfails/i915-jsl-flakes.txt | 15 +- drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt | 112 ++++ drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt | 55 +- drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt | 13 + drivers/gpu/drm/ci/xfails/i915-whl-fails.txt | 15 +- drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt | 7 + .../drm/ci/xfails/mediatek-mt8173-fails.txt | 7 +- .../drm/ci/xfails/mediatek-mt8173-flakes.txt | 7 + .../drm/ci/xfails/mediatek-mt8173-skips.txt | 5 + .../drm/ci/xfails/mediatek-mt8183-fails.txt | 12 +- .../drm/ci/xfails/mediatek-mt8183-skips.txt | 5 + .../gpu/drm/ci/xfails/msm-apq8016-fails.txt | 3 + .../msm-sc7180-trogdor-kingoftown-fails.txt | 1 - .../msm-sc7180-trogdor-kingoftown-flakes.txt | 14 + .../msm-sc7180-trogdor-kingoftown-skips.txt | 5 + ...sm-sc7180-trogdor-lazor-limozeen-fails.txt | 1 - ...sm-sc7180-trogdor-lazor-limozeen-skips.txt | 5 + .../gpu/drm/ci/xfails/msm-sdm845-fails.txt | 2 - .../gpu/drm/ci/xfails/msm-sdm845-skips.txt | 5 + .../drm/ci/xfails/msm-sm8350-hdk-fails.txt | 5 - .../drm/ci/xfails/rockchip-rk3288-fails.txt | 26 +- .../drm/ci/xfails/rockchip-rk3399-fails.txt | 13 +- .../drm/ci/xfails/rockchip-rk3399-flakes.txt | 56 ++ .../drm/ci/xfails/rockchip-rk3399-skips.txt | 5 + .../drm/ci/xfails/virtio_gpu-none-skips.txt | 5 + drivers/gpu/drm/ci/xfails/vkms-none-fails.txt | 8 - drivers/gpu/drm/ci/xfails/vkms-none-skips.txt | 543 +++++++++++++++++- 42 files changed, 1202 insertions(+), 211 deletions(-) diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml index 90bde9f00cc3..b7214718723c 100644 --- a/drivers/gpu/drm/ci/gitlab-ci.yml +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -5,7 +5,7 @@ variables: UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git TARGET_BRANCH: drm-next - IGT_VERSION: a73311079a5d8ac99eb25336a8369a2c3c6b519b + IGT_VERSION: 33adea9ebafd059ac88a5ccfec60536394f36c7c DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/mesa/deqp-runner.git DEQP_RUNNER_GIT_TAG: v0.20.0 diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt index f44dbce3151a..75374085f40f 100644 --- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt @@ -14,10 +14,16 @@ amdgpu/amd_plane@mpo-scale-nv12,Fail amdgpu/amd_plane@mpo-scale-p010,Fail amdgpu/amd_plane@mpo-scale-rgb,Crash amdgpu/amd_plane@mpo-swizzle-toggle,Fail -amdgpu/amd_uvd_dec@amdgpu_uvd_decode,Fail +amdgpu/amd_uvd_dec@amdgpu_uvd_decode,Crash kms_addfb_basic@bad-pitch-65536,Fail kms_addfb_basic@bo-too-small,Fail kms_addfb_basic@too-high,Fail +kms_async_flips@alternate-sync-async-flip,Fail +kms_async_flips@alternate-sync-async-flip-atomic,Fail +kms_async_flips@test-cursor,Fail +kms_async_flips@test-cursor-atomic,Fail +kms_async_flips@test-time-stamp,Fail +kms_async_flips@test-time-stamp-atomic,Fail kms_atomic_transition@plane-all-modeset-transition-internal-panels,Fail kms_atomic_transition@plane-all-transition,Fail kms_atomic_transition@plane-all-transition-nonblocking,Fail diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt index e70bd9d447ca..adffb011298a 100644 --- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt @@ -1,10 +1,3 @@ -# Board Name: hp-11A-G6-EE-grunt -# Bug Report: https://lore.kernel.org/amd-gfx/3542730f-b8d7-404d-a947-b7a5e95d661c@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_async_flips@async-flip-with-page-flip-events - # Board Name: hp-11A-G6-EE-grunt # Bug Report: https://lore.kernel.org/amd-gfx/3542730f-b8d7-404d-a947-b7a5e95d661c@collabora.com/T/#u # Failure Rate: 50 @@ -25,3 +18,17 @@ kms_plane@pixel-format-source-clamping # IGT Version: 1.28-ga73311079 # Linux Version: 6.11.0-rc2 kms_async_flips@async-flip-with-page-flip-events + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://gitlab.freedesktop.org/drm/amd/-/issues/3835 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_async_flips@async-flip-with-page-flip-events-atomic + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://gitlab.freedesktop.org/drm/amd/-/issues/3836 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_async_flips@crc-atomic diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt index f41b3e112976..3879c4812a22 100644 --- a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt @@ -32,3 +32,8 @@ amdgpu/amd_pci_unplug@amdgpu_hotunplug_with_exported_bo amdgpu/amd_pci_unplug@amdgpu_hotunplug_with_exported_fence amdgpu/amd_vrr_range@freesync-parsing device_reset.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt index 0907cb0f6d9e..a29cea4f234c 100644 --- a/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt @@ -1,16 +1,14 @@ core_setmaster@master-drop-set-shared-fd,Fail -core_setmaster@master-drop-set-user,Fail -core_setmaster_vs_auth,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail i915_module_load@resize-bar,Fail i915_pm_rpm@gem-execbuf-stress,Timeout i915_pm_rpm@module-reload,Fail -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions,Timeout -kms_fb_coherency@memset-crc,Crash -kms_flip@busy-flip,Timeout -kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset-interruptible,Fail +kms_async_flips@test-time-stamp,Timeout +kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout +kms_flip@dpms-off-confusion-interruptible,Timeout +kms_flip@wf_vblank-ts-check-interruptible,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail @@ -33,20 +31,12 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail -kms_frontbuffer_tracking@fbc-rgb565-draw-mmap-cpu,Timeout kms_lease@lease-uevent,Fail +kms_lease@page-flip-implicit-plane,Timeout kms_plane_alpha_blend@alpha-basic,Fail kms_plane_alpha_blend@alpha-opaque-fb,Fail kms_plane_alpha_blend@alpha-transparent-fb,Fail kms_plane_alpha_blend@constant-alpha-max,Fail -kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation,Timeout -kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5,Timeout -kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout -kms_prop_blob@invalid-set-prop,Fail -kms_rotation_crc@primary-rotation-180,Timeout -kms_vblank@query-forked-hang,Timeout perf@i915-ref-count,Fail perf_pmu@module-unload,Fail perf_pmu@rc6,Crash diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt index 0207c9807bee..fee246c25b9a 100644 --- a/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt @@ -53,3 +53,17 @@ kms_pm_rpm@modeset-lpsp-stress # IGT Version: 1.28-ga73311079 # Linux Version: 6.11.0-rc2 kms_pm_rpm@drm-resources-equal + +# Board Name: asus-C433TA-AJ0005-rammus +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13346 +# Failure Rate: 60 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +i915_hangman@gt-engine-error + +# Board Name: asus-C433TA-AJ0005-rammus +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13345 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +gen9_exec_parse@unaligned-access diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt index 5186ba3dbbc6..2ef1dc35a7fa 100644 --- a/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt +++ b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt @@ -20,6 +20,7 @@ i915_pm_rc6_residency.* i915_suspend.* kms_scaling_modes.* i915_pm_rpm.* +i915_module_load@reload-with-fault-injection # Kernel panic drm_fdinfo.* diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt index 64772fedaed5..ee11999e3da1 100644 --- a/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt @@ -1,3 +1,4 @@ +core_setmaster@master-drop-set-user,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail @@ -8,13 +9,13 @@ kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail -kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail @@ -29,13 +30,10 @@ kms_plane_alpha_blend@alpha-opaque-fb,Fail kms_plane_alpha_blend@alpha-transparent-fb,Fail kms_plane_alpha_blend@constant-alpha-max,Fail kms_pm_backlight@basic-brightness,Fail +kms_pm_backlight@brightness-with-dpms,Crash kms_pm_backlight@fade,Fail kms_pm_backlight@fade-with-dpms,Fail -kms_pm_rpm@legacy-planes,Timeout -kms_pm_rpm@legacy-planes-dpms,Timeout kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout kms_sysfs_edid_timing,Fail perf@i915-ref-count,Fail perf@non-zero-reason,Timeout diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt index f352b719cf7d..47b3f1d42bb6 100644 --- a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt @@ -1,5 +1,4 @@ core_setmaster@master-drop-set-shared-fd,Fail -core_setmaster@master-drop-set-user,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail @@ -9,9 +8,8 @@ i915_pipe_stress@stress-xrgb8888-ytiled,Fail i915_pm_rpm@gem-execbuf-stress,Timeout i915_pm_rpm@module-reload,Fail i915_pm_rpm@system-suspend-execbuf,Timeout -i915_pm_rps@engine-order,Fail -kms_big_fb@linear-16bpp-rotate-180,Timeout -kms_fb_coherency@memset-crc,Crash +kms_async_flips@test-time-stamp,Timeout +kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout kms_flip@busy-flip,Timeout kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail @@ -36,25 +34,29 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail kms_lease@lease-uevent,Fail +kms_lease@page-flip-implicit-plane,Timeout kms_plane_alpha_blend@alpha-basic,Fail kms_plane_alpha_blend@alpha-opaque-fb,Fail kms_plane_alpha_blend@alpha-transparent-fb,Fail kms_plane_alpha_blend@constant-alpha-max,Fail -kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5,Timeout -kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout kms_psr2_sf@cursor-plane-update-sf,Fail kms_psr2_sf@overlay-plane-update-continuous-sf,Fail kms_psr2_sf@overlay-plane-update-sf-dmg-area,Fail kms_psr2_sf@overlay-primary-update-sf-dmg-area,Fail kms_psr2_sf@plane-move-sf-dmg-area,Fail +kms_psr2_sf@pr-cursor-plane-update-sf,Timeout kms_psr2_sf@primary-plane-update-sf-dmg-area,Fail kms_psr2_sf@primary-plane-update-sf-dmg-area-big-fb,Fail +kms_psr2_sf@psr2-cursor-plane-update-sf,Fail +kms_psr2_sf@psr2-overlay-plane-update-continuous-sf,Fail +kms_psr2_sf@psr2-overlay-plane-update-sf-dmg-area,Fail +kms_psr2_sf@psr2-overlay-primary-update-sf-dmg-area,Fail +kms_psr2_sf@psr2-plane-move-sf-dmg-area,Fail +kms_psr2_sf@psr2-primary-plane-update-sf-dmg-area,Fail +kms_psr2_sf@psr2-primary-plane-update-sf-dmg-area-big-fb,Fail kms_psr2_su@page_flip-NV12,Fail -kms_rotation_crc@primary-rotation-180,Timeout +kms_psr2_su@page_flip-P010,Fail kms_setmode@basic,Fail -kms_vblank@query-forked-hang,Timeout perf@i915-ref-count,Fail perf_pmu@module-unload,Fail perf_pmu@rc6,Crash diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt index d8401251e5f4..5343cc1c8696 100644 --- a/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt @@ -25,3 +25,10 @@ kms_plane_alpha_blend@constant-alpha-min # IGT Version: 1.28-ga73311079 # Linux Version: 6.11.0-rc2 kms_async_flips@crc + +# Board Name: asus-C436FA-Flip-hatch +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13347 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +gen9_exec_parse@unaligned-access diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt index 9d753d97c9ab..c87ff8b40e99 100644 --- a/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt +++ b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt @@ -20,6 +20,7 @@ xe_module_load.* api_intel_allocator.* kms_cursor_legacy.* i915_pm_rpm.* +kms_flip.* # Kernel panic drm_fdinfo.* diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt index 6eb64c672f7d..843c363b42f5 100644 --- a/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt @@ -1,61 +1,35 @@ +core_setmaster@master-drop-set-shared-fd,Fail core_setmaster@master-drop-set-user,Fail -core_setmaster_vs_auth,Fail +gen9_exec_parse@unaligned-access,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail i915_module_load@resize-bar,Fail -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions,Timeout kms_dirtyfb@default-dirtyfb-ioctl,Fail kms_dirtyfb@drrs-dirtyfb-ioctl,Fail kms_dirtyfb@fbc-dirtyfb-ioctl,Fail -kms_flip@blocking-wf_vblank,Fail -kms_flip@busy-flip,Timeout -kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset-interruptible,Fail -kms_flip@wf_vblank-ts-check,Fail kms_flip@wf_vblank-ts-check-interruptible,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail -kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail -kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail -kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail -kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail -kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail -kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail -kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail -kms_frontbuffer_tracking@fbc-rgb565-draw-mmap-cpu,Timeout -kms_frontbuffer_tracking@fbc-tiling-linear,Fail kms_frontbuffer_tracking@fbcdrrs-tiling-linear,Fail kms_lease@lease-uevent,Fail kms_plane_alpha_blend@alpha-opaque-fb,Fail -kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation,Timeout -kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5,Timeout -kms_pm_rpm@legacy-planes,Timeout -kms_pm_rpm@legacy-planes-dpms,Timeout -kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout -kms_prop_blob@invalid-set-prop,Fail -kms_rotation_crc@multiplane-rotation,Fail -kms_rotation_crc@multiplane-rotation-cropping-bottom,Fail kms_rotation_crc@multiplane-rotation-cropping-top,Fail -kms_rotation_crc@primary-rotation-180,Timeout -kms_vblank@query-forked-hang,Timeout perf@non-zero-reason,Timeout sysfs_heartbeat_interval@long,Timeout -sysfs_heartbeat_interval@off,Timeout sysfs_preempt_timeout@off,Timeout sysfs_timeslice_duration@off,Timeout xe_module_load@force-load,Fail @@ -63,4 +37,3 @@ xe_module_load@load,Fail xe_module_load@many-reload,Fail xe_module_load@reload,Fail xe_module_load@reload-no-display,Fail -core_setmaster@master-drop-set-shared-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt index 077886b76093..2a909c657343 100644 --- a/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt @@ -11,3 +11,10 @@ core_hotunplug@unplug-rescan # IGT Version: 1.28-g0df7b9b97 # Linux Version: 6.9.0-rc7 kms_fb_coherency@memset-crc + +# Board Name: hp-x360-12b-ca0010nr-n4020-octopus +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13348 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_flip@dpms-off-confusion-interruptible diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt index 9c64146aed90..219ae839323a 100644 --- a/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt +++ b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt @@ -27,3 +27,301 @@ perf.* # Kernel panic drm_fdinfo.* kms_plane_alpha_blend.* + +kms_async_flips.* +# [IGT] kms_async_flips: executing +# ------------[ cut here ]------------ +# i915 0000:00:02.0: [drm] drm_WARN_ON(intel_dp->pps.vdd_wakeref) +# WARNING: CPU: 0 PID: 1253 at drivers/gpu/drm/i915/display/intel_pps.c:760 intel_pps_vdd_on_unlocked+0x351/0x360 +# Modules linked in: +# CPU: 0 UID: 0 PID: 1253 Comm: kms_async_flips Tainted: G W 6.13.0-rc2-ge95c88d68ac3 #1 +# Tainted: [W]=WARN +# Hardware name: HP Bloog/Bloog, BIOS 09/19/2019 +# RIP: 0010:intel_pps_vdd_on_unlocked+0x351/0x360 +# Code: 8b 77 50 4d 85 f6 75 03 4c 8b 37 e8 19 98 03 00 48 c7 c1 10 d3 61 84 4c 89 f2 48 c7 c7 67 56 69 84 48 89 c6 e8 a0 22 63 ff 90 <0f> 0b 90 90 e9 5f fd ff ff e8 81 94 69 00 90 90 90 90 90 90 90 90 +# RSP: 0018:ffff9573c0dcfad8 EFLAGS: 00010286 +# RAX: 0000000000000000 RBX: ffff93498148e238 RCX: 00000000ffffdfff +# RDX: 0000000000000000 RSI: ffff9573c0dcf988 RDI: 0000000000000001 +# RBP: ffff934980d80000 R08: 0000000000009ffb R09: 00000000ffffdfff +# R10: 00000000ffffdfff R11: ffffffff8488c8a0 R12: ffff934980d80d68 +# R13: 0000000000000000 R14: ffff93498093c9b0 R15: 0000000000000000 +# FS: 00007fbf1f4486c0(0000) GS:ffff9349fbc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000055f260419028 CR3: 0000000106594000 CR4: 0000000000350ef0 +# Call Trace: +# +# intel_pps_vdd_on+0x78/0x150 +# intel_dp_detect+0xb9/0x7c0 +# drm_helper_probe_detect+0x47/0xb0 +# drm_helper_probe_single_connector_modes+0x40b/0x660 +# drm_mode_getconnector+0x369/0x490 +# drm_ioctl_kernel+0xad/0x110 +# drm_ioctl+0x235/0x4b0 +# __x64_sys_ioctl+0x8f/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7fbf208e4cdb +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffe061f6c60 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fbf208e4cdb +# RDX: 00007ffe061f6d00 RSI: 00000000c05064a7 RDI: 0000000000000003 +# RBP: 00007ffe061f6d00 R08: 0000000000000007 R09: 000055f260410cb0 +# R10: 650094cf40322201 R11: 0000000000000246 R12: 00000000c05064a7 +# R13: 0000000000000003 R14: 00000000c05064a7 R15: 00007ffe061f6d00 +# +# irq event stamp: 0 +# hardirqs last enabled at (0): [<0000000000000000>] 0x0 +# hardirqs last disabled at (0): [] copy_process+0xaf3/0x2eb0 +# softirqs last enabled at (0): [] copy_process+0xaf3/0x2eb0 +# softirqs last disabled at (0): [<0000000000000000>] 0x0 +# ---[ end trace 0000000000000000 ]--- +# i915 0000:00:02.0: [drm] *ERROR* PPS state mismatch +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* Failed to read DPCD register 0x92 +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# [IGT] kms_async_flips: starting subtest test-time-stamp +# i915 0000:00:02.0: [drm] *ERROR* [CRTC:70:pipe A] flip_done timed out +# i915 0000:00:02.0: [drm] *ERROR* flip_done timed out +# i915 0000:00:02.0: [drm] *ERROR* [CRTC:70:pipe A] commit wait timed out +# INFO: task kms_async_flips:1253 blocked for more than 122 seconds. +# Tainted: G W 6.13.0-rc2-ge95c88d68ac3 #1 +# "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. +# task:kms_async_flips state:D stack:12720 pid:1253 tgid:1253 ppid:164 flags:0x00004006 +# Call Trace: +# +# __schedule+0x495/0xd20 +# schedule+0x35/0x120 +# drm_vblank_work_flush+0x91/0xf0 +# intel_wait_for_vblank_workers+0x58/0x90 +# intel_atomic_commit_tail+0xa86/0x15f0 +# intel_atomic_commit+0x257/0x2a0 +# drm_atomic_commit+0xac/0xe0 +# set_property_atomic+0x91/0x160 +# drm_mode_obj_set_property_ioctl+0xc1/0x110 +# drm_ioctl_kernel+0xad/0x110 +# drm_ioctl+0x235/0x4b0 +# __x64_sys_ioctl+0x8f/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7fbf208e4cdb +# RSP: 002b:00007ffe061f6c50 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 000055f24e272330 RCX: 00007fbf208e4cdb +# RDX: 00007ffe061f6ce0 RSI: 00000000c01864ba RDI: 0000000000000003 +# RBP: 00007ffe061f6ce0 R08: 0000000000000000 R09: 000055f2604184f0 +# R10: 00007fbf209ca310 R11: 0000000000000246 R12: 00000000c01864ba +# R13: 0000000000000003 R14: 0000000000000000 R15: 000055f26040ab90 +# +# INFO: lockdep is turned off. + +kms_lease.* +# [IGT] kms_lease: executing +# ------------[ cut here ]------------ +# i915 0000:00:02.0: [drm] drm_WARN_ON(intel_dp->pps.vdd_wakeref) +# WARNING: CPU: 1 PID: 1271 at drivers/gpu/drm/i915/display/intel_pps.c:760 intel_pps_vdd_on_unlocked+0x351/0x360 +# Modules linked in: +# CPU: 1 UID: 0 PID: 1271 Comm: kms_lease Tainted: G W 6.13.0-rc2-gfb6fd142a8eb #1 +# Tainted: [W]=WARN +# Hardware name: HP Bloog/Bloog, BIOS 09/19/2019 +# RIP: 0010:intel_pps_vdd_on_unlocked+0x351/0x360 +# Code: 8b 77 50 4d 85 f6 75 03 4c 8b 37 e8 19 98 03 00 48 c7 c1 10 d3 61 a1 4c 89 f2 48 c7 c7 67 56 69 a1 48 89 c6 e8 a0 22 63 ff 90 <0f> 0b 90 90 e9 5f fd ff ff e8 81 94 69 00 90 90 90 90 90 90 90 90 +# RSP: 0018:ffffb26500ddfad8 EFLAGS: 00010286 +# RAX: 0000000000000000 RBX: ffff988580dc6238 RCX: 00000000ffffdfff +# RDX: 0000000000000000 RSI: ffffb26500ddf988 RDI: 0000000000000001 +# RBP: ffff988580d90000 R08: 0000000000009ffb R09: 00000000ffffdfff +# R10: 00000000ffffdfff R11: ffffffffa188c8a0 R12: ffff988580d90d68 +# R13: 0000000000000000 R14: ffff98858093c9b0 R15: 0000000000000000 +# FS: 00007f6b0c60d6c0(0000) GS:ffff9885fbd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000055eea27460a8 CR3: 0000000105ac6000 CR4: 0000000000350ef0 +# Call Trace: +# +# intel_pps_vdd_on+0x78/0x150 +# intel_dp_detect+0xb9/0x7c0 +# drm_helper_probe_detect+0x47/0xb0 +# drm_helper_probe_single_connector_modes+0x40b/0x660 +# drm_mode_getconnector+0x369/0x490 +# drm_ioctl_kernel+0xad/0x110 +# drm_ioctl+0x235/0x4b0 +# __x64_sys_ioctl+0x8f/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f6b0d637c5b +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffc42b13670 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f6b0d637c5b +# RDX: 00007ffc42b13710 RSI: 00000000c05064a7 RDI: 0000000000000003 +# RBP: 00007ffc42b13710 R08: 0000000000000007 R09: 000055eea2741720 +# R10: 9fb9be4e118bfa43 R11: 0000000000000246 R12: 00000000c05064a7 +# R13: 0000000000000003 R14: 00000000c05064a7 R15: 00007ffc42b13710 +# +# irq event stamp: 0 +# hardirqs last enabled at (0): [<0000000000000000>] 0x0 +# hardirqs last disabled at (0): [] copy_process+0xaf3/0x2eb0 +# softirqs last enabled at (0): [] copy_process+0xaf3/0x2eb0 +# softirqs last disabled at (0): [<0000000000000000>] 0x0 +# ---[ end trace 0000000000000000 ]--- +# i915 0000:00:02.0: [drm] *ERROR* PPS state mismatch +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* Failed to read DPCD register 0x92 +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# i915 0000:00:02.0: [drm] *ERROR* AUX A/DDI A/PHY A: not done (status 0x00000000) +# [IGT] kms_lease: starting subtest page-flip-implicit-plane +# [IGT] kms_lease: starting dynamic subtest pipe-A-eDP-1 +# i915 0000:00:02.0: [drm] *ERROR* [CRTC:70:pipe A] flip_done timed out +# i915 0000:00:02.0: [drm] *ERROR* flip_done timed out +# i915 0000:00:02.0: [drm] *ERROR* [CRTC:70:pipe A] commit wait timed out +# INFO: task kms_lease:1271 blocked for more than 122 seconds. +# Tainted: G W 6.13.0-rc2-gfb6fd142a8eb #1 +# "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. +# task:kms_lease state:D stack:12720 pid:1271 tgid:1271 ppid:184 flags:0x00004006 +# Call Trace: +# +# __schedule+0x495/0xd20 +# schedule+0x35/0x120 +# drm_vblank_work_flush+0x91/0xf0 +# intel_wait_for_vblank_workers+0x58/0x90 +# intel_atomic_commit_tail+0xa86/0x15f0 +# intel_atomic_commit+0x257/0x2a0 +# drm_atomic_commit+0xac/0xe0 +# set_property_atomic+0x91/0x160 +# drm_mode_obj_set_property_ioctl+0xc1/0x110 +# drm_ioctl_kernel+0xad/0x110 +# drm_ioctl+0x235/0x4b0 +# __x64_sys_ioctl+0x8f/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f6b0d637c5b +# RSP: 002b:00007ffc42b13650 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 00007ffc42b14270 RCX: 00007f6b0d637c5b +# RDX: 00007ffc42b136e0 RSI: 00000000c01864ba RDI: 0000000000000003 +# RBP: 00007ffc42b136e0 R08: 0000000000000000 R09: 000055eea2741750 +# R10: 00007f6b0d71d310 R11: 0000000000000246 R12: 00000000c01864ba +# R13: 0000000000000003 R14: 0000000000000000 R15: 000055eea27378d0 +# +# INFO: lockdep is turned off. + +# The test alternates between failing and timing out on reruns, causing the pipeline to fail +gen9_exec_parse@unaligned-access diff --git a/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt index ed9f7b576843..0e08fff741aa 100644 --- a/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-jsl-fails.txt @@ -1,15 +1,16 @@ -core_setmaster@master-drop-set-user,Fail +drm_fdinfo@busy-check-all,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail i915_module_load@resize-bar,Fail -i915_pm_rpm@gem-execbuf-stress,Timeout -i915_pm_rpm@module-reload,Fail -kms_flip@plain-flip-fb-recreate,Fail +kms_flip@dpms-off-confusion,Fail +kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset,Fail +kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset-interruptible,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail @@ -17,6 +18,7 @@ kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail @@ -26,19 +28,15 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail +kms_frontbuffer_tracking@fbc-rgb565-draw-blt,Timeout kms_lease@lease-uevent,Fail -kms_pm_rpm@legacy-planes,Timeout -kms_pm_rpm@legacy-planes-dpms,Timeout kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout +kms_rotation_crc@bad-pixel-format,Fail kms_rotation_crc@multiplane-rotation,Fail kms_rotation_crc@multiplane-rotation-cropping-bottom,Fail kms_rotation_crc@multiplane-rotation-cropping-top,Fail perf@i915-ref-count,Fail -perf_pmu@busy-accuracy-50,Fail perf_pmu@module-unload,Fail -perf_pmu@most-busy-idle-check-all,Fail perf_pmu@rc6,Crash sysfs_heartbeat_interval@long,Timeout sysfs_heartbeat_interval@off,Timeout diff --git a/drivers/gpu/drm/ci/xfails/i915-jsl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-jsl-flakes.txt index 5c3ef4486b9d..8b2aefd1cc52 100644 --- a/drivers/gpu/drm/ci/xfails/i915-jsl-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/i915-jsl-flakes.txt @@ -1,13 +1,6 @@ # Board Name: acer-cb317-1h-c3z6-dedede -# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12475 +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13349 # Failure Rate: 100 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.12.0-rc1 -kms_flip@flip-vs-panning-interruptible - -# Board Name: acer-cb317-1h-c3z6-dedede -# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12476 -# Failure Rate: 100 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.12.0-rc1 -kms_universal_plane@cursor-fb-leak +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_flip@flip-vs-fences-interruptible diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt index 6ec2f83ffe13..dc722d6a774e 100644 --- a/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt +++ b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt @@ -37,3 +37,115 @@ tools_test.* # GPU hang sysfs_timeslice_.* sysfs_heartbeat_.* + +device_reset@unbind-reset-rebind +# [IGT] device_reset: starting subtest unbind-reset-rebind +# i915 0000:00:02.0: [drm] Found kabylake (device ID 5917) display version 9.00 stepping C0 +# i915 0000:00:02.0: vgaarb: deactivate vga console +# i915 0000:00:02.0: [drm] *ERROR* DPLL0 not locked +# ------------[ cut here ]------------ +# i915 0000:00:02.0: [drm] drm_WARN_ON((val & (1 << 30)) == 0) +# WARNING: CPU: 4 PID: 472 at drivers/gpu/drm/i915/display/intel_cdclk.c:944 skl_get_cdclk+0x1ca/0x360 +# Modules linked in: +# CPU: 4 UID: 0 PID: 472 Comm: device_reset Not tainted 6.13.0-rc2-ge95c88d68ac3 #1 +# Hardware name: HP Nami/Nami, BIOS 09/19/2019 +# RIP: 0010:skl_get_cdclk+0x1ca/0x360 +# Code: 67 50 4d 85 e4 0f 84 89 01 00 00 e8 e0 16 13 00 48 c7 c1 a0 71 5f 9a 4c 89 e2 48 c7 c7 67 56 69 9a 48 89 c6 e8 67 a1 72 ff 90 <0f> 0b 90 90 8b 43 08 8b 53 04 89 43 0c 89 03 85 d2 0f 85 26 ff ff +# RSP: 0018:ffffacba40f5fbd0 EFLAGS: 00010282 +# RAX: 0000000000000000 RBX: ffff953182571058 RCX: 00000000ffffdfff +# RDX: 0000000000000000 RSI: ffffacba40f5fa80 RDI: 0000000000000001 +# RBP: ffff953182570d68 R08: 0000000000009ffb R09: 00000000ffffdfff +# R10: 00000000ffffdfff R11: ffffffff9a88c8a0 R12: ffff953180d3f4a0 +# R13: 0000000000000000 R14: ffff953182570d68 R15: 0000000000000000 +# FS: 00007f74989c46c0(0000) GS:ffff9534eed00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f1df530eff8 CR3: 0000000102302002 CR4: 00000000003706f0 +# DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 +# DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 +# Call Trace: +# +# intel_update_cdclk+0x1c/0x90 +# intel_cdclk_init_hw+0x46e/0x4c0 +# intel_power_domains_init_hw+0x3eb/0x770 +# intel_display_driver_probe_noirq+0x85/0x230 +# i915_driver_probe+0x66d/0xc60 +# local_pci_probe+0x3a/0x90 +# pci_device_probe+0xb5/0x180 +# really_probe+0xc9/0x2b0 +# __driver_probe_device+0x6e/0x110 +# device_driver_attach+0x42/0xa0 +# bind_store+0x71/0xc0 +# kernfs_fop_write_iter+0x121/0x1c0 +# vfs_write+0x2a8/0x530 +# ksys_write+0x6f/0xf0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f7499e662c0 +# Code: 40 00 48 8b 15 41 9b 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 80 3d 21 23 0e 00 00 74 17 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83 ec 28 48 89 +# RSP: 002b:00007ffffaf36a68 EFLAGS: 00000202 ORIG_RAX: 0000000000000001 +# RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f7499e662c0 +# RDX: 000000000000000c RSI: 00007ffffaf36b50 RDI: 0000000000000003 +# RBP: 000000000000000c R08: 0000000000000007 R09: 000055b37d0964a0 +# R10: 0000000000000000 R11: 0000000000000202 R12: 00007ffffaf36b50 +# R13: 0000000000000003 R14: 000055b3526a7220 R15: 00007f749a0dc020 +# +# irq event stamp: 28591 +# hardirqs last enabled at (28625): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (28658): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (28688): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (28701): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# ------------[ cut here ]------------ +# i915 0000:00:02.0: [drm] drm_WARN_ON((val & (1 << 30)) == 0) +# WARNING: CPU: 1 PID: 472 at drivers/gpu/drm/i915/display/intel_cdclk.c:944 skl_get_cdclk+0x1ca/0x360 +# Modules linked in: +# CPU: 1 UID: 0 PID: 472 Comm: device_reset Tainted: G W 6.13.0-rc2-ge95c88d68ac3 #1 +# Tainted: [W]=WARN +# Hardware name: HP Nami/Nami, BIOS 09/19/2019 +# RIP: 0010:skl_get_cdclk+0x1ca/0x360 +# Code: 67 50 4d 85 e4 0f 84 89 01 00 00 e8 e0 16 13 00 48 c7 c1 a0 71 5f 9a 4c 89 e2 48 c7 c7 67 56 69 9a 48 89 c6 e8 67 a1 72 ff 90 <0f> 0b 90 90 8b 43 08 8b 53 04 89 43 0c 89 03 85 d2 0f 85 26 ff ff +# RSP: 0018:ffffacba40f5fb98 EFLAGS: 00010286 +# RAX: 0000000000000000 RBX: ffffacba40f5fbbc RCX: 00000000ffffdfff +# RDX: 0000000000000000 RSI: ffffacba40f5fa48 RDI: 0000000000000001 +# RBP: ffff953182570d68 R08: 0000000000009ffb R09: 00000000ffffdfff +# R10: 00000000ffffdfff R11: ffffffff9a88c8a0 R12: ffff953180d3f4a0 +# R13: ffff953182573208 R14: ffff953182570d68 R15: 0000000000000000 +# FS: 00007f74989c46c0(0000) GS:ffff9534eec40000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f2b6598ee00 CR3: 0000000102302001 CR4: 00000000003706f0 +# DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 +# DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 +# Call Trace: +# +# gen9_disable_dc_states+0x86/0x300 +# intel_power_well_enable+0x56/0x70 +# __intel_display_power_get_domain.part.0+0x4d/0x70 +# intel_power_domains_init_hw+0x95/0x770 +# intel_display_driver_probe_noirq+0x85/0x230 +# i915_driver_probe+0x66d/0xc60 +# local_pci_probe+0x3a/0x90 +# pci_device_probe+0xb5/0x180 +# really_probe+0xc9/0x2b0 +# __driver_probe_device+0x6e/0x110 +# device_driver_attach+0x42/0xa0 +# bind_store+0x71/0xc0 +# kernfs_fop_write_iter+0x121/0x1c0 +# vfs_write+0x2a8/0x530 +# ksys_write+0x6f/0xf0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f7499e662c0 +# Code: 40 00 48 8b 15 41 9b 0d 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 80 3d 21 23 0e 00 00 74 17 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83 ec 28 48 89 +# RSP: 002b:00007ffffaf36a68 EFLAGS: 00000202 ORIG_RAX: 0000000000000001 +# RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f7499e662c0 +# RDX: 000000000000000c RSI: 00007ffffaf36b50 RDI: 0000000000000003 +# RBP: 000000000000000c R08: 0000000000000007 R09: 000055b37d0964a0 +# R10: 0000000000000000 R11: 0000000000000202 R12: 00007ffffaf36b50 +# R13: 0000000000000003 R14: 000055b3526a7220 R15: 00007f749a0dc020 +# +# irq event stamp: 29605 +# hardirqs last enabled at (29617): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (29634): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (29630): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (29625): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt index 461ef69ef08a..93d42b146df9 100644 --- a/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt @@ -1,64 +1,24 @@ -api_intel_allocator@fork-simple-stress-signal,Timeout -api_intel_allocator@open-vm,Timeout -api_intel_allocator@simple-allocator,Timeout -api_intel_bb@lot-of-buffers,Timeout -api_intel_bb@object-reloc-keep-cache,Timeout -api_intel_bb@offset-control,Timeout -api_intel_bb@render-ccs,Timeout -api_intel_bb@reset-bb,Timeout -core_auth@basic-auth,Timeout -core_hotunplug@hotrebind,Timeout -core_setmaster@master-drop-set-user,Fail -debugfs_test@read_all_entries_display_on,Timeout -drm_read@empty-block,Timeout -dumb_buffer@create-clear,Timeout -dumb_buffer@invalid-bpp,Timeout -gen3_render_tiledx_blits,Timeout -gen7_exec_parse@basic-allocation,Timeout -gen9_exec_parse@batch-invalid-length,Timeout -gen9_exec_parse@bb-secure,Timeout -gen9_exec_parse@secure-batches,Timeout -gen9_exec_parse@shadow-peek,Timeout -gen9_exec_parse@unaligned-jump,Timeout -i915_getparams_basic@basic-subslice-total,Timeout -i915_hangman@gt-engine-hang,Timeout +api_intel_allocator@reopen,Timeout +api_intel_bb@destroy-bb,Timeout +core_hotunplug@hotrebind-lateclose,Timeout +drm_read@short-buffer-block,Timeout +dumb_buffer@map-valid,Timeout i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail i915_module_load@resize-bar,Fail -i915_pciid,Timeout -i915_pipe_stress@stress-xrgb8888-ytiled,Timeout -i915_pm_rpm@gem-execbuf-stress,Timeout i915_pm_rps@engine-order,Timeout -i915_pm_rps@thresholds-idle-park,Timeout -i915_query@engine-info,Timeout -i915_query@query-topology-kernel-writes,Timeout -i915_query@test-query-geometry-subslices,Timeout kms_lease@lease-uevent,Fail kms_rotation_crc@multiplane-rotation,Fail perf@i915-ref-count,Fail perf_pmu@busy,Timeout perf_pmu@enable-race,Timeout -perf_pmu@event-wait,Timeout -perf_pmu@faulting-read,Timeout -perf_pmu@gt-awake,Timeout -perf_pmu@interrupts,Timeout perf_pmu@module-unload,Fail -perf_pmu@most-busy-idle-check-all,Timeout perf_pmu@rc6,Crash -perf_pmu@render-node-busy-idle,Fail perf_pmu@semaphore-wait-idle,Timeout -prime_busy@after,Timeout -prime_mmap@test_aperture_limit,Timeout -prime_mmap@test_map_unmap,Timeout prime_mmap@test_refcounting,Timeout -prime_self_import@basic-with_one_bo,Timeout -sriov_basic@enable-vfs-autoprobe-off,Timeout -syncobj_basic@bad-destroy,Timeout -syncobj_basic@bad-flags-fd-to-handle,Timeout -syncobj_basic@create-signaled,Timeout -syncobj_eventfd@invalid-bad-pad,Timeout -syncobj_eventfd@timeline-wait-before-signal,Timeout +sriov_basic@enable-vfs-bind-unbind-each-numvfs-all,Timeout +syncobj_basic@illegal-fd-to-handle,Timeout syncobj_wait@invalid-multi-wait-unsubmitted-signaled,Timeout syncobj_wait@invalid-signal-illegal-handle,Timeout syncobj_wait@invalid-single-wait-all-unsubmitted,Timeout @@ -66,7 +26,6 @@ syncobj_wait@multi-wait-all-submitted,Timeout syncobj_wait@multi-wait-for-submit-submitted-signaled,Timeout syncobj_wait@wait-any-complex,Timeout syncobj_wait@wait-delayed-signal,Timeout -template@A,Timeout xe_module_load@force-load,Fail xe_module_load@load,Fail xe_module_load@many-reload,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt index b47df5855e8d..938377896841 100644 --- a/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt +++ b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt @@ -35,3 +35,16 @@ syncobj_wait.* # Kernel panic and test hangs with multiple kms tests kms_.* + +# The test alternates between failing and timing out on reruns, causing the pipeline to fail +api_intel_allocator@reopen +api_intel_bb@destroy-bb +core_hotunplug@hotrebind-lateclose +dumb_buffer@map-valid +i915_pm_rps@engine-order +perf_pmu@busy +perf_pmu@enable-race +perf_pmu@semaphore-wait-idle +prime_mmap@test_refcounting +sriov_basic@enable-vfs-bind-unbind-each-numvfs-all +syncobj_basic@illegal-fd-to-handle diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt index 0ce240e3aa07..1cb6978c86dc 100644 --- a/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt +++ b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt @@ -1,5 +1,4 @@ core_setmaster@master-drop-set-shared-fd,Fail -core_setmaster@master-drop-set-user,Fail i915_module_load@load,Fail i915_module_load@reload,Fail i915_module_load@reload-no-display,Fail @@ -7,12 +6,10 @@ i915_module_load@resize-bar,Fail i915_pm_rpm@gem-execbuf-stress,Timeout i915_pm_rpm@module-reload,Fail i915_pm_rpm@system-suspend-execbuf,Timeout -i915_pm_rps@engine-order,Fail -kms_big_fb@linear-16bpp-rotate-180,Timeout -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions,Timeout +kms_async_flips@test-time-stamp,Timeout +kms_ccs@crc-sprite-planes-basic-y-tiled-ccs,Timeout kms_dirtyfb@default-dirtyfb-ioctl,Fail kms_dirtyfb@fbc-dirtyfb-ioctl,Fail -kms_fb_coherency@memset-crc,Crash kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail @@ -34,18 +31,12 @@ kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail kms_frontbuffer_tracking@fbc-tiling-linear,Fail -kms_frontbuffer_tracking@fbc-1p-indfb-fliptrack-mmap-gtt,Timeout kms_lease@lease-uevent,Fail +kms_lease@page-flip-implicit-plane,Timeout kms_plane_alpha_blend@alpha-basic,Fail kms_plane_alpha_blend@alpha-opaque-fb,Fail kms_plane_alpha_blend@alpha-transparent-fb,Fail kms_plane_alpha_blend@constant-alpha-max,Fail -kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5,Timeout -kms_pm_rpm@modeset-stress-extra-wait,Timeout -kms_pm_rpm@universal-planes,Timeout -kms_pm_rpm@universal-planes-dpms,Timeout -kms_rotation_crc@primary-rotation-180,Timeout -kms_vblank@query-forked-hang,Timeout perf@i915-ref-count,Fail perf_pmu@module-unload,Fail perf_pmu@rc6,Crash diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt index 60b8d1c64e70..19a289617080 100644 --- a/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt @@ -4,3 +4,10 @@ # IGT Version: 1.28-g0df7b9b97 # Linux Version: 6.9.0-rc7 kms_pm_rpm@modeset-lpsp-stress + +# Board Name: dell-latitude-5400-8665U-sarien +# Bug Report: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13351 +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +gen9_exec_parse@unaligned-access diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt index 8e0efc80d510..4f176c04ec4e 100644 --- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt @@ -1,8 +1,5 @@ -fbdev@eof,Fail -fbdev@read,Fail kms_3d,Fail kms_bw@connected-linear-tiling-1-displays-1920x1080p,Fail -kms_bw@connected-linear-tiling-1-displays-2160x1440p,Fail kms_bw@connected-linear-tiling-1-displays-2560x1440p,Fail kms_bw@connected-linear-tiling-1-displays-3840x2160p,Fail kms_bw@connected-linear-tiling-2-displays-1920x1080p,Fail @@ -22,6 +19,6 @@ kms_cursor_legacy@cursor-vs-flip-atomic,Fail kms_cursor_legacy@cursor-vs-flip-legacy,Fail kms_flip@flip-vs-modeset-vs-hang,Fail kms_flip@flip-vs-panning-vs-hang,Fail -kms_lease@lease-uevent,Fail -kms_rmfb@close-fd,Fail +kms_flip@flip-vs-suspend,Fail kms_flip@flip-vs-suspend-interruptible,Fail +kms_lease@lease-uevent,Fail diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt index 2e5bf6ae25f2..2956567c3048 100644 --- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt @@ -39,3 +39,10 @@ kms_cursor_legacy@cursor-vs-flip-atomic-transitions # IGT Version: 1.28-g0df7b9b97 # Linux Version: 6.9.0-rc7 kms_prop_blob@invalid-set-prop + +# Board Name: mt8173-elm-hana +# Bug Report: https://lore.kernel.org/dri-devel/c0b4cded-0971-4abb-9965-2acf4602d7cd@collabora.com/T/#u +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_bw@connected-linear-tiling-1-displays-2160x1440p diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt index 8198e06344a3..d0db51874aef 100644 --- a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-skips.txt @@ -15,3 +15,8 @@ tools_test.* # Currently fails and causes coverage loss for other tests # since core_getversion also fails. core_hotunplug.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt index 845f852bb4a0..5a063361d7f2 100644 --- a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt @@ -1,16 +1,12 @@ -core_setmaster@master-drop-set-shared-fd,Fail +core_setmaster@master-drop-set-user,Fail dumb_buffer@create-clear,Crash -fbdev@eof,Fail -fbdev@pan,Fail -fbdev@read,Fail -fbdev@unaligned-read,Fail kms_bw@connected-linear-tiling-1-displays-1920x1080p,Fail -kms_bw@connected-linear-tiling-1-displays-2160x1440p,Fail kms_bw@connected-linear-tiling-1-displays-2560x1440p,Fail +kms_bw@connected-linear-tiling-1-displays-3840x2160p,Fail kms_bw@linear-tiling-1-displays-1920x1080p,Fail +kms_bw@linear-tiling-1-displays-2160x1440p,Fail kms_bw@linear-tiling-1-displays-3840x2160p,Fail -kms_color@invalid-gamma-lut-sizes,Fail +kms_cursor_legacy@cursor-vs-flip-atomic,Fail kms_flip@flip-vs-panning-vs-hang,Fail kms_flip@flip-vs-suspend,Fail kms_lease@lease-uevent,Fail -kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt index 8198e06344a3..d0db51874aef 100644 --- a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-skips.txt @@ -15,3 +15,8 @@ tools_test.* # Currently fails and causes coverage loss for other tests # since core_getversion also fails. core_hotunplug.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt index 066d24ee3e08..7752adff05c1 100644 --- a/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt +++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt @@ -1,4 +1,7 @@ kms_3d,Fail +kms_cursor_legacy@forked-bo,Fail +kms_cursor_legacy@forked-move,Fail +kms_cursor_legacy@single-bo,Fail kms_cursor_legacy@torture-bo,Fail kms_force_connector_basic@force-edid,Fail kms_hdmi_inject@inject-4k,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-fails.txt index 6dbc2080347d..e4a8f8352cd6 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-fails.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-fails.txt @@ -15,4 +15,3 @@ kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail kms_plane_alpha_blend@alpha-7efc,Fail kms_plane_alpha_blend@coverage-7efc,Fail kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail -kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt index d74e04405e65..8910afb6acf2 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-flakes.txt @@ -18,3 +18,17 @@ msm/msm_shrink@copy-gpu-oom-32 # IGT Version: 1.28-g0df7b9b97 # Linux Version: 6.9.0-rc7 msm/msm_shrink@copy-gpu-oom-8 + +# Board Name: sc7180-trogdor-kingoftown +# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/68 +# Failure Rate: 40 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_lease@page-flip-implicit-plane + +# Board Name: sc7180-trogdor-kingoftown +# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/69 +# Failure Rate: 20 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_plane@plane-position-hole-dpms diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt index c2833eee1c4b..478d7c161616 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt @@ -23,3 +23,8 @@ kms_flip@2x-wf_vblank-ts-check # Hangs the machine kms_cursor_crc@cursor-random-max-size + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-fails.txt index 6dbc2080347d..e4a8f8352cd6 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-fails.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-fails.txt @@ -15,4 +15,3 @@ kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail kms_plane_alpha_blend@alpha-7efc,Fail kms_plane_alpha_blend@coverage-7efc,Fail kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail -kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt index 7c69c1f1d55b..ef9318afcd89 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt @@ -17,3 +17,8 @@ tools_test.* # Currently fails and causes coverage loss for other tests # since core_getversion also fails. core_hotunplug.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt index fa8c7e663858..7a2ab58b706f 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt @@ -1,4 +1,3 @@ -drm_read@invalid-buffer,Fail kms_color@ctm-0-25,Fail kms_color@ctm-0-50,Fail kms_color@ctm-0-75,Fail @@ -28,4 +27,3 @@ kms_plane_alpha_blend@coverage-7efc,Fail kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail kms_plane_cursor@overlay,Fail kms_plane_cursor@viewport,Fail -kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt index 94783cafc21a..2ce7f7e23a01 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt @@ -30,3 +30,8 @@ kms_cursor_crc.* # IGT Version: 1.28-ga73311079 # Linux Version: 6.11.0-rc2 kms_content_protection@uevent + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-fails.txt index 4892c0c70a6d..8d26b23133aa 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-fails.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sm8350-hdk-fails.txt @@ -7,9 +7,4 @@ kms_cursor_legacy@torture-bo,Fail kms_cursor_legacy@torture-move,Fail kms_hdmi_inject@inject-4k,Fail kms_lease@lease-uevent,Fail -kms_plane_alpha_blend@alpha-7efc,Fail -kms_plane_alpha_blend@alpha-basic,Fail -kms_plane_alpha_blend@alpha-opaque-fb,Fail -kms_plane_alpha_blend@alpha-transparent-fb,Fail -kms_plane_alpha_blend@constant-alpha-max,Fail msm/msm_recovery@gpu-fault-parallel,Fail diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt index 90282dfa19f4..ba9160d4d8eb 100644 --- a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt @@ -1,24 +1,10 @@ +core_setmaster@master-drop-set-root,Crash +core_setmaster@master-drop-set-shared-fd,Crash core_setmaster@master-drop-set-user,Crash +core_setmaster_vs_auth,Crash dumb_buffer@create-clear,Crash fbdev@pan,Crash -kms_bw@linear-tiling-2-displays-1920x1080p,Fail -kms_cursor_crc@cursor-onscreen-32x10,Crash -kms_cursor_crc@cursor-onscreen-32x32,Crash -kms_cursor_crc@cursor-onscreen-64x64,Crash -kms_cursor_crc@cursor-random-32x10,Crash -kms_cursor_crc@cursor-sliding-32x10,Crash -kms_cursor_crc@cursor-sliding-32x32,Crash -kms_cursor_crc@cursor-sliding-64x21,Crash -kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail -kms_cursor_legacy@cursor-vs-flip-legacy,Fail -kms_cursor_legacy@flip-vs-cursor-crc-atomic,Crash -kms_flip@flip-vs-panning-vs-hang,Crash -kms_invalid_mode@int-max-clock,Crash -kms_lease@invalid-create-leases,Fail -kms_pipe_crc_basic@read-crc-frame-sequence,Crash -kms_plane@pixel-format,Crash -kms_plane@pixel-format-source-clamping,Crash +kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail +kms_flip@flip-vs-modeset-vs-hang,Crash kms_prop_blob@invalid-set-prop,Crash -kms_properties@get_properties-sanity-atomic,Crash -kms_properties@get_properties-sanity-non-atomic,Crash -kms_rmfb@close-fd,Crash +kms_prop_blob@invalid-set-prop-any,Crash diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt index 83a38853b4af..2803d0d80192 100644 --- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt @@ -2,6 +2,7 @@ dumb_buffer@create-clear,Crash kms_atomic_transition@modeset-transition,Fail kms_atomic_transition@modeset-transition-fencing,Fail kms_atomic_transition@plane-toggle-modeset-transition,Fail +kms_bw@connected-linear-tiling-1-displays-2160x1440p,Fail kms_color@gamma,Fail kms_color@legacy-gamma,Fail kms_cursor_crc@cursor-alpha-opaque,Fail @@ -24,6 +25,7 @@ kms_cursor_crc@cursor-rapid-movement-32x32,Fail kms_cursor_crc@cursor-rapid-movement-64x21,Fail kms_cursor_crc@cursor-rapid-movement-64x64,Fail kms_cursor_crc@cursor-size-change,Fail +kms_cursor_crc@cursor-size-hints,Fail kms_cursor_crc@cursor-sliding-32x10,Fail kms_cursor_crc@cursor-sliding-32x32,Fail kms_cursor_crc@cursor-sliding-64x21,Fail @@ -32,6 +34,7 @@ kms_cursor_edge_walk@64x64-left-edge,Fail kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail kms_cursor_legacy@cursor-vs-flip-atomic,Fail +kms_cursor_legacy@cursor-vs-flip-atomic-transitions,Fail kms_cursor_legacy@cursor-vs-flip-legacy,Fail kms_cursor_legacy@cursor-vs-flip-toggle,Fail kms_cursor_legacy@flip-vs-cursor-atomic,Fail @@ -41,14 +44,11 @@ kms_cursor_legacy@flip-vs-cursor-legacy,Fail kms_cursor_legacy@long-nonblocking-modeset-vs-cursor-atomic,Fail kms_flip@basic-flip-vs-wf_vblank,Fail kms_flip@blocking-wf_vblank,Fail -kms_flip@flip-vs-absolute-wf_vblank,Fail -kms_flip@flip-vs-blocking-wf-vblank,Fail kms_flip@flip-vs-modeset-vs-hang,Fail kms_flip@flip-vs-panning,Fail kms_flip@flip-vs-panning-interruptible,Fail kms_flip@flip-vs-panning-vs-hang,Fail kms_flip@modeset-vs-vblank-race,Fail -kms_flip@modeset-vs-vblank-race-interruptible,Fail kms_flip@plain-flip-fb-recreate,Fail kms_flip@plain-flip-fb-recreate-interruptible,Fail kms_flip@plain-flip-ts-check,Fail @@ -64,14 +64,11 @@ kms_pipe_crc_basic@nonblocking-crc,Fail kms_pipe_crc_basic@nonblocking-crc-frame-sequence,Fail kms_pipe_crc_basic@read-crc,Fail kms_pipe_crc_basic@read-crc-frame-sequence,Fail -kms_plane@pixel-format,Crash -kms_plane@pixel-format-source-clamping,Crash +kms_plane@pixel-format,Fail +kms_plane@pixel-format-source-clamping,Fail kms_plane@plane-panning-bottom-right,Fail kms_plane@plane-panning-top-left,Fail kms_plane@plane-position-covered,Fail kms_plane@plane-position-hole,Fail -kms_plane@plane-position-hole-dpms,Fail kms_plane_cursor@primary,Fail -kms_plane_multiple@tiling-none,Fail -kms_rmfb@close-fd,Fail kms_universal_plane@universal-plane-functional,Fail diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt index 56f7d4f1ed15..348b4ce7eb4b 100644 --- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt @@ -74,3 +74,59 @@ kms_bw@linear-tiling-2-displays-2160x1440p # IGT Version: 1.28-ga73311079 # Linux Version: 6.11.0-rc5 kms_flip@flip-vs-expired-vblank + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/f944dd08-c88c-49ae-aff0-274374550a93@collabora.com/T/#u +# Failure Rate: 40 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_bw@linear-tiling-1-displays-2160x1440p + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/afa2d3bf-29f2-488d-8cc9-f30d461444b0@collabora.com/T/#u +# Failure Rate: 80 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_plane_multiple@tiling-none + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/6fdaa97f-c1a5-4216-831f-dbb7c5f90498@collabora.com/T/#u +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_bw@linear-tiling-1-displays-1920x1080p + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/616aa015-9574-4527-9d07-d8d698bbcc3c@collabora.com/T/#u +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_plane@plane-position-hole-dpms + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/7a1b888f-d7db-4ed7-96cd-3975ace837fb@collabora.com/T/#u +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_flip@flip-vs-absolute-wf_vblank + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/f17fffb6-abc4-464e-8465-395311b01f6a@collabora.com/T/#u +# Failure Rate: 100 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_flip@flip-vs-blocking-wf-vblank + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/9b590b26-1bf9-4951-b6a3-ef6c67e6a1c6@collabora.com/T/#u +# Failure Rate: 60 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_bw@linear-tiling-2-displays-1920x1080p + +# Board Name: hp-11A-G6-EE-grunt +# Bug Report: https://lore.kernel.org/dri-devel/059545fa-65b1-4f5c-a13e-4d2898679f51@collabora.com/T/#u +# Failure Rate: 20 +# IGT Version: 1.29-g33adea9eb +# Linux Version: 6.13.0-rc2 +kms_flip@modeset-vs-vblank-race-interruptible diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt index eb16b29dee48..e8e994d92557 100644 --- a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt @@ -18,3 +18,8 @@ tools_test.* # Currently fails and causes coverage loss for other tests # since core_getversion also fails. core_hotunplug.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt index 9c9e048725f8..adbcdd0f28d2 100644 --- a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt +++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt @@ -23,3 +23,8 @@ tools_test.* # Currently fails and causes coverage loss for other tests # since core_getversion also fails. core_hotunplug.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative diff --git a/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt b/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt index 71c02104a683..85647eca4e55 100644 --- a/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt +++ b/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt @@ -1,12 +1,5 @@ -kms_cursor_crc@cursor-rapid-movement-128x128,Fail -kms_cursor_crc@cursor-rapid-movement-128x42,Fail -kms_cursor_crc@cursor-rapid-movement-256x256,Fail kms_cursor_crc@cursor-rapid-movement-256x85,Fail kms_cursor_crc@cursor-rapid-movement-32x10,Fail -kms_cursor_crc@cursor-rapid-movement-32x32,Fail -kms_cursor_crc@cursor-rapid-movement-512x170,Fail -kms_cursor_crc@cursor-rapid-movement-512x512,Fail -kms_cursor_crc@cursor-rapid-movement-64x21,Fail kms_cursor_crc@cursor-rapid-movement-64x64,Fail kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail @@ -21,7 +14,6 @@ kms_cursor_legacy@flip-vs-cursor-legacy,Fail kms_flip@flip-vs-modeset-vs-hang,Fail kms_flip@flip-vs-panning-vs-hang,Fail kms_lease@lease-uevent,Fail -kms_pipe_crc_basic@nonblocking-crc,Fail kms_writeback@writeback-check-output,Fail kms_writeback@writeback-check-output-XRGB2101010,Fail kms_writeback@writeback-fb-id,Fail diff --git a/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt b/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt index b3d16e82e9a2..319789806271 100644 --- a/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt +++ b/drivers/gpu/drm/ci/xfails/vkms-none-skips.txt @@ -1,8 +1,9 @@ # keeps printing vkms_vblank_simulate: vblank timer overrun and never ends kms_invalid_mode@int-max-clock -# Kernel panic -kms_cursor_crc@cursor-rapid-movement-32x10 +# kernel panic seen with kms_cursor_crc tests +kms_cursor_crc.* +# kms_cursor_crc@cursor-rapid-movement-32x10 # Oops: 0000 [#1] PREEMPT SMP NOPTI # CPU: 0 PID: 2635 Comm: kworker/u8:13 Not tainted 6.9.0-rc7-g40935263a1fd #1 # Hardware name: ChromiumOS crosvm, BIOS 0 @@ -52,7 +53,7 @@ kms_cursor_crc@cursor-rapid-movement-32x10 # FS: 0000000000000000(0000) GS:ffffa2ad2bc00000(0000) knlGS:0000000000000000 # CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 -kms_cursor_crc@cursor-rapid-movement-256x85 +#kms_cursor_crc@cursor-rapid-movement-256x85 # [drm:drm_crtc_add_crc_entry] *ERROR* Overflow of CRC buffer, userspace reads too slow. # Oops: 0000 [#1] PREEMPT SMP NOPTI # CPU: 1 PID: 10 Comm: kworker/u8:0 Not tainted 6.9.0-rc7-g646381cde463 #1 @@ -104,7 +105,7 @@ kms_cursor_crc@cursor-rapid-movement-256x85 # CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 # CR2: 0000000000000078 CR3: 0000000109b38000 CR4: 0000000000350ef0 -kms_cursor_crc@cursor-onscreen-256x256 +#kms_cursor_crc@cursor-onscreen-256x256 # Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI # CPU: 1 PID: 1913 Comm: kworker/u8:6 Not tainted 6.10.0-rc5-g8a28e73ebead #1 # Hardware name: ChromiumOS crosvm, BIOS 0 @@ -258,6 +259,535 @@ kms_cursor_edge_walk@128x128-left-edge # CR2: 0000000000000018 CR3: 0000000101710000 CR4: 0000000000350ef0 # vkms_vblank_simulate: vblank timer overrun +# DEBUG - Begin test kms_cursor_crc@cursor-rapid-movement-64x64 +# ------------[ cut here ]------------ +# WARNING: CPU: 1 PID: 1250 at drivers/gpu/drm/vkms/vkms_crtc.c:139 vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Modules linked in: vkms +# CPU: 1 UID: 0 PID: 1250 Comm: kms_cursor_crc Not tainted 6.13.0-rc2-ge95c88d68ac3 #1 +# Hardware name: ChromiumOS crosvm, BIOS 0 +# RIP: 0010:vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Code: f7 48 89 f3 e8 d0 bf ee ec 48 8b 83 50 01 00 00 a8 01 75 15 48 8b bb a0 01 00 00 e8 59 05 95 ec 48 89 df 5b e9 50 05 95 ec 90 <0f> 0b 90 eb e5 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 +# RSP: 0018:ffff9fb640aafb08 EFLAGS: 00010202 +# RAX: ffff8e7240859e05 RBX: ffff8e7241cd6400 RCX: ffffffffae496b65 +# RDX: ffffffffad2d1f80 RSI: 0000000000000000 RDI: 0000000000000000 +# RBP: 0000000000000000 R08: 0000000000000034 R09: 0000000000000002 +# R10: 0000000047dd15a5 R11: 00000000547dd15a R12: ffff8e72590cc000 +# R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 +# FS: 00007f0942ad56c0(0000) GS:ffff8e726bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f0942ad0008 CR3: 0000000118d1e000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __warn+0x8c/0x190 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? report_bug+0x164/0x190 +# ? handle_bug+0x54/0x90 +# ? exc_invalid_op+0x17/0x70 +# ? asm_exc_invalid_op+0x1a/0x20 +# ? __pfx_drm_property_free_blob+0x10/0x10 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? vkms_atomic_crtc_destroy_state+0x10/0x40 [vkms] +# drm_atomic_state_default_clear+0x137/0x2f0 +# __drm_atomic_state_free+0x6c/0xb0 +# drm_atomic_helper_update_plane+0x100/0x150 +# drm_mode_cursor_universal+0x10e/0x270 +# drm_mode_cursor_common+0x115/0x240 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# drm_mode_cursor_ioctl+0x4a/0x70 +# drm_ioctl_kernel+0xb0/0x110 +# drm_ioctl+0x235/0x4b0 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# __x64_sys_ioctl+0x92/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f0943a84cdb +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007fff267d68d0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 000000000000005a RCX: 00007f0943a84cdb +# RDX: 00007fff267d6960 RSI: 00000000c01c64a3 RDI: 0000000000000005 +# RBP: 00007fff267d6960 R08: 0000000000000007 R09: 00005619a60f3450 +# R10: fe95a83851609dee R11: 0000000000000246 R12: 00000000c01c64a3 +# R13: 0000000000000005 R14: 00005619a3cd2c68 R15: 00005619a608c830 +# +# irq event stamp: 57793 +# hardirqs last enabled at (57799): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (57804): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (45586): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (45569): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI +# CPU: 0 UID: 0 PID: 119 Comm: kworker/u8:6 Tainted: G W 6.13.0-rc2-ge95c88d68ac3 #1 +# Tainted: [W]=WARN +# Hardware name: ChromiumOS crosvm, BIOS 0 +# Workqueue: vkms_composer vkms_composer_worker [vkms] +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffff9fb640efbd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff8e7241926000 RSI: ffff8e7241927ffc RDI: 000000000b7fb767 +# RBP: ffff9fb640efbde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff8e7241b09a80 R11: 0000000000000000 R12: ffff8e7241cd6200 +# R13: 0000000000000013 R14: 0000000000000000 R15: ffff8e7241b09a80 +# FS: 0000000000000000(0000) GS:ffff8e726bc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000118d1e000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __die+0x1e/0x60 +# ? page_fault_oops+0x17b/0x4a0 +# ? srso_return_thunk+0x5/0x5f +# ? exc_page_fault+0x6d/0x230 +# ? asm_exc_page_fault+0x26/0x30 +# ? compose_active_planes+0x1a3/0x760 [vkms] +# vkms_composer_worker+0x205/0x240 [vkms] +# process_one_work+0x201/0x6c0 +# ? lock_is_held_type+0x9e/0x110 +# worker_thread+0x17e/0x320 +# ? __pfx_worker_thread+0x10/0x10 +# kthread+0xce/0x100 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork+0x2f/0x50 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork_asm+0x1a/0x30 +# +# Modules linked in: vkms +# CR2: 000000000000001c +# ---[ end trace 0000000000000000 ]--- +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffff9fb640efbd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff8e7241926000 RSI: ffff8e7241927ffc RDI: 000000000b7fb767 +# RBP: ffff9fb640efbde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff8e7241b09a80 R11: 0000000000000000 R12: ffff8e7241cd6200 +# R13: 0000000000000013 R14: 0000000000000000 R15: ffff8e7241b09a80 +# FS: 0000000000000000(0000) GS:ffff8e726bc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000118d1e000 CR4: 0000000000350ef0 + +# DEBUG - Begin test kms_cursor_crc@cursor-rapid-movement-128x42 +# ------------[ cut here ]------------ +# WARNING: CPU: 0 PID: 2933 at drivers/gpu/drm/vkms/vkms_crtc.c:139 vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Modules linked in: vkms +# CPU: 0 UID: 0 PID: 2933 Comm: kms_cursor_crc Not tainted 6.13.0-rc2-g5219242748c8 #1 +# Hardware name: ChromiumOS crosvm, BIOS 0 +# RIP: 0010:vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Code: f7 48 89 f3 e8 d0 bf 6e d0 48 8b 83 50 01 00 00 a8 01 75 15 48 8b bb a0 01 00 00 e8 59 05 15 d0 48 89 df 5b e9 50 05 15 d0 90 <0f> 0b 90 eb e5 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 +# RSP: 0018:ffffa14cc08b3b08 EFLAGS: 00010202 +# RAX: ffff9b864084b605 RBX: ffff9b8641ba4600 RCX: ffffffff91c96b65 +# RDX: ffffffff90ad1f80 RSI: 0000000000000000 RDI: 0000000000000000 +# RBP: 0000000000000000 R08: 0000000000000034 R09: 0000000000000002 +# R10: 0000000047dd15a5 R11: 00000000547dd15a R12: ffff9b864099c000 +# R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 +# FS: 00007f4f437ab6c0(0000) GS:ffff9b866bc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f4f44c3cd40 CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __warn+0x8c/0x190 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? report_bug+0x164/0x190 +# ? handle_bug+0x54/0x90 +# ? exc_invalid_op+0x17/0x70 +# ? asm_exc_invalid_op+0x1a/0x20 +# ? __pfx_drm_property_free_blob+0x10/0x10 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? vkms_atomic_crtc_destroy_state+0x10/0x40 [vkms] +# drm_atomic_state_default_clear+0x137/0x2f0 +# __drm_atomic_state_free+0x6c/0xb0 +# drm_atomic_helper_update_plane+0x100/0x150 +# drm_mode_cursor_universal+0x10e/0x270 +# drm_mode_cursor_common+0x115/0x240 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# drm_mode_cursor_ioctl+0x4a/0x70 +# drm_ioctl_kernel+0xb0/0x110 +# drm_ioctl+0x235/0x4b0 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# __x64_sys_ioctl+0x92/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f4f44960c5b +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffcdfb0b560 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 0000000000000060 RCX: 00007f4f44960c5b +# RDX: 00007ffcdfb0b5f0 RSI: 00000000c01c64a3 RDI: 0000000000000005 +# RBP: 00007ffcdfb0b5f0 R08: 0000000000000007 R09: 000055c3a5801a30 +# R10: 3107764f00e1f281 R11: 0000000000000246 R12: 00000000c01c64a3 +# R13: 0000000000000005 R14: 000055c38b7e42c8 R15: 000055c3a579aab0 +# +# irq event stamp: 58747 +# hardirqs last enabled at (58753): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (58758): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (47324): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (47307): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI +# CPU: 1 UID: 0 PID: 11 Comm: kworker/u8:0 Tainted: G W 6.13.0-rc2-g5219242748c8 #1 +# Tainted: [W]=WARN +# Hardware name: ChromiumOS crosvm, BIOS 0 +# Workqueue: vkms_composer vkms_composer_worker [vkms] +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __die+0x1e/0x60 +# ? page_fault_oops+0x17b/0x4a0 +# ? __kvmalloc_node_noprof+0x3e/0xc0 +# ? exc_page_fault+0x6d/0x230 +# ? asm_exc_page_fault+0x26/0x30 +# ? compose_active_planes+0x1a3/0x760 [vkms] +# vkms_composer_worker+0x205/0x240 [vkms] +# process_one_work+0x201/0x6c0 +# ? lock_is_held_type+0x9e/0x110 +# worker_thread+0x17e/0x320 +# ? __pfx_worker_thread+0x10/0x10 +# kthread+0xce/0x100 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork+0x2f/0x50 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork_asm+0x1a/0x30 +# +# Modules linked in: vkms +# CR2: 000000000000001c +# ---[ end trace 0000000000000000 ]--- +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 + +# DEBUG - Begin test kms_cursor_crc@cursor-rapid-movement-32x32 +# ------------[ cut here ]------------ +# WARNING: CPU: 0 PID: 2933 at drivers/gpu/drm/vkms/vkms_crtc.c:139 vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Modules linked in: vkms +# CPU: 0 UID: 0 PID: 2933 Comm: kms_cursor_crc Not tainted 6.13.0-rc2-g5219242748c8 #1 +# Hardware name: ChromiumOS crosvm, BIOS 0 +# RIP: 0010:vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Code: f7 48 89 f3 e8 d0 bf 6e d0 48 8b 83 50 01 00 00 a8 01 75 15 48 8b bb a0 01 00 00 e8 59 05 15 d0 48 89 df 5b e9 50 05 15 d0 90 <0f> 0b 90 eb e5 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 +# RSP: 0018:ffffa14cc08b3b08 EFLAGS: 00010202 +# RAX: ffff9b864084b605 RBX: ffff9b8641ba4600 RCX: ffffffff91c96b65 +# RDX: ffffffff90ad1f80 RSI: 0000000000000000 RDI: 0000000000000000 +# RBP: 0000000000000000 R08: 0000000000000034 R09: 0000000000000002 +# R10: 0000000047dd15a5 R11: 00000000547dd15a R12: ffff9b864099c000 +# R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 +# FS: 00007f4f437ab6c0(0000) GS:ffff9b866bc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f4f44c3cd40 CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __warn+0x8c/0x190 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? report_bug+0x164/0x190 +# ? handle_bug+0x54/0x90 +# ? exc_invalid_op+0x17/0x70 +# ? asm_exc_invalid_op+0x1a/0x20 +# ? __pfx_drm_property_free_blob+0x10/0x10 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? vkms_atomic_crtc_destroy_state+0x10/0x40 [vkms] +# drm_atomic_state_default_clear+0x137/0x2f0 +# __drm_atomic_state_free+0x6c/0xb0 +# drm_atomic_helper_update_plane+0x100/0x150 +# drm_mode_cursor_universal+0x10e/0x270 +# drm_mode_cursor_common+0x115/0x240 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# drm_mode_cursor_ioctl+0x4a/0x70 +# drm_ioctl_kernel+0xb0/0x110 +# drm_ioctl+0x235/0x4b0 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# __x64_sys_ioctl+0x92/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f4f44960c5b +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffcdfb0b560 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 0000000000000060 RCX: 00007f4f44960c5b +# RDX: 00007ffcdfb0b5f0 RSI: 00000000c01c64a3 RDI: 0000000000000005 +# RBP: 00007ffcdfb0b5f0 R08: 0000000000000007 R09: 000055c3a5801a30 +# R10: 3107764f00e1f281 R11: 0000000000000246 R12: 00000000c01c64a3 +# R13: 0000000000000005 R14: 000055c38b7e42c8 R15: 000055c3a579aab0 +# +# irq event stamp: 58747 +# hardirqs last enabled at (58753): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (58758): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (47324): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (47307): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI +# CPU: 1 UID: 0 PID: 11 Comm: kworker/u8:0 Tainted: G W 6.13.0-rc2-g5219242748c8 #1 +# Tainted: [W]=WARN +# Hardware name: ChromiumOS crosvm, BIOS 0 +# Workqueue: vkms_composer vkms_composer_worker [vkms] +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __die+0x1e/0x60 +# ? page_fault_oops+0x17b/0x4a0 +# ? __kvmalloc_node_noprof+0x3e/0xc0 +# ? exc_page_fault+0x6d/0x230 +# ? asm_exc_page_fault+0x26/0x30 +# ? compose_active_planes+0x1a3/0x760 [vkms] +# vkms_composer_worker+0x205/0x240 [vkms] +# process_one_work+0x201/0x6c0 +# ? lock_is_held_type+0x9e/0x110 +# worker_thread+0x17e/0x320 +# ? __pfx_worker_thread+0x10/0x10 +# kthread+0xce/0x100 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork+0x2f/0x50 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork_asm+0x1a/0x30 +# +# Modules linked in: vkms +# CR2: 000000000000001c +# ---[ end trace 0000000000000000 ]--- +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 + +# DEBUG - Begin test kms_cursor_crc@cursor-rapid-movement-32x32 +# ------------[ cut here ]------------ +# WARNING: CPU: 0 PID: 2933 at drivers/gpu/drm/vkms/vkms_crtc.c:139 vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Modules linked in: vkms +# CPU: 0 UID: 0 PID: 2933 Comm: kms_cursor_crc Not tainted 6.13.0-rc2-g5219242748c8 #1 +# Hardware name: ChromiumOS crosvm, BIOS 0 +# RIP: 0010:vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Code: f7 48 89 f3 e8 d0 bf 6e d0 48 8b 83 50 01 00 00 a8 01 75 15 48 8b bb a0 01 00 00 e8 59 05 15 d0 48 89 df 5b e9 50 05 15 d0 90 <0f> 0b 90 eb e5 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 +# RSP: 0018:ffffa14cc08b3b08 EFLAGS: 00010202 +# RAX: ffff9b864084b605 RBX: ffff9b8641ba4600 RCX: ffffffff91c96b65 +# RDX: ffffffff90ad1f80 RSI: 0000000000000000 RDI: 0000000000000000 +# RBP: 0000000000000000 R08: 0000000000000034 R09: 0000000000000002 +# R10: 0000000047dd15a5 R11: 00000000547dd15a R12: ffff9b864099c000 +# R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 +# FS: 00007f4f437ab6c0(0000) GS:ffff9b866bc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 00007f4f44c3cd40 CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __warn+0x8c/0x190 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? report_bug+0x164/0x190 +# ? handle_bug+0x54/0x90 +# ? exc_invalid_op+0x17/0x70 +# ? asm_exc_invalid_op+0x1a/0x20 +# ? __pfx_drm_property_free_blob+0x10/0x10 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? vkms_atomic_crtc_destroy_state+0x10/0x40 [vkms] +# drm_atomic_state_default_clear+0x137/0x2f0 +# __drm_atomic_state_free+0x6c/0xb0 +# drm_atomic_helper_update_plane+0x100/0x150 +# drm_mode_cursor_universal+0x10e/0x270 +# drm_mode_cursor_common+0x115/0x240 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# drm_mode_cursor_ioctl+0x4a/0x70 +# drm_ioctl_kernel+0xb0/0x110 +# drm_ioctl+0x235/0x4b0 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# __x64_sys_ioctl+0x92/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f4f44960c5b +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffcdfb0b560 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 0000000000000060 RCX: 00007f4f44960c5b +# RDX: 00007ffcdfb0b5f0 RSI: 00000000c01c64a3 RDI: 0000000000000005 +# RBP: 00007ffcdfb0b5f0 R08: 0000000000000007 R09: 000055c3a5801a30 +# R10: 3107764f00e1f281 R11: 0000000000000246 R12: 00000000c01c64a3 +# R13: 0000000000000005 R14: 000055c38b7e42c8 R15: 000055c3a579aab0 +# +# irq event stamp: 58747 +# hardirqs last enabled at (58753): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (58758): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (47324): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (47307): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI +# CPU: 1 UID: 0 PID: 11 Comm: kworker/u8:0 Tainted: G W 6.13.0-rc2-g5219242748c8 #1 +# Tainted: [W]=WARN +# Hardware name: ChromiumOS crosvm, BIOS 0 +# Workqueue: vkms_composer vkms_composer_worker [vkms] +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __die+0x1e/0x60 +# ? page_fault_oops+0x17b/0x4a0 +# ? __kvmalloc_node_noprof+0x3e/0xc0 +# ? exc_page_fault+0x6d/0x230 +# ? asm_exc_page_fault+0x26/0x30 +# ? compose_active_planes+0x1a3/0x760 [vkms] +# vkms_composer_worker+0x205/0x240 [vkms] +# process_one_work+0x201/0x6c0 +# ? lock_is_held_type+0x9e/0x110 +# worker_thread+0x17e/0x320 +# ? __pfx_worker_thread+0x10/0x10 +# kthread+0xce/0x100 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork+0x2f/0x50 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork_asm+0x1a/0x30 +# +# Modules linked in: vkms +# CR2: 000000000000001c +# ---[ end trace 0000000000000000 ]--- +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa14cc005fd20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff9b8669626000 RSI: ffff9b8669627ffc RDI: 00000000348d6c39 +# RBP: ffffa14cc005fde0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff9b8645650eb0 R11: 0000000000000000 R12: ffff9b8641ba5800 +# R13: 0000000000000028 R14: 0000000000000000 R15: ffff9b8645650eb0 +# FS: 0000000000000000(0000) GS:ffff9b866bd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000108c14000 CR4: 0000000000350ef0 + +# DEBUG - Begin test kms_cursor_crc@cursor-rapid-movement-128x128 +# ------------[ cut here ]------------ +# WARNING: CPU: 0 PID: 2898 at drivers/gpu/drm/vkms/vkms_crtc.c:139 vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Modules linked in: vkms +# CPU: 0 UID: 0 PID: 2898 Comm: kms_cursor_crc Not tainted 6.13.0-rc2-g1f006005ebf8 #1 +# Hardware name: ChromiumOS crosvm, BIOS 0 +# RIP: 0010:vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# Code: f7 48 89 f3 e8 d0 bf 6e e1 48 8b 83 50 01 00 00 a8 01 75 15 48 8b bb a0 01 00 00 e8 59 05 15 e1 48 89 df 5b e9 50 05 15 e1 90 <0f> 0b 90 eb e5 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 +# RSP: 0018:ffffa35c4082fb08 EFLAGS: 00010202 +# RAX: ffff8b02c1c0f005 RBX: ffff8b02d4509600 RCX: ffffffffa2c96b65 +# RDX: ffffffffa1ad1f80 RSI: 0000000000000000 RDI: 0000000000000000 +# RBP: 0000000000000000 R08: 0000000000000034 R09: 0000000000000002 +# R10: 0000000030f11ddf R11: 00000000f30f11dd R12: ffff8b02d2528000 +# R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000 +# FS: 00007f4b071ab6c0(0000) GS:ffff8b02efc00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000055f7f8384b48 CR3: 0000000104ef0000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __warn+0x8c/0x190 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? report_bug+0x164/0x190 +# ? handle_bug+0x54/0x90 +# ? exc_invalid_op+0x17/0x70 +# ? asm_exc_invalid_op+0x1a/0x20 +# ? __pfx_drm_property_free_blob+0x10/0x10 +# ? vkms_atomic_crtc_destroy_state+0x31/0x40 [vkms] +# ? vkms_atomic_crtc_destroy_state+0x10/0x40 [vkms] +# drm_atomic_state_default_clear+0x137/0x2f0 +# __drm_atomic_state_free+0x6c/0xb0 +# drm_atomic_helper_update_plane+0x100/0x150 +# drm_mode_cursor_universal+0x10e/0x270 +# drm_mode_cursor_common+0x115/0x240 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# drm_mode_cursor_ioctl+0x4a/0x70 +# drm_ioctl_kernel+0xb0/0x110 +# drm_ioctl+0x235/0x4b0 +# ? __pfx_drm_mode_cursor_ioctl+0x10/0x10 +# __x64_sys_ioctl+0x92/0xc0 +# do_syscall_64+0xbb/0x1d0 +# entry_SYSCALL_64_after_hwframe+0x77/0x7f +# RIP: 0033:0x7f4b0815ac5b +# Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00 +# RSP: 002b:00007ffd726f75e0 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 +# RAX: ffffffffffffffda RBX: 000000000000005d RCX: 00007f4b0815ac5b +# RDX: 00007ffd726f7670 RSI: 00000000c01c64a3 RDI: 0000000000000005 +# RBP: 00007ffd726f7670 R08: 0000000000000007 R09: 000055f7f83e30c0 +# R10: 2b16893c03bd381a R11: 0000000000000246 R12: 00000000c01c64a3 +# R13: 0000000000000005 R14: 000055f7dbee72c8 R15: 000055f7f837bab0 +# +# irq event stamp: 58921 +# hardirqs last enabled at (58927): [] __up_console_sem+0x4d/0x60 +# hardirqs last disabled at (58932): [] __up_console_sem+0x32/0x60 +# softirqs last enabled at (46140): [] handle_softirqs+0x310/0x3f0 +# softirqs last disabled at (46135): [] __irq_exit_rcu+0xa1/0xc0 +# ---[ end trace 0000000000000000 ]--- +# Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI +# CPU: 1 UID: 0 PID: 1657 Comm: kworker/u8:14 Tainted: G W 6.13.0-rc2-g1f006005ebf8 #1 +# Tainted: [W]=WARN +# Hardware name: ChromiumOS crosvm, BIOS 0 +# Workqueue: vkms_composer vkms_composer_worker [vkms] +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa35c40c43d20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff8b02c52c8000 RSI: ffff8b02c52c9ffc RDI: 00000000fa4761c9 +# RBP: ffffa35c40c43de0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff8b02c1c87840 R11: 0000000000000000 R12: ffff8b02d4508800 +# R13: 0000000000000021 R14: 0000000000000000 R15: ffff8b02c1c87840 +# FS: 0000000000000000(0000) GS:ffff8b02efd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +# CR2: 000000000000001c CR3: 0000000104ef0000 CR4: 0000000000350ef0 +# Call Trace: +# +# ? __die+0x1e/0x60 +# ? page_fault_oops+0x17b/0x4a0 +# ? srso_return_thunk+0x5/0x5f +# ? exc_page_fault+0x6d/0x230 +# ? asm_exc_page_fault+0x26/0x30 +# ? compose_active_planes+0x1a3/0x760 [vkms] +# vkms_composer_worker+0x205/0x240 [vkms] +# process_one_work+0x201/0x6c0 +# ? lock_is_held_type+0x9e/0x110 +# worker_thread+0x17e/0x320 +# ? __pfx_worker_thread+0x10/0x10 +# kthread+0xce/0x100 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork+0x2f/0x50 +# ? __pfx_kthread+0x10/0x10 +# ret_from_fork_asm+0x1a/0x30 +# +# Modules linked in: vkms +# CR2: 000000000000001c +# ---[ end trace 0000000000000000 ]--- +# RIP: 0010:compose_active_planes+0x1a3/0x760 [vkms] +# Code: db 4d 89 fa 85 c9 0f 84 32 03 00 00 4d 8b 24 da 48 c7 44 24 60 00 00 00 00 48 c7 44 24 68 00 00 00 00 49 8b 84 24 48 01 00 00 <8b> 50 1c 44 39 ea 0f 8f f3 02 00 00 44 39 68 24 0f 8e e9 02 00 00 +# RSP: 0018:ffffa35c40c43d20 EFLAGS: 00010202 +# RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000002 +# RDX: ffff8b02c52c8000 RSI: ffff8b02c52c9ffc RDI: 00000000fa4761c9 +# RBP: ffffa35c40c43de0 R08: 0000000000000000 R09: 0000000000000000 +# R10: ffff8b02c1c87840 R11: 0000000000000000 R12: ffff8b02d4508800 +# R13: 0000000000000021 R14: 0000000000000000 R15: ffff8b02c1c87840 +# FS: 0000000000000000(0000) GS:ffff8b02efd00000(0000) knlGS:0000000000000000 +# CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 + # Skip driver specific tests ^amdgpu.* ^msm.* @@ -272,3 +802,8 @@ gem_.* i915_.* xe_.* tools_test.* + +# IGT issue. is_joiner_mode() should return false for non-Intel hardware. +# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 +kms_display_modes@extended-mode-basic +kms_display_modes@mst-extended-mode-negative From 7a5cd45fab0a2671aa4ea6d8fb80cea268387176 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sat, 14 Dec 2024 15:37:05 +0200 Subject: [PATCH 0006/1443] drm/encoder_slave: make mode_valid accept const struct drm_display_mode The mode_valid() callbacks of drm_encoder, drm_crtc and drm_bridge accept const struct drm_display_mode argument. Change the mode_valid callback of drm_encoder_slave to also accept const argument. Reviewed-by: Laurent Pinchart Reviewed-by: Lyude Paul Reviewed-by: Maxime Ripard Reviewed-by: Thomas Zimmermann Reviewed-by: Harry Wentland Link: https://patchwork.freedesktop.org/patch/msgid/20241214-drm-connector-mode-valid-const-v2-1-4f9498a4c822@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/i2c/ch7006_drv.c | 2 +- drivers/gpu/drm/i2c/sil164_drv.c | 2 +- drivers/gpu/drm/nouveau/dispnv04/tvnv17.c | 2 +- include/drm/drm_encoder_slave.h | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c index fcb0fcd6c897..db4f95020fae 100644 --- a/drivers/gpu/drm/i2c/ch7006_drv.c +++ b/drivers/gpu/drm/i2c/ch7006_drv.c @@ -104,7 +104,7 @@ static bool ch7006_encoder_mode_fixup(struct drm_encoder *encoder, } static int ch7006_encoder_mode_valid(struct drm_encoder *encoder, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (ch7006_lookup_mode(encoder, mode)) return MODE_OK; diff --git a/drivers/gpu/drm/i2c/sil164_drv.c b/drivers/gpu/drm/i2c/sil164_drv.c index c17afa025d9d..6e5da2c5762f 100644 --- a/drivers/gpu/drm/i2c/sil164_drv.c +++ b/drivers/gpu/drm/i2c/sil164_drv.c @@ -255,7 +255,7 @@ sil164_encoder_restore(struct drm_encoder *encoder) static int sil164_encoder_mode_valid(struct drm_encoder *encoder, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct sil164_priv *priv = to_sil164_priv(encoder); diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c index 3ecb101d23e9..35ad4e10d273 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c @@ -308,7 +308,7 @@ static int nv17_tv_get_modes(struct drm_encoder *encoder, } static int nv17_tv_mode_valid(struct drm_encoder *encoder, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct nv17_tv_norm_params *tv_norm = get_tv_norm(encoder); diff --git a/include/drm/drm_encoder_slave.h b/include/drm/drm_encoder_slave.h index 49172166a164..b526643833dc 100644 --- a/include/drm/drm_encoder_slave.h +++ b/include/drm/drm_encoder_slave.h @@ -85,7 +85,7 @@ struct drm_encoder_slave_funcs { * @mode_valid: Analogous to &drm_encoder_helper_funcs @mode_valid. */ int (*mode_valid)(struct drm_encoder *encoder, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); /** * @mode_set: Analogous to &drm_encoder_helper_funcs @mode_set * callback. Wrapped by drm_i2c_encoder_mode_set(). From b255ce4388e09f14311e7912d0ccd45a14a08d66 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sat, 14 Dec 2024 15:37:06 +0200 Subject: [PATCH 0007/1443] drm/amdgpu: don't change mode in amdgpu_dm_connector_mode_valid() Make amdgpu_dm_connector_mode_valid() duplicate the mode during the test rather than modifying the passed mode. This is a preparation to converting the mode_valid() callback of drm_connector to take a const struct drm_display_mode argument. Reviewed-by: Laurent Pinchart Reviewed-by: Maxime Ripard Reviewed-by: Thomas Zimmermann Reviewed-by: Harry Wentland Link: https://patchwork.freedesktop.org/patch/msgid/20241214-drm-connector-mode-valid-const-v2-2-4f9498a4c822@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index dea640816e19..f92ad3b77c19 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -7420,6 +7420,7 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec { int result = MODE_ERROR; struct dc_sink *dc_sink; + struct drm_display_mode *test_mode; /* TODO: Unhardcode stream count */ struct dc_stream_state *stream; struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); @@ -7444,11 +7445,16 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec goto fail; } - drm_mode_set_crtcinfo(mode, 0); + test_mode = drm_mode_duplicate(connector->dev, mode); + if (!test_mode) + goto fail; - stream = create_validate_stream_for_sink(aconnector, mode, + drm_mode_set_crtcinfo(test_mode, 0); + + stream = create_validate_stream_for_sink(aconnector, test_mode, to_dm_connector_state(connector->state), NULL); + drm_mode_destroy(connector->dev, test_mode); if (stream) { dc_stream_release(stream); result = MODE_OK; From 5f011b442006ccb29044263df10843de80fc0b14 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sat, 14 Dec 2024 15:37:07 +0200 Subject: [PATCH 0008/1443] drm/sti: hda: pass const struct drm_display_mode* to hda_get_mode_idx() Make hda_get_mode_idx() take a const struct drm_display_mode pointer instead of just raw struct drm_display_mode. This is a preparation to converting the mode_valid() callback of drm_connector to take a const struct drm_display_mode argument. Acked-by: Raphael Gallais-Pou Reviewed-by: Laurent Pinchart Reviewed-by: Maxime Ripard Reviewed-by: Thomas Zimmermann Reviewed-by: Harry Wentland Link: https://patchwork.freedesktop.org/patch/msgid/20241214-drm-connector-mode-valid-const-v2-3-4f9498a4c822@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/sti/sti_hda.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index b12863bea955..57aa260d1e46 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -280,12 +280,12 @@ static void hda_write(struct sti_hda *hda, u32 val, int offset) * * Return true if mode is found */ -static bool hda_get_mode_idx(struct drm_display_mode mode, int *idx) +static bool hda_get_mode_idx(const struct drm_display_mode *mode, int *idx) { unsigned int i; for (i = 0; i < ARRAY_SIZE(hda_supported_modes); i++) - if (drm_mode_equal(&hda_supported_modes[i].mode, &mode)) { + if (drm_mode_equal(&hda_supported_modes[i].mode, mode)) { *idx = i; return true; } @@ -443,7 +443,7 @@ static void sti_hda_pre_enable(struct drm_bridge *bridge) if (clk_prepare_enable(hda->clk_hddac)) DRM_ERROR("Failed to prepare/enable hda_hddac clk\n"); - if (!hda_get_mode_idx(hda->mode, &mode_idx)) { + if (!hda_get_mode_idx(&hda->mode, &mode_idx)) { DRM_ERROR("Undefined mode\n"); return; } @@ -526,7 +526,7 @@ static void sti_hda_set_mode(struct drm_bridge *bridge, drm_mode_copy(&hda->mode, mode); - if (!hda_get_mode_idx(hda->mode, &mode_idx)) { + if (!hda_get_mode_idx(&hda->mode, &mode_idx)) { DRM_ERROR("Undefined mode\n"); return; } @@ -614,7 +614,7 @@ sti_hda_connector_mode_valid(struct drm_connector *connector, = to_sti_hda_connector(connector); struct sti_hda *hda = hda_connector->hda; - if (!hda_get_mode_idx(*mode, &idx)) { + if (!hda_get_mode_idx(mode, &idx)) { return MODE_BAD; } else { result = clk_round_rate(hda->clk_pix, target); From 66df9debcb29d14802912ed79a9cf9ba721b51a4 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sat, 14 Dec 2024 15:37:08 +0200 Subject: [PATCH 0009/1443] drm/connector: make mode_valid_ctx take a const struct drm_display_mode The mode_valid() callbacks of drm_encoder, drm_crtc and drm_bridge take a const struct drm_display_mode argument. Change the mode_valid_ctx callback of drm_connector to also take a const argument. Acked-by: Jani Nikula Reviewed-by: Laurent Pinchart Reviewed-by: Maxime Ripard Reviewed-by: Thomas Zimmermann Reviewed-by: Harry Wentland Link: https://patchwork.freedesktop.org/patch/msgid/20241214-drm-connector-mode-valid-const-v2-4-4f9498a4c822@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/i915/display/intel_dp_mst.c | 6 +++--- include/drm/drm_modeset_helper_vtables.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 56ca571e534c..0812a3fd8b37 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -1432,9 +1432,9 @@ static int intel_dp_mst_get_modes(struct drm_connector *connector) static int intel_dp_mst_mode_valid_ctx(struct drm_connector *connector, - struct drm_display_mode *mode, - struct drm_modeset_acquire_ctx *ctx, - enum drm_mode_status *status) + const struct drm_display_mode *mode, + struct drm_modeset_acquire_ctx *ctx, + enum drm_mode_status *status) { struct drm_i915_private *dev_priv = to_i915(connector->dev); struct intel_connector *intel_connector = to_intel_connector(connector); diff --git a/include/drm/drm_modeset_helper_vtables.h b/include/drm/drm_modeset_helper_vtables.h index ec59015aec3c..fa9ee6a128be 100644 --- a/include/drm/drm_modeset_helper_vtables.h +++ b/include/drm/drm_modeset_helper_vtables.h @@ -1006,7 +1006,7 @@ struct drm_connector_helper_funcs { * */ int (*mode_valid_ctx)(struct drm_connector *connector, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_modeset_acquire_ctx *ctx, enum drm_mode_status *status); From 26d6fd81916e62d2b0568d9756e5f9c33f0f9b7a Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sat, 14 Dec 2024 15:37:09 +0200 Subject: [PATCH 0010/1443] drm/connector: make mode_valid take a const struct drm_display_mode The mode_valid() callbacks of drm_encoder, drm_crtc and drm_bridge take a const struct drm_display_mode argument. Change the mode_valid callback of drm_connector to also take a const argument. Acked-by: Jani Nikula Reviewed-by: Liviu Dudau Reviewed-by: Raphael Gallais-Pou Reviewed-by: Laurent Pinchart Reviewed-by: Lyude Paul Reviewed-by: Maxime Ripard Reviewed-by: Thomas Zimmermann Reviewed-by: Harry Wentland Link: https://patchwork.freedesktop.org/patch/msgid/20241214-drm-connector-mode-valid-const-v2-5-4f9498a4c822@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/atombios_dp.c | 2 +- drivers/gpu/drm/amd/amdgpu/atombios_dp.h | 2 +- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 2 +- .../gpu/drm/arm/display/komeda/komeda_wb_connector.c | 2 +- drivers/gpu/drm/arm/malidp_mw.c | 2 +- drivers/gpu/drm/bridge/adv7511/adv7511_drv.c | 2 +- drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c | 2 +- .../gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 7 ------- drivers/gpu/drm/display/drm_bridge_connector.c | 2 +- drivers/gpu/drm/display/drm_hdmi_state_helper.c | 2 +- drivers/gpu/drm/drm_crtc_helper_internal.h | 2 +- drivers/gpu/drm/drm_probe_helper.c | 2 +- drivers/gpu/drm/exynos/exynos_hdmi.c | 2 +- drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c | 2 +- drivers/gpu/drm/gma500/cdv_intel_crt.c | 2 +- drivers/gpu/drm/gma500/cdv_intel_dp.c | 2 +- drivers/gpu/drm/gma500/cdv_intel_hdmi.c | 2 +- drivers/gpu/drm/gma500/cdv_intel_lvds.c | 2 +- drivers/gpu/drm/gma500/oaktrail_hdmi.c | 2 +- drivers/gpu/drm/gma500/psb_intel_drv.h | 2 +- drivers/gpu/drm/gma500/psb_intel_lvds.c | 2 +- drivers/gpu/drm/gma500/psb_intel_sdvo.c | 2 +- drivers/gpu/drm/i915/display/dvo_ch7017.c | 2 +- drivers/gpu/drm/i915/display/dvo_ch7xxx.c | 2 +- drivers/gpu/drm/i915/display/dvo_ivch.c | 2 +- drivers/gpu/drm/i915/display/dvo_ns2501.c | 2 +- drivers/gpu/drm/i915/display/dvo_sil164.c | 2 +- drivers/gpu/drm/i915/display/dvo_tfp410.c | 2 +- drivers/gpu/drm/i915/display/icl_dsi.c | 2 +- drivers/gpu/drm/i915/display/intel_crt.c | 2 +- drivers/gpu/drm/i915/display/intel_dp.c | 2 +- drivers/gpu/drm/i915/display/intel_dsi.c | 2 +- drivers/gpu/drm/i915/display/intel_dsi.h | 2 +- drivers/gpu/drm/i915/display/intel_dvo.c | 2 +- drivers/gpu/drm/i915/display/intel_dvo_dev.h | 2 +- drivers/gpu/drm/i915/display/intel_hdmi.c | 2 +- drivers/gpu/drm/i915/display/intel_lvds.c | 2 +- drivers/gpu/drm/i915/display/intel_sdvo.c | 2 +- drivers/gpu/drm/i915/display/intel_tv.c | 2 +- drivers/gpu/drm/i915/display/vlv_dsi.c | 2 +- drivers/gpu/drm/imx/ipuv3/imx-tve.c | 2 +- drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c | 2 +- drivers/gpu/drm/nouveau/dispnv50/disp.c | 2 +- drivers/gpu/drm/nouveau/nouveau_connector.c | 2 +- drivers/gpu/drm/qxl/qxl_display.c | 2 +- drivers/gpu/drm/radeon/atombios_dp.c | 2 +- drivers/gpu/drm/radeon/radeon_connectors.c | 10 +++++----- drivers/gpu/drm/radeon/radeon_mode.h | 2 +- drivers/gpu/drm/rockchip/cdn-dp-core.c | 2 +- drivers/gpu/drm/rockchip/inno_hdmi.c | 4 ++-- drivers/gpu/drm/rockchip/rk3066_hdmi.c | 2 +- drivers/gpu/drm/sti/sti_dvo.c | 2 +- drivers/gpu/drm/sti/sti_hda.c | 2 +- drivers/gpu/drm/sti/sti_hdmi.c | 2 +- drivers/gpu/drm/tegra/dsi.c | 2 +- drivers/gpu/drm/tegra/hdmi.c | 2 +- drivers/gpu/drm/tegra/sor.c | 2 +- drivers/gpu/drm/vc4/vc4_txp.c | 2 +- drivers/gpu/drm/virtio/virtgpu_display.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_kms.h | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c | 2 +- include/drm/display/drm_hdmi_state_helper.h | 2 +- include/drm/drm_modeset_helper_vtables.h | 2 +- 66 files changed, 73 insertions(+), 80 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 344e0a9ee08a..5e375e9c4f5d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -674,7 +674,7 @@ static int amdgpu_connector_lvds_get_modes(struct drm_connector *connector) } static enum drm_mode_status amdgpu_connector_lvds_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_encoder *encoder = amdgpu_connector_best_single_encoder(connector); @@ -839,7 +839,7 @@ static int amdgpu_connector_vga_get_modes(struct drm_connector *connector) } static enum drm_mode_status amdgpu_connector_vga_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct amdgpu_device *adev = drm_to_adev(dev); @@ -1196,7 +1196,7 @@ static void amdgpu_connector_dvi_force(struct drm_connector *connector) } static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct amdgpu_device *adev = drm_to_adev(dev); @@ -1464,7 +1464,7 @@ amdgpu_connector_dp_detect(struct drm_connector *connector, bool force) } static enum drm_mode_status amdgpu_connector_dp_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); struct amdgpu_connector_atom_dig *amdgpu_dig_connector = amdgpu_connector->con_priv; diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c index 622634c08c7b..521b9faab180 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c @@ -430,7 +430,7 @@ void amdgpu_atombios_dp_set_link_config(struct drm_connector *connector, } int amdgpu_atombios_dp_mode_valid_helper(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); struct amdgpu_connector_atom_dig *dig_connector; diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.h b/drivers/gpu/drm/amd/amdgpu/atombios_dp.h index f59d85eaddf0..3e24acf8133f 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.h +++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.h @@ -32,7 +32,7 @@ int amdgpu_atombios_dp_get_panel_mode(struct drm_encoder *encoder, void amdgpu_atombios_dp_set_link_config(struct drm_connector *connector, const struct drm_display_mode *mode); int amdgpu_atombios_dp_mode_valid_helper(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); bool amdgpu_atombios_dp_needs_link_train(struct amdgpu_connector *amdgpu_connector); void amdgpu_atombios_dp_set_rx_power_state(struct drm_connector *connector, u8 power_state); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index f92ad3b77c19..35c486ca74f6 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -7416,7 +7416,7 @@ create_validate_stream_for_sink(struct amdgpu_dm_connector *aconnector, } enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { int result = MODE_ERROR; struct dc_sink *dc_sink; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 6464a8378387..2cf9f6ca20bc 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -945,7 +945,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, int link_index); enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); void dm_restore_drm_connector_state(struct drm_device *dev, struct drm_connector *connector); diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c b/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c index ebccb74306a7..e075a6b317cc 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c @@ -88,7 +88,7 @@ komeda_wb_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status komeda_wb_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct drm_mode_config *mode_config = &dev->mode_config; diff --git a/drivers/gpu/drm/arm/malidp_mw.c b/drivers/gpu/drm/arm/malidp_mw.c index 2577f0cef8fc..600af5ad81b1 100644 --- a/drivers/gpu/drm/arm/malidp_mw.c +++ b/drivers/gpu/drm/arm/malidp_mw.c @@ -43,7 +43,7 @@ static int malidp_mw_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status malidp_mw_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct drm_mode_config *mode_config = &dev->mode_config; diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index eb5919b38263..fda66e7876b8 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -847,7 +847,7 @@ static int adv7511_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status adv7511_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct adv7511 *adv = connector_to_adv7511(connector); diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c index d081850e3c03..25345c3274e3 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c @@ -1619,7 +1619,7 @@ bool cdns_mhdp_bandwidth_ok(struct cdns_mhdp_device *mhdp, static enum drm_mode_status cdns_mhdp_mode_valid(struct drm_connector *conn, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct cdns_mhdp_device *mhdp = connector_to_mhdp(conn); diff --git a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c index a3dcee62e7a5..a47aabf134fd 100644 --- a/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c +++ b/drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c @@ -115,16 +115,9 @@ static int ge_b850v3_lvds_get_modes(struct drm_connector *connector) return num_modes; } -static enum drm_mode_status ge_b850v3_lvds_mode_valid( - struct drm_connector *connector, struct drm_display_mode *mode) -{ - return MODE_OK; -} - static const struct drm_connector_helper_funcs ge_b850v3_lvds_connector_helper_funcs = { .get_modes = ge_b850v3_lvds_get_modes, - .mode_valid = ge_b850v3_lvds_mode_valid, }; static enum drm_connector_status ge_b850v3_lvds_bridge_detect(struct drm_bridge *bridge) diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c index 32108307de66..0397e62f9cbc 100644 --- a/drivers/gpu/drm/display/drm_bridge_connector.c +++ b/drivers/gpu/drm/display/drm_bridge_connector.c @@ -327,7 +327,7 @@ static int drm_bridge_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status drm_bridge_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_bridge_connector *bridge_connector = to_drm_bridge_connector(connector); diff --git a/drivers/gpu/drm/display/drm_hdmi_state_helper.c b/drivers/gpu/drm/display/drm_hdmi_state_helper.c index cfc2aaee1da0..2691e8b3e480 100644 --- a/drivers/gpu/drm/display/drm_hdmi_state_helper.c +++ b/drivers/gpu/drm/display/drm_hdmi_state_helper.c @@ -539,7 +539,7 @@ EXPORT_SYMBOL(drm_atomic_helper_connector_hdmi_check); */ enum drm_mode_status drm_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { unsigned long long clock; diff --git a/drivers/gpu/drm/drm_crtc_helper_internal.h b/drivers/gpu/drm/drm_crtc_helper_internal.h index 8059f65c5d6c..bae73936acf9 100644 --- a/drivers/gpu/drm/drm_crtc_helper_internal.h +++ b/drivers/gpu/drm/drm_crtc_helper_internal.h @@ -43,7 +43,7 @@ enum drm_mode_status drm_encoder_mode_valid(struct drm_encoder *encoder, const struct drm_display_mode *mode); int drm_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_modeset_acquire_ctx *ctx, enum drm_mode_status *status); diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index 96b266b37ba4..080b55200614 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -202,7 +202,7 @@ enum drm_mode_status drm_encoder_mode_valid(struct drm_encoder *encoder, int drm_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode, + const struct drm_display_mode *mode, struct drm_modeset_acquire_ctx *ctx, enum drm_mode_status *status) { diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 176fd8871759..01813e11e6c6 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -931,7 +931,7 @@ static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock) } static enum drm_mode_status hdmi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct hdmi_context *hdata = connector_to_hdmi(connector); int ret; diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c index c418e8496bdf..84eff7519e32 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c @@ -63,7 +63,7 @@ static int fsl_dcu_drm_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status fsl_dcu_drm_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->hdisplay & 0xf) return MODE_ERROR; diff --git a/drivers/gpu/drm/gma500/cdv_intel_crt.c b/drivers/gpu/drm/gma500/cdv_intel_crt.c index 5a0acd914f76..06fe7480e7af 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_crt.c +++ b/drivers/gpu/drm/gma500/cdv_intel_crt.c @@ -69,7 +69,7 @@ static void cdv_intel_crt_dpms(struct drm_encoder *encoder, int mode) } static enum drm_mode_status cdv_intel_crt_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->flags & DRM_MODE_FLAG_DBLSCAN) return MODE_NO_DBLESCAN; diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c index cc2ed9b3fd2d..53990d27c39f 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_dp.c +++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c @@ -505,7 +505,7 @@ static void cdv_intel_edp_backlight_off (struct gma_encoder *intel_encoder) static enum drm_mode_status cdv_intel_dp_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct gma_encoder *encoder = gma_attached_encoder(connector); struct cdv_intel_dp *intel_dp = encoder->dev_priv; diff --git a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c index 2d95e0471291..f2a3e37ef632 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c +++ b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c @@ -222,7 +222,7 @@ static int cdv_hdmi_get_modes(struct drm_connector *connector) } static enum drm_mode_status cdv_hdmi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->clock > 165000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/gma500/cdv_intel_lvds.c b/drivers/gpu/drm/gma500/cdv_intel_lvds.c index f3a4517bdf27..9276e3676ba0 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_lvds.c +++ b/drivers/gpu/drm/gma500/cdv_intel_lvds.c @@ -153,7 +153,7 @@ static void cdv_intel_lvds_restore(struct drm_connector *connector) } static enum drm_mode_status cdv_intel_lvds_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct drm_psb_private *dev_priv = to_drm_psb_private(dev); diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi.c b/drivers/gpu/drm/gma500/oaktrail_hdmi.c index ed8626c73541..1cf394369127 100644 --- a/drivers/gpu/drm/gma500/oaktrail_hdmi.c +++ b/drivers/gpu/drm/gma500/oaktrail_hdmi.c @@ -514,7 +514,7 @@ static void oaktrail_hdmi_dpms(struct drm_encoder *encoder, int mode) } static enum drm_mode_status oaktrail_hdmi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->clock > 165000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h index 2499fd6a80c9..9dc9dcd1b09f 100644 --- a/drivers/gpu/drm/gma500/psb_intel_drv.h +++ b/drivers/gpu/drm/gma500/psb_intel_drv.h @@ -212,7 +212,7 @@ extern bool psb_intel_lvds_mode_fixup(struct drm_encoder *encoder, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); extern enum drm_mode_status psb_intel_lvds_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); extern int psb_intel_lvds_set_property(struct drm_connector *connector, struct drm_property *property, uint64_t value); diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c index 138f153d38ba..9ad611b5956e 100644 --- a/drivers/gpu/drm/gma500/psb_intel_lvds.c +++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c @@ -331,7 +331,7 @@ static void psb_intel_lvds_restore(struct drm_connector *connector) } enum drm_mode_status psb_intel_lvds_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_psb_private *dev_priv = to_drm_psb_private(connector->dev); struct gma_encoder *gma_encoder = gma_attached_encoder(connector); diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c index 8dafff963ca8..afda40fc4494 100644 --- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c +++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c @@ -1159,7 +1159,7 @@ static void psb_intel_sdvo_dpms(struct drm_encoder *encoder, int mode) } static enum drm_mode_status psb_intel_sdvo_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct psb_intel_sdvo *psb_intel_sdvo = intel_attached_sdvo(connector); diff --git a/drivers/gpu/drm/i915/display/dvo_ch7017.c b/drivers/gpu/drm/i915/display/dvo_ch7017.c index 493e730c685b..206818f9ad49 100644 --- a/drivers/gpu/drm/i915/display/dvo_ch7017.c +++ b/drivers/gpu/drm/i915/display/dvo_ch7017.c @@ -247,7 +247,7 @@ static enum drm_connector_status ch7017_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status ch7017_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->clock > 160000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/i915/display/dvo_ch7xxx.c b/drivers/gpu/drm/i915/display/dvo_ch7xxx.c index 534b8544e0a4..10ab3cc73e58 100644 --- a/drivers/gpu/drm/i915/display/dvo_ch7xxx.c +++ b/drivers/gpu/drm/i915/display/dvo_ch7xxx.c @@ -276,7 +276,7 @@ static enum drm_connector_status ch7xxx_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status ch7xxx_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->clock > 165000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/i915/display/dvo_ivch.c b/drivers/gpu/drm/i915/display/dvo_ivch.c index 0d5cce6051b1..d9c3152d4338 100644 --- a/drivers/gpu/drm/i915/display/dvo_ivch.c +++ b/drivers/gpu/drm/i915/display/dvo_ivch.c @@ -314,7 +314,7 @@ static enum drm_connector_status ivch_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status ivch_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if (mode->clock > 112000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/i915/display/dvo_ns2501.c b/drivers/gpu/drm/i915/display/dvo_ns2501.c index 9d47f8a93e94..cb619ffc974f 100644 --- a/drivers/gpu/drm/i915/display/dvo_ns2501.c +++ b/drivers/gpu/drm/i915/display/dvo_ns2501.c @@ -524,7 +524,7 @@ static enum drm_connector_status ns2501_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status ns2501_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { DRM_DEBUG_KMS ("is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n", diff --git a/drivers/gpu/drm/i915/display/dvo_sil164.c b/drivers/gpu/drm/i915/display/dvo_sil164.c index a8dd40c00997..b42c717085f3 100644 --- a/drivers/gpu/drm/i915/display/dvo_sil164.c +++ b/drivers/gpu/drm/i915/display/dvo_sil164.c @@ -189,7 +189,7 @@ static enum drm_connector_status sil164_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status sil164_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { return MODE_OK; } diff --git a/drivers/gpu/drm/i915/display/dvo_tfp410.c b/drivers/gpu/drm/i915/display/dvo_tfp410.c index d9a0cd753a87..280699438526 100644 --- a/drivers/gpu/drm/i915/display/dvo_tfp410.c +++ b/drivers/gpu/drm/i915/display/dvo_tfp410.c @@ -217,7 +217,7 @@ static enum drm_connector_status tfp410_detect(struct intel_dvo_device *dvo) } static enum drm_mode_status tfp410_mode_valid(struct intel_dvo_device *dvo, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { return MODE_OK; } diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index 8a49f499e3fb..afe90cf0842a 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -1459,7 +1459,7 @@ static void gen11_dsi_post_disable(struct intel_atomic_state *state, } static enum drm_mode_status gen11_dsi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_i915_private *i915 = to_i915(connector->dev); enum drm_mode_status status; diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index 74c1983fe07e..69667cd49ec8 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -352,7 +352,7 @@ static void intel_enable_crt(struct intel_atomic_state *state, static enum drm_mode_status intel_crt_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_display *display = to_intel_display(connector->dev); struct drm_device *dev = connector->dev; diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index ff5ba7b3035f..ea4f775b12a1 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -1371,7 +1371,7 @@ bool intel_dp_has_dsc(const struct intel_connector *connector) static enum drm_mode_status intel_dp_mode_valid(struct drm_connector *_connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_connector *connector = to_intel_connector(_connector); struct intel_dp *intel_dp = intel_attached_dp(connector); diff --git a/drivers/gpu/drm/i915/display/intel_dsi.c b/drivers/gpu/drm/i915/display/intel_dsi.c index 0be46c6c9611..c93a3cf75c52 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi.c +++ b/drivers/gpu/drm/i915/display/intel_dsi.c @@ -58,7 +58,7 @@ int intel_dsi_get_modes(struct drm_connector *connector) } enum drm_mode_status intel_dsi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_i915_private *dev_priv = to_i915(connector->dev); struct intel_connector *intel_connector = to_intel_connector(connector); diff --git a/drivers/gpu/drm/i915/display/intel_dsi.h b/drivers/gpu/drm/i915/display/intel_dsi.h index e8ba4ccd99d3..89c7166a3860 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi.h +++ b/drivers/gpu/drm/i915/display/intel_dsi.h @@ -165,7 +165,7 @@ enum drm_panel_orientation intel_dsi_get_panel_orientation(struct intel_connector *connector); int intel_dsi_get_modes(struct drm_connector *connector); enum drm_mode_status intel_dsi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); struct intel_dsi_host *intel_dsi_host_init(struct intel_dsi *intel_dsi, const struct mipi_dsi_host_ops *funcs, enum port port); diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c index 2d5ffb37eac9..f425bea748c4 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo.c +++ b/drivers/gpu/drm/i915/display/intel_dvo.c @@ -217,7 +217,7 @@ static void intel_enable_dvo(struct intel_atomic_state *state, static enum drm_mode_status intel_dvo_mode_valid(struct drm_connector *_connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_connector *connector = to_intel_connector(_connector); struct drm_i915_private *i915 = to_i915(connector->base.dev); diff --git a/drivers/gpu/drm/i915/display/intel_dvo_dev.h b/drivers/gpu/drm/i915/display/intel_dvo_dev.h index 4bf476656b8c..d20667870e50 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo_dev.h +++ b/drivers/gpu/drm/i915/display/intel_dvo_dev.h @@ -71,7 +71,7 @@ struct intel_dvo_dev_ops { * \return MODE_OK if the mode is valid, or another MODE_* otherwise. */ enum drm_mode_status (*mode_valid)(struct intel_dvo_device *dvo, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); /* * Callback for setting up a video mode after fixups have been made. diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index c6ce6bb88d7c..ae7600ea4a19 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -2025,7 +2025,7 @@ intel_hdmi_mode_clock_valid(struct drm_connector *connector, int clock, static enum drm_mode_status intel_hdmi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_display *display = to_intel_display(connector->dev); struct intel_hdmi *hdmi = intel_attached_hdmi(to_intel_connector(connector)); diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c index 6d7637ad980a..ef2f5c8948a7 100644 --- a/drivers/gpu/drm/i915/display/intel_lvds.c +++ b/drivers/gpu/drm/i915/display/intel_lvds.c @@ -389,7 +389,7 @@ static void intel_lvds_shutdown(struct intel_encoder *encoder) static enum drm_mode_status intel_lvds_mode_valid(struct drm_connector *_connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_connector *connector = to_intel_connector(_connector); struct drm_i915_private *i915 = to_i915(connector->base.dev); diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index 7a28104f68ad..3dcb48879e81 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -1938,7 +1938,7 @@ static void intel_enable_sdvo(struct intel_atomic_state *state, static enum drm_mode_status intel_sdvo_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_i915_private *i915 = to_i915(connector->dev); struct intel_sdvo *intel_sdvo = intel_attached_sdvo(to_intel_connector(connector)); diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c index 27c530218ee6..fa2634aa574a 100644 --- a/drivers/gpu/drm/i915/display/intel_tv.c +++ b/drivers/gpu/drm/i915/display/intel_tv.c @@ -957,7 +957,7 @@ static const struct tv_mode *intel_tv_mode_find(const struct drm_connector_state static enum drm_mode_status intel_tv_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct intel_display *display = to_intel_display(connector->dev); struct drm_i915_private *i915 = to_i915(connector->dev); diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index 9383eedee2d4..e7d9f5897d87 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -1542,7 +1542,7 @@ static const struct drm_encoder_funcs intel_dsi_funcs = { }; static enum drm_mode_status vlv_dsi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_i915_private *i915 = to_i915(connector->dev); diff --git a/drivers/gpu/drm/imx/ipuv3/imx-tve.c b/drivers/gpu/drm/imx/ipuv3/imx-tve.c index 3a3c8a195119..c5629e155d25 100644 --- a/drivers/gpu/drm/imx/ipuv3/imx-tve.c +++ b/drivers/gpu/drm/imx/ipuv3/imx-tve.c @@ -217,7 +217,7 @@ static int imx_tve_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status imx_tve_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct imx_tve *tve = con_to_tve(connector); unsigned long rate; diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c index 7444b75c4215..52e728181b52 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_connector.c @@ -58,7 +58,7 @@ static int mdp4_lvds_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status mdp4_lvds_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct mdp4_lvds_connector *mdp4_lvds_connector = to_mdp4_lvds_connector(connector); diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 8097249612bc..2e09b5a6d155 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -1142,7 +1142,7 @@ nv50_mstc_atomic_best_encoder(struct drm_connector *connector, static enum drm_mode_status nv50_mstc_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct nv50_mstc *mstc = nv50_mstc(connector); struct nouveau_encoder *outp = mstc->mstm->outp; diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 8d5c9c74cbb9..2d26784bde74 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -1074,7 +1074,7 @@ get_tmds_link_bandwidth(struct drm_connector *connector) static enum drm_mode_status nouveau_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct nouveau_connector *nv_connector = nouveau_connector(connector); struct nouveau_encoder *nv_encoder = nv_connector->detected_encoder; diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index bc24af08dfcd..70aff64ced87 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -1044,7 +1044,7 @@ static int qxl_conn_get_modes(struct drm_connector *connector) } static enum drm_mode_status qxl_conn_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *ddev = connector->dev; struct qxl_device *qdev = to_qxl(ddev); diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 6328627b7c34..fa78824931cc 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -467,7 +467,7 @@ void radeon_dp_set_link_config(struct drm_connector *connector, } int radeon_dp_mode_valid_helper(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); struct radeon_connector_atom_dig *dig_connector; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index f9996304d943..9f6a3df951ba 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -806,7 +806,7 @@ static int radeon_lvds_get_modes(struct drm_connector *connector) } static enum drm_mode_status radeon_lvds_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_encoder *encoder = radeon_best_single_encoder(connector); @@ -968,7 +968,7 @@ static int radeon_vga_get_modes(struct drm_connector *connector) } static enum drm_mode_status radeon_vga_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct radeon_device *rdev = dev->dev_private; @@ -1116,7 +1116,7 @@ static int radeon_tv_get_modes(struct drm_connector *connector) } static enum drm_mode_status radeon_tv_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { if ((mode->hdisplay > 1024) || (mode->vdisplay > 768)) return MODE_CLOCK_RANGE; @@ -1447,7 +1447,7 @@ static void radeon_dvi_force(struct drm_connector *connector) } static enum drm_mode_status radeon_dvi_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct radeon_device *rdev = dev->dev_private; @@ -1723,7 +1723,7 @@ radeon_dp_detect(struct drm_connector *connector, bool force) } static enum drm_mode_status radeon_dp_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct radeon_device *rdev = dev->dev_private; diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 4063d3801e81..3102f6c2d055 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -706,7 +706,7 @@ extern int radeon_get_monitor_bpc(struct drm_connector *connector); extern void radeon_connector_hotplug(struct drm_connector *connector); extern int radeon_dp_mode_valid_helper(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); extern void radeon_dp_set_link_config(struct drm_connector *connector, const struct drm_display_mode *mode); extern void radeon_dp_link_train(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c index b17de83b988b..292c31de18f1 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.c +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c @@ -275,7 +275,7 @@ static int cdn_dp_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status cdn_dp_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct cdn_dp_device *dp = connector_to_dp(connector); struct drm_display_info *display_info = &dp->connector.display_info; diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.c b/drivers/gpu/drm/rockchip/inno_hdmi.c index 898d90155057..483ecfeaebb0 100644 --- a/drivers/gpu/drm/rockchip/inno_hdmi.c +++ b/drivers/gpu/drm/rockchip/inno_hdmi.c @@ -471,7 +471,7 @@ static int inno_hdmi_setup(struct inno_hdmi *hdmi, } static enum drm_mode_status inno_hdmi_display_mode_valid(struct inno_hdmi *hdmi, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { unsigned long mpixelclk, max_tolerance; long rounded_refclk; @@ -577,7 +577,7 @@ static int inno_hdmi_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status inno_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct inno_hdmi *hdmi = connector_to_inno_hdmi(connector); diff --git a/drivers/gpu/drm/rockchip/rk3066_hdmi.c b/drivers/gpu/drm/rockchip/rk3066_hdmi.c index 403336397214..f7a460190313 100644 --- a/drivers/gpu/drm/rockchip/rk3066_hdmi.c +++ b/drivers/gpu/drm/rockchip/rk3066_hdmi.c @@ -482,7 +482,7 @@ static int rk3066_hdmi_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status rk3066_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { u32 vic = drm_match_cea_mode(mode); diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c index c6c2abaa1891..4dcddd02629b 100644 --- a/drivers/gpu/drm/sti/sti_dvo.c +++ b/drivers/gpu/drm/sti/sti_dvo.c @@ -349,7 +349,7 @@ static int sti_dvo_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status sti_dvo_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { int target = mode->clock * 1000; int target_min = target - CLK_TOLERANCE_HZ; diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index 57aa260d1e46..14fdc00d2ba0 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -603,7 +603,7 @@ static int sti_hda_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status sti_hda_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { int target = mode->clock * 1000; int target_min = target - CLK_TOLERANCE_HZ; diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c index ca2fe17de4a5..164a34d793d8 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.c +++ b/drivers/gpu/drm/sti/sti_hdmi.c @@ -1011,7 +1011,7 @@ static int sti_hdmi_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status sti_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { int target = mode->clock * 1000; int target_min = target - CLK_TOLERANCE_HZ; diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c index 4a8cd9ed0a94..9bb077558167 100644 --- a/drivers/gpu/drm/tegra/dsi.c +++ b/drivers/gpu/drm/tegra/dsi.c @@ -812,7 +812,7 @@ static const struct drm_connector_funcs tegra_dsi_connector_funcs = { static enum drm_mode_status tegra_dsi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { return MODE_OK; } diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c index e705f8590c13..8cd2969e7d4b 100644 --- a/drivers/gpu/drm/tegra/hdmi.c +++ b/drivers/gpu/drm/tegra/hdmi.c @@ -1137,7 +1137,7 @@ static const struct drm_connector_funcs tegra_hdmi_connector_funcs = { static enum drm_mode_status tegra_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct tegra_output *output = connector_to_output(connector); struct tegra_hdmi *hdmi = to_hdmi(output); diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c index 802d2db7007a..f98f70eda906 100644 --- a/drivers/gpu/drm/tegra/sor.c +++ b/drivers/gpu/drm/tegra/sor.c @@ -1789,7 +1789,7 @@ static int tegra_sor_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status tegra_sor_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { return MODE_OK; } diff --git a/drivers/gpu/drm/vc4/vc4_txp.c b/drivers/gpu/drm/vc4/vc4_txp.c index 4eab069cda75..42acac05fe47 100644 --- a/drivers/gpu/drm/vc4/vc4_txp.c +++ b/drivers/gpu/drm/vc4/vc4_txp.c @@ -196,7 +196,7 @@ static int vc4_txp_connector_get_modes(struct drm_connector *connector) static enum drm_mode_status vc4_txp_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct drm_device *dev = connector->dev; struct drm_mode_config *mode_config = &dev->mode_config; diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c index 64baf2f22d9f..59a45e74a641 100644 --- a/drivers/gpu/drm/virtio/virtgpu_display.c +++ b/drivers/gpu/drm/virtio/virtgpu_display.c @@ -189,7 +189,7 @@ static int virtio_gpu_conn_get_modes(struct drm_connector *connector) } static enum drm_mode_status virtio_gpu_conn_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { struct virtio_gpu_output *output = drm_connector_to_virtio_gpu_output(connector); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 8db38927729b..c896de07f7b0 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -2660,7 +2660,7 @@ int vmw_du_helper_plane_update(struct vmw_du_update_plane *update) * Returns MODE_OK on success, or a drm_mode_status error code. */ enum drm_mode_status vmw_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { enum drm_mode_status ret; struct drm_device *dev = connector->dev; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h index 2a6c6d6581e0..4eab581883e2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h @@ -435,7 +435,7 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, int increment, struct vmw_kms_dirty *dirty); enum drm_mode_status vmw_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); int vmw_connector_get_modes(struct drm_connector *connector); void vmw_kms_helper_validation_finish(struct vmw_private *dev_priv, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index 114a75069e1c..f5d2ed1b0a72 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -839,7 +839,7 @@ static void vmw_stdu_connector_destroy(struct drm_connector *connector) static enum drm_mode_status vmw_stdu_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) + const struct drm_display_mode *mode) { enum drm_mode_status ret; struct drm_device *dev = connector->dev; diff --git a/include/drm/display/drm_hdmi_state_helper.h b/include/drm/display/drm_hdmi_state_helper.h index 9ae19f3caf72..dc2d1c48cc5b 100644 --- a/include/drm/display/drm_hdmi_state_helper.h +++ b/include/drm/display/drm_hdmi_state_helper.h @@ -27,6 +27,6 @@ void drm_atomic_helper_connector_hdmi_force(struct drm_connector *connector); enum drm_mode_status drm_hdmi_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); #endif // DRM_HDMI_STATE_HELPER_H_ diff --git a/include/drm/drm_modeset_helper_vtables.h b/include/drm/drm_modeset_helper_vtables.h index fa9ee6a128be..b62f41f48962 100644 --- a/include/drm/drm_modeset_helper_vtables.h +++ b/include/drm/drm_modeset_helper_vtables.h @@ -967,7 +967,7 @@ struct drm_connector_helper_funcs { * drm_mode_status. */ enum drm_mode_status (*mode_valid)(struct drm_connector *connector, - struct drm_display_mode *mode); + const struct drm_display_mode *mode); /** * @mode_valid_ctx: From 9a0dfe9bacef9c3bf6aaa92b5913945364ef805f Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Mon, 6 Jan 2025 01:00:13 +0200 Subject: [PATCH 0011/1443] drm/nouveau: incorporate I2C TV encoder drivers Chrontel CH7006 and Silicon Image sil164 drivers use drm_encoder_slave interface which is being used only by the nouveau driver. It doesn't make sense to keep this interface inside the DRM subsystem. In preparation to moving this set of helpers to the nouveau driver, move the only two I2C driver that use that interface to the nouveau driver too. Suggested-by: Laurent Pinchart Signed-off-by: Dmitry Baryshkov Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250106-nouveau-encoder-slave-v3-1-1d37d2f2c67f@linaro.org --- arch/arm/configs/multi_v7_defconfig | 4 ++-- arch/parisc/configs/generic-32bit_defconfig | 4 ++-- arch/parisc/configs/generic-64bit_defconfig | 4 ++-- drivers/gpu/drm/i2c/Kconfig | 18 ------------------ drivers/gpu/drm/i2c/Makefile | 6 ------ drivers/gpu/drm/nouveau/Kconfig | 18 ++++++++++++++++++ drivers/gpu/drm/nouveau/dispnv04/Kbuild | 2 ++ drivers/gpu/drm/nouveau/dispnv04/dfp.c | 2 +- drivers/gpu/drm/nouveau/dispnv04/i2c/Kbuild | 5 +++++ .../{ => nouveau/dispnv04}/i2c/ch7006_drv.c | 0 .../{ => nouveau/dispnv04}/i2c/ch7006_mode.c | 0 .../{ => nouveau/dispnv04}/i2c/ch7006_priv.h | 7 ++++--- .../{ => nouveau/dispnv04}/i2c/sil164_drv.c | 3 ++- drivers/gpu/drm/nouveau/dispnv04/tvnv04.c | 2 +- .../drm/nouveau/include/dispnv04}/i2c/ch7006.h | 4 ++-- .../drm/nouveau/include/dispnv04}/i2c/sil164.h | 4 ++-- 16 files changed, 43 insertions(+), 40 deletions(-) create mode 100644 drivers/gpu/drm/nouveau/dispnv04/i2c/Kbuild rename drivers/gpu/drm/{ => nouveau/dispnv04}/i2c/ch7006_drv.c (100%) rename drivers/gpu/drm/{ => nouveau/dispnv04}/i2c/ch7006_mode.c (100%) rename drivers/gpu/drm/{ => nouveau/dispnv04}/i2c/ch7006_priv.h (98%) rename drivers/gpu/drm/{ => nouveau/dispnv04}/i2c/sil164_drv.c (99%) rename {include/drm => drivers/gpu/drm/nouveau/include/dispnv04}/i2c/ch7006.h (97%) rename {include/drm => drivers/gpu/drm/nouveau/include/dispnv04}/i2c/sil164.h (96%) diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index 758276027dbc..4b100ddb0c54 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -713,10 +713,10 @@ CONFIG_VIDEO_ADV7604_CEC=y CONFIG_VIDEO_ML86V7667=m CONFIG_IMX_IPUV3_CORE=m CONFIG_DRM=y -# CONFIG_DRM_I2C_CH7006 is not set -# CONFIG_DRM_I2C_SIL164 is not set CONFIG_DRM_I2C_NXP_TDA998X=m CONFIG_DRM_NOUVEAU=m +# CONFIG_DRM_NOUVEAU_CH7006 is not set +# CONFIG_DRM_NOUVEAU_SIL164 is not set CONFIG_DRM_EXYNOS=m CONFIG_DRM_EXYNOS_FIMD=y CONFIG_DRM_EXYNOS_MIXER=y diff --git a/arch/parisc/configs/generic-32bit_defconfig b/arch/parisc/configs/generic-32bit_defconfig index 5ce258f3fffa..f5fffc24c3bc 100644 --- a/arch/parisc/configs/generic-32bit_defconfig +++ b/arch/parisc/configs/generic-32bit_defconfig @@ -132,11 +132,11 @@ CONFIG_I2C=y CONFIG_HWMON=m CONFIG_DRM=m CONFIG_DRM_DISPLAY_DP_AUX_CEC=y -# CONFIG_DRM_I2C_CH7006 is not set -# CONFIG_DRM_I2C_SIL164 is not set CONFIG_DRM_RADEON=m CONFIG_DRM_NOUVEAU=m # CONFIG_DRM_NOUVEAU_BACKLIGHT is not set +# CONFIG_DRM_NOUVEAU_CH7006 is not set +# CONFIG_DRM_NOUVEAU_SIL164 is not set CONFIG_DRM_VGEM=m CONFIG_DRM_UDL=m CONFIG_DRM_MGAG200=m diff --git a/arch/parisc/configs/generic-64bit_defconfig b/arch/parisc/configs/generic-64bit_defconfig index 19a804860ed5..704f68fbf960 100644 --- a/arch/parisc/configs/generic-64bit_defconfig +++ b/arch/parisc/configs/generic-64bit_defconfig @@ -193,11 +193,11 @@ CONFIG_MEDIA_SUPPORT=m CONFIG_AGP=y CONFIG_AGP_PARISC=y CONFIG_DRM=y -# CONFIG_DRM_I2C_CH7006 is not set -# CONFIG_DRM_I2C_SIL164 is not set CONFIG_DRM_RADEON=y CONFIG_DRM_NOUVEAU=m # CONFIG_DRM_NOUVEAU_BACKLIGHT is not set +# CONFIG_DRM_NOUVEAU_CH7006 is not set +# CONFIG_DRM_NOUVEAU_SIL164 is not set CONFIG_DRM_MGAG200=m CONFIG_FB=y CONFIG_FB_PM2=m diff --git a/drivers/gpu/drm/i2c/Kconfig b/drivers/gpu/drm/i2c/Kconfig index 6f19e1c35e30..d5200f67958e 100644 --- a/drivers/gpu/drm/i2c/Kconfig +++ b/drivers/gpu/drm/i2c/Kconfig @@ -2,24 +2,6 @@ menu "I2C encoder or helper chips" depends on DRM && DRM_KMS_HELPER && I2C -config DRM_I2C_CH7006 - tristate "Chrontel ch7006 TV encoder" - default m if DRM_NOUVEAU - help - Support for Chrontel ch7006 and similar TV encoders, found - on some nVidia video cards. - - This driver is currently only useful if you're also using - the nouveau driver. - -config DRM_I2C_SIL164 - tristate "Silicon Image sil164 TMDS transmitter" - default m if DRM_NOUVEAU - help - Support for sil164 and similar single-link (or dual-link - when used in pairs) TMDS transmitters, used in some nVidia - video cards. - config DRM_I2C_NXP_TDA998X tristate "NXP Semiconductors TDA998X HDMI encoder" default m if DRM_TILCDC diff --git a/drivers/gpu/drm/i2c/Makefile b/drivers/gpu/drm/i2c/Makefile index a962f6f08568..31fd35527d99 100644 --- a/drivers/gpu/drm/i2c/Makefile +++ b/drivers/gpu/drm/i2c/Makefile @@ -1,10 +1,4 @@ # SPDX-License-Identifier: GPL-2.0 -ch7006-y := ch7006_drv.o ch7006_mode.o -obj-$(CONFIG_DRM_I2C_CH7006) += ch7006.o - -sil164-y := sil164_drv.o -obj-$(CONFIG_DRM_I2C_SIL164) += sil164.o - tda998x-y := tda998x_drv.o obj-$(CONFIG_DRM_I2C_NXP_TDA998X) += tda998x.o obj-$(CONFIG_DRM_I2C_NXP_TDA9950) += tda9950.o diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig index ce840300578d..ac76c0787010 100644 --- a/drivers/gpu/drm/nouveau/Kconfig +++ b/drivers/gpu/drm/nouveau/Kconfig @@ -109,3 +109,21 @@ config DRM_NOUVEAU_GSP_DEFAULT help Say Y here if you want to use the GSP codepaths by default on Turing and Ampere GPUs. + +config DRM_NOUVEAU_CH7006 + tristate "Chrontel ch7006 TV encoder" + depends on DRM_NOUVEAU + default m + help + Support for Chrontel ch7006 and similar TV encoders. + + This driver is currently only useful if you're also using + the nouveau driver. + +config DRM_NOUVEAU_SIL164 + tristate "Silicon Image sil164 TMDS transmitter" + depends on DRM_NOUVEAU + default m + help + Support for sil164 and similar single-link (or dual-link + when used in pairs) TMDS transmitters. diff --git a/drivers/gpu/drm/nouveau/dispnv04/Kbuild b/drivers/gpu/drm/nouveau/dispnv04/Kbuild index 975c4e226936..949802882ebd 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/Kbuild +++ b/drivers/gpu/drm/nouveau/dispnv04/Kbuild @@ -10,3 +10,5 @@ nouveau-y += dispnv04/overlay.o nouveau-y += dispnv04/tvmodesnv17.o nouveau-y += dispnv04/tvnv04.o nouveau-y += dispnv04/tvnv17.o + +include $(src)/dispnv04/i2c/Kbuild diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c index 504c421aa176..25fa41edcb3c 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c +++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c @@ -35,7 +35,7 @@ #include "hw.h" #include "nvreg.h" -#include +#include #include diff --git a/drivers/gpu/drm/nouveau/dispnv04/i2c/Kbuild b/drivers/gpu/drm/nouveau/dispnv04/i2c/Kbuild new file mode 100644 index 000000000000..3fddfc97bcb3 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/Kbuild @@ -0,0 +1,5 @@ +ch7006-y := dispnv04/i2c/ch7006_drv.o dispnv04/i2c/ch7006_mode.o +obj-$(CONFIG_DRM_NOUVEAU_CH7006) += ch7006.o + +sil164-y := dispnv04/i2c/sil164_drv.o +obj-$(CONFIG_DRM_NOUVEAU_SIL164) += sil164.o diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c similarity index 100% rename from drivers/gpu/drm/i2c/ch7006_drv.c rename to drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c diff --git a/drivers/gpu/drm/i2c/ch7006_mode.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c similarity index 100% rename from drivers/gpu/drm/i2c/ch7006_mode.c rename to drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h similarity index 98% rename from drivers/gpu/drm/i2c/ch7006_priv.h rename to drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h index 052bdc48a339..d7a91e17510f 100644 --- a/drivers/gpu/drm/i2c/ch7006_priv.h +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h @@ -24,12 +24,13 @@ * */ -#ifndef __DRM_I2C_CH7006_PRIV_H__ -#define __DRM_I2C_CH7006_PRIV_H__ +#ifndef __NOUVEAU_I2C_CH7006_PRIV_H__ +#define __NOUVEAU_I2C_CH7006_PRIV_H__ #include #include -#include + +#include typedef int64_t fixed; #define fixed1 (1LL << 32) diff --git a/drivers/gpu/drm/i2c/sil164_drv.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c similarity index 99% rename from drivers/gpu/drm/i2c/sil164_drv.c rename to drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c index 6e5da2c5762f..39b01c897518 100644 --- a/drivers/gpu/drm/i2c/sil164_drv.c +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c @@ -30,7 +30,8 @@ #include #include #include -#include + +#include struct sil164_priv { struct sil164_encoder_params config; diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c index d3014027a812..4ffd8af6bf8b 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c @@ -32,7 +32,7 @@ #include "hw.h" #include -#include +#include static struct nvkm_i2c_bus_probe nv04_tv_encoder_info[] = { { diff --git a/include/drm/i2c/ch7006.h b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/ch7006.h similarity index 97% rename from include/drm/i2c/ch7006.h rename to drivers/gpu/drm/nouveau/include/dispnv04/i2c/ch7006.h index 5305b9797f93..1a6fa405f85b 100644 --- a/include/drm/i2c/ch7006.h +++ b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/ch7006.h @@ -24,8 +24,8 @@ * */ -#ifndef __DRM_I2C_CH7006_H__ -#define __DRM_I2C_CH7006_H__ +#ifndef __NOUVEAU_I2C_CH7006_H__ +#define __NOUVEAU_I2C_CH7006_H__ /** * struct ch7006_encoder_params diff --git a/include/drm/i2c/sil164.h b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/sil164.h similarity index 96% rename from include/drm/i2c/sil164.h rename to drivers/gpu/drm/nouveau/include/dispnv04/i2c/sil164.h index ddf248693c8b..b86750d7abe1 100644 --- a/include/drm/i2c/sil164.h +++ b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/sil164.h @@ -24,8 +24,8 @@ * */ -#ifndef __DRM_I2C_SIL164_H__ -#define __DRM_I2C_SIL164_H__ +#ifndef __NOUVEAU_I2C_SIL164_H__ +#define __NOUVEAU_I2C_SIL164_H__ /** * struct sil164_encoder_params From a73583107af9b42c0a28045e4e30ec3fab2dbcd1 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Mon, 6 Jan 2025 01:00:14 +0200 Subject: [PATCH 0012/1443] drm/nouveau: vendor in drm_encoder_slave API Nouveau driver is the only user of the drm_encoder_slave API. Rework necessary bits of drm_encoder_slave into the nouveau_i2c_encoder API and drop drm_encoder_slave.c from the DRM KMS helper. Suggested-by: Laurent Pinchart Signed-off-by: Dmitry Baryshkov Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250106-nouveau-encoder-slave-v3-2-1d37d2f2c67f@linaro.org --- drivers/gpu/drm/Makefile | 1 - drivers/gpu/drm/nouveau/dispnv04/Kbuild | 1 + drivers/gpu/drm/nouveau/dispnv04/dfp.c | 15 ++- .../gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c | 30 ++--- .../drm/nouveau/dispnv04/i2c/ch7006_mode.c | 8 +- .../drm/nouveau/dispnv04/i2c/ch7006_priv.h | 4 +- .../gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c | 30 ++--- .../dispnv04/nouveau_i2c_encoder.c} | 95 +++++--------- drivers/gpu/drm/nouveau/dispnv04/tvnv04.c | 22 ++-- drivers/gpu/drm/nouveau/dispnv04/tvnv17.c | 4 +- .../include/dispnv04/i2c/encoder_i2c.h | 121 ++++++++---------- drivers/gpu/drm/nouveau/nouveau_connector.c | 10 +- drivers/gpu/drm/nouveau/nouveau_encoder.h | 13 +- 13 files changed, 147 insertions(+), 207 deletions(-) rename drivers/gpu/drm/{drm_encoder_slave.c => nouveau/dispnv04/nouveau_i2c_encoder.c} (54%) rename include/drm/drm_encoder_slave.h => drivers/gpu/drm/nouveau/include/dispnv04/i2c/encoder_i2c.h (59%) diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 19fb370fbc56..85af94bb907d 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -135,7 +135,6 @@ drm_kms_helper-y := \ drm_atomic_state_helper.o \ drm_crtc_helper.o \ drm_damage_helper.o \ - drm_encoder_slave.o \ drm_flip_work.o \ drm_format_helper.o \ drm_gem_atomic_helper.o \ diff --git a/drivers/gpu/drm/nouveau/dispnv04/Kbuild b/drivers/gpu/drm/nouveau/dispnv04/Kbuild index 949802882ebd..4c7bc6bb81b3 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/Kbuild +++ b/drivers/gpu/drm/nouveau/dispnv04/Kbuild @@ -6,6 +6,7 @@ nouveau-y += dispnv04/dac.o nouveau-y += dispnv04/dfp.o nouveau-y += dispnv04/disp.o nouveau-y += dispnv04/hw.o +nouveau-y += dispnv04/nouveau_i2c_encoder.o nouveau-y += dispnv04/overlay.o nouveau-y += dispnv04/tvmodesnv17.o nouveau-y += dispnv04/tvnv04.o diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c index 25fa41edcb3c..c724bacc67f8 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c +++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c @@ -171,7 +171,7 @@ static struct drm_encoder *get_tmds_slave(struct drm_encoder *encoder) list_for_each_entry(slave, &dev->mode_config.encoder_list, head) { struct dcb_output *slave_dcb = nouveau_encoder(slave)->dcb; - if (slave_dcb->type == DCB_OUTPUT_TMDS && get_slave_funcs(slave) && + if (slave_dcb->type == DCB_OUTPUT_TMDS && get_encoder_i2c_funcs(slave) && slave_dcb->tmdsconf.slave_addr == dcb->tmdsconf.slave_addr) return slave; } @@ -473,8 +473,9 @@ static void nv04_dfp_commit(struct drm_encoder *encoder) /* Init external transmitters */ slave_encoder = get_tmds_slave(encoder); if (slave_encoder) - get_slave_funcs(slave_encoder)->mode_set( - slave_encoder, &nv_encoder->mode, &nv_encoder->mode); + get_encoder_i2c_funcs(slave_encoder)->mode_set(slave_encoder, + &nv_encoder->mode, + &nv_encoder->mode); helper->dpms(encoder, DRM_MODE_DPMS_ON); @@ -614,8 +615,8 @@ static void nv04_dfp_destroy(struct drm_encoder *encoder) { struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder); - if (get_slave_funcs(encoder)) - get_slave_funcs(encoder)->destroy(encoder); + if (get_encoder_i2c_funcs(encoder)) + get_encoder_i2c_funcs(encoder)->destroy(encoder); drm_encoder_cleanup(encoder); kfree(nv_encoder); @@ -649,8 +650,8 @@ static void nv04_tmds_slave_init(struct drm_encoder *encoder) if (type < 0) return; - drm_i2c_encoder_init(dev, to_encoder_slave(encoder), - &bus->i2c, &info[type].dev); + nouveau_i2c_encoder_init(dev, to_encoder_i2c(encoder), + &bus->i2c, &info[type].dev); } static const struct drm_encoder_helper_funcs nv04_lvds_helper_funcs = { diff --git a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c index db4f95020fae..fd2150e07e36 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_drv.c @@ -47,14 +47,14 @@ static void ch7006_encoder_destroy(struct drm_encoder *encoder) drm_property_destroy(encoder->dev, priv->scale_property); kfree(priv); - to_encoder_slave(encoder)->slave_priv = NULL; + to_encoder_i2c(encoder)->encoder_i2c_priv = NULL; - drm_i2c_encoder_destroy(encoder); + nouveau_i2c_encoder_destroy(encoder); } static void ch7006_encoder_dpms(struct drm_encoder *encoder, int mode) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_state *state = &priv->state; @@ -71,7 +71,7 @@ static void ch7006_encoder_dpms(struct drm_encoder *encoder, int mode) static void ch7006_encoder_save(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); ch7006_dbg(client, "\n"); @@ -81,7 +81,7 @@ static void ch7006_encoder_save(struct drm_encoder *encoder) static void ch7006_encoder_restore(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); ch7006_dbg(client, "\n"); @@ -116,7 +116,7 @@ static void ch7006_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *drm_mode, struct drm_display_mode *adjusted_mode) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_encoder_params *params = &priv->params; struct ch7006_state *state = &priv->state; @@ -179,7 +179,7 @@ static void ch7006_encoder_mode_set(struct drm_encoder *encoder, static enum drm_connector_status ch7006_encoder_detect(struct drm_encoder *encoder, struct drm_connector *connector) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_state *state = &priv->state; int det; @@ -285,7 +285,7 @@ static int ch7006_encoder_set_property(struct drm_encoder *encoder, struct drm_property *property, uint64_t val) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_state *state = &priv->state; struct drm_mode_config *conf = &encoder->dev->mode_config; @@ -370,7 +370,7 @@ static int ch7006_encoder_set_property(struct drm_encoder *encoder, return 0; } -static const struct drm_encoder_slave_funcs ch7006_encoder_funcs = { +static const struct nouveau_i2c_encoder_funcs ch7006_encoder_funcs = { .set_config = ch7006_encoder_set_config, .destroy = ch7006_encoder_destroy, .dpms = ch7006_encoder_dpms, @@ -437,7 +437,7 @@ static int ch7006_resume(struct device *dev) static int ch7006_encoder_init(struct i2c_client *client, struct drm_device *dev, - struct drm_encoder_slave *encoder) + struct nouveau_i2c_encoder *encoder) { struct ch7006_priv *priv; int i; @@ -448,8 +448,8 @@ static int ch7006_encoder_init(struct i2c_client *client, if (!priv) return -ENOMEM; - encoder->slave_priv = priv; - encoder->slave_funcs = &ch7006_encoder_funcs; + encoder->encoder_i2c_priv = priv; + encoder->encoder_i2c_funcs = &ch7006_encoder_funcs; priv->norm = TV_NORM_PAL; priv->select_subconnector = DRM_MODE_SUBCONNECTOR_Automatic; @@ -495,7 +495,7 @@ static const struct dev_pm_ops ch7006_pm_ops = { .resume = ch7006_resume, }; -static struct drm_i2c_encoder_driver ch7006_driver = { +static struct nouveau_i2c_encoder_driver ch7006_driver = { .i2c_driver = { .probe = ch7006_probe, .remove = ch7006_remove, @@ -516,12 +516,12 @@ static struct drm_i2c_encoder_driver ch7006_driver = { static int __init ch7006_init(void) { - return drm_i2c_encoder_register(THIS_MODULE, &ch7006_driver); + return i2c_add_driver(&ch7006_driver.i2c_driver); } static void __exit ch7006_exit(void) { - drm_i2c_encoder_unregister(&ch7006_driver); + i2c_del_driver(&ch7006_driver.i2c_driver); } int ch7006_debug; diff --git a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c index 6afe6d0ee630..e58d94451959 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_mode.c @@ -198,7 +198,7 @@ const struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder, void ch7006_setup_levels(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); uint8_t *regs = priv->state.regs; const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm]; @@ -229,7 +229,7 @@ void ch7006_setup_levels(struct drm_encoder *encoder) void ch7006_setup_subcarrier(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_state *state = &priv->state; const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm]; @@ -253,7 +253,7 @@ void ch7006_setup_subcarrier(struct drm_encoder *encoder) void ch7006_setup_pll(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); uint8_t *regs = priv->state.regs; const struct ch7006_mode *mode = priv->mode; @@ -324,7 +324,7 @@ void ch7006_setup_power_state(struct drm_encoder *encoder) void ch7006_setup_properties(struct drm_encoder *encoder) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); struct ch7006_priv *priv = to_ch7006_priv(encoder); struct ch7006_state *state = &priv->state; const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm]; diff --git a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h index d7a91e17510f..5ad5157a2c02 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/ch7006_priv.h @@ -27,9 +27,9 @@ #ifndef __NOUVEAU_I2C_CH7006_PRIV_H__ #define __NOUVEAU_I2C_CH7006_PRIV_H__ -#include #include +#include #include typedef int64_t fixed; @@ -100,7 +100,7 @@ struct ch7006_priv { }; #define to_ch7006_priv(x) \ - ((struct ch7006_priv *)to_encoder_slave(x)->slave_priv) + ((struct ch7006_priv *)to_encoder_i2c(x)->encoder_i2c_priv) extern int ch7006_debug; extern char *ch7006_tv_norm; diff --git a/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c b/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c index 39b01c897518..54ea8459332d 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c +++ b/drivers/gpu/drm/nouveau/dispnv04/i2c/sil164_drv.c @@ -27,10 +27,10 @@ #include #include -#include #include #include +#include #include struct sil164_priv { @@ -42,7 +42,7 @@ struct sil164_priv { }; #define to_sil164_priv(x) \ - ((struct sil164_priv *)to_encoder_slave(x)->slave_priv) + ((struct sil164_priv *)to_encoder_i2c(x)->encoder_i2c_priv) #define sil164_dbg(client, format, ...) do { \ if (drm_debug_enabled(DRM_UT_KMS)) \ @@ -222,7 +222,7 @@ sil164_encoder_dpms(struct drm_encoder *encoder, int mode) bool on = (mode == DRM_MODE_DPMS_ON); bool duallink = (on && encoder->crtc->mode.clock > 165000); - sil164_set_power_state(drm_i2c_encoder_get_client(encoder), on); + sil164_set_power_state(nouveau_i2c_encoder_get_client(encoder), on); if (priv->duallink_slave) sil164_set_power_state(priv->duallink_slave, duallink); @@ -233,7 +233,7 @@ sil164_encoder_save(struct drm_encoder *encoder) { struct sil164_priv *priv = to_sil164_priv(encoder); - sil164_save_state(drm_i2c_encoder_get_client(encoder), + sil164_save_state(nouveau_i2c_encoder_get_client(encoder), priv->saved_state); if (priv->duallink_slave) @@ -246,7 +246,7 @@ sil164_encoder_restore(struct drm_encoder *encoder) { struct sil164_priv *priv = to_sil164_priv(encoder); - sil164_restore_state(drm_i2c_encoder_get_client(encoder), + sil164_restore_state(nouveau_i2c_encoder_get_client(encoder), priv->saved_state); if (priv->duallink_slave) @@ -278,7 +278,7 @@ sil164_encoder_mode_set(struct drm_encoder *encoder, struct sil164_priv *priv = to_sil164_priv(encoder); bool duallink = adjusted_mode->clock > 165000; - sil164_init_state(drm_i2c_encoder_get_client(encoder), + sil164_init_state(nouveau_i2c_encoder_get_client(encoder), &priv->config, duallink); if (priv->duallink_slave) @@ -292,7 +292,7 @@ static enum drm_connector_status sil164_encoder_detect(struct drm_encoder *encoder, struct drm_connector *connector) { - struct i2c_client *client = drm_i2c_encoder_get_client(encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(encoder); if (sil164_read(client, SIL164_DETECT) & SIL164_DETECT_HOTPLUG_STAT) return connector_status_connected; @@ -331,10 +331,10 @@ sil164_encoder_destroy(struct drm_encoder *encoder) i2c_unregister_device(priv->duallink_slave); kfree(priv); - drm_i2c_encoder_destroy(encoder); + nouveau_i2c_encoder_destroy(encoder); } -static const struct drm_encoder_slave_funcs sil164_encoder_funcs = { +static const struct nouveau_i2c_encoder_funcs sil164_encoder_funcs = { .set_config = sil164_encoder_set_config, .destroy = sil164_encoder_destroy, .dpms = sil164_encoder_dpms, @@ -394,7 +394,7 @@ sil164_detect_slave(struct i2c_client *client) static int sil164_encoder_init(struct i2c_client *client, struct drm_device *dev, - struct drm_encoder_slave *encoder) + struct nouveau_i2c_encoder *encoder) { struct sil164_priv *priv; struct i2c_client *slave_client; @@ -403,8 +403,8 @@ sil164_encoder_init(struct i2c_client *client, if (!priv) return -ENOMEM; - encoder->slave_priv = priv; - encoder->slave_funcs = &sil164_encoder_funcs; + encoder->encoder_i2c_priv = priv; + encoder->encoder_i2c_funcs = &sil164_encoder_funcs; slave_client = sil164_detect_slave(client); if (!IS_ERR(slave_client)) @@ -419,7 +419,7 @@ static const struct i2c_device_id sil164_ids[] = { }; MODULE_DEVICE_TABLE(i2c, sil164_ids); -static struct drm_i2c_encoder_driver sil164_driver = { +static struct nouveau_i2c_encoder_driver sil164_driver = { .i2c_driver = { .probe = sil164_probe, .driver = { @@ -435,13 +435,13 @@ static struct drm_i2c_encoder_driver sil164_driver = { static int __init sil164_init(void) { - return drm_i2c_encoder_register(THIS_MODULE, &sil164_driver); + return i2c_add_driver(&sil164_driver.i2c_driver); } static void __exit sil164_exit(void) { - drm_i2c_encoder_unregister(&sil164_driver); + i2c_del_driver(&sil164_driver.i2c_driver); } MODULE_AUTHOR("Francisco Jerez "); diff --git a/drivers/gpu/drm/drm_encoder_slave.c b/drivers/gpu/drm/nouveau/dispnv04/nouveau_i2c_encoder.c similarity index 54% rename from drivers/gpu/drm/drm_encoder_slave.c rename to drivers/gpu/drm/nouveau/dispnv04/nouveau_i2c_encoder.c index e464429d32df..e2bf99c43336 100644 --- a/drivers/gpu/drm/drm_encoder_slave.c +++ b/drivers/gpu/drm/nouveau/dispnv04/nouveau_i2c_encoder.c @@ -26,10 +26,10 @@ #include -#include +#include /** - * drm_i2c_encoder_init - Initialize an I2C slave encoder + * nouveau_i2c_encoder_init - Initialize an I2C slave encoder * @dev: DRM device. * @encoder: Encoder to be attached to the I2C device. You aren't * required to have called drm_encoder_init() before. @@ -40,7 +40,7 @@ * * Create an I2C device on the specified bus (the module containing its * driver is transparently loaded) and attach it to the specified - * &drm_encoder_slave. The @slave_funcs field will be initialized with + * &nouveau_i2c_encoder. The @encoder_i2c_funcs field will be initialized with * the hooks provided by the slave driver. * * If @info.platform_data is non-NULL it will be used as the initial @@ -49,14 +49,14 @@ * Returns 0 on success or a negative errno on failure, in particular, * -ENODEV is returned when no matching driver is found. */ -int drm_i2c_encoder_init(struct drm_device *dev, - struct drm_encoder_slave *encoder, - struct i2c_adapter *adap, - const struct i2c_board_info *info) +int nouveau_i2c_encoder_init(struct drm_device *dev, + struct nouveau_i2c_encoder *encoder, + struct i2c_adapter *adap, + const struct i2c_board_info *info) { struct module *module = NULL; struct i2c_client *client; - struct drm_i2c_encoder_driver *encoder_drv; + struct nouveau_i2c_encoder_driver *encoder_drv; int err = 0; request_module("%s%s", I2C_MODULE_PREFIX, info->type); @@ -73,16 +73,16 @@ int drm_i2c_encoder_init(struct drm_device *dev, goto fail_unregister; } - encoder->bus_priv = client; + encoder->i2c_client = client; - encoder_drv = to_drm_i2c_encoder_driver(to_i2c_driver(client->dev.driver)); + encoder_drv = to_nouveau_i2c_encoder_driver(to_i2c_driver(client->dev.driver)); err = encoder_drv->encoder_init(client, dev, encoder); if (err) goto fail_module_put; if (info->platform_data) - encoder->slave_funcs->set_config(&encoder->base, + encoder->encoder_i2c_funcs->set_config(&encoder->base, info->platform_data); return 0; @@ -93,90 +93,53 @@ int drm_i2c_encoder_init(struct drm_device *dev, i2c_unregister_device(client); return err; } -EXPORT_SYMBOL(drm_i2c_encoder_init); /** - * drm_i2c_encoder_destroy - Unregister the I2C device backing an encoder + * nouveau_i2c_encoder_destroy - Unregister the I2C device backing an encoder * @drm_encoder: Encoder to be unregistered. * * This should be called from the @destroy method of an I2C slave * encoder driver once I2C access is no longer needed. */ -void drm_i2c_encoder_destroy(struct drm_encoder *drm_encoder) +void nouveau_i2c_encoder_destroy(struct drm_encoder *drm_encoder) { - struct drm_encoder_slave *encoder = to_encoder_slave(drm_encoder); - struct i2c_client *client = drm_i2c_encoder_get_client(drm_encoder); + struct nouveau_i2c_encoder *encoder = to_encoder_i2c(drm_encoder); + struct i2c_client *client = nouveau_i2c_encoder_get_client(drm_encoder); struct module *module = client->dev.driver->owner; i2c_unregister_device(client); - encoder->bus_priv = NULL; + encoder->i2c_client = NULL; module_put(module); } -EXPORT_SYMBOL(drm_i2c_encoder_destroy); +EXPORT_SYMBOL(nouveau_i2c_encoder_destroy); /* * Wrapper fxns which can be plugged in to drm_encoder_helper_funcs: */ -static inline const struct drm_encoder_slave_funcs * -get_slave_funcs(struct drm_encoder *enc) +bool nouveau_i2c_encoder_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) { - return to_encoder_slave(enc)->slave_funcs; -} - -void drm_i2c_encoder_dpms(struct drm_encoder *encoder, int mode) -{ - get_slave_funcs(encoder)->dpms(encoder, mode); -} -EXPORT_SYMBOL(drm_i2c_encoder_dpms); - -bool drm_i2c_encoder_mode_fixup(struct drm_encoder *encoder, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - if (!get_slave_funcs(encoder)->mode_fixup) + if (!get_encoder_i2c_funcs(encoder)->mode_fixup) return true; - return get_slave_funcs(encoder)->mode_fixup(encoder, mode, adjusted_mode); + return get_encoder_i2c_funcs(encoder)->mode_fixup(encoder, mode, adjusted_mode); } -EXPORT_SYMBOL(drm_i2c_encoder_mode_fixup); -void drm_i2c_encoder_prepare(struct drm_encoder *encoder) +enum drm_connector_status nouveau_i2c_encoder_detect(struct drm_encoder *encoder, + struct drm_connector *connector) { - drm_i2c_encoder_dpms(encoder, DRM_MODE_DPMS_OFF); + return get_encoder_i2c_funcs(encoder)->detect(encoder, connector); } -EXPORT_SYMBOL(drm_i2c_encoder_prepare); -void drm_i2c_encoder_commit(struct drm_encoder *encoder) +void nouveau_i2c_encoder_save(struct drm_encoder *encoder) { - drm_i2c_encoder_dpms(encoder, DRM_MODE_DPMS_ON); + get_encoder_i2c_funcs(encoder)->save(encoder); } -EXPORT_SYMBOL(drm_i2c_encoder_commit); -void drm_i2c_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) +void nouveau_i2c_encoder_restore(struct drm_encoder *encoder) { - get_slave_funcs(encoder)->mode_set(encoder, mode, adjusted_mode); + get_encoder_i2c_funcs(encoder)->restore(encoder); } -EXPORT_SYMBOL(drm_i2c_encoder_mode_set); - -enum drm_connector_status drm_i2c_encoder_detect(struct drm_encoder *encoder, - struct drm_connector *connector) -{ - return get_slave_funcs(encoder)->detect(encoder, connector); -} -EXPORT_SYMBOL(drm_i2c_encoder_detect); - -void drm_i2c_encoder_save(struct drm_encoder *encoder) -{ - get_slave_funcs(encoder)->save(encoder); -} -EXPORT_SYMBOL(drm_i2c_encoder_save); - -void drm_i2c_encoder_restore(struct drm_encoder *encoder) -{ - get_slave_funcs(encoder)->restore(encoder); -} -EXPORT_SYMBOL(drm_i2c_encoder_restore); diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c index 4ffd8af6bf8b..c61ab083f62e 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c @@ -99,7 +99,7 @@ static void nv04_tv_dpms(struct drm_encoder *encoder, int mode) NVWriteRAMDAC(dev, 0, NV_PRAMDAC_PLL_COEFF_SELECT, state->pllsel); - get_slave_funcs(encoder)->dpms(encoder, mode); + get_encoder_i2c_funcs(encoder)->dpms(encoder, mode); } static void nv04_tv_bind(struct drm_device *dev, int head, bool bind) @@ -158,7 +158,7 @@ static void nv04_tv_mode_set(struct drm_encoder *encoder, regp->tv_vskew = 1; regp->tv_vsync_delay = 1; - get_slave_funcs(encoder)->mode_set(encoder, mode, adjusted_mode); + get_encoder_i2c_funcs(encoder)->mode_set(encoder, mode, adjusted_mode); } static void nv04_tv_commit(struct drm_encoder *encoder) @@ -178,7 +178,7 @@ static void nv04_tv_commit(struct drm_encoder *encoder) static void nv04_tv_destroy(struct drm_encoder *encoder) { - get_slave_funcs(encoder)->destroy(encoder); + get_encoder_i2c_funcs(encoder)->destroy(encoder); drm_encoder_cleanup(encoder); kfree(encoder->helper_private); @@ -191,11 +191,11 @@ static const struct drm_encoder_funcs nv04_tv_funcs = { static const struct drm_encoder_helper_funcs nv04_tv_helper_funcs = { .dpms = nv04_tv_dpms, - .mode_fixup = drm_i2c_encoder_mode_fixup, + .mode_fixup = nouveau_i2c_encoder_mode_fixup, .prepare = nv04_tv_prepare, .commit = nv04_tv_commit, .mode_set = nv04_tv_mode_set, - .detect = drm_i2c_encoder_detect, + .detect = nouveau_i2c_encoder_detect, }; int @@ -226,8 +226,8 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry) NULL); drm_encoder_helper_add(encoder, &nv04_tv_helper_funcs); - nv_encoder->enc_save = drm_i2c_encoder_save; - nv_encoder->enc_restore = drm_i2c_encoder_restore; + nv_encoder->enc_save = nouveau_i2c_encoder_save; + nv_encoder->enc_restore = nouveau_i2c_encoder_restore; encoder->possible_crtcs = entry->heads; encoder->possible_clones = 0; @@ -235,14 +235,14 @@ nv04_tv_create(struct drm_connector *connector, struct dcb_output *entry) nv_encoder->or = ffs(entry->or) - 1; /* Run the slave-specific initialization */ - ret = drm_i2c_encoder_init(dev, to_encoder_slave(encoder), - &bus->i2c, - &nv04_tv_encoder_info[type].dev); + ret = nouveau_i2c_encoder_init(dev, to_encoder_i2c(encoder), + &bus->i2c, + &nv04_tv_encoder_info[type].dev); if (ret < 0) goto fail_cleanup; /* Attach it to the specified connector. */ - get_slave_funcs(encoder)->create_resources(encoder, connector); + get_encoder_i2c_funcs(encoder)->create_resources(encoder, connector); drm_connector_attach_encoder(connector, encoder); return 0; diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c index 35ad4e10d273..06de05fe5db6 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c +++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c @@ -779,7 +779,7 @@ static const struct drm_encoder_helper_funcs nv17_tv_helper_funcs = { .detect = nv17_tv_detect, }; -static const struct drm_encoder_slave_funcs nv17_tv_slave_funcs = { +static const struct nouveau_i2c_encoder_funcs nv17_tv_encoder_i2c_funcs = { .get_modes = nv17_tv_get_modes, .mode_valid = nv17_tv_mode_valid, .create_resources = nv17_tv_create_resources, @@ -818,7 +818,7 @@ nv17_tv_create(struct drm_connector *connector, struct dcb_output *entry) drm_encoder_init(dev, encoder, &nv17_tv_funcs, DRM_MODE_ENCODER_TVDAC, NULL); drm_encoder_helper_add(encoder, &nv17_tv_helper_funcs); - to_encoder_slave(encoder)->slave_funcs = &nv17_tv_slave_funcs; + to_encoder_i2c(encoder)->encoder_i2c_funcs = &nv17_tv_encoder_i2c_funcs; tv_enc->base.enc_save = nv17_tv_save; tv_enc->base.enc_restore = nv17_tv_restore; diff --git a/include/drm/drm_encoder_slave.h b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/encoder_i2c.h similarity index 59% rename from include/drm/drm_encoder_slave.h rename to drivers/gpu/drm/nouveau/include/dispnv04/i2c/encoder_i2c.h index b526643833dc..31334aa90781 100644 --- a/include/drm/drm_encoder_slave.h +++ b/drivers/gpu/drm/nouveau/include/dispnv04/i2c/encoder_i2c.h @@ -24,8 +24,8 @@ * */ -#ifndef __DRM_ENCODER_SLAVE_H__ -#define __DRM_ENCODER_SLAVE_H__ +#ifndef __NOUVEAU_ENCODER_I2C_H__ +#define __NOUVEAU_ENCODER_I2C_H__ #include @@ -33,7 +33,7 @@ #include /** - * struct drm_encoder_slave_funcs - Entry points exposed by a slave encoder driver + * struct nouveau_i2c_encoder_funcs - Entry points exposed by a I2C encoder driver * * Most of its members are analogous to the function pointers in * &drm_encoder_helper_funcs and they can optionally be used to @@ -41,7 +41,7 @@ * @set_property) will typically be wrapped around and only be called * if the encoder is the currently selected one for the connector. */ -struct drm_encoder_slave_funcs { +struct nouveau_i2c_encoder_funcs { /** * @set_config: Initialize any encoder-specific modesetting parameters. * The meaning of the @params parameter is implementation dependent. It @@ -58,24 +58,23 @@ struct drm_encoder_slave_funcs { void (*destroy)(struct drm_encoder *encoder); /** - * @dpms: Analogous to &drm_encoder_helper_funcs @dpms callback. Wrapped - * by drm_i2c_encoder_dpms(). + * @dpms: Analogous to &drm_encoder_helper_funcs @dpms callback. */ void (*dpms)(struct drm_encoder *encoder, int mode); /** - * @save: Save state. Wrapped by drm_i2c_encoder_save(). + * @save: Save state. Wrapped by nouveau_i2c_encoder_save(). */ void (*save)(struct drm_encoder *encoder); /** - * @restore: Restore state. Wrapped by drm_i2c_encoder_restore(). + * @restore: Restore state. Wrapped by nouveau_i2c_encoder_restore(). */ void (*restore)(struct drm_encoder *encoder); /** * @mode_fixup: Analogous to &drm_encoder_helper_funcs @mode_fixup - * callback. Wrapped by drm_i2c_encoder_mode_fixup(). + * callback. Wrapped by nouveau_i2c_encoder_mode_fixup(). */ bool (*mode_fixup)(struct drm_encoder *encoder, const struct drm_display_mode *mode, @@ -88,7 +87,7 @@ struct drm_encoder_slave_funcs { const struct drm_display_mode *mode); /** * @mode_set: Analogous to &drm_encoder_helper_funcs @mode_set - * callback. Wrapped by drm_i2c_encoder_mode_set(). + * callback. */ void (*mode_set)(struct drm_encoder *encoder, struct drm_display_mode *mode, @@ -96,7 +95,7 @@ struct drm_encoder_slave_funcs { /** * @detect: Analogous to &drm_encoder_helper_funcs @detect - * callback. Wrapped by drm_i2c_encoder_detect(). + * callback. Wrapped by nouveau_i2c_encoder_detect(). */ enum drm_connector_status (*detect)(struct drm_encoder *encoder, struct drm_connector *connector); @@ -120,54 +119,60 @@ struct drm_encoder_slave_funcs { }; /** - * struct drm_encoder_slave - Slave encoder struct + * struct nouveau_i2c_encoder - I2C encoder struct * - * A &drm_encoder_slave has two sets of callbacks, @slave_funcs and the + * A &nouveau_i2c_encoder has two sets of callbacks, @encoder_i2c_funcs and the * ones in @base. The former are never actually called by the common * CRTC code, it's just a convenience for splitting the encoder * functions in an upper, GPU-specific layer and a (hopefully) * GPU-agnostic lower layer: It's the GPU driver responsibility to - * call the slave methods when appropriate. + * call the nouveau_i2c_encoder methods when appropriate. * - * drm_i2c_encoder_init() provides a way to get an implementation of + * nouveau_i2c_encoder_init() provides a way to get an implementation of * this. */ -struct drm_encoder_slave { +struct nouveau_i2c_encoder { /** * @base: DRM encoder object. */ struct drm_encoder base; /** - * @slave_funcs: Slave encoder callbacks. + * @encoder_i2c_funcs: I2C encoder callbacks. */ - const struct drm_encoder_slave_funcs *slave_funcs; + const struct nouveau_i2c_encoder_funcs *encoder_i2c_funcs; /** - * @slave_priv: Slave encoder private data. + * @encoder_i2c_priv: I2C encoder private data. */ - void *slave_priv; + void *encoder_i2c_priv; /** - * @bus_priv: Bus specific data. + * @i2c_client: corresponding I2C client structure */ - void *bus_priv; + struct i2c_client *i2c_client; }; -#define to_encoder_slave(x) container_of((x), struct drm_encoder_slave, base) -int drm_i2c_encoder_init(struct drm_device *dev, - struct drm_encoder_slave *encoder, - struct i2c_adapter *adap, - const struct i2c_board_info *info); +#define to_encoder_i2c(x) container_of((x), struct nouveau_i2c_encoder, base) +int nouveau_i2c_encoder_init(struct drm_device *dev, + struct nouveau_i2c_encoder *encoder, + struct i2c_adapter *adap, + const struct i2c_board_info *info); + +static inline const struct nouveau_i2c_encoder_funcs * +get_encoder_i2c_funcs(struct drm_encoder *enc) +{ + return to_encoder_i2c(enc)->encoder_i2c_funcs; +} /** - * struct drm_i2c_encoder_driver + * struct nouveau_i2c_encoder_driver * * Describes a device driver for an encoder connected to the GPU through an I2C * bus. */ -struct drm_i2c_encoder_driver { +struct nouveau_i2c_encoder_driver { /** * @i2c_driver: I2C device driver description. */ @@ -175,67 +180,41 @@ struct drm_i2c_encoder_driver { /** * @encoder_init: Callback to allocate any per-encoder data structures - * and to initialize the @slave_funcs and (optionally) @slave_priv + * and to initialize the @encoder_i2c_funcs and (optionally) @encoder_i2c_priv * members of @encoder. */ int (*encoder_init)(struct i2c_client *client, struct drm_device *dev, - struct drm_encoder_slave *encoder); + struct nouveau_i2c_encoder *encoder); }; -#define to_drm_i2c_encoder_driver(x) container_of((x), \ - struct drm_i2c_encoder_driver, \ + +#define to_nouveau_i2c_encoder_driver(x) container_of((x), \ + struct nouveau_i2c_encoder_driver, \ i2c_driver) /** - * drm_i2c_encoder_get_client - Get the I2C client corresponding to an encoder + * nouveau_i2c_encoder_get_client - Get the I2C client corresponding to an encoder * @encoder: The encoder */ -static inline struct i2c_client *drm_i2c_encoder_get_client(struct drm_encoder *encoder) +static inline struct i2c_client *nouveau_i2c_encoder_get_client(struct drm_encoder *encoder) { - return (struct i2c_client *)to_encoder_slave(encoder)->bus_priv; + return to_encoder_i2c(encoder)->i2c_client; } -/** - * drm_i2c_encoder_register - Register an I2C encoder driver - * @owner: Module containing the driver. - * @driver: Driver to be registered. - */ -static inline int drm_i2c_encoder_register(struct module *owner, - struct drm_i2c_encoder_driver *driver) -{ - return i2c_register_driver(owner, &driver->i2c_driver); -} - -/** - * drm_i2c_encoder_unregister - Unregister an I2C encoder driver - * @driver: Driver to be unregistered. - */ -static inline void drm_i2c_encoder_unregister(struct drm_i2c_encoder_driver *driver) -{ - i2c_del_driver(&driver->i2c_driver); -} - -void drm_i2c_encoder_destroy(struct drm_encoder *encoder); - +void nouveau_i2c_encoder_destroy(struct drm_encoder *encoder); /* * Wrapper fxns which can be plugged in to drm_encoder_helper_funcs: */ -void drm_i2c_encoder_dpms(struct drm_encoder *encoder, int mode); -bool drm_i2c_encoder_mode_fixup(struct drm_encoder *encoder, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode); -void drm_i2c_encoder_prepare(struct drm_encoder *encoder); -void drm_i2c_encoder_commit(struct drm_encoder *encoder); -void drm_i2c_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode); -enum drm_connector_status drm_i2c_encoder_detect(struct drm_encoder *encoder, - struct drm_connector *connector); -void drm_i2c_encoder_save(struct drm_encoder *encoder); -void drm_i2c_encoder_restore(struct drm_encoder *encoder); +bool nouveau_i2c_encoder_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode); +enum drm_connector_status nouveau_i2c_encoder_detect(struct drm_encoder *encoder, + struct drm_connector *connector); +void nouveau_i2c_encoder_save(struct drm_encoder *encoder); +void nouveau_i2c_encoder_restore(struct drm_encoder *encoder); #endif diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 2d26784bde74..6fb9719d721f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -797,8 +797,10 @@ nouveau_connector_set_property(struct drm_connector *connector, property, value); if (ret) { if (nv_encoder && nv_encoder->dcb->type == DCB_OUTPUT_TV) - return get_slave_funcs(encoder)->set_property( - encoder, connector, property, value); + return get_encoder_i2c_funcs(encoder)->set_property(encoder, + connector, + property, + value); return ret; } @@ -1015,7 +1017,7 @@ nouveau_connector_get_modes(struct drm_connector *connector) nouveau_connector_detect_depth(connector); if (nv_encoder->dcb->type == DCB_OUTPUT_TV) - ret = get_slave_funcs(encoder)->get_modes(encoder, connector); + ret = get_encoder_i2c_funcs(encoder)->get_modes(encoder, connector); if (nv_connector->type == DCB_CONNECTOR_LVDS || nv_connector->type == DCB_CONNECTOR_LVDS_SPWG || @@ -1100,7 +1102,7 @@ nouveau_connector_mode_valid(struct drm_connector *connector, max_clock = 350000; break; case DCB_OUTPUT_TV: - return get_slave_funcs(encoder)->mode_valid(encoder, mode); + return get_encoder_i2c_funcs(encoder)->mode_valid(encoder, mode); case DCB_OUTPUT_DP: return nv50_dp_mode_valid(nv_encoder, mode, NULL); default: diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h index 333042fc493f..dce8e5d9d496 100644 --- a/drivers/gpu/drm/nouveau/nouveau_encoder.h +++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h @@ -31,7 +31,8 @@ #include #include -#include + +#include #include "dispnv04/disp.h" @@ -43,7 +44,7 @@ struct nouveau_connector; struct nvkm_i2c_port; struct nouveau_encoder { - struct drm_encoder_slave base; + struct nouveau_i2c_encoder base; struct dcb_output *dcb; struct nvif_outp outp; @@ -137,7 +138,7 @@ find_encoder(struct drm_connector *connector, int type); static inline struct nouveau_encoder *nouveau_encoder(struct drm_encoder *enc) { - struct drm_encoder_slave *slave = to_encoder_slave(enc); + struct nouveau_i2c_encoder *slave = to_encoder_i2c(enc); return container_of(slave, struct nouveau_encoder, base); } @@ -147,12 +148,6 @@ static inline struct drm_encoder *to_drm_encoder(struct nouveau_encoder *enc) return &enc->base.base; } -static inline const struct drm_encoder_slave_funcs * -get_slave_funcs(struct drm_encoder *enc) -{ - return to_encoder_slave(enc)->slave_funcs; -} - /* nouveau_dp.c */ enum nouveau_dp_status { NOUVEAU_DP_NONE, From 574f5ee2c85a00a579549d50e9fc9c6c072ee4c4 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 10 Dec 2024 15:18:46 +0100 Subject: [PATCH 0013/1443] drm/bridge: ti-sn65dsi86: Fix multiple instances Each bridge instance creates up to four auxiliary devices with different names. However, their IDs are always zero, causing duplicate filename errors when a system has multiple bridges: sysfs: cannot create duplicate filename '/bus/auxiliary/devices/ti_sn65dsi86.gpio.0' Fix this by using a unique instance ID per bridge instance. The instance ID is derived from the I2C adapter number and the bridge's I2C address, to support multiple instances on the same bus. Fixes: bf73537f411b ("drm/bridge: ti-sn65dsi86: Break GPIO and MIPI-to-eDP bridge into sub-drivers") Signed-off-by: Geert Uytterhoeven Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/7a68a0e3f927e26edca6040067fb653eb06efb79.1733840089.git.geert+renesas@glider.be --- drivers/gpu/drm/bridge/ti-sn65dsi86.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index e4d9006b59f1..b3d617505dda 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -480,6 +480,7 @@ static int ti_sn65dsi86_add_aux_device(struct ti_sn65dsi86 *pdata, const char *name) { struct device *dev = pdata->dev; + const struct i2c_client *client = to_i2c_client(dev); struct auxiliary_device *aux; int ret; @@ -488,6 +489,7 @@ static int ti_sn65dsi86_add_aux_device(struct ti_sn65dsi86 *pdata, return -ENOMEM; aux->name = name; + aux->id = (client->adapter->nr << 10) | client->addr; aux->dev.parent = dev; aux->dev.release = ti_sn65dsi86_aux_device_release; device_set_of_node_from_dev(&aux->dev, dev); From 7e182cb4f5567f53417b762ec0d679f0b6f0039d Mon Sep 17 00:00:00 2001 From: Abhinav Kumar Date: Wed, 11 Dec 2024 13:18:42 -0800 Subject: [PATCH 0014/1443] drm: allow encoder mode_set even when connectors change for crtc In certain use-cases, a CRTC could switch between two encoders and because the mode being programmed on the CRTC remains the same during this switch, the CRTC's mode_changed remains false. In such cases, the encoder's mode_set also gets skipped. Skipping mode_set on the encoder for such cases could cause an issue because even though the same CRTC mode was being used, the encoder type could have changed like the CRTC could have switched from a real time encoder to a writeback encoder OR vice-versa. Allow encoder's mode_set to happen even when connectors changed on a CRTC and not just when the mode changed. Signed-off-by: Abhinav Kumar Signed-off-by: Jessica Zhang Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241211-abhinavk-modeset-fix-v3-1-0de4bf3e7c32@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/drm_atomic_helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 5186d2114a50..32902f77f00d 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -1376,7 +1376,7 @@ crtc_set_mode(struct drm_device *dev, struct drm_atomic_state *old_state) mode = &new_crtc_state->mode; adjusted_mode = &new_crtc_state->adjusted_mode; - if (!new_crtc_state->mode_changed) + if (!new_crtc_state->mode_changed && !new_crtc_state->connectors_changed) continue; drm_dbg_atomic(dev, "modeset on [ENCODER:%d:%s]\n", From 73d934d7b6e39a3e52586467a30ca3ff3f6f9eb4 Mon Sep 17 00:00:00 2001 From: Jessica Zhang Date: Wed, 11 Dec 2024 13:18:43 -0800 Subject: [PATCH 0015/1443] drm/tests: Add test for drm_atomic_helper_commit_modeset_disables() Add a subtest to check that modeset is called when the connector is changed Signed-off-by: Jessica Zhang Reviewed-by: Dmitry Baryshkov Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241211-abhinavk-modeset-fix-v3-2-0de4bf3e7c32@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/tests/Makefile | 1 + drivers/gpu/drm/tests/drm_atomic_state_test.c | 244 ++++++++++++++++++ 2 files changed, 245 insertions(+) create mode 100644 drivers/gpu/drm/tests/drm_atomic_state_test.c diff --git a/drivers/gpu/drm/tests/Makefile b/drivers/gpu/drm/tests/Makefile index 56dab563abd7..0109bcf7faa5 100644 --- a/drivers/gpu/drm/tests/Makefile +++ b/drivers/gpu/drm/tests/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_DRM_KUNIT_TEST_HELPERS) += \ drm_kunit_helpers.o obj-$(CONFIG_DRM_KUNIT_TEST) += \ + drm_atomic_state_test.o \ drm_buddy_test.o \ drm_cmdline_parser_test.o \ drm_connector_test.o \ diff --git a/drivers/gpu/drm/tests/drm_atomic_state_test.c b/drivers/gpu/drm/tests/drm_atomic_state_test.c new file mode 100644 index 000000000000..be1f78024945 --- /dev/null +++ b/drivers/gpu/drm/tests/drm_atomic_state_test.c @@ -0,0 +1,244 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Test cases for the drm_atomic_state helpers + * + * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#include +#include +#include +#include +#include + +#define DRM_TEST_ENC_0 BIT(0) +#define DRM_TEST_ENC_1 BIT(1) +#define DRM_TEST_ENC_2 BIT(2) + +#define DRM_TEST_CONN_0 BIT(0) + +static const struct drm_display_mode drm_atomic_test_mode = { + DRM_MODE("1024x768", 0, 65000, 1024, 1048, + 1184, 1344, 0, 768, 771, 777, 806, 0, + DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) +}; + +struct drm_atomic_test_priv { + struct drm_device drm; + struct drm_plane *plane; + struct drm_crtc *crtc; + struct drm_encoder encoders[3]; + struct drm_connector connectors[2]; +}; + +static int modeset_counter; + +static void drm_test_encoder_mode_set(struct drm_encoder *encoder, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + modeset_counter++; +} + +static const struct drm_encoder_helper_funcs drm_atomic_test_encoder_funcs = { + .atomic_mode_set = drm_test_encoder_mode_set, +}; + +static const struct drm_connector_funcs dummy_connector_funcs = { + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, + .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, + .reset = drm_atomic_helper_connector_reset, +}; + +static int drm_atomic_test_dummy_get_modes(struct drm_connector *connector) +{ + return drm_connector_helper_get_modes_fixed(connector, + &drm_atomic_test_mode); +} + +static const struct drm_connector_helper_funcs dummy_connector_helper_funcs = { + .get_modes = drm_atomic_test_dummy_get_modes, +}; + +static struct drm_atomic_test_priv * +drm_atomic_test_init_drm_components(struct kunit *test, bool has_connectors) +{ + struct drm_atomic_test_priv *priv; + struct drm_encoder *enc; + struct drm_connector *conn; + struct drm_device *drm; + struct device *dev; + int ret; + + dev = drm_kunit_helper_alloc_device(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, dev); + + priv = drm_kunit_helper_alloc_drm_device(test, dev, + struct drm_atomic_test_priv, + drm, + DRIVER_MODESET | DRIVER_ATOMIC); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv); + test->priv = priv; + + drm = &priv->drm; + priv->plane = drm_kunit_helper_create_primary_plane(test, drm, + NULL, + NULL, + NULL, 0, + NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->plane); + + priv->crtc = drm_kunit_helper_create_crtc(test, drm, + priv->plane, NULL, + NULL, + NULL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, priv->crtc); + + for (int i = 0; i < ARRAY_SIZE(priv->encoders); i++) { + enc = &priv->encoders[i]; + + ret = drmm_encoder_init(drm, enc, NULL, + DRM_MODE_ENCODER_DSI, NULL); + KUNIT_ASSERT_EQ(test, ret, 0); + + enc->possible_crtcs = drm_crtc_mask(priv->crtc); + } + + priv->encoders[0].possible_clones = DRM_TEST_ENC_0 | DRM_TEST_ENC_1; + priv->encoders[1].possible_clones = DRM_TEST_ENC_0 | DRM_TEST_ENC_1; + priv->encoders[2].possible_clones = DRM_TEST_ENC_2; + + if (!has_connectors) + goto done; + + BUILD_BUG_ON(ARRAY_SIZE(priv->connectors) > ARRAY_SIZE(priv->encoders)); + + for (int i = 0; i < ARRAY_SIZE(priv->connectors); i++) { + conn = &priv->connectors[i]; + + ret = drmm_connector_init(drm, conn, &dummy_connector_funcs, + DRM_MODE_CONNECTOR_DSI, NULL); + KUNIT_ASSERT_EQ(test, ret, 0); + + drm_connector_helper_add(conn, &dummy_connector_helper_funcs); + drm_encoder_helper_add(&priv->encoders[i], + &drm_atomic_test_encoder_funcs); + + drm_connector_attach_encoder(conn, &priv->encoders[i]); + } + +done: + drm_mode_config_reset(drm); + + return priv; +} + +static int set_up_atomic_state(struct kunit *test, + struct drm_atomic_test_priv *priv, + struct drm_connector *connector, + struct drm_modeset_acquire_ctx *ctx) +{ + struct drm_device *drm = &priv->drm; + struct drm_crtc *crtc = priv->crtc; + struct drm_atomic_state *state; + struct drm_connector_state *conn_state; + struct drm_crtc_state *crtc_state; + int ret; + + state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state); + + if (connector) { + conn_state = drm_atomic_get_connector_state(state, connector); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state); + + ret = drm_atomic_set_crtc_for_connector(conn_state, crtc); + KUNIT_EXPECT_EQ(test, ret, 0); + } + + crtc_state = drm_atomic_get_crtc_state(state, crtc); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_state); + + ret = drm_atomic_set_mode_for_crtc(crtc_state, &drm_atomic_test_mode); + KUNIT_EXPECT_EQ(test, ret, 0); + + crtc_state->enable = true; + crtc_state->active = true; + + if (connector) { + ret = drm_atomic_commit(state); + KUNIT_ASSERT_EQ(test, ret, 0); + } else { + // dummy connector mask + crtc_state->connector_mask = DRM_TEST_CONN_0; + } + + return 0; +} + +/* + * Test that the DRM encoder mode_set() is called when the atomic state + * connectors are changed but the CRTC mode is not. + */ +static void drm_test_check_connector_changed_modeset(struct kunit *test) +{ + struct drm_atomic_test_priv *priv; + struct drm_modeset_acquire_ctx *ctx; + struct drm_connector *old_conn, *new_conn; + struct drm_atomic_state *state; + struct drm_device *drm; + struct drm_connector_state *new_conn_state, *old_conn_state; + int ret, initial_modeset_count; + + priv = drm_atomic_test_init_drm_components(test, true); + KUNIT_ASSERT_NOT_NULL(test, priv); + + drm = &priv->drm; + old_conn = &priv->connectors[0]; + new_conn = &priv->connectors[1]; + + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + + // first modeset to enable + ret = set_up_atomic_state(test, priv, old_conn, ctx); + KUNIT_ASSERT_EQ(test, ret, 0); + + state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state); + + new_conn_state = drm_atomic_get_connector_state(state, new_conn); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, new_conn_state); + + old_conn_state = drm_atomic_get_connector_state(state, old_conn); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, old_conn_state); + + ret = drm_atomic_set_crtc_for_connector(old_conn_state, NULL); + KUNIT_EXPECT_EQ(test, ret, 0); + + ret = drm_atomic_set_crtc_for_connector(new_conn_state, priv->crtc); + KUNIT_EXPECT_EQ(test, ret, 0); + + initial_modeset_count = modeset_counter; + + // modeset_disables is called as part of the atomic commit tail + ret = drm_atomic_commit(state); + KUNIT_ASSERT_EQ(test, ret, 0); + KUNIT_ASSERT_EQ(test, modeset_counter, initial_modeset_count + 1); +} + +static struct kunit_case drm_test_check_modeset_test[] = { + KUNIT_CASE(drm_test_check_connector_changed_modeset), + {} +}; + +static struct kunit_suite drm_test_check_modeset_test_suite = { + .name = "drm_validate_modeset", + .test_cases = drm_test_check_modeset_test, +}; + +kunit_test_suite(drm_test_check_modeset_test_suite); + +MODULE_AUTHOR("Jessica Zhang Date: Mon, 16 Dec 2024 16:43:12 -0800 Subject: [PATCH 0016/1443] drm: add clone mode check for CRTC Add a common helper to check if the given CRTC state is in clone mode. This can be used by drivers to help detect if a CRTC is being shared by multiple encoders Signed-off-by: Jessica Zhang Reviewed-by: Abhinav Kumar Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241216-concurrent-wb-v4-1-fe220297a7f0@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/drm_crtc.c | 20 ++++++++++++++++++++ include/drm/drm_crtc.h | 2 +- 2 files changed, 21 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 3488ff067c69..46655339003d 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -939,3 +939,23 @@ int drm_crtc_create_scaling_filter_property(struct drm_crtc *crtc, return 0; } EXPORT_SYMBOL(drm_crtc_create_scaling_filter_property); + +/** + * drm_crtc_in_clone_mode - check if the given CRTC state is in clone mode + * + * @crtc_state: CRTC state to check + * + * This function determines if the given CRTC state is being cloned by multiple + * encoders. + * + * RETURNS: + * True if the CRTC state is in clone mode. False otherwise + */ +bool drm_crtc_in_clone_mode(struct drm_crtc_state *crtc_state) +{ + if (!crtc_state) + return false; + + return hweight32(crtc_state->encoder_mask) > 1; +} +EXPORT_SYMBOL(drm_crtc_in_clone_mode); diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 8b48a1974da3..caa56e039da2 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -1323,5 +1323,5 @@ static inline struct drm_crtc *drm_crtc_find(struct drm_device *dev, int drm_crtc_create_scaling_filter_property(struct drm_crtc *crtc, unsigned int supported_filters); - +bool drm_crtc_in_clone_mode(struct drm_crtc_state *crtc_state); #endif /* __DRM_CRTC_H__ */ From 5a6e8c369486a79493ab300a1987cc6aad16cf6a Mon Sep 17 00:00:00 2001 From: Jessica Zhang Date: Mon, 16 Dec 2024 16:43:13 -0800 Subject: [PATCH 0017/1443] drm/tests: Add test for drm_crtc_in_clone_mode() Add kunit test to validate drm_crtc_in_clone_mode() helper Signed-off-by: Jessica Zhang Reviewed-by: Abhinav Kumar Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241216-concurrent-wb-v4-2-fe220297a7f0@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/tests/drm_atomic_state_test.c | 62 ++++++++++++++++++- 1 file changed, 61 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/tests/drm_atomic_state_test.c b/drivers/gpu/drm/tests/drm_atomic_state_test.c index be1f78024945..79bc5a9aedbf 100644 --- a/drivers/gpu/drm/tests/drm_atomic_state_test.c +++ b/drivers/gpu/drm/tests/drm_atomic_state_test.c @@ -17,6 +17,12 @@ #define DRM_TEST_CONN_0 BIT(0) +struct drm_clone_mode_test { + const char *name; + u32 encoder_mask; + int expected_result; +}; + static const struct drm_display_mode drm_atomic_test_mode = { DRM_MODE("1024x768", 0, 65000, 1024, 1048, 1184, 1344, 0, 768, 771, 777, 806, 0, @@ -227,17 +233,71 @@ static void drm_test_check_connector_changed_modeset(struct kunit *test) KUNIT_ASSERT_EQ(test, modeset_counter, initial_modeset_count + 1); } +/* + * Test that the drm_crtc_in_clone_mode() helper can detect if a given CRTC + * state is in clone mode + */ +static void drm_test_check_in_clone_mode(struct kunit *test) +{ + bool ret; + const struct drm_clone_mode_test *param = test->param_value; + struct drm_crtc_state *crtc_state; + + crtc_state = kunit_kzalloc(test, sizeof(*crtc_state), GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, crtc_state); + + crtc_state->encoder_mask = param->encoder_mask; + + ret = drm_crtc_in_clone_mode(crtc_state); + + KUNIT_ASSERT_EQ(test, ret, param->expected_result); +} + +static void drm_check_in_clone_mode_desc(const struct drm_clone_mode_test *t, + char *desc) +{ + sprintf(desc, "%s", t->name); +} + +static const struct drm_clone_mode_test drm_clone_mode_tests[] = { + { + .name = "in_clone_mode", + .encoder_mask = DRM_TEST_ENC_0 | DRM_TEST_ENC_1, + .expected_result = true, + }, + { + .name = "not_in_clone_mode", + .encoder_mask = DRM_TEST_ENC_0, + .expected_result = false, + }, +}; + +KUNIT_ARRAY_PARAM(drm_check_in_clone_mode, drm_clone_mode_tests, + drm_check_in_clone_mode_desc); + static struct kunit_case drm_test_check_modeset_test[] = { KUNIT_CASE(drm_test_check_connector_changed_modeset), {} }; +static struct kunit_case drm_in_clone_mode_check_test[] = { + KUNIT_CASE_PARAM(drm_test_check_in_clone_mode, + drm_check_in_clone_mode_gen_params), + {} +}; + static struct kunit_suite drm_test_check_modeset_test_suite = { .name = "drm_validate_modeset", .test_cases = drm_test_check_modeset_test, }; -kunit_test_suite(drm_test_check_modeset_test_suite); +static struct kunit_suite drm_in_clone_mode_check_test_suite = { + .name = "drm_validate_clone_mode", + .test_cases = drm_in_clone_mode_check_test, +}; + +kunit_test_suites(&drm_in_clone_mode_check_test_suite, + &drm_test_check_modeset_test_suite); MODULE_AUTHOR("Jessica Zhang Date: Mon, 16 Dec 2024 16:43:14 -0800 Subject: [PATCH 0018/1443] drm: Add valid clones check Check that all encoders attached to a given CRTC are valid possible_clones of each other. Signed-off-by: Jessica Zhang Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241216-concurrent-wb-v4-3-fe220297a7f0@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/drm_atomic_helper.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 32902f77f00d..40e4e1b6c911 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -574,6 +574,30 @@ mode_valid(struct drm_atomic_state *state) return 0; } +static int drm_atomic_check_valid_clones(struct drm_atomic_state *state, + struct drm_crtc *crtc) +{ + struct drm_encoder *drm_enc; + struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, + crtc); + + drm_for_each_encoder_mask(drm_enc, crtc->dev, crtc_state->encoder_mask) { + if (!drm_enc->possible_clones) { + DRM_DEBUG("enc%d possible_clones is 0\n", drm_enc->base.id); + continue; + } + + if ((crtc_state->encoder_mask & drm_enc->possible_clones) != + crtc_state->encoder_mask) { + DRM_DEBUG("crtc%d failed valid clone check for mask 0x%x\n", + crtc->base.id, crtc_state->encoder_mask); + return -EINVAL; + } + } + + return 0; +} + /** * drm_atomic_helper_check_modeset - validate state object for modeset changes * @dev: DRM device @@ -745,6 +769,10 @@ drm_atomic_helper_check_modeset(struct drm_device *dev, ret = drm_atomic_add_affected_planes(state, crtc); if (ret != 0) return ret; + + ret = drm_atomic_check_valid_clones(state, crtc); + if (ret != 0) + return ret; } /* From 88849f24e2abba8a8951aa76ea60a72fba916afe Mon Sep 17 00:00:00 2001 From: Jessica Zhang Date: Mon, 16 Dec 2024 16:43:15 -0800 Subject: [PATCH 0019/1443] drm/tests: Add test for drm_atomic_helper_check_modeset() Add a test for drm_atomic_check_modeset() specifically to validate drm_atomic_check_valid_clones() helper Signed-off-by: Jessica Zhang Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20241216-concurrent-wb-v4-4-fe220297a7f0@quicinc.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/tests/drm_atomic_state_test.c | 71 +++++++++++++++++++ 1 file changed, 71 insertions(+) diff --git a/drivers/gpu/drm/tests/drm_atomic_state_test.c b/drivers/gpu/drm/tests/drm_atomic_state_test.c index 79bc5a9aedbf..0ab5d9ec4a75 100644 --- a/drivers/gpu/drm/tests/drm_atomic_state_test.c +++ b/drivers/gpu/drm/tests/drm_atomic_state_test.c @@ -253,12 +253,59 @@ static void drm_test_check_in_clone_mode(struct kunit *test) KUNIT_ASSERT_EQ(test, ret, param->expected_result); } +/* + * Test that the atomic commit path will succeed for valid clones (or non-cloned + * states) and fail for states where the cloned encoders are not possible_clones + * of each other. + */ +static void drm_test_check_valid_clones(struct kunit *test) +{ + int ret; + const struct drm_clone_mode_test *param = test->param_value; + struct drm_atomic_test_priv *priv; + struct drm_modeset_acquire_ctx *ctx; + struct drm_device *drm; + struct drm_atomic_state *state; + struct drm_crtc_state *crtc_state; + + priv = drm_atomic_test_init_drm_components(test, false); + KUNIT_ASSERT_NOT_NULL(test, priv); + + drm = &priv->drm; + + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + + ret = set_up_atomic_state(test, priv, NULL, ctx); + KUNIT_ASSERT_EQ(test, ret, 0); + + state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state); + + crtc_state = drm_atomic_get_crtc_state(state, priv->crtc); + KUNIT_ASSERT_NOT_NULL(test, crtc_state); + + crtc_state->encoder_mask = param->encoder_mask; + + // force modeset + crtc_state->mode_changed = true; + + ret = drm_atomic_helper_check_modeset(drm, state); + KUNIT_ASSERT_EQ(test, ret, param->expected_result); +} + static void drm_check_in_clone_mode_desc(const struct drm_clone_mode_test *t, char *desc) { sprintf(desc, "%s", t->name); } +static void drm_check_valid_clones_desc(const struct drm_clone_mode_test *t, + char *desc) +{ + sprintf(desc, "%s", t->name); +} + static const struct drm_clone_mode_test drm_clone_mode_tests[] = { { .name = "in_clone_mode", @@ -272,9 +319,31 @@ static const struct drm_clone_mode_test drm_clone_mode_tests[] = { }, }; +static const struct drm_clone_mode_test drm_valid_clone_mode_tests[] = { + { + .name = "not_in_clone_mode", + .encoder_mask = DRM_TEST_ENC_0, + .expected_result = 0, + }, + + { + .name = "valid_clone", + .encoder_mask = DRM_TEST_ENC_0 | DRM_TEST_ENC_1, + .expected_result = 0, + }, + { + .name = "invalid_clone", + .encoder_mask = DRM_TEST_ENC_0 | DRM_TEST_ENC_2, + .expected_result = -EINVAL, + }, +}; + KUNIT_ARRAY_PARAM(drm_check_in_clone_mode, drm_clone_mode_tests, drm_check_in_clone_mode_desc); +KUNIT_ARRAY_PARAM(drm_check_valid_clones, drm_valid_clone_mode_tests, + drm_check_valid_clones_desc); + static struct kunit_case drm_test_check_modeset_test[] = { KUNIT_CASE(drm_test_check_connector_changed_modeset), {} @@ -283,6 +352,8 @@ static struct kunit_case drm_test_check_modeset_test[] = { static struct kunit_case drm_in_clone_mode_check_test[] = { KUNIT_CASE_PARAM(drm_test_check_in_clone_mode, drm_check_in_clone_mode_gen_params), + KUNIT_CASE_PARAM(drm_test_check_valid_clones, + drm_check_valid_clones_gen_params), {} }; From 5d40d4fae6f2fb789f48207a9d4772bbee970b5c Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Tue, 31 Dec 2024 12:44:58 +0100 Subject: [PATCH 0020/1443] drm/ssd130x: Set SPI .id_table to prevent an SPI core warning The only reason for the ssd130x-spi driver to have an spi_device_id table is that the SPI core always reports an "spi:" MODALIAS, even when the SPI device has been registered via a Device Tree Blob. Without spi_device_id table information in the module's metadata, module autoloading would not work because there won't be an alias that matches the MODALIAS reported by the SPI core. This spi_device_id table is not needed for device matching though, since the of_device_id table is always used in this case. For this reason, the struct spi_driver .id_table member is currently not set in the SPI driver. Because the spi_device_id table is always required for module autoloading, the SPI core checks during driver registration that both an of_device_id table and a spi_device_id table are present and that they contain the same entries for all the SPI devices. Not setting the .id_table member in the driver then confuses the core and leads to the following warning when the ssd130x-spi driver is registered: [ 41.091198] SPI driver ssd130x-spi has no spi_device_id for sinowealth,sh1106 [ 41.098614] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1305 [ 41.105862] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1306 [ 41.113062] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1307 [ 41.120247] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1309 [ 41.127449] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1322 [ 41.134627] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1325 [ 41.141784] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1327 [ 41.149021] SPI driver ssd130x-spi has no spi_device_id for solomon,ssd1331 To prevent the warning, set the .id_table even though it's not necessary. Since the check is done even for built-in drivers, drop the condition to only define the ID table when the driver is built as a module. Finally, rename the variable to use the "_spi_id" convention used for ID tables. Fixes: 74373977d2ca ("drm/solomon: Add SSD130x OLED displays SPI support") Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20241231114516.2063201-1-javierm@redhat.com Signed-off-by: Javier Martinez Canillas --- drivers/gpu/drm/solomon/ssd130x-spi.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/solomon/ssd130x-spi.c b/drivers/gpu/drm/solomon/ssd130x-spi.c index 08334be38694..7c935870f7d2 100644 --- a/drivers/gpu/drm/solomon/ssd130x-spi.c +++ b/drivers/gpu/drm/solomon/ssd130x-spi.c @@ -151,7 +151,6 @@ static const struct of_device_id ssd130x_of_match[] = { }; MODULE_DEVICE_TABLE(of, ssd130x_of_match); -#if IS_MODULE(CONFIG_DRM_SSD130X_SPI) /* * The SPI core always reports a MODALIAS uevent of the form "spi:", even * if the device was registered via OF. This means that the module will not be @@ -160,7 +159,7 @@ MODULE_DEVICE_TABLE(of, ssd130x_of_match); * To workaround this issue, add a SPI device ID table. Even when this should * not be needed for this driver to match the registered SPI devices. */ -static const struct spi_device_id ssd130x_spi_table[] = { +static const struct spi_device_id ssd130x_spi_id[] = { /* ssd130x family */ { "sh1106", SH1106_ID }, { "ssd1305", SSD1305_ID }, @@ -175,14 +174,14 @@ static const struct spi_device_id ssd130x_spi_table[] = { { "ssd1331", SSD1331_ID }, { /* sentinel */ } }; -MODULE_DEVICE_TABLE(spi, ssd130x_spi_table); -#endif +MODULE_DEVICE_TABLE(spi, ssd130x_spi_id); static struct spi_driver ssd130x_spi_driver = { .driver = { .name = DRIVER_NAME, .of_match_table = ssd130x_of_match, }, + .id_table = ssd130x_spi_id, .probe = ssd130x_spi_probe, .remove = ssd130x_spi_remove, .shutdown = ssd130x_spi_shutdown, From 950942b4813f8c44dbec683fdb140cf4a238516b Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:24 +0100 Subject: [PATCH 0021/1443] accel/ivpu: Separate DB ID and CMDQ ID allocations from CMDQ allocation Move doorbell ID and command queue ID XArray allocations from command queue memory allocation function. This will allow ID allocations to be done without the need for actual memory allocation. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-2-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 88 +++++++++++++++++++++++++---------- 1 file changed, 64 insertions(+), 24 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 7149312f16e1..98e53cb38ecd 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -83,23 +83,9 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) if (!cmdq) return NULL; - ret = xa_alloc_cyclic(&vdev->db_xa, &cmdq->db_id, NULL, vdev->db_limit, &vdev->db_next, - GFP_KERNEL); - if (ret < 0) { - ivpu_err(vdev, "Failed to allocate doorbell id: %d\n", ret); - goto err_free_cmdq; - } - - ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, - &file_priv->cmdq_id_next, GFP_KERNEL); - if (ret < 0) { - ivpu_err(vdev, "Failed to allocate command queue id: %d\n", ret); - goto err_erase_db_xa; - } - cmdq->mem = ivpu_bo_create_global(vdev, SZ_4K, DRM_IVPU_BO_WC | DRM_IVPU_BO_MAPPABLE); if (!cmdq->mem) - goto err_erase_cmdq_xa; + goto err_free_cmdq; ret = ivpu_preemption_buffers_create(vdev, file_priv, cmdq); if (ret) @@ -107,10 +93,6 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) return cmdq; -err_erase_cmdq_xa: - xa_erase(&file_priv->cmdq_xa, cmdq->id); -err_erase_db_xa: - xa_erase(&vdev->db_xa, cmdq->db_id); err_free_cmdq: kfree(cmdq); return NULL; @@ -233,30 +215,88 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm return 0; } +static int ivpu_db_id_alloc(struct ivpu_device *vdev, u32 *db_id) +{ + int ret; + u32 id; + + ret = xa_alloc_cyclic(&vdev->db_xa, &id, NULL, vdev->db_limit, &vdev->db_next, GFP_KERNEL); + if (ret < 0) + return ret; + + *db_id = id; + return 0; +} + +static int ivpu_cmdq_id_alloc(struct ivpu_file_priv *file_priv, u32 *cmdq_id) +{ + int ret; + u32 id; + + ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &id, NULL, file_priv->cmdq_limit, + &file_priv->cmdq_id_next, GFP_KERNEL); + if (ret < 0) + return ret; + + *cmdq_id = id; + return 0; +} + static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u8 priority) { + struct ivpu_device *vdev = file_priv->vdev; struct ivpu_cmdq *cmdq; - unsigned long cmdq_id; + unsigned long id; int ret; lockdep_assert_held(&file_priv->lock); - xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + xa_for_each(&file_priv->cmdq_xa, id, cmdq) if (cmdq->priority == priority) break; if (!cmdq) { cmdq = ivpu_cmdq_alloc(file_priv); - if (!cmdq) + if (!cmdq) { + ivpu_err(vdev, "Failed to allocate command queue\n"); return NULL; + } + + ret = ivpu_db_id_alloc(vdev, &cmdq->db_id); + if (ret) { + ivpu_err(file_priv->vdev, "Failed to allocate doorbell ID: %d\n", ret); + goto err_free_cmdq; + } + + ret = ivpu_cmdq_id_alloc(file_priv, &cmdq->id); + if (ret) { + ivpu_err(vdev, "Failed to allocate command queue ID: %d\n", ret); + goto err_erase_db_id; + } + cmdq->priority = priority; + ret = xa_err(xa_store(&file_priv->cmdq_xa, cmdq->id, cmdq, GFP_KERNEL)); + if (ret) { + ivpu_err(vdev, "Failed to store command queue in cmdq_xa: %d\n", ret); + goto err_erase_cmdq_id; + } } ret = ivpu_cmdq_init(file_priv, cmdq, priority); - if (ret) - return NULL; + if (ret) { + ivpu_err(vdev, "Failed to initialize command queue: %d\n", ret); + goto err_free_cmdq; + } return cmdq; + +err_erase_cmdq_id: + xa_erase(&file_priv->cmdq_xa, cmdq->id); +err_erase_db_id: + xa_erase(&vdev->db_xa, cmdq->db_id); +err_free_cmdq: + ivpu_cmdq_free(file_priv, cmdq); + return NULL; } void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) From 465a3914b254f82608a8dfb580ac9930722dfd88 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:25 +0100 Subject: [PATCH 0022/1443] accel/ivpu: Add API for command queue create/destroy/submit Implement support for explicit command queue management. To allow more flexible control over command queues add capabilities to create, destroy and submit jobs to specific command queues. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-3-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 6 + drivers/accel/ivpu/ivpu_job.c | 375 +++++++++++++++++++++------------- drivers/accel/ivpu/ivpu_job.h | 5 +- include/uapi/drm/ivpu_accel.h | 84 ++++++++ 4 files changed, 322 insertions(+), 148 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index 1e8ffbe25eee..f4171536640b 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -137,6 +137,9 @@ static int ivpu_get_capabilities(struct ivpu_device *vdev, struct drm_ivpu_param case DRM_IVPU_CAP_DMA_MEMORY_RANGE: args->value = 1; break; + case DRM_IVPU_CAP_MANAGE_CMDQ: + args->value = 1; + break; default: return -EINVAL; } @@ -310,6 +313,9 @@ static const struct drm_ioctl_desc ivpu_drm_ioctls[] = { DRM_IOCTL_DEF_DRV(IVPU_METRIC_STREAMER_GET_DATA, ivpu_ms_get_data_ioctl, 0), DRM_IOCTL_DEF_DRV(IVPU_METRIC_STREAMER_STOP, ivpu_ms_stop_ioctl, 0), DRM_IOCTL_DEF_DRV(IVPU_METRIC_STREAMER_GET_INFO, ivpu_ms_get_info_ioctl, 0), + DRM_IOCTL_DEF_DRV(IVPU_CMDQ_CREATE, ivpu_cmdq_create_ioctl, 0), + DRM_IOCTL_DEF_DRV(IVPU_CMDQ_DESTROY, ivpu_cmdq_destroy_ioctl, 0), + DRM_IOCTL_DEF_DRV(IVPU_CMDQ_SUBMIT, ivpu_cmdq_submit_ioctl, 0), }; static int ivpu_wait_for_ready(struct ivpu_device *vdev) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 98e53cb38ecd..43507d3aea51 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -100,15 +100,43 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) static void ivpu_cmdq_free(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) { - if (!cmdq) - return; - ivpu_preemption_buffers_free(file_priv->vdev, file_priv, cmdq); ivpu_bo_free(cmdq->mem); - xa_erase(&file_priv->vdev->db_xa, cmdq->db_id); kfree(cmdq); } +static struct ivpu_cmdq *ivpu_cmdq_create(struct ivpu_file_priv *file_priv, u8 priority, + bool is_legacy) +{ + struct ivpu_device *vdev = file_priv->vdev; + struct ivpu_cmdq *cmdq = NULL; + int ret; + + lockdep_assert_held(&file_priv->lock); + + cmdq = ivpu_cmdq_alloc(file_priv); + if (!cmdq) { + ivpu_err(vdev, "Failed to allocate command queue\n"); + return NULL; + } + + cmdq->priority = priority; + cmdq->is_legacy = is_legacy; + + ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, + &file_priv->cmdq_id_next, GFP_KERNEL); + if (ret < 0) { + ivpu_err(vdev, "Failed to allocate command queue ID: %d\n", ret); + goto err_free_cmdq; + } + + return cmdq; + +err_free_cmdq: + ivpu_cmdq_free(file_priv, cmdq); + return NULL; +} + static int ivpu_hws_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 engine, u8 priority) { @@ -134,6 +162,13 @@ static int ivpu_register_db(struct ivpu_file_priv *file_priv, struct ivpu_cmdq * struct ivpu_device *vdev = file_priv->vdev; int ret; + ret = xa_alloc_cyclic(&vdev->db_xa, &cmdq->db_id, NULL, vdev->db_limit, &vdev->db_next, + GFP_KERNEL); + if (ret < 0) { + ivpu_err(vdev, "Failed to allocate doorbell ID: %d\n", ret); + return ret; + } + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) ret = ivpu_jsm_hws_register_db(vdev, file_priv->ctx.id, cmdq->id, cmdq->db_id, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); @@ -142,41 +177,52 @@ static int ivpu_register_db(struct ivpu_file_priv *file_priv, struct ivpu_cmdq * cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); if (!ret) - ivpu_dbg(vdev, JOB, "DB %d registered to cmdq %d ctx %d\n", - cmdq->db_id, cmdq->id, file_priv->ctx.id); + ivpu_dbg(vdev, JOB, "DB %d registered to cmdq %d ctx %d priority %d\n", + cmdq->db_id, cmdq->id, file_priv->ctx.id, cmdq->priority); + else + xa_erase(&vdev->db_xa, cmdq->db_id); return ret; } -static int -ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u8 priority) +static void ivpu_cmdq_jobq_init(struct ivpu_device *vdev, struct vpu_job_queue *jobq) +{ + jobq->header.engine_idx = VPU_ENGINE_COMPUTE; + jobq->header.head = 0; + jobq->header.tail = 0; + + if (ivpu_test_mode & IVPU_TEST_MODE_TURBO) { + ivpu_dbg(vdev, JOB, "Turbo mode enabled"); + jobq->header.flags = VPU_JOB_QUEUE_FLAGS_TURBO_MODE; + } + + wmb(); /* Flush WC buffer for jobq->header */ +} + +static inline u32 ivpu_cmdq_get_entry_count(struct ivpu_cmdq *cmdq) +{ + size_t size = ivpu_bo_size(cmdq->mem) - sizeof(struct vpu_job_queue_header); + + return size / sizeof(struct vpu_job_queue_entry); +} + +static int ivpu_cmdq_register(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) { struct ivpu_device *vdev = file_priv->vdev; - struct vpu_job_queue_header *jobq_header; int ret; lockdep_assert_held(&file_priv->lock); - if (cmdq->db_registered) + if (cmdq->db_id) return 0; - cmdq->entry_count = (u32)((ivpu_bo_size(cmdq->mem) - sizeof(struct vpu_job_queue_header)) / - sizeof(struct vpu_job_queue_entry)); - + cmdq->entry_count = ivpu_cmdq_get_entry_count(cmdq); cmdq->jobq = (struct vpu_job_queue *)ivpu_bo_vaddr(cmdq->mem); - jobq_header = &cmdq->jobq->header; - jobq_header->engine_idx = VPU_ENGINE_COMPUTE; - jobq_header->head = 0; - jobq_header->tail = 0; - if (ivpu_test_mode & IVPU_TEST_MODE_TURBO) { - ivpu_dbg(vdev, JOB, "Turbo mode enabled"); - jobq_header->flags = VPU_JOB_QUEUE_FLAGS_TURBO_MODE; - } - wmb(); /* Flush WC buffer for jobq->header */ + ivpu_cmdq_jobq_init(vdev, cmdq->jobq); if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { - ret = ivpu_hws_cmdq_init(file_priv, cmdq, VPU_ENGINE_COMPUTE, priority); + ret = ivpu_hws_cmdq_init(file_priv, cmdq, VPU_ENGINE_COMPUTE, cmdq->priority); if (ret) return ret; } @@ -185,23 +231,19 @@ ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u8 prio if (ret) return ret; - cmdq->db_registered = true; - return 0; } -static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) +static int ivpu_cmdq_unregister(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) { struct ivpu_device *vdev = file_priv->vdev; int ret; lockdep_assert_held(&file_priv->lock); - if (!cmdq->db_registered) + if (!cmdq->db_id) return 0; - cmdq->db_registered = false; - if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->id); if (!ret) @@ -212,91 +254,61 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm if (!ret) ivpu_dbg(vdev, JOB, "DB %d unregistered\n", cmdq->db_id); + xa_erase(&file_priv->vdev->db_xa, cmdq->db_id); + cmdq->db_id = 0; + return 0; } -static int ivpu_db_id_alloc(struct ivpu_device *vdev, u32 *db_id) +static inline u8 ivpu_job_to_jsm_priority(u8 priority) { - int ret; - u32 id; + if (priority == DRM_IVPU_JOB_PRIORITY_DEFAULT) + return VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL; - ret = xa_alloc_cyclic(&vdev->db_xa, &id, NULL, vdev->db_limit, &vdev->db_next, GFP_KERNEL); - if (ret < 0) - return ret; - - *db_id = id; - return 0; + return priority - 1; } -static int ivpu_cmdq_id_alloc(struct ivpu_file_priv *file_priv, u32 *cmdq_id) +static void ivpu_cmdq_destroy(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) { - int ret; - u32 id; - - ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &id, NULL, file_priv->cmdq_limit, - &file_priv->cmdq_id_next, GFP_KERNEL); - if (ret < 0) - return ret; - - *cmdq_id = id; - return 0; + ivpu_cmdq_unregister(file_priv, cmdq); + xa_erase(&file_priv->cmdq_xa, cmdq->id); + ivpu_cmdq_free(file_priv, cmdq); } -static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u8 priority) +static struct ivpu_cmdq *ivpu_cmdq_acquire_legacy(struct ivpu_file_priv *file_priv, u8 priority) { - struct ivpu_device *vdev = file_priv->vdev; struct ivpu_cmdq *cmdq; unsigned long id; - int ret; lockdep_assert_held(&file_priv->lock); xa_for_each(&file_priv->cmdq_xa, id, cmdq) - if (cmdq->priority == priority) + if (cmdq->is_legacy && cmdq->priority == priority) break; if (!cmdq) { - cmdq = ivpu_cmdq_alloc(file_priv); - if (!cmdq) { - ivpu_err(vdev, "Failed to allocate command queue\n"); + cmdq = ivpu_cmdq_create(file_priv, priority, true); + if (!cmdq) return NULL; - } - - ret = ivpu_db_id_alloc(vdev, &cmdq->db_id); - if (ret) { - ivpu_err(file_priv->vdev, "Failed to allocate doorbell ID: %d\n", ret); - goto err_free_cmdq; - } - - ret = ivpu_cmdq_id_alloc(file_priv, &cmdq->id); - if (ret) { - ivpu_err(vdev, "Failed to allocate command queue ID: %d\n", ret); - goto err_erase_db_id; - } - - cmdq->priority = priority; - ret = xa_err(xa_store(&file_priv->cmdq_xa, cmdq->id, cmdq, GFP_KERNEL)); - if (ret) { - ivpu_err(vdev, "Failed to store command queue in cmdq_xa: %d\n", ret); - goto err_erase_cmdq_id; - } - } - - ret = ivpu_cmdq_init(file_priv, cmdq, priority); - if (ret) { - ivpu_err(vdev, "Failed to initialize command queue: %d\n", ret); - goto err_free_cmdq; } return cmdq; +} -err_erase_cmdq_id: - xa_erase(&file_priv->cmdq_xa, cmdq->id); -err_erase_db_id: - xa_erase(&vdev->db_xa, cmdq->db_id); -err_free_cmdq: - ivpu_cmdq_free(file_priv, cmdq); - return NULL; +static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u32 cmdq_id) +{ + struct ivpu_device *vdev = file_priv->vdev; + struct ivpu_cmdq *cmdq; + + lockdep_assert_held(&file_priv->lock); + + cmdq = xa_load(&file_priv->cmdq_xa, cmdq_id); + if (!cmdq) { + ivpu_err(vdev, "Failed to find command queue with ID: %u\n", cmdq_id); + return NULL; + } + + return cmdq; } void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) @@ -306,11 +318,8 @@ void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) lockdep_assert_held(&file_priv->lock); - xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) { - xa_erase(&file_priv->cmdq_xa, cmdq_id); - ivpu_cmdq_fini(file_priv, cmdq); - ivpu_cmdq_free(file_priv, cmdq); - } + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + ivpu_cmdq_destroy(file_priv, cmdq); } /* @@ -326,8 +335,10 @@ static void ivpu_cmdq_reset(struct ivpu_file_priv *file_priv) mutex_lock(&file_priv->lock); - xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) - cmdq->db_registered = false; + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) { + xa_erase(&file_priv->vdev->db_xa, cmdq->db_id); + cmdq->db_id = 0; + } mutex_unlock(&file_priv->lock); } @@ -345,13 +356,13 @@ void ivpu_cmdq_reset_all_contexts(struct ivpu_device *vdev) mutex_unlock(&vdev->context_list_lock); } -static void ivpu_cmdq_fini_all(struct ivpu_file_priv *file_priv) +static void ivpu_cmdq_unregister_all(struct ivpu_file_priv *file_priv) { struct ivpu_cmdq *cmdq; unsigned long cmdq_id; xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) - ivpu_cmdq_fini(file_priv, cmdq); + ivpu_cmdq_unregister(file_priv, cmdq); } void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) @@ -360,7 +371,7 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) lockdep_assert_held(&file_priv->lock); - ivpu_cmdq_fini_all(file_priv); + ivpu_cmdq_unregister_all(file_priv); if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) ivpu_jsm_context_release(vdev, file_priv->ctx.id); @@ -549,7 +560,7 @@ void ivpu_jobs_abort_all(struct ivpu_device *vdev) ivpu_job_signal_and_destroy(vdev, id, DRM_IVPU_JOB_STATUS_ABORTED); } -static int ivpu_job_submit(struct ivpu_job *job, u8 priority) +static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) { struct ivpu_file_priv *file_priv = job->file_priv; struct ivpu_device *vdev = job->vdev; @@ -563,14 +574,22 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) mutex_lock(&file_priv->lock); - cmdq = ivpu_cmdq_acquire(file_priv, priority); + if (cmdq_id == 0) + cmdq = ivpu_cmdq_acquire_legacy(file_priv, priority); + else + cmdq = ivpu_cmdq_acquire(file_priv, cmdq_id); if (!cmdq) { - ivpu_warn_ratelimited(vdev, "Failed to get job queue, ctx %d engine %d prio %d\n", - file_priv->ctx.id, job->engine_idx, priority); + ivpu_warn_ratelimited(vdev, "Failed to get job queue, ctx %d\n", file_priv->ctx.id); ret = -EINVAL; goto err_unlock_file_priv; } + ret = ivpu_cmdq_register(file_priv, cmdq); + if (ret) { + ivpu_err(vdev, "Failed to register command queue: %d\n", ret); + goto err_unlock_file_priv; + } + xa_lock(&vdev->submitted_jobs_xa); is_first_job = xa_empty(&vdev->submitted_jobs_xa); ret = __xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, @@ -599,7 +618,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) trace_job("submit", job); ivpu_dbg(vdev, JOB, "Job submitted: id %3u ctx %2d engine %d prio %d addr 0x%llx next %d\n", - job->job_id, file_priv->ctx.id, job->engine_idx, priority, + job->job_id, file_priv->ctx.id, job->engine_idx, cmdq->priority, job->cmd_buf_vpu_addr, cmdq->jobq->header.tail); xa_unlock(&vdev->submitted_jobs_xa); @@ -625,7 +644,7 @@ static int ivpu_job_prepare_bos_for_submit(struct drm_file *file, struct ivpu_job *job, u32 *buf_handles, u32 buf_count, u32 commands_offset) { - struct ivpu_file_priv *file_priv = file->driver_priv; + struct ivpu_file_priv *file_priv = job->file_priv; struct ivpu_device *vdev = file_priv->vdev; struct ww_acquire_ctx acquire_ctx; enum dma_resv_usage usage; @@ -687,49 +706,20 @@ ivpu_job_prepare_bos_for_submit(struct drm_file *file, struct ivpu_job *job, u32 return ret; } -static inline u8 ivpu_job_to_hws_priority(struct ivpu_file_priv *file_priv, u8 priority) +static int ivpu_submit(struct drm_file *file, struct ivpu_file_priv *file_priv, u32 cmdq_id, + u32 buffer_count, u32 engine, void __user *buffers_ptr, u32 cmds_offset, + u8 priority) { - if (priority == DRM_IVPU_JOB_PRIORITY_DEFAULT) - return DRM_IVPU_JOB_PRIORITY_NORMAL; - - return priority - 1; -} - -int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) -{ - struct ivpu_file_priv *file_priv = file->driver_priv; struct ivpu_device *vdev = file_priv->vdev; - struct drm_ivpu_submit *params = data; struct ivpu_job *job; u32 *buf_handles; int idx, ret; - u8 priority; - if (params->engine != DRM_IVPU_ENGINE_COMPUTE) - return -EINVAL; - - if (params->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) - return -EINVAL; - - if (params->buffer_count == 0 || params->buffer_count > JOB_MAX_BUFFER_COUNT) - return -EINVAL; - - if (!IS_ALIGNED(params->commands_offset, 8)) - return -EINVAL; - - if (!file_priv->ctx.id) - return -EINVAL; - - if (file_priv->has_mmu_faults) - return -EBADFD; - - buf_handles = kcalloc(params->buffer_count, sizeof(u32), GFP_KERNEL); + buf_handles = kcalloc(buffer_count, sizeof(u32), GFP_KERNEL); if (!buf_handles) return -ENOMEM; - ret = copy_from_user(buf_handles, - (void __user *)params->buffers_ptr, - params->buffer_count * sizeof(u32)); + ret = copy_from_user(buf_handles, buffers_ptr, buffer_count * sizeof(u32)); if (ret) { ret = -EFAULT; goto err_free_handles; @@ -740,27 +730,23 @@ int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) goto err_free_handles; } - ivpu_dbg(vdev, JOB, "Submit ioctl: ctx %u buf_count %u\n", - file_priv->ctx.id, params->buffer_count); + ivpu_dbg(vdev, JOB, "Submit ioctl: ctx %u buf_count %u\n", file_priv->ctx.id, buffer_count); - job = ivpu_job_create(file_priv, params->engine, params->buffer_count); + job = ivpu_job_create(file_priv, engine, buffer_count); if (!job) { ivpu_err(vdev, "Failed to create job\n"); ret = -ENOMEM; goto err_exit_dev; } - ret = ivpu_job_prepare_bos_for_submit(file, job, buf_handles, params->buffer_count, - params->commands_offset); + ret = ivpu_job_prepare_bos_for_submit(file, job, buf_handles, buffer_count, cmds_offset); if (ret) { ivpu_err(vdev, "Failed to prepare job: %d\n", ret); goto err_destroy_job; } - priority = ivpu_job_to_hws_priority(file_priv, params->priority); - down_read(&vdev->pm->reset_lock); - ret = ivpu_job_submit(job, priority); + ret = ivpu_job_submit(job, priority, cmdq_id); up_read(&vdev->pm->reset_lock); if (ret) goto err_signal_fence; @@ -780,6 +766,101 @@ int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) return ret; } +int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct ivpu_file_priv *file_priv = file->driver_priv; + struct drm_ivpu_submit *args = data; + u8 priority; + + if (args->engine != DRM_IVPU_ENGINE_COMPUTE) + return -EINVAL; + + if (args->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) + return -EINVAL; + + if (args->buffer_count == 0 || args->buffer_count > JOB_MAX_BUFFER_COUNT) + return -EINVAL; + + if (!IS_ALIGNED(args->commands_offset, 8)) + return -EINVAL; + + if (!file_priv->ctx.id) + return -EINVAL; + + if (file_priv->has_mmu_faults) + return -EBADFD; + + priority = ivpu_job_to_jsm_priority(args->priority); + + return ivpu_submit(file, file_priv, 0, args->buffer_count, args->engine, + (void __user *)args->buffers_ptr, args->commands_offset, priority); +} + +int ivpu_cmdq_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct ivpu_file_priv *file_priv = file->driver_priv; + struct drm_ivpu_cmdq_submit *args = data; + + if (args->cmdq_id < IVPU_CMDQ_MIN_ID || args->cmdq_id > IVPU_CMDQ_MAX_ID) + return -EINVAL; + + if (args->buffer_count == 0 || args->buffer_count > JOB_MAX_BUFFER_COUNT) + return -EINVAL; + + if (!IS_ALIGNED(args->commands_offset, 8)) + return -EINVAL; + + if (!file_priv->ctx.id) + return -EINVAL; + + if (file_priv->has_mmu_faults) + return -EBADFD; + + return ivpu_submit(file, file_priv, args->cmdq_id, args->buffer_count, VPU_ENGINE_COMPUTE, + (void __user *)args->buffers_ptr, args->commands_offset, 0); +} + +int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct ivpu_file_priv *file_priv = file->driver_priv; + struct drm_ivpu_cmdq_create *args = data; + struct ivpu_cmdq *cmdq; + + if (args->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) + return -EINVAL; + + mutex_lock(&file_priv->lock); + + cmdq = ivpu_cmdq_create(file_priv, ivpu_job_to_jsm_priority(args->priority), false); + if (cmdq) + args->cmdq_id = cmdq->id; + + mutex_unlock(&file_priv->lock); + + return cmdq ? 0 : -ENOMEM; +} + +int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct ivpu_file_priv *file_priv = file->driver_priv; + struct drm_ivpu_cmdq_destroy *args = data; + struct ivpu_cmdq *cmdq; + int ret = 0; + + mutex_lock(&file_priv->lock); + + cmdq = xa_load(&file_priv->cmdq_xa, args->cmdq_id); + if (!cmdq || cmdq->is_legacy) { + ret = -ENOENT; + goto unlock; + } + + ivpu_cmdq_destroy(file_priv, cmdq); +unlock: + mutex_unlock(&file_priv->lock); + return ret; +} + static void ivpu_job_done_callback(struct ivpu_device *vdev, struct ivpu_ipc_hdr *ipc_hdr, struct vpu_jsm_msg *jsm_msg) diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index 8b19e3f8b4cf..4d31277bcc41 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -30,8 +30,8 @@ struct ivpu_cmdq { u32 entry_count; u32 id; u32 db_id; - bool db_registered; u8 priority; + bool is_legacy; }; /** @@ -58,6 +58,9 @@ struct ivpu_job { }; int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file); +int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file *file); +int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file); +int ivpu_cmdq_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file); void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv); diff --git a/include/uapi/drm/ivpu_accel.h b/include/uapi/drm/ivpu_accel.h index a35b97b097bf..746c43bd3eb6 100644 --- a/include/uapi/drm/ivpu_accel.h +++ b/include/uapi/drm/ivpu_accel.h @@ -22,6 +22,9 @@ extern "C" { #define DRM_IVPU_METRIC_STREAMER_STOP 0x08 #define DRM_IVPU_METRIC_STREAMER_GET_DATA 0x09 #define DRM_IVPU_METRIC_STREAMER_GET_INFO 0x0a +#define DRM_IVPU_CMDQ_CREATE 0x0b +#define DRM_IVPU_CMDQ_DESTROY 0x0c +#define DRM_IVPU_CMDQ_SUBMIT 0x0d #define DRM_IOCTL_IVPU_GET_PARAM \ DRM_IOWR(DRM_COMMAND_BASE + DRM_IVPU_GET_PARAM, struct drm_ivpu_param) @@ -57,6 +60,15 @@ extern "C" { DRM_IOWR(DRM_COMMAND_BASE + DRM_IVPU_METRIC_STREAMER_GET_INFO, \ struct drm_ivpu_metric_streamer_get_data) +#define DRM_IOCTL_IVPU_CMDQ_CREATE \ + DRM_IOWR(DRM_COMMAND_BASE + DRM_IVPU_CMDQ_CREATE, struct drm_ivpu_cmdq_create) + +#define DRM_IOCTL_IVPU_CMDQ_DESTROY \ + DRM_IOW(DRM_COMMAND_BASE + DRM_IVPU_CMDQ_DESTROY, struct drm_ivpu_cmdq_destroy) + +#define DRM_IOCTL_IVPU_CMDQ_SUBMIT \ + DRM_IOW(DRM_COMMAND_BASE + DRM_IVPU_CMDQ_SUBMIT, struct drm_ivpu_cmdq_submit) + /** * DOC: contexts * @@ -107,6 +119,13 @@ extern "C" { * accessible by hardware DMA. */ #define DRM_IVPU_CAP_DMA_MEMORY_RANGE 2 +/** + * DRM_IVPU_CAP_MANAGE_CMDQ + * + * Driver supports explicit command queue operations like command queue create, + * command queue destroy and submit job on specific command queue. + */ +#define DRM_IVPU_CAP_MANAGE_CMDQ 3 /** * struct drm_ivpu_param - Get/Set VPU parameters @@ -316,6 +335,44 @@ struct drm_ivpu_submit { __u32 priority; }; +/** + * struct drm_ivpu_cmdq_submit - Submit commands to the VPU using explicit command queue + * + * Execute a single command buffer on a given command queue. + * Handles to all referenced buffer objects have to be provided in @buffers_ptr. + * + * User space may wait on job completion using %DRM_IVPU_BO_WAIT ioctl. + */ +struct drm_ivpu_cmdq_submit { + /** + * @buffers_ptr: + * + * A pointer to an u32 array of GEM handles of the BOs required for this job. + * The number of elements in the array must be equal to the value given by @buffer_count. + * + * The first BO is the command buffer. The rest of array has to contain all + * BOs referenced from the command buffer. + */ + __u64 buffers_ptr; + + /** @buffer_count: Number of elements in the @buffers_ptr */ + __u32 buffer_count; + + /** @cmdq_id: ID for the command queue where job will be submitted */ + __u32 cmdq_id; + + /** @flags: Reserved for future use - must be zero */ + __u32 flags; + + /** + * @commands_offset: + * + * Offset inside the first buffer in @buffers_ptr containing commands + * to be executed. The offset has to be 8-byte aligned. + */ + __u32 commands_offset; +}; + /* drm_ivpu_bo_wait job status codes */ #define DRM_IVPU_JOB_STATUS_SUCCESS 0 #define DRM_IVPU_JOB_STATUS_ABORTED 256 @@ -388,6 +445,33 @@ struct drm_ivpu_metric_streamer_get_data { __u64 data_size; }; +/** + * struct drm_ivpu_cmdq_create - Create command queue for job submission + */ +struct drm_ivpu_cmdq_create { + /** @cmdq_id: Returned ID of created command queue */ + __u32 cmdq_id; + /** + * @priority: + * + * Priority to be set for related job command queue, can be one of the following: + * %DRM_IVPU_JOB_PRIORITY_DEFAULT + * %DRM_IVPU_JOB_PRIORITY_IDLE + * %DRM_IVPU_JOB_PRIORITY_NORMAL + * %DRM_IVPU_JOB_PRIORITY_FOCUS + * %DRM_IVPU_JOB_PRIORITY_REALTIME + */ + __u32 priority; +}; + +/** + * struct drm_ivpu_cmdq_destroy - Destroy a command queue + */ +struct drm_ivpu_cmdq_destroy { + /** @cmdq_id: ID of command queue to destroy */ + __u32 cmdq_id; +}; + /** * struct drm_ivpu_metric_streamer_stop - Stop collecting metric data */ From 5bbccadaf33eea2b879d8326ad59ae0663be47d1 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:26 +0100 Subject: [PATCH 0023/1443] accel/ivpu: Abort all jobs after command queue unregister With hardware scheduler it is not expected to receive JOB_DONE notifications from NPU FW for the jobs aborted due to command queue destroy JSM command. Remove jobs submitted to unregistered command queue from submitted_jobs_xa to avoid triggering a TDR in such case. Add explicit submitted_jobs_lock that protects access to list of submitted jobs which is now used to find jobs to abort. Move context abort procedure to separate work queue not to slow down handling of IPCs or DCT requests in case where job abort takes longer, especially when destruction of the last job of a specific context results in context release. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-4-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 32 ++------ drivers/accel/ivpu/ivpu_drv.h | 2 + drivers/accel/ivpu/ivpu_job.c | 137 ++++++++++++++++++++++++-------- drivers/accel/ivpu/ivpu_job.h | 4 + drivers/accel/ivpu/ivpu_mmu.c | 3 +- drivers/accel/ivpu/ivpu_sysfs.c | 5 +- 6 files changed, 121 insertions(+), 62 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index f4171536640b..300eea8c305f 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -36,8 +36,6 @@ #define DRIVER_VERSION_STR "1.0.0 " UTS_RELEASE #endif -static struct lock_class_key submitted_jobs_xa_lock_class_key; - int ivpu_dbg_mask; module_param_named(dbg_mask, ivpu_dbg_mask, int, 0644); MODULE_PARM_DESC(dbg_mask, "Driver debug mask. See IVPU_DBG_* macros."); @@ -467,26 +465,6 @@ static const struct drm_driver driver = { .major = 1, }; -static void ivpu_context_abort_invalid(struct ivpu_device *vdev) -{ - struct ivpu_file_priv *file_priv; - unsigned long ctx_id; - - mutex_lock(&vdev->context_list_lock); - - xa_for_each(&vdev->context_xa, ctx_id, file_priv) { - if (!file_priv->has_mmu_faults || file_priv->aborted) - continue; - - mutex_lock(&file_priv->lock); - ivpu_context_abort_locked(file_priv); - file_priv->aborted = true; - mutex_unlock(&file_priv->lock); - } - - mutex_unlock(&vdev->context_list_lock); -} - static irqreturn_t ivpu_irq_thread_handler(int irq, void *arg) { struct ivpu_device *vdev = arg; @@ -500,9 +478,6 @@ static irqreturn_t ivpu_irq_thread_handler(int irq, void *arg) case IVPU_HW_IRQ_SRC_IPC: ivpu_ipc_irq_thread_handler(vdev); break; - case IVPU_HW_IRQ_SRC_MMU_EVTQ: - ivpu_context_abort_invalid(vdev); - break; case IVPU_HW_IRQ_SRC_DCT: ivpu_pm_dct_irq_thread_handler(vdev); break; @@ -619,16 +594,21 @@ static int ivpu_dev_init(struct ivpu_device *vdev) xa_init_flags(&vdev->context_xa, XA_FLAGS_ALLOC | XA_FLAGS_LOCK_IRQ); xa_init_flags(&vdev->submitted_jobs_xa, XA_FLAGS_ALLOC1); xa_init_flags(&vdev->db_xa, XA_FLAGS_ALLOC1); - lockdep_set_class(&vdev->submitted_jobs_xa.xa_lock, &submitted_jobs_xa_lock_class_key); INIT_LIST_HEAD(&vdev->bo_list); vdev->db_limit.min = IVPU_MIN_DB; vdev->db_limit.max = IVPU_MAX_DB; + INIT_WORK(&vdev->context_abort_work, ivpu_context_abort_thread_handler); + ret = drmm_mutex_init(&vdev->drm, &vdev->context_list_lock); if (ret) goto err_xa_destroy; + ret = drmm_mutex_init(&vdev->drm, &vdev->submitted_jobs_lock); + if (ret) + goto err_xa_destroy; + ret = drmm_mutex_init(&vdev->drm, &vdev->bo_list_lock); if (ret) goto err_xa_destroy; diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index 3fdff3f6cffd..ebfcf3e42a3d 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -137,6 +137,7 @@ struct ivpu_device { struct mutex context_list_lock; /* Protects user context addition/removal */ struct xarray context_xa; struct xa_limit context_xa_limit; + struct work_struct context_abort_work; struct xarray db_xa; struct xa_limit db_limit; @@ -145,6 +146,7 @@ struct ivpu_device { struct mutex bo_list_lock; /* Protects bo_list */ struct list_head bo_list; + struct mutex submitted_jobs_lock; /* Protects submitted_jobs */ struct xarray submitted_jobs_xa; struct ivpu_ipc_consumer job_done_consumer; diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 43507d3aea51..7fed3c8406ee 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -122,6 +122,7 @@ static struct ivpu_cmdq *ivpu_cmdq_create(struct ivpu_file_priv *file_priv, u8 p cmdq->priority = priority; cmdq->is_legacy = is_legacy; + cmdq->is_valid = true; ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, &file_priv->cmdq_id_next, GFP_KERNEL); @@ -130,6 +131,7 @@ static struct ivpu_cmdq *ivpu_cmdq_create(struct ivpu_file_priv *file_priv, u8 p goto err_free_cmdq; } + ivpu_dbg(vdev, JOB, "Command queue %d created, ctx %d\n", cmdq->id, file_priv->ctx.id); return cmdq; err_free_cmdq: @@ -247,7 +249,8 @@ static int ivpu_cmdq_unregister(struct ivpu_file_priv *file_priv, struct ivpu_cm if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->id); if (!ret) - ivpu_dbg(vdev, JOB, "Command queue %d destroyed\n", cmdq->id); + ivpu_dbg(vdev, JOB, "Command queue %d destroyed, ctx %d\n", + cmdq->id, file_priv->ctx.id); } ret = ivpu_jsm_unregister_db(vdev, cmdq->db_id); @@ -303,8 +306,8 @@ static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u32 lockdep_assert_held(&file_priv->lock); cmdq = xa_load(&file_priv->cmdq_xa, cmdq_id); - if (!cmdq) { - ivpu_err(vdev, "Failed to find command queue with ID: %u\n", cmdq_id); + if (!cmdq || !cmdq->is_valid) { + ivpu_warn_ratelimited(vdev, "Failed to find command queue with ID: %u\n", cmdq_id); return NULL; } @@ -356,25 +359,21 @@ void ivpu_cmdq_reset_all_contexts(struct ivpu_device *vdev) mutex_unlock(&vdev->context_list_lock); } -static void ivpu_cmdq_unregister_all(struct ivpu_file_priv *file_priv) -{ - struct ivpu_cmdq *cmdq; - unsigned long cmdq_id; - - xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) - ivpu_cmdq_unregister(file_priv, cmdq); -} - void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) { struct ivpu_device *vdev = file_priv->vdev; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; lockdep_assert_held(&file_priv->lock); - ivpu_cmdq_unregister_all(file_priv); + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + ivpu_cmdq_unregister(file_priv, cmdq); if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) ivpu_jsm_context_release(vdev, file_priv->ctx.id); + + file_priv->aborted = true; } static int ivpu_cmdq_push_job(struct ivpu_cmdq *cmdq, struct ivpu_job *job) @@ -513,16 +512,14 @@ static struct ivpu_job *ivpu_job_remove_from_submitted_jobs(struct ivpu_device * { struct ivpu_job *job; - xa_lock(&vdev->submitted_jobs_xa); - job = __xa_erase(&vdev->submitted_jobs_xa, job_id); + lockdep_assert_held(&vdev->submitted_jobs_lock); + job = xa_erase(&vdev->submitted_jobs_xa, job_id); if (xa_empty(&vdev->submitted_jobs_xa) && job) { vdev->busy_time = ktime_add(ktime_sub(ktime_get(), vdev->busy_start_ts), vdev->busy_time); } - xa_unlock(&vdev->submitted_jobs_xa); - return job; } @@ -530,6 +527,8 @@ static int ivpu_job_signal_and_destroy(struct ivpu_device *vdev, u32 job_id, u32 { struct ivpu_job *job; + lockdep_assert_held(&vdev->submitted_jobs_lock); + job = ivpu_job_remove_from_submitted_jobs(vdev, job_id); if (!job) return -ENOENT; @@ -548,6 +547,10 @@ static int ivpu_job_signal_and_destroy(struct ivpu_device *vdev, u32 job_id, u32 ivpu_stop_job_timeout_detection(vdev); ivpu_rpm_put(vdev); + + if (!xa_empty(&vdev->submitted_jobs_xa)) + ivpu_start_job_timeout_detection(vdev); + return 0; } @@ -556,8 +559,26 @@ void ivpu_jobs_abort_all(struct ivpu_device *vdev) struct ivpu_job *job; unsigned long id; + mutex_lock(&vdev->submitted_jobs_lock); + xa_for_each(&vdev->submitted_jobs_xa, id, job) ivpu_job_signal_and_destroy(vdev, id, DRM_IVPU_JOB_STATUS_ABORTED); + + mutex_unlock(&vdev->submitted_jobs_lock); +} + +void ivpu_cmdq_abort_all_jobs(struct ivpu_device *vdev, u32 ctx_id, u32 cmdq_id) +{ + struct ivpu_job *job; + unsigned long id; + + mutex_lock(&vdev->submitted_jobs_lock); + + xa_for_each(&vdev->submitted_jobs_xa, id, job) + if (job->file_priv->ctx.id == ctx_id && job->cmdq_id == cmdq_id) + ivpu_job_signal_and_destroy(vdev, id, DRM_IVPU_JOB_STATUS_ABORTED); + + mutex_unlock(&vdev->submitted_jobs_lock); } static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) @@ -590,15 +611,18 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) goto err_unlock_file_priv; } - xa_lock(&vdev->submitted_jobs_xa); + job->cmdq_id = cmdq->id; + + mutex_lock(&vdev->submitted_jobs_lock); + is_first_job = xa_empty(&vdev->submitted_jobs_xa); - ret = __xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, - &file_priv->job_id_next, GFP_KERNEL); + ret = xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, + &file_priv->job_id_next, GFP_KERNEL); if (ret < 0) { ivpu_dbg(vdev, JOB, "Too many active jobs in ctx %d\n", file_priv->ctx.id); ret = -EBUSY; - goto err_unlock_submitted_jobs_xa; + goto err_unlock_submitted_jobs; } ret = ivpu_cmdq_push_job(cmdq, job); @@ -621,19 +645,21 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) job->job_id, file_priv->ctx.id, job->engine_idx, cmdq->priority, job->cmd_buf_vpu_addr, cmdq->jobq->header.tail); - xa_unlock(&vdev->submitted_jobs_xa); - + mutex_unlock(&vdev->submitted_jobs_lock); mutex_unlock(&file_priv->lock); - if (unlikely(ivpu_test_mode & IVPU_TEST_MODE_NULL_HW)) + if (unlikely(ivpu_test_mode & IVPU_TEST_MODE_NULL_HW)) { + mutex_lock(&vdev->submitted_jobs_lock); ivpu_job_signal_and_destroy(vdev, job->job_id, VPU_JSM_STATUS_SUCCESS); + mutex_unlock(&vdev->submitted_jobs_lock); + } return 0; err_erase_xa: - __xa_erase(&vdev->submitted_jobs_xa, job->job_id); -err_unlock_submitted_jobs_xa: - xa_unlock(&vdev->submitted_jobs_xa); + xa_erase(&vdev->submitted_jobs_xa, job->job_id); +err_unlock_submitted_jobs: + mutex_unlock(&vdev->submitted_jobs_lock); err_unlock_file_priv: mutex_unlock(&file_priv->lock); ivpu_rpm_put(vdev); @@ -843,19 +869,33 @@ int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file * int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { struct ivpu_file_priv *file_priv = file->driver_priv; + struct ivpu_device *vdev = file_priv->vdev; struct drm_ivpu_cmdq_destroy *args = data; struct ivpu_cmdq *cmdq; + u32 cmdq_id; int ret = 0; mutex_lock(&file_priv->lock); cmdq = xa_load(&file_priv->cmdq_xa, args->cmdq_id); - if (!cmdq || cmdq->is_legacy) { + if (!cmdq || !cmdq->is_valid || cmdq->is_legacy) { ret = -ENOENT; goto unlock; } + /* + * There is no way to stop executing jobs per command queue + * in OS scheduling mode, mark command queue as invalid instead + * and it will be freed together with context release. + */ + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) { + cmdq->is_valid = false; + goto unlock; + } + + cmdq_id = cmdq->id; ivpu_cmdq_destroy(file_priv, cmdq); + ivpu_cmdq_abort_all_jobs(vdev, file_priv->ctx.id, cmdq_id); unlock: mutex_unlock(&file_priv->lock); return ret; @@ -866,7 +906,6 @@ ivpu_job_done_callback(struct ivpu_device *vdev, struct ivpu_ipc_hdr *ipc_hdr, struct vpu_jsm_msg *jsm_msg) { struct vpu_ipc_msg_payload_job_done *payload; - int ret; if (!jsm_msg) { ivpu_err(vdev, "IPC message has no JSM payload\n"); @@ -879,9 +918,10 @@ ivpu_job_done_callback(struct ivpu_device *vdev, struct ivpu_ipc_hdr *ipc_hdr, } payload = (struct vpu_ipc_msg_payload_job_done *)&jsm_msg->payload; - ret = ivpu_job_signal_and_destroy(vdev, payload->job_id, payload->job_status); - if (!ret && !xa_empty(&vdev->submitted_jobs_xa)) - ivpu_start_job_timeout_detection(vdev); + + mutex_lock(&vdev->submitted_jobs_lock); + ivpu_job_signal_and_destroy(vdev, payload->job_id, payload->job_status); + mutex_unlock(&vdev->submitted_jobs_lock); } void ivpu_job_done_consumer_init(struct ivpu_device *vdev) @@ -894,3 +934,36 @@ void ivpu_job_done_consumer_fini(struct ivpu_device *vdev) { ivpu_ipc_consumer_del(vdev, &vdev->job_done_consumer); } + +void ivpu_context_abort_thread_handler(struct work_struct *work) +{ + struct ivpu_device *vdev = container_of(work, struct ivpu_device, context_abort_work); + struct ivpu_file_priv *file_priv; + unsigned long ctx_id; + struct ivpu_job *job; + unsigned long id; + + mutex_lock(&vdev->context_list_lock); + xa_for_each(&vdev->context_xa, ctx_id, file_priv) { + if (!file_priv->has_mmu_faults || file_priv->aborted) + continue; + + mutex_lock(&file_priv->lock); + ivpu_context_abort_locked(file_priv); + mutex_unlock(&file_priv->lock); + } + mutex_unlock(&vdev->context_list_lock); + + if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) + return; + /* + * In hardware scheduling mode NPU already has stopped processing jobs + * and won't send us any further notifications, thus we have to free job related resources + * and notify userspace + */ + mutex_lock(&vdev->submitted_jobs_lock); + xa_for_each(&vdev->submitted_jobs_xa, id, job) + if (job->file_priv->aborted) + ivpu_job_signal_and_destroy(vdev, job->job_id, DRM_IVPU_JOB_STATUS_ABORTED); + mutex_unlock(&vdev->submitted_jobs_lock); +} diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index 4d31277bcc41..fef8aed1fc88 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -31,6 +31,7 @@ struct ivpu_cmdq { u32 id; u32 db_id; u8 priority; + bool is_valid; bool is_legacy; }; @@ -51,6 +52,7 @@ struct ivpu_job { struct ivpu_file_priv *file_priv; struct dma_fence *done_fence; u64 cmd_buf_vpu_addr; + u32 cmdq_id; u32 job_id; u32 engine_idx; size_t bo_count; @@ -66,9 +68,11 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv); void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv); void ivpu_cmdq_reset_all_contexts(struct ivpu_device *vdev); +void ivpu_cmdq_abort_all_jobs(struct ivpu_device *vdev, u32 ctx_id, u32 cmdq_id); void ivpu_job_done_consumer_init(struct ivpu_device *vdev); void ivpu_job_done_consumer_fini(struct ivpu_device *vdev); +void ivpu_context_abort_thread_handler(struct work_struct *work); void ivpu_jobs_abort_all(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index 26ef52fbb93e..21f820dd0c65 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -890,8 +890,7 @@ void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev) REGV_WR32(IVPU_MMU_REG_EVTQ_CONS_SEC, vdev->mmu->evtq.cons); } - if (!kfifo_put(&vdev->hw->irq.fifo, IVPU_HW_IRQ_SRC_MMU_EVTQ)) - ivpu_err_ratelimited(vdev, "IRQ FIFO full\n"); + queue_work(system_wq, &vdev->context_abort_work); } void ivpu_mmu_evtq_dump(struct ivpu_device *vdev) diff --git a/drivers/accel/ivpu/ivpu_sysfs.c b/drivers/accel/ivpu/ivpu_sysfs.c index 616477fc17fa..8a616791c32f 100644 --- a/drivers/accel/ivpu/ivpu_sysfs.c +++ b/drivers/accel/ivpu/ivpu_sysfs.c @@ -30,11 +30,12 @@ npu_busy_time_us_show(struct device *dev, struct device_attribute *attr, char *b struct ivpu_device *vdev = to_ivpu_device(drm); ktime_t total, now = 0; - xa_lock(&vdev->submitted_jobs_xa); + mutex_lock(&vdev->submitted_jobs_lock); + total = vdev->busy_time; if (!xa_empty(&vdev->submitted_jobs_xa)) now = ktime_sub(ktime_get(), vdev->busy_start_ts); - xa_unlock(&vdev->submitted_jobs_xa); + mutex_unlock(&vdev->submitted_jobs_lock); return sysfs_emit(buf, "%lld\n", ktime_to_us(ktime_add(total, now))); } From 7bfc9fa99580b9b94da1c4cdcdebe3f792c1fe40 Mon Sep 17 00:00:00 2001 From: Maciej Falkowski Date: Tue, 7 Jan 2025 18:32:27 +0100 Subject: [PATCH 0024/1443] accel/ivpu: Expose NPU memory utilization info in sysfs Expose NPU memory utilization info in sysfs in bytes to show total memory used by NPU (FW + runtime). Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-5-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_sysfs.c | 30 +++++++++++++++++++++++++++++- 1 file changed, 29 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_sysfs.c b/drivers/accel/ivpu/ivpu_sysfs.c index 8a616791c32f..97102feaf8dd 100644 --- a/drivers/accel/ivpu/ivpu_sysfs.c +++ b/drivers/accel/ivpu/ivpu_sysfs.c @@ -7,11 +7,14 @@ #include #include "ivpu_drv.h" +#include "ivpu_gem.h" #include "ivpu_fw.h" #include "ivpu_hw.h" #include "ivpu_sysfs.h" -/* +/** + * DOC: npu_busy_time_us + * * npu_busy_time_us is the time that the device spent executing jobs. * The time is counted when and only when there are jobs submitted to firmware. * @@ -42,6 +45,30 @@ npu_busy_time_us_show(struct device *dev, struct device_attribute *attr, char *b static DEVICE_ATTR_RO(npu_busy_time_us); +/** + * DOC: npu_memory_utilization + * + * The npu_memory_utilization is used to report in bytes a current NPU memory utilization. + * + */ +static ssize_t +npu_memory_utilization_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_device *drm = dev_get_drvdata(dev); + struct ivpu_device *vdev = to_ivpu_device(drm); + struct ivpu_bo *bo; + u64 total_npu_memory = 0; + + mutex_lock(&vdev->bo_list_lock); + list_for_each_entry(bo, &vdev->bo_list, bo_list_node) + total_npu_memory += bo->base.base.size; + mutex_unlock(&vdev->bo_list_lock); + + return sysfs_emit(buf, "%lld\n", total_npu_memory); +} + +static DEVICE_ATTR_RO(npu_memory_utilization); + /** * DOC: sched_mode * @@ -65,6 +92,7 @@ static DEVICE_ATTR_RO(sched_mode); static struct attribute *ivpu_dev_attrs[] = { &dev_attr_npu_busy_time_us.attr, + &dev_attr_npu_memory_utilization.attr, &dev_attr_sched_mode.attr, NULL, }; From bc3e5f48b7ee021371dc37297678f7089be6ce28 Mon Sep 17 00:00:00 2001 From: Maciej Falkowski Date: Tue, 7 Jan 2025 18:32:28 +0100 Subject: [PATCH 0025/1443] accel/ivpu: Use workqueue for IRQ handling Convert IRQ bottom half from the thread handler into workqueue. This increases a stability in rare scenarios where driver on debugging/hardening kernels processes IRQ too slow and misses some interrupts due to it. Workqueue handler also gives a very minor performance increase. Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-6-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 39 ++++++++----------------------- drivers/accel/ivpu/ivpu_drv.h | 5 +++- drivers/accel/ivpu/ivpu_hw.c | 5 ---- drivers/accel/ivpu/ivpu_hw.h | 9 ------- drivers/accel/ivpu/ivpu_hw_btrs.c | 3 +-- drivers/accel/ivpu/ivpu_ipc.c | 7 +++--- drivers/accel/ivpu/ivpu_ipc.h | 2 +- drivers/accel/ivpu/ivpu_job.c | 2 +- drivers/accel/ivpu/ivpu_job.h | 2 +- drivers/accel/ivpu/ivpu_pm.c | 3 ++- drivers/accel/ivpu/ivpu_pm.h | 2 +- 11 files changed, 24 insertions(+), 55 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index 300eea8c305f..9b0d99873fb3 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include @@ -421,6 +422,9 @@ void ivpu_prepare_for_reset(struct ivpu_device *vdev) { ivpu_hw_irq_disable(vdev); disable_irq(vdev->irq); + cancel_work_sync(&vdev->irq_ipc_work); + cancel_work_sync(&vdev->irq_dct_work); + cancel_work_sync(&vdev->context_abort_work); ivpu_ipc_disable(vdev); ivpu_mmu_disable(vdev); } @@ -465,31 +469,6 @@ static const struct drm_driver driver = { .major = 1, }; -static irqreturn_t ivpu_irq_thread_handler(int irq, void *arg) -{ - struct ivpu_device *vdev = arg; - u8 irq_src; - - if (kfifo_is_empty(&vdev->hw->irq.fifo)) - return IRQ_NONE; - - while (kfifo_get(&vdev->hw->irq.fifo, &irq_src)) { - switch (irq_src) { - case IVPU_HW_IRQ_SRC_IPC: - ivpu_ipc_irq_thread_handler(vdev); - break; - case IVPU_HW_IRQ_SRC_DCT: - ivpu_pm_dct_irq_thread_handler(vdev); - break; - default: - ivpu_err_ratelimited(vdev, "Unknown IRQ source: %u\n", irq_src); - break; - } - } - - return IRQ_HANDLED; -} - static int ivpu_irq_init(struct ivpu_device *vdev) { struct pci_dev *pdev = to_pci_dev(vdev->drm.dev); @@ -501,12 +480,16 @@ static int ivpu_irq_init(struct ivpu_device *vdev) return ret; } + INIT_WORK(&vdev->irq_ipc_work, ivpu_ipc_irq_work_fn); + INIT_WORK(&vdev->irq_dct_work, ivpu_pm_irq_dct_work_fn); + INIT_WORK(&vdev->context_abort_work, ivpu_context_abort_work_fn); + ivpu_irq_handlers_init(vdev); vdev->irq = pci_irq_vector(pdev, 0); - ret = devm_request_threaded_irq(vdev->drm.dev, vdev->irq, ivpu_hw_irq_handler, - ivpu_irq_thread_handler, IRQF_NO_AUTOEN, DRIVER_NAME, vdev); + ret = devm_request_irq(vdev->drm.dev, vdev->irq, ivpu_hw_irq_handler, + IRQF_NO_AUTOEN, DRIVER_NAME, vdev); if (ret) ivpu_err(vdev, "Failed to request an IRQ %d\n", ret); @@ -599,8 +582,6 @@ static int ivpu_dev_init(struct ivpu_device *vdev) vdev->db_limit.min = IVPU_MIN_DB; vdev->db_limit.max = IVPU_MAX_DB; - INIT_WORK(&vdev->context_abort_work, ivpu_context_abort_thread_handler); - ret = drmm_mutex_init(&vdev->drm, &vdev->context_list_lock); if (ret) goto err_xa_destroy; diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index ebfcf3e42a3d..b57d878f2fcd 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -137,12 +137,15 @@ struct ivpu_device { struct mutex context_list_lock; /* Protects user context addition/removal */ struct xarray context_xa; struct xa_limit context_xa_limit; - struct work_struct context_abort_work; struct xarray db_xa; struct xa_limit db_limit; u32 db_next; + struct work_struct irq_ipc_work; + struct work_struct irq_dct_work; + struct work_struct context_abort_work; + struct mutex bo_list_lock; /* Protects bo_list */ struct list_head bo_list; diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 4e1054f3466e..1b691375ee4d 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -285,8 +285,6 @@ void ivpu_hw_profiling_freq_drive(struct ivpu_device *vdev, bool enable) void ivpu_irq_handlers_init(struct ivpu_device *vdev) { - INIT_KFIFO(vdev->hw->irq.fifo); - if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) vdev->hw->irq.ip_irq_handler = ivpu_hw_ip_irq_handler_37xx; else @@ -300,7 +298,6 @@ void ivpu_irq_handlers_init(struct ivpu_device *vdev) void ivpu_hw_irq_enable(struct ivpu_device *vdev) { - kfifo_reset(&vdev->hw->irq.fifo); ivpu_hw_ip_irq_enable(vdev); ivpu_hw_btrs_irq_enable(vdev); } @@ -327,8 +324,6 @@ irqreturn_t ivpu_hw_irq_handler(int irq, void *ptr) /* Re-enable global interrupts to re-trigger MSI for pending interrupts */ ivpu_hw_btrs_global_int_enable(vdev); - if (!kfifo_is_empty(&vdev->hw->irq.fifo)) - return IRQ_WAKE_THREAD; if (ip_handled || btrs_handled) return IRQ_HANDLED; return IRQ_NONE; diff --git a/drivers/accel/ivpu/ivpu_hw.h b/drivers/accel/ivpu/ivpu_hw.h index fc4dbfc980c8..fbef9816b9d0 100644 --- a/drivers/accel/ivpu/ivpu_hw.h +++ b/drivers/accel/ivpu/ivpu_hw.h @@ -6,18 +6,10 @@ #ifndef __IVPU_HW_H__ #define __IVPU_HW_H__ -#include - #include "ivpu_drv.h" #include "ivpu_hw_btrs.h" #include "ivpu_hw_ip.h" -#define IVPU_HW_IRQ_FIFO_LENGTH 1024 - -#define IVPU_HW_IRQ_SRC_IPC 1 -#define IVPU_HW_IRQ_SRC_MMU_EVTQ 2 -#define IVPU_HW_IRQ_SRC_DCT 3 - struct ivpu_addr_range { resource_size_t start; resource_size_t end; @@ -27,7 +19,6 @@ struct ivpu_hw_info { struct { bool (*btrs_irq_handler)(struct ivpu_device *vdev, int irq); bool (*ip_irq_handler)(struct ivpu_device *vdev, int irq); - DECLARE_KFIFO(fifo, u8, IVPU_HW_IRQ_FIFO_LENGTH); } irq; struct { struct ivpu_addr_range global; diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c index 3212c99f3682..3753b00ed2d6 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.c +++ b/drivers/accel/ivpu/ivpu_hw_btrs.c @@ -630,8 +630,7 @@ bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq) if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, SURV_ERR, status)) { ivpu_dbg(vdev, IRQ, "Survivability IRQ\n"); - if (!kfifo_put(&vdev->hw->irq.fifo, IVPU_HW_IRQ_SRC_DCT)) - ivpu_err_ratelimited(vdev, "IRQ FIFO full\n"); + queue_work(system_wq, &vdev->irq_dct_work); } if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, FREQ_CHANGE, status)) diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c index 01ebf88fe6ef..0e096fd9b95d 100644 --- a/drivers/accel/ivpu/ivpu_ipc.c +++ b/drivers/accel/ivpu/ivpu_ipc.c @@ -459,13 +459,12 @@ void ivpu_ipc_irq_handler(struct ivpu_device *vdev) } } - if (!list_empty(&ipc->cb_msg_list)) - if (!kfifo_put(&vdev->hw->irq.fifo, IVPU_HW_IRQ_SRC_IPC)) - ivpu_err_ratelimited(vdev, "IRQ FIFO full\n"); + queue_work(system_wq, &vdev->irq_ipc_work); } -void ivpu_ipc_irq_thread_handler(struct ivpu_device *vdev) +void ivpu_ipc_irq_work_fn(struct work_struct *work) { + struct ivpu_device *vdev = container_of(work, struct ivpu_device, irq_ipc_work); struct ivpu_ipc_info *ipc = vdev->ipc; struct ivpu_ipc_rx_msg *rx_msg, *r; struct list_head cb_msg_list; diff --git a/drivers/accel/ivpu/ivpu_ipc.h b/drivers/accel/ivpu/ivpu_ipc.h index b4dfb504679b..b524a1985b9d 100644 --- a/drivers/accel/ivpu/ivpu_ipc.h +++ b/drivers/accel/ivpu/ivpu_ipc.h @@ -90,7 +90,7 @@ void ivpu_ipc_disable(struct ivpu_device *vdev); void ivpu_ipc_reset(struct ivpu_device *vdev); void ivpu_ipc_irq_handler(struct ivpu_device *vdev); -void ivpu_ipc_irq_thread_handler(struct ivpu_device *vdev); +void ivpu_ipc_irq_work_fn(struct work_struct *work); void ivpu_ipc_consumer_add(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, u32 channel, ivpu_ipc_rx_callback_t callback); diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 7fed3c8406ee..c678dcddb8d8 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -935,7 +935,7 @@ void ivpu_job_done_consumer_fini(struct ivpu_device *vdev) ivpu_ipc_consumer_del(vdev, &vdev->job_done_consumer); } -void ivpu_context_abort_thread_handler(struct work_struct *work) +void ivpu_context_abort_work_fn(struct work_struct *work) { struct ivpu_device *vdev = container_of(work, struct ivpu_device, context_abort_work); struct ivpu_file_priv *file_priv; diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index fef8aed1fc88..ff77ee1fcee2 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -72,7 +72,7 @@ void ivpu_cmdq_abort_all_jobs(struct ivpu_device *vdev, u32 ctx_id, u32 cmdq_id) void ivpu_job_done_consumer_init(struct ivpu_device *vdev); void ivpu_job_done_consumer_fini(struct ivpu_device *vdev); -void ivpu_context_abort_thread_handler(struct work_struct *work); +void ivpu_context_abort_work_fn(struct work_struct *work); void ivpu_jobs_abort_all(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index 6821051dfa3a..f41b3bfe40af 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -452,8 +452,9 @@ int ivpu_pm_dct_disable(struct ivpu_device *vdev) return 0; } -void ivpu_pm_dct_irq_thread_handler(struct ivpu_device *vdev) +void ivpu_pm_irq_dct_work_fn(struct work_struct *work) { + struct ivpu_device *vdev = container_of(work, struct ivpu_device, irq_dct_work); bool enable; int ret; diff --git a/drivers/accel/ivpu/ivpu_pm.h b/drivers/accel/ivpu/ivpu_pm.h index b70efe6c36e4..89b264cc0e3e 100644 --- a/drivers/accel/ivpu/ivpu_pm.h +++ b/drivers/accel/ivpu/ivpu_pm.h @@ -45,6 +45,6 @@ void ivpu_stop_job_timeout_detection(struct ivpu_device *vdev); int ivpu_pm_dct_init(struct ivpu_device *vdev); int ivpu_pm_dct_enable(struct ivpu_device *vdev, u8 active_percent); int ivpu_pm_dct_disable(struct ivpu_device *vdev); -void ivpu_pm_dct_irq_thread_handler(struct ivpu_device *vdev); +void ivpu_pm_irq_dct_work_fn(struct work_struct *work); #endif /* __IVPU_PM_H__ */ From 0240fa18d247c99a1967f2fed025296a89a1c5f5 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:29 +0100 Subject: [PATCH 0026/1443] accel/ivpu: Dump only first MMU fault from single context Stop dumping consecutive faults from an already faulty context immediately, instead of waiting for the context abort thread handler (IRQ handler bottom half) to abort currently executing jobs. Remove 'R' (record events) bit from context descriptor of a faulty context to prevent future faults generation. This change speeds up the IRQ handler by eliminating the need to print the fault content repeatedly. Additionally, it prevents flooding dmesg with errors, which was occurring due to the delay in the bottom half of the handler stopping fault-generating jobs. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-7-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_mmu.c | 51 ++++++++++++++++++++++++--- drivers/accel/ivpu/ivpu_mmu_context.c | 13 ------- drivers/accel/ivpu/ivpu_mmu_context.h | 2 -- 3 files changed, 46 insertions(+), 20 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index 21f820dd0c65..5ee4df892b3e 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -870,23 +870,64 @@ static u32 *ivpu_mmu_get_event(struct ivpu_device *vdev) return evt; } +static int ivpu_mmu_disable_events(struct ivpu_device *vdev, u32 ssid) +{ + struct ivpu_mmu_info *mmu = vdev->mmu; + struct ivpu_mmu_cdtab *cdtab = &mmu->cdtab; + u64 *entry; + u64 val; + + if (ssid > IVPU_MMU_CDTAB_ENT_COUNT) + return -EINVAL; + + entry = cdtab->base + (ssid * IVPU_MMU_CDTAB_ENT_SIZE); + + val = READ_ONCE(entry[0]); + val &= ~IVPU_MMU_CD_0_R; + WRITE_ONCE(entry[0], val); + + if (!ivpu_is_force_snoop_enabled(vdev)) + clflush_cache_range(entry, IVPU_MMU_CDTAB_ENT_SIZE); + + ivpu_mmu_cmdq_write_cfgi_all(vdev); + + return 0; +} + void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev) { + struct ivpu_file_priv *file_priv; + u32 last_ssid = -1; u32 *event; u32 ssid; ivpu_dbg(vdev, IRQ, "MMU event queue\n"); - while ((event = ivpu_mmu_get_event(vdev)) != NULL) { - ivpu_mmu_dump_event(vdev, event); - + while ((event = ivpu_mmu_get_event(vdev))) { ssid = FIELD_GET(IVPU_MMU_EVT_SSID_MASK, event[0]); + + if (ssid == last_ssid) + continue; + + xa_lock(&vdev->context_xa); + file_priv = xa_load(&vdev->context_xa, ssid); + if (file_priv) { + if (file_priv->has_mmu_faults) { + event = NULL; + } else { + ivpu_mmu_disable_events(vdev, ssid); + file_priv->has_mmu_faults = true; + } + } + xa_unlock(&vdev->context_xa); + + if (event) + ivpu_mmu_dump_event(vdev, event); + if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) { ivpu_pm_trigger_recovery(vdev, "MMU event"); return; } - - ivpu_mmu_user_context_mark_invalid(vdev, ssid); REGV_WR32(IVPU_MMU_REG_EVTQ_CONS_SEC, vdev->mmu->evtq.cons); } diff --git a/drivers/accel/ivpu/ivpu_mmu_context.c b/drivers/accel/ivpu/ivpu_mmu_context.c index 891967a95bc3..d373443bbc83 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.c +++ b/drivers/accel/ivpu/ivpu_mmu_context.c @@ -631,16 +631,3 @@ void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev) ivpu_mmu_cd_clear(vdev, vdev->rctx.id); ivpu_mmu_context_fini(vdev, &vdev->rctx); } - -void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid) -{ - struct ivpu_file_priv *file_priv; - - xa_lock(&vdev->context_xa); - - file_priv = xa_load(&vdev->context_xa, ssid); - if (file_priv) - file_priv->has_mmu_faults = true; - - xa_unlock(&vdev->context_xa); -} diff --git a/drivers/accel/ivpu/ivpu_mmu_context.h b/drivers/accel/ivpu/ivpu_mmu_context.h index 8042fc067062..f255310968cf 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.h +++ b/drivers/accel/ivpu/ivpu_mmu_context.h @@ -37,8 +37,6 @@ void ivpu_mmu_global_context_fini(struct ivpu_device *vdev); int ivpu_mmu_reserved_context_init(struct ivpu_device *vdev); void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev); -void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid); - int ivpu_mmu_context_insert_node(struct ivpu_mmu_context *ctx, const struct ivpu_addr_range *range, u64 size, struct drm_mm_node *node); void ivpu_mmu_context_remove_node(struct ivpu_mmu_context *ctx, struct drm_mm_node *node); From 4480912f3f8b8a1fbb5ae12c5c547fd094ec4197 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:30 +0100 Subject: [PATCH 0027/1443] accel/ivpu: Move parts of MMU event IRQ handling to thread handler To prevent looping infinitely in MMU event handler we stop generating new events by removing 'R' (record) bit from context descriptor, but to ensure this change has effect KMD has to perform configuration invalidation followed by sync command. Because of that move parts of the interrupt handler that can take longer to a thread not to block in interrupt handler for too long. This includes: * disabling event queue for the time KMD updates MMU event queue consumer to ensure proper synchronization between MMU and KMD * removal of 'R' (record) bit from context descriptor to ensure no more faults are recorded until that context is destroyed Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-8-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 7 ++- drivers/accel/ivpu/ivpu_mmu.c | 95 +++++++++++++++++++++++------------ drivers/accel/ivpu/ivpu_mmu.h | 2 + 3 files changed, 70 insertions(+), 34 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index c678dcddb8d8..c55de9736d84 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -17,6 +17,7 @@ #include "ivpu_ipc.h" #include "ivpu_job.h" #include "ivpu_jsm_msg.h" +#include "ivpu_mmu.h" #include "ivpu_pm.h" #include "ivpu_trace.h" #include "vpu_boot_api.h" @@ -366,6 +367,7 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) unsigned long cmdq_id; lockdep_assert_held(&file_priv->lock); + ivpu_dbg(vdev, JOB, "Context ID: %u abort\n", file_priv->ctx.id); xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) ivpu_cmdq_unregister(file_priv, cmdq); @@ -373,6 +375,9 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) ivpu_jsm_context_release(vdev, file_priv->ctx.id); + ivpu_mmu_disable_ssid_events(vdev, file_priv->ctx.id); + ivpu_mmu_discard_events(vdev); + file_priv->aborted = true; } @@ -939,8 +944,8 @@ void ivpu_context_abort_work_fn(struct work_struct *work) { struct ivpu_device *vdev = container_of(work, struct ivpu_device, context_abort_work); struct ivpu_file_priv *file_priv; - unsigned long ctx_id; struct ivpu_job *job; + unsigned long ctx_id; unsigned long id; mutex_lock(&vdev->context_list_lock); diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index 5ee4df892b3e..ae1dcd04051c 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -20,6 +20,12 @@ #define IVPU_MMU_REG_CR0 0x00200020u #define IVPU_MMU_REG_CR0ACK 0x00200024u #define IVPU_MMU_REG_CR0ACK_VAL_MASK GENMASK(31, 0) +#define IVPU_MMU_REG_CR0_ATSCHK_MASK BIT(4) +#define IVPU_MMU_REG_CR0_CMDQEN_MASK BIT(3) +#define IVPU_MMU_REG_CR0_EVTQEN_MASK BIT(2) +#define IVPU_MMU_REG_CR0_PRIQEN_MASK BIT(1) +#define IVPU_MMU_REG_CR0_SMMUEN_MASK BIT(0) + #define IVPU_MMU_REG_CR1 0x00200028u #define IVPU_MMU_REG_CR2 0x0020002cu #define IVPU_MMU_REG_IRQ_CTRL 0x00200050u @@ -141,12 +147,6 @@ #define IVPU_MMU_IRQ_EVTQ_EN BIT(2) #define IVPU_MMU_IRQ_GERROR_EN BIT(0) -#define IVPU_MMU_CR0_ATSCHK BIT(4) -#define IVPU_MMU_CR0_CMDQEN BIT(3) -#define IVPU_MMU_CR0_EVTQEN BIT(2) -#define IVPU_MMU_CR0_PRIQEN BIT(1) -#define IVPU_MMU_CR0_SMMUEN BIT(0) - #define IVPU_MMU_CR1_TABLE_SH GENMASK(11, 10) #define IVPU_MMU_CR1_TABLE_OC GENMASK(9, 8) #define IVPU_MMU_CR1_TABLE_IC GENMASK(7, 6) @@ -596,7 +596,7 @@ static int ivpu_mmu_reset(struct ivpu_device *vdev) REGV_WR32(IVPU_MMU_REG_CMDQ_PROD, 0); REGV_WR32(IVPU_MMU_REG_CMDQ_CONS, 0); - val = IVPU_MMU_CR0_CMDQEN; + val = REG_SET_FLD(IVPU_MMU_REG_CR0, CMDQEN, 0); ret = ivpu_mmu_reg_write_cr0(vdev, val); if (ret) return ret; @@ -617,12 +617,12 @@ static int ivpu_mmu_reset(struct ivpu_device *vdev) REGV_WR32(IVPU_MMU_REG_EVTQ_PROD_SEC, 0); REGV_WR32(IVPU_MMU_REG_EVTQ_CONS_SEC, 0); - val |= IVPU_MMU_CR0_EVTQEN; + val = REG_SET_FLD(IVPU_MMU_REG_CR0, EVTQEN, val); ret = ivpu_mmu_reg_write_cr0(vdev, val); if (ret) return ret; - val |= IVPU_MMU_CR0_ATSCHK; + val = REG_SET_FLD(IVPU_MMU_REG_CR0, ATSCHK, val); ret = ivpu_mmu_reg_write_cr0(vdev, val); if (ret) return ret; @@ -631,7 +631,7 @@ static int ivpu_mmu_reset(struct ivpu_device *vdev) if (ret) return ret; - val |= IVPU_MMU_CR0_SMMUEN; + val = REG_SET_FLD(IVPU_MMU_REG_CR0, SMMUEN, val); return ivpu_mmu_reg_write_cr0(vdev, val); } @@ -870,7 +870,47 @@ static u32 *ivpu_mmu_get_event(struct ivpu_device *vdev) return evt; } -static int ivpu_mmu_disable_events(struct ivpu_device *vdev, u32 ssid) +static int ivpu_mmu_evtq_set(struct ivpu_device *vdev, bool enable) +{ + u32 val = REGV_RD32(IVPU_MMU_REG_CR0); + + if (enable) + val = REG_SET_FLD(IVPU_MMU_REG_CR0, EVTQEN, val); + else + val = REG_CLR_FLD(IVPU_MMU_REG_CR0, EVTQEN, val); + REGV_WR32(IVPU_MMU_REG_CR0, val); + + return REGV_POLL_FLD(IVPU_MMU_REG_CR0ACK, VAL, val, IVPU_MMU_REG_TIMEOUT_US); +} + +static int ivpu_mmu_evtq_enable(struct ivpu_device *vdev) +{ + return ivpu_mmu_evtq_set(vdev, true); +} + +static int ivpu_mmu_evtq_disable(struct ivpu_device *vdev) +{ + return ivpu_mmu_evtq_set(vdev, false); +} + +void ivpu_mmu_discard_events(struct ivpu_device *vdev) +{ + /* + * Disable event queue (stop MMU from updating the producer) + * to allow synchronization of consumer and producer indexes + */ + ivpu_mmu_evtq_disable(vdev); + + vdev->mmu->evtq.cons = REGV_RD32(IVPU_MMU_REG_EVTQ_PROD_SEC); + REGV_WR32(IVPU_MMU_REG_EVTQ_CONS_SEC, vdev->mmu->evtq.cons); + vdev->mmu->evtq.prod = REGV_RD32(IVPU_MMU_REG_EVTQ_PROD_SEC); + + ivpu_mmu_evtq_enable(vdev); + + drm_WARN_ON_ONCE(&vdev->drm, vdev->mmu->evtq.cons != vdev->mmu->evtq.prod); +} + +int ivpu_mmu_disable_ssid_events(struct ivpu_device *vdev, u32 ssid) { struct ivpu_mmu_info *mmu = vdev->mmu; struct ivpu_mmu_cdtab *cdtab = &mmu->cdtab; @@ -890,6 +930,7 @@ static int ivpu_mmu_disable_events(struct ivpu_device *vdev, u32 ssid) clflush_cache_range(entry, IVPU_MMU_CDTAB_ENT_SIZE); ivpu_mmu_cmdq_write_cfgi_all(vdev); + ivpu_mmu_cmdq_sync(vdev); return 0; } @@ -897,38 +938,26 @@ static int ivpu_mmu_disable_events(struct ivpu_device *vdev, u32 ssid) void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev) { struct ivpu_file_priv *file_priv; - u32 last_ssid = -1; u32 *event; u32 ssid; ivpu_dbg(vdev, IRQ, "MMU event queue\n"); while ((event = ivpu_mmu_get_event(vdev))) { - ssid = FIELD_GET(IVPU_MMU_EVT_SSID_MASK, event[0]); - - if (ssid == last_ssid) - continue; - - xa_lock(&vdev->context_xa); - file_priv = xa_load(&vdev->context_xa, ssid); - if (file_priv) { - if (file_priv->has_mmu_faults) { - event = NULL; - } else { - ivpu_mmu_disable_events(vdev, ssid); - file_priv->has_mmu_faults = true; - } - } - xa_unlock(&vdev->context_xa); - - if (event) - ivpu_mmu_dump_event(vdev, event); - + ssid = FIELD_GET(IVPU_MMU_EVT_SSID_MASK, *event); if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) { + ivpu_mmu_dump_event(vdev, event); ivpu_pm_trigger_recovery(vdev, "MMU event"); return; } - REGV_WR32(IVPU_MMU_REG_EVTQ_CONS_SEC, vdev->mmu->evtq.cons); + + file_priv = xa_load(&vdev->context_xa, ssid); + if (file_priv) { + if (!READ_ONCE(file_priv->has_mmu_faults)) { + ivpu_mmu_dump_event(vdev, event); + WRITE_ONCE(file_priv->has_mmu_faults, true); + } + } } queue_work(system_wq, &vdev->context_abort_work); diff --git a/drivers/accel/ivpu/ivpu_mmu.h b/drivers/accel/ivpu/ivpu_mmu.h index 7afea9cd8731..1ce7529746ad 100644 --- a/drivers/accel/ivpu/ivpu_mmu.h +++ b/drivers/accel/ivpu/ivpu_mmu.h @@ -47,5 +47,7 @@ int ivpu_mmu_invalidate_tlb(struct ivpu_device *vdev, u16 ssid); void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev); void ivpu_mmu_irq_gerr_handler(struct ivpu_device *vdev); void ivpu_mmu_evtq_dump(struct ivpu_device *vdev); +void ivpu_mmu_discard_events(struct ivpu_device *vdev); +int ivpu_mmu_disable_ssid_events(struct ivpu_device *vdev, u32 ssid); #endif /* __IVPU_MMU_H__ */ From 353b8f48390d36b39276ff6af61464ec64cd4d5c Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:31 +0100 Subject: [PATCH 0028/1443] accel/ivpu: Fix missing MMU events from reserved SSID Generate recovery when fault from reserved context is detected. Add Abort (A) bit to reserved (1) SSID to ensure NPU also receives a fault. There is no way to create a file_priv with reserved SSID but it is still possible to receive MMU faults from that SSID as it is a default NPU HW setting. Such situation will occur if FW freed context related resources but still performed access to DRAM. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-9-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_mmu.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index ae1dcd04051c..b80bdded9fd7 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -725,8 +725,8 @@ static int ivpu_mmu_cdtab_entry_set(struct ivpu_device *vdev, u32 ssid, u64 cd_d cd[2] = 0; cd[3] = 0x0000000000007444; - /* For global context generate memory fault on VPU */ - if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) + /* For global and reserved contexts generate memory fault on VPU */ + if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID || ssid == IVPU_RESERVED_CONTEXT_MMU_SSID) cd[0] |= IVPU_MMU_CD_0_A; if (valid) @@ -945,7 +945,8 @@ void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev) while ((event = ivpu_mmu_get_event(vdev))) { ssid = FIELD_GET(IVPU_MMU_EVT_SSID_MASK, *event); - if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) { + if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID || + ssid == IVPU_RESERVED_CONTEXT_MMU_SSID) { ivpu_mmu_dump_event(vdev, event); ivpu_pm_trigger_recovery(vdev, "MMU event"); return; From ae06e0b3bfe07321c6a865df1b701e64194ec66e Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:32 +0100 Subject: [PATCH 0029/1443] accel/ivpu: Set command queue management capability based on HWS Control explicit command queue management capability bit based on scheduling mode. Capability will be available only when hardware scheduling mode is set. There is no point of allowing user space to create and destroy command queues with OS schedling mode because FW does not support all required functionalities for correct command queue management with OS scheduling. Return -ENODEV from command queue create/destroy/submit IOCTLs. Remove is_valid field from struct ivpu_job_cmdq Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-10-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 19 +++++++------------ drivers/accel/ivpu/ivpu_drv.h | 1 + drivers/accel/ivpu/ivpu_job.c | 24 +++++++++++------------- drivers/accel/ivpu/ivpu_job.h | 1 - 4 files changed, 19 insertions(+), 26 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index 9b0d99873fb3..6a80d626d609 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -127,23 +127,18 @@ void ivpu_file_priv_put(struct ivpu_file_priv **link) kref_put(&file_priv->ref, file_priv_release); } -static int ivpu_get_capabilities(struct ivpu_device *vdev, struct drm_ivpu_param *args) +bool ivpu_is_capable(struct ivpu_device *vdev, u32 capability) { - switch (args->index) { + switch (capability) { case DRM_IVPU_CAP_METRIC_STREAMER: - args->value = 1; - break; + return true; case DRM_IVPU_CAP_DMA_MEMORY_RANGE: - args->value = 1; - break; + return true; case DRM_IVPU_CAP_MANAGE_CMDQ: - args->value = 1; - break; + return vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW; default: - return -EINVAL; + return false; } - - return 0; } static int ivpu_get_param_ioctl(struct drm_device *dev, void *data, struct drm_file *file) @@ -203,7 +198,7 @@ static int ivpu_get_param_ioctl(struct drm_device *dev, void *data, struct drm_f args->value = vdev->hw->sku; break; case DRM_IVPU_PARAM_CAPABILITIES: - ret = ivpu_get_capabilities(vdev, args); + args->value = ivpu_is_capable(vdev, args->index); break; default: ret = -EINVAL; diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index b57d878f2fcd..d53902b34070 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -213,6 +213,7 @@ void ivpu_file_priv_put(struct ivpu_file_priv **link); int ivpu_boot(struct ivpu_device *vdev); int ivpu_shutdown(struct ivpu_device *vdev); void ivpu_prepare_for_reset(struct ivpu_device *vdev); +bool ivpu_is_capable(struct ivpu_device *vdev, u32 capability); static inline u8 ivpu_revision(struct ivpu_device *vdev) { diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index c55de9736d84..37ea92eb4b25 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -123,7 +123,6 @@ static struct ivpu_cmdq *ivpu_cmdq_create(struct ivpu_file_priv *file_priv, u8 p cmdq->priority = priority; cmdq->is_legacy = is_legacy; - cmdq->is_valid = true; ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, &file_priv->cmdq_id_next, GFP_KERNEL); @@ -307,7 +306,7 @@ static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u32 lockdep_assert_held(&file_priv->lock); cmdq = xa_load(&file_priv->cmdq_xa, cmdq_id); - if (!cmdq || !cmdq->is_valid) { + if (!cmdq) { ivpu_warn_ratelimited(vdev, "Failed to find command queue with ID: %u\n", cmdq_id); return NULL; } @@ -832,6 +831,9 @@ int ivpu_cmdq_submit_ioctl(struct drm_device *dev, void *data, struct drm_file * struct ivpu_file_priv *file_priv = file->driver_priv; struct drm_ivpu_cmdq_submit *args = data; + if (!ivpu_is_capable(file_priv->vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) + return -ENODEV; + if (args->cmdq_id < IVPU_CMDQ_MIN_ID || args->cmdq_id > IVPU_CMDQ_MAX_ID) return -EINVAL; @@ -857,6 +859,9 @@ int ivpu_cmdq_create_ioctl(struct drm_device *dev, void *data, struct drm_file * struct drm_ivpu_cmdq_create *args = data; struct ivpu_cmdq *cmdq; + if (!ivpu_is_capable(file_priv->vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) + return -ENODEV; + if (args->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) return -EINVAL; @@ -880,24 +885,17 @@ int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file u32 cmdq_id; int ret = 0; + if (!ivpu_is_capable(vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) + return -ENODEV; + mutex_lock(&file_priv->lock); cmdq = xa_load(&file_priv->cmdq_xa, args->cmdq_id); - if (!cmdq || !cmdq->is_valid || cmdq->is_legacy) { + if (!cmdq || cmdq->is_legacy) { ret = -ENOENT; goto unlock; } - /* - * There is no way to stop executing jobs per command queue - * in OS scheduling mode, mark command queue as invalid instead - * and it will be freed together with context release. - */ - if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) { - cmdq->is_valid = false; - goto unlock; - } - cmdq_id = cmdq->id; ivpu_cmdq_destroy(file_priv, cmdq); ivpu_cmdq_abort_all_jobs(vdev, file_priv->ctx.id, cmdq_id); diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index ff77ee1fcee2..2e301c2eea7b 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -31,7 +31,6 @@ struct ivpu_cmdq { u32 id; u32 db_id; u8 priority; - bool is_valid; bool is_legacy; }; From e52443608934952fc978234cf7d639d6aa3f1856 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:33 +0100 Subject: [PATCH 0030/1443] accel/ivpu: Fix locking order in ivpu_cmdq_destroy_ioctl Fix deadlock caused by inversed locking order in ivpu_job_submit() and ivpu_cmdq_destroy_ioctl(). Both functions operate locking file_priv->lock and submitted_jobs_lock. Unlock file_priv->lock in ivpu_cmdq_destroy_ioctl() before calling ivpu_cmdq_abort_all_jobs() function which locks submitted_jobs_lock. That way locking order is maintained: 1) global submitted_jobs_lock first 2) per context file_priv->lock second Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-11-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 37ea92eb4b25..c694822a14bf 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -883,7 +883,7 @@ int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file struct drm_ivpu_cmdq_destroy *args = data; struct ivpu_cmdq *cmdq; u32 cmdq_id; - int ret = 0; + int ret; if (!ivpu_is_capable(vdev, DRM_IVPU_CAP_MANAGE_CMDQ)) return -ENODEV; @@ -893,13 +893,16 @@ int ivpu_cmdq_destroy_ioctl(struct drm_device *dev, void *data, struct drm_file cmdq = xa_load(&file_priv->cmdq_xa, args->cmdq_id); if (!cmdq || cmdq->is_legacy) { ret = -ENOENT; - goto unlock; + goto err_unlock; } cmdq_id = cmdq->id; ivpu_cmdq_destroy(file_priv, cmdq); + mutex_unlock(&file_priv->lock); ivpu_cmdq_abort_all_jobs(vdev, file_priv->ctx.id, cmdq_id); -unlock: + return 0; + +err_unlock: mutex_unlock(&file_priv->lock); return ret; } From ab680dc6c78aa035e944ecc8c48a1caab9f39924 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:34 +0100 Subject: [PATCH 0031/1443] accel/ivpu: Fix locking order in ivpu_job_submit Fix deadlock in job submission and abort handling. When a thread aborts currently executing jobs due to a fault, it first locks the global lock protecting submitted_jobs (#1). After the last job is destroyed, it proceeds to release the related context and locks file_priv (#2). Meanwhile, in the job submission thread, the file_priv lock (#2) is taken first, and then the submitted_jobs lock (#1) is obtained when a job is added to the submitted jobs list. CPU0 CPU1 ---- ---- (for example due to a fault) (jobs submissions keep coming) lock(&vdev->submitted_jobs_lock) #1 ivpu_jobs_abort_all() job_destroy() lock(&file_priv->lock) #2 lock(&vdev->submitted_jobs_lock) #1 file_priv_release() lock(&vdev->context_list_lock) lock(&file_priv->lock) #2 This order of locking causes a deadlock. To resolve this issue, change the order of locking in ivpu_job_submit(). Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-12-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index c694822a14bf..c93ea37062d7 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -597,6 +597,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) if (ret < 0) return ret; + mutex_lock(&vdev->submitted_jobs_lock); mutex_lock(&file_priv->lock); if (cmdq_id == 0) @@ -606,19 +607,17 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) if (!cmdq) { ivpu_warn_ratelimited(vdev, "Failed to get job queue, ctx %d\n", file_priv->ctx.id); ret = -EINVAL; - goto err_unlock_file_priv; + goto err_unlock; } ret = ivpu_cmdq_register(file_priv, cmdq); if (ret) { ivpu_err(vdev, "Failed to register command queue: %d\n", ret); - goto err_unlock_file_priv; + goto err_unlock; } job->cmdq_id = cmdq->id; - mutex_lock(&vdev->submitted_jobs_lock); - is_first_job = xa_empty(&vdev->submitted_jobs_xa); ret = xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, &file_priv->job_id_next, GFP_KERNEL); @@ -626,7 +625,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) ivpu_dbg(vdev, JOB, "Too many active jobs in ctx %d\n", file_priv->ctx.id); ret = -EBUSY; - goto err_unlock_submitted_jobs; + goto err_unlock; } ret = ivpu_cmdq_push_job(cmdq, job); @@ -649,22 +648,20 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority, u32 cmdq_id) job->job_id, file_priv->ctx.id, job->engine_idx, cmdq->priority, job->cmd_buf_vpu_addr, cmdq->jobq->header.tail); - mutex_unlock(&vdev->submitted_jobs_lock); mutex_unlock(&file_priv->lock); if (unlikely(ivpu_test_mode & IVPU_TEST_MODE_NULL_HW)) { - mutex_lock(&vdev->submitted_jobs_lock); ivpu_job_signal_and_destroy(vdev, job->job_id, VPU_JSM_STATUS_SUCCESS); - mutex_unlock(&vdev->submitted_jobs_lock); } + mutex_unlock(&vdev->submitted_jobs_lock); + return 0; err_erase_xa: xa_erase(&vdev->submitted_jobs_xa, job->job_id); -err_unlock_submitted_jobs: +err_unlock: mutex_unlock(&vdev->submitted_jobs_lock); -err_unlock_file_priv: mutex_unlock(&file_priv->lock); ivpu_rpm_put(vdev); return ret; From dad945c27a42dfadddff1049cf5ae417209a8996 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:35 +0100 Subject: [PATCH 0032/1443] accel/ivpu: Add handling of VPU_JSM_STATUS_MVNCI_CONTEXT_VIOLATION_HW Mark as invalid context of a job that returned HW context violation error and queue work that aborts jobs from faulty context. Add engine reset to the context abort thread handler to not only abort currently executing jobs but also to ensure NPU invalid state recovery. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-13-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index c93ea37062d7..3c162ac41a1d 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -533,6 +533,26 @@ static int ivpu_job_signal_and_destroy(struct ivpu_device *vdev, u32 job_id, u32 lockdep_assert_held(&vdev->submitted_jobs_lock); + job = xa_load(&vdev->submitted_jobs_xa, job_id); + if (!job) + return -ENOENT; + + if (job_status == VPU_JSM_STATUS_MVNCI_CONTEXT_VIOLATION_HW) { + guard(mutex)(&job->file_priv->lock); + + if (job->file_priv->has_mmu_faults) + return 0; + + /* + * Mark context as faulty and defer destruction of the job to jobs abort thread + * handler to synchronize between both faults and jobs returning context violation + * status and ensure both are handled in the same way + */ + job->file_priv->has_mmu_faults = true; + queue_work(system_wq, &vdev->context_abort_work); + return 0; + } + job = ivpu_job_remove_from_submitted_jobs(vdev, job_id); if (!job) return -ENOENT; @@ -946,6 +966,9 @@ void ivpu_context_abort_work_fn(struct work_struct *work) unsigned long ctx_id; unsigned long id; + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) + ivpu_jsm_reset_engine(vdev, 0); + mutex_lock(&vdev->context_list_lock); xa_for_each(&vdev->context_xa, ctx_id, file_priv) { if (!file_priv->has_mmu_faults || file_priv->aborted) @@ -959,6 +982,8 @@ void ivpu_context_abort_work_fn(struct work_struct *work) if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) return; + + ivpu_jsm_hws_resume_engine(vdev, 0); /* * In hardware scheduling mode NPU already has stopped processing jobs * and won't send us any further notifications, thus we have to free job related resources From dd4f78ec6a6f82d02e59d6cadde6b92ed0507a4d Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 7 Jan 2025 18:32:36 +0100 Subject: [PATCH 0033/1443] accel/ivpu: Add platform detection for presilicon Use highest buttress VPU_STATUS register bits(15:13) that encode platform type as follows: 0 - Silicon 2 - Simics 3 - FPGA 4 - Hybrid SLE Remove old DMI based method. Signed-off-by: Karol Wachowski Signed-off-by: Maciej Falkowski Reviewed-by: Jacek Lawrynowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-14-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.h | 4 ++- drivers/accel/ivpu/ivpu_hw.c | 41 ++++++++++------------- drivers/accel/ivpu/ivpu_hw_btrs.c | 7 ++++ drivers/accel/ivpu/ivpu_hw_btrs.h | 1 + drivers/accel/ivpu/ivpu_hw_btrs_lnl_reg.h | 1 + 5 files changed, 29 insertions(+), 25 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index d53902b34070..ca21102ca366 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -58,6 +58,7 @@ #define IVPU_PLATFORM_SILICON 0 #define IVPU_PLATFORM_SIMICS 2 #define IVPU_PLATFORM_FPGA 3 +#define IVPU_PLATFORM_HSLE 4 #define IVPU_PLATFORM_INVALID 8 #define IVPU_SCHED_MODE_AUTO -1 @@ -288,7 +289,8 @@ static inline bool ivpu_is_simics(struct ivpu_device *vdev) static inline bool ivpu_is_fpga(struct ivpu_device *vdev) { - return ivpu_get_platform(vdev) == IVPU_PLATFORM_FPGA; + return ivpu_get_platform(vdev) == IVPU_PLATFORM_FPGA || + ivpu_get_platform(vdev) == IVPU_PLATFORM_HSLE; } static inline bool ivpu_is_force_snoop_enabled(struct ivpu_device *vdev) diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 1b691375ee4d..e332f19ab51d 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -19,38 +19,31 @@ static char *platform_to_str(u32 platform) return "SIMICS"; case IVPU_PLATFORM_FPGA: return "FPGA"; + case IVPU_PLATFORM_HSLE: + return "HSLE"; default: return "Invalid platform"; } } -static const struct dmi_system_id dmi_platform_simulation[] = { - { - .ident = "Intel Simics", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "lnlrvp"), - DMI_MATCH(DMI_BOARD_VERSION, "1.0"), - DMI_MATCH(DMI_BOARD_SERIAL, "123456789"), - }, - }, - { - .ident = "Intel Simics", - .matches = { - DMI_MATCH(DMI_BOARD_NAME, "Simics"), - }, - }, - { } -}; - static void platform_init(struct ivpu_device *vdev) { - if (dmi_check_system(dmi_platform_simulation)) - vdev->platform = IVPU_PLATFORM_SIMICS; - else - vdev->platform = IVPU_PLATFORM_SILICON; + int platform = ivpu_hw_btrs_platform_read(vdev); - ivpu_dbg(vdev, MISC, "Platform type: %s (%d)\n", - platform_to_str(vdev->platform), vdev->platform); + ivpu_dbg(vdev, MISC, "Platform type: %s (%d)\n", platform_to_str(platform), platform); + + switch (platform) { + case IVPU_PLATFORM_SILICON: + case IVPU_PLATFORM_SIMICS: + case IVPU_PLATFORM_FPGA: + case IVPU_PLATFORM_HSLE: + vdev->platform = platform; + break; + + default: + ivpu_err(vdev, "Invalid platform type: %d\n", platform); + break; + } } static void wa_init(struct ivpu_device *vdev) diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c index 3753b00ed2d6..56c56012b980 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.c +++ b/drivers/accel/ivpu/ivpu_hw_btrs.c @@ -887,3 +887,10 @@ void ivpu_hw_btrs_diagnose_failure(struct ivpu_device *vdev) else return diagnose_failure_lnl(vdev); } + +int ivpu_hw_btrs_platform_read(struct ivpu_device *vdev) +{ + u32 reg = REGB_RD32(VPU_HW_BTRS_LNL_VPU_STATUS); + + return REG_GET_FLD(VPU_HW_BTRS_LNL_VPU_STATUS, PLATFORM, reg); +} diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.h b/drivers/accel/ivpu/ivpu_hw_btrs.h index 04f14f50fed6..1fd71b4d4ab0 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.h +++ b/drivers/accel/ivpu/ivpu_hw_btrs.h @@ -46,5 +46,6 @@ void ivpu_hw_btrs_global_int_disable(struct ivpu_device *vdev); void ivpu_hw_btrs_irq_enable(struct ivpu_device *vdev); void ivpu_hw_btrs_irq_disable(struct ivpu_device *vdev); void ivpu_hw_btrs_diagnose_failure(struct ivpu_device *vdev); +int ivpu_hw_btrs_platform_read(struct ivpu_device *vdev); #endif /* __IVPU_HW_BTRS_H__ */ diff --git a/drivers/accel/ivpu/ivpu_hw_btrs_lnl_reg.h b/drivers/accel/ivpu/ivpu_hw_btrs_lnl_reg.h index fc51f3098f97..fff2ef2cada6 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs_lnl_reg.h +++ b/drivers/accel/ivpu/ivpu_hw_btrs_lnl_reg.h @@ -86,6 +86,7 @@ #define VPU_HW_BTRS_LNL_VPU_STATUS_POWER_RESOURCE_OWN_ACK_MASK BIT_MASK(7) #define VPU_HW_BTRS_LNL_VPU_STATUS_PERF_CLK_MASK BIT_MASK(11) #define VPU_HW_BTRS_LNL_VPU_STATUS_DISABLE_CLK_RELINQUISH_MASK BIT_MASK(12) +#define VPU_HW_BTRS_LNL_VPU_STATUS_PLATFORM_MASK GENMASK(31, 29) #define VPU_HW_BTRS_LNL_IP_RESET 0x00000160u #define VPU_HW_BTRS_LNL_IP_RESET_TRIGGER_MASK BIT_MASK(0) From 74509d54ebf1ecfbdf5f7edec32c490fefa01b8b Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Tue, 7 Jan 2025 18:32:37 +0100 Subject: [PATCH 0034/1443] accel/ivpu: Enable HWS by default on all platforms Enable HWS on selected platforms if FW API version is above 3.19. Signed-off-by: Maciej Falkowski Reviewed-by: Karol Wachowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250107173238.381120-15-maciej.falkowski@linux.intel.com --- drivers/accel/ivpu/ivpu_fw.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index 6037ec0b3096..6cf1fb826d1b 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -145,7 +145,18 @@ ivpu_fw_sched_mode_select(struct ivpu_device *vdev, const struct vpu_firmware_he if (ivpu_sched_mode != IVPU_SCHED_MODE_AUTO) return ivpu_sched_mode; - return VPU_SCHEDULING_MODE_OS; + if (IVPU_FW_CHECK_API_VER_LT(vdev, fw_hdr, JSM, 3, 24)) + return VPU_SCHEDULING_MODE_OS; + + switch (ivpu_device_id(vdev)) { + case PCI_DEVICE_ID_MTL: + case PCI_DEVICE_ID_ARL: + case PCI_DEVICE_ID_LNL: + case PCI_DEVICE_ID_PTL_P: + return VPU_SCHEDULING_MODE_HW; + default: + return VPU_SCHEDULING_MODE_OS; + } } static int ivpu_fw_parse(struct ivpu_device *vdev) From f7dfd3db3e0459765176124b4b7e4b4b93533676 Mon Sep 17 00:00:00 2001 From: Vivek Kasireddy Date: Wed, 11 Dec 2024 21:54:21 -0800 Subject: [PATCH 0035/1443] drm/virtio: Fix UAF in virtgpu_dma_buf_free_obj() Fix the following issues identified by Smatch static checker: - The call to dma_buf_put(attach->dmabuf) after dma_buf_detach() leads to a UAF bug as dma_buf_detach() frees the attach object. Fix this by extracting the dmabuf object from attach and using that in the call to dma_buf_put(). - The resv object is extracted from attach before checking to see if attach is valid (that is !NULL) or not. Although, attach would very likely be valid, fix this by making sure that the resv object is used only after ensuring that attach is valid. Fixes: 2885e575abc7 ("drm/virtio: Add helpers to initialize and free the imported object") Fixes: ca77f27a2665 ("drm/virtio: Import prime buffers from other devices as guest blobs") Cc: Gerd Hoffmann Cc: Dmitry Osipenko Cc: Gurchetan Singh Cc: Chia-I Wu Reported-by: Dan Carpenter Signed-off-by: Vivek Kasireddy Link: https://patchwork.freedesktop.org/patch/msgid/20241212055421.775759-1-vivek.kasireddy@intel.com Reviewed-by: Dmitry Osipenko Tested-by: Dmitry Osipenko Signed-off-by: Dmitry Osipenko [dmitry.osipenko@collabora.com: Edited commit title] --- drivers/gpu/drm/virtio/virtgpu_prime.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c index b3664c12843d..f92133a01195 100644 --- a/drivers/gpu/drm/virtio/virtgpu_prime.c +++ b/drivers/gpu/drm/virtio/virtgpu_prime.c @@ -189,10 +189,11 @@ static void virtgpu_dma_buf_free_obj(struct drm_gem_object *obj) struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); struct virtio_gpu_device *vgdev = obj->dev->dev_private; struct dma_buf_attachment *attach = obj->import_attach; - struct dma_resv *resv = attach->dmabuf->resv; if (attach) { - dma_resv_lock(resv, NULL); + struct dma_buf *dmabuf = attach->dmabuf; + + dma_resv_lock(dmabuf->resv, NULL); virtio_gpu_detach_object_fenced(bo); @@ -200,10 +201,10 @@ static void virtgpu_dma_buf_free_obj(struct drm_gem_object *obj) dma_buf_unmap_attachment(attach, bo->sgt, DMA_BIDIRECTIONAL); - dma_resv_unlock(resv); + dma_resv_unlock(dmabuf->resv); - dma_buf_detach(attach->dmabuf, attach); - dma_buf_put(attach->dmabuf); + dma_buf_detach(dmabuf, attach); + dma_buf_put(dmabuf); } if (bo->created) { From 5dd8b536bbdaee00df1e8dbd8dc4e9fc2f7fadcb Mon Sep 17 00:00:00 2001 From: Vivek Kasireddy Date: Tue, 10 Dec 2024 22:43:43 -0800 Subject: [PATCH 0036/1443] drm/virtio: Lock the VGA resources during initialization If another driver for a VGA compatible GPU (that is passthrough'd) locks the VGA resources (by calling vga_get()), then virtio_gpu driver would encounter the following errors and fail to load during probe and initialization: Invalid read at addr 0x7200005014, size 1, region '(null)', reason: rejected Invalid write at addr 0x7200005014, size 1, region '(null)', reason: rejected virtio_gpu virtio0: virtio: device uses modern interface but does not have VIRTIO_F_VERSION_1 virtio_gpu virtio0: probe with driver virtio_gpu failed with error -22 This issue is only seen if virtio-gpu and the other GPU are on different PCI buses, which can happen if the user includes an additional PCIe port and associates the VGA compatible GPU with it while launching Qemu: qemu-system-x86_64... -device virtio-vga,max_outputs=1,xres=1920,yres=1080,blob=true -device pcie-root-port,id=pcie.1,bus=pcie.0,addr=1c.0,slot=1,chassis=1,multifunction=on -device vfio-pci,host=03:00.0,bus=pcie.1,addr=00.0 ... In the above example, the device 03:00.0 is an Intel DG2 card and this issue is seen when both i915 driver and virtio_gpu driver are loading (or initializing) concurrently or when i915 is loaded first. Note that during initalization, i915 driver does the following in intel_vga_reset_io_mem(): vga_get_uninterruptible(pdev, VGA_RSRC_LEGACY_IO); outb(inb(VGA_MIS_R), VGA_MIS_W); vga_put(pdev, VGA_RSRC_LEGACY_IO); Although, virtio-gpu might own the VGA resources initially, the above call (in i915) to vga_get_uninterruptible() would result in these resources being taken away, which means that virtio-gpu would not be able to decode VGA anymore. This happens in __vga_tryget() when it calls pci_set_vga_state(conflict->pdev, false, pci_bits, flags); where pci_bits = PCI_COMMAND_MEMORY | PCI_COMMAND_IO flags = PCI_VGA_STATE_CHANGE_DECODES | PCI_VGA_STATE_CHANGE_BRIDGE Therefore, to solve this issue, virtio-gpu driver needs to call vga_get() whenever it needs to reclaim and access VGA resources, which is during initial probe and setup. After that, a call to vga_put() would release the lock to allow other VGA compatible devices to access these shared VGA resources. Cc: Gerd Hoffmann Cc: Gurchetan Singh Cc: Chia-I Wu Reported-by: Dmitry Osipenko Signed-off-by: Vivek Kasireddy Link: https://patchwork.freedesktop.org/patch/msgid/20241211064343.550153-1-vivek.kasireddy@intel.com Tested-by: Dmitry Osipenko Signed-off-by: Dmitry Osipenko --- drivers/gpu/drm/virtio/virtgpu_drv.c | 40 +++++++++++++++++++++++++++- 1 file changed, 39 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c index 6a67c6297d58..d4309dba557b 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.c +++ b/drivers/gpu/drm/virtio/virtgpu_drv.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -41,6 +42,8 @@ #include "virtgpu_drv.h" +#define PCI_DEVICE_ID_VIRTIO_GPU 0x1050 + static const struct drm_driver driver; static int virtio_gpu_modeset = -1; @@ -162,7 +165,42 @@ static struct virtio_driver virtio_gpu_driver = { .config_changed = virtio_gpu_config_changed }; -module_virtio_driver(virtio_gpu_driver); +static int __init virtio_gpu_driver_init(void) +{ + struct pci_dev *pdev; + int ret; + + pdev = pci_get_device(PCI_VENDOR_ID_REDHAT_QUMRANET, + PCI_DEVICE_ID_VIRTIO_GPU, + NULL); + if (!pdev) + return -ENODEV; + + if (pci_is_vga(pdev)) { + ret = vga_get_interruptible(pdev, + VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM); + if (ret) + goto error; + } + + ret = register_virtio_driver(&virtio_gpu_driver); + + if (pci_is_vga(pdev)) + vga_put(pdev, VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM); + +error: + pci_dev_put(pdev); + + return ret; +} + +static void __exit virtio_gpu_driver_exit(void) +{ + unregister_virtio_driver(&virtio_gpu_driver); +} + +module_init(virtio_gpu_driver_init); +module_exit(virtio_gpu_driver_exit); MODULE_DEVICE_TABLE(virtio, id_table); MODULE_DESCRIPTION("Virtio GPU driver"); From ffda6454267d0b870f3a09945a7ce88137b914a6 Mon Sep 17 00:00:00 2001 From: Dmitry Osipenko Date: Fri, 29 Nov 2024 18:53:57 +0300 Subject: [PATCH 0037/1443] drm/virtio: Set missing bo->attached flag VirtIO-GPU driver now supports detachment of shmem BOs from host, but doing it only for imported dma-bufs. Mark all shmem BOs as attached, not just dma-bufs. This is a minor correction since detachment of a non-dmabuf BOs not supported today. Signed-off-by: Dmitry Osipenko Acked-by: Vivek Kasireddy Link: https://patchwork.freedesktop.org/patch/msgid/20241129155357.2265357-1-dmitry.osipenko@collabora.com --- drivers/gpu/drm/virtio/virtgpu_prime.c | 1 - drivers/gpu/drm/virtio/virtgpu_vq.c | 3 +++ 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c index f92133a01195..483707630861 100644 --- a/drivers/gpu/drm/virtio/virtgpu_prime.c +++ b/drivers/gpu/drm/virtio/virtgpu_prime.c @@ -250,7 +250,6 @@ static int virtgpu_dma_buf_init_obj(struct drm_device *dev, virtio_gpu_cmd_resource_create_blob(vgdev, bo, ¶ms, ents, nents); bo->guest_blob = true; - bo->attached = true; dma_buf_unpin(attach); dma_resv_unlock(resv); diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c index ad91624df42d..062639250a4e 100644 --- a/drivers/gpu/drm/virtio/virtgpu_vq.c +++ b/drivers/gpu/drm/virtio/virtgpu_vq.c @@ -1300,6 +1300,9 @@ virtio_gpu_cmd_resource_create_blob(struct virtio_gpu_device *vgdev, virtio_gpu_queue_ctrl_buffer(vgdev, vbuf); bo->created = true; + + if (nents) + bo->attached = true; } void virtio_gpu_cmd_set_scanout_blob(struct virtio_gpu_device *vgdev, From 2ef1c8c5de40316ba65051940e587157385e10c0 Mon Sep 17 00:00:00 2001 From: Dmitry Osipenko Date: Mon, 2 Dec 2024 08:39:55 +0300 Subject: [PATCH 0038/1443] drm/virtio: Factor out common dmabuf unmapping code Move out dmabuf detachment and unmapping into separate function. This removes duplicated code and there is no need to check the GEM's kref now, since both bo->attached and bo->sgt are unset under held reservation lock. Signed-off-by: Dmitry Osipenko Acked-by: Vivek Kasireddy Link: https://patchwork.freedesktop.org/patch/msgid/20241202053955.2451321-1-dmitry.osipenko@collabora.com --- drivers/gpu/drm/virtio/virtgpu_prime.c | 35 ++++++++++++++------------ 1 file changed, 19 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c index 483707630861..fe6a0b018571 100644 --- a/drivers/gpu/drm/virtio/virtgpu_prime.c +++ b/drivers/gpu/drm/virtio/virtgpu_prime.c @@ -184,6 +184,23 @@ int virtgpu_dma_buf_import_sgt(struct virtio_gpu_mem_entry **ents, return 0; } +static void virtgpu_dma_buf_unmap(struct virtio_gpu_object *bo) +{ + struct dma_buf_attachment *attach = bo->base.base.import_attach; + + dma_resv_assert_held(attach->dmabuf->resv); + + if (bo->created) { + virtio_gpu_detach_object_fenced(bo); + + if (bo->sgt) + dma_buf_unmap_attachment(attach, bo->sgt, + DMA_BIDIRECTIONAL); + + bo->sgt = NULL; + } +} + static void virtgpu_dma_buf_free_obj(struct drm_gem_object *obj) { struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); @@ -194,13 +211,7 @@ static void virtgpu_dma_buf_free_obj(struct drm_gem_object *obj) struct dma_buf *dmabuf = attach->dmabuf; dma_resv_lock(dmabuf->resv, NULL); - - virtio_gpu_detach_object_fenced(bo); - - if (bo->sgt) - dma_buf_unmap_attachment(attach, bo->sgt, - DMA_BIDIRECTIONAL); - + virtgpu_dma_buf_unmap(bo); dma_resv_unlock(dmabuf->resv); dma_buf_detach(dmabuf, attach); @@ -273,15 +284,7 @@ static void virtgpu_dma_buf_move_notify(struct dma_buf_attachment *attach) struct drm_gem_object *obj = attach->importer_priv; struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); - if (bo->created && kref_read(&obj->refcount)) { - virtio_gpu_detach_object_fenced(bo); - - if (bo->sgt) - dma_buf_unmap_attachment(attach, bo->sgt, - DMA_BIDIRECTIONAL); - - bo->sgt = NULL; - } + virtgpu_dma_buf_unmap(bo); } static const struct dma_buf_attach_ops virtgpu_dma_buf_attach_ops = { From 178ada9d6e90637667a7410dd7fe75fbbe2cd4c5 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Fri, 15 Nov 2024 10:21:51 +0000 Subject: [PATCH 0039/1443] dma-fence: Add a single fence fast path for fence merging MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Testing some workloads in two different scenarios, such as games running under Gamescope on a Steam Deck, or vkcube under a Plasma desktop, shows that in a significant portion of calls the dma_fence_unwrap_merge helper is called with just a single unsignalled fence. Therefore it is worthile to add a fast path for that case and so bypass the memory allocation and insertion sort attempts. Tested scenarios: 1) Hogwarts Legacy under Gamescope ~1500 calls per second to __dma_fence_unwrap_merge. Percentages per number of fences buckets, before and after checking for signalled status, sorting and flattening: N Before After 0 0.85% 1 69.80% -> The new fast path. 2-9 29.36% 9% (Ie. 91% of this bucket flattened to 1 fence) 10-19 20-40 50+ 2) Cyberpunk 2077 under Gamescope ~2400 calls per second. N Before After 0 0.71% 1 52.53% -> The new fast path. 2-9 44.38% 50.60% (Ie. half resolved to a single fence) 10-19 2.34% 20-40 0.06% 50+ 3) vkcube under Plasma 90 calls per second. N Before After 0 1 2-9 100% 0% (Ie. all resolved to a single fence) 10-19 20-40 50+ In the case of vkcube all invocations in the 2-9 bucket were actually just two input fences. v2: * Correct local variable name and hold on to unsignaled reference. (Chistian) Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Friedrich Vock Reviewed-by: Christian König Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20241115102153.1980-4-tursulin@igalia.com --- drivers/dma-buf/dma-fence-unwrap.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/dma-buf/dma-fence-unwrap.c b/drivers/dma-buf/dma-fence-unwrap.c index 6345062731f1..2a059ac0ed27 100644 --- a/drivers/dma-buf/dma-fence-unwrap.c +++ b/drivers/dma-buf/dma-fence-unwrap.c @@ -84,8 +84,8 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences, struct dma_fence **fences, struct dma_fence_unwrap *iter) { + struct dma_fence *tmp, *unsignaled = NULL, **array; struct dma_fence_array *result; - struct dma_fence *tmp, **array; ktime_t timestamp; int i, j, count; @@ -94,6 +94,8 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences, for (i = 0; i < num_fences; ++i) { dma_fence_unwrap_for_each(tmp, &iter[i], fences[i]) { if (!dma_fence_is_signaled(tmp)) { + dma_fence_put(unsignaled); + unsignaled = dma_fence_get(tmp); ++count; } else { ktime_t t = dma_fence_timestamp(tmp); @@ -107,9 +109,16 @@ struct dma_fence *__dma_fence_unwrap_merge(unsigned int num_fences, /* * If we couldn't find a pending fence just return a private signaled * fence with the timestamp of the last signaled one. + * + * Or if there was a single unsignaled fence left we can return it + * directly and early since that is a major path on many workloads. */ if (count == 0) return dma_fence_allocate_private_stub(timestamp); + else if (count == 1) + return unsignaled; + + dma_fence_put(unsignaled); array = kmalloc_array(count, sizeof(*array), GFP_KERNEL); if (!array) From b1cce631e61fcf3e1bc77ace05f10d00c737af9a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Fri, 15 Nov 2024 10:21:52 +0000 Subject: [PATCH 0040/1443] dma-buf: add selftest for fence order after merge MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a test which double checks that fences are in the expected order after a merge. While at it also switch to using a mock array for the complex test instead of a merge. Signed-off-by: Christian König Signed-off-by: Tvrtko Ursulin Reviewed-by: Tvrtko Ursulin Link: https://patchwork.freedesktop.org/patch/msgid/20241115102153.1980-5-tursulin@igalia.com --- drivers/dma-buf/st-dma-fence-unwrap.c | 69 ++++++++++++++++++++++++++- 1 file changed, 68 insertions(+), 1 deletion(-) diff --git a/drivers/dma-buf/st-dma-fence-unwrap.c b/drivers/dma-buf/st-dma-fence-unwrap.c index f0cee984b6c7..876eabddb08f 100644 --- a/drivers/dma-buf/st-dma-fence-unwrap.c +++ b/drivers/dma-buf/st-dma-fence-unwrap.c @@ -304,6 +304,72 @@ static int unwrap_merge(void *arg) return err; } +static int unwrap_merge_order(void *arg) +{ + struct dma_fence *fence, *f1, *f2, *a1, *a2, *c1, *c2; + struct dma_fence_unwrap iter; + int err = 0; + + f1 = mock_fence(); + if (!f1) + return -ENOMEM; + + dma_fence_enable_sw_signaling(f1); + + f2 = mock_fence(); + if (!f2) { + dma_fence_put(f1); + return -ENOMEM; + } + + dma_fence_enable_sw_signaling(f2); + + a1 = mock_array(2, f1, f2); + if (!a1) + return -ENOMEM; + + c1 = mock_chain(NULL, dma_fence_get(f1)); + if (!c1) + goto error_put_a1; + + c2 = mock_chain(c1, dma_fence_get(f2)); + if (!c2) + goto error_put_a1; + + /* + * The fences in the chain are the same as in a1 but in oposite order, + * the dma_fence_merge() function should be able to handle that. + */ + a2 = dma_fence_unwrap_merge(a1, c2); + + dma_fence_unwrap_for_each(fence, &iter, a2) { + if (fence == f1) { + f1 = NULL; + if (!f2) + pr_err("Unexpected order!\n"); + } else if (fence == f2) { + f2 = NULL; + if (f1) + pr_err("Unexpected order!\n"); + } else { + pr_err("Unexpected fence!\n"); + err = -EINVAL; + } + } + + if (f1 || f2) { + pr_err("Not all fences seen!\n"); + err = -EINVAL; + } + + dma_fence_put(a2); + return err; + +error_put_a1: + dma_fence_put(a1); + return -ENOMEM; +} + static int unwrap_merge_complex(void *arg) { struct dma_fence *fence, *f1, *f2, *f3, *f4, *f5; @@ -327,7 +393,7 @@ static int unwrap_merge_complex(void *arg) goto error_put_f2; /* The resulting array has the fences in reverse */ - f4 = dma_fence_unwrap_merge(f2, f1); + f4 = mock_array(2, dma_fence_get(f2), dma_fence_get(f1)); if (!f4) goto error_put_f3; @@ -375,6 +441,7 @@ int dma_fence_unwrap(void) SUBTEST(unwrap_chain), SUBTEST(unwrap_chain_array), SUBTEST(unwrap_merge), + SUBTEST(unwrap_merge_order), SUBTEST(unwrap_merge_complex), }; From d2a0b1bc5773d3124ef7bee886fce1bb59fd9b76 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Fri, 15 Nov 2024 10:21:53 +0000 Subject: [PATCH 0041/1443] dma-fence: Add some more fence-merge-unwrap tests MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far all tests use seqno one and only vary the context. Lets add some tests which vary the seqno too. Signed-off-by: Tvrtko Ursulin Reviewed-by: Christian König Signed-off-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20241115102153.1980-6-tursulin@igalia.com --- drivers/dma-buf/st-dma-fence-unwrap.c | 199 +++++++++++++++++++++++++- 1 file changed, 196 insertions(+), 3 deletions(-) diff --git a/drivers/dma-buf/st-dma-fence-unwrap.c b/drivers/dma-buf/st-dma-fence-unwrap.c index 876eabddb08f..a3be888ae2e8 100644 --- a/drivers/dma-buf/st-dma-fence-unwrap.c +++ b/drivers/dma-buf/st-dma-fence-unwrap.c @@ -28,7 +28,7 @@ static const struct dma_fence_ops mock_ops = { .get_timeline_name = mock_name, }; -static struct dma_fence *mock_fence(void) +static struct dma_fence *__mock_fence(u64 context, u64 seqno) { struct mock_fence *f; @@ -37,12 +37,16 @@ static struct dma_fence *mock_fence(void) return NULL; spin_lock_init(&f->lock); - dma_fence_init(&f->base, &mock_ops, &f->lock, - dma_fence_context_alloc(1), 1); + dma_fence_init(&f->base, &mock_ops, &f->lock, context, seqno); return &f->base; } +static struct dma_fence *mock_fence(void) +{ + return __mock_fence(dma_fence_context_alloc(1), 1); +} + static struct dma_fence *mock_array(unsigned int num_fences, ...) { struct dma_fence_array *array; @@ -304,6 +308,111 @@ static int unwrap_merge(void *arg) return err; } +static int unwrap_merge_duplicate(void *arg) +{ + struct dma_fence *fence, *f1, *f2; + struct dma_fence_unwrap iter; + int err = 0; + + f1 = mock_fence(); + if (!f1) + return -ENOMEM; + + dma_fence_enable_sw_signaling(f1); + + f2 = dma_fence_unwrap_merge(f1, f1); + if (!f2) { + err = -ENOMEM; + goto error_put_f1; + } + + dma_fence_unwrap_for_each(fence, &iter, f2) { + if (fence == f1) { + dma_fence_put(f1); + f1 = NULL; + } else { + pr_err("Unexpected fence!\n"); + err = -EINVAL; + } + } + + if (f1) { + pr_err("Not all fences seen!\n"); + err = -EINVAL; + } + + dma_fence_put(f2); +error_put_f1: + dma_fence_put(f1); + return err; +} + +static int unwrap_merge_seqno(void *arg) +{ + struct dma_fence *fence, *f1, *f2, *f3, *f4; + struct dma_fence_unwrap iter; + int err = 0; + u64 ctx[2]; + + ctx[0] = dma_fence_context_alloc(1); + ctx[1] = dma_fence_context_alloc(1); + + f1 = __mock_fence(ctx[1], 1); + if (!f1) + return -ENOMEM; + + dma_fence_enable_sw_signaling(f1); + + f2 = __mock_fence(ctx[1], 2); + if (!f2) { + err = -ENOMEM; + goto error_put_f1; + } + + dma_fence_enable_sw_signaling(f2); + + f3 = __mock_fence(ctx[0], 1); + if (!f3) { + err = -ENOMEM; + goto error_put_f2; + } + + dma_fence_enable_sw_signaling(f3); + + f4 = dma_fence_unwrap_merge(f1, f2, f3); + if (!f4) { + err = -ENOMEM; + goto error_put_f3; + } + + dma_fence_unwrap_for_each(fence, &iter, f4) { + if (fence == f3 && f2) { + dma_fence_put(f3); + f3 = NULL; + } else if (fence == f2 && !f3) { + dma_fence_put(f2); + f2 = NULL; + } else { + pr_err("Unexpected fence!\n"); + err = -EINVAL; + } + } + + if (f2 || f3) { + pr_err("Not all fences seen!\n"); + err = -EINVAL; + } + + dma_fence_put(f4); +error_put_f3: + dma_fence_put(f3); +error_put_f2: + dma_fence_put(f2); +error_put_f1: + dma_fence_put(f1); + return err; +} + static int unwrap_merge_order(void *arg) { struct dma_fence *fence, *f1, *f2, *a1, *a2, *c1, *c2; @@ -433,6 +542,87 @@ static int unwrap_merge_complex(void *arg) return err; } +static int unwrap_merge_complex_seqno(void *arg) +{ + struct dma_fence *fence, *f1, *f2, *f3, *f4, *f5, *f6, *f7; + struct dma_fence_unwrap iter; + int err = -ENOMEM; + u64 ctx[2]; + + ctx[0] = dma_fence_context_alloc(1); + ctx[1] = dma_fence_context_alloc(1); + + f1 = __mock_fence(ctx[0], 2); + if (!f1) + return -ENOMEM; + + dma_fence_enable_sw_signaling(f1); + + f2 = __mock_fence(ctx[1], 1); + if (!f2) + goto error_put_f1; + + dma_fence_enable_sw_signaling(f2); + + f3 = __mock_fence(ctx[0], 1); + if (!f3) + goto error_put_f2; + + dma_fence_enable_sw_signaling(f3); + + f4 = __mock_fence(ctx[1], 2); + if (!f4) + goto error_put_f3; + + dma_fence_enable_sw_signaling(f4); + + f5 = mock_array(2, dma_fence_get(f1), dma_fence_get(f2)); + if (!f5) + goto error_put_f4; + + f6 = mock_array(2, dma_fence_get(f3), dma_fence_get(f4)); + if (!f6) + goto error_put_f5; + + f7 = dma_fence_unwrap_merge(f5, f6); + if (!f7) + goto error_put_f6; + + err = 0; + dma_fence_unwrap_for_each(fence, &iter, f7) { + if (fence == f1 && f4) { + dma_fence_put(f1); + f1 = NULL; + } else if (fence == f4 && !f1) { + dma_fence_put(f4); + f4 = NULL; + } else { + pr_err("Unexpected fence!\n"); + err = -EINVAL; + } + } + + if (f1 || f4) { + pr_err("Not all fences seen!\n"); + err = -EINVAL; + } + + dma_fence_put(f7); +error_put_f6: + dma_fence_put(f6); +error_put_f5: + dma_fence_put(f5); +error_put_f4: + dma_fence_put(f4); +error_put_f3: + dma_fence_put(f3); +error_put_f2: + dma_fence_put(f2); +error_put_f1: + dma_fence_put(f1); + return err; +} + int dma_fence_unwrap(void) { static const struct subtest tests[] = { @@ -441,8 +631,11 @@ int dma_fence_unwrap(void) SUBTEST(unwrap_chain), SUBTEST(unwrap_chain_array), SUBTEST(unwrap_merge), + SUBTEST(unwrap_merge_duplicate), + SUBTEST(unwrap_merge_seqno), SUBTEST(unwrap_merge_order), SUBTEST(unwrap_merge_complex), + SUBTEST(unwrap_merge_complex_seqno), }; return subtests(tests, NULL); From 97395ce76edcce4d39419e90a65d84960fd48aee Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Wed, 8 Jan 2025 17:43:29 -0600 Subject: [PATCH 0042/1443] drm/nouveau: fix kernel-doc comments Fix some malformed kernel-doc comments that were added in a recent commit. Also, kernel-doc does not support global variables, so change those kernel-doc comments into regular comments. Fixes: 214c9539cf2f ("drm/nouveau: expose GSP-RM logging buffers via debugfs") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202412310834.jtCJj4oz-lkp@intel.com/ Signed-off-by: Timur Tabi Reviewed-by: Ben Skeggs Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250108234329.842256-1-ttabi@nvidia.com --- drivers/gpu/drm/nouveau/nouveau_drm.c | 2 +- drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 12 ++++++++---- 2 files changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 21d2d9ca5e85..8c970f018c00 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -118,7 +118,7 @@ static struct drm_driver driver_platform; #ifdef CONFIG_DEBUG_FS struct dentry *nouveau_debugfs_root; -/** +/* * gsp_logs - list of nvif_log GSP-RM logging buffers * * Head pointer to a a list of nvif_log buffers that is created for each GPU diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 58502102926b..10fe2d15b5ce 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -1111,7 +1111,7 @@ enum registry_type { #define REGISTRY_MAX_KEY_LENGTH 64 /** - * registry_list_entry - linked list member for a registry key/value + * struct registry_list_entry - linked list member for a registry key/value * @head: list_head struct * @type: dword, binary, or string * @klen: the length of name of the key @@ -1327,7 +1327,7 @@ struct nv_gsp_registry_entries { u32 value; }; -/** +/* * r535_registry_entries - required registry entries for GSP-RM * * This array lists registry entries that are required for GSP-RM to @@ -2101,7 +2101,7 @@ MODULE_PARM_DESC(keep_gsp_logging, #define NV_GSP_MSG_EVENT_UCODE_LIBOS_CLASS_PMU 0xf3d722 /** - * rpc_ucode_libos_print_v1E_08 - RPC payload for libos print buffers + * struct rpc_ucode_libos_print_v1e_08 - RPC payload for libos print buffers * @ucode_eng_desc: the engine descriptor * @libos_print_buf_size: the size of the libos_print_buf[] * @libos_print_buf: the actual buffer @@ -2162,7 +2162,7 @@ r535_gsp_msg_libos_print(void *priv, u32 fn, void *repv, u32 repc) } /** - * create_debufgs - create a blob debugfs entry + * create_debugfs - create a blob debugfs entry * @gsp: gsp pointer * @name: name of this dentry * @blob: blob wrapper @@ -2788,6 +2788,10 @@ static bool is_empty(const struct debugfs_blob_wrapper *b) /** * r535_gsp_copy_log - preserve the logging buffers in a blob + * @parent: the top-level dentry for this GPU + * @name: name of debugfs entry to create + * @s: original wrapper object to copy from + * @t: new wrapper object to copy to * * When GSP shuts down, the nvkm_gsp object and all its memory is deleted. * To preserve the logging buffers, the buffers need to be copied, but only From 0c2768bf818904db705ff18674f771c3c4d8bbca Mon Sep 17 00:00:00 2001 From: Lizhi Hou Date: Thu, 9 Jan 2025 11:48:11 -0800 Subject: [PATCH 0043/1443] accel/amdxdna: Return error when setting clock failed for npu1 Due to miss returning error when setting clock, the smatch static checker reports warning: drivers/accel/amdxdna/aie2_smu.c:68 npu1_set_dpm() error: uninitialized symbol 'freq'. Fixes: f4d7b8a6bc8c ("accel/amdxdna: Enhance power management settings") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/dri-devel/202267d0-882e-4593-b58d-be9274592f9b@stanley.mountain/ Signed-off-by: Lizhi Hou Reviewed-by: Mario Limonciello Signed-off-by: Mario Limonciello Link: https://patchwork.freedesktop.org/patch/msgid/20250109194811.499505-1-lizhi.hou@amd.com --- drivers/accel/amdxdna/aie2_smu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/accel/amdxdna/aie2_smu.c b/drivers/accel/amdxdna/aie2_smu.c index 73388443c676..d303701b0ded 100644 --- a/drivers/accel/amdxdna/aie2_smu.c +++ b/drivers/accel/amdxdna/aie2_smu.c @@ -64,6 +64,7 @@ int npu1_set_dpm(struct amdxdna_dev_hdl *ndev, u32 dpm_level) if (ret) { XDNA_ERR(ndev->xdna, "Set npu clock to %d failed, ret %d\n", ndev->priv->dpm_clk_tbl[dpm_level].npuclk, ret); + return ret; } ndev->npuclk_freq = freq; @@ -72,6 +73,7 @@ int npu1_set_dpm(struct amdxdna_dev_hdl *ndev, u32 dpm_level) if (ret) { XDNA_ERR(ndev->xdna, "Set h clock to %d failed, ret %d\n", ndev->priv->dpm_clk_tbl[dpm_level].hclk, ret); + return ret; } ndev->hclk_freq = freq; ndev->dpm_level = dpm_level; From 1854df7087be70ad54e24b2e308d7558ebea9f27 Mon Sep 17 00:00:00 2001 From: Derek Foreman Date: Tue, 17 Dec 2024 14:17:07 -0600 Subject: [PATCH 0044/1443] drm/rockchip: Don't change hdmi reference clock rate The code that changes hdmi->ref_clk was accidentally copied from downstream code that sets a different clock. We don't actually want to set any clock here at all. Setting this clock incorrectly leads to incorrect timings for DDC, CEC, and HDCP signal generation. No Fixes listed, as the theoretical timing error in DDC appears to still be within tolerances and harmless - and HDCP and CEC are not yet supported. Signed-off-by: Derek Foreman Reviewed-by: Cristian Ciocaltea Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20241217201708.3320673-1-derek.foreman@collabora.com --- drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c | 13 ------------- 1 file changed, 13 deletions(-) diff --git a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c index 9916843a6ff3..f41151d49fca 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi_qp-rockchip.c @@ -89,7 +89,6 @@ struct rockchip_hdmi_qp { struct regmap *regmap; struct regmap *vo_regmap; struct rockchip_encoder encoder; - struct clk *ref_clk; struct dw_hdmi_qp *hdmi; struct phy *phy; struct gpio_desc *enable_gpio; @@ -122,7 +121,6 @@ static void dw_hdmi_qp_rockchip_encoder_enable(struct drm_encoder *encoder) if (crtc && crtc->state) { rate = drm_hdmi_compute_mode_clock(&crtc->state->adjusted_mode, 8, HDMI_COLORSPACE_RGB); - clk_set_rate(hdmi->ref_clk, rate); /* * FIXME: Temporary workaround to pass pixel clock rate * to the PHY driver until phy_configure_opts_hdmi @@ -515,17 +513,6 @@ static int dw_hdmi_qp_rockchip_bind(struct device *dev, struct device *master, return ret; } - for (i = 0; i < ret; i++) { - if (!strcmp(clks[i].id, "ref")) { - hdmi->ref_clk = clks[1].clk; - break; - } - } - if (!hdmi->ref_clk) { - drm_err(hdmi, "Missing ref clock\n"); - return -EINVAL; - } - hdmi->enable_gpio = devm_gpiod_get_optional(hdmi->dev, "enable", GPIOD_OUT_HIGH); if (IS_ERR(hdmi->enable_gpio)) { From d995dc60e0e9611cc11dc869b176c66dc16b7245 Mon Sep 17 00:00:00 2001 From: Florent Tomasin Date: Tue, 7 Jan 2025 17:33:09 +0000 Subject: [PATCH 0045/1443] drm/panthor: Remove dead code Remove unused function declaration in panthor_gem.h: - `panthor_gem_prime_import_sg_table()` Remove duplicate macro definitions: - `MAX_CSG_PRIO` - `MIN_CS_PER_CSG` - `MIN_CSGS` Signed-off-by: Florent Tomasin Reviewed-by: Boris Brezillon Reviewed-by: Steven Price Signed-off-by: Steven Price Link: https://patchwork.freedesktop.org/patch/msgid/20250107173310.88329-1-florent.tomasin@arm.com --- drivers/gpu/drm/panthor/panthor_fw.c | 1 - drivers/gpu/drm/panthor/panthor_gem.h | 5 ----- drivers/gpu/drm/panthor/panthor_sched.c | 3 --- 3 files changed, 9 deletions(-) diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c index 68eb4fb4d3a8..4a9c4afa9ad7 100644 --- a/drivers/gpu/drm/panthor/panthor_fw.c +++ b/drivers/gpu/drm/panthor/panthor_fw.c @@ -201,7 +201,6 @@ struct panthor_fw_section { #define MIN_CS_PER_CSG 8 #define MIN_CSGS 3 -#define MAX_CSG_PRIO 0xf #define CSF_IFACE_VERSION(major, minor, patch) \ (((major) << 24) | ((minor) << 16) | (patch)) diff --git a/drivers/gpu/drm/panthor/panthor_gem.h b/drivers/gpu/drm/panthor/panthor_gem.h index e43021cf6d45..5749ef2ebe03 100644 --- a/drivers/gpu/drm/panthor/panthor_gem.h +++ b/drivers/gpu/drm/panthor/panthor_gem.h @@ -85,11 +85,6 @@ struct panthor_gem_object *to_panthor_bo(struct drm_gem_object *obj) struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t size); -struct drm_gem_object * -panthor_gem_prime_import_sg_table(struct drm_device *ddev, - struct dma_buf_attachment *attach, - struct sg_table *sgt); - int panthor_gem_create_with_handle(struct drm_file *file, struct drm_device *ddev, diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c index 77b184c3fb0c..5844a7f639e0 100644 --- a/drivers/gpu/drm/panthor/panthor_sched.c +++ b/drivers/gpu/drm/panthor/panthor_sched.c @@ -88,9 +88,6 @@ #define JOB_TIMEOUT_MS 5000 -#define MIN_CS_PER_CSG 8 - -#define MIN_CSGS 3 #define MAX_CSG_PRIO 0xf #define NUM_INSTRS_PER_CACHE_LINE (64 / sizeof(u64)) From 9104ee0868ff0a944f9f7f5ac30bfa88eecaa289 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 10 Jan 2025 22:28:32 -0800 Subject: [PATCH 0046/1443] drm/panthor: fix all mmu kernel-doc comments Use the correct format for all kernel-doc comments. Use structname.membername for named structs. Don't precede function names in kernel-doc with '@' sign. Use the correct function parameter names in kernel-doc comments. This fixes around 80 kernel-doc warnings. Signed-off-by: Randy Dunlap Cc: Boris Brezillon Cc: Steven Price Cc: Liviu Dudau Cc: Maarten Lankhorst Cc: Maxime Ripard Cc: Thomas Zimmermann Cc: David Airlie Cc: Simona Vetter Reviewed-by: Steven Price Signed-off-by: Steven Price Link: https://patchwork.freedesktop.org/patch/msgid/20250111062832.910495-1-rdunlap@infradead.org --- drivers/gpu/drm/panthor/panthor_mmu.c | 71 ++++++++++++++------------- 1 file changed, 37 insertions(+), 34 deletions(-) diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c index c39e3eb1c15d..5ce80e2532d5 100644 --- a/drivers/gpu/drm/panthor/panthor_mmu.c +++ b/drivers/gpu/drm/panthor/panthor_mmu.c @@ -53,26 +53,27 @@ struct panthor_mmu { /** @irq: The MMU irq. */ struct panthor_irq irq; - /** @as: Address space related fields. + /** + * @as: Address space related fields. * * The GPU has a limited number of address spaces (AS) slots, forcing * us to re-assign them to re-assign slots on-demand. */ struct { - /** @slots_lock: Lock protecting access to all other AS fields. */ + /** @as.slots_lock: Lock protecting access to all other AS fields. */ struct mutex slots_lock; - /** @alloc_mask: Bitmask encoding the allocated slots. */ + /** @as.alloc_mask: Bitmask encoding the allocated slots. */ unsigned long alloc_mask; - /** @faulty_mask: Bitmask encoding the faulty slots. */ + /** @as.faulty_mask: Bitmask encoding the faulty slots. */ unsigned long faulty_mask; - /** @slots: VMs currently bound to the AS slots. */ + /** @as.slots: VMs currently bound to the AS slots. */ struct panthor_as_slot slots[MAX_AS_SLOTS]; /** - * @lru_list: List of least recently used VMs. + * @as.lru_list: List of least recently used VMs. * * We use this list to pick a VM to evict when all slots are * used. @@ -87,16 +88,16 @@ struct panthor_mmu { /** @vm: VMs management fields */ struct { - /** @lock: Lock protecting access to list. */ + /** @vm.lock: Lock protecting access to list. */ struct mutex lock; - /** @list: List containing all VMs. */ + /** @vm.list: List containing all VMs. */ struct list_head list; - /** @reset_in_progress: True if a reset is in progress. */ + /** @vm.reset_in_progress: True if a reset is in progress. */ bool reset_in_progress; - /** @wq: Workqueue used for the VM_BIND queues. */ + /** @vm.wq: Workqueue used for the VM_BIND queues. */ struct workqueue_struct *wq; } vm; }; @@ -143,14 +144,14 @@ struct panthor_vma { struct panthor_vm_op_ctx { /** @rsvd_page_tables: Pages reserved for the MMU page table update. */ struct { - /** @count: Number of pages reserved. */ + /** @rsvd_page_tables.count: Number of pages reserved. */ u32 count; - /** @ptr: Point to the first unused page in the @pages table. */ + /** @rsvd_page_tables.ptr: Point to the first unused page in the @pages table. */ u32 ptr; /** - * @page: Array of pages that can be used for an MMU page table update. + * @rsvd_page_tables.pages: Array of pages to be used for an MMU page table update. * * After an VM operation, there might be free pages left in this array. * They should be returned to the pt_cache as part of the op_ctx cleanup. @@ -172,10 +173,10 @@ struct panthor_vm_op_ctx { /** @va: Virtual range targeted by the VM operation. */ struct { - /** @addr: Start address. */ + /** @va.addr: Start address. */ u64 addr; - /** @range: Range size. */ + /** @va.range: Range size. */ u64 range; } va; @@ -195,14 +196,14 @@ struct panthor_vm_op_ctx { /** @map: Fields specific to a map operation. */ struct { - /** @vm_bo: Buffer object to map. */ + /** @map.vm_bo: Buffer object to map. */ struct drm_gpuvm_bo *vm_bo; - /** @bo_offset: Offset in the buffer object. */ + /** @map.bo_offset: Offset in the buffer object. */ u64 bo_offset; /** - * @sgt: sg-table pointing to pages backing the GEM object. + * @map.sgt: sg-table pointing to pages backing the GEM object. * * This is gathered at job creation time, such that we don't have * to allocate in ::run_job(). @@ -210,7 +211,7 @@ struct panthor_vm_op_ctx { struct sg_table *sgt; /** - * @new_vma: The new VMA object that will be inserted to the VA tree. + * @map.new_vma: The new VMA object that will be inserted to the VA tree. */ struct panthor_vma *new_vma; } map; @@ -304,27 +305,27 @@ struct panthor_vm { /** @kernel_auto_va: Automatic VA-range for kernel BOs. */ struct { - /** @start: Start of the automatic VA-range for kernel BOs. */ + /** @kernel_auto_va.start: Start of the automatic VA-range for kernel BOs. */ u64 start; - /** @size: Size of the automatic VA-range for kernel BOs. */ + /** @kernel_auto_va.size: Size of the automatic VA-range for kernel BOs. */ u64 end; } kernel_auto_va; /** @as: Address space related fields. */ struct { /** - * @id: ID of the address space this VM is bound to. + * @as.id: ID of the address space this VM is bound to. * * A value of -1 means the VM is inactive/not bound. */ int id; - /** @active_cnt: Number of active users of this VM. */ + /** @as.active_cnt: Number of active users of this VM. */ refcount_t active_cnt; /** - * @lru_node: Used to instead the VM in the panthor_mmu::as::lru_list. + * @as.lru_node: Used to instead the VM in the panthor_mmu::as::lru_list. * * Active VMs should not be inserted in the LRU list. */ @@ -336,13 +337,13 @@ struct panthor_vm { */ struct { /** - * @pool: The heap pool attached to this VM. + * @heaps.pool: The heap pool attached to this VM. * * Will stay NULL until someone creates a heap context on this VM. */ struct panthor_heap_pool *pool; - /** @lock: Lock used to protect access to @pool. */ + /** @heaps.lock: Lock used to protect access to @pool. */ struct mutex lock; } heaps; @@ -408,7 +409,7 @@ struct panthor_vm_bind_job { struct panthor_vm_op_ctx ctx; }; -/** +/* * @pt_cache: Cache used to allocate MMU page tables. * * The pre-allocation pattern forces us to over-allocate to plan for @@ -478,7 +479,7 @@ static void *alloc_pt(void *cookie, size_t size, gfp_t gfp) } /** - * @free_pt() - Custom page table free function + * free_pt() - Custom page table free function * @cookie: Cookie passed at page table allocation time. * @data: Page table to free. * @size: Size of the page table. This size should be fixed, @@ -697,7 +698,7 @@ static void panthor_vm_release_as_locked(struct panthor_vm *vm) /** * panthor_vm_active() - Flag a VM as active - * @VM: VM to flag as active. + * @vm: VM to flag as active. * * Assigns an address space to a VM so it can be used by the GPU/MCU. * @@ -801,7 +802,7 @@ int panthor_vm_active(struct panthor_vm *vm) /** * panthor_vm_idle() - Flag a VM idle - * @VM: VM to flag as idle. + * @vm: VM to flag as idle. * * When we know the GPU is done with the VM (no more jobs to process), * we can relinquish the AS slot attached to this VM, if any. @@ -1017,7 +1018,7 @@ static int flags_to_prot(u32 flags) /** * panthor_vm_alloc_va() - Allocate a region in the auto-va space - * @VM: VM to allocate a region on. + * @vm: VM to allocate a region on. * @va: start of the VA range. Can be PANTHOR_VM_KERNEL_AUTO_VA if the user * wants the VA to be automatically allocated from the auto-VA range. * @size: size of the VA range. @@ -1063,7 +1064,7 @@ panthor_vm_alloc_va(struct panthor_vm *vm, u64 va, u64 size, /** * panthor_vm_free_va() - Free a region allocated with panthor_vm_alloc_va() - * @VM: VM to free the region on. + * @vm: VM to free the region on. * @va_node: Memory node representing the region to free. */ void panthor_vm_free_va(struct panthor_vm *vm, struct drm_mm_node *va_node) @@ -1492,9 +1493,9 @@ panthor_vm_create_check_args(const struct panthor_device *ptdev, /** * panthor_vm_pool_create_vm() - Create a VM + * @ptdev: The panthor device * @pool: The VM to create this VM on. - * @kernel_va_start: Start of the region reserved for kernel objects. - * @kernel_va_range: Size of the region reserved for kernel objects. + * @args: VM creation args. * * Return: a positive VM ID on success, a negative error code otherwise. */ @@ -1558,6 +1559,8 @@ static void panthor_vm_destroy(struct panthor_vm *vm) * * The VM resources are freed when the last reference on the VM object is * dropped. + * + * Return: %0 for success, negative errno value for failure */ int panthor_vm_pool_destroy_vm(struct panthor_vm_pool *pool, u32 handle) { From 573b73e5ac2ce0d58859eace8218f3a7e9212186 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Fri, 10 Jan 2025 11:13:01 +0000 Subject: [PATCH 0047/1443] drm/sched: Delete unused update_job_credits MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit No driver is using the update_job_credits() schduler vfunc so lets remove it. Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Danilo Krummrich Cc: Matthew Brost Cc: Philipp Stanner Acked-by: Danilo Krummrich Acked-by: Boris Brezillon Acked-by: Matt Coster Signed-off-by: Philipp Stanner Link: https://patchwork.freedesktop.org/patch/msgid/20250110111301.76909-1-tvrtko.ursulin@igalia.com --- drivers/gpu/drm/scheduler/sched_main.c | 13 ------------- include/drm/gpu_scheduler.h | 13 ------------- 2 files changed, 26 deletions(-) diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 57da84908752..81d69c2ff6ab 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -64,12 +64,6 @@ * credit limit, the job won't be executed. Instead, the scheduler will wait * until the credit count has decreased enough to not overflow its credit limit. * This implies waiting for previously executed jobs. - * - * Optionally, drivers may register a callback (update_job_credits) provided by - * struct drm_sched_backend_ops to update the job's credits dynamically. The - * scheduler executes this callback every time the scheduler considers a job for - * execution and subsequently checks whether the job fits the scheduler's credit - * limit. */ #include @@ -133,13 +127,6 @@ static bool drm_sched_can_queue(struct drm_gpu_scheduler *sched, if (!s_job) return false; - if (sched->ops->update_job_credits) { - s_job->credits = sched->ops->update_job_credits(s_job); - - drm_WARN(sched, !s_job->credits, - "Jobs with zero credits bypass job-flow control.\n"); - } - /* If a job exceeds the credit limit, truncate it to the credit limit * itself to guarantee forward progress. */ diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index 95e17504e46a..e2e6af8849c6 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -476,19 +476,6 @@ struct drm_sched_backend_ops { * and it's time to clean it up. */ void (*free_job)(struct drm_sched_job *sched_job); - - /** - * @update_job_credits: Called when the scheduler is considering this - * job for execution. - * - * This callback returns the number of credits the job would take if - * pushed to the hardware. Drivers may use this to dynamically update - * the job's credit count. For instance, deduct the number of credits - * for already signalled native fences. - * - * This callback is optional. - */ - u32 (*update_job_credits)(struct drm_sched_job *sched_job); }; /** From 155c77f45f63dd58a37eeb0896b0b140ab785836 Mon Sep 17 00:00:00 2001 From: Maciej Patelczyk Date: Wed, 11 Dec 2024 12:17:26 +0100 Subject: [PATCH 0048/1443] drm/xe: introduce xe_gt_reset and xe_gt_wait_for_reset Add synchronous version gt reset as there are few places where it is expected. Also add a wait helper to wait until gt reset is done. Signed-off-by: Maciej Patelczyk Reviewed-by: Lucas De Marchi Fixes: f3bc5bb4d53d ("drm/xe: Allow userspace to configure CCS mode") Reviewed-by: Nirmoy Das Link: https://patchwork.freedesktop.org/patch/msgid/20241211111727.1481476-2-maciej.patelczyk@intel.com Signed-off-by: Nirmoy Das --- drivers/gpu/drm/xe/tests/xe_bo.c | 7 +++---- drivers/gpu/drm/xe/tests/xe_mocs.c | 3 +-- drivers/gpu/drm/xe/xe_gt.h | 25 +++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_gt_debugfs.c | 4 +--- 4 files changed, 30 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c index 405ff904153e..6795d1d916e4 100644 --- a/drivers/gpu/drm/xe/tests/xe_bo.c +++ b/drivers/gpu/drm/xe/tests/xe_bo.c @@ -264,10 +264,9 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc * however seems quite fragile not to also restart the GT. Try * to do that here by triggering a GT reset. */ - for_each_gt(__gt, xe, id) { - xe_gt_reset_async(__gt); - flush_work(&__gt->reset.worker); - } + for_each_gt(__gt, xe, id) + xe_gt_reset(__gt); + if (err) { KUNIT_FAIL(test, "restore kernel err=%pe\n", ERR_PTR(err)); diff --git a/drivers/gpu/drm/xe/tests/xe_mocs.c b/drivers/gpu/drm/xe/tests/xe_mocs.c index d3f71d13eb81..ef1e5256c56a 100644 --- a/drivers/gpu/drm/xe/tests/xe_mocs.c +++ b/drivers/gpu/drm/xe/tests/xe_mocs.c @@ -162,8 +162,7 @@ static int mocs_reset_test_run_device(struct xe_device *xe) if (flags & HAS_LNCF_MOCS) read_l3cc_table(gt, &mocs.table); - xe_gt_reset_async(gt); - flush_work(>->reset.worker); + xe_gt_reset(gt); kunit_info(test, "mocs_reset_test after reset\n"); if (flags & HAS_GLOBAL_MOCS) diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h index 4e4e8e103419..e504cc33ade4 100644 --- a/drivers/gpu/drm/xe/xe_gt.h +++ b/drivers/gpu/drm/xe/xe_gt.h @@ -56,6 +56,31 @@ void xe_gt_sanitize(struct xe_gt *gt); int xe_gt_sanitize_freq(struct xe_gt *gt); void xe_gt_remove(struct xe_gt *gt); +/** + * xe_gt_wait_for_reset - wait for gt's async reset to finalize. + * @gt: GT structure + * Return: + * %true if it waited for the work to finish execution, + * %false if there was no scheduled reset or it was done. + */ +static inline bool xe_gt_wait_for_reset(struct xe_gt *gt) +{ + return flush_work(>->reset.worker); +} + +/** + * xe_gt_reset - perform synchronous reset + * @gt: GT structure + * Return: + * %true if it waited for the reset to finish, + * %false if there was no scheduled reset. + */ +static inline bool xe_gt_reset(struct xe_gt *gt) +{ + xe_gt_reset_async(gt); + return xe_gt_wait_for_reset(gt); +} + /** * xe_gt_any_hw_engine_by_reset_domain - scan the list of engines and return the * first that matches the same reset domain as @class diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c index 3e8c351a0eab..e7792858b1e4 100644 --- a/drivers/gpu/drm/xe/xe_gt_debugfs.c +++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c @@ -132,11 +132,9 @@ static int force_reset(struct xe_gt *gt, struct drm_printer *p) static int force_reset_sync(struct xe_gt *gt, struct drm_printer *p) { xe_pm_runtime_get(gt_to_xe(gt)); - xe_gt_reset_async(gt); + xe_gt_reset(gt); xe_pm_runtime_put(gt_to_xe(gt)); - flush_work(>->reset.worker); - return 0; } From 480fb9806e2e073532f7786166287114c696b340 Mon Sep 17 00:00:00 2001 From: Maciej Patelczyk Date: Wed, 11 Dec 2024 12:17:27 +0100 Subject: [PATCH 0049/1443] drm/xe: make change ccs_mode a synchronous action If ccs_mode is being modified via /sys/class/drm/cardX/device/tileY/gtY/ccs_mode the asynchronous reset is triggered and the write returns immediately. With that some test receive false information about number of CCS engines or even fail if they proceed without delay after changing the ccs_mode. Changing the ccs_mode change from async to sync to prevent failures in tests. Signed-off-by: Maciej Patelczyk Reviewed-by: Lucas De Marchi Fixes: f3bc5bb4d53d ("drm/xe: Allow userspace to configure CCS mode") Reviewed-by: Nirmoy Das Link: https://patchwork.freedesktop.org/patch/msgid/20241211111727.1481476-3-maciej.patelczyk@intel.com Signed-off-by: Nirmoy Das --- drivers/gpu/drm/xe/xe_gt_ccs_mode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c index b6adfb9f2030..50fffc9ebf62 100644 --- a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c +++ b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c @@ -150,7 +150,7 @@ ccs_mode_store(struct device *kdev, struct device_attribute *attr, xe_gt_info(gt, "Setting compute mode to %d\n", num_engines); gt->ccs_mode = num_engines; xe_gt_record_user_engines(gt); - xe_gt_reset_async(gt); + xe_gt_reset(gt); } mutex_unlock(&xe->drm.filelist_mutex); From 92029e0baa5313ba208103f90086f59070bbf93b Mon Sep 17 00:00:00 2001 From: Nirmoy Das Date: Wed, 8 Jan 2025 15:13:23 +0100 Subject: [PATCH 0050/1443] drm/xe/ptl: Apply Wa_14023061436 Enable WMTP for the BTD kernel to address Wa14023061436 by setting the proper TDL Chicken Bit. v2: Apply it on engine_was[] as this register is not part of LRC(Matt) Apply it for first_render_or_compute in case this gets extended to compute only platforms(Matt). Cc: Gustavo Sousa Cc: Matt Roper Reviewed-by: Matt Roper Link: https://patchwork.freedesktop.org/patch/msgid/20250108141323.311601-1-nirmoy.das@intel.com Signed-off-by: Nirmoy Das --- drivers/gpu/drm/xe/regs/xe_gt_regs.h | 3 +++ drivers/gpu/drm/xe/xe_wa.c | 5 +++++ 2 files changed, 8 insertions(+) diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h index 162f18e975da..b4283ac030f4 100644 --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h @@ -500,6 +500,9 @@ #define LSC_L1_FLUSH_CTL_3D_DATAPORT_FLUSH_EVENTS_MASK REG_GENMASK(13, 11) #define DIS_ATOMIC_CHAINING_TYPED_WRITES REG_BIT(3) +#define TDL_CHICKEN XE_REG_MCR(0xe5f4, XE_REG_OPTION_MASKED) +#define QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE REG_BIT(12) + #define LSC_CHICKEN_BIT_0 XE_REG_MCR(0xe7c8) #define DISABLE_D8_D16_COASLESCE REG_BIT(30) #define WR_REQ_CHAINING_DIS REG_BIT(26) diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c index 570fe0376402..744dba4fdb58 100644 --- a/drivers/gpu/drm/xe/xe_wa.c +++ b/drivers/gpu/drm/xe/xe_wa.c @@ -613,6 +613,11 @@ static const struct xe_rtp_entry_sr engine_was[] = { XE_RTP_ACTIONS(FIELD_SET(SAMPLER_MODE, SMP_WAIT_FETCH_MERGING_COUNTER, SMP_FORCE_128B_OVERFETCH)) }, + { XE_RTP_NAME("14023061436"), + XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001), + FUNC(xe_rtp_match_first_render_or_compute)), + XE_RTP_ACTIONS(SET(TDL_CHICKEN, QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE)) + }, {} }; From 57e233c3bd63f32d2c7e937db2e16b98f723ce2f Mon Sep 17 00:00:00 2001 From: Boris Brezillon Date: Tue, 17 Dec 2024 10:24:57 +0100 Subject: [PATCH 0051/1443] drm/panthor: Fix a race between the reset and suspend path If a reset is scheduled when the suspend happens, we drop the reset-pending info on the floor assuming the resume will fix things, but the resume logic might try a fast reset. If we're lucky, the fast reset fails and we fallback to a slow reset, but if the FW was corrupted in a way that makes it partially functional (it boots but doesn't quite do what it's expected to do), we won't notice immediately that things are not working correctly, leading to a new reset further down the road. Fixes: 5fe909cae118 ("drm/panthor: Add the device logical block") Signed-off-by: Boris Brezillon Reviewed-by: Liviu Dudau Reviewed-by: Steven Price Signed-off-by: Steven Price Link: https://patchwork.freedesktop.org/patch/msgid/20241217092457.1582053-1-boris.brezillon@collabora.com --- drivers/gpu/drm/panthor/panthor_device.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c index 0a37cfeeb181..a9da1d1eeb70 100644 --- a/drivers/gpu/drm/panthor/panthor_device.c +++ b/drivers/gpu/drm/panthor/panthor_device.c @@ -128,14 +128,11 @@ static void panthor_device_reset_work(struct work_struct *work) struct panthor_device *ptdev = container_of(work, struct panthor_device, reset.work); int ret = 0, cookie; - if (atomic_read(&ptdev->pm.state) != PANTHOR_DEVICE_PM_STATE_ACTIVE) { - /* - * No need for a reset as the device has been (or will be) - * powered down - */ - atomic_set(&ptdev->reset.pending, 0); + /* If the device is entering suspend, we don't reset. A slow reset will + * be forced at resume time instead. + */ + if (atomic_read(&ptdev->pm.state) != PANTHOR_DEVICE_PM_STATE_ACTIVE) return; - } if (!drm_dev_enter(&ptdev->base, &cookie)) return; @@ -477,6 +474,14 @@ int panthor_device_resume(struct device *dev) if (panthor_device_is_initialized(ptdev) && drm_dev_enter(&ptdev->base, &cookie)) { + /* If there was a reset pending at the time we suspended the + * device, we force a slow reset. + */ + if (atomic_read(&ptdev->reset.pending)) { + ptdev->reset.fast = false; + atomic_set(&ptdev->reset.pending, 0); + } + ret = panthor_device_resume_hw_components(ptdev); if (ret && ptdev->reset.fast) { drm_err(&ptdev->base, "Fast reset failed, trying a slow reset"); @@ -493,9 +498,6 @@ int panthor_device_resume(struct device *dev) goto err_suspend_devfreq; } - if (atomic_read(&ptdev->reset.pending)) - queue_work(ptdev->reset.wq, &ptdev->reset.work); - /* Clear all IOMEM mappings pointing to this device after we've * resumed. This way the fake mappings pointing to the dummy pages * are removed and the real iomem mapping will be restored on next From 1f463794097dcdf4c64fffd31de2177681525e35 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Tue, 17 Dec 2024 15:58:44 +0100 Subject: [PATCH 0052/1443] drm/ttm: Balance ttm_resource_cursor_init() and ttm_resource_cursor_fini() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Make the interface more symmetric by providing and using a ttm_resource_cursor_init(). v10: - Fix a stray newline (Matthew Brost) - Update kerneldoc (Matthew Brost) Signed-off-by: Thomas Hellström Reviewed-by: Matthew Brost Reviewed-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20241217145852.37342-2-thomas.hellstrom@linux.intel.com Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_bo.c | 3 ++- drivers/gpu/drm/ttm/ttm_bo_util.c | 3 ++- drivers/gpu/drm/ttm/ttm_resource.c | 35 ++++++++++++++++++++---------- include/drm/ttm/ttm_resource.h | 11 +++++----- 4 files changed, 34 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 48c5365efca1..06d6a452c4f4 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -450,7 +450,8 @@ int ttm_bo_evict_first(struct ttm_device *bdev, struct ttm_resource_manager *man int ret = 0; spin_lock(&bdev->lru_lock); - res = ttm_resource_manager_first(man, &cursor); + ttm_resource_cursor_init(&cursor, man); + res = ttm_resource_manager_first(&cursor); ttm_resource_cursor_fini(&cursor); if (!res) { ret = -ENOENT; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index d939925efa81..917096bd5f68 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -865,7 +865,8 @@ s64 ttm_lru_walk_for_evict(struct ttm_lru_walk *walk, struct ttm_device *bdev, s64 lret; spin_lock(&bdev->lru_lock); - ttm_resource_manager_for_each_res(man, &cursor, res) { + ttm_resource_cursor_init(&cursor, man); + ttm_resource_manager_for_each_res(&cursor, res) { struct ttm_buffer_object *bo = res->bo; bool bo_needs_unlock = false; bool bo_locked = false; diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c index a87665eb28a6..e19360cc7930 100644 --- a/drivers/gpu/drm/ttm/ttm_resource.c +++ b/drivers/gpu/drm/ttm/ttm_resource.c @@ -81,6 +81,23 @@ static void ttm_bulk_move_drop_cursors(struct ttm_lru_bulk_move *bulk) ttm_resource_cursor_clear_bulk(cursor); } +/** + * ttm_resource_cursor_init() - Initialize a struct ttm_resource_cursor + * @cursor: The cursor to initialize. + * @man: The resource manager. + * + * Initialize the cursor before using it for iteration. + */ +void ttm_resource_cursor_init(struct ttm_resource_cursor *cursor, + struct ttm_resource_manager *man) +{ + cursor->priority = 0; + cursor->man = man; + ttm_lru_item_init(&cursor->hitch, TTM_LRU_HITCH); + INIT_LIST_HEAD(&cursor->bulk_link); + INIT_LIST_HEAD(&cursor->hitch.link); +} + /** * ttm_resource_cursor_fini() - Finalize the LRU list cursor usage * @cursor: The struct ttm_resource_cursor to finalize. @@ -593,7 +610,6 @@ ttm_resource_cursor_check_bulk(struct ttm_resource_cursor *cursor, /** * ttm_resource_manager_first() - Start iterating over the resources * of a resource manager - * @man: resource manager to iterate over * @cursor: cursor to record the position * * Initializes the cursor and starts iterating. When done iterating, @@ -602,17 +618,16 @@ ttm_resource_cursor_check_bulk(struct ttm_resource_cursor *cursor, * Return: The first resource from the resource manager. */ struct ttm_resource * -ttm_resource_manager_first(struct ttm_resource_manager *man, - struct ttm_resource_cursor *cursor) +ttm_resource_manager_first(struct ttm_resource_cursor *cursor) { + struct ttm_resource_manager *man = cursor->man; + + if (WARN_ON_ONCE(!man)) + return NULL; + lockdep_assert_held(&man->bdev->lru_lock); - cursor->priority = 0; - cursor->man = man; - ttm_lru_item_init(&cursor->hitch, TTM_LRU_HITCH); - INIT_LIST_HEAD(&cursor->bulk_link); - list_add(&cursor->hitch.link, &man->lru[cursor->priority]); - + list_move(&cursor->hitch.link, &man->lru[cursor->priority]); return ttm_resource_manager_next(cursor); } @@ -648,8 +663,6 @@ ttm_resource_manager_next(struct ttm_resource_cursor *cursor) ttm_resource_cursor_clear_bulk(cursor); } - ttm_resource_cursor_fini(cursor); - return NULL; } diff --git a/include/drm/ttm/ttm_resource.h b/include/drm/ttm/ttm_resource.h index be034be56ba1..e1f3b95d73b6 100644 --- a/include/drm/ttm/ttm_resource.h +++ b/include/drm/ttm/ttm_resource.h @@ -325,6 +325,9 @@ struct ttm_resource_cursor { unsigned int priority; }; +void ttm_resource_cursor_init(struct ttm_resource_cursor *cursor, + struct ttm_resource_manager *man); + void ttm_resource_cursor_fini(struct ttm_resource_cursor *cursor); /** @@ -456,8 +459,7 @@ void ttm_resource_manager_debug(struct ttm_resource_manager *man, struct drm_printer *p); struct ttm_resource * -ttm_resource_manager_first(struct ttm_resource_manager *man, - struct ttm_resource_cursor *cursor); +ttm_resource_manager_first(struct ttm_resource_cursor *cursor); struct ttm_resource * ttm_resource_manager_next(struct ttm_resource_cursor *cursor); @@ -466,14 +468,13 @@ ttm_lru_first_res_or_null(struct list_head *head); /** * ttm_resource_manager_for_each_res - iterate over all resources - * @man: the resource manager * @cursor: struct ttm_resource_cursor for the current position * @res: the current resource * * Iterate over all the evictable resources in a resource manager. */ -#define ttm_resource_manager_for_each_res(man, cursor, res) \ - for (res = ttm_resource_manager_first(man, cursor); res; \ +#define ttm_resource_manager_for_each_res(cursor, res) \ + for (res = ttm_resource_manager_first(cursor); res; \ res = ttm_resource_manager_next(cursor)) struct ttm_kmap_iter * From 1d3160c7cba37b48be453f29ec2faede77002d14 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Tue, 17 Dec 2024 15:58:46 +0100 Subject: [PATCH 0053/1443] drm/ttm/pool: Restructure the pool allocation code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Simplify the pool allocation code somewhat by merging loop arguments used by multiple functions together in a struct and simplifying the loop. Also add documentation. This hopefully makes the behaviour of the allocation loop simplier to understand, but above all paves the way for upcoming restore-while-allocating functionality. There are no functional changes, but the "allow_pools" bool introduced to keep current functionality could be removed as a follow up, which would enable using write-back cached pools when allocating memory for other caching modes, rather than to resort to allocating from the system directly. v15: - Introduce this patch to simplify the upcoming patch that introduces restore while allocating. Signed-off-by: Thomas Hellström Link: https://patchwork.freedesktop.org/patch/msgid/20241217145852.37342-4-thomas.hellstrom@linux.intel.com Reviewed-by: Christian König Reviewed-by: Matthew Brost Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_pool.c | 183 +++++++++++++++++++-------------- 1 file changed, 108 insertions(+), 75 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_pool.c b/drivers/gpu/drm/ttm/ttm_pool.c index 8504dbe19c1a..c9eba76d5143 100644 --- a/drivers/gpu/drm/ttm/ttm_pool.c +++ b/drivers/gpu/drm/ttm/ttm_pool.c @@ -58,6 +58,23 @@ struct ttm_pool_dma { unsigned long vaddr; }; +/** + * struct ttm_pool_alloc_state - Current state of the tt page allocation process + * @pages: Pointer to the next tt page pointer to populate. + * @caching_divide: Pointer to the first page pointer whose page has a staged but + * not committed caching transition from write-back to @tt_caching. + * @dma_addr: Pointer to the next tt dma_address entry to populate if any. + * @remaining_pages: Remaining pages to populate. + * @tt_caching: The requested cpu-caching for the pages allocated. + */ +struct ttm_pool_alloc_state { + struct page **pages; + struct page **caching_divide; + dma_addr_t *dma_addr; + pgoff_t remaining_pages; + enum ttm_caching tt_caching; +}; + static unsigned long page_pool_size; MODULE_PARM_DESC(page_pool_size, "Number of pages in the WC/UC/DMA pool"); @@ -160,25 +177,25 @@ static void ttm_pool_free_page(struct ttm_pool *pool, enum ttm_caching caching, kfree(dma); } -/* Apply a new caching to an array of pages */ -static int ttm_pool_apply_caching(struct page **first, struct page **last, - enum ttm_caching caching) +/* Apply any cpu-caching deferred during page allocation */ +static int ttm_pool_apply_caching(struct ttm_pool_alloc_state *alloc) { #ifdef CONFIG_X86 - unsigned int num_pages = last - first; + unsigned int num_pages = alloc->pages - alloc->caching_divide; if (!num_pages) return 0; - switch (caching) { + switch (alloc->tt_caching) { case ttm_cached: break; case ttm_write_combined: - return set_pages_array_wc(first, num_pages); + return set_pages_array_wc(alloc->caching_divide, num_pages); case ttm_uncached: - return set_pages_array_uc(first, num_pages); + return set_pages_array_uc(alloc->caching_divide, num_pages); } #endif + alloc->caching_divide = alloc->pages; return 0; } @@ -354,24 +371,41 @@ static unsigned int ttm_pool_page_order(struct ttm_pool *pool, struct page *p) return p->private; } -/* Called when we got a page, either from a pool or newly allocated */ +/* + * Called when we got a page, either from a pool or newly allocated. + * if needed, dma map the page and populate the dma address array. + * Populate the page address array. + * If the caching is consistent, update any deferred caching. Otherwise + * stage this page for an upcoming deferred caching update. + */ static int ttm_pool_page_allocated(struct ttm_pool *pool, unsigned int order, - struct page *p, dma_addr_t **dma_addr, - unsigned long *num_pages, - struct page ***pages) + struct page *p, enum ttm_caching page_caching, + struct ttm_pool_alloc_state *alloc) { - unsigned int i; - int r; + pgoff_t i, nr = 1UL << order; + bool caching_consistent; + int r = 0; - if (*dma_addr) { - r = ttm_pool_map(pool, order, p, dma_addr); + caching_consistent = (page_caching == alloc->tt_caching) || PageHighMem(p); + + if (caching_consistent) { + r = ttm_pool_apply_caching(alloc); if (r) return r; } - *num_pages -= 1 << order; - for (i = 1 << order; i; --i, ++(*pages), ++p) - **pages = p; + if (alloc->dma_addr) { + r = ttm_pool_map(pool, order, p, &alloc->dma_addr); + if (r) + return r; + } + + alloc->remaining_pages -= nr; + for (i = 0; i < nr; ++i) + *alloc->pages++ = p++; + + if (caching_consistent) + alloc->caching_divide = alloc->pages; return 0; } @@ -413,6 +447,26 @@ static void ttm_pool_free_range(struct ttm_pool *pool, struct ttm_tt *tt, } } +static void ttm_pool_alloc_state_init(const struct ttm_tt *tt, + struct ttm_pool_alloc_state *alloc) +{ + alloc->pages = tt->pages; + alloc->caching_divide = tt->pages; + alloc->dma_addr = tt->dma_address; + alloc->remaining_pages = tt->num_pages; + alloc->tt_caching = tt->caching; +} + +/* + * Find a suitable allocation order based on highest desired order + * and number of remaining pages + */ +static unsigned int ttm_pool_alloc_find_order(unsigned int highest, + const struct ttm_pool_alloc_state *alloc) +{ + return min_t(unsigned int, highest, __fls(alloc->remaining_pages)); +} + /** * ttm_pool_alloc - Fill a ttm_tt object * @@ -428,19 +482,19 @@ static void ttm_pool_free_range(struct ttm_pool *pool, struct ttm_tt *tt, int ttm_pool_alloc(struct ttm_pool *pool, struct ttm_tt *tt, struct ttm_operation_ctx *ctx) { - pgoff_t num_pages = tt->num_pages; - dma_addr_t *dma_addr = tt->dma_address; - struct page **caching = tt->pages; - struct page **pages = tt->pages; + struct ttm_pool_alloc_state alloc; enum ttm_caching page_caching; gfp_t gfp_flags = GFP_USER; pgoff_t caching_divide; unsigned int order; + bool allow_pools; struct page *p; int r; - WARN_ON(!num_pages || ttm_tt_is_populated(tt)); - WARN_ON(dma_addr && !pool->dev); + ttm_pool_alloc_state_init(tt, &alloc); + + WARN_ON(!alloc.remaining_pages || ttm_tt_is_populated(tt)); + WARN_ON(alloc.dma_addr && !pool->dev); if (tt->page_flags & TTM_TT_FLAG_ZERO_ALLOC) gfp_flags |= __GFP_ZERO; @@ -453,67 +507,46 @@ int ttm_pool_alloc(struct ttm_pool *pool, struct ttm_tt *tt, else gfp_flags |= GFP_HIGHUSER; - for (order = min_t(unsigned int, MAX_PAGE_ORDER, __fls(num_pages)); - num_pages; - order = min_t(unsigned int, order, __fls(num_pages))) { + page_caching = tt->caching; + allow_pools = true; + for (order = ttm_pool_alloc_find_order(MAX_PAGE_ORDER, &alloc); + alloc.remaining_pages; + order = ttm_pool_alloc_find_order(order, &alloc)) { struct ttm_pool_type *pt; - page_caching = tt->caching; - pt = ttm_pool_select_type(pool, tt->caching, order); - p = pt ? ttm_pool_type_take(pt) : NULL; - if (p) { - r = ttm_pool_apply_caching(caching, pages, - tt->caching); - if (r) - goto error_free_page; - - caching = pages; - do { - r = ttm_pool_page_allocated(pool, order, p, - &dma_addr, - &num_pages, - &pages); - if (r) - goto error_free_page; - - caching = pages; - if (num_pages < (1 << order)) - break; - - p = ttm_pool_type_take(pt); - } while (p); + /* First, try to allocate a page from a pool if one exists. */ + p = NULL; + pt = ttm_pool_select_type(pool, page_caching, order); + if (pt && allow_pools) + p = ttm_pool_type_take(pt); + /* + * If that fails or previously failed, allocate from system. + * Note that this also disallows additional pool allocations using + * write-back cached pools of the same order. Consider removing + * that behaviour. + */ + if (!p) { + page_caching = ttm_cached; + allow_pools = false; + p = ttm_pool_alloc_page(pool, gfp_flags, order); } - - page_caching = ttm_cached; - while (num_pages >= (1 << order) && - (p = ttm_pool_alloc_page(pool, gfp_flags, order))) { - - if (PageHighMem(p)) { - r = ttm_pool_apply_caching(caching, pages, - tt->caching); - if (r) - goto error_free_page; - caching = pages; - } - r = ttm_pool_page_allocated(pool, order, p, &dma_addr, - &num_pages, &pages); - if (r) - goto error_free_page; - if (PageHighMem(p)) - caching = pages; - } - + /* If that fails, lower the order if possible and retry. */ if (!p) { if (order) { --order; + page_caching = tt->caching; + allow_pools = true; continue; } r = -ENOMEM; goto error_free_all; } + r = ttm_pool_page_allocated(pool, order, p, page_caching, &alloc); + if (r) + goto error_free_page; } - r = ttm_pool_apply_caching(caching, pages, tt->caching); + r = ttm_pool_apply_caching(&alloc); if (r) goto error_free_all; @@ -523,10 +556,10 @@ int ttm_pool_alloc(struct ttm_pool *pool, struct ttm_tt *tt, ttm_pool_free_page(pool, page_caching, order, p); error_free_all: - num_pages = tt->num_pages - num_pages; - caching_divide = caching - tt->pages; + caching_divide = alloc.caching_divide - tt->pages; ttm_pool_free_range(pool, tt, tt->caching, 0, caching_divide); - ttm_pool_free_range(pool, tt, ttm_cached, caching_divide, num_pages); + ttm_pool_free_range(pool, tt, ttm_cached, caching_divide, + tt->num_pages - alloc.remaining_pages); return r; } From 0d6c9edf9e5b25361bb23a151ad162878ecd81a5 Mon Sep 17 00:00:00 2001 From: Tejas Vipin Date: Mon, 6 Jan 2025 09:41:29 +0530 Subject: [PATCH 0054/1443] drm/panel: ebbg-ft8719: transition to mipi_dsi wrapped functions Changes the ebbg-ft8719 panel to use multi style functions for improved error handling. Signed-off-by: Tejas Vipin Reviewed-by: Neil Armstrong Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250106041129.114867-1-tejasvipin76@gmail.com --- drivers/gpu/drm/panel/panel-ebbg-ft8719.c | 67 ++++++----------------- 1 file changed, 16 insertions(+), 51 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-ebbg-ft8719.c b/drivers/gpu/drm/panel/panel-ebbg-ft8719.c index e85d63a176d0..0bfed0ec0bbc 100644 --- a/drivers/gpu/drm/panel/panel-ebbg-ft8719.c +++ b/drivers/gpu/drm/panel/panel-ebbg-ft8719.c @@ -57,65 +57,39 @@ static void ebbg_ft8719_reset(struct ebbg_ft8719 *ctx) static int ebbg_ft8719_on(struct ebbg_ft8719 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dsi->mode_flags |= MIPI_DSI_MODE_LPM; - ret = mipi_dsi_dcs_set_display_brightness(dsi, 0x00ff); - if (ret < 0) { - dev_err(dev, "Failed to set display brightness: %d\n", ret); - return ret; - } + mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 0x00ff); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x24); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, 0x00); - mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY, 0x24); - mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, 0x00); + mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 90); + mipi_dsi_dcs_set_display_on_multi(&dsi_ctx); - ret = mipi_dsi_dcs_exit_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to exit sleep mode: %d\n", ret); - return ret; - } - msleep(90); - - ret = mipi_dsi_dcs_set_display_on(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display on: %d\n", ret); - return ret; - } - - return 0; + return dsi_ctx.accum_err; } static int ebbg_ft8719_off(struct ebbg_ft8719 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dsi->mode_flags &= ~MIPI_DSI_MODE_LPM; - ret = mipi_dsi_dcs_set_display_off(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display off: %d\n", ret); - return ret; - } - usleep_range(10000, 11000); + mipi_dsi_dcs_set_display_off_multi(&dsi_ctx); + mipi_dsi_usleep_range(&dsi_ctx, 10000, 11000); + mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 90); - ret = mipi_dsi_dcs_enter_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to enter sleep mode: %d\n", ret); - return ret; - } - msleep(90); - - return 0; + return dsi_ctx.accum_err; } static int ebbg_ft8719_prepare(struct drm_panel *panel) { struct ebbg_ft8719 *ctx = to_ebbg_ft8719(panel); - struct device *dev = &ctx->dsi->dev; int ret; ret = regulator_bulk_enable(ARRAY_SIZE(ctx->supplies), ctx->supplies); @@ -126,7 +100,6 @@ static int ebbg_ft8719_prepare(struct drm_panel *panel) ret = ebbg_ft8719_on(ctx); if (ret < 0) { - dev_err(dev, "Failed to initialize panel: %d\n", ret); gpiod_set_value_cansleep(ctx->reset_gpio, 1); return ret; } @@ -137,18 +110,10 @@ static int ebbg_ft8719_prepare(struct drm_panel *panel) static int ebbg_ft8719_unprepare(struct drm_panel *panel) { struct ebbg_ft8719 *ctx = to_ebbg_ft8719(panel); - struct device *dev = &ctx->dsi->dev; - int ret; - - ret = ebbg_ft8719_off(ctx); - if (ret < 0) - dev_err(dev, "Failed to un-initialize panel: %d\n", ret); + ebbg_ft8719_off(ctx); gpiod_set_value_cansleep(ctx->reset_gpio, 1); - - ret = regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies); - if (ret) - dev_err(panel->dev, "Failed to disable regulators: %d\n", ret); + regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies); return 0; } From 8025f23728e9bc817495d48a04954cdee27462bd Mon Sep 17 00:00:00 2001 From: Tejas Vipin Date: Tue, 7 Jan 2025 14:55:10 +0530 Subject: [PATCH 0055/1443] drm/panel: xinpeng-xpp055c272: transition to mipi_dsi wrapped functions Changes the xinpeng-xpp055c272 panel to use multi style functions for improved error handling. Reviewed-by: Neil Armstrong Signed-off-by: Tejas Vipin Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250107092510.174089-1-tejasvipin76@gmail.com --- .../gpu/drm/panel/panel-xinpeng-xpp055c272.c | 174 ++++++++---------- 1 file changed, 78 insertions(+), 96 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c b/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c index 22a14006765e..2b91414c2829 100644 --- a/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c +++ b/drivers/gpu/drm/panel/panel-xinpeng-xpp055c272.c @@ -59,91 +59,80 @@ static inline struct xpp055c272 *panel_to_xpp055c272(struct drm_panel *panel) return container_of(panel, struct xpp055c272, panel); } -static int xpp055c272_init_sequence(struct xpp055c272 *ctx) +static void xpp055c272_init_sequence(struct mipi_dsi_multi_context *dsi_ctx) { - struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); - struct device *dev = ctx->dev; - /* * Init sequence was supplied by the panel vendor without much * documentation. */ - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETEXTC, 0xf1, 0x12, 0x83); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETMIPI, - 0x33, 0x81, 0x05, 0xf9, 0x0e, 0x0e, 0x00, 0x00, - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x44, 0x25, - 0x00, 0x91, 0x0a, 0x00, 0x00, 0x02, 0x4f, 0x01, - 0x00, 0x00, 0x37); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETPOWER_EXT, 0x25); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETPCR, 0x02, 0x11, 0x00); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETRGBIF, - 0x0c, 0x10, 0x0a, 0x50, 0x03, 0xff, 0x00, 0x00, - 0x00, 0x00); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETSCR, - 0x73, 0x73, 0x50, 0x50, 0x00, 0x00, 0x08, 0x70, - 0x00); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETVDC, 0x46); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETPANEL, 0x0b); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETCYC, 0x80); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETDISP, 0xc8, 0x12, 0x30); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETEQ, - 0x07, 0x07, 0x0B, 0x0B, 0x03, 0x0B, 0x00, 0x00, - 0x00, 0x00, 0xFF, 0x00, 0xC0, 0x10); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETPOWER, - 0x53, 0x00, 0x1e, 0x1e, 0x77, 0xe1, 0xcc, 0xdd, - 0x67, 0x77, 0x33, 0x33); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETECO, 0x00, 0x00, 0xff, - 0xff, 0x01, 0xff); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETBGP, 0x09, 0x09); - msleep(20); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETEXTC, 0xf1, 0x12, 0x83); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETMIPI, + 0x33, 0x81, 0x05, 0xf9, 0x0e, 0x0e, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x44, 0x25, + 0x00, 0x91, 0x0a, 0x00, 0x00, 0x02, 0x4f, 0x01, + 0x00, 0x00, 0x37); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETPOWER_EXT, 0x25); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETPCR, 0x02, 0x11, 0x00); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETRGBIF, + 0x0c, 0x10, 0x0a, 0x50, 0x03, 0xff, 0x00, 0x00, + 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETSCR, + 0x73, 0x73, 0x50, 0x50, 0x00, 0x00, 0x08, 0x70, + 0x00); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETVDC, 0x46); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETPANEL, 0x0b); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETCYC, 0x80); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETDISP, 0xc8, 0x12, 0x30); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETEQ, + 0x07, 0x07, 0x0b, 0x0b, 0x03, 0x0b, 0x00, 0x00, + 0x00, 0x00, 0xff, 0x00, 0xC0, 0x10); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETPOWER, + 0x53, 0x00, 0x1e, 0x1e, 0x77, 0xe1, 0xcc, 0xdd, + 0x67, 0x77, 0x33, 0x33); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETECO, 0x00, 0x00, 0xff, + 0xff, 0x01, 0xff); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETBGP, 0x09, 0x09); + mipi_dsi_msleep(dsi_ctx, 20); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETVCOM, 0x87, 0x95); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETGIP1, - 0xc2, 0x10, 0x05, 0x05, 0x10, 0x05, 0xa0, 0x12, - 0x31, 0x23, 0x3f, 0x81, 0x0a, 0xa0, 0x37, 0x18, - 0x00, 0x80, 0x01, 0x00, 0x00, 0x00, 0x00, 0x80, - 0x01, 0x00, 0x00, 0x00, 0x48, 0xf8, 0x86, 0x42, - 0x08, 0x88, 0x88, 0x80, 0x88, 0x88, 0x88, 0x58, - 0xf8, 0x87, 0x53, 0x18, 0x88, 0x88, 0x81, 0x88, - 0x88, 0x88, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETGIP2, - 0x00, 0x1a, 0x00, 0x00, 0x00, 0x00, 0x02, 0x00, - 0x00, 0x00, 0x00, 0x00, 0x1f, 0x88, 0x81, 0x35, - 0x78, 0x88, 0x88, 0x85, 0x88, 0x88, 0x88, 0x0f, - 0x88, 0x80, 0x24, 0x68, 0x88, 0x88, 0x84, 0x88, - 0x88, 0x88, 0x23, 0x10, 0x00, 0x00, 0x1c, 0x00, - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x30, 0x05, - 0xa0, 0x00, 0x00, 0x00, 0x00); - mipi_dsi_dcs_write_seq(dsi, XPP055C272_CMD_SETGAMMA, - 0x00, 0x06, 0x08, 0x2a, 0x31, 0x3f, 0x38, 0x36, - 0x07, 0x0c, 0x0d, 0x11, 0x13, 0x12, 0x13, 0x11, - 0x18, 0x00, 0x06, 0x08, 0x2a, 0x31, 0x3f, 0x38, - 0x36, 0x07, 0x0c, 0x0d, 0x11, 0x13, 0x12, 0x13, - 0x11, 0x18); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETVCOM, 0x87, 0x95); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETGIP1, + 0xc2, 0x10, 0x05, 0x05, 0x10, 0x05, 0xa0, 0x12, + 0x31, 0x23, 0x3f, 0x81, 0x0a, 0xa0, 0x37, 0x18, + 0x00, 0x80, 0x01, 0x00, 0x00, 0x00, 0x00, 0x80, + 0x01, 0x00, 0x00, 0x00, 0x48, 0xf8, 0x86, 0x42, + 0x08, 0x88, 0x88, 0x80, 0x88, 0x88, 0x88, 0x58, + 0xf8, 0x87, 0x53, 0x18, 0x88, 0x88, 0x81, 0x88, + 0x88, 0x88, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETGIP2, + 0x00, 0x1a, 0x00, 0x00, 0x00, 0x00, 0x02, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x1f, 0x88, 0x81, 0x35, + 0x78, 0x88, 0x88, 0x85, 0x88, 0x88, 0x88, 0x0f, + 0x88, 0x80, 0x24, 0x68, 0x88, 0x88, 0x84, 0x88, + 0x88, 0x88, 0x23, 0x10, 0x00, 0x00, 0x1c, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x30, 0x05, + 0xa0, 0x00, 0x00, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(dsi_ctx, XPP055C272_CMD_SETGAMMA, + 0x00, 0x06, 0x08, 0x2a, 0x31, 0x3f, 0x38, 0x36, + 0x07, 0x0c, 0x0d, 0x11, 0x13, 0x12, 0x13, 0x11, + 0x18, 0x00, 0x06, 0x08, 0x2a, 0x31, 0x3f, 0x38, + 0x36, 0x07, 0x0c, 0x0d, 0x11, 0x13, 0x12, 0x13, + 0x11, 0x18); - msleep(60); - - dev_dbg(dev, "Panel init sequence done\n"); - return 0; + mipi_dsi_msleep(dsi_ctx, 60); } static int xpp055c272_unprepare(struct drm_panel *panel) { struct xpp055c272 *ctx = panel_to_xpp055c272(panel); struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; - ret = mipi_dsi_dcs_set_display_off(dsi); - if (ret < 0) - dev_err(ctx->dev, "failed to set display off: %d\n", ret); - - mipi_dsi_dcs_enter_sleep_mode(dsi); - if (ret < 0) { - dev_err(ctx->dev, "failed to enter sleep mode: %d\n", ret); - return ret; - } + mipi_dsi_dcs_set_display_off_multi(&dsi_ctx); + mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx); + if (dsi_ctx.accum_err) + return dsi_ctx.accum_err; regulator_disable(ctx->iovcc); regulator_disable(ctx->vci); @@ -155,17 +144,19 @@ static int xpp055c272_prepare(struct drm_panel *panel) { struct xpp055c272 *ctx = panel_to_xpp055c272(panel); struct mipi_dsi_device *dsi = to_mipi_dsi_device(ctx->dev); - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dev_dbg(ctx->dev, "Resetting the panel\n"); - ret = regulator_enable(ctx->vci); - if (ret < 0) { - dev_err(ctx->dev, "Failed to enable vci supply: %d\n", ret); - return ret; + dsi_ctx.accum_err = regulator_enable(ctx->vci); + if (dsi_ctx.accum_err) { + dev_err(ctx->dev, "Failed to enable vci supply: %d\n", + dsi_ctx.accum_err); + return dsi_ctx.accum_err; } - ret = regulator_enable(ctx->iovcc); - if (ret < 0) { - dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", ret); + dsi_ctx.accum_err = regulator_enable(ctx->iovcc); + if (dsi_ctx.accum_err) { + dev_err(ctx->dev, "Failed to enable iovcc supply: %d\n", + dsi_ctx.accum_err); goto disable_vci; } @@ -177,26 +168,17 @@ static int xpp055c272_prepare(struct drm_panel *panel) /* T8: 20ms */ msleep(20); - ret = xpp055c272_init_sequence(ctx); - if (ret < 0) { - dev_err(ctx->dev, "Panel init sequence failed: %d\n", ret); - goto disable_iovcc; - } - - ret = mipi_dsi_dcs_exit_sleep_mode(dsi); - if (ret < 0) { - dev_err(ctx->dev, "Failed to exit sleep mode: %d\n", ret); - goto disable_iovcc; - } + xpp055c272_init_sequence(&dsi_ctx); + if (!dsi_ctx.accum_err) + dev_dbg(ctx->dev, "Panel init sequence done\n"); + mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx); /* T9: 120ms */ - msleep(120); + mipi_dsi_msleep(&dsi_ctx, 120); + mipi_dsi_dcs_set_display_on_multi(&dsi_ctx); - ret = mipi_dsi_dcs_set_display_on(dsi); - if (ret < 0) { - dev_err(ctx->dev, "Failed to set display on: %d\n", ret); + if (dsi_ctx.accum_err) goto disable_iovcc; - } msleep(50); @@ -206,7 +188,7 @@ static int xpp055c272_prepare(struct drm_panel *panel) regulator_disable(ctx->iovcc); disable_vci: regulator_disable(ctx->vci); - return ret; + return dsi_ctx.accum_err; } static const struct drm_display_mode default_mode = { From 0ca6d6058852857c628b479f1e7aad3386036bdb Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Wed, 25 Dec 2024 10:01:09 +0800 Subject: [PATCH 0056/1443] drm/panel-edp: Add B140UAN04.4 and MNE007QS3-7 The raw edid for B140UAN04.4 panel is: 00 ff ff ff ff ff ff 00 06 af b3 a7 00 00 00 00 0c 22 01 04 a5 1e 13 78 03 cb 55 91 57 5a 91 29 1c 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 f4 3c 80 b8 70 b0 24 40 10 10 3e 00 2d bc 10 00 00 18 00 00 00 0f 00 00 00 00 00 00 00 00 00 00 00 00 00 20 00 00 00 fe 00 41 55 4f 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe 00 42 31 34 30 55 41 4e 30 34 2e 34 20 0a 01 46 70 20 79 02 00 22 00 14 87 61 02 85 7f 07 b7 00 0f 80 0f 00 af 04 23 00 02 00 0d 00 25 01 09 87 61 02 87 61 02 28 3c 80 81 00 15 74 1a 00 00 03 01 28 3c 00 00 53 50 53 50 3c 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0b 90 The raw edid for MNE007QS3-7 panel is: 00 ff ff ff ff ff ff 00 0e 77 48 14 00 00 00 00 34 20 01 04 a5 1e 13 78 03 2c c5 94 5c 59 95 29 1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 ea 3d 80 c8 70 b0 2e 40 30 20 36 00 2e bd 10 00 00 1a 00 00 00 fd 00 28 3c 4b 4b 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43 53 4f 54 20 54 39 0a 20 20 20 20 20 00 00 00 fe 00 4d 4e 45 30 30 37 51 53 33 2d 37 0a 20 00 df Signed-off-by: Langyan Ye [dianders: adjusted sort ordering] Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20241225020109.1051449-1-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-edp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index f8511fe5fb0d..2080f400bb65 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1862,6 +1862,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('A', 'U', 'O', 0x73aa, &delay_200_500_e50, "B116XTN02.3"), EDP_PANEL_ENTRY('A', 'U', 'O', 0x8594, &delay_200_500_e50, "B133UAN01.0"), EDP_PANEL_ENTRY('A', 'U', 'O', 0xa199, &delay_200_500_e50, "B116XAN06.1"), + EDP_PANEL_ENTRY('A', 'U', 'O', 0xa7b3, &delay_200_500_e50, "B140UAN04.4"), EDP_PANEL_ENTRY('A', 'U', 'O', 0xc4b4, &delay_200_500_e50, "B116XAT04.1"), EDP_PANEL_ENTRY('A', 'U', 'O', 0xd497, &delay_200_500_e50, "B120XAN01.0"), EDP_PANEL_ENTRY('A', 'U', 'O', 0xf390, &delay_200_500_e50, "B140XTN07.7"), @@ -1954,6 +1955,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('C', 'S', 'W', 0x1100, &delay_200_500_e80_d50, "MNB601LS1-1"), EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50, "MNB601LS1-4"), + EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"), EDP_PANEL_ENTRY('H', 'K', 'C', 0x2d51, &delay_200_500_e200, "Unknown"), EDP_PANEL_ENTRY('H', 'K', 'C', 0x2d5b, &delay_200_500_e200, "MB116AN01"), From 29e93d0a04b6c546a2557486029219e5f6146e65 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Fri, 27 Dec 2024 18:19:13 +0800 Subject: [PATCH 0057/1443] drm/panel-edp: Add CSW MNB601LS1-3 Add support for the CSW MNB601LS1-3, pleace the EDID here for subsequent reference. 00 ff ff ff ff ff ff 00 0e 77 03 11 00 00 00 00 00 22 01 04 95 1a 0e 78 03 a1 35 9b 5e 58 91 25 1c 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 09 1e 56 dc 50 00 28 30 30 20 36 00 00 90 10 00 00 1a 00 00 00 fd 00 28 3c 30 30 08 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43 53 4f 54 20 54 39 0a 20 20 20 20 20 00 00 00 fe 00 4d 4e 42 36 30 31 4c 53 31 2d 33 0a 20 00 32 Signed-off-by: Langyan Ye Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20241227101913.1711071-1-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-edp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 2080f400bb65..2e1081caff93 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1954,6 +1954,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('C', 'S', 'O', 0x1200, &delay_200_500_e50_p2e200, "MNC207QS1-1"), EDP_PANEL_ENTRY('C', 'S', 'W', 0x1100, &delay_200_500_e80_d50, "MNB601LS1-1"), + EDP_PANEL_ENTRY('C', 'S', 'W', 0x1103, &delay_200_500_e80_d50, "MNB601LS1-3"), EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50, "MNB601LS1-4"), EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"), From 749b5b279e5636cdcef51e15d67b77162cca6caa Mon Sep 17 00:00:00 2001 From: Douglas Anderson Date: Thu, 9 Jan 2025 14:28:53 -0800 Subject: [PATCH 0058/1443] drm/panel-edp: Add Starry 116KHD024006 We have a few reports of sc7180-trogdor-pompom devices that have a panel in them that IDs as STA 0x0004 and has the following raw EDID: 00 ff ff ff ff ff ff 00 4e 81 04 00 00 00 00 00 10 20 01 04 a5 1a 0e 78 0a dc dd 96 5b 5b 91 28 1f 52 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 8e 1c 56 a0 50 00 1e 30 28 20 55 00 00 90 10 00 00 18 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fe 00 31 31 36 4b 48 44 30 32 34 30 30 36 0a 00 e6 We've been unable to locate a datasheet for this panel and our partner has not been responsive, but all Starry eDP datasheets that we can find agree on the same timing (delay_100_500_e200) so it should be safe to use that here instead of the super conservative timings. We'll still go a little extra conservative and allow `hpd_absent` of 200 instead of 100 because that won't add any real-world delay in most cases. We'll associate the string from the EDID ("116KHD024006") with this panel. Given that the ID is the suspicious value of 0x0004 it seems likely that Starry doesn't always update their IDs but the string will still work to differentiate if we ever need to in the future. Reviewed-by: Neil Armstrong Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250109142853.1.Ibcc3009933fd19507cc9c713ad0c99c7a9e4fe17@changeid --- drivers/gpu/drm/panel/panel-edp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 2e1081caff93..319ebc184c16 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1996,6 +1996,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('S', 'H', 'P', 0x154c, &delay_200_500_p2e100, "LQ116M1JW10"), EDP_PANEL_ENTRY('S', 'H', 'P', 0x1593, &delay_200_500_p2e100, "LQ134N1"), + EDP_PANEL_ENTRY('S', 'T', 'A', 0x0004, &delay_200_500_e200, "116KHD024006"), EDP_PANEL_ENTRY('S', 'T', 'A', 0x0100, &delay_100_500_e200, "2081116HHD028001-51D"), { /* sentinal */ } From 412576293cca1ec2e9c1532acb7a66e811dd8ceb Mon Sep 17 00:00:00 2001 From: Lizhi Hou Date: Mon, 13 Jan 2025 10:26:16 -0800 Subject: [PATCH 0059/1443] accel/amdxdna: Remove casting mailbox payload pointer The mailbox payload pointer is void __iomem *. Casting it to u32 * is incorrect and causes sparse warning. cast removes address space '__iomem' of expression Fixes: b87f920b9344 ("accel/amdxdna: Support hardware mailbox") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501130921.ktqwsMLH-lkp@intel.com/ Signed-off-by: Lizhi Hou Reviewed-by: Mario Limonciello Signed-off-by: Mario Limonciello Link: https://patchwork.freedesktop.org/patch/msgid/20250113182617.1256094-1-lizhi.hou@amd.com --- drivers/accel/amdxdna/aie2_ctx.c | 22 +++++++++---------- drivers/accel/amdxdna/aie2_error.c | 8 +++---- drivers/accel/amdxdna/aie2_message.c | 10 ++++----- drivers/accel/amdxdna/aie2_pci.h | 10 ++++----- drivers/accel/amdxdna/amdxdna_mailbox.c | 6 ++--- drivers/accel/amdxdna/amdxdna_mailbox.h | 2 +- .../accel/amdxdna/amdxdna_mailbox_helper.c | 6 ++--- .../accel/amdxdna/amdxdna_mailbox_helper.h | 2 +- 8 files changed, 32 insertions(+), 34 deletions(-) diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c index 5f43db02b240..5c69ae3c40f8 100644 --- a/drivers/accel/amdxdna/aie2_ctx.c +++ b/drivers/accel/amdxdna/aie2_ctx.c @@ -185,7 +185,7 @@ aie2_sched_notify(struct amdxdna_sched_job *job) } static int -aie2_sched_resp_handler(void *handle, const u32 *data, size_t size) +aie2_sched_resp_handler(void *handle, void __iomem *data, size_t size) { struct amdxdna_sched_job *job = handle; struct amdxdna_gem_obj *cmd_abo; @@ -203,7 +203,7 @@ aie2_sched_resp_handler(void *handle, const u32 *data, size_t size) goto out; } - status = *data; + status = readl(data); XDNA_DBG(job->hwctx->client->xdna, "Resp status 0x%x", status); if (status == AIE2_STATUS_SUCCESS) amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_COMPLETED); @@ -216,7 +216,7 @@ aie2_sched_resp_handler(void *handle, const u32 *data, size_t size) } static int -aie2_sched_nocmd_resp_handler(void *handle, const u32 *data, size_t size) +aie2_sched_nocmd_resp_handler(void *handle, void __iomem *data, size_t size) { struct amdxdna_sched_job *job = handle; u32 ret = 0; @@ -230,7 +230,7 @@ aie2_sched_nocmd_resp_handler(void *handle, const u32 *data, size_t size) goto out; } - status = *data; + status = readl(data); XDNA_DBG(job->hwctx->client->xdna, "Resp status 0x%x", status); out: @@ -239,14 +239,14 @@ aie2_sched_nocmd_resp_handler(void *handle, const u32 *data, size_t size) } static int -aie2_sched_cmdlist_resp_handler(void *handle, const u32 *data, size_t size) +aie2_sched_cmdlist_resp_handler(void *handle, void __iomem *data, size_t size) { struct amdxdna_sched_job *job = handle; struct amdxdna_gem_obj *cmd_abo; - struct cmd_chain_resp *resp; struct amdxdna_dev *xdna; u32 fail_cmd_status; u32 fail_cmd_idx; + u32 cmd_status; u32 ret = 0; cmd_abo = job->cmd_bo; @@ -256,17 +256,17 @@ aie2_sched_cmdlist_resp_handler(void *handle, const u32 *data, size_t size) goto out; } - resp = (struct cmd_chain_resp *)data; + cmd_status = readl(data + offsetof(struct cmd_chain_resp, status)); xdna = job->hwctx->client->xdna; - XDNA_DBG(xdna, "Status 0x%x", resp->status); - if (resp->status == AIE2_STATUS_SUCCESS) { + XDNA_DBG(xdna, "Status 0x%x", cmd_status); + if (cmd_status == AIE2_STATUS_SUCCESS) { amdxdna_cmd_set_state(cmd_abo, ERT_CMD_STATE_COMPLETED); goto out; } /* Slow path to handle error, read from ringbuf on BAR */ - fail_cmd_idx = resp->fail_cmd_idx; - fail_cmd_status = resp->fail_cmd_status; + fail_cmd_idx = readl(data + offsetof(struct cmd_chain_resp, fail_cmd_idx)); + fail_cmd_status = readl(data + offsetof(struct cmd_chain_resp, fail_cmd_status)); XDNA_DBG(xdna, "Failed cmd idx %d, status 0x%x", fail_cmd_idx, fail_cmd_status); diff --git a/drivers/accel/amdxdna/aie2_error.c b/drivers/accel/amdxdna/aie2_error.c index b1defaa8513b..5ee905632a39 100644 --- a/drivers/accel/amdxdna/aie2_error.c +++ b/drivers/accel/amdxdna/aie2_error.c @@ -209,16 +209,14 @@ static u32 aie2_error_backtrack(struct amdxdna_dev_hdl *ndev, void *err_info, u3 return err_col; } -static int aie2_error_async_cb(void *handle, const u32 *data, size_t size) +static int aie2_error_async_cb(void *handle, void __iomem *data, size_t size) { - struct async_event_msg_resp *resp; struct async_event *e = handle; if (data) { - resp = (struct async_event_msg_resp *)data; - e->resp.type = resp->type; + e->resp.type = readl(data + offsetof(struct async_event_msg_resp, type)); wmb(); /* Update status in the end, so that no lock for here */ - e->resp.status = resp->status; + e->resp.status = readl(data + offsetof(struct async_event_msg_resp, status)); } queue_work(e->wq, &e->work); return 0; diff --git a/drivers/accel/amdxdna/aie2_message.c b/drivers/accel/amdxdna/aie2_message.c index 9e2c9a44f76a..bf4219e32cc1 100644 --- a/drivers/accel/amdxdna/aie2_message.c +++ b/drivers/accel/amdxdna/aie2_message.c @@ -356,7 +356,7 @@ int aie2_query_status(struct amdxdna_dev_hdl *ndev, char __user *buf, } int aie2_register_asyn_event_msg(struct amdxdna_dev_hdl *ndev, dma_addr_t addr, u32 size, - void *handle, int (*cb)(void*, const u32 *, size_t)) + void *handle, int (*cb)(void*, void __iomem *, size_t)) { struct async_event_msg_req req = { 0 }; struct xdna_mailbox_msg msg = { @@ -435,7 +435,7 @@ int aie2_config_cu(struct amdxdna_hwctx *hwctx) } int aie2_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)) + int (*notify_cb)(void *, void __iomem *, size_t)) { struct mailbox_channel *chann = hwctx->priv->mbox_chann; struct amdxdna_dev *xdna = hwctx->client->xdna; @@ -640,7 +640,7 @@ aie2_cmd_op_to_msg_op(u32 op) int aie2_cmdlist_multi_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)) + int (*notify_cb)(void *, void __iomem *, size_t)) { struct amdxdna_gem_obj *cmdbuf_abo = aie2_cmdlist_get_cmd_buf(job); struct mailbox_channel *chann = hwctx->priv->mbox_chann; @@ -705,7 +705,7 @@ int aie2_cmdlist_multi_execbuf(struct amdxdna_hwctx *hwctx, int aie2_cmdlist_single_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)) + int (*notify_cb)(void *, void __iomem *, size_t)) { struct amdxdna_gem_obj *cmdbuf_abo = aie2_cmdlist_get_cmd_buf(job); struct mailbox_channel *chann = hwctx->priv->mbox_chann; @@ -740,7 +740,7 @@ int aie2_cmdlist_single_execbuf(struct amdxdna_hwctx *hwctx, } int aie2_sync_bo(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)) + int (*notify_cb)(void *, void __iomem *, size_t)) { struct mailbox_channel *chann = hwctx->priv->mbox_chann; struct amdxdna_gem_obj *abo = to_xdna_obj(job->bos[0]); diff --git a/drivers/accel/amdxdna/aie2_pci.h b/drivers/accel/amdxdna/aie2_pci.h index f2d95531ddc2..385914840eaa 100644 --- a/drivers/accel/amdxdna/aie2_pci.h +++ b/drivers/accel/amdxdna/aie2_pci.h @@ -271,18 +271,18 @@ int aie2_destroy_context(struct amdxdna_dev_hdl *ndev, struct amdxdna_hwctx *hwc int aie2_map_host_buf(struct amdxdna_dev_hdl *ndev, u32 context_id, u64 addr, u64 size); int aie2_query_status(struct amdxdna_dev_hdl *ndev, char __user *buf, u32 size, u32 *cols_filled); int aie2_register_asyn_event_msg(struct amdxdna_dev_hdl *ndev, dma_addr_t addr, u32 size, - void *handle, int (*cb)(void*, const u32 *, size_t)); + void *handle, int (*cb)(void*, void __iomem *, size_t)); int aie2_config_cu(struct amdxdna_hwctx *hwctx); int aie2_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)); + int (*notify_cb)(void *, void __iomem *, size_t)); int aie2_cmdlist_single_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)); + int (*notify_cb)(void *, void __iomem *, size_t)); int aie2_cmdlist_multi_execbuf(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)); + int (*notify_cb)(void *, void __iomem *, size_t)); int aie2_sync_bo(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, - int (*notify_cb)(void *, const u32 *, size_t)); + int (*notify_cb)(void *, void __iomem *, size_t)); /* aie2_hwctx.c */ int aie2_hwctx_init(struct amdxdna_hwctx *hwctx); diff --git a/drivers/accel/amdxdna/amdxdna_mailbox.c b/drivers/accel/amdxdna/amdxdna_mailbox.c index 814b16bb1953..de7bf0fb4594 100644 --- a/drivers/accel/amdxdna/amdxdna_mailbox.c +++ b/drivers/accel/amdxdna/amdxdna_mailbox.c @@ -90,7 +90,7 @@ struct mailbox_pkg { struct mailbox_msg { void *handle; - int (*notify_cb)(void *handle, const u32 *data, size_t size); + int (*notify_cb)(void *handle, void __iomem *data, size_t size); size_t pkg_size; /* package size in bytes */ struct mailbox_pkg pkg; }; @@ -243,7 +243,7 @@ mailbox_send_msg(struct mailbox_channel *mb_chann, struct mailbox_msg *mb_msg) static int mailbox_get_resp(struct mailbox_channel *mb_chann, struct xdna_msg_header *header, - void *data) + void __iomem *data) { struct mailbox_msg *mb_msg; int msg_id; @@ -331,7 +331,7 @@ static int mailbox_get_msg(struct mailbox_channel *mb_chann) memcpy_fromio((u32 *)&header + 1, read_addr, rest); read_addr += rest; - ret = mailbox_get_resp(mb_chann, &header, (u32 *)read_addr); + ret = mailbox_get_resp(mb_chann, &header, read_addr); mailbox_set_headptr(mb_chann, head + msg_size); /* After update head, it can equal to ringbuf_size. This is expected. */ diff --git a/drivers/accel/amdxdna/amdxdna_mailbox.h b/drivers/accel/amdxdna/amdxdna_mailbox.h index 57954c303bdd..ea367f2fb738 100644 --- a/drivers/accel/amdxdna/amdxdna_mailbox.h +++ b/drivers/accel/amdxdna/amdxdna_mailbox.h @@ -25,7 +25,7 @@ struct mailbox_channel; struct xdna_mailbox_msg { u32 opcode; void *handle; - int (*notify_cb)(void *handle, const u32 *data, size_t size); + int (*notify_cb)(void *handle, void __iomem *data, size_t size); u8 *send_data; size_t send_size; }; diff --git a/drivers/accel/amdxdna/amdxdna_mailbox_helper.c b/drivers/accel/amdxdna/amdxdna_mailbox_helper.c index 5139a9c96a91..6d0c24513476 100644 --- a/drivers/accel/amdxdna/amdxdna_mailbox_helper.c +++ b/drivers/accel/amdxdna/amdxdna_mailbox_helper.c @@ -16,7 +16,7 @@ #include "amdxdna_mailbox_helper.h" #include "amdxdna_pci_drv.h" -int xdna_msg_cb(void *handle, const u32 *data, size_t size) +int xdna_msg_cb(void *handle, void __iomem *data, size_t size) { struct xdna_notify *cb_arg = handle; int ret; @@ -29,9 +29,9 @@ int xdna_msg_cb(void *handle, const u32 *data, size_t size) goto out; } + memcpy_fromio(cb_arg->data, data, cb_arg->size); print_hex_dump_debug("resp data: ", DUMP_PREFIX_OFFSET, - 16, 4, data, cb_arg->size, true); - memcpy(cb_arg->data, data, cb_arg->size); + 16, 4, cb_arg->data, cb_arg->size, true); out: ret = cb_arg->error; complete(&cb_arg->comp); diff --git a/drivers/accel/amdxdna/amdxdna_mailbox_helper.h b/drivers/accel/amdxdna/amdxdna_mailbox_helper.h index 23e1317b79fe..710ff8873d61 100644 --- a/drivers/accel/amdxdna/amdxdna_mailbox_helper.h +++ b/drivers/accel/amdxdna/amdxdna_mailbox_helper.h @@ -35,7 +35,7 @@ struct xdna_notify { .notify_cb = xdna_msg_cb, \ } -int xdna_msg_cb(void *handle, const u32 *data, size_t size); +int xdna_msg_cb(void *handle, void __iomem *data, size_t size); int xdna_send_msg_wait(struct amdxdna_dev *xdna, struct mailbox_channel *chann, struct xdna_mailbox_msg *msg); From b3dff598e72f58888f39588d621913eddb9f9313 Mon Sep 17 00:00:00 2001 From: Lizhi Hou Date: Mon, 13 Jan 2025 10:26:17 -0800 Subject: [PATCH 0060/1443] accel/amdxdna: Declare sched_ops as static Fix sparse warning: symbol 'sched_ops' was not declared. Should it be static? Fixes: aac243092b70 ("accel/amdxdna: Add command execution") Signed-off-by: Lizhi Hou Reviewed-by: Mario Limonciello Signed-off-by: Mario Limonciello Link: https://patchwork.freedesktop.org/patch/msgid/20250113182617.1256094-2-lizhi.hou@amd.com --- drivers/accel/amdxdna/aie2_ctx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c index 5c69ae3c40f8..2799729801f6 100644 --- a/drivers/accel/amdxdna/aie2_ctx.c +++ b/drivers/accel/amdxdna/aie2_ctx.c @@ -361,7 +361,7 @@ aie2_sched_job_timedout(struct drm_sched_job *sched_job) return DRM_GPU_SCHED_STAT_NOMINAL; } -const struct drm_sched_backend_ops sched_ops = { +static const struct drm_sched_backend_ops sched_ops = { .run_job = aie2_sched_job_run, .free_job = aie2_sched_job_free, .timedout_job = aie2_sched_job_timedout, From c26f22dac3449d8a687237cdfc59a6445eb8f75a Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Fri, 10 Jan 2025 18:15:39 -0800 Subject: [PATCH 0061/1443] drm/xe/oa: Add missing VISACTL mux registers Add missing VISACTL mux registers required for some OA config's (e.g. RenderPipeCtrl). Fixes: cdf02fe1a94a ("drm/xe/oa/uapi: Add/remove OA config perf ops") Cc: stable@vger.kernel.org Signed-off-by: Ashutosh Dixit Reviewed-by: Umesh Nerlige Ramappa Link: https://patchwork.freedesktop.org/patch/msgid/20250111021539.2920346-1-ashutosh.dixit@intel.com --- drivers/gpu/drm/xe/xe_oa.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 4e00a77289c5..eeb96b5f49e2 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -2163,6 +2163,7 @@ static const struct xe_mmio_range xe2_oa_mux_regs[] = { { .start = 0x5194, .end = 0x5194 }, /* SYS_MEM_LAT_MEASURE_MERTF_GRP_3D */ { .start = 0x8704, .end = 0x8704 }, /* LMEM_LAT_MEASURE_MCFG_GRP */ { .start = 0xB1BC, .end = 0xB1BC }, /* L3_BANK_LAT_MEASURE_LBCF_GFX */ + { .start = 0xD0E0, .end = 0xD0F4 }, /* VISACTL */ { .start = 0xE18C, .end = 0xE18C }, /* SAMPLER_MODE */ { .start = 0xE590, .end = 0xE590 }, /* TDL_LSC_LAT_MEASURE_TDL_GFX */ { .start = 0x13000, .end = 0x137FC }, /* PES_0_PESL0 - PES_63_UPPER_PESL3 */ From 6e11ce84c514f3ad8c8c766e1328bf49d80a0325 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Mon, 15 Jul 2024 11:38:57 +0200 Subject: [PATCH 0062/1443] drm/probe-helper: Call connector detect functions in single helper Move the logic to call the connector's detect helper into a single internal function. Reduces code dupliction. Signed-off-by: Thomas Zimmermann Acked-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20240715093936.793552-2-tzimmermann@suse.de --- drivers/gpu/drm/drm_probe_helper.c | 33 +++++++++++++++--------------- 1 file changed, 17 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index 080b55200614..7ba16323e7c2 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -338,10 +338,23 @@ void drm_kms_helper_poll_reschedule(struct drm_device *dev) } EXPORT_SYMBOL(drm_kms_helper_poll_reschedule); +static int detect_connector_status(struct drm_connector *connector, + struct drm_modeset_acquire_ctx *ctx, + bool force) +{ + const struct drm_connector_helper_funcs *funcs = connector->helper_private; + + if (funcs->detect_ctx) + return funcs->detect_ctx(connector, ctx, force); + else if (connector->funcs->detect) + return connector->funcs->detect(connector, force); + + return connector_status_connected; +} + static enum drm_connector_status drm_helper_probe_detect_ctx(struct drm_connector *connector, bool force) { - const struct drm_connector_helper_funcs *funcs = connector->helper_private; struct drm_modeset_acquire_ctx ctx; int ret; @@ -349,14 +362,8 @@ drm_helper_probe_detect_ctx(struct drm_connector *connector, bool force) retry: ret = drm_modeset_lock(&connector->dev->mode_config.connection_mutex, &ctx); - if (!ret) { - if (funcs->detect_ctx) - ret = funcs->detect_ctx(connector, &ctx, force); - else if (connector->funcs->detect) - ret = connector->funcs->detect(connector, force); - else - ret = connector_status_connected; - } + if (!ret) + ret = detect_connector_status(connector, &ctx, force); if (ret == -EDEADLK) { drm_modeset_backoff(&ctx); @@ -390,7 +397,6 @@ drm_helper_probe_detect(struct drm_connector *connector, struct drm_modeset_acquire_ctx *ctx, bool force) { - const struct drm_connector_helper_funcs *funcs = connector->helper_private; struct drm_device *dev = connector->dev; int ret; @@ -401,12 +407,7 @@ drm_helper_probe_detect(struct drm_connector *connector, if (ret) return ret; - if (funcs->detect_ctx) - ret = funcs->detect_ctx(connector, ctx, force); - else if (connector->funcs->detect) - ret = connector->funcs->detect(connector, force); - else - ret = connector_status_connected; + ret = detect_connector_status(connector, ctx, force); if (ret != connector->status) connector->epoch_counter += 1; From d160dc6f53914d729be7fcb7afbd0e9e6a3725b2 Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Fri, 10 Jan 2025 09:33:09 -0800 Subject: [PATCH 0063/1443] drm/xe: Add locks in gtidle code The update of the residency values needs to be protected by a lock to avoid multiple entrypoints, for example when multiple userspace clients read the sysfs file. Other in-kernel clients are going to be added to sample these values, making the problem worse. Protect those updates with a raw_spinlock so it can be called by future integration with perf pmu. Suggested-by: Lucas De Marchi Cc: Rodrigo Vivi Cc: Lucas De Marchi Signed-off-by: Vinay Belgaumkar Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20250110173308.2412232-2-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_gt_idle.c | 23 ++++++++++++++++++++--- drivers/gpu/drm/xe/xe_gt_idle.h | 1 + drivers/gpu/drm/xe/xe_gt_idle_types.h | 3 +++ 3 files changed, 24 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c index ffd3ba7f6656..fbbace7b0b12 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.c +++ b/drivers/gpu/drm/xe/xe_gt_idle.c @@ -69,6 +69,8 @@ static u64 get_residency_ms(struct xe_gt_idle *gtidle, u64 cur_residency) { u64 delta, overflow_residency, prev_residency; + lockdep_assert_held(>idle->lock); + overflow_residency = BIT_ULL(32); /* @@ -275,8 +277,21 @@ static ssize_t idle_status_show(struct device *dev, return sysfs_emit(buff, "%s\n", gt_idle_state_to_string(state)); } -static DEVICE_ATTR_RO(idle_status); +u64 xe_gt_idle_residency_msec(struct xe_gt_idle *gtidle) +{ + struct xe_guc_pc *pc = gtidle_to_pc(gtidle); + u64 residency; + unsigned long flags; + + raw_spin_lock_irqsave(>idle->lock, flags); + residency = get_residency_ms(gtidle, gtidle->idle_residency(pc)); + raw_spin_unlock_irqrestore(>idle->lock, flags); + + return residency; +} + +static DEVICE_ATTR_RO(idle_status); static ssize_t idle_residency_ms_show(struct device *dev, struct device_attribute *attr, char *buff) { @@ -285,10 +300,10 @@ static ssize_t idle_residency_ms_show(struct device *dev, u64 residency; xe_pm_runtime_get(pc_to_xe(pc)); - residency = gtidle->idle_residency(pc); + residency = xe_gt_idle_residency_msec(gtidle); xe_pm_runtime_put(pc_to_xe(pc)); - return sysfs_emit(buff, "%llu\n", get_residency_ms(gtidle, residency)); + return sysfs_emit(buff, "%llu\n", residency); } static DEVICE_ATTR_RO(idle_residency_ms); @@ -331,6 +346,8 @@ int xe_gt_idle_init(struct xe_gt_idle *gtidle) if (!kobj) return -ENOMEM; + raw_spin_lock_init(>idle->lock); + if (xe_gt_is_media_type(gt)) { snprintf(gtidle->name, sizeof(gtidle->name), "gt%d-mc", gt->info.id); gtidle->idle_residency = xe_guc_pc_mc6_residency; diff --git a/drivers/gpu/drm/xe/xe_gt_idle.h b/drivers/gpu/drm/xe/xe_gt_idle.h index 4455a6501cb0..591a01e181bc 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.h +++ b/drivers/gpu/drm/xe/xe_gt_idle.h @@ -17,5 +17,6 @@ void xe_gt_idle_disable_c6(struct xe_gt *gt); void xe_gt_idle_enable_pg(struct xe_gt *gt); void xe_gt_idle_disable_pg(struct xe_gt *gt); int xe_gt_idle_pg_print(struct xe_gt *gt, struct drm_printer *p); +u64 xe_gt_idle_residency_msec(struct xe_gt_idle *gtidle); #endif /* _XE_GT_IDLE_H_ */ diff --git a/drivers/gpu/drm/xe/xe_gt_idle_types.h b/drivers/gpu/drm/xe/xe_gt_idle_types.h index b8b297a3f884..a3667c567f8a 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle_types.h +++ b/drivers/gpu/drm/xe/xe_gt_idle_types.h @@ -6,6 +6,7 @@ #ifndef _XE_GT_IDLE_SYSFS_TYPES_H_ #define _XE_GT_IDLE_SYSFS_TYPES_H_ +#include #include struct xe_guc_pc; @@ -31,6 +32,8 @@ struct xe_gt_idle { u64 cur_residency; /** @prev_residency: previous residency counter */ u64 prev_residency; + /** @lock: Lock protecting idle residency counters */ + raw_spinlock_t lock; /** @idle_status: get the current idle state */ enum xe_gt_idle_state (*idle_status)(struct xe_guc_pc *pc); /** @idle_residency: get idle residency counter */ From 467dce3817bd2b62ccd6fcfd7aae76f242ac907e Mon Sep 17 00:00:00 2001 From: "Lin.Cao" Date: Thu, 26 Dec 2024 12:31:15 +0530 Subject: [PATCH 0064/1443] drm/buddy: fix issue that force_merge cannot free all roots If buddy manager have more than one roots and each root have sub-block need to be free. When drm_buddy_fini called, the first loop of force_merge will merge and free all of the sub block of first root, which offset is 0x0 and size is biggest(more than have of the mm size). In subsequent force_merge rounds, if we use 0 as start and use remaining mm size as end, the block of other roots will be skipped in __force_merge function. It will cause the other roots can not be freed. Solution: use roots' offset as the start could fix this issue. Signed-off-by: Lin.Cao Signed-off-by: Arunpravin Paneer Selvam Reviewed-by: Matthew Auld Link: https://patchwork.freedesktop.org/patch/msgid/20241226070116.309290-1-Arunpravin.PaneerSelvam@amd.com --- drivers/gpu/drm/drm_buddy.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c index 103c185bb1c8..ca42e6081d27 100644 --- a/drivers/gpu/drm/drm_buddy.c +++ b/drivers/gpu/drm/drm_buddy.c @@ -324,7 +324,7 @@ EXPORT_SYMBOL(drm_buddy_init); */ void drm_buddy_fini(struct drm_buddy *mm) { - u64 root_size, size; + u64 root_size, size, start; unsigned int order; int i; @@ -332,7 +332,8 @@ void drm_buddy_fini(struct drm_buddy *mm) for (i = 0; i < mm->n_roots; ++i) { order = ilog2(size) - ilog2(mm->chunk_size); - __force_merge(mm, 0, size, order); + start = drm_buddy_block_offset(mm->roots[i]); + __force_merge(mm, start, start + size, order); WARN_ON(!drm_buddy_block_is_free(mm->roots[i])); drm_block_free(mm, mm->roots[i]); From 8cb3a1e2b3502341f7e5253842a74e511c13e5c9 Mon Sep 17 00:00:00 2001 From: Arunpravin Paneer Selvam Date: Thu, 26 Dec 2024 12:31:16 +0530 Subject: [PATCH 0065/1443] drm/buddy: Add a testcase to verify the multiroot fini - Added a testcase to verify the multiroot force merge fini. - Added a new field in_use to track the mm freed status. v2:(Matthew) - Add kunit_fail_current_test() when WARN_ON is true. Signed-off-by: Arunpravin Paneer Selvam Signed-off-by: Lin.Cao Reviewed-by: Matthew Auld Link: https://patchwork.freedesktop.org/patch/msgid/20241226070116.309290-2-Arunpravin.PaneerSelvam@amd.com --- drivers/gpu/drm/drm_buddy.c | 6 +++++- drivers/gpu/drm/tests/drm_buddy_test.c | 29 ++++++++++++++++++-------- 2 files changed, 25 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c index ca42e6081d27..241c855f891f 100644 --- a/drivers/gpu/drm/drm_buddy.c +++ b/drivers/gpu/drm/drm_buddy.c @@ -3,6 +3,8 @@ * Copyright © 2021 Intel Corporation */ +#include + #include #include #include @@ -335,7 +337,9 @@ void drm_buddy_fini(struct drm_buddy *mm) start = drm_buddy_block_offset(mm->roots[i]); __force_merge(mm, start, start + size, order); - WARN_ON(!drm_buddy_block_is_free(mm->roots[i])); + if (WARN_ON(!drm_buddy_block_is_free(mm->roots[i]))) + kunit_fail_current_test("buddy_fini() root"); + drm_block_free(mm, mm->roots[i]); root_size = mm->chunk_size << order; diff --git a/drivers/gpu/drm/tests/drm_buddy_test.c b/drivers/gpu/drm/tests/drm_buddy_test.c index 9662c949d0e3..4b5818f9f2a9 100644 --- a/drivers/gpu/drm/tests/drm_buddy_test.c +++ b/drivers/gpu/drm/tests/drm_buddy_test.c @@ -385,17 +385,28 @@ static void drm_test_buddy_alloc_clear(struct kunit *test) drm_buddy_fini(&mm); /* - * Create a new mm with a non power-of-two size. Allocate a random size, free as - * cleared and then call fini. This will ensure the multi-root force merge during - * fini. + * Create a new mm with a non power-of-two size. Allocate a random size from each + * root, free as cleared and then call fini. This will ensure the multi-root + * force merge during fini. */ - mm_size = 12 * SZ_4K; - size = max(round_up(prandom_u32_state(&prng) % mm_size, ps), ps); + mm_size = (SZ_4K << max_order) + (SZ_4K << (max_order - 2)); + KUNIT_EXPECT_FALSE(test, drm_buddy_init(&mm, mm_size, ps)); - KUNIT_ASSERT_FALSE_MSG(test, drm_buddy_alloc_blocks(&mm, 0, mm_size, - size, ps, &allocated, - DRM_BUDDY_TOPDOWN_ALLOCATION), - "buddy_alloc hit an error size=%u\n", size); + KUNIT_EXPECT_EQ(test, mm.max_order, max_order); + KUNIT_ASSERT_FALSE_MSG(test, drm_buddy_alloc_blocks(&mm, 0, SZ_4K << max_order, + 4 * ps, ps, &allocated, + DRM_BUDDY_RANGE_ALLOCATION), + "buddy_alloc hit an error size=%lu\n", 4 * ps); + drm_buddy_free_list(&mm, &allocated, DRM_BUDDY_CLEARED); + KUNIT_ASSERT_FALSE_MSG(test, drm_buddy_alloc_blocks(&mm, 0, SZ_4K << max_order, + 2 * ps, ps, &allocated, + DRM_BUDDY_CLEAR_ALLOCATION), + "buddy_alloc hit an error size=%lu\n", 2 * ps); + drm_buddy_free_list(&mm, &allocated, DRM_BUDDY_CLEARED); + KUNIT_ASSERT_FALSE_MSG(test, drm_buddy_alloc_blocks(&mm, SZ_4K << max_order, mm_size, + ps, ps, &allocated, + DRM_BUDDY_RANGE_ALLOCATION), + "buddy_alloc hit an error size=%lu\n", ps); drm_buddy_free_list(&mm, &allocated, DRM_BUDDY_CLEARED); drm_buddy_fini(&mm); } From 3c47c2ccd5a29c78780ccfd0227a805f3873ab1c Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 14 Jan 2025 07:35:08 -0800 Subject: [PATCH 0066/1443] nvmet: fix rw control endian access Fixes: 3ec5c62cfcf060e ("nvmet: handle rw's limited retry flag") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501142128.WexgyMTv-lkp@intel.com/ Cc: Guixin Liu Signed-off-by: Keith Busch --- drivers/nvme/target/io-cmd-bdev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/target/io-cmd-bdev.c b/drivers/nvme/target/io-cmd-bdev.c index 6380b60fd490..2b09b2c69857 100644 --- a/drivers/nvme/target/io-cmd-bdev.c +++ b/drivers/nvme/target/io-cmd-bdev.c @@ -272,7 +272,7 @@ static void nvmet_bdev_execute_rw(struct nvmet_req *req) iter_flags = SG_MITER_FROM_SG; } - if (req->cmd->rw.control & NVME_RW_LR) + if (req->cmd->rw.control & cpu_to_le16(NVME_RW_LR)) opf |= REQ_FAILFAST_DEV; if (is_pci_p2pdma_page(sg_page(req->sg))) From c5e3306a424b52e38ad2c28c7f3399fcd03e383d Mon Sep 17 00:00:00 2001 From: Simona Vetter Date: Wed, 8 Jan 2025 18:24:16 +0100 Subject: [PATCH 0067/1443] drm/atomic: clarify the rules around drm_atomic_state->allow_modeset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit msm is automagically upgrading normal commits to full modesets, and that's a big no-no: - for one this results in full on->off->on transitions on all these crtc, at least if you're using the usual helpers. Which seems to be the case, and is breaking uapi - further even if the ctm change itself would not result in flicker, this can hide modesets for other reasons. Which again breaks the uapi v2: I forgot the case of adding unrelated crtc state. Add that case and link to the existing kerneldoc explainers. This has come up in an irc discussion with Manasi and Ville about intel's bigjoiner mode. Also cc everyone involved in the msm irc discussion, more people joined after I sent out v1. v3: Wording polish from Pekka and Thomas Acked-by: Pekka Paalanen Acked-by: Dmitry Baryshkov Cc: Maarten Lankhorst Cc: Maxime Ripard Cc: Thomas Zimmermann Cc: David Airlie Cc: Daniel Vetter Cc: Pekka Paalanen Cc: Rob Clark Cc: Simon Ser Cc: Manasi Navare Cc: Ville Syrjälä Cc: Abhinav Kumar Cc: Dmitry Baryshkov Signed-off-by: Simona Vetter Signed-off-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250108172417.160831-1-simona.vetter@ffwll.ch --- include/drm/drm_atomic.h | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) diff --git a/include/drm/drm_atomic.h b/include/drm/drm_atomic.h index 31ca88deb10d..1ded9a8d4e84 100644 --- a/include/drm/drm_atomic.h +++ b/include/drm/drm_atomic.h @@ -376,8 +376,27 @@ struct drm_atomic_state { * * Allow full modeset. This is used by the ATOMIC IOCTL handler to * implement the DRM_MODE_ATOMIC_ALLOW_MODESET flag. Drivers should - * never consult this flag, instead looking at the output of - * drm_atomic_crtc_needs_modeset(). + * generally not consult this flag, but instead look at the output of + * drm_atomic_crtc_needs_modeset(). The detailed rules are: + * + * - Drivers must not consult @allow_modeset in the atomic commit path. + * Use drm_atomic_crtc_needs_modeset() instead. + * + * - Drivers must consult @allow_modeset before adding unrelated struct + * drm_crtc_state to this commit by calling + * drm_atomic_get_crtc_state(). See also the warning in the + * documentation for that function. + * + * - Drivers must never change this flag, it is under the exclusive + * control of userspace. + * + * - Drivers may consult @allow_modeset in the atomic check path, if + * they have the choice between an optimal hardware configuration + * which requires a modeset, and a less optimal configuration which + * can be committed without a modeset. An example would be suboptimal + * scanout FIFO allocation resulting in increased idle power + * consumption. This allows userspace to avoid flickering and delays + * for the normal composition loop at reasonable cost. */ bool allow_modeset : 1; /** From dc4afc0de9654f88676d77094a38f9451d519011 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Mon, 13 Jan 2025 12:47:41 -0300 Subject: [PATCH 0068/1443] drm/v3d: Remove `v3d->cpu_job` MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CPU jobs, like Cache Clean jobs, execute synchronously once the DRM scheduler starts running them. Consequently, a global `v3d->cpu_job` variable is unnecessary, as everything is managed within the `v3d_cpu_job_run()` function. This commit removes the `v3d->cpu_job` pointer, as it is not needed. Signed-off-by: Maíra Canal Reviewed-by: Jose Maria Casanova Crespo Link: https://patchwork.freedesktop.org/patch/msgid/20250113154741.67520-2-mcanal@igalia.com --- drivers/gpu/drm/v3d/v3d_drv.h | 1 - drivers/gpu/drm/v3d/v3d_sched.c | 2 -- 2 files changed, 3 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h index dc1cfe2e14be..9deaefa0f95b 100644 --- a/drivers/gpu/drm/v3d/v3d_drv.h +++ b/drivers/gpu/drm/v3d/v3d_drv.h @@ -147,7 +147,6 @@ struct v3d_dev { struct v3d_render_job *render_job; struct v3d_tfu_job *tfu_job; struct v3d_csd_job *csd_job; - struct v3d_cpu_job *cpu_job; struct v3d_queue_state queue[V3D_MAX_QUEUES]; diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c index da08ddb01d21..961465128d80 100644 --- a/drivers/gpu/drm/v3d/v3d_sched.c +++ b/drivers/gpu/drm/v3d/v3d_sched.c @@ -656,8 +656,6 @@ v3d_cpu_job_run(struct drm_sched_job *sched_job) struct v3d_cpu_job *job = to_cpu_job(sched_job); struct v3d_dev *v3d = job->base.v3d; - v3d->cpu_job = job; - if (job->job_type >= ARRAY_SIZE(cpu_job_function)) { DRM_DEBUG_DRIVER("Unknown CPU job: %d\n", job->job_type); return NULL; From 3318ef9888d5b4f5c5a9473180fd0b16e9ef266d Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Mon, 6 Jan 2025 15:43:13 -0800 Subject: [PATCH 0069/1443] drm/xe: Remove unused "mmio_ext" code The "mmio_ext" and 'REG_EXT" code is currently unused on any existing platform. Going forward, this also isn't the design we want to use for any future platforms/features either, so we should just go ahead and remove the dead code to avoid confusion. mmio_ext was originally added in an attempt to hack around the early (mis)design of the Xe driver, which used xe_gt as the target for all register MMIO access, even those completely unrelated to the GT subunit of the hardware. With the introduction of commit 34953ee349dd ("drm/xe: Create dedicated xe_mmio structure") and its follow-up patches, that misdesign has been corrected and access to register MMIO regions specific to hardware units is now done through xe_mmio structures which encapsulate an iomap, region size, and some other metadata. Although all of the registers used by the driver today happen to fall within one specific PCI BAR region, and thus re-use a single device-wide iomap, there's no requirement that this stay true for future platforms or features. I.e., if a future platform adds a new 'foo' hardware unit that exists at a different area in the BAR, or even in a completely different BAR, then that would be handled by doing a separate iomap of that unit's register region and wrapping it in its own 'struct xe_mmio foo_regs' structure. The pointer to the new 'foo_regs' could be placed within the xe_device, xe_tile, xe_gt, etc., according to where the new hardware unit falls within the current hardware hierarchy. This effectively reverts the following commits, although parts of these commits had already vanished or changed with the earlier xe_mmio refactor work: - commit 399a13323f0d ("drm/xe: add 28-bit address support in struct xe_reg") - commit fdef72e02e20 ("drm/xe: add a flag to bypass multi-tile config from MTCFG reg") - commit 866b2b176434 ("drm/xe: add MMIO extension support flags") - commit ef29b390c734 ("drm/xe: map MMIO BAR according to the num of tiles in device desc") - commit a4e2f3a299ea ("drm/xe: refactor xe_mmio_probe_tiles to support MMIO extension") Cc: Lucas De Marchi Cc: Rodrigo Vivi Cc: Koby Elbaz Acked-by: Maciej Patelczyk Reviewed-by: Reviewed-by: Lucas De Marchi Reviewed-by: Stuart Summers Link: https://patchwork.freedesktop.org/patch/msgid/20250106234312.2986065-2-matthew.d.roper@intel.com Signed-off-by: Matt Roper --- drivers/gpu/drm/xe/regs/xe_reg_defs.h | 16 +---------- drivers/gpu/drm/xe/xe_device_types.h | 11 -------- drivers/gpu/drm/xe/xe_mmio.c | 39 --------------------------- drivers/gpu/drm/xe/xe_pci.c | 3 --- drivers/gpu/drm/xe/xe_pci_types.h | 2 -- 5 files changed, 1 insertion(+), 70 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_reg_defs.h b/drivers/gpu/drm/xe/regs/xe_reg_defs.h index 0eedd6c26b1b..89716172fbb8 100644 --- a/drivers/gpu/drm/xe/regs/xe_reg_defs.h +++ b/drivers/gpu/drm/xe/regs/xe_reg_defs.h @@ -21,7 +21,7 @@ struct xe_reg { union { struct { /** @addr: address */ - u32 addr:28; + u32 addr:22; /** * @masked: register is "masked", with upper 16bits used * to identify the bits that are updated on the lower @@ -41,10 +41,6 @@ struct xe_reg { * @vf: register is accessible from the Virtual Function. */ u32 vf:1; - /** - * @ext: access MMIO extension space for current register. - */ - u32 ext:1; }; /** @raw: Raw value with both address and options */ u32 raw; @@ -111,16 +107,6 @@ struct xe_reg_mcr { */ #define XE_REG(r_, ...) ((const struct xe_reg)XE_REG_INITIALIZER(r_, ##__VA_ARGS__)) -/** - * XE_REG_EXT - Create a struct xe_reg from extension offset and additional - * flags - * @r_: Register extension offset - * @...: Additional options like access mode. See struct xe_reg for available - * options. - */ -#define XE_REG_EXT(r_, ...) \ - ((const struct xe_reg)XE_REG_INITIALIZER(r_, ##__VA_ARGS__, .ext = 1)) - /** * XE_REG_MCR - Create a struct xe_reg_mcr from offset and additional flags * @r_: Register offset diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 8a7b15972413..16ebb2859877 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -186,13 +186,6 @@ struct xe_tile { */ struct xe_mmio mmio; - /** - * @mmio_ext: MMIO-extension info for a tile. - * - * Each tile has its own additional 256MB (28-bit) MMIO-extension space. - */ - struct xe_mmio mmio_ext; - /** @mem: memory management info for tile */ struct { /** @@ -263,8 +256,6 @@ struct xe_device { const char *graphics_name; /** @info.media_name: media IP name */ const char *media_name; - /** @info.tile_mmio_ext_size: size of MMIO extension space, per-tile */ - u32 tile_mmio_ext_size; /** @info.graphics_verx100: graphics IP version */ u32 graphics_verx100; /** @info.media_verx100: media IP version */ @@ -314,8 +305,6 @@ struct xe_device { u8 has_heci_gscfi:1; /** @info.has_llc: Device has a shared CPU+GPU last level cache */ u8 has_llc:1; - /** @info.has_mmio_ext: Device has extra MMIO address range */ - u8 has_mmio_ext:1; /** @info.has_range_tlb_invalidation: Has range based TLB invalidations */ u8 has_range_tlb_invalidation:1; /** @info.has_sriov: Supports SR-IOV */ diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c index a48f239cad1c..d321a21aacf0 100644 --- a/drivers/gpu/drm/xe/xe_mmio.c +++ b/drivers/gpu/drm/xe/xe_mmio.c @@ -103,50 +103,11 @@ static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) } } -/* - * On top of all the multi-tile MMIO space there can be a platform-dependent - * extension for each tile, resulting in a layout like below: - * - * .----------------------. <- ext_base + tile_count * tile_mmio_ext_size - * | .... | - * |----------------------| <- ext_base + 2 * tile_mmio_ext_size - * | tile1->mmio_ext.regs | - * |----------------------| <- ext_base + 1 * tile_mmio_ext_size - * | tile0->mmio_ext.regs | - * |======================| <- ext_base = tile_count * tile_mmio_size - * | | - * | mmio.regs | - * | | - * '----------------------' <- 0MB - * - * Set up the tile[]->mmio_ext pointers/sizes. - */ -static void mmio_extension_setup(struct xe_device *xe, size_t tile_mmio_size, - size_t tile_mmio_ext_size) -{ - struct xe_tile *tile; - void __iomem *regs; - u8 id; - - if (!xe->info.has_mmio_ext) - return; - - regs = xe->mmio.regs + tile_mmio_size * xe->info.tile_count; - for_each_tile(tile, xe, id) { - tile->mmio_ext.regs_size = tile_mmio_ext_size; - tile->mmio_ext.regs = regs; - tile->mmio_ext.tile = tile; - regs += tile_mmio_ext_size; - } -} - int xe_mmio_probe_tiles(struct xe_device *xe) { size_t tile_mmio_size = SZ_16M; - size_t tile_mmio_ext_size = xe->info.tile_mmio_ext_size; mmio_multi_tile_setup(xe, tile_mmio_size); - mmio_extension_setup(xe, tile_mmio_size, tile_mmio_ext_size); return devm_add_action_or_reset(xe->drm.dev, tiles_fini, xe); } diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 39be74848e44..48d1c81d441e 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -61,7 +61,6 @@ struct xe_device_desc { u8 has_heci_gscfi:1; u8 has_heci_cscfi:1; u8 has_llc:1; - u8 has_mmio_ext:1; u8 has_sriov:1; u8 skip_guc_pc:1; u8 skip_mtcfg:1; @@ -617,7 +616,6 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.has_heci_gscfi = desc->has_heci_gscfi; xe->info.has_heci_cscfi = desc->has_heci_cscfi; xe->info.has_llc = desc->has_llc; - xe->info.has_mmio_ext = desc->has_mmio_ext; xe->info.has_sriov = desc->has_sriov; xe->info.skip_guc_pc = desc->skip_guc_pc; xe->info.skip_mtcfg = desc->skip_mtcfg; @@ -677,7 +675,6 @@ static int xe_info_init(struct xe_device *xe, xe->info.graphics_name = graphics_desc->name; xe->info.media_name = media_desc ? media_desc->name : "none"; - xe->info.tile_mmio_ext_size = graphics_desc->tile_mmio_ext_size; xe->info.dma_mask_size = graphics_desc->dma_mask_size; xe->info.vram_flags = graphics_desc->vram_flags; diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index 79b0f80376a4..873efec5cdee 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -20,8 +20,6 @@ struct xe_graphics_desc { u64 hw_engine_mask; /* hardware engines provided by graphics IP */ - u32 tile_mmio_ext_size; /* size of MMIO extension space, per-tile */ - u8 max_remote_tiles:2; u8 has_asid:1; From 0af944f0e3082ff517958b1cea76fb9b8cb379dd Mon Sep 17 00:00:00 2001 From: Oak Zeng Date: Fri, 10 Jan 2025 16:01:37 -0500 Subject: [PATCH 0070/1443] drm/xe: Reject BO eviction if BO is bound to current VM MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is a follow up fix for https://patchwork.freedesktop.org/patch/msgid/20241203021929.1919730-1-oak.zeng@intel.com The overall goal is to fail vm_bind when there is memory pressure. See more details in the commit message of above patch. Abbove patch fixes the issue when user pass in a vm_id parameter during gem_create. If user doesn't pass in a vm_id during gem_create, above patch doesn't help. This patch further reject BO eviction (which could be triggered by bo validation) if BO is bound to the current VM. vm_bind could fail due to the eviction failure. The BO to VM reverse mapping structure is used to determine whether BO is bound to VM. v2: Move vm_bo definition from function scope to if(evict) clause (Thomas) Further constraint the condition by adding ctx->resv (Thomas) Add a short comment describe the change. Suggested-by: Thomas Hellström Signed-off-by: Oak Zeng Reviewed-by: Thomas Hellström Signed-off-by: Thomas Hellström Link: https://patchwork.freedesktop.org/patch/msgid/20250110210137.3181576-1-oak.zeng@intel.com --- drivers/gpu/drm/xe/xe_bo.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 3f5391d416d4..4f077c11e8e2 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -713,6 +713,21 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, goto out; } + /* Reject BO eviction if BO is bound to current VM. */ + if (evict && ctx->resv) { + struct drm_gpuvm_bo *vm_bo; + + drm_gem_for_each_gpuvm_bo(vm_bo, &bo->ttm.base) { + struct xe_vm *vm = gpuvm_to_vm(vm_bo->vm); + + if (xe_vm_resv(vm) == ctx->resv && + xe_vm_in_preempt_fence_mode(vm)) { + ret = -EBUSY; + goto out; + } + } + } + /* * Failed multi-hop where the old_mem is still marked as * TTM_PL_FLAG_TEMPORARY, should just be a dummy move. From 00728273bdf1001f8d2f7b65bc398000d7defe0b Mon Sep 17 00:00:00 2001 From: Matthew Auld Date: Wed, 15 Jan 2025 12:49:42 +0000 Subject: [PATCH 0071/1443] drm/tests/buddy: fix build with unused prng MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We no longer use the prng, which leads to the following warning: drivers/gpu/drm/tests/drm_buddy_test.c: In function ‘drm_test_buddy_alloc_clear’: drivers/gpu/drm/tests/drm_buddy_test.c:264:23: error: unused variable ‘prng’ [-Werror=unused-variable] 264 | DRM_RND_STATE(prng, random_seed); v2 (Thomas) - Add Closes links Reported-by: Jani Nikula Closes: https://lore.kernel.org/dri-devel/875xmggvcs.fsf@intel.com/ Reported-by: Thomas Hellström Closes: https://lore.kernel.org/dri-devel/3f816555e6913fdbcb254523f65c98088d70581b.camel@intel.com/ Fixes: 8cb3a1e2b350 ("drm/buddy: Add a testcase to verify the multiroot fini") Signed-off-by: Matthew Auld Cc: Arunpravin Paneer Selvam Cc: Christian König Reviewed-by: Thomas Hellström Reviewed-by: Arunpravin Paneer Selvam Signed-off-by: Arunpravin Paneer Selvam Link: https://patchwork.freedesktop.org/patch/msgid/20250115124941.155990-2-matthew.auld@intel.com --- drivers/gpu/drm/tests/drm_buddy_test.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/tests/drm_buddy_test.c b/drivers/gpu/drm/tests/drm_buddy_test.c index 4b5818f9f2a9..7a0e523651f0 100644 --- a/drivers/gpu/drm/tests/drm_buddy_test.c +++ b/drivers/gpu/drm/tests/drm_buddy_test.c @@ -261,7 +261,6 @@ static void drm_test_buddy_alloc_range_bias(struct kunit *test) static void drm_test_buddy_alloc_clear(struct kunit *test) { unsigned long n_pages, total, i = 0; - DRM_RND_STATE(prng, random_seed); const unsigned long ps = SZ_4K; struct drm_buddy_block *block; const int max_order = 12; From aaab5404b16f19b06c7d88787d7ba18d91eeb854 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Wed, 15 Jan 2025 09:50:52 -0500 Subject: [PATCH 0072/1443] drm/xe: Introduce GuC PC debugfs Allows the visualization of the current GuC power conservation status and policies. v2: Fix DCC msg (Vinay) v3: Simplify pc_get_state_string (Jonathan) Reviewed-by: Vinay Belgaumkar Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20250115145053.1142023-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_debugfs.c | 15 ++++++++ drivers/gpu/drm/xe/xe_guc_pc.c | 59 +++++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_guc_pc.h | 2 + 3 files changed, 76 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_guc_debugfs.c b/drivers/gpu/drm/xe/xe_guc_debugfs.c index 995b306aced7..0aff1d462bc0 100644 --- a/drivers/gpu/drm/xe/xe_guc_debugfs.c +++ b/drivers/gpu/drm/xe/xe_guc_debugfs.c @@ -13,6 +13,7 @@ #include "xe_guc.h" #include "xe_guc_ct.h" #include "xe_guc_log.h" +#include "xe_guc_pc.h" #include "xe_macros.h" #include "xe_pm.h" @@ -60,10 +61,24 @@ static int guc_ctb(struct seq_file *m, void *data) return 0; } +static int guc_pc(struct seq_file *m, void *data) +{ + struct xe_guc *guc = node_to_guc(m->private); + struct xe_device *xe = guc_to_xe(guc); + struct drm_printer p = drm_seq_file_printer(m); + + xe_pm_runtime_get(xe); + xe_guc_pc_print(&guc->pc, &p); + xe_pm_runtime_put(xe); + + return 0; +} + static const struct drm_info_list debugfs_list[] = { {"guc_info", guc_info, 0}, {"guc_log", guc_log, 0}, {"guc_ctb", guc_ctb, 0}, + {"guc_pc", guc_pc, 0}, }; void xe_guc_debugfs_register(struct xe_guc *guc, struct dentry *parent) diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index df7f130fb663..43f9617baba2 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -8,6 +8,7 @@ #include #include +#include #include #include "abi/guc_actions_slpc_abi.h" @@ -1131,3 +1132,61 @@ int xe_guc_pc_init(struct xe_guc_pc *pc) return devm_add_action_or_reset(xe->drm.dev, xe_guc_pc_fini_hw, pc); } + +static const char *pc_get_state_string(struct xe_guc_pc *pc) +{ + switch (slpc_shared_data_read(pc, header.global_state)) { + case SLPC_GLOBAL_STATE_NOT_RUNNING: + return "not running"; + case SLPC_GLOBAL_STATE_INITIALIZING: + return "initializing"; + case SLPC_GLOBAL_STATE_RESETTING: + return "resetting"; + case SLPC_GLOBAL_STATE_RUNNING: + return "running"; + case SLPC_GLOBAL_STATE_SHUTTING_DOWN: + return "shutting down"; + case SLPC_GLOBAL_STATE_ERROR: + return "error"; + default: + return "unknown"; + } +} + +/** + * xe_guc_pc_print - Print GuC's Power Conservation information for debug + * @pc: Xe_GuC_PC instance + * @p: drm_printer + */ +void xe_guc_pc_print(struct xe_guc_pc *pc, struct drm_printer *p) +{ + drm_printf(p, "SLPC Shared Data Header:\n"); + drm_printf(p, "\tSize: %x\n", slpc_shared_data_read(pc, header.size)); + drm_printf(p, "\tGlobal State: %s\n", pc_get_state_string(pc)); + + if (pc_action_query_task_state(pc)) + return; + + drm_printf(p, "\nSLPC Tasks Status:\n"); + drm_printf(p, "\tGTPERF enabled: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_GTPERF_TASK_ENABLED)); + drm_printf(p, "\tDCC enabled: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_DCC_TASK_ENABLED)); + drm_printf(p, "\tDCC in use: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_IN_DCC)); + drm_printf(p, "\tBalancer enabled: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_BALANCER_ENABLED)); + drm_printf(p, "\tIBC enabled: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_IBC_TASK_ENABLED)); + drm_printf(p, "\tBalancer IA LMT enabled: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_BALANCER_IA_LMT_ENABLED)); + drm_printf(p, "\tBalancer IA LMT active: %s\n", + str_yes_no(slpc_shared_data_read(pc, task_state_data.status) & + SLPC_BALANCER_IA_LMT_ACTIVE)); +} diff --git a/drivers/gpu/drm/xe/xe_guc_pc.h b/drivers/gpu/drm/xe/xe_guc_pc.h index 619f59cd633c..39102b79602f 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.h +++ b/drivers/gpu/drm/xe/xe_guc_pc.h @@ -10,6 +10,7 @@ struct xe_guc_pc; enum slpc_gucrc_mode; +struct drm_printer; int xe_guc_pc_init(struct xe_guc_pc *pc); int xe_guc_pc_start(struct xe_guc_pc *pc); @@ -17,6 +18,7 @@ int xe_guc_pc_stop(struct xe_guc_pc *pc); int xe_guc_pc_gucrc_disable(struct xe_guc_pc *pc); int xe_guc_pc_override_gucrc_mode(struct xe_guc_pc *pc, enum slpc_gucrc_mode mode); int xe_guc_pc_unset_gucrc_mode(struct xe_guc_pc *pc); +void xe_guc_pc_print(struct xe_guc_pc *pc, struct drm_printer *p); u32 xe_guc_pc_get_act_freq(struct xe_guc_pc *pc); int xe_guc_pc_get_cur_freq(struct xe_guc_pc *pc, u32 *freq); From 50554bf3e56dd0c78ef1eedb685d0ab36c9c9987 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Wed, 15 Jan 2025 09:50:53 -0500 Subject: [PATCH 0073/1443] drm/xe/lnl: Enable GuC SLPC DCC task Enable DCC (Duty Cycle Control) in Lunar Lake. DCC is the SLPC task that tries to keep the GT from operating inefficiently when thermally constrained. Although the recommendation is to enable it, LNL GuC is leaving it disabled by default on LNL. It would minimize the GT frequency oscillation on throttled scenarios, which could potentially reduce latencies. v2: Move set_policies call after wait for running state, so we ensure it is not overwritten. (Vinay) v3: Fix English in the commit message (Jonathan) v4: Also set disable to 0 so DCC can really get into effect. v5: Avoid lnl_ prefix (Vinay) v6: Finish renaming... Reviewed-by: Vinay Belgaumkar Reviewed-by: Jonathan Cavitt #v3 Link: https://patchwork.freedesktop.org/patch/msgid/20250115145053.1142023-2-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_pc.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 43f9617baba2..44b5211066ef 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -993,6 +993,27 @@ static int pc_init_freqs(struct xe_guc_pc *pc) return ret; } +static int slpc_enable_dcc(struct xe_guc_pc *pc) +{ + int ret; + + ret = pc_action_set_param(pc, SLPC_PARAM_TASK_ENABLE_DCC, 1); + if (ret) + return ret; + + return pc_action_set_param(pc, SLPC_PARAM_TASK_DISABLE_DCC, 0); +} + +static int slpc_set_policies(struct xe_guc_pc *pc) +{ + struct xe_device *xe = pc_to_xe(pc); + + if (xe->info.platform == XE_LUNARLAKE) + return slpc_enable_dcc(pc); + + return 0; +} + /** * xe_guc_pc_start - Start GuC's Power Conservation component * @pc: Xe_GuC_PC instance @@ -1037,6 +1058,10 @@ int xe_guc_pc_start(struct xe_guc_pc *pc) goto out; } + ret = slpc_set_policies(pc); + if (ret) + goto out; + ret = pc_init_freqs(pc); if (ret) goto out; From 11a64adcdbcc3028b96e440bc33fa76e2e825c10 Mon Sep 17 00:00:00 2001 From: Francois Dugast Date: Tue, 14 Jan 2025 12:38:53 -0800 Subject: [PATCH 0074/1443] drm/xe/xe3: Generate and store the L3 bank mask On Xe3, the register used to indicate which L3 banks are enabled on the system is a new one called MIRROR_L3BANK_ENABLE. Each bit represents one bank enabled in each node. Extend the existing topology code for Xe3 to read this register and generate the correct L3 bank mask, which can be read by user space throug the topology query. Bspec: 72573, 73439 Signed-off-by: Francois Dugast Signed-off-by: Matt Atwood Reviewed-by: Matt Roper Link: https://patchwork.freedesktop.org/patch/msgid/20250114203853.35055-1-matthew.s.atwood@intel.com Signed-off-by: Matt Roper --- drivers/gpu/drm/xe/regs/xe_gt_regs.h | 3 +++ drivers/gpu/drm/xe/xe_gt_topology.c | 16 +++++++++++++--- 2 files changed, 16 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h index b4283ac030f4..096859072396 100644 --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h @@ -221,6 +221,9 @@ #define MIRROR_FUSE1 XE_REG(0x911c) +#define MIRROR_L3BANK_ENABLE XE_REG(0x9130) +#define XE3_L3BANK_ENABLE REG_GENMASK(31, 0) + #define XELP_EU_ENABLE XE_REG(0x9134) /* "_DISABLE" on Xe_LP */ #define XELP_EU_MASK REG_GENMASK(7, 0) #define XELP_GT_SLICE_ENABLE XE_REG(0x9138) diff --git a/drivers/gpu/drm/xe/xe_gt_topology.c b/drivers/gpu/drm/xe/xe_gt_topology.c index df2042db7ee6..516c81e3b8dd 100644 --- a/drivers/gpu/drm/xe/xe_gt_topology.c +++ b/drivers/gpu/drm/xe/xe_gt_topology.c @@ -129,7 +129,8 @@ static void load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask) { struct xe_device *xe = gt_to_xe(gt); - u32 fuse3 = xe_mmio_read32(>->mmio, MIRROR_FUSE3); + struct xe_mmio *mmio = >->mmio; + u32 fuse3 = xe_mmio_read32(mmio, MIRROR_FUSE3); /* * PTL platforms with media version 30.00 do not provide proper values @@ -143,7 +144,16 @@ load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask) if (XE_WA(gt, no_media_l3)) return; - if (GRAPHICS_VER(xe) >= 20) { + if (GRAPHICS_VER(xe) >= 30) { + xe_l3_bank_mask_t per_node = {}; + u32 meml3_en = REG_FIELD_GET(XE2_NODE_ENABLE_MASK, fuse3); + u32 mirror_l3bank_enable = xe_mmio_read32(mmio, MIRROR_L3BANK_ENABLE); + u32 bank_val = REG_FIELD_GET(XE3_L3BANK_ENABLE, mirror_l3bank_enable); + + bitmap_from_arr32(per_node, &bank_val, 32); + gen_l3_mask_from_pattern(xe, l3_bank_mask, per_node, 32, + meml3_en); + } else if (GRAPHICS_VER(xe) >= 20) { xe_l3_bank_mask_t per_node = {}; u32 meml3_en = REG_FIELD_GET(XE2_NODE_ENABLE_MASK, fuse3); u32 bank_val = REG_FIELD_GET(XE2_GT_L3_MODE_MASK, fuse3); @@ -155,7 +165,7 @@ load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask) xe_l3_bank_mask_t per_node = {}; xe_l3_bank_mask_t per_mask_bit = {}; u32 meml3_en = REG_FIELD_GET(MEML3_EN_MASK, fuse3); - u32 fuse4 = xe_mmio_read32(>->mmio, XEHP_FUSE4); + u32 fuse4 = xe_mmio_read32(mmio, XEHP_FUSE4); u32 bank_val = REG_FIELD_GET(GT_L3_EXC_MASK, fuse4); bitmap_set_value8(per_mask_bit, 0x3, 0); From 174e9ce0daf6af791386e96e76e743eb59e8a401 Mon Sep 17 00:00:00 2001 From: John Harrison Date: Mon, 13 Jan 2025 11:44:04 -0800 Subject: [PATCH 0075/1443] drm/xe/guc: Drop error messages about missing GuC logs The GuC log snapshot code would complain loudly if there was no GuC log to take a snapshot of or if the snapshot alloc failed. Originally, this code was only called on demand when a user (or developer) explicitly requested a dump of the log. Hence an error message was useful. However, it is now part of the general devcoredump file and is called for any GPU hang. Most people don't care about GuC logs and GPU hangs do not generally mean a kernel/GuC bug. More importantly, there are valid situations where there is no GuC log, e.g. SRIOV VFs. So drop the error message. Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/3958 Signed-off-by: John Harrison Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250113194405.2033085-1-John.C.Harrison@Intel.com --- drivers/gpu/drm/xe/xe_guc_log.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index df4cfb698cdb..80151ff6a71f 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -149,16 +149,12 @@ struct xe_guc_log_snapshot *xe_guc_log_snapshot_capture(struct xe_guc_log *log, size_t remain; int i; - if (!log->bo) { - xe_gt_err(gt, "GuC log buffer not allocated\n"); + if (!log->bo) return NULL; - } snapshot = xe_guc_log_snapshot_alloc(log, atomic); - if (!snapshot) { - xe_gt_err(gt, "GuC log snapshot not allocated\n"); + if (!snapshot) return NULL; - } remain = snapshot->size; for (i = 0; i < snapshot->num_chunks; i++) { From 1e14484677c8e87548f5f0d4eb8800e408004404 Mon Sep 17 00:00:00 2001 From: John Keeping Date: Wed, 15 Jan 2025 11:01:37 +0000 Subject: [PATCH 0076/1443] drm/ssd130x: fix ssd132x encoding The ssd132x buffer is encoded one pixel per nibble, with two pixels in each byte. When encoding an 8-bit greyscale input, take the top 4-bits as the value and ensure the two pixels are distinct and do not overwrite each other. Fixes: fdd591e00a9c ("drm/ssd130x: Add support for the SSD132x OLED controller family") Signed-off-by: John Keeping Reviewed-by: Javier Martinez Canillas Link: https://patchwork.freedesktop.org/patch/msgid/20250115110139.1672488-2-jkeeping@inmusicbrands.com Signed-off-by: Javier Martinez Canillas --- drivers/gpu/drm/solomon/ssd130x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/solomon/ssd130x.c b/drivers/gpu/drm/solomon/ssd130x.c index b777690fd660..4bb663f984ce 100644 --- a/drivers/gpu/drm/solomon/ssd130x.c +++ b/drivers/gpu/drm/solomon/ssd130x.c @@ -880,7 +880,7 @@ static int ssd132x_update_rect(struct ssd130x_device *ssd130x, u8 n1 = buf[i * width + j]; u8 n2 = buf[i * width + j + 1]; - data_array[array_idx++] = (n2 << 4) | n1; + data_array[array_idx++] = (n2 & 0xf0) | (n1 >> 4); } } From 229adcffdb54b13332d2afd2dc5d203418d50908 Mon Sep 17 00:00:00 2001 From: John Keeping Date: Wed, 15 Jan 2025 11:01:38 +0000 Subject: [PATCH 0077/1443] drm/ssd130x: ensure ssd132x pitch is correct The bounding rectangle is adjusted to ensure it aligns to SSD132X_SEGMENT_WIDTH, which may adjust the pitch. Calculate the pitch after aligning the left and right edge. Fixes: fdd591e00a9c ("drm/ssd130x: Add support for the SSD132x OLED controller family") Signed-off-by: John Keeping Reviewed-by: Javier Martinez Canillas Link: https://patchwork.freedesktop.org/patch/msgid/20250115110139.1672488-3-jkeeping@inmusicbrands.com Signed-off-by: Javier Martinez Canillas --- drivers/gpu/drm/solomon/ssd130x.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/solomon/ssd130x.c b/drivers/gpu/drm/solomon/ssd130x.c index 4bb663f984ce..dd2006d51c7a 100644 --- a/drivers/gpu/drm/solomon/ssd130x.c +++ b/drivers/gpu/drm/solomon/ssd130x.c @@ -1037,7 +1037,7 @@ static int ssd132x_fb_blit_rect(struct drm_framebuffer *fb, struct drm_format_conv_state *fmtcnv_state) { struct ssd130x_device *ssd130x = drm_to_ssd130x(fb->dev); - unsigned int dst_pitch = drm_rect_width(rect); + unsigned int dst_pitch; struct iosys_map dst; int ret = 0; @@ -1046,6 +1046,8 @@ static int ssd132x_fb_blit_rect(struct drm_framebuffer *fb, rect->x2 = min_t(unsigned int, round_up(rect->x2, SSD132X_SEGMENT_WIDTH), ssd130x->width); + dst_pitch = drm_rect_width(rect); + ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE); if (ret) return ret; From 440aaf479c9aaf5ecea9a463eb826ec243d5f1cf Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Tue, 14 Jan 2025 10:59:42 +0000 Subject: [PATCH 0078/1443] drm/sched: Remove weak paused submission checks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is no need to check the boolean in the work item's prologues since the boolean can be set at any later time anyway. The helper which pauses submission sets it and synchronously cancels the work and helpers which queue the work check for the flag so all should be good. Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Danilo Krummrich Cc: Matthew Brost Cc: Philipp Stanner Signed-off-by: Philipp Stanner Link: https://patchwork.freedesktop.org/patch/msgid/20250114105942.64832-1-tvrtko.ursulin@igalia.com --- drivers/gpu/drm/scheduler/sched_main.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 81d69c2ff6ab..d172e3158880 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -1153,9 +1153,6 @@ static void drm_sched_free_job_work(struct work_struct *w) container_of(w, struct drm_gpu_scheduler, work_free_job); struct drm_sched_job *job; - if (READ_ONCE(sched->pause_submit)) - return; - job = drm_sched_get_finished_job(sched); if (job) sched->ops->free_job(job); @@ -1179,9 +1176,6 @@ static void drm_sched_run_job_work(struct work_struct *w) struct drm_sched_job *sched_job; int r; - if (READ_ONCE(sched->pause_submit)) - return; - /* Find entity with a ready job */ entity = drm_sched_select_entity(sched); if (!entity) From 75d37750a753e7ae079e470ea9699caeae756e3d Mon Sep 17 00:00:00 2001 From: Tejas Upadhyay Date: Mon, 13 Jan 2025 17:12:01 +0530 Subject: [PATCH 0079/1443] drm/xe/mmap: Add mmap support for PCI memory barrier In order to avoid having userspace to use MI_MEM_FENCE, we are adding a mechanism for userspace to generate a PCI memory barrier with low overhead (avoiding IOCTL call as well as writing to VRAM will adds some overhead). This is implemented by memory-mapping a page as uncached that is backed by MMIO on the dGPU and thus allowing userspace to do memory write to the page without invoking an IOCTL. We are selecting the MMIO so that it is not accessible from the PCI bus so that the MMIO writes themselves are ignored, but the PCI memory barrier will still take action as the MMIO filtering will happen after the memory barrier effect. When we detect special defined offset in mmap(), We are mapping 4K page which contains the last of page of doorbell MMIO range to userspace for same purpose. For user to query special offset we are adding special flag in mmap_offset ioctl which needs to be passed as follows, struct drm_xe_gem_mmap_offset mmo = { .handle = 0, /* this must be 0 */ .flags = DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER, }; igt_ioctl(fd, DRM_IOCTL_XE_GEM_MMAP_OFFSET, &mmo); map = mmap(NULL, size, PROT_WRITE, MAP_SHARED, fd, mmo); IGT : https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/commit/b2dbc6f22815128c0dd5c737504f42e1f1a6ad62 UMD : https://github.com/intel/compute-runtime/pull/772 V7: - Dgpu filter added V6(MAuld) - Move physical mmap to fault handler - Modify kernel-doc and attach UMD PR when ready V5(MAuld) - Return invalid early in case of non 4K PAGE_SIZE - Format kernel-doc and add note for 4K PAGE_SIZE HW limit V4(MAuld) - Add kernel-doc for uapi change - Restrict page size to 4K V3(MAuld) - Remove offset defination from UAPI to be able to change later - Edit commit message for special flag addition V2(MAuld) - Add fault handler with dummy page to handle unplug device - Add Build check for special offset to be below normal start page - Test d3hot, mapping seems to be valid in d3hot as well - Add more info to commit message Cc: Matthew Auld Acked-by: Michal Mrozek Reviewed-by: Matthew Auld Signed-off-by: Tejas Upadhyay Signed-off-by: Matthew Auld Link: https://patchwork.freedesktop.org/patch/msgid/20250113114201.3178806-1-tejas.upadhyay@intel.com --- drivers/gpu/drm/xe/xe_bo.c | 19 +++++- drivers/gpu/drm/xe/xe_bo.h | 2 + drivers/gpu/drm/xe/xe_device.c | 107 ++++++++++++++++++++++++++++++++- include/uapi/drm/xe_drm.h | 29 ++++++++- 4 files changed, 154 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 4f077c11e8e2..78d19fd4670a 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -2278,9 +2278,26 @@ int xe_gem_mmap_offset_ioctl(struct drm_device *dev, void *data, XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1])) return -EINVAL; - if (XE_IOCTL_DBG(xe, args->flags)) + if (XE_IOCTL_DBG(xe, args->flags & + ~DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER)) return -EINVAL; + if (args->flags & DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER) { + if (XE_IOCTL_DBG(xe, !IS_DGFX(xe))) + return -EINVAL; + + if (XE_IOCTL_DBG(xe, args->handle)) + return -EINVAL; + + if (XE_IOCTL_DBG(xe, PAGE_SIZE > SZ_4K)) + return -EINVAL; + + BUILD_BUG_ON(((XE_PCI_BARRIER_MMAP_OFFSET >> XE_PTE_SHIFT) + + SZ_4K) >= DRM_FILE_PAGE_OFFSET_START); + args->offset = XE_PCI_BARRIER_MMAP_OFFSET; + return 0; + } + gem_obj = drm_gem_object_lookup(file, args->handle); if (XE_IOCTL_DBG(xe, !gem_obj)) return -ENOENT; diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h index d9386ab03140..04995c5ced32 100644 --- a/drivers/gpu/drm/xe/xe_bo.h +++ b/drivers/gpu/drm/xe/xe_bo.h @@ -75,6 +75,8 @@ #define XE_BO_PROPS_INVALID (-1) +#define XE_PCI_BARRIER_MMAP_OFFSET (0x50 << XE_PTE_SHIFT) + struct sg_table; struct xe_bo *xe_bo_alloc(void); diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 5cbc96b214fe..6ecbf7dd396c 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -232,12 +232,117 @@ static long xe_drm_compat_ioctl(struct file *file, unsigned int cmd, unsigned lo #define xe_drm_compat_ioctl NULL #endif +static void barrier_open(struct vm_area_struct *vma) +{ + drm_dev_get(vma->vm_private_data); +} + +static void barrier_close(struct vm_area_struct *vma) +{ + drm_dev_put(vma->vm_private_data); +} + +static void barrier_release_dummy_page(struct drm_device *dev, void *res) +{ + struct page *dummy_page = (struct page *)res; + + __free_page(dummy_page); +} + +static vm_fault_t barrier_fault(struct vm_fault *vmf) +{ + struct drm_device *dev = vmf->vma->vm_private_data; + struct vm_area_struct *vma = vmf->vma; + vm_fault_t ret = VM_FAULT_NOPAGE; + pgprot_t prot; + int idx; + + prot = vm_get_page_prot(vma->vm_flags); + + if (drm_dev_enter(dev, &idx)) { + unsigned long pfn; + +#define LAST_DB_PAGE_OFFSET 0x7ff001 + pfn = PHYS_PFN(pci_resource_start(to_pci_dev(dev->dev), 0) + + LAST_DB_PAGE_OFFSET); + ret = vmf_insert_pfn_prot(vma, vma->vm_start, pfn, + pgprot_noncached(prot)); + drm_dev_exit(idx); + } else { + struct page *page; + + /* Allocate new dummy page to map all the VA range in this VMA to it*/ + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + return VM_FAULT_OOM; + + /* Set the page to be freed using drmm release action */ + if (drmm_add_action_or_reset(dev, barrier_release_dummy_page, page)) + return VM_FAULT_OOM; + + ret = vmf_insert_pfn_prot(vma, vma->vm_start, page_to_pfn(page), + prot); + } + + return ret; +} + +static const struct vm_operations_struct vm_ops_barrier = { + .open = barrier_open, + .close = barrier_close, + .fault = barrier_fault, +}; + +static int xe_pci_barrier_mmap(struct file *filp, + struct vm_area_struct *vma) +{ + struct drm_file *priv = filp->private_data; + struct drm_device *dev = priv->minor->dev; + struct xe_device *xe = to_xe_device(dev); + + if (!IS_DGFX(xe)) + return -EINVAL; + + if (vma->vm_end - vma->vm_start > SZ_4K) + return -EINVAL; + + if (is_cow_mapping(vma->vm_flags)) + return -EINVAL; + + if (vma->vm_flags & (VM_READ | VM_EXEC)) + return -EINVAL; + + vm_flags_clear(vma, VM_MAYREAD | VM_MAYEXEC); + vm_flags_set(vma, VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP | VM_IO); + vma->vm_ops = &vm_ops_barrier; + vma->vm_private_data = dev; + drm_dev_get(vma->vm_private_data); + + return 0; +} + +static int xe_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_file *priv = filp->private_data; + struct drm_device *dev = priv->minor->dev; + + if (drm_dev_is_unplugged(dev)) + return -ENODEV; + + switch (vma->vm_pgoff) { + case XE_PCI_BARRIER_MMAP_OFFSET >> XE_PTE_SHIFT: + return xe_pci_barrier_mmap(filp, vma); + } + + return drm_gem_mmap(filp, vma); +} + static const struct file_operations xe_driver_fops = { .owner = THIS_MODULE, .open = drm_open, .release = drm_release_noglobal, .unlocked_ioctl = xe_drm_ioctl, - .mmap = drm_gem_mmap, + .mmap = xe_mmap, .poll = drm_poll, .read = drm_read, .compat_ioctl = xe_drm_compat_ioctl, diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index f62689ca861a..cac607a30f6d 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -811,6 +811,32 @@ struct drm_xe_gem_create { /** * struct drm_xe_gem_mmap_offset - Input of &DRM_IOCTL_XE_GEM_MMAP_OFFSET + * + * The @flags can be: + * - %DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER - For user to query special offset + * for use in mmap ioctl. Writing to the returned mmap address will generate a + * PCI memory barrier with low overhead (avoiding IOCTL call as well as writing + * to VRAM which would also add overhead), acting like an MI_MEM_FENCE + * instruction. + * + * Note: The mmap size can be at most 4K, due to HW limitations. As a result + * this interface is only supported on CPU architectures that support 4K page + * size. The mmap_offset ioctl will detect this and gracefully return an + * error, where userspace is expected to have a different fallback method for + * triggering a barrier. + * + * Roughly the usage would be as follows: + * + * .. code-block:: C + * + * struct drm_xe_gem_mmap_offset mmo = { + * .handle = 0, // must be set to 0 + * .flags = DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER, + * }; + * + * err = ioctl(fd, DRM_IOCTL_XE_GEM_MMAP_OFFSET, &mmo); + * map = mmap(NULL, size, PROT_WRITE, MAP_SHARED, fd, mmo.offset); + * map[i] = 0xdeadbeaf; // issue barrier */ struct drm_xe_gem_mmap_offset { /** @extensions: Pointer to the first extension struct, if any */ @@ -819,7 +845,8 @@ struct drm_xe_gem_mmap_offset { /** @handle: Handle for the object being mapped. */ __u32 handle; - /** @flags: Must be zero */ +#define DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER (1 << 0) + /** @flags: Flags */ __u32 flags; /** @offset: The fake offset to use for subsequent mmap call */ From 758debf35b9cda5450e40996991a6e4b222899bd Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Mon, 13 Jan 2025 16:25:07 -0800 Subject: [PATCH 0080/1443] drm/xe: Mark ComputeCS read mode as UC on iGPU MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit RING_CMD_CCTL read index should be UC on iGPU parts due to L3 caching structure. Having this as WB blocks ULLS from being enabled. Change to UC to unblock ULLS on iGPU. v2: - Drop internal communications commnet, bspec is updated Cc: Balasubramani Vivekanandan Cc: Michal Mrozek Cc: Paulo Zanoni Cc: José Roberto de Souza Cc: stable@vger.kernel.org Fixes: 328e089bfb37 ("drm/xe: Leverage ComputeCS read L3 caching") Signed-off-by: Matthew Brost Acked-by: Michal Mrozek Reviewed-by: Stuart Summers Reviewed-by: Matt Roper Link: https://patchwork.freedesktop.org/patch/msgid/20250114002507.114087-1-matthew.brost@intel.com --- drivers/gpu/drm/xe/xe_hw_engine.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c index ac9c666a9652..fc447751fe78 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine.c +++ b/drivers/gpu/drm/xe/xe_hw_engine.c @@ -422,7 +422,7 @@ hw_engine_setup_default_state(struct xe_hw_engine *hwe) * Bspec: 72161 */ const u8 mocs_write_idx = gt->mocs.uc_index; - const u8 mocs_read_idx = hwe->class == XE_ENGINE_CLASS_COMPUTE && + const u8 mocs_read_idx = hwe->class == XE_ENGINE_CLASS_COMPUTE && IS_DGFX(xe) && (GRAPHICS_VER(xe) >= 20 || xe->info.platform == XE_PVC) ? gt->mocs.wb_index : gt->mocs.uc_index; u32 ring_cmd_cctl_val = REG_FIELD_PREP(CMD_CCTL_WRITE_OVERRIDE_MASK, mocs_write_idx) | From 63060df6f709cbe494f0cfcaa613655862ba479a Mon Sep 17 00:00:00 2001 From: Oak Zeng Date: Wed, 18 Dec 2024 11:48:31 -0500 Subject: [PATCH 0081/1443] drm/xe: trace bo create Add a tracepoint to trace bo create. Signed-off-by: Oak Zeng Reviewed-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://patchwork.freedesktop.org/patch/msgid/20241218164833.2364049-2-oak.zeng@intel.com Signed-off-by: Himal Prasad Ghimiray --- drivers/gpu/drm/xe/xe_bo.c | 1 + drivers/gpu/drm/xe/xe_trace_bo.h | 5 +++++ 2 files changed, 6 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 78d19fd4670a..cf0dc9e9c53e 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -1659,6 +1659,7 @@ __xe_bo_create_locked(struct xe_device *xe, } } + trace_xe_bo_create(bo); return bo; err_unlock_put_bo: diff --git a/drivers/gpu/drm/xe/xe_trace_bo.h b/drivers/gpu/drm/xe/xe_trace_bo.h index ea50fee50c7d..3d7e6c80b0aa 100644 --- a/drivers/gpu/drm/xe/xe_trace_bo.h +++ b/drivers/gpu/drm/xe/xe_trace_bo.h @@ -53,6 +53,11 @@ DEFINE_EVENT(xe_bo, xe_bo_validate, TP_ARGS(bo) ); +DEFINE_EVENT(xe_bo, xe_bo_create, + TP_PROTO(struct xe_bo *bo), + TP_ARGS(bo) +); + TRACE_EVENT(xe_bo_move, TP_PROTO(struct xe_bo *bo, uint32_t new_placement, uint32_t old_placement, bool move_lacks_source), From 861b27584d9055e4e1763341474ce8ce9dc6a55d Mon Sep 17 00:00:00 2001 From: Oak Zeng Date: Wed, 18 Dec 2024 11:48:32 -0500 Subject: [PATCH 0082/1443] drm/xe: Print vm flags in xe_vm trace print Print vm flags in xe_vm trace print. This is helpful to diagnosis the VM mode of operation. Signed-off-by: Oak Zeng Reviewed-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://patchwork.freedesktop.org/patch/msgid/20241218164833.2364049-3-oak.zeng@intel.com Signed-off-by: Himal Prasad Ghimiray --- drivers/gpu/drm/xe/xe_trace_bo.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_trace_bo.h b/drivers/gpu/drm/xe/xe_trace_bo.h index 3d7e6c80b0aa..082fadb5f99b 100644 --- a/drivers/gpu/drm/xe/xe_trace_bo.h +++ b/drivers/gpu/drm/xe/xe_trace_bo.h @@ -190,16 +190,19 @@ DECLARE_EVENT_CLASS(xe_vm, __string(dev, __dev_name_vm(vm)) __field(struct xe_vm *, vm) __field(u32, asid) + __field(u32, flags) ), TP_fast_assign( __assign_str(dev); __entry->vm = vm; __entry->asid = vm->usm.asid; + __entry->flags = vm->flags; ), - TP_printk("dev=%s, vm=%p, asid=0x%05x", __get_str(dev), - __entry->vm, __entry->asid) + TP_printk("dev=%s, vm=%p, asid=0x%05x, vm flags=0x%05x", + __get_str(dev), __entry->vm, __entry->asid, + __entry->flags) ); DEFINE_EVENT(xe_vm, xe_vm_kill, From 22b1a53f282b1ad6692c6238a7446275854f0afb Mon Sep 17 00:00:00 2001 From: Oak Zeng Date: Wed, 18 Dec 2024 11:48:33 -0500 Subject: [PATCH 0083/1443] drm/xe: Print vm parameter in xe_vma trace Print the vm that the vma belongs to in the vma trace. This is useful to correlate VMA operations to the VM. Signed-off-by: Oak Zeng Reviewed-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://patchwork.freedesktop.org/patch/msgid/20241218164833.2364049-4-oak.zeng@intel.com Signed-off-by: Himal Prasad Ghimiray --- drivers/gpu/drm/xe/xe_trace_bo.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_trace_bo.h b/drivers/gpu/drm/xe/xe_trace_bo.h index 082fadb5f99b..ccebd5f0878e 100644 --- a/drivers/gpu/drm/xe/xe_trace_bo.h +++ b/drivers/gpu/drm/xe/xe_trace_bo.h @@ -92,6 +92,7 @@ DECLARE_EVENT_CLASS(xe_vma, TP_STRUCT__entry( __string(dev, __dev_name_vma(vma)) __field(struct xe_vma *, vma) + __field(struct xe_vm *, vm) __field(u32, asid) __field(u64, start) __field(u64, end) @@ -101,14 +102,16 @@ DECLARE_EVENT_CLASS(xe_vma, TP_fast_assign( __assign_str(dev); __entry->vma = vma; + __entry->vm = xe_vma_vm(vma); __entry->asid = xe_vma_vm(vma)->usm.asid; __entry->start = xe_vma_start(vma); __entry->end = xe_vma_end(vma) - 1; __entry->ptr = xe_vma_userptr(vma); ), - TP_printk("dev=%s, vma=%p, asid=0x%05x, start=0x%012llx, end=0x%012llx, userptr=0x%012llx,", - __get_str(dev), __entry->vma, __entry->asid, __entry->start, + TP_printk("dev=%s, vma=%p, vm=%p, asid=0x%05x, start=0x%012llx, end=0x%012llx, userptr=0x%012llx", + __get_str(dev), __entry->vma, __entry->vm, + __entry->asid, __entry->start, __entry->end, __entry->ptr) ) From b824709ee1d0dbfed4b1757279c97fc0edad1e1a Mon Sep 17 00:00:00 2001 From: Oak Zeng Date: Mon, 13 Jan 2025 16:23:24 -0500 Subject: [PATCH 0084/1443] drm/xe: Fix a typo in xe_vm_doc.h s/vm->ttm.base.resv->lock/vm->gpuvm.r_obj->resv->lock Signed-off-by: Oak Zeng Reviewed-by: Maciej Patelczyk Link: https://patchwork.freedesktop.org/patch/msgid/20250113212324.3264218-1-oak.zeng@intel.com Signed-off-by: Himal Prasad Ghimiray --- drivers/gpu/drm/xe/xe_vm_doc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_vm_doc.h b/drivers/gpu/drm/xe/xe_vm_doc.h index 078786958403..1030ce214032 100644 --- a/drivers/gpu/drm/xe/xe_vm_doc.h +++ b/drivers/gpu/drm/xe/xe_vm_doc.h @@ -431,7 +431,7 @@ * bind path also acquires this lock in write while the exec / compute mode * rebind worker acquires this lock in read mode. * - * VM dma-resv lock (vm->ttm.base.resv->lock) - WW lock. Protects VM dma-resv + * VM dma-resv lock (vm->gpuvm.r_obj->resv->lock) - WW lock. Protects VM dma-resv * slots which is shared with any private BO in the VM. Expected to be acquired * during VM binds, execs, and compute mode rebind worker. This lock is also * held when private BOs are being evicted. From 35286a67d69cf5738f6f0e7f35c1a77b3d3a2921 Mon Sep 17 00:00:00 2001 From: Philipp Stanner Date: Mon, 13 Jan 2025 13:18:50 +0100 Subject: [PATCH 0085/1443] MAINTAINERS: Update DRM GPU Scheduler section Luben has not been active and has not responded to mails since summer 2024. Remove him from MAINTAINERS and add an entry in CREDITS. Philipp has a new email address and an ACK to commit work time to the scheduler. Thus, set the state to 'Supported'. Cc: Matthew Brost Cc: Danilo Krummrich Reviewed-by: Dave Airlie Signed-off-by: Philipp Stanner Link: https://patchwork.freedesktop.org/patch/msgid/20250113121851.31382-1-phasta@kernel.org --- CREDITS | 4 ++++ MAINTAINERS | 5 ++--- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/CREDITS b/CREDITS index b1777b53c63a..a548de2702b8 100644 --- a/CREDITS +++ b/CREDITS @@ -3936,6 +3936,10 @@ S: 1 Amherst Street S: Cambridge, Massachusetts 02139 S: USA +N: Luben Tuikov +E: Luben Tuikov +D: Maintainer of the DRM GPU Scheduler + N: Simmule Turner E: sturner@tele-tv.com D: Added swapping to filesystem diff --git a/MAINTAINERS b/MAINTAINERS index 33fe500bf3ae..3fdc1e60e8b0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7905,12 +7905,11 @@ F: Documentation/gpu/zynqmp.rst F: drivers/gpu/drm/xlnx/ DRM GPU SCHEDULER -M: Luben Tuikov M: Matthew Brost M: Danilo Krummrich -M: Philipp Stanner +M: Philipp Stanner L: dri-devel@lists.freedesktop.org -S: Maintained +S: Supported T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: drivers/gpu/drm/scheduler/ F: include/drm/gpu_scheduler.h From 9b22441acdc3aaf77d5e10ae7a8240fa3f53b9eb Mon Sep 17 00:00:00 2001 From: Philipp Stanner Date: Mon, 13 Jan 2025 13:18:51 +0100 Subject: [PATCH 0086/1443] MAINTAINERS: Add DRM GPU Scheduler reviewer MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Christian König is the original author of much of the scheduler's code and, thus, well suited to do reviews. Cc: Matthew Brost Cc: Danilo Krummrich Cc: Christian König Signed-off-by: Philipp Stanner Acked-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20250113121851.31382-2-phasta@kernel.org --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 3fdc1e60e8b0..b6d06fdaced3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7908,6 +7908,7 @@ DRM GPU SCHEDULER M: Matthew Brost M: Danilo Krummrich M: Philipp Stanner +R: Christian König L: dri-devel@lists.freedesktop.org S: Supported T: git https://gitlab.freedesktop.org/drm/misc/kernel.git From 713a341836942d2b3ea4463377cb8c8f7fc37a26 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Tue, 14 Jan 2025 09:44:36 +0100 Subject: [PATCH 0087/1443] MAINTAINERS: Update intel_vpu maintainer list Slawek moved to another project and Maciej will be replacing him. Reviewed-by: Stanislaw Gruszka Reviewed-by: Oded Gabbay Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250114084436.1326127-1-jacek.lawrynowicz@linux.intel.com --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index b6d06fdaced3..4b038382481f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7054,7 +7054,7 @@ F: include/linux/power/smartreflex.h DRM ACCEL DRIVERS FOR INTEL VPU M: Jacek Lawrynowicz -M: Stanislaw Gruszka +M: Maciej Falkowski L: dri-devel@lists.freedesktop.org S: Supported T: git https://gitlab.freedesktop.org/drm/misc/kernel.git From 474c4dd29f666145dee7b5dce56d024a26e9550c Mon Sep 17 00:00:00 2001 From: Francois Dugast Date: Thu, 16 Jan 2025 13:45:32 +0100 Subject: [PATCH 0088/1443] drm/xe: Add missing SPDX license identifiers Ensure all Xe driver files have a proper SPDX license identifier, add it in files where it was missing. Link: https://patchwork.freedesktop.org/patch/msgid/20250116124532.1480351-1-francois.dugast@intel.com Signed-off-by: Francois Dugast Reviewed-by: Lucas De Marchi --- drivers/gpu/drm/xe/Kconfig.profile | 1 + drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h | 5 +++++ 2 files changed, 6 insertions(+) diff --git a/drivers/gpu/drm/xe/Kconfig.profile b/drivers/gpu/drm/xe/Kconfig.profile index ba17a25e8db3..7530df998148 100644 --- a/drivers/gpu/drm/xe/Kconfig.profile +++ b/drivers/gpu/drm/xe/Kconfig.profile @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only config DRM_XE_JOB_TIMEOUT_MAX int "Default max job timeout (ms)" default 10000 # milliseconds diff --git a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h index 9c4cf050059a..41d39d67817a 100644 --- a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h +++ b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h @@ -1,3 +1,8 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2024 Intel Corporation + */ + #ifndef _I915_GEM_STOLEN_H_ #define _I915_GEM_STOLEN_H_ From 931a765c2d95c46cb818eccc641006c86697486f Mon Sep 17 00:00:00 2001 From: Youssef Samir Date: Fri, 13 Dec 2024 11:51:10 -0700 Subject: [PATCH 0089/1443] accel/qaic: Change aic100_image_table definition aic100_image_table is currently defined as a "const char *" array, this can potentially lead to the accidental modification of the pointers inside. Also, checkpatch.pl gives a warning about it. Change the type to a "const char * const" array to make the pointers immutable, preventing accidental modification of the images' paths. Signed-off-by: Youssef Samir Reviewed-by: Carl Vanderlip Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20241213185110.2469159-1-quic_jhugo@quicinc.com --- drivers/accel/qaic/sahara.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/accel/qaic/sahara.c b/drivers/accel/qaic/sahara.c index 21d58aed0deb..09c8b055aa81 100644 --- a/drivers/accel/qaic/sahara.c +++ b/drivers/accel/qaic/sahara.c @@ -160,7 +160,7 @@ struct sahara_context { struct work_struct fw_work; struct work_struct dump_work; struct mhi_device *mhi_dev; - const char **image_table; + const char * const *image_table; u32 table_size; u32 active_image_id; const struct firmware *firmware; @@ -177,7 +177,7 @@ struct sahara_context { bool is_mem_dump_mode; }; -static const char *aic100_image_table[] = { +static const char * const aic100_image_table[] = { [1] = "qcom/aic100/fw1.bin", [2] = "qcom/aic100/fw2.bin", [4] = "qcom/aic100/fw4.bin", From c03dd019f0b56c40f42fa0e7785eaa109e0bdcd6 Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Fri, 17 Jan 2025 09:00:29 +0800 Subject: [PATCH 0090/1443] drm/panel-edp: Add BOE NV140FHM-NZ panel entry Add an eDP panel entry for BOE NV140FHM-NZ. No datasheet found for this panel, so the timing is based on a similar NV140FHM-N41 datasheet that I can find on internet[0]. edid: 00 ff ff ff ff ff ff 00 09 e5 09 0b 00 00 00 00 01 20 01 04 a5 1f 11 78 03 9b 75 99 5b 5d 8f 2a 23 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 c8 37 80 cc 70 38 28 40 6c 30 aa 00 35 ae 10 00 00 1a 00 00 00 fd 00 30 3c 43 43 8f 01 0a 20 20 20 20 20 20 00 00 00 fe 00 42 4f 45 20 48 46 0a 20 20 20 20 20 20 00 00 00 fe 00 4e 56 31 34 30 46 48 4d 2d 4e 34 5a 0a 00 35 [0]:http://www.tfinno.com/PIC/PIC/20215121628440.pdf Signed-off-by: Andy Yan Reviewed-by: Thomas Zimmermann Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250117010039.1815430-1-andyshrk@163.com --- drivers/gpu/drm/panel/panel-edp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 319ebc184c16..1721a3397dbf 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1915,6 +1915,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a5d, &delay_200_500_e50, "NV116WHM-N45"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ac5, &delay_200_500_e50, "NV116WHM-N4C"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ae8, &delay_200_500_e50_p2e80, "NV140WUM-N41"), + EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b09, &delay_200_500_e50_po2e200, "NV140FHM-NZ"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b34, &delay_200_500_e80, "NV122WUM-N41"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b43, &delay_200_500_e200, "NV140FHM-T09"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b56, &delay_200_500_e80, "NT140FHM-N47"), From d68fc95a771e0a7edd876ede7913d61276be77fd Mon Sep 17 00:00:00 2001 From: Francis Pravin Date: Fri, 17 Jan 2025 05:12:09 +0530 Subject: [PATCH 0091/1443] nvme-pci: remove redundant dma frees in hmb The value of size is 0 when there is no dma buffer allocated. The value of i also remains 0. So, no need to free the dma buffer in out_free_bufs. Hence, remove the redundant dma frees. Signed-off-by: Francis Pravin Reviewed-by: Christoph Hellwig Signed-off-by: Keith Busch --- drivers/nvme/host/pci.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index fe0795e16e25..a14f3c74b717 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2153,14 +2153,6 @@ static int nvme_alloc_host_mem_multi(struct nvme_dev *dev, u64 preferred, return 0; out_free_bufs: - while (--i >= 0) { - size_t size = le32_to_cpu(descs[i].size) * NVME_CTRL_PAGE_SIZE; - - dma_free_attrs(dev->dev, size, bufs[i], - le64_to_cpu(descs[i].addr), - DMA_ATTR_NO_KERNEL_MAPPING | DMA_ATTR_NO_WARN); - } - kfree(bufs); out_free_descs: dma_free_coherent(dev->dev, descs_size, descs, descs_dma); From dbf2bb1a1319b7c7d8828905378a6696cca6b0f2 Mon Sep 17 00:00:00 2001 From: Georg Gottleuber Date: Mon, 16 Dec 2024 23:28:03 +0100 Subject: [PATCH 0092/1443] nvme-pci: Add TUXEDO InfinityFlex to Samsung sleep quirk On the TUXEDO InfinityFlex, a Samsung 990 Evo NVMe leads to a high power consumption in s2idle sleep (4 watts). This patch applies 'Force No Simple Suspend' quirk to achieve a sleep with a lower power consumption, typically around 1.4 watts. Signed-off-by: Georg Gottleuber Cc: stable@vger.kernel.org Signed-off-by: Werner Sembach Reviewed-by: Christoph Hellwig Signed-off-by: Keith Busch --- drivers/nvme/host/pci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index a14f3c74b717..60afffc917b7 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -3134,7 +3134,8 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev) * because of high power consumption (> 2 Watt) in s2idle * sleep. Only some boards with Intel CPU are affected. */ - if (dmi_match(DMI_BOARD_NAME, "GMxPXxx") || + if (dmi_match(DMI_BOARD_NAME, "DN50Z-140HC-YD") || + dmi_match(DMI_BOARD_NAME, "GMxPXxx") || dmi_match(DMI_BOARD_NAME, "PH4PG31") || dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1") || dmi_match(DMI_BOARD_NAME, "PH6PG01_PH6PG71")) From 11cb3529d18514f7d28ad2190533192aedefd761 Mon Sep 17 00:00:00 2001 From: Georg Gottleuber Date: Mon, 16 Dec 2024 23:28:04 +0100 Subject: [PATCH 0093/1443] nvme-pci: Add TUXEDO IBP Gen9 to Samsung sleep quirk On the TUXEDO InfinityBook Pro Gen9 Intel, a Samsung 990 Evo NVMe leads to a high power consumption in s2idle sleep (4 watts). This patch applies 'Force No Simple Suspend' quirk to achieve a sleep with a lower power consumption, typically around 1.2 watts. Signed-off-by: Georg Gottleuber Cc: stable@vger.kernel.org Signed-off-by: Werner Sembach Reviewed-by: Christoph Hellwig Signed-off-by: Keith Busch --- drivers/nvme/host/pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 60afffc917b7..ac708169efed 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -3136,6 +3136,7 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev) */ if (dmi_match(DMI_BOARD_NAME, "DN50Z-140HC-YD") || dmi_match(DMI_BOARD_NAME, "GMxPXxx") || + dmi_match(DMI_BOARD_NAME, "GXxMRXx") || dmi_match(DMI_BOARD_NAME, "PH4PG31") || dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1") || dmi_match(DMI_BOARD_NAME, "PH6PG01_PH6PG71")) From bbd8429264baf8bc3c40cefda048560ae0eb7890 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Thu, 14 Nov 2024 18:59:54 +0100 Subject: [PATCH 0094/1443] drm/xe: Always setup GT MMIO adjustment data While we believed that xe_gt_mmio_init() will be called just once per GT, this might not be a case due to some tweaks that need to performed by the VF driver during early probe. To avoid leaving any stale data in case of the re-run, reset the GT MMIO adjustment data for the non-media GT case. Signed-off-by: Michal Wajdeczko Cc: Matt Roper Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20241114175955.2299-2-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 26e64530ada2..b5c313a3e946 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -643,6 +643,9 @@ void xe_gt_mmio_init(struct xe_gt *gt) if (gt->info.type == XE_GT_TYPE_MEDIA) { gt->mmio.adj_offset = MEDIA_GT_GSI_OFFSET; gt->mmio.adj_limit = MEDIA_GT_GSI_LENGTH; + } else { + gt->mmio.adj_offset = 0; + gt->mmio.adj_limit = 0; } if (IS_SRIOV_VF(gt_to_xe(gt))) From 13265fe7426ec9ba5aa86baab913417ca361e8a4 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Tue, 14 Jan 2025 22:13:47 +0100 Subject: [PATCH 0095/1443] drm/xe/vf: Perform early GT MMIO initialization to read GMDID MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit VFs need to communicate with the GuC to obtain the GMDID value and existing GuC functions used for that assume that the GT has it's MMIO members already setup. However, due to recent refactoring the gt->mmio is initialized later, and any attempt by the VF to use xe_mmio_read|write() from GuC functions will lead to NPD crash due to unset MMIO register address: [] xe 0000:00:02.1: [drm] Running in SR-IOV VF mode [] xe 0000:00:02.1: [drm] GT0: sending H2G MMIO 0x5507 [] BUG: unable to handle page fault for address: 0000000000190240 Since we are already tweaking the id and type of the primary GT to mimic it's a Media GT before initializing the GuC communication, we can also call xe_gt_mmio_init() to perform early setup of the gt->mmio which will make those GuC functions work again. Signed-off-by: Michal Wajdeczko Cc: Matt Roper Cc: Piotr Piórkowski Reviewed-by: Piotr Piórkowski Link: https://patchwork.freedesktop.org/patch/msgid/20250114211347.1083-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 48d1c81d441e..bf35a18bf5e7 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -501,6 +501,7 @@ static void read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, u32 *ver, gt->info.type = XE_GT_TYPE_MAIN; } + xe_gt_mmio_init(gt); xe_guc_comm_init_early(>->uc.guc); /* Don't bother with GMDID if failed to negotiate the GuC ABI */ From 9cd3f4efc870463f17f6c29114c61fb6bfcaa291 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:54 +0100 Subject: [PATCH 0096/1443] drm/xe/sa: Always call drm_suballoc_manager_fini() After successful call to drm_suballoc_manager_init() we should make sure to call drm_suballoc_manager_fini() as it may include some cleanup code even if we didn't start using it for real. As we can abort init() early due to kvzalloc() failure, we should either explicitly call drm_suballoc_manager_fini() or, even better, postpone drm_suballoc_manager_init() once we finish all other preparation steps, so we can rely on fini() that will do cleanup. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-2-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index e055bed7ae55..4e7aba445ebc 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -57,8 +57,6 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 } sa_manager->bo = bo; sa_manager->is_iomem = bo->vmap.is_iomem; - - drm_suballoc_manager_init(&sa_manager->base, managed_size, align); sa_manager->gpu_addr = xe_bo_ggtt_addr(bo); if (bo->vmap.is_iomem) { @@ -72,6 +70,7 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 memset(sa_manager->cpu_ptr, 0, bo->ttm.base.size); } + drm_suballoc_manager_init(&sa_manager->base, managed_size, align); ret = drmm_add_action_or_reset(&xe->drm, xe_sa_bo_manager_fini, sa_manager); if (ret) From d29cddd49bed2c880e7c17724bcf3604e865c23a Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:55 +0100 Subject: [PATCH 0097/1443] drm/xe/sa: Drop redundant NULL assignments The sa_manager is drmm_kzalloc'ed so all members are already zero. And in case of kvzalloc() failure we are not returning pointer to the sa_manager at all, so no point in resetting .bo member. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-3-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index 4e7aba445ebc..eb314ca75355 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -44,8 +44,6 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 if (!sa_manager) return ERR_PTR(-ENOMEM); - sa_manager->bo = NULL; - bo = xe_managed_bo_create_pin_map(xe, tile, size, XE_BO_FLAG_VRAM_IF_DGFX(tile) | XE_BO_FLAG_GGTT | @@ -61,10 +59,8 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 if (bo->vmap.is_iomem) { sa_manager->cpu_ptr = kvzalloc(managed_size, GFP_KERNEL); - if (!sa_manager->cpu_ptr) { - sa_manager->bo = NULL; + if (!sa_manager->cpu_ptr) return ERR_PTR(-ENOMEM); - } } else { sa_manager->cpu_ptr = bo->vmap.vaddr; memset(sa_manager->cpu_ptr, 0, bo->ttm.base.size); From 97ee0e351f6ebbcb2a2dccdff726f75f728fede8 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:56 +0100 Subject: [PATCH 0098/1443] drm/xe/sa: Improve error message on init failure Instead of raw errno value we can print friendly error code and also print size of the buffer object that we fail to prepare. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-4-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index eb314ca75355..0b87599759d7 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -49,8 +49,8 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 XE_BO_FLAG_GGTT | XE_BO_FLAG_GGTT_INVALIDATE); if (IS_ERR(bo)) { - drm_err(&xe->drm, "failed to allocate bo for sa manager: %ld\n", - PTR_ERR(bo)); + drm_err(&xe->drm, "Failed to prepare %uKiB BO for SA manager (%pe)\n", + size / SZ_1K, bo); return ERR_CAST(bo); } sa_manager->bo = bo; From 7e937cdf18164ea276ce0f4bbc5755e0031280e0 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:57 +0100 Subject: [PATCH 0099/1443] drm/xe/sa: Tidy up coding style in init() There is no need to use tile_to_xe() since we already got the xe. And we should keep all variable declarations together, no need for separate sa_manager declaration. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-5-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index 0b87599759d7..0e78ae46667e 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -34,13 +34,12 @@ static void xe_sa_bo_manager_fini(struct drm_device *drm, void *arg) struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align) { struct xe_device *xe = tile_to_xe(tile); + struct xe_sa_manager *sa_manager; u32 managed_size = size - SZ_4K; struct xe_bo *bo; int ret; - struct xe_sa_manager *sa_manager = drmm_kzalloc(&tile_to_xe(tile)->drm, - sizeof(*sa_manager), - GFP_KERNEL); + sa_manager = drmm_kzalloc(&xe->drm, sizeof(*sa_manager), GFP_KERNEL); if (!sa_manager) return ERR_PTR(-ENOMEM); From 0e1871f61e71d7611196b04d1b133f18fef666dd Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:58 +0100 Subject: [PATCH 0100/1443] drm/xe/sa: Allow making suballocations using custom gfp flags Actual xe_sa_manager implementation uses hardcoded GFP_KERNEL flag during creation of suballocations but in upcoming patch we want to reuse the xe_sa_manager in places where GFP_KERNEL is not allowed. Add another variant of the xe_sa_bo_new() function that accepts arbitrary gfp flags. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-6-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 15 ++++++++++++--- drivers/gpu/drm/xe/xe_sa.h | 19 +++++++++++++++++-- 2 files changed, 29 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index 0e78ae46667e..5f89e32b0640 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -74,8 +74,17 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 return sa_manager; } -struct drm_suballoc *xe_sa_bo_new(struct xe_sa_manager *sa_manager, - unsigned int size) +/** + * __xe_sa_bo_new() - Make a suballocation but use custom gfp flags. + * @sa_manager: the &xe_sa_manager + * @size: number of bytes we want to suballocate + * @gfp: gfp flags used for memory allocation. Typically GFP_KERNEL. + * + * Try to make a suballocation of size @size. + * + * Return: a &drm_suballoc, or an ERR_PTR. + */ +struct drm_suballoc *__xe_sa_bo_new(struct xe_sa_manager *sa_manager, u32 size, gfp_t gfp) { /* * BB to large, return -ENOBUFS indicating user should split @@ -84,7 +93,7 @@ struct drm_suballoc *xe_sa_bo_new(struct xe_sa_manager *sa_manager, if (size > sa_manager->base.size) return ERR_PTR(-ENOBUFS); - return drm_suballoc_new(&sa_manager->base, size, GFP_KERNEL, true, 0); + return drm_suballoc_new(&sa_manager->base, size, gfp, true, 0); } void xe_sa_bo_flush_write(struct drm_suballoc *sa_bo) diff --git a/drivers/gpu/drm/xe/xe_sa.h b/drivers/gpu/drm/xe/xe_sa.h index 4e96483057d7..a0341eafbe77 100644 --- a/drivers/gpu/drm/xe/xe_sa.h +++ b/drivers/gpu/drm/xe/xe_sa.h @@ -5,6 +5,7 @@ #ifndef _XE_SA_H_ #define _XE_SA_H_ +#include #include "xe_sa_types.h" struct dma_fence; @@ -13,8 +14,22 @@ struct xe_tile; struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align); -struct drm_suballoc *xe_sa_bo_new(struct xe_sa_manager *sa_manager, - u32 size); +struct drm_suballoc *__xe_sa_bo_new(struct xe_sa_manager *sa_manager, u32 size, gfp_t gfp); + +/** + * xe_sa_bo_new() - Make a suballocation. + * @sa_manager: the &xe_sa_manager + * @size: number of bytes we want to suballocate + * + * Try to make a suballocation of size @size. + * + * Return: a &drm_suballoc, or an ERR_PTR. + */ +static inline struct drm_suballoc *xe_sa_bo_new(struct xe_sa_manager *sa_manager, u32 size) +{ + return __xe_sa_bo_new(sa_manager, size, GFP_KERNEL); +} + void xe_sa_bo_flush_write(struct drm_suballoc *sa_bo); void xe_sa_bo_free(struct drm_suballoc *sa_bo, struct dma_fence *fence); From ae8b507fb8bbea2aa30783184d5728b14ce40c8f Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:41:59 +0100 Subject: [PATCH 0101/1443] drm/xe/sa: Allow creating suballocator with custom guard size Actual xe_sa_manager implementation uses hardcoded 4K to exclude it from making suballocations but in upcoming patch we want to reuse the xe_sa_manager where such 4K guard is not needed. Add another variant of the xe_sa_bo_manager_init() function that accepts arbitrary guard size. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-7-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.c | 18 ++++++++++++++++-- drivers/gpu/drm/xe/xe_sa.h | 9 +++++++-- 2 files changed, 23 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c index 5f89e32b0640..f8fe61e25518 100644 --- a/drivers/gpu/drm/xe/xe_sa.c +++ b/drivers/gpu/drm/xe/xe_sa.c @@ -31,14 +31,28 @@ static void xe_sa_bo_manager_fini(struct drm_device *drm, void *arg) sa_manager->bo = NULL; } -struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align) +/** + * __xe_sa_bo_manager_init() - Create and initialize the suballocator + * @tile: the &xe_tile where allocate + * @size: number of bytes to allocate + * @guard: number of bytes to exclude from suballocations + * @align: alignment for each suballocated chunk + * + * Prepares the suballocation manager for suballocations. + * + * Return: a pointer to the &xe_sa_manager or an ERR_PTR on failure. + */ +struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 guard, u32 align) { struct xe_device *xe = tile_to_xe(tile); struct xe_sa_manager *sa_manager; - u32 managed_size = size - SZ_4K; + u32 managed_size; struct xe_bo *bo; int ret; + xe_tile_assert(tile, size > guard); + managed_size = size - guard; + sa_manager = drmm_kzalloc(&xe->drm, sizeof(*sa_manager), GFP_KERNEL); if (!sa_manager) return ERR_PTR(-ENOMEM); diff --git a/drivers/gpu/drm/xe/xe_sa.h b/drivers/gpu/drm/xe/xe_sa.h index a0341eafbe77..de0330eb36d4 100644 --- a/drivers/gpu/drm/xe/xe_sa.h +++ b/drivers/gpu/drm/xe/xe_sa.h @@ -5,6 +5,7 @@ #ifndef _XE_SA_H_ #define _XE_SA_H_ +#include #include #include "xe_sa_types.h" @@ -12,10 +13,14 @@ struct dma_fence; struct xe_bo; struct xe_tile; -struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align); - +struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 guard, u32 align); struct drm_suballoc *__xe_sa_bo_new(struct xe_sa_manager *sa_manager, u32 size, gfp_t gfp); +static inline struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align) +{ + return __xe_sa_bo_manager_init(tile, size, SZ_4K, align); +} + /** * xe_sa_bo_new() - Make a suballocation. * @sa_manager: the &xe_sa_manager From c49ca671818a325f2221f0bda8af96e339272a5e Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:42:00 +0100 Subject: [PATCH 0102/1443] drm/xe/sa: Minor header cleanups Drop unused struct xe_bo forward declaration and, while around, fix unnecessary line split in xe_sa_bo_free() declaration. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-8-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_sa.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_sa.h b/drivers/gpu/drm/xe/xe_sa.h index de0330eb36d4..1170ee5a81a8 100644 --- a/drivers/gpu/drm/xe/xe_sa.h +++ b/drivers/gpu/drm/xe/xe_sa.h @@ -10,7 +10,6 @@ #include "xe_sa_types.h" struct dma_fence; -struct xe_bo; struct xe_tile; struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 guard, u32 align); @@ -36,8 +35,7 @@ static inline struct drm_suballoc *xe_sa_bo_new(struct xe_sa_manager *sa_manager } void xe_sa_bo_flush_write(struct drm_suballoc *sa_bo); -void xe_sa_bo_free(struct drm_suballoc *sa_bo, - struct dma_fence *fence); +void xe_sa_bo_free(struct drm_suballoc *sa_bo, struct dma_fence *fence); static inline struct xe_sa_manager * to_xe_sa_manager(struct drm_suballoc_manager *mng) From 696bfdf273eab9ce3dd2ff51d26ca30f7924a4bb Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:42:01 +0100 Subject: [PATCH 0103/1443] drm/xe/guc: Introduce the GuC Buffer Cache The purpose of the GuC Buffer Cache is to maintain a set ofreusable buffers that could be used while sending some of the CTB H2G actions that require separate buffer with indirect data. Currently only few PF actions need this so initialize it only when running as a PF. Signed-off-by: Michal Wajdeczko Cc: Matthew Brost Cc: Rodrigo Vivi Acked-by: Rodrigo Vivi Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-9-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/Makefile | 1 + drivers/gpu/drm/xe/xe_guc.c | 5 + drivers/gpu/drm/xe/xe_guc_buf.c | 172 ++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_guc_buf.h | 47 +++++++ drivers/gpu/drm/xe/xe_guc_buf_types.h | 28 +++++ drivers/gpu/drm/xe/xe_guc_types.h | 3 + 6 files changed, 256 insertions(+) create mode 100644 drivers/gpu/drm/xe/xe_guc_buf.c create mode 100644 drivers/gpu/drm/xe/xe_guc_buf.h create mode 100644 drivers/gpu/drm/xe/xe_guc_buf_types.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 7730e0596299..2e50a697f549 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -56,6 +56,7 @@ xe-y += xe_bb.o \ xe_gt_topology.o \ xe_guc.o \ xe_guc_ads.o \ + xe_guc_buf.o \ xe_guc_capture.o \ xe_guc_ct.o \ xe_guc_db_mgr.o \ diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index 408365dfe4ee..1619c0a52db9 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -23,6 +23,7 @@ #include "xe_gt_sriov_vf.h" #include "xe_gt_throttle.h" #include "xe_guc_ads.h" +#include "xe_guc_buf.h" #include "xe_guc_capture.h" #include "xe_guc_ct.h" #include "xe_guc_db_mgr.h" @@ -743,6 +744,10 @@ int xe_guc_init_post_hwconfig(struct xe_guc *guc) if (ret) return ret; + ret = xe_guc_buf_cache_init(&guc->buf); + if (ret) + return ret; + return xe_guc_ads_init_post_hwconfig(&guc->ads); } diff --git a/drivers/gpu/drm/xe/xe_guc_buf.c b/drivers/gpu/drm/xe/xe_guc_buf.c new file mode 100644 index 000000000000..261c7c74417f --- /dev/null +++ b/drivers/gpu/drm/xe/xe_guc_buf.c @@ -0,0 +1,172 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2024 Intel Corporation + */ + +#include +#include + +#include "xe_assert.h" +#include "xe_bo.h" +#include "xe_gt_printk.h" +#include "xe_guc.h" +#include "xe_guc_buf.h" +#include "xe_sa.h" + +static struct xe_guc *cache_to_guc(struct xe_guc_buf_cache *cache) +{ + return container_of(cache, struct xe_guc, buf); +} + +static struct xe_gt *cache_to_gt(struct xe_guc_buf_cache *cache) +{ + return guc_to_gt(cache_to_guc(cache)); +} + +/** + * xe_guc_buf_cache_init() - Initialize the GuC Buffer Cache. + * @cache: the &xe_guc_buf_cache to initialize + * + * The Buffer Cache allows to obtain a reusable buffer that can be used to pass + * indirect H2G data to GuC without a need to create a ad-hoc allocation. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_guc_buf_cache_init(struct xe_guc_buf_cache *cache) +{ + struct xe_gt *gt = cache_to_gt(cache); + struct xe_sa_manager *sam; + + /* XXX: currently it's useful only for the PF actions */ + if (!IS_SRIOV_PF(gt_to_xe(gt))) + return 0; + + sam = __xe_sa_bo_manager_init(gt_to_tile(gt), SZ_8K, 0, sizeof(u32)); + if (IS_ERR(sam)) + return PTR_ERR(sam); + cache->sam = sam; + + xe_gt_dbg(gt, "reusable buffer with %u dwords at %#x for %ps\n", + xe_guc_buf_cache_dwords(cache), xe_bo_ggtt_addr(sam->bo), + __builtin_return_address(0)); + return 0; +} + +/** + * xe_guc_buf_cache_dwords() - Number of dwords the GuC Buffer Cache supports. + * @cache: the &xe_guc_buf_cache to query + * + * Return: a size of the largest reusable buffer (in dwords) + */ +u32 xe_guc_buf_cache_dwords(struct xe_guc_buf_cache *cache) +{ + return cache->sam ? cache->sam->base.size / sizeof(u32) : 0; +} + +/** + * xe_guc_buf_reserve() - Reserve a new sub-allocation. + * @cache: the &xe_guc_buf_cache where reserve sub-allocation + * @dwords: the requested size of the buffer in dwords + * + * Use xe_guc_buf_is_valid() to check if returned buffer reference is valid. + * Must use xe_guc_buf_release() to release a sub-allocation. + * + * Return: a &xe_guc_buf of new sub-allocation. + */ +struct xe_guc_buf xe_guc_buf_reserve(struct xe_guc_buf_cache *cache, u32 dwords) +{ + struct drm_suballoc *sa; + + if (cache->sam) + sa = __xe_sa_bo_new(cache->sam, dwords * sizeof(32), GFP_ATOMIC); + else + sa = ERR_PTR(-EOPNOTSUPP); + + return (struct xe_guc_buf){ .sa = sa }; +} + +/** + * xe_guc_buf_from_data() - Reserve a new sub-allocation using data. + * @cache: the &xe_guc_buf_cache where reserve sub-allocation + * @data: the data to flush the sub-allocation + * @size: the size of the data + * + * Similar to xe_guc_buf_reserve() but flushes @data to the GPU memory. + * + * Return: a &xe_guc_buf of new sub-allocation. + */ +struct xe_guc_buf xe_guc_buf_from_data(struct xe_guc_buf_cache *cache, + const void *data, size_t size) +{ + struct drm_suballoc *sa; + + sa = __xe_sa_bo_new(cache->sam, size, GFP_ATOMIC); + if (!IS_ERR(sa)) + memcpy(xe_sa_bo_cpu_addr(sa), data, size); + + return (struct xe_guc_buf){ .sa = sa }; +} + +/** + * xe_guc_buf_release() - Release a sub-allocation. + * @buf: the &xe_guc_buf to release + * + * Releases a sub-allocation reserved by the xe_guc_buf_reserve(). + */ +void xe_guc_buf_release(const struct xe_guc_buf buf) +{ + if (xe_guc_buf_is_valid(buf)) + xe_sa_bo_free(buf.sa, NULL); +} + +/** + * xe_guc_buf_flush() - Copy the data from the sub-allocation to the GPU memory. + * @buf: the &xe_guc_buf to flush + * + * Return: a GPU address of the sub-allocation. + */ +u64 xe_guc_buf_flush(const struct xe_guc_buf buf) +{ + xe_sa_bo_flush_write(buf.sa); + return xe_sa_bo_gpu_addr(buf.sa); +} + +/** + * xe_guc_buf_cpu_ptr() - Obtain a CPU pointer to the sub-allocation. + * @buf: the &xe_guc_buf to query + * + * Return: a CPU pointer of the sub-allocation. + */ +void *xe_guc_buf_cpu_ptr(const struct xe_guc_buf buf) +{ + return xe_sa_bo_cpu_addr(buf.sa); +} + +/** + * xe_guc_buf_gpu_addr() - Obtain a GPU address of the sub-allocation. + * @buf: the &xe_guc_buf to query + * + * Return: a GPU address of the sub-allocation. + */ +u64 xe_guc_buf_gpu_addr(const struct xe_guc_buf buf) +{ + return xe_sa_bo_gpu_addr(buf.sa); +} + +/** + * xe_guc_cache_gpu_addr_from_ptr() - Lookup a GPU address using the pointer. + * @cache: the &xe_guc_buf_cache with sub-allocations + * @ptr: the CPU pointer of the sub-allocation + * @size: the size of the data + * + * Return: a GPU address on success or 0 if the pointer was unrelated. + */ +u64 xe_guc_cache_gpu_addr_from_ptr(struct xe_guc_buf_cache *cache, const void *ptr, u32 size) +{ + ptrdiff_t offset = ptr - cache->sam->cpu_ptr; + + if (offset < 0 || offset + size > cache->sam->base.size) + return 0; + + return cache->sam->gpu_addr + offset; +} diff --git a/drivers/gpu/drm/xe/xe_guc_buf.h b/drivers/gpu/drm/xe/xe_guc_buf.h new file mode 100644 index 000000000000..0d67604d96bd --- /dev/null +++ b/drivers/gpu/drm/xe/xe_guc_buf.h @@ -0,0 +1,47 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2024 Intel Corporation + */ + +#ifndef _XE_GUC_BUF_H_ +#define _XE_GUC_BUF_H_ + +#include +#include + +#include "xe_guc_buf_types.h" + +int xe_guc_buf_cache_init(struct xe_guc_buf_cache *cache); +u32 xe_guc_buf_cache_dwords(struct xe_guc_buf_cache *cache); +struct xe_guc_buf xe_guc_buf_reserve(struct xe_guc_buf_cache *cache, u32 dwords); +struct xe_guc_buf xe_guc_buf_from_data(struct xe_guc_buf_cache *cache, + const void *data, size_t size); +void xe_guc_buf_release(const struct xe_guc_buf buf); + +/** + * xe_guc_buf_is_valid() - Check if a buffer reference is valid. + * @buf: the &xe_guc_buf reference to check + * + * Return: true if @ref represents a valid sub-allication. + */ +static inline bool xe_guc_buf_is_valid(const struct xe_guc_buf buf) +{ + return !IS_ERR_OR_NULL(buf.sa); +} + +void *xe_guc_buf_cpu_ptr(const struct xe_guc_buf buf); +u64 xe_guc_buf_flush(const struct xe_guc_buf buf); +u64 xe_guc_buf_gpu_addr(const struct xe_guc_buf buf); +u64 xe_guc_cache_gpu_addr_from_ptr(struct xe_guc_buf_cache *cache, const void *ptr, u32 size); + +DEFINE_CLASS(xe_guc_buf, struct xe_guc_buf, + xe_guc_buf_release(_T), + xe_guc_buf_reserve(cache, num), + struct xe_guc_buf_cache *cache, u32 num); + +DEFINE_CLASS(xe_guc_buf_from_data, struct xe_guc_buf, + xe_guc_buf_release(_T), + xe_guc_buf_from_data(cache, data, size), + struct xe_guc_buf_cache *cache, const void *data, size_t size); + +#endif diff --git a/drivers/gpu/drm/xe/xe_guc_buf_types.h b/drivers/gpu/drm/xe/xe_guc_buf_types.h new file mode 100644 index 000000000000..9e123d71c064 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_guc_buf_types.h @@ -0,0 +1,28 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2024 Intel Corporation + */ + +#ifndef _XE_GUC_BUF_TYPES_H_ +#define _XE_GUC_BUF_TYPES_H_ + +struct drm_suballoc; +struct xe_sa_manager; + +/** + * struct xe_guc_buf_cache - GuC Data Buffer Cache. + */ +struct xe_guc_buf_cache { + /* private: internal sub-allocation manager */ + struct xe_sa_manager *sam; +}; + +/** + * struct xe_guc_buf - GuC Data Buffer Reference. + */ +struct xe_guc_buf { + /* private: internal sub-allocation reference */ + struct drm_suballoc *sa; +}; + +#endif diff --git a/drivers/gpu/drm/xe/xe_guc_types.h b/drivers/gpu/drm/xe/xe_guc_types.h index 83a41ebcdc91..573aa6308380 100644 --- a/drivers/gpu/drm/xe/xe_guc_types.h +++ b/drivers/gpu/drm/xe/xe_guc_types.h @@ -11,6 +11,7 @@ #include "regs/xe_reg_defs.h" #include "xe_guc_ads_types.h" +#include "xe_guc_buf_types.h" #include "xe_guc_ct_types.h" #include "xe_guc_fwif.h" #include "xe_guc_log_types.h" @@ -58,6 +59,8 @@ struct xe_guc { struct xe_guc_ads ads; /** @ct: GuC ct */ struct xe_guc_ct ct; + /** @buf: GuC Buffer Cache manager */ + struct xe_guc_buf_cache buf; /** @capture: the error-state-capture module's data and objects */ struct xe_guc_state_capture *capture; /** @pc: GuC Power Conservation */ From d8b2149ba8f184cd138b482289b16d8558787e99 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:42:02 +0100 Subject: [PATCH 0104/1443] drm/xe/pf: Use GuC Buffer Cache during VFs provisioning Start using GuC buffer cache for the VF's configuration actions. Signed-off-by: Michal Wajdeczko Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-10-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 68 +++++++++++----------- 1 file changed, 35 insertions(+), 33 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 878e96281c03..9db1c920219d 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -20,6 +20,7 @@ #include "xe_gt_sriov_pf_policy.h" #include "xe_gt_sriov_printk.h" #include "xe_guc.h" +#include "xe_guc_buf.h" #include "xe_guc_ct.h" #include "xe_guc_db_mgr.h" #include "xe_guc_fwif.h" @@ -71,48 +72,27 @@ static int pf_send_vf_cfg_reset(struct xe_gt *gt, u32 vfid) * Return: number of KLVs that were successfully parsed and saved, * negative error code on failure. */ -static int pf_send_vf_cfg_klvs(struct xe_gt *gt, u32 vfid, const u32 *klvs, u32 num_dwords) +static int pf_send_vf_buf_klvs(struct xe_gt *gt, u32 vfid, struct xe_guc_buf buf, u32 num_dwords) { - const u32 bytes = num_dwords * sizeof(u32); - struct xe_tile *tile = gt_to_tile(gt); - struct xe_device *xe = tile_to_xe(tile); struct xe_guc *guc = >->uc.guc; - struct xe_bo *bo; - int ret; - bo = xe_bo_create_pin_map(xe, tile, NULL, - ALIGN(bytes, PAGE_SIZE), - ttm_bo_type_kernel, - XE_BO_FLAG_VRAM_IF_DGFX(tile) | - XE_BO_FLAG_GGTT | - XE_BO_FLAG_GGTT_INVALIDATE); - if (IS_ERR(bo)) - return PTR_ERR(bo); - - xe_map_memcpy_to(xe, &bo->vmap, 0, klvs, bytes); - - ret = guc_action_update_vf_cfg(guc, vfid, xe_bo_ggtt_addr(bo), num_dwords); - - xe_bo_unpin_map_no_vm(bo); - - return ret; + return guc_action_update_vf_cfg(guc, vfid, xe_guc_buf_flush(buf), num_dwords); } /* * Return: 0 on success, -ENOKEY if some KLVs were not updated, -EPROTO if reply was malformed, * negative error code on failure. */ -static int pf_push_vf_cfg_klvs(struct xe_gt *gt, unsigned int vfid, u32 num_klvs, - const u32 *klvs, u32 num_dwords) +static int pf_push_vf_buf_klvs(struct xe_gt *gt, unsigned int vfid, u32 num_klvs, + struct xe_guc_buf buf, u32 num_dwords) { int ret; - xe_gt_assert(gt, num_klvs == xe_guc_klv_count(klvs, num_dwords)); - - ret = pf_send_vf_cfg_klvs(gt, vfid, klvs, num_dwords); + ret = pf_send_vf_buf_klvs(gt, vfid, buf, num_dwords); if (ret != num_klvs) { int err = ret < 0 ? ret : ret < num_klvs ? -ENOKEY : -EPROTO; + void *klvs = xe_guc_buf_cpu_ptr(buf); struct drm_printer p = xe_gt_info_printer(gt); char name[8]; @@ -125,13 +105,35 @@ static int pf_push_vf_cfg_klvs(struct xe_gt *gt, unsigned int vfid, u32 num_klvs if (IS_ENABLED(CONFIG_DRM_XE_DEBUG_SRIOV)) { struct drm_printer p = xe_gt_info_printer(gt); + void *klvs = xe_guc_buf_cpu_ptr(buf); + char name[8]; + xe_gt_sriov_info(gt, "pushed %s config with %u KLV%s:\n", + xe_sriov_function_name(vfid, name, sizeof(name)), + num_klvs, str_plural(num_klvs)); xe_guc_klv_print(klvs, num_dwords, &p); } return 0; } +/* + * Return: 0 on success, -ENOBUFS if no free buffer for the indirect data, + * negative error code on failure. + */ +static int pf_push_vf_cfg_klvs(struct xe_gt *gt, unsigned int vfid, u32 num_klvs, + const u32 *klvs, u32 num_dwords) +{ + CLASS(xe_guc_buf_from_data, buf)(>->uc.guc.buf, klvs, num_dwords * sizeof(u32)); + + xe_gt_assert(gt, num_klvs == xe_guc_klv_count(klvs, num_dwords)); + + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + + return pf_push_vf_buf_klvs(gt, vfid, num_klvs, buf, num_dwords); +} + static int pf_push_vf_cfg_u32(struct xe_gt *gt, unsigned int vfid, u16 key, u32 value) { u32 klv[] = { @@ -304,16 +306,17 @@ static u32 encode_config(u32 *cfg, const struct xe_gt_sriov_config *config, bool static int pf_push_full_vf_config(struct xe_gt *gt, unsigned int vfid) { struct xe_gt_sriov_config *config = pf_pick_vf_config(gt, vfid); - u32 max_cfg_dwords = SZ_4K / sizeof(u32); + u32 max_cfg_dwords = xe_guc_buf_cache_dwords(>->uc.guc.buf); + CLASS(xe_guc_buf, buf)(>->uc.guc.buf, max_cfg_dwords); u32 num_dwords; int num_klvs; u32 *cfg; int err; - cfg = kcalloc(max_cfg_dwords, sizeof(u32), GFP_KERNEL); - if (!cfg) - return -ENOMEM; + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + cfg = xe_guc_buf_cpu_ptr(buf); num_dwords = encode_config(cfg, config, true); xe_gt_assert(gt, num_dwords <= max_cfg_dwords); @@ -330,9 +333,8 @@ static int pf_push_full_vf_config(struct xe_gt *gt, unsigned int vfid) xe_gt_assert(gt, num_dwords <= max_cfg_dwords); num_klvs = xe_guc_klv_count(cfg, num_dwords); - err = pf_push_vf_cfg_klvs(gt, vfid, num_klvs, cfg, num_dwords); + err = pf_push_vf_buf_klvs(gt, vfid, num_klvs, buf, num_dwords); - kfree(cfg); return err; } From f90b552dcbb4e142f2a15d2b4458ea601248b8e8 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 20 Dec 2024 20:42:03 +0100 Subject: [PATCH 0105/1443] drm/xe/kunit: Allow to replace xe_managed_bo_create_pin_map() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We want to use replacement functions in upcoming kunit tests. Signed-off-by: Michal Wajdeczko Reviewed-by: Michał Winiarski Link: https://patchwork.freedesktop.org/patch/msgid/20241220194205.995-11-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_bo.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index cf0dc9e9c53e..c32201123d44 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -15,6 +15,8 @@ #include #include +#include + #include "xe_device.h" #include "xe_dma_buf.h" #include "xe_drm_client.h" @@ -1797,6 +1799,8 @@ struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile struct xe_bo *bo; int ret; + KUNIT_STATIC_STUB_REDIRECT(xe_managed_bo_create_pin_map, xe, tile, size, flags); + bo = xe_bo_create_pin_map(xe, tile, NULL, size, ttm_bo_type_kernel, flags); if (IS_ERR(bo)) return bo; From 238f96315ada9e2183b04df90c9714b1da68455c Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Tue, 14 Jan 2025 20:21:40 +0100 Subject: [PATCH 0106/1443] drm/xe/kunit: Add KUnit tests for GuC Buffer Cache MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add tests to make sure that recently added GuC Buffer Cache component is working as expected. Signed-off-by: Michal Wajdeczko Reviewed-by: Michał Winiarski Link: https://patchwork.freedesktop.org/patch/msgid/20250114192140.1039-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c | 334 ++++++++++++++++++++ drivers/gpu/drm/xe/xe_guc_buf.c | 4 + 2 files changed, 338 insertions(+) create mode 100644 drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c diff --git a/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c b/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c new file mode 100644 index 000000000000..6faffcd74869 --- /dev/null +++ b/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c @@ -0,0 +1,334 @@ +// SPDX-License-Identifier: GPL-2.0 AND MIT +/* + * Copyright © 2024 Intel Corporation + */ + +#include +#include +#include + +#include "xe_device.h" +#include "xe_ggtt.h" +#include "xe_guc_ct.h" +#include "xe_kunit_helpers.h" +#include "xe_pci_test.h" + +#define DUT_GGTT_START SZ_1M +#define DUT_GGTT_SIZE SZ_2M + +static struct xe_bo *replacement_xe_managed_bo_create_pin_map(struct xe_device *xe, + struct xe_tile *tile, + size_t size, u32 flags) +{ + struct kunit *test = kunit_get_current_test(); + struct xe_bo *bo; + void *buf; + + bo = drmm_kzalloc(&xe->drm, sizeof(*bo), GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, bo); + + buf = drmm_kzalloc(&xe->drm, size, GFP_KERNEL); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, buf); + + bo->tile = tile; + bo->ttm.bdev = &xe->ttm; + bo->size = size; + iosys_map_set_vaddr(&bo->vmap, buf); + + if (flags & XE_BO_FLAG_GGTT) { + struct xe_ggtt *ggtt = tile->mem.ggtt; + + bo->ggtt_node[tile->id] = xe_ggtt_node_init(ggtt); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, bo->ggtt_node[tile->id]); + + KUNIT_ASSERT_EQ(test, 0, + drm_mm_insert_node_in_range(&ggtt->mm, + &bo->ggtt_node[tile->id]->base, + bo->size, SZ_4K, + 0, 0, U64_MAX, 0)); + } + + return bo; +} + +static int guc_buf_test_init(struct kunit *test) +{ + struct xe_pci_fake_data fake = { + .sriov_mode = XE_SRIOV_MODE_PF, + .platform = XE_TIGERLAKE, /* some random platform */ + .subplatform = XE_SUBPLATFORM_NONE, + }; + struct xe_ggtt *ggtt; + struct xe_guc *guc; + + test->priv = &fake; + xe_kunit_helper_xe_device_test_init(test); + + ggtt = xe_device_get_root_tile(test->priv)->mem.ggtt; + guc = &xe_device_get_gt(test->priv, 0)->uc.guc; + + drm_mm_init(&ggtt->mm, DUT_GGTT_START, DUT_GGTT_SIZE); + mutex_init(&ggtt->lock); + + kunit_activate_static_stub(test, xe_managed_bo_create_pin_map, + replacement_xe_managed_bo_create_pin_map); + + KUNIT_ASSERT_EQ(test, 0, xe_guc_buf_cache_init(&guc->buf)); + + test->priv = &guc->buf; + return 0; +} + +static void test_smallest(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + + buf = xe_guc_buf_reserve(cache, 1); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_EXPECT_NOT_NULL(test, xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_NE(test, 0, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_LE(test, DUT_GGTT_START, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_GT(test, DUT_GGTT_START + DUT_GGTT_SIZE, xe_guc_buf_gpu_addr(buf)); + xe_guc_buf_release(buf); +} + +static void test_largest(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + + buf = xe_guc_buf_reserve(cache, xe_guc_buf_cache_dwords(cache)); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_EXPECT_NOT_NULL(test, xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_NE(test, 0, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_LE(test, DUT_GGTT_START, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_GT(test, DUT_GGTT_START + DUT_GGTT_SIZE, xe_guc_buf_gpu_addr(buf)); + xe_guc_buf_release(buf); +} + +static void test_granular(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf *bufs; + int n, dwords; + + dwords = xe_guc_buf_cache_dwords(cache); + bufs = kunit_kcalloc(test, dwords, sizeof(*bufs), GFP_KERNEL); + KUNIT_EXPECT_NOT_NULL(test, bufs); + + for (n = 0; n < dwords; n++) + bufs[n] = xe_guc_buf_reserve(cache, 1); + + for (n = 0; n < dwords; n++) + KUNIT_EXPECT_TRUE_MSG(test, xe_guc_buf_is_valid(bufs[n]), "n=%d", n); + + for (n = 0; n < dwords; n++) + xe_guc_buf_release(bufs[n]); +} + +static void test_unique(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf *bufs; + int n, m, dwords; + + dwords = xe_guc_buf_cache_dwords(cache); + bufs = kunit_kcalloc(test, dwords, sizeof(*bufs), GFP_KERNEL); + KUNIT_EXPECT_NOT_NULL(test, bufs); + + for (n = 0; n < dwords; n++) + bufs[n] = xe_guc_buf_reserve(cache, 1); + + for (n = 0; n < dwords; n++) { + for (m = n + 1; m < dwords; m++) { + KUNIT_EXPECT_PTR_NE_MSG(test, xe_guc_buf_cpu_ptr(bufs[n]), + xe_guc_buf_cpu_ptr(bufs[m]), "n=%d, m=%d", n, m); + KUNIT_ASSERT_NE_MSG(test, xe_guc_buf_gpu_addr(bufs[n]), + xe_guc_buf_gpu_addr(bufs[m]), "n=%d, m=%d", n, m); + } + } + + for (n = 0; n < dwords; n++) + xe_guc_buf_release(bufs[n]); +} + +static void test_overlap(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf b1, b2; + u32 dwords = xe_guc_buf_cache_dwords(cache) / 2; + u32 bytes = dwords * sizeof(u32); + void *p1, *p2; + u64 a1, a2; + + b1 = xe_guc_buf_reserve(cache, dwords); + b2 = xe_guc_buf_reserve(cache, dwords); + + p1 = xe_guc_buf_cpu_ptr(b1); + p2 = xe_guc_buf_cpu_ptr(b2); + + a1 = xe_guc_buf_gpu_addr(b1); + a2 = xe_guc_buf_gpu_addr(b2); + + KUNIT_EXPECT_PTR_NE(test, p1, p2); + if (p1 < p2) + KUNIT_EXPECT_LT(test, (uintptr_t)(p1 + bytes - 1), (uintptr_t)p2); + else + KUNIT_EXPECT_LT(test, (uintptr_t)(p2 + bytes - 1), (uintptr_t)p1); + + KUNIT_EXPECT_NE(test, a1, a2); + if (a1 < a2) + KUNIT_EXPECT_LT(test, a1 + bytes - 1, a2); + else + KUNIT_EXPECT_LT(test, a2 + bytes - 1, a1); + + xe_guc_buf_release(b1); + xe_guc_buf_release(b2); +} + +static void test_reusable(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf b1, b2; + void *p1; + u64 a1; + + b1 = xe_guc_buf_reserve(cache, xe_guc_buf_cache_dwords(cache)); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(b1)); + KUNIT_EXPECT_NOT_NULL(test, p1 = xe_guc_buf_cpu_ptr(b1)); + KUNIT_EXPECT_NE(test, 0, a1 = xe_guc_buf_gpu_addr(b1)); + xe_guc_buf_release(b1); + + b2 = xe_guc_buf_reserve(cache, xe_guc_buf_cache_dwords(cache)); + KUNIT_EXPECT_PTR_EQ(test, p1, xe_guc_buf_cpu_ptr(b2)); + KUNIT_EXPECT_EQ(test, a1, xe_guc_buf_gpu_addr(b2)); + xe_guc_buf_release(b2); +} + +static void test_too_big(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + + buf = xe_guc_buf_reserve(cache, xe_guc_buf_cache_dwords(cache) + 1); + KUNIT_EXPECT_FALSE(test, xe_guc_buf_is_valid(buf)); + xe_guc_buf_release(buf); /* shouldn't crash */ +} + +static void test_flush(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + const u32 dwords = xe_guc_buf_cache_dwords(cache); + const u32 bytes = dwords * sizeof(u32); + u32 *s, *p, *d; + int n; + + KUNIT_ASSERT_NOT_NULL(test, s = kunit_kcalloc(test, dwords, sizeof(u32), GFP_KERNEL)); + KUNIT_ASSERT_NOT_NULL(test, d = kunit_kcalloc(test, dwords, sizeof(u32), GFP_KERNEL)); + + for (n = 0; n < dwords; n++) + s[n] = n; + + buf = xe_guc_buf_reserve(cache, dwords); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_ASSERT_NOT_NULL(test, p = xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_PTR_NE(test, p, s); + KUNIT_EXPECT_PTR_NE(test, p, d); + + memcpy(p, s, bytes); + KUNIT_EXPECT_NE(test, 0, xe_guc_buf_flush(buf)); + + iosys_map_memcpy_from(d, &cache->sam->bo->vmap, 0, bytes); + KUNIT_EXPECT_MEMEQ(test, s, d, bytes); + + xe_guc_buf_release(buf); +} + +static void test_lookup(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + u32 dwords; + u64 addr; + u32 *p; + int n; + + dwords = xe_guc_buf_cache_dwords(cache); + buf = xe_guc_buf_reserve(cache, dwords); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_ASSERT_NOT_NULL(test, p = xe_guc_buf_cpu_ptr(buf)); + KUNIT_ASSERT_NE(test, 0, addr = xe_guc_buf_gpu_addr(buf)); + + KUNIT_EXPECT_EQ(test, 0, xe_guc_cache_gpu_addr_from_ptr(cache, p - 1, sizeof(u32))); + KUNIT_EXPECT_EQ(test, 0, xe_guc_cache_gpu_addr_from_ptr(cache, p + dwords, sizeof(u32))); + + for (n = 0; n < dwords; n++) + KUNIT_EXPECT_EQ_MSG(test, xe_guc_cache_gpu_addr_from_ptr(cache, p + n, sizeof(u32)), + addr + n * sizeof(u32), "n=%d", n); + + xe_guc_buf_release(buf); +} + +static void test_data(struct kunit *test) +{ + static const u32 data[] = { 1, 2, 3, 4, 5, 6 }; + struct xe_guc_buf_cache *cache = test->priv; + struct xe_guc_buf buf; + void *p; + + buf = xe_guc_buf_from_data(cache, data, sizeof(data)); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_ASSERT_NOT_NULL(test, p = xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_MEMEQ(test, p, data, sizeof(data)); + + xe_guc_buf_release(buf); +} + +static void test_class(struct kunit *test) +{ + struct xe_guc_buf_cache *cache = test->priv; + u32 dwords = xe_guc_buf_cache_dwords(cache); + + { + CLASS(xe_guc_buf, buf)(cache, dwords); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_EXPECT_NOT_NULL(test, xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_NE(test, 0, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_LE(test, DUT_GGTT_START, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_GT(test, DUT_GGTT_START + DUT_GGTT_SIZE, xe_guc_buf_gpu_addr(buf)); + } + + { + CLASS(xe_guc_buf, buf)(cache, dwords); + KUNIT_ASSERT_TRUE(test, xe_guc_buf_is_valid(buf)); + KUNIT_EXPECT_NOT_NULL(test, xe_guc_buf_cpu_ptr(buf)); + KUNIT_EXPECT_NE(test, 0, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_LE(test, DUT_GGTT_START, xe_guc_buf_gpu_addr(buf)); + KUNIT_EXPECT_GT(test, DUT_GGTT_START + DUT_GGTT_SIZE, xe_guc_buf_gpu_addr(buf)); + } +} + +static struct kunit_case guc_buf_test_cases[] = { + KUNIT_CASE(test_smallest), + KUNIT_CASE(test_largest), + KUNIT_CASE(test_granular), + KUNIT_CASE(test_unique), + KUNIT_CASE(test_overlap), + KUNIT_CASE(test_reusable), + KUNIT_CASE(test_too_big), + KUNIT_CASE(test_flush), + KUNIT_CASE(test_lookup), + KUNIT_CASE(test_data), + KUNIT_CASE(test_class), + {} +}; + +static struct kunit_suite guc_buf_suite = { + .name = "guc_buf", + .test_cases = guc_buf_test_cases, + .init = guc_buf_test_init, +}; + +kunit_test_suites(&guc_buf_suite); diff --git a/drivers/gpu/drm/xe/xe_guc_buf.c b/drivers/gpu/drm/xe/xe_guc_buf.c index 261c7c74417f..ce6d9830e13b 100644 --- a/drivers/gpu/drm/xe/xe_guc_buf.c +++ b/drivers/gpu/drm/xe/xe_guc_buf.c @@ -170,3 +170,7 @@ u64 xe_guc_cache_gpu_addr_from_ptr(struct xe_guc_buf_cache *cache, const void *p return cache->sam->gpu_addr + offset; } + +#if IS_BUILTIN(CONFIG_DRM_XE_KUNIT_TEST) +#include "tests/xe_guc_buf_kunit.c" +#endif From 173baa1b2dc44e3551d9414f4919a48fe5da4880 Mon Sep 17 00:00:00 2001 From: Satyanarayana K V P Date: Thu, 16 Jan 2025 11:26:17 +0530 Subject: [PATCH 0107/1443] drm/xe: Suppress printing of mode when running in non-sriov mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The xe_sriov_probe_early() function prints the sriov pf/vf mode on driver probe. When running in non-sriov mode, the below debug message is seen. "Running in none mode". This print does not convey any information. This commit suppresses this debug message and shows only when running in PF/VF mode. Signed-off-by: Satyanarayana K V P Cc: Michał Wajdeczko Reviewed-by: Lucas De Marchi Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250116055617.20611-1-satyanarayana.k.v.p@intel.com --- drivers/gpu/drm/xe/xe_sriov.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c index 04e2f539ccd9..a0eab44c0e76 100644 --- a/drivers/gpu/drm/xe/xe_sriov.c +++ b/drivers/gpu/drm/xe/xe_sriov.c @@ -81,7 +81,7 @@ void xe_sriov_probe_early(struct xe_device *xe) xe->sriov.__mode = mode; xe_assert(xe, xe->sriov.__mode); - if (has_sriov) + if (IS_SRIOV(xe)) drm_info(&xe->drm, "Running in %s mode\n", xe_sriov_mode_to_string(xe_device_sriov_mode(xe))); } From 6568d423cb3000789b2b8493049883fcec09c329 Mon Sep 17 00:00:00 2001 From: Vivek Kasireddy Date: Mon, 13 Jan 2025 23:57:59 -0800 Subject: [PATCH 0108/1443] drm/virtio: Don't return error if virtio-gpu PCI dev is not found While fixing a shared VGA resource ownership issue, commit 5dd8b536bbda ("drm/virtio: Lock the VGA resources during initialization") wrongly assumed that there is always a PCI device associated with virtio-gpu and it would return error if this device is not found during init. This is incorrect, as virtio-gpu can be operated in MMIO mode (M68K) where a PCI device would probably not be created for it. Therefore, fix this issue by not erroring out if the associated PCI device is not found during initialization. Fixes: 5dd8b536bbda ("drm/virtio: Lock the VGA resources during initialization") Suggested-by: Dmitry Osipenko Cc: Gerd Hoffmann Cc: Gurchetan Singh Cc: Chia-I Wu Signed-off-by: Vivek Kasireddy Link: https://patchwork.freedesktop.org/patch/msgid/20250114075759.2616551-1-vivek.kasireddy@intel.com Reviewed-by: Dmitry Osipenko Signed-off-by: Dmitry Osipenko Link: https://patchwork.freedesktop.org/patch/msgid/20250114075759.2616551-1-vivek.kasireddy@intel.com --- drivers/gpu/drm/virtio/virtgpu_drv.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c index d4309dba557b..2d88e390feb4 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.c +++ b/drivers/gpu/drm/virtio/virtgpu_drv.c @@ -173,23 +173,24 @@ static int __init virtio_gpu_driver_init(void) pdev = pci_get_device(PCI_VENDOR_ID_REDHAT_QUMRANET, PCI_DEVICE_ID_VIRTIO_GPU, NULL); - if (!pdev) - return -ENODEV; - - if (pci_is_vga(pdev)) { + if (pdev && pci_is_vga(pdev)) { ret = vga_get_interruptible(pdev, VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM); - if (ret) - goto error; + if (ret) { + pci_dev_put(pdev); + return ret; + } } ret = register_virtio_driver(&virtio_gpu_driver); - if (pci_is_vga(pdev)) - vga_put(pdev, VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM); + if (pdev) { + if (pci_is_vga(pdev)) + vga_put(pdev, + VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM); -error: - pci_dev_put(pdev); + pci_dev_put(pdev); + } return ret; } From 04630796c437a9285643097825cbd3cd06603f47 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Fri, 20 Dec 2024 10:37:07 -0300 Subject: [PATCH 0109/1443] drm/vc4: Use DRM Execution Contexts MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit VC4 has internal copies of `drm_gem_lock_reservations()` and `drm_gem_unlock_reservations()` within the driver. Ideally, these hard-coded functions should be replaced with the generic functions provided by DRM common code. However, instead of using the DRM GEM functions to (un)lock reservations, transition to the new DRM Execution Contexts API. Signed-off-by: Maíra Canal Acked-by: Christian König Reviewed-by: Melissa Wen Link: https://patchwork.freedesktop.org/patch/msgid/20241220134204.634577-2-mcanal@igalia.com --- drivers/gpu/drm/vc4/Kconfig | 1 + drivers/gpu/drm/vc4/vc4_gem.c | 99 ++++++++--------------------------- 2 files changed, 22 insertions(+), 78 deletions(-) diff --git a/drivers/gpu/drm/vc4/Kconfig b/drivers/gpu/drm/vc4/Kconfig index 6cc7b7e6294a..123ab0ce1781 100644 --- a/drivers/gpu/drm/vc4/Kconfig +++ b/drivers/gpu/drm/vc4/Kconfig @@ -14,6 +14,7 @@ config DRM_VC4 select DRM_DISPLAY_HDMI_HELPER select DRM_DISPLAY_HDMI_STATE_HELPER select DRM_DISPLAY_HELPER + select DRM_EXEC select DRM_KMS_HELPER select DRM_GEM_DMA_HELPER select DRM_PANEL_BRIDGE diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index 22bccd69eb62..1021f45cb53c 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -29,6 +29,7 @@ #include #include +#include #include #include "uapi/drm/vc4_drm.h" @@ -578,19 +579,6 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno) } } -static void -vc4_unlock_bo_reservations(struct drm_device *dev, - struct vc4_exec_info *exec, - struct ww_acquire_ctx *acquire_ctx) -{ - int i; - - for (i = 0; i < exec->bo_count; i++) - dma_resv_unlock(exec->bo[i]->resv); - - ww_acquire_fini(acquire_ctx); -} - /* Takes the reservation lock on all the BOs being referenced, so that * at queue submit time we can update the reservations. * @@ -599,70 +587,23 @@ vc4_unlock_bo_reservations(struct drm_device *dev, * to vc4, so we don't attach dma-buf fences to them. */ static int -vc4_lock_bo_reservations(struct drm_device *dev, - struct vc4_exec_info *exec, - struct ww_acquire_ctx *acquire_ctx) +vc4_lock_bo_reservations(struct vc4_exec_info *exec, + struct drm_exec *exec_ctx) { - int contended_lock = -1; - int i, ret; - struct drm_gem_object *bo; - - ww_acquire_init(acquire_ctx, &reservation_ww_class); - -retry: - if (contended_lock != -1) { - bo = exec->bo[contended_lock]; - ret = dma_resv_lock_slow_interruptible(bo->resv, acquire_ctx); - if (ret) { - ww_acquire_done(acquire_ctx); - return ret; - } - } - - for (i = 0; i < exec->bo_count; i++) { - if (i == contended_lock) - continue; - - bo = exec->bo[i]; - - ret = dma_resv_lock_interruptible(bo->resv, acquire_ctx); - if (ret) { - int j; - - for (j = 0; j < i; j++) { - bo = exec->bo[j]; - dma_resv_unlock(bo->resv); - } - - if (contended_lock != -1 && contended_lock >= i) { - bo = exec->bo[contended_lock]; - - dma_resv_unlock(bo->resv); - } - - if (ret == -EDEADLK) { - contended_lock = i; - goto retry; - } - - ww_acquire_done(acquire_ctx); - return ret; - } - } - - ww_acquire_done(acquire_ctx); + int ret; /* Reserve space for our shared (read-only) fence references, * before we commit the CL to the hardware. */ - for (i = 0; i < exec->bo_count; i++) { - bo = exec->bo[i]; + drm_exec_init(exec_ctx, DRM_EXEC_INTERRUPTIBLE_WAIT, exec->bo_count); + drm_exec_until_all_locked(exec_ctx) { + ret = drm_exec_prepare_array(exec_ctx, exec->bo, + exec->bo_count, 1); + } - ret = dma_resv_reserve_fences(bo->resv, 1); - if (ret) { - vc4_unlock_bo_reservations(dev, exec, acquire_ctx); - return ret; - } + if (ret) { + drm_exec_fini(exec_ctx); + return ret; } return 0; @@ -679,7 +620,7 @@ vc4_lock_bo_reservations(struct drm_device *dev, */ static int vc4_queue_submit(struct drm_device *dev, struct vc4_exec_info *exec, - struct ww_acquire_ctx *acquire_ctx, + struct drm_exec *exec_ctx, struct drm_syncobj *out_sync) { struct vc4_dev *vc4 = to_vc4_dev(dev); @@ -708,7 +649,7 @@ vc4_queue_submit(struct drm_device *dev, struct vc4_exec_info *exec, vc4_update_bo_seqnos(exec, seqno); - vc4_unlock_bo_reservations(dev, exec, acquire_ctx); + drm_exec_fini(exec_ctx); list_add_tail(&exec->head, &vc4->bin_job_list); @@ -1123,7 +1064,7 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data, struct drm_vc4_submit_cl *args = data; struct drm_syncobj *out_sync = NULL; struct vc4_exec_info *exec; - struct ww_acquire_ctx acquire_ctx; + struct drm_exec exec_ctx; struct dma_fence *in_fence; int ret = 0; @@ -1216,7 +1157,7 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data, if (ret) goto fail; - ret = vc4_lock_bo_reservations(dev, exec, &acquire_ctx); + ret = vc4_lock_bo_reservations(exec, &exec_ctx); if (ret) goto fail; @@ -1224,7 +1165,7 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data, out_sync = drm_syncobj_find(file_priv, args->out_sync); if (!out_sync) { ret = -EINVAL; - goto fail; + goto fail_unreserve; } /* We replace the fence in out_sync in vc4_queue_submit since @@ -1239,7 +1180,7 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data, */ exec->args = NULL; - ret = vc4_queue_submit(dev, exec, &acquire_ctx, out_sync); + ret = vc4_queue_submit(dev, exec, &exec_ctx, out_sync); /* The syncobj isn't part of the exec data and we need to free our * reference even if job submission failed. @@ -1248,13 +1189,15 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data, drm_syncobj_put(out_sync); if (ret) - goto fail; + goto fail_unreserve; /* Return the seqno for our job. */ args->seqno = vc4->emit_seqno; return 0; +fail_unreserve: + drm_exec_fini(&exec_ctx); fail: vc4_complete_exec(&vc4->base, exec); From 21bd85f82ec20ddfc65ae999f49356c86fb68a8d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Fri, 20 Dec 2024 10:37:08 -0300 Subject: [PATCH 0110/1443] drm/vc4: Use DMA Resv to implement VC4 wait BO IOCTL MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since the BOs used by VC4 have DMA Reservation Objects attached to them, waiting for seqnos to check BO availability is unnecessary. Instead, `drm_gem_dma_resv_wait()` can be used. Signed-off-by: Maíra Canal Reviewed-by: Melissa Wen Link: https://patchwork.freedesktop.org/patch/msgid/20241220134204.634577-3-mcanal@igalia.com --- drivers/gpu/drm/vc4/vc4_gem.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index 1021f45cb53c..de7be9942c13 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -1020,8 +1020,10 @@ vc4_wait_bo_ioctl(struct drm_device *dev, void *data, struct vc4_dev *vc4 = to_vc4_dev(dev); int ret; struct drm_vc4_wait_bo *args = data; - struct drm_gem_object *gem_obj; - struct vc4_bo *bo; + unsigned long timeout_jiffies = + usecs_to_jiffies(div_u64(args->timeout_ns, 1000)); + ktime_t start = ktime_get(); + u64 delta_ns; if (WARN_ON_ONCE(vc4->gen > VC4_GEN_4)) return -ENODEV; @@ -1029,17 +1031,18 @@ vc4_wait_bo_ioctl(struct drm_device *dev, void *data, if (args->pad != 0) return -EINVAL; - gem_obj = drm_gem_object_lookup(file_priv, args->handle); - if (!gem_obj) { - DRM_DEBUG("Failed to look up GEM BO %d\n", args->handle); - return -EINVAL; - } - bo = to_vc4_bo(gem_obj); + ret = drm_gem_dma_resv_wait(file_priv, args->handle, + true, timeout_jiffies); - ret = vc4_wait_for_seqno_ioctl_helper(dev, bo->seqno, - &args->timeout_ns); + /* Decrement the user's timeout, in case we got interrupted + * such that the ioctl will be restarted. + */ + delta_ns = ktime_to_ns(ktime_sub(ktime_get(), start)); + if (delta_ns < args->timeout_ns) + args->timeout_ns -= delta_ns; + else + args->timeout_ns = 0; - drm_gem_object_put(gem_obj); return ret; } From e8941ac97f281ffa06d6ed058b3bba676f3f8183 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Fri, 20 Dec 2024 10:37:09 -0300 Subject: [PATCH 0111/1443] drm/vc4: Remove BOs seqnos MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit `bo->seqno`, `bo->write_seqno`, and `exec->bin_dep_seqno` are leftovers from a time when VC4 didn't support DMA Reservation Objects. Before DMA Resv was introduced, tracking the correspondence between BOs and jobs through the job's seqno made sense. However, this is no longer needed, as VC4 now supports DMA Reservation Objects and attaches the "job done" fence to the BOs. Therefore, remove the BOs seqnos in favor of using DMA Resv Objects. Signed-off-by: Maíra Canal Reviewed-by: Maxime Ripard Reviewed-by: Melissa Wen Link: https://patchwork.freedesktop.org/patch/msgid/20241220134204.634577-4-mcanal@igalia.com --- drivers/gpu/drm/vc4/vc4_crtc.c | 33 ++++++++--------- drivers/gpu/drm/vc4/vc4_drv.h | 27 -------------- drivers/gpu/drm/vc4/vc4_gem.c | 59 +----------------------------- drivers/gpu/drm/vc4/vc4_validate.c | 11 ------ 4 files changed, 17 insertions(+), 113 deletions(-) diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index cf40a53ad42e..2a48038abe7a 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -884,11 +884,7 @@ struct vc4_async_flip_state { struct drm_framebuffer *fb; struct drm_framebuffer *old_fb; struct drm_pending_vblank_event *event; - - union { - struct dma_fence_cb fence; - struct vc4_seqno_cb seqno; - } cb; + struct dma_fence_cb cb; }; /* Called when the V3D execution for the BO being flipped to is done, so that @@ -919,10 +915,11 @@ vc4_async_page_flip_complete(struct vc4_async_flip_state *flip_state) kfree(flip_state); } -static void vc4_async_page_flip_seqno_complete(struct vc4_seqno_cb *cb) +static void vc4_async_page_flip_complete_with_cleanup(struct dma_fence *fence, + struct dma_fence_cb *cb) { struct vc4_async_flip_state *flip_state = - container_of(cb, struct vc4_async_flip_state, cb.seqno); + container_of(cb, struct vc4_async_flip_state, cb); struct vc4_bo *bo = NULL; if (flip_state->old_fb) { @@ -932,6 +929,7 @@ static void vc4_async_page_flip_seqno_complete(struct vc4_seqno_cb *cb) } vc4_async_page_flip_complete(flip_state); + dma_fence_put(fence); /* * Decrement the BO usecnt in order to keep the inc/dec @@ -950,7 +948,7 @@ static void vc4_async_page_flip_fence_complete(struct dma_fence *fence, struct dma_fence_cb *cb) { struct vc4_async_flip_state *flip_state = - container_of(cb, struct vc4_async_flip_state, cb.fence); + container_of(cb, struct vc4_async_flip_state, cb); vc4_async_page_flip_complete(flip_state); dma_fence_put(fence); @@ -961,16 +959,15 @@ static int vc4_async_set_fence_cb(struct drm_device *dev, { struct drm_framebuffer *fb = flip_state->fb; struct drm_gem_dma_object *dma_bo = drm_fb_dma_get_gem_obj(fb, 0); + dma_fence_func_t async_page_flip_complete_function; struct vc4_dev *vc4 = to_vc4_dev(dev); struct dma_fence *fence; int ret; - if (vc4->gen == VC4_GEN_4) { - struct vc4_bo *bo = to_vc4_bo(&dma_bo->base); - - return vc4_queue_seqno_cb(dev, &flip_state->cb.seqno, bo->seqno, - vc4_async_page_flip_seqno_complete); - } + if (vc4->gen == VC4_GEN_4) + async_page_flip_complete_function = vc4_async_page_flip_complete_with_cleanup; + else + async_page_flip_complete_function = vc4_async_page_flip_fence_complete; ret = dma_resv_get_singleton(dma_bo->base.resv, DMA_RESV_USAGE_READ, &fence); if (ret) @@ -978,14 +975,14 @@ static int vc4_async_set_fence_cb(struct drm_device *dev, /* If there's no fence, complete the page flip immediately */ if (!fence) { - vc4_async_page_flip_fence_complete(fence, &flip_state->cb.fence); + async_page_flip_complete_function(fence, &flip_state->cb); return 0; } /* If the fence has already been completed, complete the page flip */ - if (dma_fence_add_callback(fence, &flip_state->cb.fence, - vc4_async_page_flip_fence_complete)) - vc4_async_page_flip_fence_complete(fence, &flip_state->cb.fence); + if (dma_fence_add_callback(fence, &flip_state->cb, + async_page_flip_complete_function)) + async_page_flip_complete_function(fence, &flip_state->cb); return 0; } diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 4a078ffd9f82..221d8e01d539 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -186,11 +186,6 @@ struct vc4_dev { */ struct vc4_perfmon *active_perfmon; - /* List of struct vc4_seqno_cb for callbacks to be made from a - * workqueue when the given seqno is passed. - */ - struct list_head seqno_cb_list; - /* The memory used for storing binner tile alloc, tile state, * and overflow memory allocations. This is freed when V3D * powers down. @@ -247,16 +242,6 @@ struct vc4_dev { struct vc4_bo { struct drm_gem_dma_object base; - /* seqno of the last job to render using this BO. */ - uint64_t seqno; - - /* seqno of the last job to use the RCL to write to this BO. - * - * Note that this doesn't include binner overflow memory - * writes. - */ - uint64_t write_seqno; - bool t_format; /* List entry for the BO's position in either @@ -304,12 +289,6 @@ struct vc4_fence { #define to_vc4_fence(_fence) \ container_of_const(_fence, struct vc4_fence, base) -struct vc4_seqno_cb { - struct work_struct work; - uint64_t seqno; - void (*func)(struct vc4_seqno_cb *cb); -}; - struct vc4_v3d { struct vc4_dev *vc4; struct platform_device *pdev; @@ -695,9 +674,6 @@ struct vc4_exec_info { /* Sequence number for this bin/render job. */ uint64_t seqno; - /* Latest write_seqno of any BO that binning depends on. */ - uint64_t bin_dep_seqno; - struct dma_fence *fence; /* Last current addresses the hardware was processing when the @@ -1025,9 +1001,6 @@ void vc4_move_job_to_render(struct drm_device *dev, struct vc4_exec_info *exec); int vc4_wait_for_seqno(struct drm_device *dev, uint64_t seqno, uint64_t timeout_ns, bool interruptible); void vc4_job_handle_completed(struct vc4_dev *vc4); -int vc4_queue_seqno_cb(struct drm_device *dev, - struct vc4_seqno_cb *cb, uint64_t seqno, - void (*func)(struct vc4_seqno_cb *cb)); int vc4_gem_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index de7be9942c13..8125f87edc60 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -553,27 +553,19 @@ vc4_move_job_to_render(struct drm_device *dev, struct vc4_exec_info *exec) } static void -vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno) +vc4_attach_fences(struct vc4_exec_info *exec) { struct vc4_bo *bo; unsigned i; for (i = 0; i < exec->bo_count; i++) { bo = to_vc4_bo(exec->bo[i]); - bo->seqno = seqno; - dma_resv_add_fence(bo->base.base.resv, exec->fence, DMA_RESV_USAGE_READ); } - list_for_each_entry(bo, &exec->unref_list, unref_head) { - bo->seqno = seqno; - } - for (i = 0; i < exec->rcl_write_bo_count; i++) { bo = to_vc4_bo(&exec->rcl_write_bo[i]->base); - bo->write_seqno = seqno; - dma_resv_add_fence(bo->base.base.resv, exec->fence, DMA_RESV_USAGE_WRITE); } @@ -647,7 +639,7 @@ vc4_queue_submit(struct drm_device *dev, struct vc4_exec_info *exec, if (out_sync) drm_syncobj_replace_fence(out_sync, exec->fence); - vc4_update_bo_seqnos(exec, seqno); + vc4_attach_fences(exec); drm_exec_fini(exec_ctx); @@ -845,12 +837,6 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec) goto fail; } - /* Block waiting on any previous rendering into the CS's VBO, - * IB, or textures, so that pixels are actually written by the - * time we try to read them. - */ - ret = vc4_wait_for_seqno(dev, exec->bin_dep_seqno, ~0ull, true); - fail: kvfree(temp); return ret; @@ -909,7 +895,6 @@ void vc4_job_handle_completed(struct vc4_dev *vc4) { unsigned long irqflags; - struct vc4_seqno_cb *cb, *cb_temp; if (WARN_ON_ONCE(vc4->gen > VC4_GEN_4)) return; @@ -926,48 +911,9 @@ vc4_job_handle_completed(struct vc4_dev *vc4) spin_lock_irqsave(&vc4->job_lock, irqflags); } - list_for_each_entry_safe(cb, cb_temp, &vc4->seqno_cb_list, work.entry) { - if (cb->seqno <= vc4->finished_seqno) { - list_del_init(&cb->work.entry); - schedule_work(&cb->work); - } - } - spin_unlock_irqrestore(&vc4->job_lock, irqflags); } -static void vc4_seqno_cb_work(struct work_struct *work) -{ - struct vc4_seqno_cb *cb = container_of(work, struct vc4_seqno_cb, work); - - cb->func(cb); -} - -int vc4_queue_seqno_cb(struct drm_device *dev, - struct vc4_seqno_cb *cb, uint64_t seqno, - void (*func)(struct vc4_seqno_cb *cb)) -{ - struct vc4_dev *vc4 = to_vc4_dev(dev); - unsigned long irqflags; - - if (WARN_ON_ONCE(vc4->gen > VC4_GEN_4)) - return -ENODEV; - - cb->func = func; - INIT_WORK(&cb->work, vc4_seqno_cb_work); - - spin_lock_irqsave(&vc4->job_lock, irqflags); - if (seqno > vc4->finished_seqno) { - cb->seqno = seqno; - list_add_tail(&cb->work.entry, &vc4->seqno_cb_list); - } else { - schedule_work(&cb->work); - } - spin_unlock_irqrestore(&vc4->job_lock, irqflags); - - return 0; -} - /* Scheduled when any job has been completed, this walks the list of * jobs that had completed and unrefs their BOs and frees their exec * structs. @@ -1221,7 +1167,6 @@ int vc4_gem_init(struct drm_device *dev) INIT_LIST_HEAD(&vc4->bin_job_list); INIT_LIST_HEAD(&vc4->render_job_list); INIT_LIST_HEAD(&vc4->job_done_list); - INIT_LIST_HEAD(&vc4->seqno_cb_list); spin_lock_init(&vc4->job_lock); INIT_WORK(&vc4->hangcheck.reset_work, vc4_reset_work); diff --git a/drivers/gpu/drm/vc4/vc4_validate.c b/drivers/gpu/drm/vc4/vc4_validate.c index 5bf134968ade..1e7bdda55698 100644 --- a/drivers/gpu/drm/vc4/vc4_validate.c +++ b/drivers/gpu/drm/vc4/vc4_validate.c @@ -284,9 +284,6 @@ validate_indexed_prim_list(VALIDATE_ARGS) if (!ib) return -EINVAL; - exec->bin_dep_seqno = max(exec->bin_dep_seqno, - to_vc4_bo(&ib->base)->write_seqno); - if (offset > ib->base.size || (ib->base.size - offset) / index_size < length) { DRM_DEBUG("IB access overflow (%d + %d*%d > %zd)\n", @@ -738,11 +735,6 @@ reloc_tex(struct vc4_exec_info *exec, *validated_p0 = tex->dma_addr + p0; - if (is_cs) { - exec->bin_dep_seqno = max(exec->bin_dep_seqno, - to_vc4_bo(&tex->base)->write_seqno); - } - return true; fail: DRM_INFO("Texture p0 at %d: 0x%08x\n", sample->p_offset[0], p0); @@ -904,9 +896,6 @@ validate_gl_shader_rec(struct drm_device *dev, uint32_t stride = *(uint8_t *)(pkt_u + o + 5); uint32_t max_index; - exec->bin_dep_seqno = max(exec->bin_dep_seqno, - to_vc4_bo(&vbo->base)->write_seqno); - if (state->addr & 0x8) stride |= (*(uint32_t *)(pkt_u + 100 + i * 4)) & ~0xff; From ccb7276a6d26d6f8416e315b43b45e15ee7f29e2 Mon Sep 17 00:00:00 2001 From: Andy Strohman Date: Thu, 9 Jan 2025 02:27:56 +0000 Subject: [PATCH 0112/1443] batman-adv: fix panic during interface removal Reference counting is used to ensure that batadv_hardif_neigh_node and batadv_hard_iface are not freed before/during batadv_v_elp_throughput_metric_update work is finished. But there isn't a guarantee that the hard if will remain associated with a soft interface up until the work is finished. This fixes a crash triggered by reboot that looks like this: Call trace: batadv_v_mesh_free+0xd0/0x4dc [batman_adv] batadv_v_elp_throughput_metric_update+0x1c/0xa4 process_one_work+0x178/0x398 worker_thread+0x2e8/0x4d0 kthread+0xd8/0xdc ret_from_fork+0x10/0x20 (the batadv_v_mesh_free call is misleading, and does not actually happen) I was able to make the issue happen more reliably by changing hardif_neigh->bat_v.metric_work work to be delayed work. This allowed me to track down and confirm the fix. Cc: stable@vger.kernel.org Fixes: c833484e5f38 ("batman-adv: ELP - compute the metric based on the estimated throughput") Signed-off-by: Andy Strohman [sven@narfation.org: prevent entering batadv_v_elp_get_throughput without soft_iface] Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/bat_v_elp.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index 1d704574e6bf..fbf499bcc671 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -66,12 +66,19 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface) static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) { struct batadv_hard_iface *hard_iface = neigh->if_incoming; + struct net_device *soft_iface = hard_iface->soft_iface; struct ethtool_link_ksettings link_settings; struct net_device *real_netdev; struct station_info sinfo; u32 throughput; int ret; + /* don't query throughput when no longer associated with any + * batman-adv interface + */ + if (!soft_iface) + return BATADV_THROUGHPUT_DEFAULT_VALUE; + /* if the user specified a customised value for this interface, then * return it directly */ @@ -141,7 +148,7 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) default_throughput: if (!(hard_iface->bat_v.flags & BATADV_WARNING_DEFAULT)) { - batadv_info(hard_iface->soft_iface, + batadv_info(soft_iface, "WiFi driver or ethtool info does not provide information about link speeds on interface %s, therefore defaulting to hardcoded throughput values of %u.%1u Mbps. Consider overriding the throughput manually or checking your driver.\n", hard_iface->net_dev->name, BATADV_THROUGHPUT_DEFAULT_VALUE / 10, From 51678bb9a7fb25e44f38a4f0b1bd283fec809917 Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Mon, 13 Jan 2025 10:33:40 +0000 Subject: [PATCH 0113/1443] drm/sched: Add helper to check job dependencies MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Lets isolate scheduler internals from drivers such as pvr which currently walks the dependency array to look for fences. Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Danilo Krummrich Cc: Matthew Brost Cc: Philipp Stanner Reviewed-by: Matt Coster Acked-by: Danilo Krummrich Signed-off-by: Philipp Stanner Link: https://patchwork.freedesktop.org/patch/msgid/20250113103341.43914-1-tvrtko.ursulin@igalia.com --- drivers/gpu/drm/scheduler/sched_main.c | 23 +++++++++++++++++++++++ include/drm/gpu_scheduler.h | 3 ++- 2 files changed, 25 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index d172e3158880..a48be16ab84f 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -984,6 +984,29 @@ int drm_sched_job_add_implicit_dependencies(struct drm_sched_job *job, } EXPORT_SYMBOL(drm_sched_job_add_implicit_dependencies); +/** + * drm_sched_job_has_dependency - check whether fence is the job's dependency + * @job: scheduler job to check + * @fence: fence to look for + * + * Returns: + * True if @fence is found within the job's dependencies, or otherwise false. + */ +bool drm_sched_job_has_dependency(struct drm_sched_job *job, + struct dma_fence *fence) +{ + struct dma_fence *f; + unsigned long index; + + xa_for_each(&job->dependencies, index, f) { + if (f == fence) + return true; + } + + return false; +} +EXPORT_SYMBOL(drm_sched_job_has_dependency); + /** * drm_sched_job_cleanup - clean up scheduler job resources * @job: scheduler job to clean up diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index e2e6af8849c6..a0ff08123f07 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -564,7 +564,8 @@ int drm_sched_job_add_resv_dependencies(struct drm_sched_job *job, int drm_sched_job_add_implicit_dependencies(struct drm_sched_job *job, struct drm_gem_object *obj, bool write); - +bool drm_sched_job_has_dependency(struct drm_sched_job *job, + struct dma_fence *fence); void drm_sched_entity_modify_sched(struct drm_sched_entity *entity, struct drm_gpu_scheduler **sched_list, From cdb73451bc675392c9c76779e2fec720edafd7ab Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Mon, 13 Jan 2025 10:33:41 +0000 Subject: [PATCH 0114/1443] drm/imagination: Use the drm_sched_job_has_dependency helper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of manually peeking into the DRM scheduler implementation details lets use the previously added helper. Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Danilo Krummrich Cc: Matthew Brost Cc: Philipp Stanner Cc: Frank Binns Cc: Matt Coster Reviewed-by: Matt Coster Signed-off-by: Philipp Stanner Link: https://patchwork.freedesktop.org/patch/msgid/20250113103341.43914-2-tvrtko.ursulin@igalia.com --- drivers/gpu/drm/imagination/pvr_job.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/imagination/pvr_job.c b/drivers/gpu/drm/imagination/pvr_job.c index 618503a212a7..1cdb3cfd058d 100644 --- a/drivers/gpu/drm/imagination/pvr_job.c +++ b/drivers/gpu/drm/imagination/pvr_job.c @@ -597,8 +597,6 @@ update_job_resvs_for_each(struct pvr_job_data *job_data, u32 job_count) static bool can_combine_jobs(struct pvr_job *a, struct pvr_job *b) { struct pvr_job *geom_job = a, *frag_job = b; - struct dma_fence *fence; - unsigned long index; /* Geometry and fragment jobs can be combined if they are queued to the * same context and targeting the same HWRT. @@ -609,13 +607,9 @@ static bool can_combine_jobs(struct pvr_job *a, struct pvr_job *b) a->hwrt != b->hwrt) return false; - xa_for_each(&frag_job->base.dependencies, index, fence) { - /* We combine when we see an explicit geom -> frag dep. */ - if (&geom_job->base.s_fence->scheduled == fence) - return true; - } - - return false; + /* We combine when we see an explicit geom -> frag dep. */ + return drm_sched_job_has_dependency(&frag_job->base, + &geom_job->base.s_fence->scheduled); } static struct dma_fence * From 9bcbb6104a344d3526e185ee1e7b985509914e90 Mon Sep 17 00:00:00 2001 From: Lokesh Vutla Date: Tue, 21 Jan 2025 04:40:16 +0000 Subject: [PATCH 0115/1443] KVM: arm64: Flush hyp bss section after initialization of variables in bss To determine CPU features during initialization, the nVHE hypervisor utilizes sanitized values of the host's CPU features registers. These values, stored in u64 idaa64*_el1_sys_val variables are updated by the kvm_hyp_init_symbols() function at EL1. To ensure EL2 visibility with the MMU off, the data cache needs to be flushed after these updates. However, individually flushing each variable using kvm_flush_dcache_to_poc() is inefficient. These cpu feature variables would be part of the bss section of the hypervisor. Hence, flush the entire bss section of hypervisor once the initialization is complete. Fixes: 6c30bfb18d0b ("KVM: arm64: Add handlers for protected VM System Registers") Suggested-by: Fuad Tabba Signed-off-by: Lokesh Vutla Link: https://lore.kernel.org/r/20250121044016.2219256-1-lokeshvutla@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/arm.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index bcc4f7e92634..0725a0b50a3e 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -2400,6 +2400,13 @@ static void kvm_hyp_init_symbols(void) kvm_nvhe_sym(id_aa64smfr0_el1_sys_val) = read_sanitised_ftr_reg(SYS_ID_AA64SMFR0_EL1); kvm_nvhe_sym(__icache_flags) = __icache_flags; kvm_nvhe_sym(kvm_arm_vmid_bits) = kvm_arm_vmid_bits; + + /* + * Flush entire BSS since part of its data containing init symbols is read + * while the MMU is off. + */ + kvm_flush_dcache_to_poc(kvm_ksym_ref(__hyp_bss_start), + kvm_ksym_ref(__hyp_bss_end) - kvm_ksym_ref(__hyp_bss_start)); } static int __init kvm_hyp_init_protection(u32 hyp_va_bits) From 0a3f3f7c5da8e45e1d4adf4ed6f4e2b05912785a Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:13 +0100 Subject: [PATCH 0116/1443] drm/vkms: Switch to managed for connector MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current VKMS driver uses non-managed function to create connectors. It is not an issue yet, but in order to support multiple devices easily, convert this code to use drm and device managed helpers. Reviewed-by: Maxime Ripard Reviewed-by: Maíra Canal Reviewed-by: Thomas Zimmermann Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-1-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_output.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index 8f4bd5aef087..570823ecb28f 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -3,11 +3,11 @@ #include "vkms_drv.h" #include #include +#include #include static const struct drm_connector_funcs vkms_connector_funcs = { .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = drm_connector_cleanup, .reset = drm_atomic_helper_connector_reset, .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, @@ -75,8 +75,8 @@ int vkms_output_init(struct vkms_device *vkmsdev) } } - ret = drm_connector_init(dev, connector, &vkms_connector_funcs, - DRM_MODE_CONNECTOR_VIRTUAL); + ret = drmm_connector_init(dev, connector, &vkms_connector_funcs, + DRM_MODE_CONNECTOR_VIRTUAL, NULL); if (ret) { DRM_ERROR("Failed to init connector\n"); return ret; @@ -88,7 +88,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) DRM_MODE_ENCODER_VIRTUAL, NULL); if (ret) { DRM_ERROR("Failed to init encoder\n"); - goto err_encoder; + return ret; } encoder->possible_crtcs = drm_crtc_mask(crtc); @@ -110,9 +110,5 @@ int vkms_output_init(struct vkms_device *vkmsdev) err_attach: drm_encoder_cleanup(encoder); - -err_encoder: - drm_connector_cleanup(connector); - return ret; } From 16d22ba2debda22907b3eda4cc8ec1229136f424 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:14 +0100 Subject: [PATCH 0117/1443] drm/vkms: Switch to managed for encoder MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current VKMS driver uses non-managed function to create encoders. It is not an issue yet, but in order to support multiple devices easily, convert this code to use drm and device managed helpers. Reviewed-by: Maxime Ripard Reviewed-by: Maíra Canal Reviewed-by: Thomas Zimmermann Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-2-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_output.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index 570823ecb28f..ab9affa75b66 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -13,10 +13,6 @@ static const struct drm_connector_funcs vkms_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static const struct drm_encoder_funcs vkms_encoder_funcs = { - .destroy = drm_encoder_cleanup, -}; - static int vkms_conn_get_modes(struct drm_connector *connector) { int count; @@ -84,8 +80,8 @@ int vkms_output_init(struct vkms_device *vkmsdev) drm_connector_helper_add(connector, &vkms_conn_helper_funcs); - ret = drm_encoder_init(dev, encoder, &vkms_encoder_funcs, - DRM_MODE_ENCODER_VIRTUAL, NULL); + ret = drmm_encoder_init(dev, encoder, NULL, + DRM_MODE_ENCODER_VIRTUAL, NULL); if (ret) { DRM_ERROR("Failed to init encoder\n"); return ret; @@ -95,7 +91,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) ret = drm_connector_attach_encoder(connector, encoder); if (ret) { DRM_ERROR("Failed to attach connector to encoder\n"); - goto err_attach; + return ret; } if (vkmsdev->config->writeback) { @@ -108,7 +104,5 @@ int vkms_output_init(struct vkms_device *vkmsdev) return 0; -err_attach: - drm_encoder_cleanup(encoder); return ret; } From c367b772e6d89d8c7b560c7df7e3803ce6b8bcea Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:15 +0100 Subject: [PATCH 0118/1443] drm/managed: Add DRM-managed alloc_ordered_workqueue Add drmm_alloc_ordered_workqueue(), a helper that provides managed ordered workqueue cleanup. The workqueue will be destroyed with the final reference of the DRM device. Reviewed-by: Thomas Zimmermann Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-3-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/drm_managed.c | 8 ++++++++ include/drm/drm_managed.h | 12 ++++++++++++ 2 files changed, 20 insertions(+) diff --git a/drivers/gpu/drm/drm_managed.c b/drivers/gpu/drm/drm_managed.c index 79ce86a5bd67..cc4c463daae7 100644 --- a/drivers/gpu/drm/drm_managed.c +++ b/drivers/gpu/drm/drm_managed.c @@ -310,3 +310,11 @@ void __drmm_mutex_release(struct drm_device *dev, void *res) mutex_destroy(lock); } EXPORT_SYMBOL(__drmm_mutex_release); + +void __drmm_workqueue_release(struct drm_device *device, void *res) +{ + struct workqueue_struct *wq = res; + + destroy_workqueue(wq); +} +EXPORT_SYMBOL(__drmm_workqueue_release); diff --git a/include/drm/drm_managed.h b/include/drm/drm_managed.h index f547b09ca023..53017cc609ac 100644 --- a/include/drm/drm_managed.h +++ b/include/drm/drm_managed.h @@ -127,4 +127,16 @@ void __drmm_mutex_release(struct drm_device *dev, void *res); drmm_add_action_or_reset(dev, __drmm_mutex_release, lock); \ }) \ +void __drmm_workqueue_release(struct drm_device *device, void *wq); + +#define drmm_alloc_ordered_workqueue(dev, fmt, flags, args...) \ + ({ \ + struct workqueue_struct *wq = alloc_ordered_workqueue(fmt, flags, ##args); \ + wq ? ({ \ + int ret = drmm_add_action_or_reset(dev, __drmm_workqueue_release, wq); \ + ret ? ERR_PTR(ret) : wq; \ + }) : \ + wq; \ + }) + #endif From 8dd92e6eee590179818ba155cf02dc172e0c88c9 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:16 +0100 Subject: [PATCH 0119/1443] drm/vkms: Switch to managed for crtc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current VKMS driver uses managed function to create crtc, but don't use it to properly clean the crtc workqueue. It is not an issue yet, but in order to support multiple devices easily, convert this code to use drm and device managed helpers. Acked-by: Maxime Ripard Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-4-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_crtc.c | 5 ++++- drivers/gpu/drm/vkms/vkms_drv.c | 9 --------- 2 files changed, 4 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c index 28a57ae109fc..434c35d5e947 100644 --- a/drivers/gpu/drm/vkms/vkms_crtc.c +++ b/drivers/gpu/drm/vkms/vkms_crtc.c @@ -4,6 +4,7 @@ #include #include +#include #include #include @@ -296,7 +297,9 @@ int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, spin_lock_init(&vkms_out->lock); spin_lock_init(&vkms_out->composer_lock); - vkms_out->composer_workq = alloc_ordered_workqueue("vkms_composer", 0); + vkms_out->composer_workq = drmm_alloc_ordered_workqueue(dev, "vkms_composer", 0); + if (IS_ERR(vkms_out->composer_workq)) + return PTR_ERR(vkms_out->composer_workq); if (!vkms_out->composer_workq) return -ENOMEM; diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c index e0409aba9349..7c142bfc3bd9 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.c +++ b/drivers/gpu/drm/vkms/vkms_drv.c @@ -53,14 +53,6 @@ MODULE_PARM_DESC(enable_overlay, "Enable/Disable overlay support"); DEFINE_DRM_GEM_FOPS(vkms_driver_fops); -static void vkms_release(struct drm_device *dev) -{ - struct vkms_device *vkms = drm_device_to_vkms_device(dev); - - if (vkms->output.composer_workq) - destroy_workqueue(vkms->output.composer_workq); -} - static void vkms_atomic_commit_tail(struct drm_atomic_state *old_state) { struct drm_device *dev = old_state->dev; @@ -108,7 +100,6 @@ static const struct drm_debugfs_info vkms_config_debugfs_list[] = { static const struct drm_driver vkms_driver = { .driver_features = DRIVER_MODESET | DRIVER_ATOMIC | DRIVER_GEM, - .release = vkms_release, .fops = &vkms_driver_fops, DRM_GEM_SHMEM_DRIVER_OPS, DRM_FBDEV_SHMEM_DRIVER_OPS, From 135d8fc7af44c52083e18ccb24d56383d301f741 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:17 +0100 Subject: [PATCH 0120/1443] drm: writeback: Create an helper for drm_writeback_connector initialization As the old drm and the new drmm variants of drm_writeback_connector requires almost the same initialization, create an internal helper to do most of the initialization work. Currently there is no cleanup function for writeback connectors. To allows implementation of drmm variant of writeback connector, create a cleanup function that can be used to properly remove all the writeback-specific properties and allocations. This also introduce an helper to cleanup only the drm_writeback_connector properties, so it can be used during initialization to cleanup in case of failure. Reviewed-by: Maxime Ripard Acked-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-5-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/drm_writeback.c | 127 ++++++++++++++++++++------------ 1 file changed, 81 insertions(+), 46 deletions(-) diff --git a/drivers/gpu/drm/drm_writeback.c b/drivers/gpu/drm/drm_writeback.c index 33a3c98a962d..057af96dafeb 100644 --- a/drivers/gpu/drm/drm_writeback.c +++ b/drivers/gpu/drm/drm_writeback.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -195,6 +196,77 @@ int drm_writeback_connector_init(struct drm_device *dev, } EXPORT_SYMBOL(drm_writeback_connector_init); +/** + * drm_writeback_connector_init_with_encoder - Initialize a writeback connector with + * a custom encoder + * + * @dev: DRM device + * @wb_connector: Writeback connector to initialize + * @enc: handle to the already initialized drm encoder + * @formats: Array of supported pixel formats for the writeback engine + * @n_formats: Length of the formats array + * + * This function creates the writeback-connector-specific properties if they + * have not been already created, initializes the connector as + * type DRM_MODE_CONNECTOR_WRITEBACK, and correctly initializes the property + * values. + * + * This function assumes that the drm_writeback_connector's encoder has already been + * created and initialized before invoking this function. + * + * In addition, this function also assumes that callers of this API will manage + * assigning the encoder helper functions, possible_crtcs and any other encoder + * specific operation. + * + * Returns: 0 on success, or a negative error code + */ +static int __drm_writeback_connector_init(struct drm_device *dev, + struct drm_writeback_connector *wb_connector, + struct drm_encoder *enc, const u32 *formats, + int n_formats) +{ + struct drm_connector *connector = &wb_connector->base; + struct drm_mode_config *config = &dev->mode_config; + struct drm_property_blob *blob; + int ret = create_writeback_properties(dev); + + if (ret != 0) + return ret; + + connector->interlace_allowed = 0; + + ret = drm_connector_attach_encoder(connector, enc); + if (ret) + return ret; + + blob = drm_property_create_blob(dev, n_formats * sizeof(*formats), + formats); + if (IS_ERR(blob)) + return PTR_ERR(blob); + + INIT_LIST_HEAD(&wb_connector->job_queue); + spin_lock_init(&wb_connector->job_lock); + + wb_connector->fence_context = dma_fence_context_alloc(1); + spin_lock_init(&wb_connector->fence_lock); + snprintf(wb_connector->timeline_name, + sizeof(wb_connector->timeline_name), + "CONNECTOR:%d-%s", connector->base.id, connector->name); + + drm_object_attach_property(&connector->base, + config->writeback_out_fence_ptr_property, 0); + + drm_object_attach_property(&connector->base, + config->writeback_fb_id_property, 0); + + drm_object_attach_property(&connector->base, + config->writeback_pixel_formats_property, + blob->base.id); + wb_connector->pixel_formats_blob_ptr = blob; + + return 0; +} + /** * drm_writeback_connector_init_with_encoder - Initialize a writeback connector with * a custom encoder @@ -225,61 +297,24 @@ EXPORT_SYMBOL(drm_writeback_connector_init); * Returns: 0 on success, or a negative error code */ int drm_writeback_connector_init_with_encoder(struct drm_device *dev, - struct drm_writeback_connector *wb_connector, struct drm_encoder *enc, - const struct drm_connector_funcs *con_funcs, const u32 *formats, - int n_formats) + struct drm_writeback_connector *wb_connector, + struct drm_encoder *enc, + const struct drm_connector_funcs *con_funcs, + const u32 *formats, int n_formats) { - struct drm_property_blob *blob; struct drm_connector *connector = &wb_connector->base; - struct drm_mode_config *config = &dev->mode_config; - int ret = create_writeback_properties(dev); - - if (ret != 0) - return ret; - - blob = drm_property_create_blob(dev, n_formats * sizeof(*formats), - formats); - if (IS_ERR(blob)) - return PTR_ERR(blob); - - - connector->interlace_allowed = 0; + int ret; ret = drm_connector_init(dev, connector, con_funcs, DRM_MODE_CONNECTOR_WRITEBACK); if (ret) - goto connector_fail; + return ret; - ret = drm_connector_attach_encoder(connector, enc); + ret = __drm_writeback_connector_init(dev, wb_connector, enc, formats, + n_formats); if (ret) - goto attach_fail; + drm_connector_cleanup(connector); - INIT_LIST_HEAD(&wb_connector->job_queue); - spin_lock_init(&wb_connector->job_lock); - - wb_connector->fence_context = dma_fence_context_alloc(1); - spin_lock_init(&wb_connector->fence_lock); - snprintf(wb_connector->timeline_name, - sizeof(wb_connector->timeline_name), - "CONNECTOR:%d-%s", connector->base.id, connector->name); - - drm_object_attach_property(&connector->base, - config->writeback_out_fence_ptr_property, 0); - - drm_object_attach_property(&connector->base, - config->writeback_fb_id_property, 0); - - drm_object_attach_property(&connector->base, - config->writeback_pixel_formats_property, - blob->base.id); - wb_connector->pixel_formats_blob_ptr = blob; - - return 0; - -attach_fail: - drm_connector_cleanup(connector); -connector_fail: - drm_property_blob_put(blob); return ret; } EXPORT_SYMBOL(drm_writeback_connector_init_with_encoder); From 2f3f4a73631b160e44ab13d497f7be62264d47ac Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:18 +0100 Subject: [PATCH 0121/1443] drm: writeback: Add missing cleanup in case of initialization failure The current implementation of drm_writeback_connector initialization does not properly clean up all resources in case of failure (allocated properties and possible_encoders). Add this cleaning in case of failure. Acked-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-6-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/drm_writeback.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/drm_writeback.c b/drivers/gpu/drm/drm_writeback.c index 057af96dafeb..4cdc6cdcf76b 100644 --- a/drivers/gpu/drm/drm_writeback.c +++ b/drivers/gpu/drm/drm_writeback.c @@ -196,6 +196,22 @@ int drm_writeback_connector_init(struct drm_device *dev, } EXPORT_SYMBOL(drm_writeback_connector_init); +static void delete_writeback_properties(struct drm_device *dev) +{ + if (dev->mode_config.writeback_pixel_formats_property) { + drm_property_destroy(dev, dev->mode_config.writeback_pixel_formats_property); + dev->mode_config.writeback_pixel_formats_property = NULL; + } + if (dev->mode_config.writeback_out_fence_ptr_property) { + drm_property_destroy(dev, dev->mode_config.writeback_out_fence_ptr_property); + dev->mode_config.writeback_out_fence_ptr_property = NULL; + } + if (dev->mode_config.writeback_fb_id_property) { + drm_property_destroy(dev, dev->mode_config.writeback_fb_id_property); + dev->mode_config.writeback_fb_id_property = NULL; + } +} + /** * drm_writeback_connector_init_with_encoder - Initialize a writeback connector with * a custom encoder @@ -231,18 +247,20 @@ static int __drm_writeback_connector_init(struct drm_device *dev, int ret = create_writeback_properties(dev); if (ret != 0) - return ret; + goto failed_properties; connector->interlace_allowed = 0; ret = drm_connector_attach_encoder(connector, enc); if (ret) - return ret; + goto failed_properties; blob = drm_property_create_blob(dev, n_formats * sizeof(*formats), formats); - if (IS_ERR(blob)) - return PTR_ERR(blob); + if (IS_ERR(blob)) { + ret = PTR_ERR(blob); + goto failed_properties; + } INIT_LIST_HEAD(&wb_connector->job_queue); spin_lock_init(&wb_connector->job_lock); @@ -265,6 +283,9 @@ static int __drm_writeback_connector_init(struct drm_device *dev, wb_connector->pixel_formats_blob_ptr = blob; return 0; +failed_properties: + delete_writeback_properties(dev); + return ret; } /** From 1914ba2b91ea8eff674e2369f610bb6bb9056745 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:19 +0100 Subject: [PATCH 0122/1443] drm: writeback: Create drmm variants for drm_writeback_connector initialization To allows driver to only use drmm objects, add helper to create drm_writeback_connectors with automated lifetime management. Acked-by: Thomas Zimmermann Acked-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-7-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/drm_writeback.c | 74 +++++++++++++++++++++++++++++++++ include/drm/drm_writeback.h | 6 +++ 2 files changed, 80 insertions(+) diff --git a/drivers/gpu/drm/drm_writeback.c b/drivers/gpu/drm/drm_writeback.c index 4cdc6cdcf76b..3628fbef7752 100644 --- a/drivers/gpu/drm/drm_writeback.c +++ b/drivers/gpu/drm/drm_writeback.c @@ -340,6 +340,80 @@ int drm_writeback_connector_init_with_encoder(struct drm_device *dev, } EXPORT_SYMBOL(drm_writeback_connector_init_with_encoder); +/** + * drm_writeback_connector_cleanup - Cleanup the writeback connector + * @dev: DRM device + * @wb_connector: Pointer to the writeback connector to clean up + * + * This will decrement the reference counter of blobs and destroy properties. It + * will also clean the remaining jobs in this writeback connector. Caution: This helper will not + * clean up the attached encoder and the drm_connector. + */ +static void drm_writeback_connector_cleanup(struct drm_device *dev, + struct drm_writeback_connector *wb_connector) +{ + unsigned long flags; + struct drm_writeback_job *pos, *n; + + delete_writeback_properties(dev); + drm_property_blob_put(wb_connector->pixel_formats_blob_ptr); + + spin_lock_irqsave(&wb_connector->job_lock, flags); + list_for_each_entry_safe(pos, n, &wb_connector->job_queue, list_entry) { + drm_writeback_cleanup_job(pos); + list_del(&pos->list_entry); + } + spin_unlock_irqrestore(&wb_connector->job_lock, flags); +} + +/** + * drmm_writeback_connector_init - Initialize a writeback connector with + * a custom encoder + * + * @dev: DRM device + * @wb_connector: Writeback connector to initialize + * @con_funcs: Connector funcs vtable + * @enc: Encoder to connect this writeback connector + * @formats: Array of supported pixel formats for the writeback engine + * @n_formats: Length of the formats array + * + * This function initialize a writeback connector and register its cleanup. + * + * This function creates the writeback-connector-specific properties if they + * have not been already created, initializes the connector as + * type DRM_MODE_CONNECTOR_WRITEBACK, and correctly initializes the property + * values. + * + * Returns: 0 on success, or a negative error code + */ +int drmm_writeback_connector_init(struct drm_device *dev, + struct drm_writeback_connector *wb_connector, + const struct drm_connector_funcs *con_funcs, + struct drm_encoder *enc, + const u32 *formats, int n_formats) +{ + struct drm_connector *connector = &wb_connector->base; + int ret; + + ret = drmm_connector_init(dev, connector, con_funcs, + DRM_MODE_CONNECTOR_WRITEBACK, NULL); + if (ret) + return ret; + + ret = __drm_writeback_connector_init(dev, wb_connector, enc, formats, + n_formats); + if (ret) + return ret; + + ret = drmm_add_action_or_reset(dev, (void *)drm_writeback_connector_cleanup, + wb_connector); + if (ret) + return ret; + + return 0; +} +EXPORT_SYMBOL(drmm_writeback_connector_init); + int drm_writeback_set_fb(struct drm_connector_state *conn_state, struct drm_framebuffer *fb) { diff --git a/include/drm/drm_writeback.h b/include/drm/drm_writeback.h index 17e576c80169..c380a7b8f55a 100644 --- a/include/drm/drm_writeback.h +++ b/include/drm/drm_writeback.h @@ -161,6 +161,12 @@ int drm_writeback_connector_init_with_encoder(struct drm_device *dev, const struct drm_connector_funcs *con_funcs, const u32 *formats, int n_formats); +int drmm_writeback_connector_init(struct drm_device *dev, + struct drm_writeback_connector *wb_connector, + const struct drm_connector_funcs *con_funcs, + struct drm_encoder *enc, + const u32 *formats, int n_formats); + int drm_writeback_set_fb(struct drm_connector_state *conn_state, struct drm_framebuffer *fb); From 23fdf4308988b8aee2bb7cf8b77153f822d1fb3a Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Thu, 16 Jan 2025 18:47:20 +0100 Subject: [PATCH 0123/1443] drm/vkms: Switch to managed for writeback connector MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current VKMS driver uses non-managed function to create writeback connectors. It is not an issue yet, but in order to support multiple devices easily, convert this code to use drm and device managed helpers. Acked-by: Thomas Zimmermann Acked-by: Maxime Ripard Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250116-google-vkms-managed-v9-8-3e4ae1bd05a0@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_drv.h | 3 ++- drivers/gpu/drm/vkms/vkms_output.c | 2 +- drivers/gpu/drm/vkms/vkms_writeback.c | 21 +++++++++++++-------- 3 files changed, 16 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index 00541eff3d1b..46ac36aebb27 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -179,6 +179,7 @@ struct vkms_output { struct drm_encoder encoder; struct drm_connector connector; struct drm_writeback_connector wb_connector; + struct drm_encoder wb_encoder; struct hrtimer vblank_hrtimer; ktime_t period_ns; struct workqueue_struct *composer_workq; @@ -275,6 +276,6 @@ void vkms_set_composer(struct vkms_output *out, bool enabled); void vkms_writeback_row(struct vkms_writeback_job *wb, const struct line_buffer *src_buffer, int y); /* Writeback */ -int vkms_enable_writeback_connector(struct vkms_device *vkmsdev); +int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, struct drm_crtc *crtc); #endif /* _VKMS_DRV_H_ */ diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index ab9affa75b66..de817e279486 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -95,7 +95,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) } if (vkmsdev->config->writeback) { - writeback = vkms_enable_writeback_connector(vkmsdev); + writeback = vkms_enable_writeback_connector(vkmsdev, crtc); if (writeback) DRM_ERROR("Failed to init writeback connector\n"); } diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c index 79918b44fedd..981975c2b0a0 100644 --- a/drivers/gpu/drm/vkms/vkms_writeback.c +++ b/drivers/gpu/drm/vkms/vkms_writeback.c @@ -24,7 +24,6 @@ static const u32 vkms_wb_formats[] = { static const struct drm_connector_funcs vkms_wb_connector_funcs = { .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = drm_connector_cleanup, .reset = drm_atomic_helper_connector_reset, .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, @@ -163,16 +162,22 @@ static const struct drm_connector_helper_funcs vkms_wb_conn_helper_funcs = { .atomic_check = vkms_wb_atomic_check, }; -int vkms_enable_writeback_connector(struct vkms_device *vkmsdev) +int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, struct drm_crtc *crtc) { struct drm_writeback_connector *wb = &vkmsdev->output.wb_connector; + int ret; + + ret = drmm_encoder_init(&vkmsdev->drm, &vkmsdev->output.wb_encoder, + NULL, DRM_MODE_ENCODER_VIRTUAL, NULL); + if (ret) + return ret; + vkmsdev->output.wb_encoder.possible_crtcs |= drm_crtc_mask(crtc); drm_connector_helper_add(&wb->base, &vkms_wb_conn_helper_funcs); - return drm_writeback_connector_init(&vkmsdev->drm, wb, - &vkms_wb_connector_funcs, - NULL, - vkms_wb_formats, - ARRAY_SIZE(vkms_wb_formats), - 1); + return drmm_writeback_connector_init(&vkmsdev->drm, wb, + &vkms_wb_connector_funcs, + &vkmsdev->output.wb_encoder, + vkms_wb_formats, + ARRAY_SIZE(vkms_wb_formats)); } From b0a76faea6b1492e480a69ef1a6cd19e30e7d60d Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Fri, 17 Jan 2025 10:04:27 +0100 Subject: [PATCH 0124/1443] drm/vkms: Switch to dynamic allocation for connector MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A specific allocation for the connector is not strictly necessary at this point, but in order to implement dynamic configuration of VKMS (configFS), it will be easier to have one allocation per connector. Reviewed-by: Maxime Ripard Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250117-b4-vkms-allocated-v4-1-8ec8fd21aaf6@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_drv.h | 1 - drivers/gpu/drm/vkms/vkms_output.c | 8 +++++++- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index 46ac36aebb27..afa625457b61 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -177,7 +177,6 @@ struct vkms_crtc_state { struct vkms_output { struct drm_crtc crtc; struct drm_encoder encoder; - struct drm_connector connector; struct drm_writeback_connector wb_connector; struct drm_encoder wb_encoder; struct hrtimer vblank_hrtimer; diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index de817e279486..a41d7a29a377 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -32,7 +32,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) { struct vkms_output *output = &vkmsdev->output; struct drm_device *dev = &vkmsdev->drm; - struct drm_connector *connector = &output->connector; + struct drm_connector *connector; struct drm_encoder *encoder = &output->encoder; struct drm_crtc *crtc = &output->crtc; struct vkms_plane *primary, *overlay, *cursor = NULL; @@ -71,6 +71,12 @@ int vkms_output_init(struct vkms_device *vkmsdev) } } + connector = drmm_kzalloc(dev, sizeof(*connector), GFP_KERNEL); + if (!connector) { + DRM_ERROR("Failed to allocate connector\n"); + return -ENOMEM; + } + ret = drmm_connector_init(dev, connector, &vkms_connector_funcs, DRM_MODE_CONNECTOR_VIRTUAL, NULL); if (ret) { From 45a4778415736fe4649a9fac2323091cdd710d86 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Fri, 17 Jan 2025 10:04:28 +0100 Subject: [PATCH 0125/1443] drm/vkms: Switch to dynamic allocation for encoder MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A specific allocation for the encoder is not strictly necessary at this point, but in order to implement dynamic configuration of VKMS (configFS), it will be easier to have one allocation per encoder. Reviewed-by: Maxime Ripard Reviewed-by: José Expósito Link: https://patchwork.freedesktop.org/patch/msgid/20250117-b4-vkms-allocated-v4-2-8ec8fd21aaf6@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_drv.h | 1 - drivers/gpu/drm/vkms/vkms_output.c | 7 ++++++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index afa625457b61..333983bcf8d4 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -176,7 +176,6 @@ struct vkms_crtc_state { */ struct vkms_output { struct drm_crtc crtc; - struct drm_encoder encoder; struct drm_writeback_connector wb_connector; struct drm_encoder wb_encoder; struct hrtimer vblank_hrtimer; diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index a41d7a29a377..21ca975e424d 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -33,7 +33,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) struct vkms_output *output = &vkmsdev->output; struct drm_device *dev = &vkmsdev->drm; struct drm_connector *connector; - struct drm_encoder *encoder = &output->encoder; + struct drm_encoder *encoder; struct drm_crtc *crtc = &output->crtc; struct vkms_plane *primary, *overlay, *cursor = NULL; int ret; @@ -86,6 +86,11 @@ int vkms_output_init(struct vkms_device *vkmsdev) drm_connector_helper_add(connector, &vkms_conn_helper_funcs); + encoder = drmm_kzalloc(dev, sizeof(*encoder), GFP_KERNEL); + if (!encoder) { + DRM_ERROR("Failed to allocate encoder\n"); + return -ENOMEM; + } ret = drmm_encoder_init(dev, encoder, NULL, DRM_MODE_ENCODER_VIRTUAL, NULL); if (ret) { From 49a167c393b0ceb592b9d2e65cc4f46bcc707108 Mon Sep 17 00:00:00 2001 From: Louis Chauvet Date: Fri, 17 Jan 2025 10:04:29 +0100 Subject: [PATCH 0126/1443] drm/vkms: Switch to dynamic allocation for CRTC A specific allocation for the CRTC is not strictly necessary at this point, but in order to implement dynamic configuration of VKMS (configFS), it will be easier to have one allocation per CRTC. Reviewed-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20250117-b4-vkms-allocated-v4-3-8ec8fd21aaf6@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_crtc.c | 32 ++++++++++++++------------- drivers/gpu/drm/vkms/vkms_drv.h | 8 +++---- drivers/gpu/drm/vkms/vkms_output.c | 22 +++++++++--------- drivers/gpu/drm/vkms/vkms_writeback.c | 23 ++++++++++--------- 4 files changed, 45 insertions(+), 40 deletions(-) diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c index 434c35d5e947..cf229aec71c3 100644 --- a/drivers/gpu/drm/vkms/vkms_crtc.c +++ b/drivers/gpu/drm/vkms/vkms_crtc.c @@ -84,9 +84,7 @@ static bool vkms_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error, ktime_t *vblank_time, bool in_vblank_irq) { - struct drm_device *dev = crtc->dev; - struct vkms_device *vkmsdev = drm_device_to_vkms_device(dev); - struct vkms_output *output = &vkmsdev->output; + struct vkms_output *output = drm_crtc_to_vkms_output(crtc); struct drm_vblank_crtc *vblank = drm_crtc_vblank_crtc(crtc); if (!READ_ONCE(vblank->enabled)) { @@ -271,25 +269,29 @@ static const struct drm_crtc_helper_funcs vkms_crtc_helper_funcs = { .atomic_disable = vkms_crtc_atomic_disable, }; -int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, - struct drm_plane *primary, struct drm_plane *cursor) +struct vkms_output *vkms_crtc_init(struct drm_device *dev, struct drm_plane *primary, + struct drm_plane *cursor) { - struct vkms_output *vkms_out = drm_crtc_to_vkms_output(crtc); + struct vkms_output *vkms_out; + struct drm_crtc *crtc; int ret; - ret = drmm_crtc_init_with_planes(dev, crtc, primary, cursor, - &vkms_crtc_funcs, NULL); - if (ret) { - DRM_ERROR("Failed to init CRTC\n"); - return ret; + vkms_out = drmm_crtc_alloc_with_planes(dev, struct vkms_output, crtc, + primary, cursor, + &vkms_crtc_funcs, NULL); + if (IS_ERR(vkms_out)) { + DRM_DEV_ERROR(dev->dev, "Failed to init CRTC\n"); + return vkms_out; } + crtc = &vkms_out->crtc; + drm_crtc_helper_add(crtc, &vkms_crtc_helper_funcs); ret = drm_mode_crtc_set_gamma_size(crtc, VKMS_LUT_SIZE); if (ret) { DRM_ERROR("Failed to set gamma size\n"); - return ret; + return ERR_PTR(ret); } drm_crtc_enable_color_mgmt(crtc, 0, false, VKMS_LUT_SIZE); @@ -299,9 +301,9 @@ int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, vkms_out->composer_workq = drmm_alloc_ordered_workqueue(dev, "vkms_composer", 0); if (IS_ERR(vkms_out->composer_workq)) - return PTR_ERR(vkms_out->composer_workq); + return ERR_CAST(vkms_out->composer_workq); if (!vkms_out->composer_workq) - return -ENOMEM; + return ERR_PTR(-ENOMEM); - return ret; + return vkms_out; } diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index 333983bcf8d4..abbb652be2b5 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -215,7 +215,6 @@ struct vkms_config { struct vkms_device { struct drm_device drm; struct platform_device *platform; - struct vkms_output output; const struct vkms_config *config; }; @@ -242,8 +241,9 @@ struct vkms_device { * @primary: primary plane to attach to the CRTC * @cursor: plane to attach to the CRTC */ -int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, - struct drm_plane *primary, struct drm_plane *cursor); +struct vkms_output *vkms_crtc_init(struct drm_device *dev, + struct drm_plane *primary, + struct drm_plane *cursor); /** * vkms_output_init() - Initialize all sub-components needed for a VKMS device. @@ -274,6 +274,6 @@ void vkms_set_composer(struct vkms_output *out, bool enabled); void vkms_writeback_row(struct vkms_writeback_job *wb, const struct line_buffer *src_buffer, int y); /* Writeback */ -int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, struct drm_crtc *crtc); +int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, struct vkms_output *vkms_out); #endif /* _VKMS_DRV_H_ */ diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index 21ca975e424d..22f0d678af3a 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -30,11 +30,10 @@ static const struct drm_connector_helper_funcs vkms_conn_helper_funcs = { int vkms_output_init(struct vkms_device *vkmsdev) { - struct vkms_output *output = &vkmsdev->output; struct drm_device *dev = &vkmsdev->drm; struct drm_connector *connector; struct drm_encoder *encoder; - struct drm_crtc *crtc = &output->crtc; + struct vkms_output *output; struct vkms_plane *primary, *overlay, *cursor = NULL; int ret; int writeback; @@ -56,9 +55,12 @@ int vkms_output_init(struct vkms_device *vkmsdev) return PTR_ERR(cursor); } - ret = vkms_crtc_init(dev, crtc, &primary->base, &cursor->base); - if (ret) - return ret; + output = vkms_crtc_init(dev, &primary->base, + cursor ? &cursor->base : NULL); + if (IS_ERR(output)) { + DRM_ERROR("Failed to allocate CRTC\n"); + return PTR_ERR(output); + } if (vkmsdev->config->overlay) { for (n = 0; n < NUM_OVERLAY_PLANES; n++) { @@ -67,7 +69,7 @@ int vkms_output_init(struct vkms_device *vkmsdev) DRM_DEV_ERROR(dev->dev, "Failed to init vkms plane\n"); return PTR_ERR(overlay); } - overlay->base.possible_crtcs = drm_crtc_mask(crtc); + overlay->base.possible_crtcs = drm_crtc_mask(&output->crtc); } } @@ -97,23 +99,23 @@ int vkms_output_init(struct vkms_device *vkmsdev) DRM_ERROR("Failed to init encoder\n"); return ret; } - encoder->possible_crtcs = drm_crtc_mask(crtc); + encoder->possible_crtcs = drm_crtc_mask(&output->crtc); + /* Attach the encoder and the connector */ ret = drm_connector_attach_encoder(connector, encoder); if (ret) { DRM_ERROR("Failed to attach connector to encoder\n"); return ret; } + /* Initialize the writeback component */ if (vkmsdev->config->writeback) { - writeback = vkms_enable_writeback_connector(vkmsdev, crtc); + writeback = vkms_enable_writeback_connector(vkmsdev, output); if (writeback) DRM_ERROR("Failed to init writeback connector\n"); } drm_mode_config_reset(dev); - return 0; - return ret; } diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c index 981975c2b0a0..e9b5c74d7c58 100644 --- a/drivers/gpu/drm/vkms/vkms_writeback.c +++ b/drivers/gpu/drm/vkms/vkms_writeback.c @@ -105,7 +105,9 @@ static void vkms_wb_cleanup_job(struct drm_writeback_connector *connector, struct drm_writeback_job *job) { struct vkms_writeback_job *vkmsjob = job->priv; - struct vkms_device *vkmsdev; + struct vkms_output *vkms_output = container_of(connector, + struct vkms_output, + wb_connector); if (!job->fb) return; @@ -114,8 +116,7 @@ static void vkms_wb_cleanup_job(struct drm_writeback_connector *connector, drm_framebuffer_put(vkmsjob->wb_frame_info.fb); - vkmsdev = drm_device_to_vkms_device(job->fb->dev); - vkms_set_composer(&vkmsdev->output, false); + vkms_set_composer(vkms_output, false); kfree(vkmsjob); } @@ -124,8 +125,7 @@ static void vkms_wb_atomic_commit(struct drm_connector *conn, { struct drm_connector_state *connector_state = drm_atomic_get_new_connector_state(state, conn); - struct vkms_device *vkmsdev = drm_device_to_vkms_device(conn->dev); - struct vkms_output *output = &vkmsdev->output; + struct vkms_output *output = drm_crtc_to_vkms_output(connector_state->crtc); struct drm_writeback_connector *wb_conn = &output->wb_connector; struct drm_connector_state *conn_state = wb_conn->base.state; struct vkms_crtc_state *crtc_state = output->composer_state; @@ -139,7 +139,7 @@ static void vkms_wb_atomic_commit(struct drm_connector *conn, if (!conn_state) return; - vkms_set_composer(&vkmsdev->output, true); + vkms_set_composer(output, true); active_wb = conn_state->writeback_job->priv; wb_frame_info = &active_wb->wb_frame_info; @@ -162,22 +162,23 @@ static const struct drm_connector_helper_funcs vkms_wb_conn_helper_funcs = { .atomic_check = vkms_wb_atomic_check, }; -int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, struct drm_crtc *crtc) +int vkms_enable_writeback_connector(struct vkms_device *vkmsdev, + struct vkms_output *vkms_output) { - struct drm_writeback_connector *wb = &vkmsdev->output.wb_connector; + struct drm_writeback_connector *wb = &vkms_output->wb_connector; int ret; - ret = drmm_encoder_init(&vkmsdev->drm, &vkmsdev->output.wb_encoder, + ret = drmm_encoder_init(&vkmsdev->drm, &vkms_output->wb_encoder, NULL, DRM_MODE_ENCODER_VIRTUAL, NULL); if (ret) return ret; - vkmsdev->output.wb_encoder.possible_crtcs |= drm_crtc_mask(crtc); + vkms_output->wb_encoder.possible_crtcs |= drm_crtc_mask(&vkms_output->crtc); drm_connector_helper_add(&wb->base, &vkms_wb_conn_helper_funcs); return drmm_writeback_connector_init(&vkmsdev->drm, wb, &vkms_wb_connector_funcs, - &vkmsdev->output.wb_encoder, + &vkms_output->wb_encoder, vkms_wb_formats, ARRAY_SIZE(vkms_wb_formats)); } From 3429dd57f0deb1a602c2624a1dd7c4c11b6c4734 Mon Sep 17 00:00:00 2001 From: K Prateek Nayak Date: Fri, 17 Jan 2025 10:58:52 +0000 Subject: [PATCH 0127/1443] sched/fair: Fix inaccurate h_nr_runnable accounting with delayed dequeue set_delayed() adjusts cfs_rq->h_nr_runnable for the hierarchy when an entity is delayed irrespective of whether the entity corresponds to a task or a cfs_rq. Consider the following scenario: root / \ A B (*) delayed since B is no longer eligible on root | | Task0 Task1 <--- dequeue_task_fair() - task blocks When Task1 blocks (dequeue_entity() for task's se returns true), dequeue_entities() will continue adjusting cfs_rq->h_nr_* for the hierarchy of Task1. However, when the sched_entity corresponding to cfs_rq B is delayed, set_delayed() will adjust the h_nr_runnable for the hierarchy too leading to both dequeue_entity() and set_delayed() decrementing h_nr_runnable for the dequeue of the same task. A SCHED_WARN_ON() to inspect h_nr_runnable post its update in dequeue_entities() like below: cfs_rq->h_nr_runnable -= h_nr_runnable; SCHED_WARN_ON(((int) cfs_rq->h_nr_runnable) < 0); is consistently tripped when running wakeup intensive workloads like hackbench in a cgroup. This error is self correcting since cfs_rq are per-cpu and cannot migrate. The entitiy is either picked for full dequeue or is requeued when a task wakes up below it. Both those paths call clear_delayed() which again increments h_nr_runnable of the hierarchy without considering if the entity corresponds to a task or not. h_nr_runnable will eventually reflect the correct value however in the interim, the incorrect values can still influence PELT calculation which uses se->runnable_weight or cfs_rq->h_nr_runnable. Since only delayed tasks take the early return path in dequeue_entities() and enqueue_task_fair(), adjust the h_nr_runnable in {set,clear}_delayed() only when a task is delayed as this path skips the h_nr_* update loops and returns early. For entities corresponding to cfs_rq, the h_nr_* update loop in the caller will do the right thing. Fixes: 76f2f783294d ("sched/eevdf: More PELT vs DELAYED_DEQUEUE") Signed-off-by: K Prateek Nayak Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Gautham R. Shenoy Tested-by: Swapnil Sapkal Link: https://lkml.kernel.org/r/20250117105852.23908-1-kprateek.nayak@amd.com --- kernel/sched/fair.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 26958431deb7..f4e4d3ed943c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5372,6 +5372,15 @@ static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); static void set_delayed(struct sched_entity *se) { se->sched_delayed = 1; + + /* + * Delayed se of cfs_rq have no tasks queued on them. + * Do not adjust h_nr_runnable since dequeue_entities() + * will account it for blocked tasks. + */ + if (!entity_is_task(se)) + return; + for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); @@ -5384,6 +5393,16 @@ static void set_delayed(struct sched_entity *se) static void clear_delayed(struct sched_entity *se) { se->sched_delayed = 0; + + /* + * Delayed se of cfs_rq have no tasks queued on them. + * Do not adjust h_nr_runnable since a dequeue has + * already accounted for it or an enqueue of a task + * below it will account for it in enqueue_task_fair(). + */ + if (!entity_is_task(se)) + return; + for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); From f3b59457808f61d88178b0afa67cbd017d7ce79e Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 10 Dec 2024 09:31:11 +0100 Subject: [PATCH 0128/1443] drm/xe: Do not attempt to bootstrap VF in execlists mode It was mentioned in a review that there is a possibility of choosing to load the module with VF in execlists mode. Of course this doesn't work, just bomb out as hard as possible. Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20241210083111.230484-12-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c index cca5d5732802..6671030439fd 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c @@ -213,6 +213,9 @@ int xe_gt_sriov_vf_bootstrap(struct xe_gt *gt) { int err; + if (!xe_device_uc_enabled(gt_to_xe(gt))) + return -ENODEV; + err = vf_reset_guc_state(gt); if (unlikely(err)) return err; From a46ea12eca59fd3741ddfec3042d43f87fadf58f Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Fri, 17 Jan 2025 14:38:27 -0500 Subject: [PATCH 0129/1443] drm/xe/uapi: Fix documentation indentation Fix these issues: Documentation/gpu/driver-uapi:29: include/uapi/drm/xe_drm.h:817: WARNING: +Bullet list ends without a blank line; unexpected unindent. Documentation/gpu/driver-uapi:29: include/uapi/drm/xe_drm.h:835: WARNING: +Definition list ends without a blank line; unexpected unindent. Fixes: 75d37750a753 ("drm/xe/mmap: Add mmap support for PCI memory barrier") Reported-by: Stephen Rothwell Closes: https://lore.kernel.org/intel-xe/20250117164023.3fdc00b9@canb.auug.org.au/ Cc: Tejas Upadhyay Tested-by: Bagas Sanjaya Reviewed-by: Tejas Upadhyay Link: https://patchwork.freedesktop.org/patch/msgid/20250117193827.91779-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi --- include/uapi/drm/xe_drm.h | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index cac607a30f6d..e2160330ad01 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -814,29 +814,29 @@ struct drm_xe_gem_create { * * The @flags can be: * - %DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER - For user to query special offset - * for use in mmap ioctl. Writing to the returned mmap address will generate a - * PCI memory barrier with low overhead (avoiding IOCTL call as well as writing - * to VRAM which would also add overhead), acting like an MI_MEM_FENCE - * instruction. + * for use in mmap ioctl. Writing to the returned mmap address will generate a + * PCI memory barrier with low overhead (avoiding IOCTL call as well as writing + * to VRAM which would also add overhead), acting like an MI_MEM_FENCE + * instruction. * - * Note: The mmap size can be at most 4K, due to HW limitations. As a result - * this interface is only supported on CPU architectures that support 4K page - * size. The mmap_offset ioctl will detect this and gracefully return an - * error, where userspace is expected to have a different fallback method for - * triggering a barrier. + * Note: The mmap size can be at most 4K, due to HW limitations. As a result + * this interface is only supported on CPU architectures that support 4K page + * size. The mmap_offset ioctl will detect this and gracefully return an + * error, where userspace is expected to have a different fallback method for + * triggering a barrier. * - * Roughly the usage would be as follows: + * Roughly the usage would be as follows: * - * .. code-block:: C + * .. code-block:: C * - * struct drm_xe_gem_mmap_offset mmo = { - * .handle = 0, // must be set to 0 - * .flags = DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER, - * }; + * struct drm_xe_gem_mmap_offset mmo = { + * .handle = 0, // must be set to 0 + * .flags = DRM_XE_MMAP_OFFSET_FLAG_PCI_BARRIER, + * }; * - * err = ioctl(fd, DRM_IOCTL_XE_GEM_MMAP_OFFSET, &mmo); - * map = mmap(NULL, size, PROT_WRITE, MAP_SHARED, fd, mmo.offset); - * map[i] = 0xdeadbeaf; // issue barrier + * err = ioctl(fd, DRM_IOCTL_XE_GEM_MMAP_OFFSET, &mmo); + * map = mmap(NULL, size, PROT_WRITE, MAP_SHARED, fd, mmo.offset); + * map[i] = 0xdeadbeaf; // issue barrier */ struct drm_xe_gem_mmap_offset { /** @extensions: Pointer to the first extension struct, if any */ From 380b0cdaa76bc8f5c16db16eaf48751e792ff041 Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 10 Dec 2024 09:31:03 +0100 Subject: [PATCH 0130/1443] drm/xe: Move suballocator init to after display init No allocations should be done before we have had a chance to preserve the display fb. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20241210083111.230484-4-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/xe/xe_device.c | 6 ++++++ drivers/gpu/drm/xe/xe_tile.c | 12 ++++++++---- drivers/gpu/drm/xe/xe_tile.h | 1 + 3 files changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 6ecbf7dd396c..af9b2bb62dee 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -840,6 +840,12 @@ int xe_device_probe(struct xe_device *xe) if (err) goto err; + for_each_tile(tile, xe, id) { + err = xe_tile_init(tile); + if (err) + goto err; + } + for_each_gt(gt, xe, id) { last_gt = id; diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c index 07cf7cfe4abd..2825553b568f 100644 --- a/drivers/gpu/drm/xe/xe_tile.c +++ b/drivers/gpu/drm/xe/xe_tile.c @@ -170,10 +170,6 @@ int xe_tile_init_noalloc(struct xe_tile *tile) if (err) return err; - tile->mem.kernel_bb_pool = xe_sa_bo_manager_init(tile, SZ_1M, 16); - if (IS_ERR(tile->mem.kernel_bb_pool)) - return PTR_ERR(tile->mem.kernel_bb_pool); - xe_wa_apply_tile_workarounds(tile); err = xe_tile_sysfs_init(tile); @@ -181,6 +177,14 @@ int xe_tile_init_noalloc(struct xe_tile *tile) return 0; } +int xe_tile_init(struct xe_tile *tile) +{ + tile->mem.kernel_bb_pool = xe_sa_bo_manager_init(tile, SZ_1M, 16); + if (IS_ERR(tile->mem.kernel_bb_pool)) + return PTR_ERR(tile->mem.kernel_bb_pool); + + return 0; +} void xe_tile_migrate_wait(struct xe_tile *tile) { xe_migrate_wait(tile->migrate); diff --git a/drivers/gpu/drm/xe/xe_tile.h b/drivers/gpu/drm/xe/xe_tile.h index 1c9e42ade6b0..eb939316d55b 100644 --- a/drivers/gpu/drm/xe/xe_tile.h +++ b/drivers/gpu/drm/xe/xe_tile.h @@ -12,6 +12,7 @@ struct xe_tile; int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id); int xe_tile_init_noalloc(struct xe_tile *tile); +int xe_tile_init(struct xe_tile *tile); void xe_tile_migrate_wait(struct xe_tile *tile); From b893d7ff853e27aa6000fc4ca12e0ffda3318bfc Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 13 Jan 2025 12:07:57 -0600 Subject: [PATCH 0131/1443] scsi: core: Add passthrough tests for success and no failure definitions This patch adds scsi_check_passthrough() tests for the cases where a command completes successfully and when the command failed but the caller did not pass in a list of failures. Signed-off-by: Mike Christie Link: https://lore.kernel.org/r/20250113180757.16691-1-michael.christie@oracle.com Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_lib_test.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/scsi/scsi_lib_test.c b/drivers/scsi/scsi_lib_test.c index 99834426a100..ae8af0e0047a 100644 --- a/drivers/scsi/scsi_lib_test.c +++ b/drivers/scsi/scsi_lib_test.c @@ -67,6 +67,13 @@ static void scsi_lib_test_multiple_sense(struct kunit *test) }; int i; + /* Success */ + sc.result = 0; + KUNIT_EXPECT_EQ(test, 0, scsi_check_passthrough(&sc, &failures)); + KUNIT_EXPECT_EQ(test, 0, scsi_check_passthrough(&sc, NULL)); + /* Command failed but caller did not pass in a failures array */ + scsi_build_sense(&sc, 0, ILLEGAL_REQUEST, 0x91, 0x36); + KUNIT_EXPECT_EQ(test, 0, scsi_check_passthrough(&sc, NULL)); /* Match end of array */ scsi_build_sense(&sc, 0, ILLEGAL_REQUEST, 0x91, 0x36); KUNIT_EXPECT_EQ(test, -EAGAIN, scsi_check_passthrough(&sc, &failures)); From 1b3e2d4ec0c5848776cc56d2624998aa5b2f0d27 Mon Sep 17 00:00:00 2001 From: "Bao D. Nguyen" Date: Mon, 13 Jan 2025 10:32:07 -0800 Subject: [PATCH 0132/1443] scsi: ufs: core: Fix the HIGH/LOW_TEMP Bit Definitions According to the UFS Device Specification, the dExtendedUFSFeaturesSupport defines the support for TOO_HIGH_TEMPERATURE as bit[4] and the TOO_LOW_TEMPERATURE as bit[5]. Correct the code to match with the UFS device specification definition. Cc: stable@vger.kernel.org Fixes: e88e2d32200a ("scsi: ufs: core: Probe for temperature notification support") Signed-off-by: Bao D. Nguyen Link: https://lore.kernel.org/r/69992b3e3e3434a5c7643be5a64de48be892ca46.1736793068.git.quic_nguyenb@quicinc.com Reviewed-by: Avri Altman Reviewed-by: Peter Wang Signed-off-by: Martin K. Petersen --- include/ufs/ufs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/ufs/ufs.h b/include/ufs/ufs.h index 89672ad8c3bb..f151feb0ca8c 100644 --- a/include/ufs/ufs.h +++ b/include/ufs/ufs.h @@ -385,8 +385,8 @@ enum { /* Possible values for dExtendedUFSFeaturesSupport */ enum { - UFS_DEV_LOW_TEMP_NOTIF = BIT(4), - UFS_DEV_HIGH_TEMP_NOTIF = BIT(5), + UFS_DEV_HIGH_TEMP_NOTIF = BIT(4), + UFS_DEV_LOW_TEMP_NOTIF = BIT(5), UFS_DEV_EXT_TEMP_NOTIF = BIT(6), UFS_DEV_HPB_SUPPORT = BIT(7), UFS_DEV_WRITE_BOOSTER_SUP = BIT(8), From c9d2782988df354b5a2db00be93920b4ecdde7a2 Mon Sep 17 00:00:00 2001 From: Guixin Liu Date: Tue, 14 Jan 2025 10:50:41 +0800 Subject: [PATCH 0133/1443] scsi: target: core: Add line break to status show To ensure the output is not tangled with the shell prompt, add a line break to clearly display the status. Signed-off-by: Guixin Liu Link: https://lore.kernel.org/r/20250114025041.97301-1-kanie@linux.alibaba.com Reviewed-by: Mike Christie Signed-off-by: Martin K. Petersen --- drivers/target/target_core_stat.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_stat.c b/drivers/target/target_core_stat.c index c42cbde8a31b..210648a0092e 100644 --- a/drivers/target/target_core_stat.c +++ b/drivers/target/target_core_stat.c @@ -117,9 +117,9 @@ static ssize_t target_stat_tgt_status_show(struct config_item *item, char *page) { if (to_stat_tgt_dev(item)->export_count) - return snprintf(page, PAGE_SIZE, "activated"); + return snprintf(page, PAGE_SIZE, "activated\n"); else - return snprintf(page, PAGE_SIZE, "deactivated"); + return snprintf(page, PAGE_SIZE, "deactivated\n"); } static ssize_t target_stat_tgt_non_access_lus_show(struct config_item *item, From 8c09f612b2937da109ed0df583ace3a29fc95a93 Mon Sep 17 00:00:00 2001 From: Avri Altman Date: Tue, 14 Jan 2025 20:12:05 +0200 Subject: [PATCH 0134/1443] scsi: ufs: core: Simplify temperature exception event handling This commit simplifies the temperature exception event handling by removing the ufshcd_temp_exception_event_handler() function and directly calling ufs_hwmon_notify_event() in ufshcd_exception_event_handler(). The ufshcd_temp_exception_event_handler() function contained a placeholder comment for platform vendors to add additional steps if required. However, since its introduction a few years ago, no vendor has added any additional steps. Therefore, the placeholder function is removed to streamline the code. Signed-off-by: Avri Altman Link: https://lore.kernel.org/r/20250114181205.153760-1-avri.altman@wdc.com Reviewed-by: Bean Huo Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 20 +------------------- 1 file changed, 1 insertion(+), 19 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 0920a443588c..f6c38cf10382 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -5976,24 +5976,6 @@ static void ufshcd_bkops_exception_event_handler(struct ufs_hba *hba) __func__, err); } -static void ufshcd_temp_exception_event_handler(struct ufs_hba *hba, u16 status) -{ - u32 value; - - if (ufshcd_query_attr_retry(hba, UPIU_QUERY_OPCODE_READ_ATTR, - QUERY_ATTR_IDN_CASE_ROUGH_TEMP, 0, 0, &value)) - return; - - dev_info(hba->dev, "exception Tcase %d\n", value - 80); - - ufs_hwmon_notify_event(hba, status & MASK_EE_URGENT_TEMP); - - /* - * A placeholder for the platform vendors to add whatever additional - * steps required - */ -} - static int __ufshcd_wb_toggle(struct ufs_hba *hba, bool set, enum flag_idn idn) { u8 index; @@ -6214,7 +6196,7 @@ static void ufshcd_exception_event_handler(struct work_struct *work) ufshcd_bkops_exception_event_handler(hba); if (status & hba->ee_drv_mask & MASK_EE_URGENT_TEMP) - ufshcd_temp_exception_event_handler(hba, status); + ufs_hwmon_notify_event(hba, status & MASK_EE_URGENT_TEMP); ufs_debugfs_exception_event(hba, status); } From cfa9d40db8c30d894171010fe765d96e9bc6a47e Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Thu, 16 Jan 2025 19:21:55 -0800 Subject: [PATCH 0135/1443] drm/xe/oa: Preserve oa_ctrl unused bits UMD's have interest in setting unused bits of the oa_ctrl register "out of band" for certain experiments. To facilitate this, don't clobber previous oa_ctrl unused bits, i.e. rmw the values rather than simply write them. Fixes: e936f885f1e9 ("drm/xe/oa/uapi: Expose OA stream fd") Signed-off-by: Ashutosh Dixit Reviewed-by: Umesh Nerlige Ramappa Link: https://patchwork.freedesktop.org/patch/msgid/20250117032155.3048063-1-ashutosh.dixit@intel.com --- drivers/gpu/drm/xe/regs/xe_oa_regs.h | 6 ++++++ drivers/gpu/drm/xe/xe_oa.c | 12 ++++++++++-- 2 files changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_oa_regs.h b/drivers/gpu/drm/xe/regs/xe_oa_regs.h index a49561e9f3c3..a79ad2da070c 100644 --- a/drivers/gpu/drm/xe/regs/xe_oa_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_oa_regs.h @@ -51,6 +51,10 @@ /* Common to all OA units */ #define OA_OACONTROL_REPORT_BC_MASK REG_GENMASK(9, 9) #define OA_OACONTROL_COUNTER_SIZE_MASK REG_GENMASK(8, 8) +#define OAG_OACONTROL_USED_BITS \ + (OAG_OACONTROL_OA_PES_DISAG_EN | OAG_OACONTROL_OA_CCS_SELECT_MASK | \ + OAG_OACONTROL_OA_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE | \ + OA_OACONTROL_REPORT_BC_MASK | OA_OACONTROL_COUNTER_SIZE_MASK) #define OAG_OA_DEBUG XE_REG(0xdaf8, XE_REG_OPTION_MASKED) #define OAG_OA_DEBUG_DISABLE_MMIO_TRG REG_BIT(14) @@ -78,6 +82,8 @@ #define OAM_CONTEXT_CONTROL_OFFSET (0x1bc) #define OAM_CONTROL_OFFSET (0x194) #define OAM_CONTROL_COUNTER_SEL_MASK REG_GENMASK(3, 1) +#define OAM_OACONTROL_USED_BITS \ + (OAM_CONTROL_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE) #define OAM_DEBUG_OFFSET (0x198) #define OAM_STATUS_OFFSET (0x19c) #define OAM_MMIO_TRG_OFFSET (0x1d0) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index eeb96b5f49e2..6a08e6c92835 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -452,6 +452,12 @@ static u32 __oa_ccs_select(struct xe_oa_stream *stream) return val; } +static u32 __oactrl_used_bits(struct xe_oa_stream *stream) +{ + return stream->hwe->oa_unit->type == DRM_XE_OA_UNIT_TYPE_OAG ? + OAG_OACONTROL_USED_BITS : OAM_OACONTROL_USED_BITS; +} + static void xe_oa_enable(struct xe_oa_stream *stream) { const struct xe_oa_format *format = stream->oa_buffer.format; @@ -472,14 +478,14 @@ static void xe_oa_enable(struct xe_oa_stream *stream) stream->hwe->oa_unit->type == DRM_XE_OA_UNIT_TYPE_OAG) val |= OAG_OACONTROL_OA_PES_DISAG_EN; - xe_mmio_write32(&stream->gt->mmio, regs->oa_ctrl, val); + xe_mmio_rmw32(&stream->gt->mmio, regs->oa_ctrl, __oactrl_used_bits(stream), val); } static void xe_oa_disable(struct xe_oa_stream *stream) { struct xe_mmio *mmio = &stream->gt->mmio; - xe_mmio_write32(mmio, __oa_regs(stream)->oa_ctrl, 0); + xe_mmio_rmw32(mmio, __oa_regs(stream)->oa_ctrl, __oactrl_used_bits(stream), 0); if (xe_mmio_wait32(mmio, __oa_regs(stream)->oa_ctrl, OAG_OACONTROL_OA_COUNTER_ENABLE, 0, 50000, NULL, false)) drm_err(&stream->oa->xe->drm, @@ -2534,6 +2540,8 @@ static void __xe_oa_init_oa_units(struct xe_gt *gt) u->type = DRM_XE_OA_UNIT_TYPE_OAM; } + xe_mmio_write32(>->mmio, u->regs.oa_ctrl, 0); + /* Ensure MMIO trigger remains disabled till there is a stream */ xe_mmio_write32(>->mmio, u->regs.oa_debug, oag_configure_mmio_trigger(NULL, false)); From 9ebb5846e1a3b1705f8a7cbc528888a1aa0b163e Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Tue, 21 Jan 2025 00:24:43 +0100 Subject: [PATCH 0136/1443] drm/xe/pf: Fix migration initialization MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The migration support only needs to be initialized once, but it was incorrectly called from the xe_gt_sriov_pf_init_hw(), which is part of the reset flow and may be called multiple times. Fixes: d86e3737c7ab ("drm/xe/pf: Add functions to save and restore VF GuC state") Signed-off-by: Michal Wajdeczko Cc: Michał Winiarski Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250120232443.544-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt.c | 4 +++- drivers/gpu/drm/xe/xe_gt_sriov_pf.c | 14 +++++++++++++- drivers/gpu/drm/xe/xe_gt_sriov_pf.h | 6 ++++++ 3 files changed, 22 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index b5c313a3e946..01a4a852b8f4 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -532,8 +532,10 @@ static int all_fw_domain_init(struct xe_gt *gt) if (IS_SRIOV_PF(gt_to_xe(gt)) && !xe_gt_is_media_type(gt)) xe_lmtt_init_hw(>_to_tile(gt)->sriov.pf.lmtt); - if (IS_SRIOV_PF(gt_to_xe(gt))) + if (IS_SRIOV_PF(gt_to_xe(gt))) { + xe_gt_sriov_pf_init(gt); xe_gt_sriov_pf_init_hw(gt); + } xe_force_wake_put(gt_to_fw(gt), fw_ref); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c index e71fc3d2bda2..6f906c8e8108 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c @@ -68,6 +68,19 @@ int xe_gt_sriov_pf_init_early(struct xe_gt *gt) return 0; } +/** + * xe_gt_sriov_pf_init - Prepare SR-IOV PF data structures on PF. + * @gt: the &xe_gt to initialize + * + * Late one-time initialization of the PF data. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_gt_sriov_pf_init(struct xe_gt *gt) +{ + return xe_gt_sriov_pf_migration_init(gt); +} + static bool pf_needs_enable_ggtt_guest_update(struct xe_device *xe) { return GRAPHICS_VERx100(xe) == 1200; @@ -90,7 +103,6 @@ void xe_gt_sriov_pf_init_hw(struct xe_gt *gt) pf_enable_ggtt_guest_update(gt); xe_gt_sriov_pf_service_update(gt); - xe_gt_sriov_pf_migration_init(gt); } static u32 pf_get_vf_regs_stride(struct xe_device *xe) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h index 96fab779a906..f474509411c0 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h @@ -10,6 +10,7 @@ struct xe_gt; #ifdef CONFIG_PCI_IOV int xe_gt_sriov_pf_init_early(struct xe_gt *gt); +int xe_gt_sriov_pf_init(struct xe_gt *gt); void xe_gt_sriov_pf_init_hw(struct xe_gt *gt); void xe_gt_sriov_pf_sanitize_hw(struct xe_gt *gt, unsigned int vfid); void xe_gt_sriov_pf_restart(struct xe_gt *gt); @@ -19,6 +20,11 @@ static inline int xe_gt_sriov_pf_init_early(struct xe_gt *gt) return 0; } +static inline int xe_gt_sriov_pf_init(struct xe_gt *gt) +{ + return 0; +} + static inline void xe_gt_sriov_pf_init_hw(struct xe_gt *gt) { } From 5994018ecffc9e70e192bede2270819d8c93b2d4 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Tue, 21 Jan 2025 10:48:32 +0100 Subject: [PATCH 0137/1443] drm/xe/guc: Fix sizeof(32) typo A small typo leads to the following static code checker warning: drivers/gpu/drm/xe/xe_guc_buf.c:81 xe_guc_buf_reserve() warn: sizeof(NUMBER)? Reported-by: Dan Carpenter Closes: https://lore.kernel.org/intel-xe/0d5bcbf1-79f9-4a10-a221-ddbaec9f6122@stanley.mountain/ Fixes: 696bfdf273ea ("drm/xe/guc: Introduce the GuC Buffer Cache") Signed-off-by: Michal Wajdeczko Cc: Dan Carpenter Cc: Matthew Brost Reviewed-by: Stuart Summers Link: https://patchwork.freedesktop.org/patch/msgid/20250121094832.588-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_guc_buf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_guc_buf.c b/drivers/gpu/drm/xe/xe_guc_buf.c index ce6d9830e13b..0193c94dd6a0 100644 --- a/drivers/gpu/drm/xe/xe_guc_buf.c +++ b/drivers/gpu/drm/xe/xe_guc_buf.c @@ -78,7 +78,7 @@ struct xe_guc_buf xe_guc_buf_reserve(struct xe_guc_buf_cache *cache, u32 dwords) struct drm_suballoc *sa; if (cache->sam) - sa = __xe_sa_bo_new(cache->sam, dwords * sizeof(32), GFP_ATOMIC); + sa = __xe_sa_bo_new(cache->sam, dwords * sizeof(u32), GFP_ATOMIC); else sa = ERR_PTR(-EOPNOTSUPP); From e0a4cd6aceca0e9164d7304ac1a9b061e1d96529 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Fri, 17 Jan 2025 08:45:29 -0800 Subject: [PATCH 0138/1443] MAINTAINERS: Also exclude xe for drm-misc When the xe driver was added, it didn't extend the exclude entries for drm-misc, as done in commit 5a44d50f0072 ("MAINTAINERS: Update drm-misc entry to match all drivers"). Exclude it like is done for i915 and other drivers with dedicated maintainers. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250117164529.393503-1-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 17daa9ee9384..6664d39ab4c6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7598,6 +7598,7 @@ X: drivers/gpu/drm/nouveau/ X: drivers/gpu/drm/radeon/ X: drivers/gpu/drm/renesas/rcar-du/ X: drivers/gpu/drm/tegra/ +X: drivers/gpu/drm/xe/ DRM DRIVERS FOR ALLWINNER A10 M: Maxime Ripard From 3d09b2718969f6db5b9e50daa9a033a78f065522 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:06 +0100 Subject: [PATCH 0139/1443] drm/ast: Detect wide-screen support before creating modeset pipeline Wide-screen support is relevant for mode validation. Do not detect it before setting up the mode-setting pipeline. Gets the function call out of the way of other initialization code. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-2-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index bc37c65305d4..037d389ab630 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -290,7 +290,6 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, ast->regs = regs; ast->ioregs = ioregs; - ast_detect_widescreen(ast); ast_detect_tx_chip(ast, need_post); ret = ast_get_dram_info(ast); @@ -315,6 +314,8 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, drm_info(dev, "failed to map reserved buffer!\n"); } + ast_detect_widescreen(ast); + ret = ast_mode_config_init(ast); if (ret) return ERR_PTR(ret); From be1c00b180f1c580c93e585058e64df51fcfd4c2 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:07 +0100 Subject: [PATCH 0140/1443] drm/ast: Detect DRAM before TX-chip Move DRAM detection before TX-chip detection. Both steps are independent from each other. Detection of the TX-chip is now next to posting those chips, which can be done in a single step. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-3-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 037d389ab630..456230bef273 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -290,15 +290,13 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, ast->regs = regs; ast->ioregs = ioregs; - ast_detect_tx_chip(ast, need_post); - ret = ast_get_dram_info(ast); if (ret) return ERR_PTR(ret); - drm_info(dev, "dram MCLK=%u Mhz type=%d bus_width=%d\n", ast->mclk, ast->dram_type, ast->dram_bus_width); + ast_detect_tx_chip(ast, need_post); if (need_post) ast_post_gpu(ast); From b40e209130bff435c8dcd17660cb4614ae62a3fb Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:08 +0100 Subject: [PATCH 0141/1443] drm/ast: Refactor ast_post_gpu() by Gen Reorganize ast_post_gpu() so that it first branches by Gen and then by config mode and TX chip. This will later make it possible to split up the function by Gen. The helper ast_init_3rdtx() only handles Gen4 and Gen5, so leave it out from the other Gens. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-4-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_post.c | 36 ++++++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 364030f97571..49f661760f9e 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -344,21 +344,37 @@ void ast_post_gpu(struct ast_device *ast) { ast_set_def_ext_reg(ast); - if (IS_AST_GEN7(ast)) { + if (AST_GEN(ast) >= 7) { if (ast->tx_chip == AST_TX_ASTDP) ast_dp_launch(ast); - } else if (ast->config_mode == ast_use_p2a) { - if (IS_AST_GEN6(ast)) + } else if (AST_GEN(ast) >= 6) { + if (ast->config_mode == ast_use_p2a) { ast_post_chip_2500(ast); - else if (IS_AST_GEN5(ast) || IS_AST_GEN4(ast)) + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + } else if (AST_GEN(ast) >= 4) { + if (ast->config_mode == ast_use_p2a) { ast_post_chip_2300(ast); - else + ast_init_3rdtx(ast); + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + } else { + if (ast->config_mode == ast_use_p2a) { ast_init_dram_reg(ast); - - ast_init_3rdtx(ast); - } else { - if (ast->tx_chip == AST_TX_SIL164) - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); /* Enable DVO */ + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } } } From 1c6220a3950bd8e60126f5ea234f3f7bd86321b1 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:09 +0100 Subject: [PATCH 0142/1443] drm/ast: Initialize ASTDP in ast_post_gpu() Remove the call to ast_dp_launch() from ast_detect_tx_chip() and perform it unconditionally in ast_post_gpu(). Also add error handling: the detection code apparently used ast_dp_launch() to test for a working ASTDP, falling back to VGA on errors. As the VBIOS reports ASTDP, silently ignoring errors is questionable behavior. With the refactoring, failing to initialize the ASTDP will also fail probing the driver. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-5-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_drv.c | 6 +++++- drivers/gpu/drm/ast/ast_drv.h | 2 +- drivers/gpu/drm/ast/ast_main.c | 19 +++++++++++++------ drivers/gpu/drm/ast/ast_post.c | 13 ++++++++++--- 4 files changed, 29 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index ff3bcdd1cff2..cddd69972e89 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -393,11 +393,15 @@ static int ast_drm_freeze(struct drm_device *dev) static int ast_drm_thaw(struct drm_device *dev) { struct ast_device *ast = to_ast_device(dev); + int ret; ast_enable_vga(ast->ioregs); ast_open_key(ast->ioregs); ast_enable_mmio(dev->dev, ast->ioregs); - ast_post_gpu(ast); + + ret = ast_post_gpu(ast); + if (ret) + return ret; return drm_mode_config_helper_resume(dev); } diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 6b4305ac07d4..cf9edef8fca6 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -445,7 +445,7 @@ int ast_mode_config_init(struct ast_device *ast); int ast_mm_init(struct ast_device *ast); /* ast post */ -void ast_post_gpu(struct ast_device *ast); +int ast_post_gpu(struct ast_device *ast); u32 ast_mindwm(struct ast_device *ast, u32 r); void ast_moutdwm(struct ast_device *ast, u32 r, u32 v); void ast_patch_ahb_2500(void __iomem *regs); diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 456230bef273..474eb255b325 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -138,10 +138,7 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) } else if (IS_AST_GEN7(ast)) { if (ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK) == AST_IO_VGACRD1_TX_ASTDP) { - int ret = ast_dp_launch(ast); - - if (!ret) - ast->tx_chip = AST_TX_ASTDP; + ast->tx_chip = AST_TX_ASTDP; } } @@ -297,8 +294,18 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, ast->mclk, ast->dram_type, ast->dram_bus_width); ast_detect_tx_chip(ast, need_post); - if (need_post) - ast_post_gpu(ast); + switch (ast->tx_chip) { + case AST_TX_ASTDP: + ret = ast_post_gpu(ast); + break; + default: + ret = 0; + if (need_post) + ret = ast_post_gpu(ast); + break; + } + if (ret) + return ERR_PTR(ret); ret = ast_mm_init(ast); if (ret) diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 49f661760f9e..0daa8e52a092 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -340,13 +340,18 @@ static void ast_init_dram_reg(struct ast_device *ast) } while ((j & 0x40) == 0); } -void ast_post_gpu(struct ast_device *ast) +int ast_post_gpu(struct ast_device *ast) { + int ret; + ast_set_def_ext_reg(ast); if (AST_GEN(ast) >= 7) { - if (ast->tx_chip == AST_TX_ASTDP) - ast_dp_launch(ast); + if (ast->tx_chip == AST_TX_ASTDP) { + ret = ast_dp_launch(ast); + if (ret) + return ret; + } } else if (AST_GEN(ast) >= 6) { if (ast->config_mode == ast_use_p2a) { ast_post_chip_2500(ast); @@ -376,6 +381,8 @@ void ast_post_gpu(struct ast_device *ast) } } } + + return 0; } /* AST 2300 DRAM settings */ From 87478ba50a05a1f44508316ae109622e8a85adc9 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:10 +0100 Subject: [PATCH 0143/1443] drm/ast: Hide Gens 1 to 3 TX detection in branch Gen7 only supports ASTDP. Gens 4 to 6 support various TX chips, except ASTDP. These boards detect the TX chips by reading the SoC scratch register as VGACRD1. Gens 1 to 3 only support SIL164. These boards read the DVO bit from VGACRA3. Hence move this test behind a branch, so that it does not run on later generations. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-6-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 474eb255b325..50b57bc15d53 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -96,21 +96,21 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) /* Check 3rd Tx option (digital output afaik) */ ast->tx_chip = AST_TX_NONE; - /* - * VGACRA3 Enhanced Color Mode Register, check if DVO is already - * enabled, in that case, assume we have a SIL164 TMDS transmitter - * - * Don't make that assumption if we the chip wasn't enabled and - * is at power-on reset, otherwise we'll incorrectly "detect" a - * SIL164 when there is none. - */ - if (!need_post) { - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xff); - if (jreg & 0x80) - ast->tx_chip = AST_TX_SIL164; - } - - if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) { + if (AST_GEN(ast) <= 3) { + /* + * VGACRA3 Enhanced Color Mode Register, check if DVO is already + * enabled, in that case, assume we have a SIL164 TMDS transmitter + * + * Don't make that assumption if we the chip wasn't enabled and + * is at power-on reset, otherwise we'll incorrectly "detect" a + * SIL164 when there is none. + */ + if (!need_post) { + jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xff); + if (jreg & 0x80) + ast->tx_chip = AST_TX_SIL164; + } + } else if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) { /* * On AST GEN4+, look the configuration set by the SoC in * the SOC scratch register #1 bits 11:8 (interestingly marked From 2eede6f1d2b1d3c36e77555f7b98ab8a45f0527c Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:11 +0100 Subject: [PATCH 0144/1443] drm/ast: Align Gen1 DVO detection to register manual Align variable names and register constants for TX-chip detection to the names in the register manual. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-7-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 6 +++--- drivers/gpu/drm/ast/ast_reg.h | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 50b57bc15d53..40d3b7770cf1 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -76,7 +76,7 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) }; struct drm_device *dev = &ast->base; - u8 jreg, vgacrd1; + u8 vgacra3, vgacrd1; /* * Several of the listed TX chips are not explicitly supported @@ -106,8 +106,8 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) * SIL164 when there is none. */ if (!need_post) { - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xff); - if (jreg & 0x80) + vgacra3 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xff); + if (vgacra3 & AST_IO_VGACRA3_DVO_ENABLED) ast->tx_chip = AST_TX_SIL164; } } else if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) { diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h index 2aadf07d135a..0745d58e5b45 100644 --- a/drivers/gpu/drm/ast/ast_reg.h +++ b/drivers/gpu/drm/ast/ast_reg.h @@ -32,6 +32,7 @@ #define AST_IO_VGACR80_PASSWORD (0xa8) #define AST_IO_VGACRA1_VGAIO_DISABLED BIT(1) #define AST_IO_VGACRA1_MMIO_ENABLED BIT(2) +#define AST_IO_VGACRA3_DVO_ENABLED BIT(7) #define AST_IO_VGACRB6_HSYNC_OFF BIT(0) #define AST_IO_VGACRB6_VSYNC_OFF BIT(1) #define AST_IO_VGACRCB_HWC_16BPP BIT(0) /* set: ARGB4444, cleared: 2bpp palette */ From dc80fde7947fd9d90d229e2b007cda2066943fb7 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:12 +0100 Subject: [PATCH 0145/1443] drm/ast: Merge TX-chip detection code for Gen4 and later Gens 4 to 6 and Gen7 use the same pattern for detecting the installed TX chips. Merge the code into a single branch. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-8-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 40d3b7770cf1..b0d1b99ed532 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -110,15 +110,18 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) if (vgacra3 & AST_IO_VGACRA3_DVO_ENABLED) ast->tx_chip = AST_TX_SIL164; } - } else if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) { + } else { /* - * On AST GEN4+, look the configuration set by the SoC in + * On AST GEN4+, look at the configuration set by the SoC in * the SOC scratch register #1 bits 11:8 (interestingly marked * as "reserved" in the spec) */ jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK); switch (jreg) { + /* + * GEN4 to GEN6 + */ case AST_IO_VGACRD1_TX_SIL164_VBIOS: ast->tx_chip = AST_TX_SIL164; break; @@ -134,11 +137,13 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) fallthrough; case AST_IO_VGACRD1_TX_FW_EMBEDDED_FW: ast->tx_chip = AST_TX_DP501; - } - } else if (IS_AST_GEN7(ast)) { - if (ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK) == - AST_IO_VGACRD1_TX_ASTDP) { + break; + /* + * GEN7+ + */ + case AST_IO_VGACRD1_TX_ASTDP: ast->tx_chip = AST_TX_ASTDP; + break; } } From a9301e5bef12f8989a02d886109f13e89e1e51b0 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 17 Jan 2025 11:29:13 +0100 Subject: [PATCH 0146/1443] drm/ast: Only warn about unsupported TX chips on Gen4 and later Only Gen4 and later read the installed TX chip from the SoC. So only warn on those generations about unsupported chips. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250117103450.28692-9-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 40 +++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index b0d1b99ed532..ba69280b33e7 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -78,21 +78,6 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) struct drm_device *dev = &ast->base; u8 vgacra3, vgacrd1; - /* - * Several of the listed TX chips are not explicitly supported - * by the ast driver. If these exist in real-world devices, they - * are most likely reported as VGA or SIL164 outputs. We warn here - * to get bug reports for these devices. If none come in for some - * time, we can begin to fail device probing on these values. - */ - vgacrd1 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK); - drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ITE66121_VBIOS, - "ITE IT66121 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); - drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_CH7003_VBIOS, - "Chrontel CH7003 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); - drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ANX9807_VBIOS, - "Analogix ANX9807 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); - /* Check 3rd Tx option (digital output afaik) */ ast->tx_chip = AST_TX_NONE; @@ -116,9 +101,9 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) * the SOC scratch register #1 bits 11:8 (interestingly marked * as "reserved" in the spec) */ - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, - AST_IO_VGACRD1_TX_TYPE_MASK); - switch (jreg) { + vgacrd1 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, + AST_IO_VGACRD1_TX_TYPE_MASK); + switch (vgacrd1) { /* * GEN4 to GEN6 */ @@ -144,6 +129,25 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) case AST_IO_VGACRD1_TX_ASTDP: ast->tx_chip = AST_TX_ASTDP; break; + /* + * Several of the listed TX chips are not explicitly supported + * by the ast driver. If these exist in real-world devices, they + * are most likely reported as VGA or SIL164 outputs. We warn here + * to get bug reports for these devices. If none come in for some + * time, we can begin to fail device probing on these values. + */ + case AST_IO_VGACRD1_TX_ITE66121_VBIOS: + drm_warn(dev, "ITE IT66121 detected, 0x%x, Gen%lu\n", + vgacrd1, AST_GEN(ast)); + break; + case AST_IO_VGACRD1_TX_CH7003_VBIOS: + drm_warn(dev, "Chrontel CH7003 detected, 0x%x, Gen%lu\n", + vgacrd1, AST_GEN(ast)); + break; + case AST_IO_VGACRD1_TX_ANX9807_VBIOS: + drm_warn(dev, "Analogix ANX9807 detected, 0x%x, Gen%lu\n", + vgacrd1, AST_GEN(ast)); + break; } } From e7e34ffc976aaae4f465b7898303241b81ceefc3 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Mon, 20 Jan 2025 20:35:28 +0100 Subject: [PATCH 0147/1443] batman-adv: Ignore neighbor throughput metrics in error case If a temporary error happened in the evaluation of the neighbor throughput information, then the invalid throughput result should not be stored in the throughtput EWMA. Cc: stable@vger.kernel.org Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/bat_v_elp.c | 50 ++++++++++++++++++++++++++------------ 1 file changed, 34 insertions(+), 16 deletions(-) diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index fbf499bcc671..65e52de52bcd 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -59,11 +59,13 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface) /** * batadv_v_elp_get_throughput() - get the throughput towards a neighbour * @neigh: the neighbour for which the throughput has to be obtained + * @pthroughput: calculated throughput towards the given neighbour in multiples + * of 100kpbs (a value of '1' equals 0.1Mbps, '10' equals 1Mbps, etc). * - * Return: The throughput towards the given neighbour in multiples of 100kpbs - * (a value of '1' equals 0.1Mbps, '10' equals 1Mbps, etc). + * Return: true when value behind @pthroughput was set */ -static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) +static bool batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh, + u32 *pthroughput) { struct batadv_hard_iface *hard_iface = neigh->if_incoming; struct net_device *soft_iface = hard_iface->soft_iface; @@ -77,14 +79,16 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) * batman-adv interface */ if (!soft_iface) - return BATADV_THROUGHPUT_DEFAULT_VALUE; + return false; /* if the user specified a customised value for this interface, then * return it directly */ throughput = atomic_read(&hard_iface->bat_v.throughput_override); - if (throughput != 0) - return throughput; + if (throughput != 0) { + *pthroughput = throughput; + return true; + } /* if this is a wireless device, then ask its throughput through * cfg80211 API @@ -111,19 +115,24 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) * possible to delete this neighbor. For now set * the throughput metric to 0. */ - return 0; + *pthroughput = 0; + return true; } if (ret) goto default_throughput; - if (sinfo.filled & BIT(NL80211_STA_INFO_EXPECTED_THROUGHPUT)) - return sinfo.expected_throughput / 100; + if (sinfo.filled & BIT(NL80211_STA_INFO_EXPECTED_THROUGHPUT)) { + *pthroughput = sinfo.expected_throughput / 100; + return true; + } /* try to estimate the expected throughput based on reported tx * rates */ - if (sinfo.filled & BIT(NL80211_STA_INFO_TX_BITRATE)) - return cfg80211_calculate_bitrate(&sinfo.txrate) / 3; + if (sinfo.filled & BIT(NL80211_STA_INFO_TX_BITRATE)) { + *pthroughput = cfg80211_calculate_bitrate(&sinfo.txrate) / 3; + return true; + } goto default_throughput; } @@ -142,8 +151,10 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) hard_iface->bat_v.flags &= ~BATADV_FULL_DUPLEX; throughput = link_settings.base.speed; - if (throughput && throughput != SPEED_UNKNOWN) - return throughput * 10; + if (throughput && throughput != SPEED_UNKNOWN) { + *pthroughput = throughput * 10; + return true; + } } default_throughput: @@ -157,7 +168,8 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh) } /* if none of the above cases apply, return the base_throughput */ - return BATADV_THROUGHPUT_DEFAULT_VALUE; + *pthroughput = BATADV_THROUGHPUT_DEFAULT_VALUE; + return true; } /** @@ -169,15 +181,21 @@ void batadv_v_elp_throughput_metric_update(struct work_struct *work) { struct batadv_hardif_neigh_node_bat_v *neigh_bat_v; struct batadv_hardif_neigh_node *neigh; + u32 throughput; + bool valid; neigh_bat_v = container_of(work, struct batadv_hardif_neigh_node_bat_v, metric_work); neigh = container_of(neigh_bat_v, struct batadv_hardif_neigh_node, bat_v); - ewma_throughput_add(&neigh->bat_v.throughput, - batadv_v_elp_get_throughput(neigh)); + valid = batadv_v_elp_get_throughput(neigh, &throughput); + if (!valid) + goto put_neigh; + ewma_throughput_add(&neigh->bat_v.throughput, throughput); + +put_neigh: /* decrement refcounter to balance increment performed before scheduling * this task */ From 8c8ecc98f5c65947b0070a24bac11e12e47cc65d Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Mon, 20 Jan 2025 00:06:11 +0100 Subject: [PATCH 0148/1443] batman-adv: Drop unmanaged ELP metric worker The ELP worker needs to calculate new metric values for all neighbors "reachable" over an interface. Some of the used metric sources require locks which might need to sleep. This sleep is incompatible with the RCU list iterator used for the recorded neighbors. The initial approach to work around of this problem was to queue another work item per neighbor and then run this in a new context. Even when this solved the RCU vs might_sleep() conflict, it has a major problems: Nothing was stopping the work item in case it is not needed anymore - for example because one of the related interfaces was removed or the batman-adv module was unloaded - resulting in potential invalid memory accesses. Directly canceling the metric worker also has various problems: * cancel_work_sync for a to-be-deactivated interface is called with rtnl_lock held. But the code in the ELP metric worker also tries to use rtnl_lock() - which will never return in this case. This also means that cancel_work_sync would never return because it is waiting for the worker to finish. * iterating over the neighbor list for the to-be-deactivated interface is currently done using the RCU specific methods. Which means that it is possible to miss items when iterating over it without the associated spinlock - a behaviour which is acceptable for a periodic metric check but not for a cleanup routine (which must "stop" all still running workers) The better approch is to get rid of the per interface neighbor metric worker and handle everything in the interface worker. The original problems are solved by: * creating a list of neighbors which require new metric information inside the RCU protected context, gathering the metric according to the new list outside the RCU protected context * only use rcu_trylock inside metric gathering code to avoid a deadlock when the cancel_delayed_work_sync is called in the interface removal code (which is called with the rtnl_lock held) Cc: stable@vger.kernel.org Fixes: c833484e5f38 ("batman-adv: ELP - compute the metric based on the estimated throughput") Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/bat_v.c | 2 -- net/batman-adv/bat_v_elp.c | 71 ++++++++++++++++++++++++++------------ net/batman-adv/bat_v_elp.h | 2 -- net/batman-adv/types.h | 3 -- 4 files changed, 48 insertions(+), 30 deletions(-) diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c index ac11f1f08db0..d35479c465e2 100644 --- a/net/batman-adv/bat_v.c +++ b/net/batman-adv/bat_v.c @@ -113,8 +113,6 @@ static void batadv_v_hardif_neigh_init(struct batadv_hardif_neigh_node *hardif_neigh) { ewma_throughput_init(&hardif_neigh->bat_v.throughput); - INIT_WORK(&hardif_neigh->bat_v.metric_work, - batadv_v_elp_throughput_metric_update); } /** diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index 65e52de52bcd..b065578b4436 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -26,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -41,6 +43,18 @@ #include "routing.h" #include "send.h" +/** + * struct batadv_v_metric_queue_entry - list of hardif neighbors which require + * and metric update + */ +struct batadv_v_metric_queue_entry { + /** @hardif_neigh: hardif neighbor scheduled for metric update */ + struct batadv_hardif_neigh_node *hardif_neigh; + + /** @list: list node for metric_queue */ + struct list_head list; +}; + /** * batadv_v_elp_start_timer() - restart timer for ELP periodic work * @hard_iface: the interface for which the timer has to be reset @@ -137,10 +151,17 @@ static bool batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh, goto default_throughput; } + /* only use rtnl_trylock because the elp worker will be cancelled while + * the rntl_lock is held. the cancel_delayed_work_sync() would otherwise + * wait forever when the elp work_item was started and it is then also + * trying to rtnl_lock + */ + if (!rtnl_trylock()) + return false; + /* if not a wifi interface, check if this device provides data via * ethtool (e.g. an Ethernet adapter) */ - rtnl_lock(); ret = __ethtool_get_link_ksettings(hard_iface->net_dev, &link_settings); rtnl_unlock(); if (ret == 0) { @@ -175,31 +196,19 @@ static bool batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh, /** * batadv_v_elp_throughput_metric_update() - worker updating the throughput * metric of a single hop neighbour - * @work: the work queue item + * @neigh: the neighbour to probe */ -void batadv_v_elp_throughput_metric_update(struct work_struct *work) +static void +batadv_v_elp_throughput_metric_update(struct batadv_hardif_neigh_node *neigh) { - struct batadv_hardif_neigh_node_bat_v *neigh_bat_v; - struct batadv_hardif_neigh_node *neigh; u32 throughput; bool valid; - neigh_bat_v = container_of(work, struct batadv_hardif_neigh_node_bat_v, - metric_work); - neigh = container_of(neigh_bat_v, struct batadv_hardif_neigh_node, - bat_v); - valid = batadv_v_elp_get_throughput(neigh, &throughput); if (!valid) - goto put_neigh; + return; ewma_throughput_add(&neigh->bat_v.throughput, throughput); - -put_neigh: - /* decrement refcounter to balance increment performed before scheduling - * this task - */ - batadv_hardif_neigh_put(neigh); } /** @@ -273,14 +282,16 @@ batadv_v_elp_wifi_neigh_probe(struct batadv_hardif_neigh_node *neigh) */ static void batadv_v_elp_periodic_work(struct work_struct *work) { + struct batadv_v_metric_queue_entry *metric_entry; + struct batadv_v_metric_queue_entry *metric_safe; struct batadv_hardif_neigh_node *hardif_neigh; struct batadv_hard_iface *hard_iface; struct batadv_hard_iface_bat_v *bat_v; struct batadv_elp_packet *elp_packet; + struct list_head metric_queue; struct batadv_priv *bat_priv; struct sk_buff *skb; u32 elp_interval; - bool ret; bat_v = container_of(work, struct batadv_hard_iface_bat_v, elp_wq.work); hard_iface = container_of(bat_v, struct batadv_hard_iface, bat_v); @@ -316,6 +327,8 @@ static void batadv_v_elp_periodic_work(struct work_struct *work) atomic_inc(&hard_iface->bat_v.elp_seqno); + INIT_LIST_HEAD(&metric_queue); + /* The throughput metric is updated on each sent packet. This way, if a * node is dead and no longer sends packets, batman-adv is still able to * react timely to its death. @@ -340,16 +353,28 @@ static void batadv_v_elp_periodic_work(struct work_struct *work) /* Reading the estimated throughput from cfg80211 is a task that * may sleep and that is not allowed in an rcu protected - * context. Therefore schedule a task for that. + * context. Therefore add it to metric_queue and process it + * outside rcu protected context. */ - ret = queue_work(batadv_event_workqueue, - &hardif_neigh->bat_v.metric_work); - - if (!ret) + metric_entry = kzalloc(sizeof(*metric_entry), GFP_ATOMIC); + if (!metric_entry) { batadv_hardif_neigh_put(hardif_neigh); + continue; + } + + metric_entry->hardif_neigh = hardif_neigh; + list_add(&metric_entry->list, &metric_queue); } rcu_read_unlock(); + list_for_each_entry_safe(metric_entry, metric_safe, &metric_queue, list) { + batadv_v_elp_throughput_metric_update(metric_entry->hardif_neigh); + + batadv_hardif_neigh_put(metric_entry->hardif_neigh); + list_del(&metric_entry->list); + kfree(metric_entry); + } + restart_timer: batadv_v_elp_start_timer(hard_iface); out: diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h index 9e2740195fa2..c9cb0a307100 100644 --- a/net/batman-adv/bat_v_elp.h +++ b/net/batman-adv/bat_v_elp.h @@ -10,7 +10,6 @@ #include "main.h" #include -#include int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface); void batadv_v_elp_iface_disable(struct batadv_hard_iface *hard_iface); @@ -19,6 +18,5 @@ void batadv_v_elp_iface_activate(struct batadv_hard_iface *primary_iface, void batadv_v_elp_primary_iface_set(struct batadv_hard_iface *primary_iface); int batadv_v_elp_packet_recv(struct sk_buff *skb, struct batadv_hard_iface *if_incoming); -void batadv_v_elp_throughput_metric_update(struct work_struct *work); #endif /* _NET_BATMAN_ADV_BAT_V_ELP_H_ */ diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 04f6398b3a40..85a50096f5b2 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -596,9 +596,6 @@ struct batadv_hardif_neigh_node_bat_v { * neighbor */ unsigned long last_unicast_tx; - - /** @metric_work: work queue callback item for metric update */ - struct work_struct metric_work; }; /** From dddc53806dd2a10e210d5ea08caec6d3f92440b2 Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Thu, 16 Jan 2025 10:46:59 -0800 Subject: [PATCH 0149/1443] drm/xe/ptl: Apply Wa_13011645652 Extend Wa_13011645652 to PTL. Signed-off-by: Vinay Belgaumkar Reviewed-by: Stuart Summers Signed-off-by: Daniele Ceraolo Spurio Link: https://patchwork.freedesktop.org/patch/msgid/20250116184659.384874-1-vinay.belgaumkar@intel.com --- drivers/gpu/drm/xe/xe_wa_oob.rules | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules index 40438c3d9b72..228436532282 100644 --- a/drivers/gpu/drm/xe/xe_wa_oob.rules +++ b/drivers/gpu/drm/xe/xe_wa_oob.rules @@ -28,6 +28,7 @@ 16022287689 GRAPHICS_VERSION(2001) GRAPHICS_VERSION(2004) 13011645652 GRAPHICS_VERSION(2004) + GRAPHICS_VERSION(3001) 14022293748 GRAPHICS_VERSION(2001) GRAPHICS_VERSION(2004) 22019794406 GRAPHICS_VERSION(2001) From 3fafa6a02be219ddd05d6201911534a34135cb82 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 20 Jan 2025 15:35:01 +0100 Subject: [PATCH 0150/1443] dt-bindings: interrupt-controller: microchip,lan966x-oic: Clarify endpoint use Reword the description, to make it clear that the LAN966x Outbound Interrupt Controller is used only in PCI endpoint mode. Signed-off-by: Geert Uytterhoeven Signed-off-by: Thomas Gleixner Acked-by: Krzysztof Kozlowski Acked-by: Herve Codina Link: https://lore.kernel.org/all/247b1185c93610100f3f8c9e0ab2c1506e53e1f4.1737383314.git.geert+renesas@glider.be --- .../bindings/interrupt-controller/microchip,lan966x-oic.yaml | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/interrupt-controller/microchip,lan966x-oic.yaml b/Documentation/devicetree/bindings/interrupt-controller/microchip,lan966x-oic.yaml index b2adc7174177..dca16e202da9 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/microchip,lan966x-oic.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/microchip,lan966x-oic.yaml @@ -14,9 +14,8 @@ allOf: description: | The Microchip LAN966x outband interrupt controller (OIC) maps the internal - interrupt sources of the LAN966x device to an external interrupt. - When the LAN966x device is used as a PCI device, the external interrupt is - routed to the PCI interrupt. + interrupt sources of the LAN966x device to a PCI interrupt when the LAN966x + device is used as a PCI device. properties: compatible: From e06c9e3682f58fbeb632b7b866bb4fe66a4a4b42 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 20 Jan 2025 15:35:02 +0100 Subject: [PATCH 0151/1443] irqchip/lan966x-oic: Make CONFIG_LAN966X_OIC depend on CONFIG_MCHP_LAN966X_PCI The Microchip LAN966x outband interrupt controller is only present on Microchip LAN966x SoCs, and only used in PCI endpoint mode. Hence add a dependency on MCHP_LAN966X_PCI, to prevent asking the user about this driver when configuring a kernel without Microchip LAN966x PCIe support. Fixes: 3e3a7b35332924c8 ("irqchip: Add support for LAN966x OIC") Signed-off-by: Geert Uytterhoeven Signed-off-by: Thomas Gleixner Acked-by: Herve Codina Link: https://lore.kernel.org/all/28e8a605e72ee45e27f0d06b2b71366159a9c782.1737383314.git.geert+renesas@glider.be --- drivers/irqchip/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig index be063bfb50c4..c11b9965c4ad 100644 --- a/drivers/irqchip/Kconfig +++ b/drivers/irqchip/Kconfig @@ -169,6 +169,7 @@ config IXP4XX_IRQ config LAN966X_OIC tristate "Microchip LAN966x OIC Support" + depends on MCHP_LAN966X_PCI || COMPILE_TEST select GENERIC_IRQ_CHIP select IRQ_DOMAIN help From ed868bcb4f5cecbb61c4f057aa7550650643ac3b Mon Sep 17 00:00:00 2001 From: Jesse Van Gavere Date: Mon, 6 Jan 2025 13:10:54 +0100 Subject: [PATCH 0152/1443] drm/bridge: adv7511: Switch to atomic operations Use the atomic version of enable/disable. To support bridges where bus format negotiation is needed such as TIDSS we need to implement atomic_get_input_bus_fmts, prepare the driver for this by switching the existing operations to it's atomic variants. Signed-off-by: Jesse Van Gavere Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20250106121054.96739-1-jesseevg@gmail.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/bridge/adv7511/adv7511_drv.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index fda66e7876b8..1a76aea6a945 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -910,14 +910,16 @@ static struct adv7511 *bridge_to_adv7511(struct drm_bridge *bridge) return container_of(bridge, struct adv7511, bridge); } -static void adv7511_bridge_enable(struct drm_bridge *bridge) +static void adv7511_bridge_atomic_enable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { struct adv7511 *adv = bridge_to_adv7511(bridge); adv7511_power_on(adv); } -static void adv7511_bridge_disable(struct drm_bridge *bridge) +static void adv7511_bridge_atomic_disable(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state) { struct adv7511 *adv = bridge_to_adv7511(bridge); @@ -996,14 +998,18 @@ static void adv7511_bridge_hpd_notify(struct drm_bridge *bridge, } static const struct drm_bridge_funcs adv7511_bridge_funcs = { - .enable = adv7511_bridge_enable, - .disable = adv7511_bridge_disable, .mode_set = adv7511_bridge_mode_set, .mode_valid = adv7511_bridge_mode_valid, .attach = adv7511_bridge_attach, .detect = adv7511_bridge_detect, .edid_read = adv7511_bridge_edid_read, .hpd_notify = adv7511_bridge_hpd_notify, + + .atomic_enable = adv7511_bridge_atomic_enable, + .atomic_disable = adv7511_bridge_atomic_disable, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, }; /* ----------------------------------------------------------------------------- From 0936f0e54426177b0f0263ddf806ed5e13487db6 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Sun, 22 Dec 2024 07:00:41 +0200 Subject: [PATCH 0153/1443] drm/atomic-helper: document drm_atomic_helper_check() restrictions The drm_atomic_helper_check() calls drm_atomic_helper_check_modeset() insternally. Document that corresponding restrictions also apply to the drivers that call the former function (as it's easy to miss the documentation for the latter function). Reviewed-by: Simona Vetter Reviewed-by: Abhinav Kumar Link: https://patchwork.freedesktop.org/patch/msgid/20241222-drm-dirty-modeset-v1-1-0e76a53eceb9@linaro.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/drm_atomic_helper.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 40e4e1b6c911..3034ba09c0ee 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -1087,6 +1087,15 @@ EXPORT_SYMBOL(drm_atomic_helper_check_planes); * For example enable/disable of a cursor plane which have fixed zpos value * would trigger all other enabled planes to be forced to the state change. * + * IMPORTANT: + * + * As this function calls drm_atomic_helper_check_modeset() internally, its + * restrictions also apply: + * Drivers which set &drm_crtc_state.mode_changed (e.g. in their + * &drm_plane_helper_funcs.atomic_check hooks if a plane update can't be done + * without a full modeset) _must_ call drm_atomic_helper_check_modeset() + * function again after that change. + * * RETURNS: * Zero for success or -errno */ From 6ce24b3450b8e8132b74d4f0b43a48f4e370e825 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Thu, 23 Jan 2025 19:20:55 +0800 Subject: [PATCH 0154/1443] drm/panel-edp: Add STA 116QHD024002 Add support for the STA 116QHD024002, pleace the EDID here for subsequent reference. 00 ff ff ff ff ff ff 00 4e 81 09 00 00 00 00 00 26 21 01 04 a5 1a 0e 78 02 1e b5 9a 5f 57 94 26 0f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 8e 1c 56 a0 50 00 1e 30 28 20 55 00 00 90 10 00 00 18 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fe 00 20 20 20 20 20 20 0a 20 20 20 20 20 20 00 00 00 fe 00 31 31 36 51 48 44 30 32 34 30 30 32 0a 00 3b Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250123112055.1521471-1-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-edp.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 1721a3397dbf..25bb7e454fc9 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1808,6 +1808,12 @@ static const struct panel_delay delay_200_150_e50 = { .enable = 50, }; +static const struct panel_delay delay_200_500_e250 = { + .hpd_absent = 200, + .unprepare = 500, + .enable = 250, +}; + #define EDP_PANEL_ENTRY(vend_chr_0, vend_chr_1, vend_chr_2, product_id, _delay, _name) \ { \ .ident = { \ @@ -1998,6 +2004,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('S', 'H', 'P', 0x1593, &delay_200_500_p2e100, "LQ134N1"), EDP_PANEL_ENTRY('S', 'T', 'A', 0x0004, &delay_200_500_e200, "116KHD024006"), + EDP_PANEL_ENTRY('S', 'T', 'A', 0x0009, &delay_200_500_e250, "116QHD024002"), EDP_PANEL_ENTRY('S', 'T', 'A', 0x0100, &delay_100_500_e200, "2081116HHD028001-51D"), { /* sentinal */ } From d3d380eded7ee5fc2fc53b3b0e72365ded025c4a Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Thu, 9 Jan 2025 14:30:47 +0100 Subject: [PATCH 0155/1443] nvme-fc: go straight to connecting state when initializing The initial controller initialization mimiks the reconnect loop behavior by switching from NEW to RESETTING and then to CONNECTING. The transition from NEW to CONNECTING is a valid transition, so there is no point entering the RESETTING state. TCP and RDMA also transition directly to CONNECTING state. Reviewed-by: Sagi Grimberg Reviewed-by: Hannes Reinecke Reviewed-by: Christoph Hellwig Signed-off-by: Daniel Wagner Signed-off-by: Keith Busch --- drivers/nvme/host/fc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 094be164ffdc..7409da42b9ee 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3578,8 +3578,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, list_add_tail(&ctrl->ctrl_list, &rport->ctrl_list); spin_unlock_irqrestore(&rport->lock, flags); - if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_RESETTING) || - !nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) { dev_err(ctrl->ctrl.device, "NVME-FC{%d}: failed to init ctrl state\n", ctrl->cnum); goto fail_ctrl; From 294b2b7516fd06a8dd82e4a6118f318ec521e706 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Thu, 9 Jan 2025 14:30:48 +0100 Subject: [PATCH 0156/1443] nvme: handle connectivity loss in nvme_set_queue_count When the set feature attempts fails with any NVME status code set in nvme_set_queue_count, the function still report success. Though the numbers of queues set to 0. This is done to support controllers in degraded state (the admin queue is still up and running but no IO queues). Though there is an exception. When nvme_set_features reports an host path error, nvme_set_queue_count should propagate this error as the connectivity is lost, which means also the admin queue is not working anymore. Fixes: 9a0be7abb62f ("nvme: refactor set_queue_count") Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Sagi Grimberg Signed-off-by: Daniel Wagner Signed-off-by: Keith Busch --- drivers/nvme/host/core.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 0d21258e2283..2bcd9f710cb6 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -1695,7 +1695,13 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count) status = nvme_set_features(ctrl, NVME_FEAT_NUM_QUEUES, q_count, NULL, 0, &result); - if (status < 0) + + /* + * It's either a kernel error or the host observed a connection + * lost. In either case it's not possible communicate with the + * controller and thus enter the error code path. + */ + if (status < 0 || status == NVME_SC_HOST_PATH_ERROR) return status; /* From ee59e3820ca92a9f4307ae23dfc7229dc8b8d400 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Thu, 9 Jan 2025 14:30:49 +0100 Subject: [PATCH 0157/1443] nvme-fc: do not ignore connectivity loss during connecting When a connectivity loss occurs while nvme_fc_create_assocation is being executed, it's possible that the ctrl ends up stuck in the LIVE state: 1) nvme nvme10: NVME-FC{10}: create association : ... 2) nvme nvme10: NVME-FC{10}: controller connectivity lost. Awaiting Reconnect nvme nvme10: queue_size 128 > ctrl maxcmd 32, reducing to maxcmd 3) nvme nvme10: Could not set queue count (880) nvme nvme10: Failed to configure AEN (cfg 900) 4) nvme nvme10: NVME-FC{10}: controller connect complete 5) nvme nvme10: failed nvme_keep_alive_end_io error=4 A connection attempt starts 1) and the ctrl is in state CONNECTING. Shortly after the LLDD driver detects a connection lost event and calls nvme_fc_ctrl_connectivity_loss 2). Because we are still in CONNECTING state, this event is ignored. nvme_fc_create_association continues to run in parallel and tries to communicate with the controller and these commands will fail. Though these errors are filtered out, e.g in 3) setting the I/O queues numbers fails which leads to an early exit in nvme_fc_create_io_queues. Because the number of IO queues is 0 at this point, there is nothing left in nvme_fc_create_association which could detected the connection drop. Thus the ctrl enters LIVE state 4). Eventually the keep alive handler times out 5) but because nothing is being done, the ctrl stays in LIVE state. There is already the ASSOC_FAILED flag to track connectivity loss event but this bit is set too late in the recovery code path. Move this into the connectivity loss event handler and synchronize it with the state change. This ensures that the ASSOC_FAILED flag is seen by nvme_fc_create_io_queues and it does not enter the LIVE state after a connectivity loss event. If the connectivity loss event happens after we entered the LIVE state the normal error recovery path is executed. Signed-off-by: Daniel Wagner Reviewed-by: Hannes Reinecke Reviewed-by: Sagi Grimberg Signed-off-by: Keith Busch --- drivers/nvme/host/fc.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 7409da42b9ee..55884d3df6f2 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -781,11 +781,19 @@ nvme_fc_abort_lsops(struct nvme_fc_rport *rport) static void nvme_fc_ctrl_connectivity_loss(struct nvme_fc_ctrl *ctrl) { + enum nvme_ctrl_state state; + unsigned long flags; + dev_info(ctrl->ctrl.device, "NVME-FC{%d}: controller connectivity lost. Awaiting " "Reconnect", ctrl->cnum); - switch (nvme_ctrl_state(&ctrl->ctrl)) { + spin_lock_irqsave(&ctrl->lock, flags); + set_bit(ASSOC_FAILED, &ctrl->flags); + state = nvme_ctrl_state(&ctrl->ctrl); + spin_unlock_irqrestore(&ctrl->lock, flags); + + switch (state) { case NVME_CTRL_NEW: case NVME_CTRL_LIVE: /* @@ -2542,7 +2550,6 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) */ if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { __nvme_fc_abort_outstanding_ios(ctrl, true); - set_bit(ASSOC_FAILED, &ctrl->flags); dev_warn(ctrl->ctrl.device, "NVME-FC{%d}: transport error during (re)connect\n", ctrl->cnum); @@ -3167,12 +3174,18 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) else ret = nvme_fc_recreate_io_queues(ctrl); } - if (!ret && test_bit(ASSOC_FAILED, &ctrl->flags)) - ret = -EIO; if (ret) goto out_term_aen_ops; - changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); + spin_lock_irqsave(&ctrl->lock, flags); + if (!test_bit(ASSOC_FAILED, &ctrl->flags)) + changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); + else + ret = -EIO; + spin_unlock_irqrestore(&ctrl->lock, flags); + + if (ret) + goto out_term_aen_ops; ctrl->ctrl.nr_reconnects = 0; From 27af31e44949fa85550176520ef7086a0d00fd7b Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 16 Jan 2025 18:07:45 +0200 Subject: [PATCH 0158/1443] hrtimers: Mark is_migration_base() with __always_inline When is_migration_base() is unused, it prevents kernel builds with clang, `make W=1` and CONFIG_WERROR=y: kernel/time/hrtimer.c:156:20: error: unused function 'is_migration_base' [-Werror,-Wunused-function] 156 | static inline bool is_migration_base(struct hrtimer_clock_base *base) | ^~~~~~~~~~~~~~~~~ Fix this by marking it with __always_inline. [ tglx: Use __always_inline instead of __maybe_unused and move it into the usage sites conditional ] Signed-off-by: Andy Shevchenko Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/20250116160745.243358-1-andriy.shevchenko@linux.intel.com --- kernel/time/hrtimer.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index f6d8df94045c..4fb81f8c6f1c 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -145,11 +145,6 @@ static struct hrtimer_cpu_base migration_cpu_base = { #define migration_base migration_cpu_base.clock_base[0] -static inline bool is_migration_base(struct hrtimer_clock_base *base) -{ - return base == &migration_base; -} - /* * We are using hashed locking: holding per_cpu(hrtimer_bases)[n].lock * means that all timers which are tied to this base via timer->base are @@ -275,11 +270,6 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, #else /* CONFIG_SMP */ -static inline bool is_migration_base(struct hrtimer_clock_base *base) -{ - return false; -} - static inline struct hrtimer_clock_base * lock_hrtimer_base(const struct hrtimer *timer, unsigned long *flags) __acquires(&timer->base->cpu_base->lock) @@ -1370,6 +1360,18 @@ static void hrtimer_sync_wait_running(struct hrtimer_cpu_base *cpu_base, } } +#ifdef CONFIG_SMP +static __always_inline bool is_migration_base(struct hrtimer_clock_base *base) +{ + return base == &migration_base; +} +#else +static __always_inline bool is_migration_base(struct hrtimer_clock_base *base) +{ + return false; +} +#endif + /* * This function is called on PREEMPT_RT kernels when the fast path * deletion of a timer failed because the timer callback function was From 53dac345395c0d2493cbc2f4c85fe38aef5b63f5 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 18 Jan 2025 00:24:33 +0100 Subject: [PATCH 0159/1443] hrtimers: Force migrate away hrtimers queued after CPUHP_AP_HRTIMERS_DYING hrtimers are migrated away from the dying CPU to any online target at the CPUHP_AP_HRTIMERS_DYING stage in order not to delay bandwidth timers handling tasks involved in the CPU hotplug forward progress. However wakeups can still be performed by the outgoing CPU after CPUHP_AP_HRTIMERS_DYING. Those can result again in bandwidth timers being armed. Depending on several considerations (crystal ball power management based election, earliest timer already enqueued, timer migration enabled or not), the target may eventually be the current CPU even if offline. If that happens, the timer is eventually ignored. The most notable example is RCU which had to deal with each and every of those wake-ups by deferring them to an online CPU, along with related workarounds: _ e787644caf76 (rcu: Defer RCU kthreads wakeup when CPU is dying) _ 9139f93209d1 (rcu/nocb: Fix RT throttling hrtimer armed from offline CPU) _ f7345ccc62a4 (rcu/nocb: Fix rcuog wake-up from offline softirq) The problem isn't confined to RCU though as the stop machine kthread (which runs CPUHP_AP_HRTIMERS_DYING) reports its completion at the end of its work through cpu_stop_signal_done() and performs a wake up that eventually arms the deadline server timer: WARNING: CPU: 94 PID: 588 at kernel/time/hrtimer.c:1086 hrtimer_start_range_ns+0x289/0x2d0 CPU: 94 UID: 0 PID: 588 Comm: migration/94 Not tainted Stopper: multi_cpu_stop+0x0/0x120 <- stop_machine_cpuslocked+0x66/0xc0 RIP: 0010:hrtimer_start_range_ns+0x289/0x2d0 Call Trace: start_dl_timer enqueue_dl_entity dl_server_start enqueue_task_fair enqueue_task ttwu_do_activate try_to_wake_up complete cpu_stopper_thread Instead of providing yet another bandaid to work around the situation, fix it in the hrtimers infrastructure instead: always migrate away a timer to an online target whenever it is enqueued from an offline CPU. This will also allow to revert all the above RCU disgraceful hacks. Fixes: 5c0930ccaad5 ("hrtimers: Push pending hrtimers away from outgoing CPU earlier") Reported-by: Vlad Poenaru Reported-by: Usama Arif Signed-off-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Tested-by: Paul E. McKenney Link: https://lore.kernel.org/all/20250117232433.24027-1-frederic@kernel.org Closes: 20241213203739.1519801-1-usamaarif642@gmail.com --- include/linux/hrtimer_defs.h | 1 + kernel/time/hrtimer.c | 103 ++++++++++++++++++++++++++++------- 2 files changed, 83 insertions(+), 21 deletions(-) diff --git a/include/linux/hrtimer_defs.h b/include/linux/hrtimer_defs.h index c3b4b7ed7c16..84a5045f80f3 100644 --- a/include/linux/hrtimer_defs.h +++ b/include/linux/hrtimer_defs.h @@ -125,6 +125,7 @@ struct hrtimer_cpu_base { ktime_t softirq_expires_next; struct hrtimer *softirq_next_timer; struct hrtimer_clock_base clock_base[HRTIMER_MAX_CLOCK_BASES]; + call_single_data_t csd; } ____cacheline_aligned; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 4fb81f8c6f1c..deb1aa32814e 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -58,6 +58,8 @@ #define HRTIMER_ACTIVE_SOFT (HRTIMER_ACTIVE_HARD << MASK_SHIFT) #define HRTIMER_ACTIVE_ALL (HRTIMER_ACTIVE_SOFT | HRTIMER_ACTIVE_HARD) +static void retrigger_next_event(void *arg); + /* * The timer bases: * @@ -111,7 +113,8 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = .clockid = CLOCK_TAI, .get_time = &ktime_get_clocktai, }, - } + }, + .csd = CSD_INIT(retrigger_next_event, NULL) }; static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { @@ -124,6 +127,14 @@ static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { [CLOCK_TAI] = HRTIMER_BASE_TAI, }; +static inline bool hrtimer_base_is_online(struct hrtimer_cpu_base *base) +{ + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) + return true; + else + return likely(base->online); +} + /* * Functions and macros which are different for UP/SMP systems are kept in a * single place @@ -178,27 +189,54 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer, } /* - * We do not migrate the timer when it is expiring before the next - * event on the target cpu. When high resolution is enabled, we cannot - * reprogram the target cpu hardware and we would cause it to fire - * late. To keep it simple, we handle the high resolution enabled and - * disabled case similar. + * Check if the elected target is suitable considering its next + * event and the hotplug state of the current CPU. + * + * If the elected target is remote and its next event is after the timer + * to queue, then a remote reprogram is necessary. However there is no + * guarantee the IPI handling the operation would arrive in time to meet + * the high resolution deadline. In this case the local CPU becomes a + * preferred target, unless it is offline. + * + * High and low resolution modes are handled the same way for simplicity. * * Called with cpu_base->lock of target cpu held. */ -static int -hrtimer_check_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base) +static bool hrtimer_suitable_target(struct hrtimer *timer, struct hrtimer_clock_base *new_base, + struct hrtimer_cpu_base *new_cpu_base, + struct hrtimer_cpu_base *this_cpu_base) { ktime_t expires; + /* + * The local CPU clockevent can be reprogrammed. Also get_target_base() + * guarantees it is online. + */ + if (new_cpu_base == this_cpu_base) + return true; + + /* + * The offline local CPU can't be the default target if the + * next remote target event is after this timer. Keep the + * elected new base. An IPI will we issued to reprogram + * it as a last resort. + */ + if (!hrtimer_base_is_online(this_cpu_base)) + return true; + expires = ktime_sub(hrtimer_get_expires(timer), new_base->offset); - return expires < new_base->cpu_base->expires_next; + + return expires >= new_base->cpu_base->expires_next; } -static inline -struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, - int pinned) +static inline struct hrtimer_cpu_base *get_target_base(struct hrtimer_cpu_base *base, int pinned) { + if (!hrtimer_base_is_online(base)) { + int cpu = cpumask_any_and(cpu_online_mask, housekeeping_cpumask(HK_TYPE_TIMER)); + + return &per_cpu(hrtimer_bases, cpu); + } + #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) if (static_branch_likely(&timers_migration_enabled) && !pinned) return &per_cpu(hrtimer_bases, get_nohz_timer_target()); @@ -249,8 +287,8 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, raw_spin_unlock(&base->cpu_base->lock); raw_spin_lock(&new_base->cpu_base->lock); - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, + this_cpu_base)) { raw_spin_unlock(&new_base->cpu_base->lock); raw_spin_lock(&base->cpu_base->lock); new_cpu_base = this_cpu_base; @@ -259,8 +297,7 @@ switch_hrtimer_base(struct hrtimer *timer, struct hrtimer_clock_base *base, } WRITE_ONCE(timer->base, new_base); } else { - if (new_cpu_base != this_cpu_base && - hrtimer_check_target(timer, new_base)) { + if (!hrtimer_suitable_target(timer, new_base, new_cpu_base, this_cpu_base)) { new_cpu_base = this_cpu_base; goto again; } @@ -706,8 +743,6 @@ static inline int hrtimer_is_hres_enabled(void) return hrtimer_hres_enabled; } -static void retrigger_next_event(void *arg); - /* * Switch to high resolution mode */ @@ -1195,6 +1230,7 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, u64 delta_ns, const enum hrtimer_mode mode, struct hrtimer_clock_base *base) { + struct hrtimer_cpu_base *this_cpu_base = this_cpu_ptr(&hrtimer_bases); struct hrtimer_clock_base *new_base; bool force_local, first; @@ -1206,9 +1242,15 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, * and enforce reprogramming after it is queued no matter whether * it is the new first expiring timer again or not. */ - force_local = base->cpu_base == this_cpu_ptr(&hrtimer_bases); + force_local = base->cpu_base == this_cpu_base; force_local &= base->cpu_base->next_timer == timer; + /* + * Don't force local queuing if this enqueue happens on a unplugged + * CPU after hrtimer_cpu_dying() has been invoked. + */ + force_local &= this_cpu_base->online; + /* * Remove an active timer from the queue. In case it is not queued * on the current CPU, make sure that remove_hrtimer() updates the @@ -1238,8 +1280,27 @@ static int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, } first = enqueue_hrtimer(timer, new_base, mode); - if (!force_local) - return first; + if (!force_local) { + /* + * If the current CPU base is online, then the timer is + * never queued on a remote CPU if it would be the first + * expiring timer there. + */ + if (hrtimer_base_is_online(this_cpu_base)) + return first; + + /* + * Timer was enqueued remote because the current base is + * already offline. If the timer is the first to expire, + * kick the remote CPU to reprogram the clock event. + */ + if (first) { + struct hrtimer_cpu_base *new_cpu_base = new_base->cpu_base; + + smp_call_function_single_async(new_cpu_base->cpu, &new_cpu_base->csd); + } + return 0; + } /* * Timer was forced to stay on the current CPU to avoid From 93c66fbc280747ea700bd6199633d661e3c819b3 Mon Sep 17 00:00:00 2001 From: Joe Hattori Date: Fri, 10 Jan 2025 10:05:54 +0900 Subject: [PATCH 0160/1443] powercap: call put_device() on an error path in powercap_register_control_type() powercap_register_control_type() calls device_register(), but does not release the refcount of the device when it fails. Call put_device() before returning an error to balance the refcount. Since the kfree(control_type) will be done by powercap_release(), remove the lines in powercap_register_control_type() before returning the error. This bug was found by an experimental verifier that I am developing. Signed-off-by: Joe Hattori Link: https://patch.msgid.link/20250110010554.1583411-1-joe@pf.is.s.u-tokyo.ac.jp [ rjw: Changelog edits ] Signed-off-by: Rafael J. Wysocki --- drivers/powercap/powercap_sys.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/powercap/powercap_sys.c b/drivers/powercap/powercap_sys.c index 52c32dcbf7d8..4112a0097338 100644 --- a/drivers/powercap/powercap_sys.c +++ b/drivers/powercap/powercap_sys.c @@ -627,8 +627,7 @@ struct powercap_control_type *powercap_register_control_type( dev_set_name(&control_type->dev, "%s", name); result = device_register(&control_type->dev); if (result) { - if (control_type->allocated) - kfree(control_type); + put_device(&control_type->dev); return ERR_PTR(result); } idr_init(&control_type->idr); From a216542027b892e6651c1b4e076012140d04afaf Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 10 Jan 2025 15:22:24 +0000 Subject: [PATCH 0161/1443] btrfs: fix lockdep splat while merging a relocation root When COWing a relocation tree path, at relocation.c:replace_path(), we can trigger a lockdep splat while we are in the btrfs_search_slot() call against the relocation root. This happens in that callchain at ctree.c:read_block_for_search() when we happen to find a child extent buffer already loaded through the fs tree with a lockdep class set to the fs tree. So when we attempt to lock that extent buffer through a relocation tree we have to reset the lockdep class to the class for a relocation tree, since a relocation tree has extent buffers that used to belong to a fs tree and may currently be already loaded (we swap extent buffers between the two trees at the end of replace_path()). However we are missing calls to btrfs_maybe_reset_lockdep_class() to reset the lockdep class at ctree.c:read_block_for_search() before we read lock an extent buffer, just like we did for btrfs_search_slot() in commit b40130b23ca4 ("btrfs: fix lockdep splat with reloc root extent buffers"). So add the missing btrfs_maybe_reset_lockdep_class() calls before the attempts to read lock an extent buffer at ctree.c:read_block_for_search(). The lockdep splat was reported by syzbot and it looks like this: ====================================================== WARNING: possible circular locking dependency detected 6.13.0-rc5-syzkaller-00163-gab75170520d4 #0 Not tainted ------------------------------------------------------ syz.0.0/5335 is trying to acquire lock: ffff8880545dbc38 (btrfs-tree-01){++++}-{4:4}, at: btrfs_tree_read_lock_nested+0x2f/0x250 fs/btrfs/locking.c:146 but task is already holding lock: ffff8880545dba58 (btrfs-treloc-02/1){+.+.}-{4:4}, at: btrfs_tree_lock_nested+0x2f/0x250 fs/btrfs/locking.c:189 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (btrfs-treloc-02/1){+.+.}-{4:4}: reacquire_held_locks+0x3eb/0x690 kernel/locking/lockdep.c:5374 __lock_release kernel/locking/lockdep.c:5563 [inline] lock_release+0x396/0xa30 kernel/locking/lockdep.c:5870 up_write+0x79/0x590 kernel/locking/rwsem.c:1629 btrfs_force_cow_block+0x14b3/0x1fd0 fs/btrfs/ctree.c:660 btrfs_cow_block+0x371/0x830 fs/btrfs/ctree.c:755 btrfs_search_slot+0xc01/0x3180 fs/btrfs/ctree.c:2153 replace_path+0x1243/0x2740 fs/btrfs/relocation.c:1224 merge_reloc_root+0xc46/0x1ad0 fs/btrfs/relocation.c:1692 merge_reloc_roots+0x3b3/0x980 fs/btrfs/relocation.c:1942 relocate_block_group+0xb0a/0xd40 fs/btrfs/relocation.c:3754 btrfs_relocate_block_group+0x77d/0xd90 fs/btrfs/relocation.c:4087 btrfs_relocate_chunk+0x12c/0x3b0 fs/btrfs/volumes.c:3494 __btrfs_balance+0x1b0f/0x26b0 fs/btrfs/volumes.c:4278 btrfs_balance+0xbdc/0x10c0 fs/btrfs/volumes.c:4655 btrfs_ioctl_balance+0x493/0x7c0 fs/btrfs/ioctl.c:3670 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl+0xf5/0x170 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f -> #1 (btrfs-tree-01/1){+.+.}-{4:4}: lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5849 down_write_nested+0xa2/0x220 kernel/locking/rwsem.c:1693 btrfs_tree_lock_nested+0x2f/0x250 fs/btrfs/locking.c:189 btrfs_init_new_buffer fs/btrfs/extent-tree.c:5052 [inline] btrfs_alloc_tree_block+0x41c/0x1440 fs/btrfs/extent-tree.c:5132 btrfs_force_cow_block+0x526/0x1fd0 fs/btrfs/ctree.c:573 btrfs_cow_block+0x371/0x830 fs/btrfs/ctree.c:755 btrfs_search_slot+0xc01/0x3180 fs/btrfs/ctree.c:2153 btrfs_insert_empty_items+0x9c/0x1a0 fs/btrfs/ctree.c:4351 btrfs_insert_empty_item fs/btrfs/ctree.h:688 [inline] btrfs_insert_inode_ref+0x2bb/0xf80 fs/btrfs/inode-item.c:330 btrfs_rename_exchange fs/btrfs/inode.c:7990 [inline] btrfs_rename2+0xcb7/0x2b90 fs/btrfs/inode.c:8374 vfs_rename+0xbdb/0xf00 fs/namei.c:5067 do_renameat2+0xd94/0x13f0 fs/namei.c:5224 __do_sys_renameat2 fs/namei.c:5258 [inline] __se_sys_renameat2 fs/namei.c:5255 [inline] __x64_sys_renameat2+0xce/0xe0 fs/namei.c:5255 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f -> #0 (btrfs-tree-01){++++}-{4:4}: check_prev_add kernel/locking/lockdep.c:3161 [inline] check_prevs_add kernel/locking/lockdep.c:3280 [inline] validate_chain+0x18ef/0x5920 kernel/locking/lockdep.c:3904 __lock_acquire+0x1397/0x2100 kernel/locking/lockdep.c:5226 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5849 down_read_nested+0xb5/0xa50 kernel/locking/rwsem.c:1649 btrfs_tree_read_lock_nested+0x2f/0x250 fs/btrfs/locking.c:146 btrfs_tree_read_lock fs/btrfs/locking.h:188 [inline] read_block_for_search+0x718/0xbb0 fs/btrfs/ctree.c:1610 btrfs_search_slot+0x1274/0x3180 fs/btrfs/ctree.c:2237 replace_path+0x1243/0x2740 fs/btrfs/relocation.c:1224 merge_reloc_root+0xc46/0x1ad0 fs/btrfs/relocation.c:1692 merge_reloc_roots+0x3b3/0x980 fs/btrfs/relocation.c:1942 relocate_block_group+0xb0a/0xd40 fs/btrfs/relocation.c:3754 btrfs_relocate_block_group+0x77d/0xd90 fs/btrfs/relocation.c:4087 btrfs_relocate_chunk+0x12c/0x3b0 fs/btrfs/volumes.c:3494 __btrfs_balance+0x1b0f/0x26b0 fs/btrfs/volumes.c:4278 btrfs_balance+0xbdc/0x10c0 fs/btrfs/volumes.c:4655 btrfs_ioctl_balance+0x493/0x7c0 fs/btrfs/ioctl.c:3670 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl+0xf5/0x170 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f other info that might help us debug this: Chain exists of: btrfs-tree-01 --> btrfs-tree-01/1 --> btrfs-treloc-02/1 Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(btrfs-treloc-02/1); lock(btrfs-tree-01/1); lock(btrfs-treloc-02/1); rlock(btrfs-tree-01); *** DEADLOCK *** 8 locks held by syz.0.0/5335: #0: ffff88801e3ae420 (sb_writers#13){.+.+}-{0:0}, at: mnt_want_write_file+0x5e/0x200 fs/namespace.c:559 #1: ffff888052c760d0 (&fs_info->reclaim_bgs_lock){+.+.}-{4:4}, at: __btrfs_balance+0x4c2/0x26b0 fs/btrfs/volumes.c:4183 #2: ffff888052c74850 (&fs_info->cleaner_mutex){+.+.}-{4:4}, at: btrfs_relocate_block_group+0x775/0xd90 fs/btrfs/relocation.c:4086 #3: ffff88801e3ae610 (sb_internal#2){.+.+}-{0:0}, at: merge_reloc_root+0xf11/0x1ad0 fs/btrfs/relocation.c:1659 #4: ffff888052c76470 (btrfs_trans_num_writers){++++}-{0:0}, at: join_transaction+0x405/0xda0 fs/btrfs/transaction.c:288 #5: ffff888052c76498 (btrfs_trans_num_extwriters){++++}-{0:0}, at: join_transaction+0x405/0xda0 fs/btrfs/transaction.c:288 #6: ffff8880545db878 (btrfs-tree-01/1){+.+.}-{4:4}, at: btrfs_tree_lock_nested+0x2f/0x250 fs/btrfs/locking.c:189 #7: ffff8880545dba58 (btrfs-treloc-02/1){+.+.}-{4:4}, at: btrfs_tree_lock_nested+0x2f/0x250 fs/btrfs/locking.c:189 stack backtrace: CPU: 0 UID: 0 PID: 5335 Comm: syz.0.0 Not tainted 6.13.0-rc5-syzkaller-00163-gab75170520d4 #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 print_circular_bug+0x13a/0x1b0 kernel/locking/lockdep.c:2074 check_noncircular+0x36a/0x4a0 kernel/locking/lockdep.c:2206 check_prev_add kernel/locking/lockdep.c:3161 [inline] check_prevs_add kernel/locking/lockdep.c:3280 [inline] validate_chain+0x18ef/0x5920 kernel/locking/lockdep.c:3904 __lock_acquire+0x1397/0x2100 kernel/locking/lockdep.c:5226 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5849 down_read_nested+0xb5/0xa50 kernel/locking/rwsem.c:1649 btrfs_tree_read_lock_nested+0x2f/0x250 fs/btrfs/locking.c:146 btrfs_tree_read_lock fs/btrfs/locking.h:188 [inline] read_block_for_search+0x718/0xbb0 fs/btrfs/ctree.c:1610 btrfs_search_slot+0x1274/0x3180 fs/btrfs/ctree.c:2237 replace_path+0x1243/0x2740 fs/btrfs/relocation.c:1224 merge_reloc_root+0xc46/0x1ad0 fs/btrfs/relocation.c:1692 merge_reloc_roots+0x3b3/0x980 fs/btrfs/relocation.c:1942 relocate_block_group+0xb0a/0xd40 fs/btrfs/relocation.c:3754 btrfs_relocate_block_group+0x77d/0xd90 fs/btrfs/relocation.c:4087 btrfs_relocate_chunk+0x12c/0x3b0 fs/btrfs/volumes.c:3494 __btrfs_balance+0x1b0f/0x26b0 fs/btrfs/volumes.c:4278 btrfs_balance+0xbdc/0x10c0 fs/btrfs/volumes.c:4655 btrfs_ioctl_balance+0x493/0x7c0 fs/btrfs/ioctl.c:3670 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl+0xf5/0x170 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f1ac6985d29 Code: ff ff c3 (...) RSP: 002b:00007f1ac63fe038 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007f1ac6b76160 RCX: 00007f1ac6985d29 RDX: 0000000020000180 RSI: 00000000c4009420 RDI: 0000000000000007 RBP: 00007f1ac6a01b08 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000001 R14: 00007f1ac6b76160 R15: 00007fffda145a88 Reported-by: syzbot+63913e558c084f7f8fdc@syzkaller.appspotmail.com Link: https://lore.kernel.org/linux-btrfs/677b3014.050a0220.3b53b0.0064.GAE@google.com/ Fixes: 99785998ed1c ("btrfs: reduce lock contention when eb cache miss for btree search") Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ctree.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 92071ca0655f..3dc5a35dd19b 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -1496,6 +1496,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, if (!p->skip_locking) { btrfs_unlock_up_safe(p, parent_level + 1); + btrfs_maybe_reset_lockdep_class(root, tmp); tmp_locked = true; btrfs_tree_read_lock(tmp); btrfs_release_path(p); @@ -1539,6 +1540,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p, if (!p->skip_locking) { ASSERT(ret == -EAGAIN); + btrfs_maybe_reset_lockdep_class(root, tmp); tmp_locked = true; btrfs_tree_read_lock(tmp); btrfs_release_path(p); From 0d85f5c2dd91df6b5da454406756f463ba923b69 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 13 Jan 2025 15:01:08 +0000 Subject: [PATCH 0162/1443] btrfs: fix assertion failure when splitting ordered extent after transaction abort If while we are doing a direct IO write a transaction abort happens, we mark all existing ordered extents with the BTRFS_ORDERED_IOERR flag (done at btrfs_destroy_ordered_extents()), and then after that if we enter btrfs_split_ordered_extent() and the ordered extent has bytes left (meaning we have a bio that doesn't cover the whole ordered extent, see details at btrfs_extract_ordered_extent()), we will fail on the following assertion at btrfs_split_ordered_extent(): ASSERT(!(flags & ~BTRFS_ORDERED_TYPE_FLAGS)); because the BTRFS_ORDERED_IOERR flag is set and the definition of BTRFS_ORDERED_TYPE_FLAGS is just the union of all flags that identify the type of write (regular, nocow, prealloc, compressed, direct IO, encoded). Fix this by returning an error from btrfs_extract_ordered_extent() if we find the BTRFS_ORDERED_IOERR flag in the ordered extent. The error will be the error that resulted in the transaction abort or -EIO if no transaction abort happened. This was recently reported by syzbot with the following trace: FAULT_INJECTION: forcing a failure. name failslab, interval 1, probability 0, space 0, times 1 CPU: 0 UID: 0 PID: 5321 Comm: syz.0.0 Not tainted 6.13.0-rc5-syzkaller #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 fail_dump lib/fault-inject.c:53 [inline] should_fail_ex+0x3b0/0x4e0 lib/fault-inject.c:154 should_failslab+0xac/0x100 mm/failslab.c:46 slab_pre_alloc_hook mm/slub.c:4072 [inline] slab_alloc_node mm/slub.c:4148 [inline] __do_kmalloc_node mm/slub.c:4297 [inline] __kmalloc_noprof+0xdd/0x4c0 mm/slub.c:4310 kmalloc_noprof include/linux/slab.h:905 [inline] kzalloc_noprof include/linux/slab.h:1037 [inline] btrfs_chunk_alloc_add_chunk_item+0x244/0x1100 fs/btrfs/volumes.c:5742 reserve_chunk_space+0x1ca/0x2c0 fs/btrfs/block-group.c:4292 check_system_chunk fs/btrfs/block-group.c:4319 [inline] do_chunk_alloc fs/btrfs/block-group.c:3891 [inline] btrfs_chunk_alloc+0x77b/0xf80 fs/btrfs/block-group.c:4187 find_free_extent_update_loop fs/btrfs/extent-tree.c:4166 [inline] find_free_extent+0x42d1/0x5810 fs/btrfs/extent-tree.c:4579 btrfs_reserve_extent+0x422/0x810 fs/btrfs/extent-tree.c:4672 btrfs_new_extent_direct fs/btrfs/direct-io.c:186 [inline] btrfs_get_blocks_direct_write+0x706/0xfa0 fs/btrfs/direct-io.c:321 btrfs_dio_iomap_begin+0xbb7/0x1180 fs/btrfs/direct-io.c:525 iomap_iter+0x697/0xf60 fs/iomap/iter.c:90 __iomap_dio_rw+0xeb9/0x25b0 fs/iomap/direct-io.c:702 btrfs_dio_write fs/btrfs/direct-io.c:775 [inline] btrfs_direct_write+0x610/0xa30 fs/btrfs/direct-io.c:880 btrfs_do_write_iter+0x2a0/0x760 fs/btrfs/file.c:1397 do_iter_readv_writev+0x600/0x880 vfs_writev+0x376/0xba0 fs/read_write.c:1050 do_pwritev fs/read_write.c:1146 [inline] __do_sys_pwritev2 fs/read_write.c:1204 [inline] __se_sys_pwritev2+0x196/0x2b0 fs/read_write.c:1195 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f1281f85d29 RSP: 002b:00007f12819fe038 EFLAGS: 00000246 ORIG_RAX: 0000000000000148 RAX: ffffffffffffffda RBX: 00007f1282176080 RCX: 00007f1281f85d29 RDX: 0000000000000001 RSI: 0000000020000240 RDI: 0000000000000005 RBP: 00007f12819fe090 R08: 0000000000000000 R09: 0000000000000003 R10: 0000000000007000 R11: 0000000000000246 R12: 0000000000000002 R13: 0000000000000000 R14: 00007f1282176080 R15: 00007ffcb9e23328 BTRFS error (device loop0 state A): Transaction aborted (error -12) BTRFS: error (device loop0 state A) in btrfs_chunk_alloc_add_chunk_item:5745: errno=-12 Out of memory BTRFS info (device loop0 state EA): forced readonly assertion failed: !(flags & ~BTRFS_ORDERED_TYPE_FLAGS), in fs/btrfs/ordered-data.c:1234 ------------[ cut here ]------------ kernel BUG at fs/btrfs/ordered-data.c:1234! Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN NOPTI CPU: 0 UID: 0 PID: 5321 Comm: syz.0.0 Not tainted 6.13.0-rc5-syzkaller #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 RIP: 0010:btrfs_split_ordered_extent+0xd8d/0xe20 fs/btrfs/ordered-data.c:1234 RSP: 0018:ffffc9000d1df2b8 EFLAGS: 00010246 RAX: 0000000000000057 RBX: 000000000006a000 RCX: 9ce21886c4195300 RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000 RBP: 0000000000000091 R08: ffffffff817f0a3c R09: 1ffff92001a3bdf4 R10: dffffc0000000000 R11: fffff52001a3bdf5 R12: 1ffff1100a45f401 R13: ffff8880522fa018 R14: dffffc0000000000 R15: 000000000006a000 FS: 00007f12819fe6c0(0000) GS:ffff88801fc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000557750bd7da8 CR3: 00000000400ea000 CR4: 0000000000352ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: btrfs_extract_ordered_extent fs/btrfs/direct-io.c:702 [inline] btrfs_dio_submit_io+0x4be/0x6d0 fs/btrfs/direct-io.c:737 iomap_dio_submit_bio fs/iomap/direct-io.c:85 [inline] iomap_dio_bio_iter+0x1022/0x1740 fs/iomap/direct-io.c:447 __iomap_dio_rw+0x13b7/0x25b0 fs/iomap/direct-io.c:703 btrfs_dio_write fs/btrfs/direct-io.c:775 [inline] btrfs_direct_write+0x610/0xa30 fs/btrfs/direct-io.c:880 btrfs_do_write_iter+0x2a0/0x760 fs/btrfs/file.c:1397 do_iter_readv_writev+0x600/0x880 vfs_writev+0x376/0xba0 fs/read_write.c:1050 do_pwritev fs/read_write.c:1146 [inline] __do_sys_pwritev2 fs/read_write.c:1204 [inline] __se_sys_pwritev2+0x196/0x2b0 fs/read_write.c:1195 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f1281f85d29 RSP: 002b:00007f12819fe038 EFLAGS: 00000246 ORIG_RAX: 0000000000000148 RAX: ffffffffffffffda RBX: 00007f1282176080 RCX: 00007f1281f85d29 RDX: 0000000000000001 RSI: 0000000020000240 RDI: 0000000000000005 RBP: 00007f12819fe090 R08: 0000000000000000 R09: 0000000000000003 R10: 0000000000007000 R11: 0000000000000246 R12: 0000000000000002 R13: 0000000000000000 R14: 00007f1282176080 R15: 00007ffcb9e23328 Modules linked in: ---[ end trace 0000000000000000 ]--- RIP: 0010:btrfs_split_ordered_extent+0xd8d/0xe20 fs/btrfs/ordered-data.c:1234 RSP: 0018:ffffc9000d1df2b8 EFLAGS: 00010246 RAX: 0000000000000057 RBX: 000000000006a000 RCX: 9ce21886c4195300 RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000 RBP: 0000000000000091 R08: ffffffff817f0a3c R09: 1ffff92001a3bdf4 R10: dffffc0000000000 R11: fffff52001a3bdf5 R12: 1ffff1100a45f401 R13: ffff8880522fa018 R14: dffffc0000000000 R15: 000000000006a000 FS: 00007f12819fe6c0(0000) GS:ffff88801fc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000557750bd7da8 CR3: 00000000400ea000 CR4: 0000000000352ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 In this case the transaction abort was due to (an injected) memory allocation failure when attempting to allocate a new chunk. Reported-by: syzbot+f60d8337a5c8e8d92a77@syzkaller.appspotmail.com Link: https://lore.kernel.org/linux-btrfs/6777f2dd.050a0220.178762.0045.GAE@google.com/ Fixes: 52b1fdca23ac ("btrfs: handle completed ordered extents in btrfs_split_ordered_extent") Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/ordered-data.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 30eceaf829a7..4aca7475fd82 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -1229,6 +1229,18 @@ struct btrfs_ordered_extent *btrfs_split_ordered_extent( */ if (WARN_ON_ONCE(len >= ordered->num_bytes)) return ERR_PTR(-EINVAL); + /* + * If our ordered extent had an error there's no point in continuing. + * The error may have come from a transaction abort done either by this + * task or some other concurrent task, and the transaction abort path + * iterates over all existing ordered extents and sets the flag + * BTRFS_ORDERED_IOERR on them. + */ + if (unlikely(flags & (1U << BTRFS_ORDERED_IOERR))) { + const int fs_error = BTRFS_FS_ERROR(fs_info); + + return fs_error ? ERR_PTR(fs_error) : ERR_PTR(-EIO); + } /* We cannot split partially completed ordered extents. */ if (ordered->bytes_left) { ASSERT(!(flags & ~BTRFS_ORDERED_TYPE_FLAGS)); From c9c863793395cf0a66c2778a29d72c48c02fbb66 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 20 Jan 2025 09:40:43 +1030 Subject: [PATCH 0163/1443] btrfs: do not output error message if a qgroup has been already cleaned up [BUG] There is a bug report that btrfs outputs the following error message: BTRFS info (device nvme0n1p2): qgroup scan completed (inconsistency flag cleared) BTRFS warning (device nvme0n1p2): failed to cleanup qgroup 0/1179: -2 [CAUSE] The error itself is pretty harmless, and the end user should ignore it. When a subvolume is fully dropped, btrfs will call btrfs_qgroup_cleanup_dropped_subvolume() to delete the qgroup. However if a qgroup rescan happened before a subvolume fully dropped, qgroup for that subvolume will not be re-created, as rescan will only create new qgroup if there is a BTRFS_ROOT_REF_KEY found. But before we drop a subvolume, the subvolume is unlinked thus there is no BTRFS_ROOT_REF_KEY. In that case, btrfs_remove_qgroup() will fail with -ENOENT and trigger the above error message. [FIX] Just ignore -ENOENT error from btrfs_remove_qgroup() inside btrfs_qgroup_cleanup_dropped_subvolume(). Reported-by: John Shand Link: https://bugzilla.suse.com/show_bug.cgi?id=1236056 Fixes: 839d6ea4f86d ("btrfs: automatically remove the subvolume qgroup") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index b90fabe302e6..aaf16019d829 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1897,8 +1897,11 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su /* * It's squota and the subvolume still has numbers needed for future * accounting, in this case we can not delete it. Just skip it. + * + * Or the qgroup is already removed by a qgroup rescan. For both cases we're + * safe to ignore them. */ - if (ret == -EBUSY) + if (ret == -EBUSY || ret == -ENOENT) ret = 0; return ret; } From e2f0943cf37305dbdeaf9846e3c941451bcdef63 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 20 Jan 2025 17:26:10 +0000 Subject: [PATCH 0164/1443] btrfs: fix use-after-free when attempting to join an aborted transaction When we are trying to join the current transaction and if it's aborted, we read its 'aborted' field after unlocking fs_info->trans_lock and without holding any extra reference count on it. This means that a concurrent task that is aborting the transaction may free the transaction before we read its 'aborted' field, leading to a use-after-free. Fix this by reading the 'aborted' field while holding fs_info->trans_lock since any freeing task must first acquire that lock and set fs_info->running_transaction to NULL before freeing the transaction. This was reported by syzbot and Dmitry with the following stack traces from KASAN: ================================================================== BUG: KASAN: slab-use-after-free in join_transaction+0xd9b/0xda0 fs/btrfs/transaction.c:278 Read of size 4 at addr ffff888011839024 by task kworker/u4:9/1128 CPU: 0 UID: 0 PID: 1128 Comm: kworker/u4:9 Not tainted 6.13.0-rc7-syzkaller-00019-gc45323b7560e #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 Workqueue: events_unbound btrfs_async_reclaim_data_space Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0x169/0x550 mm/kasan/report.c:489 kasan_report+0x143/0x180 mm/kasan/report.c:602 join_transaction+0xd9b/0xda0 fs/btrfs/transaction.c:278 start_transaction+0xaf8/0x1670 fs/btrfs/transaction.c:697 flush_space+0x448/0xcf0 fs/btrfs/space-info.c:803 btrfs_async_reclaim_data_space+0x159/0x510 fs/btrfs/space-info.c:1321 process_one_work kernel/workqueue.c:3236 [inline] process_scheduled_works+0xa66/0x1840 kernel/workqueue.c:3317 worker_thread+0x870/0xd30 kernel/workqueue.c:3398 kthread+0x2f0/0x390 kernel/kthread.c:389 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Allocated by task 5315: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 poison_kmalloc_redzone mm/kasan/common.c:377 [inline] __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:394 kasan_kmalloc include/linux/kasan.h:260 [inline] __kmalloc_cache_noprof+0x243/0x390 mm/slub.c:4329 kmalloc_noprof include/linux/slab.h:901 [inline] join_transaction+0x144/0xda0 fs/btrfs/transaction.c:308 start_transaction+0xaf8/0x1670 fs/btrfs/transaction.c:697 btrfs_create_common+0x1b2/0x2e0 fs/btrfs/inode.c:6572 lookup_open fs/namei.c:3649 [inline] open_last_lookups fs/namei.c:3748 [inline] path_openat+0x1c03/0x3590 fs/namei.c:3984 do_filp_open+0x27f/0x4e0 fs/namei.c:4014 do_sys_openat2+0x13e/0x1d0 fs/open.c:1402 do_sys_open fs/open.c:1417 [inline] __do_sys_creat fs/open.c:1495 [inline] __se_sys_creat fs/open.c:1489 [inline] __x64_sys_creat+0x123/0x170 fs/open.c:1489 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task 5336: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:582 poison_slab_object mm/kasan/common.c:247 [inline] __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264 kasan_slab_free include/linux/kasan.h:233 [inline] slab_free_hook mm/slub.c:2353 [inline] slab_free mm/slub.c:4613 [inline] kfree+0x196/0x430 mm/slub.c:4761 cleanup_transaction fs/btrfs/transaction.c:2063 [inline] btrfs_commit_transaction+0x2c97/0x3720 fs/btrfs/transaction.c:2598 insert_balance_item+0x1284/0x20b0 fs/btrfs/volumes.c:3757 btrfs_balance+0x992/0x10c0 fs/btrfs/volumes.c:4633 btrfs_ioctl_balance+0x493/0x7c0 fs/btrfs/ioctl.c:3670 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:906 [inline] __se_sys_ioctl+0xf5/0x170 fs/ioctl.c:892 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f The buggy address belongs to the object at ffff888011839000 which belongs to the cache kmalloc-2k of size 2048 The buggy address is located 36 bytes inside of freed 2048-byte region [ffff888011839000, ffff888011839800) The buggy address belongs to the physical page: page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x11838 head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0 flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff) page_type: f5(slab) raw: 00fff00000000040 ffff88801ac42000 ffffea0000493400 dead000000000002 raw: 0000000000000000 0000000000080008 00000001f5000000 0000000000000000 head: 00fff00000000040 ffff88801ac42000 ffffea0000493400 dead000000000002 head: 0000000000000000 0000000000080008 00000001f5000000 0000000000000000 head: 00fff00000000003 ffffea0000460e01 ffffffffffffffff 0000000000000000 head: 0000000000000008 0000000000000000 00000000ffffffff 0000000000000000 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 57, tgid 57 (kworker/0:2), ts 67248182943, free_ts 67229742023 set_page_owner include/linux/page_owner.h:32 [inline] post_alloc_hook+0x1f3/0x230 mm/page_alloc.c:1558 prep_new_page mm/page_alloc.c:1566 [inline] get_page_from_freelist+0x365c/0x37a0 mm/page_alloc.c:3476 __alloc_pages_noprof+0x292/0x710 mm/page_alloc.c:4753 alloc_pages_mpol_noprof+0x3e1/0x780 mm/mempolicy.c:2269 alloc_slab_page+0x6a/0x110 mm/slub.c:2423 allocate_slab+0x5a/0x2b0 mm/slub.c:2589 new_slab mm/slub.c:2642 [inline] ___slab_alloc+0xc27/0x14a0 mm/slub.c:3830 __slab_alloc+0x58/0xa0 mm/slub.c:3920 __slab_alloc_node mm/slub.c:3995 [inline] slab_alloc_node mm/slub.c:4156 [inline] __do_kmalloc_node mm/slub.c:4297 [inline] __kmalloc_node_track_caller_noprof+0x2e9/0x4c0 mm/slub.c:4317 kmalloc_reserve+0x111/0x2a0 net/core/skbuff.c:609 __alloc_skb+0x1f3/0x440 net/core/skbuff.c:678 alloc_skb include/linux/skbuff.h:1323 [inline] alloc_skb_with_frags+0xc3/0x820 net/core/skbuff.c:6612 sock_alloc_send_pskb+0x91a/0xa60 net/core/sock.c:2884 sock_alloc_send_skb include/net/sock.h:1803 [inline] mld_newpack+0x1c3/0xaf0 net/ipv6/mcast.c:1747 add_grhead net/ipv6/mcast.c:1850 [inline] add_grec+0x1492/0x19a0 net/ipv6/mcast.c:1988 mld_send_cr net/ipv6/mcast.c:2114 [inline] mld_ifc_work+0x691/0xd90 net/ipv6/mcast.c:2651 page last free pid 5300 tgid 5300 stack trace: reset_page_owner include/linux/page_owner.h:25 [inline] free_pages_prepare mm/page_alloc.c:1127 [inline] free_unref_page+0xd3f/0x1010 mm/page_alloc.c:2659 __slab_free+0x2c2/0x380 mm/slub.c:4524 qlink_free mm/kasan/quarantine.c:163 [inline] qlist_free_all+0x9a/0x140 mm/kasan/quarantine.c:179 kasan_quarantine_reduce+0x14f/0x170 mm/kasan/quarantine.c:286 __kasan_slab_alloc+0x23/0x80 mm/kasan/common.c:329 kasan_slab_alloc include/linux/kasan.h:250 [inline] slab_post_alloc_hook mm/slub.c:4119 [inline] slab_alloc_node mm/slub.c:4168 [inline] __do_kmalloc_node mm/slub.c:4297 [inline] __kmalloc_noprof+0x236/0x4c0 mm/slub.c:4310 kmalloc_noprof include/linux/slab.h:905 [inline] kzalloc_noprof include/linux/slab.h:1037 [inline] fib_create_info+0xc14/0x25b0 net/ipv4/fib_semantics.c:1435 fib_table_insert+0x1f6/0x1f20 net/ipv4/fib_trie.c:1231 fib_magic+0x3d8/0x620 net/ipv4/fib_frontend.c:1112 fib_add_ifaddr+0x40c/0x5e0 net/ipv4/fib_frontend.c:1156 fib_netdev_event+0x375/0x490 net/ipv4/fib_frontend.c:1494 notifier_call_chain+0x1a5/0x3f0 kernel/notifier.c:85 __dev_notify_flags+0x207/0x400 dev_change_flags+0xf0/0x1a0 net/core/dev.c:9045 do_setlink+0xc90/0x4210 net/core/rtnetlink.c:3109 rtnl_changelink net/core/rtnetlink.c:3723 [inline] __rtnl_newlink net/core/rtnetlink.c:3875 [inline] rtnl_newlink+0x1bb6/0x2210 net/core/rtnetlink.c:4012 Memory state around the buggy address: ffff888011838f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff888011838f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc >ffff888011839000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff888011839080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff888011839100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== Reported-by: syzbot+45212e9d87a98c3f5b42@syzkaller.appspotmail.com Link: https://lore.kernel.org/linux-btrfs/678e7da5.050a0220.303755.007c.GAE@google.com/ Reported-by: Dmitry Vyukov Link: https://lore.kernel.org/linux-btrfs/CACT4Y+ZFBdo7pT8L2AzM=vegZwjp-wNkVJZQf0Ta3vZqtExaSw@mail.gmail.com/ Fixes: 871383be592b ("btrfs: add missing unlocks to transaction abort paths") Reviewed-by: Johannes Thumshirn Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/transaction.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 15312013f2a3..aca83a98b75a 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -274,8 +274,10 @@ static noinline int join_transaction(struct btrfs_fs_info *fs_info, cur_trans = fs_info->running_transaction; if (cur_trans) { if (TRANS_ABORTED(cur_trans)) { + const int abort_error = cur_trans->aborted; + spin_unlock(&fs_info->trans_lock); - return cur_trans->aborted; + return abort_error; } if (btrfs_blocked_trans_types[cur_trans->state] & type) { spin_unlock(&fs_info->trans_lock); From fdef89ce6fada462aef9cb90a140c93c8c209f0f Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Tue, 21 Jan 2025 12:24:39 +0000 Subject: [PATCH 0165/1443] btrfs: avoid starting new transaction when cleaning qgroup during subvolume drop At btrfs_qgroup_cleanup_dropped_subvolume() all we want to commit the current transaction in order to have all the qgroup rfer/excl numbers up to date. However we are using btrfs_start_transaction(), which joins the current transaction if there is one that is not yet committing, but also starts a new one if there is none or if the current one is already committing (its state is >= TRANS_STATE_COMMIT_START). This later case results in unnecessary IO, wasting time and a pointless rotation of the backup roots in the super block. So instead of using btrfs_start_transaction() followed by a btrfs_commit_transaction(), use btrfs_commit_current_transaction() which achieves our purpose and avoids starting and committing new transactions. Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index aaf16019d829..f9d3766c809b 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1880,11 +1880,7 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su * Commit current transaction to make sure all the rfer/excl numbers * get updated. */ - trans = btrfs_start_transaction(fs_info->quota_root, 0); - if (IS_ERR(trans)) - return PTR_ERR(trans); - - ret = btrfs_commit_transaction(trans); + ret = btrfs_commit_current_transaction(fs_info->quota_root); if (ret < 0) return ret; From d3fedff828bb7e4a422c42caeafd5d974e24ee43 Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Wed, 15 Jan 2025 14:20:29 -0800 Subject: [PATCH 0166/1443] drm/xe/oa: Set stream->pollin in xe_oa_buffer_check_unlocked We rely on stream->pollin to decide whether or not to block during poll/read calls. However, currently there are blocking read code paths which don't even set stream->pollin. The best place to consistently set stream->pollin for all code paths is therefore to set it in xe_oa_buffer_check_unlocked. Fixes: e936f885f1e9 ("drm/xe/oa/uapi: Expose OA stream fd") Signed-off-by: Ashutosh Dixit Acked-by: Rodrigo Vivi Reviewed-by: Jonathan Cavitt Reviewed-by: Umesh Nerlige Ramappa Link: https://patchwork.freedesktop.org/patch/msgid/20250115222029.3002103-1-ashutosh.dixit@intel.com --- drivers/gpu/drm/xe/xe_oa.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 6a08e6c92835..fa873f3d0a9d 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -237,7 +237,6 @@ static bool xe_oa_buffer_check_unlocked(struct xe_oa_stream *stream) u32 tail, hw_tail, partial_report_size, available; int report_size = stream->oa_buffer.format->size; unsigned long flags; - bool pollin; spin_lock_irqsave(&stream->oa_buffer.ptr_lock, flags); @@ -282,11 +281,11 @@ static bool xe_oa_buffer_check_unlocked(struct xe_oa_stream *stream) stream->oa_buffer.tail = tail; available = xe_oa_circ_diff(stream, stream->oa_buffer.tail, stream->oa_buffer.head); - pollin = available >= stream->wait_num_reports * report_size; + stream->pollin = available >= stream->wait_num_reports * report_size; spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags); - return pollin; + return stream->pollin; } static enum hrtimer_restart xe_oa_poll_check_timer_cb(struct hrtimer *hrtimer) @@ -294,10 +293,8 @@ static enum hrtimer_restart xe_oa_poll_check_timer_cb(struct hrtimer *hrtimer) struct xe_oa_stream *stream = container_of(hrtimer, typeof(*stream), poll_check_timer); - if (xe_oa_buffer_check_unlocked(stream)) { - stream->pollin = true; + if (xe_oa_buffer_check_unlocked(stream)) wake_up(&stream->poll_wq); - } hrtimer_forward_now(hrtimer, ns_to_ktime(stream->poll_period_ns)); From 2654d4711375ec32ffb8c83644e3af716d009b63 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 14 Jan 2025 15:58:39 +0100 Subject: [PATCH 0167/1443] drm/omap/dss: Use of_property_present() to test existence of DT property of_property_read_bool() should be used only on boolean properties. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20250114145840.505459-1-krzysztof.kozlowski@linaro.org --- drivers/gpu/drm/omapdrm/dss/dss.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c index 7b2df3185de4..319f0a1d23a7 100644 --- a/drivers/gpu/drm/omapdrm/dss/dss.c +++ b/drivers/gpu/drm/omapdrm/dss/dss.c @@ -1236,7 +1236,7 @@ static int dss_video_pll_probe(struct dss_device *dss) if (!np) return 0; - if (of_property_read_bool(np, "syscon-pll-ctrl")) { + if (of_property_present(np, "syscon-pll-ctrl")) { dss->syscon_pll_ctrl = syscon_regmap_lookup_by_phandle(np, "syscon-pll-ctrl"); if (IS_ERR(dss->syscon_pll_ctrl)) { From 07efb761b672753a5ce911e41d2be7c171b4f18f Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 14 Jan 2025 15:58:40 +0100 Subject: [PATCH 0168/1443] drm/omap/dss: Use syscon_regmap_lookup_by_phandle_args Use syscon_regmap_lookup_by_phandle_args() which is a wrapper over syscon_regmap_lookup_by_phandle() combined with getting the syscon argument. Except simpler code this annotates within one line that given phandle has arguments, so grepping for code would be easier. There is also no real benefit in printing errors on missing syscon argument, because this is done just too late: runtime check on static/build-time data. Dtschema and Devicetree bindings offer the static/build-time check for this already. Reviewed-by: Tomi Valkeinen Signed-off-by: Krzysztof Kozlowski Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20250114145840.505459-2-krzysztof.kozlowski@linaro.org --- drivers/gpu/drm/omapdrm/dss/dss.c | 12 +++--------- 1 file changed, 3 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c index 319f0a1d23a7..692df747e2ae 100644 --- a/drivers/gpu/drm/omapdrm/dss/dss.c +++ b/drivers/gpu/drm/omapdrm/dss/dss.c @@ -1237,20 +1237,14 @@ static int dss_video_pll_probe(struct dss_device *dss) return 0; if (of_property_present(np, "syscon-pll-ctrl")) { - dss->syscon_pll_ctrl = syscon_regmap_lookup_by_phandle(np, - "syscon-pll-ctrl"); + dss->syscon_pll_ctrl = + syscon_regmap_lookup_by_phandle_args(np, "syscon-pll-ctrl", + 1, &dss->syscon_pll_ctrl_offset); if (IS_ERR(dss->syscon_pll_ctrl)) { dev_err(&pdev->dev, "failed to get syscon-pll-ctrl regmap\n"); return PTR_ERR(dss->syscon_pll_ctrl); } - - if (of_property_read_u32_index(np, "syscon-pll-ctrl", 1, - &dss->syscon_pll_ctrl_offset)) { - dev_err(&pdev->dev, - "failed to get syscon-pll-ctrl offset\n"); - return -EINVAL; - } } pll_regulator = devm_regulator_get(&pdev->dev, "vdda_video"); From c905d149c506a498f8f0819c77e159d3b9e06a64 Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Wed, 18 Dec 2024 02:01:24 +0000 Subject: [PATCH 0169/1443] drm/omap: Remove hdmi5_core_handle_irqs() hdmi5_core_handle_irqs() is a function that was copied from omapdss by commit f76ee892a99e ("omapfb: copy omapdss & displays for omapfb") but it wasn't used in the original anyway. Remove it. Signed-off-by: Dr. David Alan Gilbert Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20241218020124.278925-1-linux@treblig.org --- drivers/gpu/drm/omapdrm/dss/hdmi5_core.c | 17 ----------------- drivers/gpu/drm/omapdrm/dss/hdmi5_core.h | 1 - 2 files changed, 18 deletions(-) diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c index 21564c38234f..12ef47cd232b 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c @@ -545,23 +545,6 @@ static void hdmi_core_enable_interrupts(struct hdmi_core_data *core) REG_FLD_MOD(core->base, HDMI_CORE_IH_MUTE, 0x0, 1, 0); } -int hdmi5_core_handle_irqs(struct hdmi_core_data *core) -{ - void __iomem *base = core->base; - - REG_FLD_MOD(base, HDMI_CORE_IH_FC_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_FC_STAT1, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_FC_STAT2, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_AS_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_PHY_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_I2CM_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_CEC_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_VP_STAT0, 0xff, 7, 0); - REG_FLD_MOD(base, HDMI_CORE_IH_I2CMPHY_STAT0, 0xff, 7, 0); - - return 0; -} - void hdmi5_configure(struct hdmi_core_data *core, struct hdmi_wp_data *wp, struct hdmi_config *cfg) { diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h index 070cbf5fb57d..b8ed21156e8c 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.h @@ -286,7 +286,6 @@ int hdmi5_core_ddc_read(void *data, u8 *buf, unsigned int block, size_t len); void hdmi5_core_ddc_uninit(struct hdmi_core_data *core); void hdmi5_core_dump(struct hdmi_core_data *core, struct seq_file *s); -int hdmi5_core_handle_irqs(struct hdmi_core_data *core); void hdmi5_configure(struct hdmi_core_data *core, struct hdmi_wp_data *wp, struct hdmi_config *cfg); int hdmi5_core_init(struct platform_device *pdev, struct hdmi_core_data *core); From 798047e63ac970f105c49c22e6d44df901c486b2 Mon Sep 17 00:00:00 2001 From: Andrew Kreimer Date: Thu, 12 Sep 2024 15:57:24 +0300 Subject: [PATCH 0170/1443] drm/tidss: Fix typos Fix typos in comments. Reported-by: Matthew Wilcox Signed-off-by: Andrew Kreimer Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20240912125735.45114-1-algonell@gmail.com --- drivers/gpu/drm/tidss/tidss_dispc_regs.h | 2 +- drivers/gpu/drm/tidss/tidss_plane.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/tidss/tidss_dispc_regs.h b/drivers/gpu/drm/tidss/tidss_dispc_regs.h index 13feedfe5d6d..e88148e44937 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc_regs.h +++ b/drivers/gpu/drm/tidss/tidss_dispc_regs.h @@ -49,7 +49,7 @@ enum dispc_common_regs { /* * dispc_common_regmap should be defined as const u16 * and pointing * to a valid dss common register map for the platform, before the - * macros bellow can be used. + * macros below can be used. */ #define REG(r) (dispc_common_regmap[r ## _OFF]) diff --git a/drivers/gpu/drm/tidss/tidss_plane.c b/drivers/gpu/drm/tidss/tidss_plane.c index 116de124bddb..719412e6c346 100644 --- a/drivers/gpu/drm/tidss/tidss_plane.c +++ b/drivers/gpu/drm/tidss/tidss_plane.c @@ -67,7 +67,7 @@ static int tidss_plane_atomic_check(struct drm_plane *plane, /* * The HW is only able to start drawing at subpixel boundary - * (the two first checks bellow). At the end of a row the HW + * (the two first checks below). At the end of a row the HW * can only jump integer number of subpixels forward to the * beginning of the next row. So we can only show picture with * integer subpixel width (the third check). However, after From 5e0e02f0d7e52cfc8b1adfc778dd02181d8b47b4 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 15 Jan 2025 09:05:15 -0700 Subject: [PATCH 0171/1443] futex: Pass in task to futex_queue() futex_queue() -> __futex_queue() uses 'current' as the task to store in the struct futex_q->task field. This is fine for synchronous usage of the futex infrastructure, but it's not always correct when used by io_uring where the task doing the initial futex_queue() might not be available later on. This doesn't lead to any issues currently, as the io_uring side doesn't support PI futexes, but it does leave a potentially dangling pointer which is never a good idea. Have futex_queue() take a task_struct argument, and have the regular callers pass in 'current' for that. Meanwhile io_uring can just pass in NULL, as the task should never be used off that path. In theory req->tctx->task could be used here, but there's no point populating it with a task field that will never be used anyway. Reported-by: Jann Horn Signed-off-by: Jens Axboe Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/22484a23-542c-4003-b721-400688a0d055@kernel.dk --- io_uring/futex.c | 2 +- kernel/futex/core.c | 5 +++-- kernel/futex/futex.h | 11 ++++++++--- kernel/futex/pi.c | 2 +- kernel/futex/waitwake.c | 4 ++-- 5 files changed, 15 insertions(+), 9 deletions(-) diff --git a/io_uring/futex.c b/io_uring/futex.c index 30139cc150f2..e5cc208810ad 100644 --- a/io_uring/futex.c +++ b/io_uring/futex.c @@ -338,7 +338,7 @@ int io_futex_wait(struct io_kiocb *req, unsigned int issue_flags) hlist_add_head(&req->hash_node, &ctx->futex_list); io_ring_submit_unlock(ctx, issue_flags); - futex_queue(&ifd->q, hb); + futex_queue(&ifd->q, hb, NULL); return IOU_ISSUE_SKIP_COMPLETE; } diff --git a/kernel/futex/core.c b/kernel/futex/core.c index ebdd76b4ecbb..3db8567f5a44 100644 --- a/kernel/futex/core.c +++ b/kernel/futex/core.c @@ -532,7 +532,8 @@ void futex_q_unlock(struct futex_hash_bucket *hb) futex_hb_waiters_dec(hb); } -void __futex_queue(struct futex_q *q, struct futex_hash_bucket *hb) +void __futex_queue(struct futex_q *q, struct futex_hash_bucket *hb, + struct task_struct *task) { int prio; @@ -548,7 +549,7 @@ void __futex_queue(struct futex_q *q, struct futex_hash_bucket *hb) plist_node_init(&q->list, prio); plist_add(&q->list, &hb->chain); - q->task = current; + q->task = task; } /** diff --git a/kernel/futex/futex.h b/kernel/futex/futex.h index 99b32e728c4a..6b2f4c7eb720 100644 --- a/kernel/futex/futex.h +++ b/kernel/futex/futex.h @@ -285,13 +285,15 @@ static inline int futex_get_value_locked(u32 *dest, u32 __user *from) } extern void __futex_unqueue(struct futex_q *q); -extern void __futex_queue(struct futex_q *q, struct futex_hash_bucket *hb); +extern void __futex_queue(struct futex_q *q, struct futex_hash_bucket *hb, + struct task_struct *task); extern int futex_unqueue(struct futex_q *q); /** * futex_queue() - Enqueue the futex_q on the futex_hash_bucket * @q: The futex_q to enqueue * @hb: The destination hash bucket + * @task: Task queueing this futex * * The hb->lock must be held by the caller, and is released here. A call to * futex_queue() is typically paired with exactly one call to futex_unqueue(). The @@ -299,11 +301,14 @@ extern int futex_unqueue(struct futex_q *q); * or nothing if the unqueue is done as part of the wake process and the unqueue * state is implicit in the state of woken task (see futex_wait_requeue_pi() for * an example). + * + * Note that @task may be NULL, for async usage of futexes. */ -static inline void futex_queue(struct futex_q *q, struct futex_hash_bucket *hb) +static inline void futex_queue(struct futex_q *q, struct futex_hash_bucket *hb, + struct task_struct *task) __releases(&hb->lock) { - __futex_queue(q, hb); + __futex_queue(q, hb, task); spin_unlock(&hb->lock); } diff --git a/kernel/futex/pi.c b/kernel/futex/pi.c index daea650b16f5..7a941845f7ee 100644 --- a/kernel/futex/pi.c +++ b/kernel/futex/pi.c @@ -982,7 +982,7 @@ int futex_lock_pi(u32 __user *uaddr, unsigned int flags, ktime_t *time, int tryl /* * Only actually queue now that the atomic ops are done: */ - __futex_queue(&q, hb); + __futex_queue(&q, hb, current); if (trylock) { ret = rt_mutex_futex_trylock(&q.pi_state->pi_mutex); diff --git a/kernel/futex/waitwake.c b/kernel/futex/waitwake.c index 3a10375d9521..a9056acb75ee 100644 --- a/kernel/futex/waitwake.c +++ b/kernel/futex/waitwake.c @@ -350,7 +350,7 @@ void futex_wait_queue(struct futex_hash_bucket *hb, struct futex_q *q, * access to the hash list and forcing another memory barrier. */ set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); - futex_queue(q, hb); + futex_queue(q, hb, current); /* Arm the timer */ if (timeout) @@ -461,7 +461,7 @@ int futex_wait_multiple_setup(struct futex_vector *vs, int count, int *woken) * next futex. Queue each futex at this moment so hb can * be unlocked. */ - futex_queue(q, hb); + futex_queue(q, hb, current); continue; } From 915175b49f65d9edeb81659e82cbb27b621dbc17 Mon Sep 17 00:00:00 2001 From: Jinliang Zheng Date: Wed, 15 Jan 2025 20:35:25 +0800 Subject: [PATCH 0172/1443] xfs: fix the entry condition of exact EOF block allocation optimization When we call create(), lseek() and write() sequentially, offset != 0 cannot be used as a judgment condition for whether the file already has extents. Furthermore, when xfs_bmap_adjacent() has not given a better blkno, it is not necessary to use exact EOF block allocation. Suggested-by: Dave Chinner Signed-off-by: Jinliang Zheng Reviewed-by: Dave Chinner Signed-off-by: Carlos Maiolino --- fs/xfs/libxfs/xfs_bmap.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 40ad22fb808b..0ef19f1469ec 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -3563,12 +3563,12 @@ xfs_bmap_btalloc_at_eof( int error; /* - * If there are already extents in the file, try an exact EOF block - * allocation to extend the file as a contiguous extent. If that fails, - * or it's the first allocation in a file, just try for a stripe aligned - * allocation. + * If there are already extents in the file, and xfs_bmap_adjacent() has + * given a better blkno, try an exact EOF block allocation to extend the + * file as a contiguous extent. If that fails, or it's the first + * allocation in a file, just try for a stripe aligned allocation. */ - if (ap->offset) { + if (ap->eof) { xfs_extlen_t nextminlen = 0; /* @@ -3736,7 +3736,8 @@ xfs_bmap_btalloc_best_length( int error; ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino); - xfs_bmap_adjacent(ap); + if (!xfs_bmap_adjacent(ap)) + ap->eof = false; /* * Search for an allocation group with a single extent large enough for From 89841b23809f5fb12cbead142204064739fef25a Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 16 Jan 2025 07:03:35 +0100 Subject: [PATCH 0173/1443] xfs: remove an out of data comment in _xfs_buf_alloc There hasn't been anything like an io_length for a long time. Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_buf.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 7fbdd4b30676..f1252ed8bd0a 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -232,11 +232,6 @@ _xfs_buf_alloc( bp->b_mount = target->bt_mount; bp->b_flags = flags; - /* - * Set length and io_length to the same value initially. - * I/O routines should use io_length, which will be the same in - * most cases but may be reset (e.g. XFS recovery). - */ error = xfs_buf_get_maps(bp, nmaps); if (error) { kmem_cache_free(xfs_buf_cache, bp); From f5f0ed89f13e3e5246404a322ee85169a226bfb5 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 22 Jan 2025 06:43:21 +0100 Subject: [PATCH 0174/1443] xfs: don't call remap_verify_area with sb write protection held The XFS_IOC_EXCHANGE_RANGE ioctl with the XFS_EXCHANGE_RANGE_TO_EOF flag operates on a range bounded by the end of the file. This means the actual amount of blocks exchanged is derived from the inode size, which is only stable with the IOLOCK (i_rwsem) held. Do that, it currently calls remap_verify_area from inside the sb write protection which nests outside the IOLOCK. But this makes fsnotify_file_area_perm which is called from remap_verify_area unhappy when the kernel is built with lockdep and the recently added CONFIG_FANOTIFY_ACCESS_PERMISSIONS option. Fix this by always calling remap_verify_area before taking the write protection, and passing a 0 size to remap_verify_area similar to the FICLONE/FICLONERANGE ioctls when they are asked to clone until the file end. (Note: the size argument gets passed to fsnotify_file_area_perm, but then isn't actually used there). Fixes: 9a64d9b3109d ("xfs: introduce new file range exchange ioctl") Cc: # v6.10 Signed-off-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_exchrange.c | 71 ++++++++++++++++-------------------------- 1 file changed, 27 insertions(+), 44 deletions(-) diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c index f340a2015c4c..0b41bdfecdfb 100644 --- a/fs/xfs/xfs_exchrange.c +++ b/fs/xfs/xfs_exchrange.c @@ -329,22 +329,6 @@ xfs_exchrange_mappings( * successfully but before locks are dropped. */ -/* Verify that we have security clearance to perform this operation. */ -static int -xfs_exchange_range_verify_area( - struct xfs_exchrange *fxr) -{ - int ret; - - ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length, - true); - if (ret) - return ret; - - return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length, - true); -} - /* * Performs necessary checks before doing a range exchange, having stabilized * mutable inode attributes via i_rwsem. @@ -355,11 +339,13 @@ xfs_exchange_range_checks( unsigned int alloc_unit) { struct inode *inode1 = file_inode(fxr->file1); + loff_t size1 = i_size_read(inode1); struct inode *inode2 = file_inode(fxr->file2); + loff_t size2 = i_size_read(inode2); uint64_t allocmask = alloc_unit - 1; int64_t test_len; uint64_t blen; - loff_t size1, size2, tmp; + loff_t tmp; int error; /* Don't touch certain kinds of inodes */ @@ -368,24 +354,25 @@ xfs_exchange_range_checks( if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2)) return -ETXTBSY; - size1 = i_size_read(inode1); - size2 = i_size_read(inode2); - /* Ranges cannot start after EOF. */ if (fxr->file1_offset > size1 || fxr->file2_offset > size2) return -EINVAL; - /* - * If the caller said to exchange to EOF, we set the length of the - * request large enough to cover everything to the end of both files. - */ if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) { + /* + * If the caller said to exchange to EOF, we set the length of + * the request large enough to cover everything to the end of + * both files. + */ fxr->length = max_t(int64_t, size1 - fxr->file1_offset, size2 - fxr->file2_offset); - - error = xfs_exchange_range_verify_area(fxr); - if (error) - return error; + } else { + /* + * Otherwise we require both ranges to end within EOF. + */ + if (fxr->file1_offset + fxr->length > size1 || + fxr->file2_offset + fxr->length > size2) + return -EINVAL; } /* @@ -401,15 +388,6 @@ xfs_exchange_range_checks( check_add_overflow(fxr->file2_offset, fxr->length, &tmp)) return -EINVAL; - /* - * We require both ranges to end within EOF, unless we're exchanging - * to EOF. - */ - if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) && - (fxr->file1_offset + fxr->length > size1 || - fxr->file2_offset + fxr->length > size2)) - return -EINVAL; - /* * Make sure we don't hit any file size limits. If we hit any size * limits such that test_length was adjusted, we abort the whole @@ -747,6 +725,7 @@ xfs_exchange_range( { struct inode *inode1 = file_inode(fxr->file1); struct inode *inode2 = file_inode(fxr->file2); + loff_t check_len = fxr->length; int ret; BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS & @@ -779,14 +758,18 @@ xfs_exchange_range( return -EBADF; /* - * If we're not exchanging to EOF, we can check the areas before - * stabilizing both files' i_size. + * If we're exchanging to EOF we can't calculate the length until taking + * the iolock. Pass a 0 length to remap_verify_area similar to the + * FICLONE and FICLONERANGE ioctls that support cloning to EOF as well. */ - if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) { - ret = xfs_exchange_range_verify_area(fxr); - if (ret) - return ret; - } + if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) + check_len = 0; + ret = remap_verify_area(fxr->file1, fxr->file1_offset, check_len, true); + if (ret) + return ret; + ret = remap_verify_area(fxr->file2, fxr->file2_offset, check_len, true); + if (ret) + return ret; /* Update cmtime if the fd/inode don't forbid it. */ if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1)) From 3ab334814dc7dff39075e055e12847d51878916e Mon Sep 17 00:00:00 2001 From: "Eric R. Smith" Date: Thu, 19 Dec 2024 13:49:28 -0400 Subject: [PATCH 0175/1443] drm: add modifiers for MediaTek tiled formats MediaTek (MTK) uses some unique tiled memory formats for video decoding. Add these to the uapi drm_fourcc.h so that we can use them in Mesa, GStreamer, and other tools/libraries. v2: - Classify the modifier bits into categories and provide room for expansion (Daniel S.) Signed-off-by: Eric R. Smith Reviewed-by: Daniel Stone Link: https://patchwork.freedesktop.org/patch/msgid/20241219174929.126205-1-eric.smith@collabora.com Signed-off-by: Daniel Stone --- include/uapi/drm/drm_fourcc.h | 41 +++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index 70f3b00b0681..e41a3cec6a9e 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -421,6 +421,7 @@ extern "C" { #define DRM_FORMAT_MOD_VENDOR_ARM 0x08 #define DRM_FORMAT_MOD_VENDOR_ALLWINNER 0x09 #define DRM_FORMAT_MOD_VENDOR_AMLOGIC 0x0a +#define DRM_FORMAT_MOD_VENDOR_MTK 0x0b /* add more to the end as needed */ @@ -1453,6 +1454,46 @@ drm_fourcc_canonicalize_nvidia_format_mod(__u64 modifier) */ #define AMLOGIC_FBC_OPTION_MEM_SAVING (1ULL << 0) +/* MediaTek modifiers + * Bits Parameter Notes + * ----- ------------------------ --------------------------------------------- + * 7: 0 TILE LAYOUT Values are MTK_FMT_MOD_TILE_* + * 15: 8 COMPRESSION Values are MTK_FMT_MOD_COMPRESS_* + * 23:16 10 BIT LAYOUT Values are MTK_FMT_MOD_10BIT_LAYOUT_* + * + */ + +#define DRM_FORMAT_MOD_MTK(__flags) fourcc_mod_code(MTK, __flags) + +/* + * MediaTek Tiled Modifier + * The lowest 8 bits of the modifier is used to specify the tiling + * layout. Only the 16L_32S tiling is used for now, but we define an + * "untiled" version and leave room for future expansion. + */ +#define MTK_FMT_MOD_TILE_MASK 0xf +#define MTK_FMT_MOD_TILE_NONE 0x0 +#define MTK_FMT_MOD_TILE_16L32S 0x1 + +/* + * Bits 8-15 specify compression options + */ +#define MTK_FMT_MOD_COMPRESS_MASK (0xf << 8) +#define MTK_FMT_MOD_COMPRESS_NONE (0x0 << 8) +#define MTK_FMT_MOD_COMPRESS_V1 (0x1 << 8) + +/* + * Bits 16-23 specify how the bits of 10 bit formats are + * stored out in memory + */ +#define MTK_FMT_MOD_10BIT_LAYOUT_MASK (0xf << 16) +#define MTK_FMT_MOD_10BIT_LAYOUT_PACKED (0x0 << 16) +#define MTK_FMT_MOD_10BIT_LAYOUT_LSBTILED (0x1 << 16) +#define MTK_FMT_MOD_10BIT_LAYOUT_LSBRASTER (0x2 << 16) + +/* alias for the most common tiling format */ +#define DRM_FORMAT_MOD_MTK_16L_32S_TILE DRM_FORMAT_MOD_MTK(MTK_FMT_MOD_TILE_16L32S) + /* * AMD modifiers * From 58f5c8d5ca07a2f9fa93fb073f5b1646ec482ff2 Mon Sep 17 00:00:00 2001 From: Sagi Grimberg Date: Fri, 24 Jan 2025 13:00:33 +0200 Subject: [PATCH 0176/1443] nvmet: fix a memory leak in controller identify Simply free an allocated buffer once we copied its content to the request sgl. kmemleak complaint: unreferenced object 0xffff8cd40c388000 (size 4096): comm "kworker/2:2H", pid 14739, jiffies 4401313113 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace (crc 0): [] kmemleak_alloc+0x4a/0x90 [] __kmalloc_cache_noprof+0x35a/0x420 [] nvmet_execute_identify+0x912/0x9f0 [nvmet] [] nvmet_tcp_try_recv_pdu+0x84c/0xc90 [nvmet_tcp] [] nvmet_tcp_io_work+0x82/0x8b0 [nvmet_tcp] [] process_one_work+0x178/0x3e0 [] worker_thread+0x2ec/0x420 [] kthread+0xf0/0x120 [] ret_from_fork+0x44/0x70 [] ret_from_fork_asm+0x1a/0x30 Fixes: 84909f7decbd ("nvmet: use kzalloc instead of ZERO_PAGE in nvme_execute_identify_ns_nvm()") Signed-off-by: Sagi Grimberg Reviewed-by: Nilay Shroff Signed-off-by: Keith Busch --- drivers/nvme/target/admin-cmd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/nvme/target/admin-cmd.c b/drivers/nvme/target/admin-cmd.c index 3ddd8e44e148..ec7f70be6daa 100644 --- a/drivers/nvme/target/admin-cmd.c +++ b/drivers/nvme/target/admin-cmd.c @@ -1067,6 +1067,7 @@ static void nvme_execute_identify_ns_nvm(struct nvmet_req *req) goto out; } status = nvmet_copy_to_sgl(req, 0, id, sizeof(*id)); + kfree(id); out: nvmet_req_complete(req, status); } From be8ee18152b0523752f3a44900363838bd1573bb Mon Sep 17 00:00:00 2001 From: Atul Kumar Pant Date: Sat, 18 Jan 2025 14:09:27 +0530 Subject: [PATCH 0177/1443] sched_ext: Fixes typos in comments Fixes some spelling errors in the comments. Signed-off-by: Atul Kumar Pant Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 8857c0709bdd..283d7f1addc5 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -416,7 +416,7 @@ struct sched_ext_ops { /** * @update_idle: Update the idle state of a CPU - * @cpu: CPU to udpate the idle state for + * @cpu: CPU to update the idle state for * @idle: whether entering or exiting the idle state * * This operation is called when @rq's CPU goes or leaves the idle @@ -1214,7 +1214,7 @@ static bool scx_kf_allowed_if_unlocked(void) /** * nldsq_next_task - Iterate to the next task in a non-local DSQ - * @dsq: user dsq being interated + * @dsq: user dsq being iterated * @cur: current position, %NULL to start iteration * @rev: walk backwards * @@ -2078,7 +2078,7 @@ static void set_task_runnable(struct rq *rq, struct task_struct *p) /* * list_add_tail() must be used. scx_ops_bypass() depends on tasks being - * appened to the runnable_list. + * appended to the runnable_list. */ list_add_tail(&p->scx.runnable_node, &rq->scx.runnable_list); } @@ -2480,7 +2480,7 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, /* * A poorly behaving BPF scheduler can live-lock the system by e.g. incessantly * banging on the same DSQ on a large NUMA system to the point where switching - * to the bypass mode can take a long time. Inject artifical delays while the + * to the bypass mode can take a long time. Inject artificial delays while the * bypass mode is switching to guarantee timely completion. */ static void scx_ops_breather(struct rq *rq) @@ -3144,7 +3144,7 @@ static struct task_struct *pick_task_scx(struct rq *rq) * * Unless overridden by ops.core_sched_before(), @p->scx.core_sched_at is used * to implement the default task ordering. The older the timestamp, the higher - * prority the task - the global FIFO ordering matching the default scheduling + * priority the task - the global FIFO ordering matching the default scheduling * behavior. * * When ops.core_sched_before() is enabled, @p->scx.core_sched_at is used to @@ -4590,7 +4590,7 @@ static int scx_cgroup_init(void) cgroup_warned_missing_idle = false; /* - * scx_tg_on/offline() are excluded thorugh scx_cgroup_rwsem. If we walk + * scx_tg_on/offline() are excluded through scx_cgroup_rwsem. If we walk * cgroups and init, all online cgroups are initialized. */ rcu_read_lock(); From 2279563e3a8cac367b267b09c15cf1e39c06c5cc Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Wed, 22 Jan 2025 10:05:25 +0100 Subject: [PATCH 0178/1443] sched_ext: Include task weight in the error state dump Report the task weight when dumping the task state during an error exit. Moreover, adjust the output format to display dsq_vtime, slice, and weight on the same line. This can help identify whether certain tasks were excessively prioritized or de-prioritized due to large niceness gaps. Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 283d7f1addc5..7081c7be5f62 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -5277,9 +5277,10 @@ static void scx_dump_task(struct seq_buf *s, struct scx_dump_ctx *dctx, scx_get_task_state(p), p->scx.flags & ~SCX_TASK_STATE_MASK, p->scx.dsq_flags, ops_state & SCX_OPSS_STATE_MASK, ops_state >> SCX_OPSS_QSEQ_SHIFT); - dump_line(s, " sticky/holding_cpu=%d/%d dsq_id=%s dsq_vtime=%llu slice=%llu", - p->scx.sticky_cpu, p->scx.holding_cpu, dsq_id_buf, - p->scx.dsq_vtime, p->scx.slice); + dump_line(s, " sticky/holding_cpu=%d/%d dsq_id=%s", + p->scx.sticky_cpu, p->scx.holding_cpu, dsq_id_buf); + dump_line(s, " dsq_vtime=%llu slice=%llu weight=%u", + p->scx.dsq_vtime, p->scx.slice, p->scx.weight); dump_line(s, " cpus=%*pb", cpumask_pr_args(p->cpus_ptr)); if (SCX_HAS_OP(dump_task)) { From 74ca334338a4489173d9e50775b13fa20cbd5958 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Thu, 23 Jan 2025 13:46:06 +0100 Subject: [PATCH 0179/1443] selftests/sched_ext: Fix enum resolution All scx enums are now automatically generated from vmlinux.h and they must be initialized using the SCX_ENUM_INIT() macro. Fix the scx selftests to use this macro to properly initialize these values. Fixes: 8da7bf2cee27 ("tools/sched_ext: Receive updates from SCX repo") Reported-by: Ihor Solodrai Closes: https://lore.kernel.org/all/Z2tNK2oFDX1OPp8C@slm.duckdns.org/ Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- .../testing/selftests/sched_ext/create_dsq.c | 10 ++++---- .../selftests/sched_ext/ddsp_bogus_dsq_fail.c | 7 ++++-- .../sched_ext/ddsp_vtimelocal_fail.c | 7 ++++-- .../selftests/sched_ext/dsp_local_on.c | 1 + .../sched_ext/enq_last_no_enq_fails.c | 10 ++++---- .../sched_ext/enq_select_cpu_fails.c | 10 ++++---- tools/testing/selftests/sched_ext/exit.c | 1 + tools/testing/selftests/sched_ext/hotplug.c | 6 +++-- .../selftests/sched_ext/init_enable_count.c | 25 ++++++------------- tools/testing/selftests/sched_ext/maximal.c | 7 ++++-- tools/testing/selftests/sched_ext/minimal.c | 10 ++++---- tools/testing/selftests/sched_ext/prog_run.c | 10 ++++---- .../testing/selftests/sched_ext/reload_loop.c | 9 +++---- .../selftests/sched_ext/select_cpu_dfl.c | 7 ++++-- .../sched_ext/select_cpu_dfl_nodispatch.c | 7 ++++-- .../selftests/sched_ext/select_cpu_dispatch.c | 7 ++++-- .../sched_ext/select_cpu_dispatch_bad_dsq.c | 7 ++++-- .../sched_ext/select_cpu_dispatch_dbl_dsp.c | 7 ++++-- .../selftests/sched_ext/select_cpu_vtime.c | 7 ++++-- 19 files changed, 88 insertions(+), 67 deletions(-) diff --git a/tools/testing/selftests/sched_ext/create_dsq.c b/tools/testing/selftests/sched_ext/create_dsq.c index fa946d9146d4..d67431f57ac6 100644 --- a/tools/testing/selftests/sched_ext/create_dsq.c +++ b/tools/testing/selftests/sched_ext/create_dsq.c @@ -14,11 +14,11 @@ static enum scx_test_status setup(void **ctx) { struct create_dsq *skel; - skel = create_dsq__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = create_dsq__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(create_dsq__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.c b/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.c index e65d22f23f3b..b6d13496b24e 100644 --- a/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.c +++ b/tools/testing/selftests/sched_ext/ddsp_bogus_dsq_fail.c @@ -15,8 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct ddsp_bogus_dsq_fail *skel; - skel = ddsp_bogus_dsq_fail__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = ddsp_bogus_dsq_fail__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(ddsp_bogus_dsq_fail__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.c b/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.c index abafee587cd6..af9ce4ee8baa 100644 --- a/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.c +++ b/tools/testing/selftests/sched_ext/ddsp_vtimelocal_fail.c @@ -14,8 +14,11 @@ static enum scx_test_status setup(void **ctx) { struct ddsp_vtimelocal_fail *skel; - skel = ddsp_vtimelocal_fail__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = ddsp_vtimelocal_fail__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(ddsp_vtimelocal_fail__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/dsp_local_on.c b/tools/testing/selftests/sched_ext/dsp_local_on.c index 0ff27e57fe43..e1f2ce4abfe6 100644 --- a/tools/testing/selftests/sched_ext/dsp_local_on.c +++ b/tools/testing/selftests/sched_ext/dsp_local_on.c @@ -15,6 +15,7 @@ static enum scx_test_status setup(void **ctx) skel = dsp_local_on__open(); SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); skel->rodata->nr_cpus = libbpf_num_possible_cpus(); SCX_FAIL_IF(dsp_local_on__load(skel), "Failed to load skel"); diff --git a/tools/testing/selftests/sched_ext/enq_last_no_enq_fails.c b/tools/testing/selftests/sched_ext/enq_last_no_enq_fails.c index 73e679953e27..d3387ae03679 100644 --- a/tools/testing/selftests/sched_ext/enq_last_no_enq_fails.c +++ b/tools/testing/selftests/sched_ext/enq_last_no_enq_fails.c @@ -15,11 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct enq_last_no_enq_fails *skel; - skel = enq_last_no_enq_fails__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = enq_last_no_enq_fails__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(enq_last_no_enq_fails__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/enq_select_cpu_fails.c b/tools/testing/selftests/sched_ext/enq_select_cpu_fails.c index dd1350e5f002..a80e3a3b3698 100644 --- a/tools/testing/selftests/sched_ext/enq_select_cpu_fails.c +++ b/tools/testing/selftests/sched_ext/enq_select_cpu_fails.c @@ -15,11 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct enq_select_cpu_fails *skel; - skel = enq_select_cpu_fails__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = enq_select_cpu_fails__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(enq_select_cpu_fails__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/exit.c b/tools/testing/selftests/sched_ext/exit.c index 31bcd06e21cd..9451782689de 100644 --- a/tools/testing/selftests/sched_ext/exit.c +++ b/tools/testing/selftests/sched_ext/exit.c @@ -23,6 +23,7 @@ static enum scx_test_status run(void *ctx) char buf[16]; skel = exit__open(); + SCX_ENUM_INIT(skel); skel->rodata->exit_point = tc; exit__load(skel); link = bpf_map__attach_struct_ops(skel->maps.exit_ops); diff --git a/tools/testing/selftests/sched_ext/hotplug.c b/tools/testing/selftests/sched_ext/hotplug.c index 87bf220b1bce..1c9ceb661c43 100644 --- a/tools/testing/selftests/sched_ext/hotplug.c +++ b/tools/testing/selftests/sched_ext/hotplug.c @@ -49,8 +49,10 @@ static enum scx_test_status test_hotplug(bool onlining, bool cbs_defined) SCX_ASSERT(is_cpu_online()); - skel = hotplug__open_and_load(); - SCX_ASSERT(skel); + skel = hotplug__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(hotplug__load(skel), "Failed to load skel"); /* Testing the offline -> online path, so go offline before starting */ if (onlining) diff --git a/tools/testing/selftests/sched_ext/init_enable_count.c b/tools/testing/selftests/sched_ext/init_enable_count.c index 97d45f1e5597..0f3eddc7a17a 100644 --- a/tools/testing/selftests/sched_ext/init_enable_count.c +++ b/tools/testing/selftests/sched_ext/init_enable_count.c @@ -15,22 +15,6 @@ #define SCHED_EXT 7 -static struct init_enable_count * -open_load_prog(bool global) -{ - struct init_enable_count *skel; - - skel = init_enable_count__open(); - SCX_BUG_ON(!skel, "Failed to open skel"); - - if (!global) - skel->struct_ops.init_enable_count_ops->flags |= SCX_OPS_SWITCH_PARTIAL; - - SCX_BUG_ON(init_enable_count__load(skel), "Failed to load skel"); - - return skel; -} - static enum scx_test_status run_test(bool global) { struct init_enable_count *skel; @@ -40,7 +24,14 @@ static enum scx_test_status run_test(bool global) struct sched_param param = {}; pid_t pids[num_pre_forks]; - skel = open_load_prog(global); + skel = init_enable_count__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + + if (!global) + skel->struct_ops.init_enable_count_ops->flags |= SCX_OPS_SWITCH_PARTIAL; + + SCX_FAIL_IF(init_enable_count__load(skel), "Failed to load skel"); /* * Fork a bunch of children before we attach the scheduler so that we diff --git a/tools/testing/selftests/sched_ext/maximal.c b/tools/testing/selftests/sched_ext/maximal.c index f38fc973c380..c6be50a9941d 100644 --- a/tools/testing/selftests/sched_ext/maximal.c +++ b/tools/testing/selftests/sched_ext/maximal.c @@ -14,8 +14,11 @@ static enum scx_test_status setup(void **ctx) { struct maximal *skel; - skel = maximal__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = maximal__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(maximal__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/minimal.c b/tools/testing/selftests/sched_ext/minimal.c index 6c5db8ebbf8a..89f7261757ff 100644 --- a/tools/testing/selftests/sched_ext/minimal.c +++ b/tools/testing/selftests/sched_ext/minimal.c @@ -15,11 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct minimal *skel; - skel = minimal__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = minimal__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(minimal__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/prog_run.c b/tools/testing/selftests/sched_ext/prog_run.c index 3cd57ef8daaa..05974820ca69 100644 --- a/tools/testing/selftests/sched_ext/prog_run.c +++ b/tools/testing/selftests/sched_ext/prog_run.c @@ -15,11 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct prog_run *skel; - skel = prog_run__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = prog_run__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(prog_run__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/reload_loop.c b/tools/testing/selftests/sched_ext/reload_loop.c index 5cfba2d6e056..308211d80436 100644 --- a/tools/testing/selftests/sched_ext/reload_loop.c +++ b/tools/testing/selftests/sched_ext/reload_loop.c @@ -18,11 +18,10 @@ bool force_exit = false; static enum scx_test_status setup(void **ctx) { - skel = maximal__open_and_load(); - if (!skel) { - SCX_ERR("Failed to open and load skel"); - return SCX_TEST_FAIL; - } + skel = maximal__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(maximal__load(skel), "Failed to load skel"); return SCX_TEST_PASS; } diff --git a/tools/testing/selftests/sched_ext/select_cpu_dfl.c b/tools/testing/selftests/sched_ext/select_cpu_dfl.c index a53a40c2d2f0..5b6e045e1109 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dfl.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dfl.c @@ -17,8 +17,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_dfl *skel; - skel = select_cpu_dfl__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_dfl__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_dfl__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.c b/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.c index 1d85bf4bf3a3..9b5d232efb7f 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dfl_nodispatch.c @@ -17,8 +17,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_dfl_nodispatch *skel; - skel = select_cpu_dfl_nodispatch__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_dfl_nodispatch__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_dfl_nodispatch__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch.c index 0309ca8785b3..80283dbc41b7 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch.c @@ -17,8 +17,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_dispatch *skel; - skel = select_cpu_dispatch__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_dispatch__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_dispatch__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.c index 47eb6ed7627d..5e72ebbc90a5 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch_bad_dsq.c @@ -15,8 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_dispatch_bad_dsq *skel; - skel = select_cpu_dispatch_bad_dsq__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_dispatch_bad_dsq__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_dispatch_bad_dsq__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.c b/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.c index 48ff028a3c46..aa85949478bc 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.c +++ b/tools/testing/selftests/sched_ext/select_cpu_dispatch_dbl_dsp.c @@ -15,8 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_dispatch_dbl_dsp *skel; - skel = select_cpu_dispatch_dbl_dsp__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_dispatch_dbl_dsp__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_dispatch_dbl_dsp__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; diff --git a/tools/testing/selftests/sched_ext/select_cpu_vtime.c b/tools/testing/selftests/sched_ext/select_cpu_vtime.c index b4629c2364f5..1e9b5c9bfff1 100644 --- a/tools/testing/selftests/sched_ext/select_cpu_vtime.c +++ b/tools/testing/selftests/sched_ext/select_cpu_vtime.c @@ -15,8 +15,11 @@ static enum scx_test_status setup(void **ctx) { struct select_cpu_vtime *skel; - skel = select_cpu_vtime__open_and_load(); - SCX_FAIL_IF(!skel, "Failed to open and load skel"); + skel = select_cpu_vtime__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(select_cpu_vtime__load(skel), "Failed to load skel"); + *ctx = skel; return SCX_TEST_PASS; From 64a1ba4072b34af1b76bf15fca5c2075b8cc4d64 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Thu, 23 Jan 2025 21:51:38 +0530 Subject: [PATCH 0180/1443] wifi: ath12k: fix handling of 6 GHz rules In the US country code, to avoid including 6 GHz rules in the 5 GHz rules list, the number of 5 GHz rules is set to a default constant value of 4 (REG_US_5G_NUM_REG_RULES). However, if there are more than 4 valid 5 GHz rules, the current logic will bypass the legitimate 6 GHz rules. For example, if there are 5 valid 5 GHz rules and 1 valid 6 GHz rule, the current logic will only consider 4 of the 5 GHz rules, treating the last valid rule as a 6 GHz rule. Consequently, the actual 6 GHz rule is never processed, leading to the eventual disabling of 6 GHz channels. To fix this issue, instead of hardcoding the value to 4, use a helper function to determine the number of 6 GHz rules present in the 5 GHz rules list and ignore only those rules. Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1 Cc: stable@vger.kernel.org Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices") Signed-off-by: Aditya Kumar Singh Link: https://patch.msgid.link/20250123-fix_6ghz_rules_handling-v1-1-d734bfa58ff4@oss.qualcomm.com Signed-off-by: Jeff Johnson --- drivers/net/wireless/ath/ath12k/wmi.c | 61 ++++++++++++++++++++------- drivers/net/wireless/ath/ath12k/wmi.h | 1 - 2 files changed, 45 insertions(+), 17 deletions(-) diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c index dced2aa9ba1a..d953742b67e1 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.c +++ b/drivers/net/wireless/ath/ath12k/wmi.c @@ -4681,6 +4681,22 @@ static struct ath12k_reg_rule return reg_rule_ptr; } +static u8 ath12k_wmi_ignore_num_extra_rules(struct ath12k_wmi_reg_rule_ext_params *rule, + u32 num_reg_rules) +{ + u8 num_invalid_5ghz_rules = 0; + u32 count, start_freq; + + for (count = 0; count < num_reg_rules; count++) { + start_freq = le32_get_bits(rule[count].freq_info, REG_RULE_START_FREQ); + + if (start_freq >= ATH12K_MIN_6G_FREQ) + num_invalid_5ghz_rules++; + } + + return num_invalid_5ghz_rules; +} + static int ath12k_pull_reg_chan_list_ext_update_ev(struct ath12k_base *ab, struct sk_buff *skb, struct ath12k_reg_info *reg_info) @@ -4691,6 +4707,7 @@ static int ath12k_pull_reg_chan_list_ext_update_ev(struct ath12k_base *ab, u32 num_2g_reg_rules, num_5g_reg_rules; u32 num_6g_reg_rules_ap[WMI_REG_CURRENT_MAX_AP_TYPE]; u32 num_6g_reg_rules_cl[WMI_REG_CURRENT_MAX_AP_TYPE][WMI_REG_MAX_CLIENT_TYPE]; + u8 num_invalid_5ghz_ext_rules; u32 total_reg_rules = 0; int ret, i, j; @@ -4784,20 +4801,6 @@ static int ath12k_pull_reg_chan_list_ext_update_ev(struct ath12k_base *ab, memcpy(reg_info->alpha2, &ev->alpha2, REG_ALPHA2_LEN); - /* FIXME: Currently FW includes 6G reg rule also in 5G rule - * list for country US. - * Having same 6G reg rule in 5G and 6G rules list causes - * intersect check to be true, and same rules will be shown - * multiple times in iw cmd. So added hack below to avoid - * parsing 6G rule from 5G reg rule list, and this can be - * removed later, after FW updates to remove 6G reg rule - * from 5G rules list. - */ - if (memcmp(reg_info->alpha2, "US", 2) == 0) { - reg_info->num_5g_reg_rules = REG_US_5G_NUM_REG_RULES; - num_5g_reg_rules = reg_info->num_5g_reg_rules; - } - reg_info->dfs_region = le32_to_cpu(ev->dfs_region); reg_info->phybitmap = le32_to_cpu(ev->phybitmap); reg_info->num_phy = le32_to_cpu(ev->num_phy); @@ -4900,8 +4903,29 @@ static int ath12k_pull_reg_chan_list_ext_update_ev(struct ath12k_base *ab, } } + ext_wmi_reg_rule += num_2g_reg_rules; + + /* Firmware might include 6 GHz reg rule in 5 GHz rule list + * for few countries along with separate 6 GHz rule. + * Having same 6 GHz reg rule in 5 GHz and 6 GHz rules list + * causes intersect check to be true, and same rules will be + * shown multiple times in iw cmd. + * Hence, avoid parsing 6 GHz rule from 5 GHz reg rule list + */ + num_invalid_5ghz_ext_rules = ath12k_wmi_ignore_num_extra_rules(ext_wmi_reg_rule, + num_5g_reg_rules); + + if (num_invalid_5ghz_ext_rules) { + ath12k_dbg(ab, ATH12K_DBG_WMI, + "CC: %s 5 GHz reg rules number %d from fw, %d number of invalid 5 GHz rules", + reg_info->alpha2, reg_info->num_5g_reg_rules, + num_invalid_5ghz_ext_rules); + + num_5g_reg_rules = num_5g_reg_rules - num_invalid_5ghz_ext_rules; + reg_info->num_5g_reg_rules = num_5g_reg_rules; + } + if (num_5g_reg_rules) { - ext_wmi_reg_rule += num_2g_reg_rules; reg_info->reg_rules_5g_ptr = create_ext_reg_rules_from_wmi(num_5g_reg_rules, ext_wmi_reg_rule); @@ -4913,7 +4937,12 @@ static int ath12k_pull_reg_chan_list_ext_update_ev(struct ath12k_base *ab, } } - ext_wmi_reg_rule += num_5g_reg_rules; + /* We have adjusted the number of 5 GHz reg rules above. But still those + * many rules needs to be adjusted in ext_wmi_reg_rule. + * + * NOTE: num_invalid_5ghz_ext_rules will be 0 for rest other cases. + */ + ext_wmi_reg_rule += (num_5g_reg_rules + num_invalid_5ghz_ext_rules); for (i = 0; i < WMI_REG_CURRENT_MAX_AP_TYPE; i++) { reg_info->reg_rules_6g_ap_ptr[i] = diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h index 6f55dbdf629d..b16615b116ae 100644 --- a/drivers/net/wireless/ath/ath12k/wmi.h +++ b/drivers/net/wireless/ath/ath12k/wmi.h @@ -3943,7 +3943,6 @@ struct ath12k_wmi_eht_rate_set_params { #define MAX_REG_RULES 10 #define REG_ALPHA2_LEN 2 #define MAX_6G_REG_RULES 5 -#define REG_US_5G_NUM_REG_RULES 4 enum wmi_start_event_param { WMI_VDEV_START_RESP_EVENT = 0, From e76946110137703c16423baf6ee177b751a34b7e Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 23 Jan 2025 16:25:35 +0800 Subject: [PATCH 0181/1443] workqueue: Put the pwq after detaching the rescuer from the pool MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The commit 68f83057b913("workqueue: Reap workers via kthread_stop() and remove detach_completion") adds code to reap the normal workers but mistakenly does not handle the rescuer and also removes the code waiting for the rescuer in put_unbound_pool(), which caused a use-after-free bug reported by Cheung Wall. To avoid the use-after-free bug, the pool’s reference must be held until the detachment is complete. Therefore, move the code that puts the pwq after detaching the rescuer from the pool. Reported-by: cheung wall Cc: cheung wall Link: https://lore.kernel.org/lkml/CAKHoSAvP3iQW+GwmKzWjEAOoPvzeWeoMO0Gz7Pp3_4kxt-RMoA@mail.gmail.com/ Fixes: 68f83057b913("workqueue: Reap workers via kthread_stop() and remove detach_completion") Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo --- kernel/workqueue.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 33a23c7b2274..ccad33001c58 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3516,12 +3516,6 @@ static int rescuer_thread(void *__rescuer) } } - /* - * Put the reference grabbed by send_mayday(). @pool won't - * go away while we're still attached to it. - */ - put_pwq(pwq); - /* * Leave this pool. Notify regular workers; otherwise, we end up * with 0 concurrency and stalling the execution. @@ -3532,6 +3526,12 @@ static int rescuer_thread(void *__rescuer) worker_detach_from_pool(rescuer); + /* + * Put the reference grabbed by send_mayday(). @pool might + * go away any time after it. + */ + put_pwq_unlocked(pwq); + raw_spin_lock_irq(&wq_mayday_lock); } From e9fe182772dcb2630964724fd93e9c90b68ea0fd Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 24 Jan 2025 10:48:25 -1000 Subject: [PATCH 0182/1443] sched_ext: selftests/dsp_local_on: Fix sporadic failures dsp_local_on has several incorrect assumptions, one of which is that p->nr_cpus_allowed always tracks p->cpus_ptr. This is not true when a task is scheduled out while migration is disabled - p->cpus_ptr is temporarily overridden to the previous CPU while p->nr_cpus_allowed remains unchanged. This led to sporadic test faliures when dsp_local_on_dispatch() tries to put a migration disabled task to a different CPU. Fix it by keeping the previous CPU when migration is disabled. There are SCX schedulers that make use of p->nr_cpus_allowed. They should also implement explicit handling for p->migration_disabled. Signed-off-by: Tejun Heo Reported-by: Ihor Solodrai Cc: Andrea Righi Cc: Changwoo Min --- tools/testing/selftests/sched_ext/dsp_local_on.bpf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c index fbda6bf54671..758b479bd1ee 100644 --- a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c +++ b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c @@ -43,7 +43,7 @@ void BPF_STRUCT_OPS(dsp_local_on_dispatch, s32 cpu, struct task_struct *prev) if (!p) return; - if (p->nr_cpus_allowed == nr_cpus) + if (p->nr_cpus_allowed == nr_cpus && !p->migration_disabled) target = bpf_get_prandom_u32() % nr_cpus; else target = scx_bpf_task_cpu(p); From 7f3b11934f9684038555fd9160e188d6893949b9 Mon Sep 17 00:00:00 2001 From: Sasha Finkelstein Date: Thu, 9 Jan 2025 22:45:46 +0100 Subject: [PATCH 0183/1443] drm/virtio: Support partial maps of GEM objects Those are useful to implement coherent cross-vm mmap. Signed-off-by: Sasha Finkelstein Reviewed-by: Dmitry Osipenko Tested-by: Dmitry Osipenko Link: https://patchwork.freedesktop.org/patch/msgid/20250109-virtgpu-gem-partial-map-v1-1-a914b48776bd@gmail.com Signed-off-by: Dmitry Osipenko [dmitry.osipenko@collabora.com: Corrected vm_size check to account vm_pgoff] --- drivers/gpu/drm/virtio/virtgpu_vram.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/virtio/virtgpu_vram.c b/drivers/gpu/drm/virtio/virtgpu_vram.c index 25df81c02783..5ad3b7c6f73c 100644 --- a/drivers/gpu/drm/virtio/virtgpu_vram.c +++ b/drivers/gpu/drm/virtio/virtgpu_vram.c @@ -37,6 +37,7 @@ static int virtio_gpu_vram_mmap(struct drm_gem_object *obj, struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(obj); struct virtio_gpu_object_vram *vram = to_virtio_gpu_vram(bo); unsigned long vm_size = vma->vm_end - vma->vm_start; + unsigned long vm_end; if (!(bo->blob_flags & VIRTGPU_BLOB_FLAG_USE_MAPPABLE)) return -EINVAL; @@ -56,12 +57,14 @@ static int virtio_gpu_vram_mmap(struct drm_gem_object *obj, else if (vram->map_info == VIRTIO_GPU_MAP_CACHE_UNCACHED) vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); - /* Partial mappings of GEM buffers don't happen much in practice. */ - if (vm_size != vram->vram_node.size) + if (check_add_overflow(vma->vm_pgoff << PAGE_SHIFT, vm_size, &vm_end)) + return -EINVAL; + + if (vm_end > vram->vram_node.size) return -EINVAL; ret = io_remap_pfn_range(vma, vma->vm_start, - vram->vram_node.start >> PAGE_SHIFT, + (vram->vram_node.start >> PAGE_SHIFT) + vma->vm_pgoff, vm_size, vma->vm_page_prot); return ret; } From ce8c69ec90ca83fd4eacb246e2be699e6b439e97 Mon Sep 17 00:00:00 2001 From: Tejas Vipin Date: Wed, 22 Jan 2025 08:41:10 +0530 Subject: [PATCH 0184/1443] drm/panel: samsung-s6e88a0-ams452ef01: transition to mipi_dsi wrapped functions Changes the samsung-s6e88a0-ams452ef01 panel to use multi style functions for improved error handling. Reviewed-by: Douglas Anderson Signed-off-by: Tejas Vipin Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250122031110.286079-1-tejasvipin76@gmail.com --- .../panel/panel-samsung-s6e88a0-ams452ef01.c | 91 +++++++------------ 1 file changed, 31 insertions(+), 60 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c index d2df227abbea..57b1a899bbdc 100644 --- a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c +++ b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams452ef01.c @@ -39,91 +39,66 @@ static void s6e88a0_ams452ef01_reset(struct s6e88a0_ams452ef01 *ctx) static int s6e88a0_ams452ef01_on(struct s6e88a0_ams452ef01 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dsi->mode_flags |= MIPI_DSI_MODE_LPM; - mipi_dsi_dcs_write_seq(dsi, 0xf0, 0x5a, 0x5a); // enable LEVEL2 commands - mipi_dsi_dcs_write_seq(dsi, 0xcc, 0x4c); // set Pixel Clock Divider polarity + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0x5a, 0x5a); // enable LEVEL2 commands + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xcc, 0x4c); // set Pixel Clock Divider polarity - ret = mipi_dsi_dcs_exit_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to exit sleep mode: %d\n", ret); - return ret; - } - msleep(120); + mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 120); // set default brightness/gama - mipi_dsi_dcs_write_seq(dsi, 0xca, - 0x01, 0x00, 0x01, 0x00, 0x01, 0x00, // V255 RR,GG,BB - 0x80, 0x80, 0x80, // V203 R,G,B - 0x80, 0x80, 0x80, // V151 R,G,B - 0x80, 0x80, 0x80, // V87 R,G,B - 0x80, 0x80, 0x80, // V51 R,G,B - 0x80, 0x80, 0x80, // V35 R,G,B - 0x80, 0x80, 0x80, // V23 R,G,B - 0x80, 0x80, 0x80, // V11 R,G,B - 0x6b, 0x68, 0x71, // V3 R,G,B - 0x00, 0x00, 0x00); // V1 R,G,B + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xca, + 0x01, 0x00, 0x01, 0x00, 0x01, 0x00,// V255 RR,GG,BB + 0x80, 0x80, 0x80, // V203 R,G,B + 0x80, 0x80, 0x80, // V151 R,G,B + 0x80, 0x80, 0x80, // V87 R,G,B + 0x80, 0x80, 0x80, // V51 R,G,B + 0x80, 0x80, 0x80, // V35 R,G,B + 0x80, 0x80, 0x80, // V23 R,G,B + 0x80, 0x80, 0x80, // V11 R,G,B + 0x6b, 0x68, 0x71, // V3 R,G,B + 0x00, 0x00, 0x00); // V1 R,G,B // set default Amoled Off Ratio - mipi_dsi_dcs_write_seq(dsi, 0xb2, 0x40, 0x0a, 0x17, 0x00, 0x0a); - mipi_dsi_dcs_write_seq(dsi, 0xb6, 0x2c, 0x0b); // set default elvss voltage - mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_POWER_SAVE, 0x00); - mipi_dsi_dcs_write_seq(dsi, 0xf7, 0x03); // gamma/aor update - mipi_dsi_dcs_write_seq(dsi, 0xf0, 0xa5, 0xa5); // disable LEVEL2 commands + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb2, 0x40, 0x0a, 0x17, 0x00, 0x0a); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xb6, 0x2c, 0x0b); // set default elvss voltage + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_POWER_SAVE, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf7, 0x03); // gamma/aor update + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf0, 0xa5, 0xa5); // disable LEVEL2 commands - ret = mipi_dsi_dcs_set_display_on(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display on: %d\n", ret); - return ret; - } + mipi_dsi_dcs_set_display_on_multi(&dsi_ctx); - return 0; + return dsi_ctx.accum_err; } -static int s6e88a0_ams452ef01_off(struct s6e88a0_ams452ef01 *ctx) +static void s6e88a0_ams452ef01_off(struct s6e88a0_ams452ef01 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi}; dsi->mode_flags &= ~MIPI_DSI_MODE_LPM; - ret = mipi_dsi_dcs_set_display_off(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display off: %d\n", ret); - return ret; - } - msleep(35); - - ret = mipi_dsi_dcs_enter_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to enter sleep mode: %d\n", ret); - return ret; - } - msleep(120); - - return 0; + mipi_dsi_dcs_set_display_off_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 35); + mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 120); } static int s6e88a0_ams452ef01_prepare(struct drm_panel *panel) { struct s6e88a0_ams452ef01 *ctx = to_s6e88a0_ams452ef01(panel); - struct device *dev = &ctx->dsi->dev; int ret; ret = regulator_bulk_enable(ARRAY_SIZE(ctx->supplies), ctx->supplies); - if (ret < 0) { - dev_err(dev, "Failed to enable regulators: %d\n", ret); + if (ret < 0) return ret; - } s6e88a0_ams452ef01_reset(ctx); ret = s6e88a0_ams452ef01_on(ctx); if (ret < 0) { - dev_err(dev, "Failed to initialize panel: %d\n", ret); gpiod_set_value_cansleep(ctx->reset_gpio, 0); regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies); @@ -136,12 +111,8 @@ static int s6e88a0_ams452ef01_prepare(struct drm_panel *panel) static int s6e88a0_ams452ef01_unprepare(struct drm_panel *panel) { struct s6e88a0_ams452ef01 *ctx = to_s6e88a0_ams452ef01(panel); - struct device *dev = &ctx->dsi->dev; - int ret; - ret = s6e88a0_ams452ef01_off(ctx); - if (ret < 0) - dev_err(dev, "Failed to un-initialize panel: %d\n", ret); + s6e88a0_ams452ef01_off(ctx); gpiod_set_value_cansleep(ctx->reset_gpio, 0); regulator_bulk_disable(ARRAY_SIZE(ctx->supplies), ctx->supplies); From 6bbce873a9c97cb12f5455c497be279ac58e707f Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Mon, 13 Jan 2025 17:10:59 +0800 Subject: [PATCH 0185/1443] drm/dp_mst: Fix drm RAD print MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [Why] The RAD of sideband message printed today is incorrect. For RAD stored within MST branch - If MST branch LCT is 1, it's RAD array is untouched and remained as 0. - If MST branch LCT is larger than 1, use nibble to store the up facing port number in cascaded sequence as illustrated below: u8 RAD[0] = (LCT_2_UFP << 4) | LCT_3_UFP RAD[1] = (LCT_4_UFP << 4) | LCT_5_UFP ... In drm_dp_mst_rad_to_str(), it wrongly to use BIT_MASK(4) to fetch the port number of one nibble. [How] Adjust the code by: - RAD array items are valuable only for LCT >= 1. - Use 0xF as the mask to replace BIT_MASK(4) V2: - Document how RAD is constructed (Imre) V3: - Adjust the comment for rad[] so kdoc formats it properly (Lyude) Fixes: 2f015ec6eab6 ("drm/dp_mst: Add sideband down request tracing + selftests") Cc: Imre Deak Cc: Ville Syrjälä Cc: Harry Wentland Cc: Lyude Paul Reviewed-by: Lyude Paul Signed-off-by: Wayne Lin Signed-off-by: Lyude Paul Link: https://patchwork.freedesktop.org/patch/msgid/20250113091100.3314533-2-Wayne.Lin@amd.com --- drivers/gpu/drm/display/drm_dp_mst_topology.c | 8 ++++---- include/drm/display/drm_dp_mst_helper.h | 7 +++++++ 2 files changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c index f8cd094efa3c..271a00b81264 100644 --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c @@ -175,13 +175,13 @@ static int drm_dp_mst_rad_to_str(const u8 rad[8], u8 lct, char *out, size_t len) { int i; - u8 unpacked_rad[16]; + u8 unpacked_rad[16] = {}; - for (i = 0; i < lct; i++) { + for (i = 1; i < lct; i++) { if (i % 2) - unpacked_rad[i] = rad[i / 2] >> 4; + unpacked_rad[i] = rad[(i - 1) / 2] >> 4; else - unpacked_rad[i] = rad[i / 2] & BIT_MASK(4); + unpacked_rad[i] = rad[(i - 1) / 2] & 0xF; } /* TODO: Eventually add something to printk so we can format the rad diff --git a/include/drm/display/drm_dp_mst_helper.h b/include/drm/display/drm_dp_mst_helper.h index a80ba457a858..6398a6b50bd1 100644 --- a/include/drm/display/drm_dp_mst_helper.h +++ b/include/drm/display/drm_dp_mst_helper.h @@ -222,6 +222,13 @@ struct drm_dp_mst_branch { */ struct list_head destroy_next; + /** + * @rad: Relative Address of the MST branch. + * For &drm_dp_mst_topology_mgr.mst_primary, it's rad[8] are all 0, + * unset and unused. For MST branches connected after mst_primary, + * in each element of rad[] the nibbles are ordered by the most + * signifcant 4 bits first and the least significant 4 bits second. + */ u8 rad[8]; u8 lct; int num_ports; From 580aa2782c5b9eb50cb3bf936bc57ec0bf7d55db Mon Sep 17 00:00:00 2001 From: Wayne Lin Date: Mon, 13 Jan 2025 17:11:00 +0800 Subject: [PATCH 0186/1443] drm/dp_mst: Add helper to get port number at specific LCT from RAD MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a helper drm_dp_mst_get_ufp_num_at_lct_from_rad() to extract the up facing port number at specific link count from the RAD. Use the added helper in drm_dp_mst_rad_to_str() & drm_dp_get_mst_branch_device() to unify the implementation. V2: - Adjust the code format (Lyude) V3: - Adjust parameter "rad" of drm_dp_mst_get_ufp_num_at_lct_from_rad() to be constant (Kernel test robot) Cc: Imre Deak Cc: Ville Syrjälä Cc: Harry Wentland Cc: Lyude Paul Reviewed-by: Lyude Paul Signed-off-by: Wayne Lin [fixed commit message typo] Signed-off-by: Lyude Paul Link: https://patchwork.freedesktop.org/patch/msgid/20250113091100.3314533-3-Wayne.Lin@amd.com --- drivers/gpu/drm/display/drm_dp_mst_topology.c | 29 +++++++++++++------ 1 file changed, 20 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c index 271a00b81264..e79f4b76ec2b 100644 --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c @@ -171,18 +171,30 @@ static const char *drm_dp_mst_sideband_tx_state_str(int state) return sideband_reason_str[state]; } +static inline u8 +drm_dp_mst_get_ufp_num_at_lct_from_rad(u8 lct, const u8 *rad) +{ + int idx = (lct / 2) - 1; + int shift = (lct % 2) ? 0 : 4; + u8 ufp_num; + + /* mst_primary, it's rad is unset*/ + if (lct == 1) + return 0; + + ufp_num = (rad[idx] >> shift) & 0xf; + + return ufp_num; +} + static int drm_dp_mst_rad_to_str(const u8 rad[8], u8 lct, char *out, size_t len) { int i; u8 unpacked_rad[16] = {}; - for (i = 1; i < lct; i++) { - if (i % 2) - unpacked_rad[i] = rad[(i - 1) / 2] >> 4; - else - unpacked_rad[i] = rad[(i - 1) / 2] & 0xF; - } + for (i = 0; i < lct; i++) + unpacked_rad[i] = drm_dp_mst_get_ufp_num_at_lct_from_rad(i + 1, rad); /* TODO: Eventually add something to printk so we can format the rad * like this: 1.2.3 @@ -2544,9 +2556,8 @@ static struct drm_dp_mst_branch *drm_dp_get_mst_branch_device(struct drm_dp_mst_ if (!mstb) goto out; - for (i = 0; i < lct - 1; i++) { - int shift = (i % 2) ? 0 : 4; - int port_num = (rad[i / 2] >> shift) & 0xf; + for (i = 1; i < lct; i++) { + int port_num = drm_dp_mst_get_ufp_num_at_lct_from_rad(i + 1, rad); list_for_each_entry(port, &mstb->ports, next) { if (port->port_num == port_num) { From 22807d30fa854bbbe26e53774c9c899c39b62a71 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:44 -0800 Subject: [PATCH 0187/1443] drm/nouveau: add a kernel doc to introduce the GSP RPC In order to explain the name clean-ups in GSP RPC routines, a kernel doc to explain the memory layout and terms is required. Add a kernel doc to introduce the GSP RPC. Cc: Danilo Krummrich Signed-off-by: Zhi Wang [ Fix bullet list indentation; add SPDX-License-Identifier. - Danilo ] Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-2-zhiw@nvidia.com --- Documentation/gpu/drivers.rst | 1 + Documentation/gpu/nouveau.rst | 29 ++++++++++++ .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 45 +++++++++++++++++++ 3 files changed, 75 insertions(+) create mode 100644 Documentation/gpu/nouveau.rst diff --git a/Documentation/gpu/drivers.rst b/Documentation/gpu/drivers.rst index 1f17ad0790d7..74fc2cbf1b6f 100644 --- a/Documentation/gpu/drivers.rst +++ b/Documentation/gpu/drivers.rst @@ -10,6 +10,7 @@ GPU Driver Documentation imagination/index mcde meson + nouveau pl111 tegra tve200 diff --git a/Documentation/gpu/nouveau.rst b/Documentation/gpu/nouveau.rst new file mode 100644 index 000000000000..0f34131ccc27 --- /dev/null +++ b/Documentation/gpu/nouveau.rst @@ -0,0 +1,29 @@ +.. SPDX-License-Identifier: (GPL-2.0+ OR MIT) + +=============================== + drm/nouveau NVIDIA GPU Driver +=============================== + +The drm/nouveau driver provides support for a wide range of NVIDIA GPUs, +covering GeForce, Quadro, and Tesla series, from the NV04 architecture up +to the latest Turing, Ampere, Ada families. + +NVKM: NVIDIA Kernel Manager +=========================== + +The NVKM component serves as the core abstraction layer within the nouveau +driver, responsible for managing NVIDIA GPU hardware at the kernel level. +NVKM provides a unified interface for handling various GPU architectures. + +It enables resource management, power control, memory handling, and command +submission required for the proper functioning of NVIDIA GPUs under the +nouveau driver. + +NVKM plays a critical role in abstracting hardware complexities and +providing a consistent API to upper layers of the driver stack. + +GSP Support +------------------------ + +.. kernel-doc:: drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c + :doc: GSP message queue element diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 10fe2d15b5ce..b0d623611f88 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -63,6 +63,51 @@ extern struct dentry *nouveau_debugfs_root; #define GSP_MSG_MIN_SIZE GSP_PAGE_SIZE #define GSP_MSG_MAX_SIZE GSP_PAGE_MIN_SIZE * 16 +/** + * DOC: GSP message queue element + * + * https://github.com/NVIDIA/open-gpu-kernel-modules/blob/535/src/nvidia/inc/kernel/gpu/gsp/message_queue_priv.h + * + * The GSP command queue and status queue are message queues for the + * communication between software and GSP. The software submits the GSP + * RPC via the GSP command queue, GSP writes the status of the submitted + * RPC in the status queue. + * + * A GSP message queue element consists of three parts: + * + * - message element header (struct r535_gsp_msg), which mostly maintains + * the metadata for queuing the element. + * + * - RPC message header (struct nvfw_gsp_rpc), which maintains the info + * of the RPC. E.g., the RPC function number. + * + * - The payload, where the RPC message stays. E.g. the params of a + * specific RPC function. Some RPC functions also have their headers + * in the payload. E.g. rm_alloc, rm_control. + * + * The memory layout of a GSP message element can be illustrated below:: + * + * +------------------------+ + * | Message Element Header | + * | (r535_gsp_msg) | + * | | + * | (r535_gsp_msg.data) | + * | | | + * |----------V-------------| + * | GSP RPC Header | + * | (nvfw_gsp_rpc) | + * | | + * | (nvfw_gsp_rpc.data) | + * | | | + * |----------V-------------| + * | Payload | + * | | + * | header(optional) | + * | params | + * +------------------------+ + * + */ + struct r535_gsp_msg { u8 auth_tag_buffer[16]; u8 aad_buffer[16]; From f98ed88eb98b84f3e12ff09ec41a36a39dc43a98 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:45 -0800 Subject: [PATCH 0188/1443] drm/nouveau: rename "repc" to "gsp_rpc_len" on the GSP message recv path The name "repc" has different meanings in different contexts. To improve the readability, it's better to refine it to a name that reflects what it actually represents. Rename "repc" to "gsp_rpc_len" in the GSP message recv path. Add an section in the doc to explain the terms. No functional change is intended. Cc: Danilo Krummrich Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-3-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 46 +++++++++++-------- 1 file changed, 27 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index b0d623611f88..2f61e1a4c9ba 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -106,6 +106,9 @@ extern struct dentry *nouveau_debugfs_root; * | params | * +------------------------+ * + * Terminology: + * + * - gsp_rpc_len: size of (GSP RPC header + payload) */ struct r535_gsp_msg { @@ -135,7 +138,8 @@ r535_rpc_status_to_errno(uint32_t rpc_status) } static void * -r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 repc, u32 *prepc, int *ptime) +r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, u32 *prepc, + int *ptime) { struct r535_gsp_msg *mqe; u32 size, rptr = *gsp->msgq.rptr; @@ -143,7 +147,8 @@ r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 repc, u32 *prepc, int *ptime) u8 *msg; u32 len; - size = DIV_ROUND_UP(GSP_MSG_HDR_SIZE + repc, GSP_PAGE_SIZE); + size = DIV_ROUND_UP(GSP_MSG_HDR_SIZE + gsp_rpc_len, + GSP_PAGE_SIZE); if (WARN_ON(!size || size >= gsp->msgq.cnt)) return ERR_PTR(-EINVAL); @@ -169,21 +174,21 @@ r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 repc, u32 *prepc, int *ptime) return mqe->data; } - size = ALIGN(repc + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); + size = ALIGN(gsp_rpc_len + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); - msg = kvmalloc(repc, GFP_KERNEL); + msg = kvmalloc(gsp_rpc_len, GFP_KERNEL); if (!msg) return ERR_PTR(-ENOMEM); len = ((gsp->msgq.cnt - rptr) * GSP_PAGE_SIZE) - sizeof(*mqe); - len = min_t(u32, repc, len); + len = min_t(u32, gsp_rpc_len, len); memcpy(msg, mqe->data, len); - repc -= len; + gsp_rpc_len -= len; - if (repc) { + if (gsp_rpc_len) { mqe = (void *)((u8 *)gsp->shm.msgq.ptr + 0x1000 + 0 * 0x1000); - memcpy(msg + len, mqe, repc); + memcpy(msg + len, mqe, gsp_rpc_len); } rptr = (rptr + DIV_ROUND_UP(size, GSP_PAGE_SIZE)) % gsp->msgq.cnt; @@ -194,9 +199,9 @@ r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 repc, u32 *prepc, int *ptime) } static void * -r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 repc, int *ptime) +r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime) { - return r535_gsp_msgq_wait(gsp, repc, NULL, ptime); + return r535_gsp_msgq_wait(gsp, gsp_rpc_len, NULL, ptime); } static int @@ -317,7 +322,7 @@ r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl) } static struct nvfw_gsp_rpc * -r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 repc) +r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len) { struct nvkm_subdev *subdev = &gsp->subdev; struct nvfw_gsp_rpc *msg; @@ -342,10 +347,11 @@ r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 repc) r535_gsp_msg_dump(gsp, msg, NV_DBG_TRACE); if (fn && msg->function == fn) { - if (repc) { - if (msg->length < sizeof(*msg) + repc) { + if (gsp_rpc_len) { + if (msg->length < sizeof(*msg) + gsp_rpc_len) { nvkm_error(subdev, "msg len %d < %zd\n", - msg->length, sizeof(*msg) + repc); + msg->length, sizeof(*msg) + + gsp_rpc_len); r535_gsp_msg_dump(gsp, msg, NV_DBG_ERROR); r535_gsp_msg_done(gsp, msg); return ERR_PTR(-EIO); @@ -414,7 +420,8 @@ r535_gsp_rpc_poll(struct nvkm_gsp *gsp, u32 fn) } static void * -r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc) +r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *argv, bool wait, + u32 gsp_rpc_len) { struct nvfw_gsp_rpc *rpc = container_of(argv, typeof(*rpc), data); struct nvfw_gsp_rpc *msg; @@ -434,7 +441,7 @@ r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc) return ERR_PTR(ret); if (wait) { - msg = r535_gsp_msg_recv(gsp, fn, repc); + msg = r535_gsp_msg_recv(gsp, fn, gsp_rpc_len); if (!IS_ERR_OR_NULL(msg)) repv = msg->data; else @@ -770,7 +777,8 @@ r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 argc) } static void * -r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc) +r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, + u32 gsp_rpc_len) { struct nvfw_gsp_rpc *rpc = container_of(argv, typeof(*rpc), data); struct r535_gsp_msg *cmd = container_of((void *)rpc, typeof(*cmd), data); @@ -817,7 +825,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc) /* Wait for reply. */ if (wait) { - rpc = r535_gsp_msg_recv(gsp, fn, repc); + rpc = r535_gsp_msg_recv(gsp, fn, gsp_rpc_len); if (!IS_ERR_OR_NULL(rpc)) repv = rpc->data; else @@ -826,7 +834,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, u32 repc) repv = NULL; } } else { - repv = r535_gsp_rpc_send(gsp, argv, wait, repc); + repv = r535_gsp_rpc_send(gsp, argv, wait, gsp_rpc_len); } done: From 2c6a79af3fe0bc5db6f1097d0f64380f557830af Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:46 -0800 Subject: [PATCH 0189/1443] drm/nouveau: rename "argv" to what it represents on the GSP message send path The name "argv" has different meanings in different functions. To improve the readability, it's better to refine it to a name that reflects what it represents. Rename "repc" to what it represents in the GSP message send path. Wrap the long container_of() into to_gsp_hdr() to make checkpatch.pl happy. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-4-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 27 ++++++++++--------- 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 2f61e1a4c9ba..06676c17faab 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -123,6 +123,9 @@ struct r535_gsp_msg { #define GSP_MSG_HDR_SIZE offsetof(struct r535_gsp_msg, data) +#define to_gsp_hdr(p, header) \ + container_of((void *)p, typeof(*header), data) + static int r535_rpc_status_to_errno(uint32_t rpc_status) { @@ -205,9 +208,9 @@ r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime) } static int -r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *argv) +r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) { - struct r535_gsp_msg *cmd = container_of(argv, typeof(*cmd), data); + struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); struct r535_gsp_msg *cqe; u32 argc = cmd->checksum; u64 *ptr = (void *)cmd; @@ -420,10 +423,10 @@ r535_gsp_rpc_poll(struct nvkm_gsp *gsp, u32 fn) } static void * -r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *argv, bool wait, +r535_gsp_rpc_send(struct nvkm_gsp *gsp, void *payload, bool wait, u32 gsp_rpc_len) { - struct nvfw_gsp_rpc *rpc = container_of(argv, typeof(*rpc), data); + struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc); struct nvfw_gsp_rpc *msg; u32 fn = rpc->function; void *repv = NULL; @@ -777,11 +780,11 @@ r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 argc) } static void * -r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, +r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, u32 gsp_rpc_len) { - struct nvfw_gsp_rpc *rpc = container_of(argv, typeof(*rpc), data); - struct r535_gsp_msg *cmd = container_of((void *)rpc, typeof(*cmd), data); + struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc); + struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); const u32 max_msg_size = (16 * 0x1000) - sizeof(struct r535_gsp_msg); const u32 max_rpc_size = max_msg_size - sizeof(*rpc); u32 rpc_size = rpc->length - sizeof(*rpc); @@ -795,11 +798,11 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, rpc->length = sizeof(*rpc) + max_rpc_size; cmd->checksum = rpc->length; - repv = r535_gsp_rpc_send(gsp, argv, false, 0); + repv = r535_gsp_rpc_send(gsp, payload, false, 0); if (IS_ERR(repv)) goto done; - argv += max_rpc_size; + payload += max_rpc_size; rpc_size -= max_rpc_size; /* Remaining chunks sent as CONTINUATION_RECORD RPCs. */ @@ -813,13 +816,13 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, goto done; } - memcpy(next, argv, size); + memcpy(next, payload, size); repv = r535_gsp_rpc_send(gsp, next, false, 0); if (IS_ERR(repv)) goto done; - argv += size; + payload += size; rpc_size -= size; } @@ -834,7 +837,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *argv, bool wait, repv = NULL; } } else { - repv = r535_gsp_rpc_send(gsp, argv, wait, gsp_rpc_len); + repv = r535_gsp_rpc_send(gsp, payload, wait, gsp_rpc_len); } done: From a15b5379764be947001ab81bedc3ad6ee93919f8 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:47 -0800 Subject: [PATCH 0190/1443] drm/nouveau: remove unused param repc in *rm_alloc_push() The user of *rm_alloc_push() always pass 0 in repc. Remove unused param repc since no user actually uses it. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-5-zhiw@nvidia.com --- drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h | 8 ++++---- drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 8 +++----- 2 files changed, 7 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h index 5c5f4607fcc9..746e126c3ecf 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h @@ -196,7 +196,7 @@ struct nvkm_gsp { void (*rm_ctrl_done)(struct nvkm_gsp_object *, void *repv); void *(*rm_alloc_get)(struct nvkm_gsp_object *, u32 oclass, u32 argc); - void *(*rm_alloc_push)(struct nvkm_gsp_object *, void *argv, u32 repc); + void *(*rm_alloc_push)(struct nvkm_gsp_object *, void *argv); void (*rm_alloc_done)(struct nvkm_gsp_object *, void *repv); int (*rm_free)(struct nvkm_gsp_object *); @@ -353,9 +353,9 @@ nvkm_gsp_rm_alloc_get(struct nvkm_gsp_object *parent, u32 handle, u32 oclass, u3 } static inline void * -nvkm_gsp_rm_alloc_push(struct nvkm_gsp_object *object, void *argv, u32 repc) +nvkm_gsp_rm_alloc_push(struct nvkm_gsp_object *object, void *argv) { - void *repv = object->client->gsp->rm->rm_alloc_push(object, argv, repc); + void *repv = object->client->gsp->rm->rm_alloc_push(object, argv); if (IS_ERR(repv)) object->client = NULL; @@ -366,7 +366,7 @@ nvkm_gsp_rm_alloc_push(struct nvkm_gsp_object *object, void *argv, u32 repc) static inline int nvkm_gsp_rm_alloc_wr(struct nvkm_gsp_object *object, void *argv) { - void *repv = nvkm_gsp_rm_alloc_push(object, argv, 0); + void *repv = nvkm_gsp_rm_alloc_push(object, argv); if (IS_ERR(repv)) return PTR_ERR(repv); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 06676c17faab..4d15a9dea934 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -647,13 +647,13 @@ r535_gsp_rpc_rm_alloc_done(struct nvkm_gsp_object *object, void *repv) } static void * -r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *argv, u32 repc) +r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *argv) { rpc_gsp_rm_alloc_v03_00 *rpc = container_of(argv, typeof(*rpc), params); struct nvkm_gsp *gsp = object->client->gsp; - void *ret; + void *ret = NULL; - rpc = nvkm_gsp_rpc_push(gsp, rpc, true, sizeof(*rpc) + repc); + rpc = nvkm_gsp_rpc_push(gsp, rpc, true, sizeof(*rpc)); if (IS_ERR_OR_NULL(rpc)) return rpc; @@ -661,8 +661,6 @@ r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *argv, u32 repc) ret = ERR_PTR(r535_rpc_status_to_errno(rpc->status)); if (PTR_ERR(ret) != -EAGAIN && PTR_ERR(ret) != -EBUSY) nvkm_error(&gsp->subdev, "RM_ALLOC: 0x%x\n", rpc->status); - } else { - ret = repc ? rpc->params : NULL; } nvkm_gsp_rpc_done(gsp, rpc); From 0c2f211b66051f97f8e3628be3a486a5841aa1b8 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:48 -0800 Subject: [PATCH 0191/1443] drm/nouveau: rename "argv" to what it represents in *rm_{alloc, ctrl}_*() The name "argv" has different meanings in different functions. To improve the readability, it's better to refine it to a name that reflects what it represents. Rename "argv" to what it represents. Wrap the long container_of() into to_payload_header() to denote a clear meaning and make checkpatch.pl happy. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-6-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 25 +++++++++++-------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 4d15a9dea934..d3b2b545841b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -126,6 +126,9 @@ struct r535_gsp_msg { #define to_gsp_hdr(p, header) \ container_of((void *)p, typeof(*header), data) +#define to_payload_hdr(p, header) \ + container_of((void *)p, typeof(*header), params) + static int r535_rpc_status_to_errno(uint32_t rpc_status) { @@ -639,17 +642,17 @@ r535_gsp_rpc_rm_free(struct nvkm_gsp_object *object) } static void -r535_gsp_rpc_rm_alloc_done(struct nvkm_gsp_object *object, void *repv) +r535_gsp_rpc_rm_alloc_done(struct nvkm_gsp_object *object, void *params) { - rpc_gsp_rm_alloc_v03_00 *rpc = container_of(repv, typeof(*rpc), params); + rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc); nvkm_gsp_rpc_done(object->client->gsp, rpc); } static void * -r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *argv) +r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *params) { - rpc_gsp_rm_alloc_v03_00 *rpc = container_of(argv, typeof(*rpc), params); + rpc_gsp_rm_alloc_v03_00 *rpc = to_payload_hdr(params, rpc); struct nvkm_gsp *gsp = object->client->gsp; void *ret = NULL; @@ -692,25 +695,25 @@ r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass, u32 argc) } static void -r535_gsp_rpc_rm_ctrl_done(struct nvkm_gsp_object *object, void *repv) +r535_gsp_rpc_rm_ctrl_done(struct nvkm_gsp_object *object, void *params) { - rpc_gsp_rm_control_v03_00 *rpc = container_of(repv, typeof(*rpc), params); + rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr(params, rpc); - if (!repv) + if (!params) return; nvkm_gsp_rpc_done(object->client->gsp, rpc); } static int -r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **argv, u32 repc) +r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **params, u32 repc) { - rpc_gsp_rm_control_v03_00 *rpc = container_of((*argv), typeof(*rpc), params); + rpc_gsp_rm_control_v03_00 *rpc = to_payload_hdr((*params), rpc); struct nvkm_gsp *gsp = object->client->gsp; int ret = 0; rpc = nvkm_gsp_rpc_push(gsp, rpc, true, repc); if (IS_ERR_OR_NULL(rpc)) { - *argv = NULL; + *params = NULL; return PTR_ERR(rpc); } @@ -722,7 +725,7 @@ r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **argv, u32 repc) } if (repc) - *argv = rpc->params; + *params = rpc->params; else nvkm_gsp_rpc_done(gsp, rpc); From bda6fe811fba6742c4ec2f3c0a962b02d7d6e87f Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:49 -0800 Subject: [PATCH 0192/1443] drm/nouveau: rename "argc" to what it represents in GSP RPC routines The name "argc" has different meanings in different functions. To improve the readability, it's better to refine it to a name that reflects what it represents. Rename "argc" to what it represents. Add terms in the decoder section to explain their meaning. No functional change is intended. Signed-off-by: Zhi Wang [ Fix indentation. - Danilo ] Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-7-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 58 +++++++++++-------- 1 file changed, 34 insertions(+), 24 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index d3b2b545841b..0753fe07e914 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -109,6 +109,8 @@ extern struct dentry *nouveau_debugfs_root; * Terminology: * * - gsp_rpc_len: size of (GSP RPC header + payload) + * - params_size: size of params in the payload + * - payload_size: size of (header if exists + params) in the payload */ struct r535_gsp_msg { @@ -215,21 +217,21 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) { struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); struct r535_gsp_msg *cqe; - u32 argc = cmd->checksum; + u32 gsp_rpc_len = cmd->checksum; u64 *ptr = (void *)cmd; u64 *end; u64 csum = 0; int free, time = 1000000; - u32 wptr, size, step; + u32 wptr, size, step, len; u32 off = 0; - argc = ALIGN(GSP_MSG_HDR_SIZE + argc, GSP_PAGE_SIZE); + len = ALIGN(GSP_MSG_HDR_SIZE + gsp_rpc_len, GSP_PAGE_SIZE); - end = (u64 *)((char *)ptr + argc); + end = (u64 *)((char *)ptr + len); cmd->pad = 0; cmd->checksum = 0; cmd->sequence = gsp->cmdq.seq++; - cmd->elem_count = DIV_ROUND_UP(argc, 0x1000); + cmd->elem_count = DIV_ROUND_UP(len, 0x1000); while (ptr < end) csum ^= *ptr++; @@ -255,7 +257,7 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) cqe = (void *)((u8 *)gsp->shm.cmdq.ptr + 0x1000 + wptr * 0x1000); step = min_t(u32, free, (gsp->cmdq.cnt - wptr)); - size = min_t(u32, argc, step * GSP_PAGE_SIZE); + size = min_t(u32, len, step * GSP_PAGE_SIZE); memcpy(cqe, (u8 *)cmd + off, size); @@ -264,8 +266,8 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) wptr = 0; off += size; - argc -= size; - } while(argc); + len -= size; + } while (len); nvkm_trace(&gsp->subdev, "cmdq: wptr %d\n", wptr); wmb(); @@ -279,17 +281,17 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) } static void * -r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 argc) +r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 gsp_rpc_len) { struct r535_gsp_msg *cmd; - u32 size = GSP_MSG_HDR_SIZE + argc; + u32 size = GSP_MSG_HDR_SIZE + gsp_rpc_len; size = ALIGN(size, GSP_MSG_MIN_SIZE); cmd = kvzalloc(size, GFP_KERNEL); if (!cmd) return ERR_PTR(-ENOMEM); - cmd->checksum = argc; + cmd->checksum = gsp_rpc_len; return cmd->data; } @@ -672,16 +674,22 @@ r535_gsp_rpc_rm_alloc_push(struct nvkm_gsp_object *object, void *params) } static void * -r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass, u32 argc) +r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass, + u32 params_size) { struct nvkm_gsp_client *client = object->client; struct nvkm_gsp *gsp = client->gsp; rpc_gsp_rm_alloc_v03_00 *rpc; - nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x new obj:0x%08x cls:0x%08x argc:%d\n", - client->object.handle, object->parent->handle, object->handle, oclass, argc); + nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x new obj:0x%08x\n", + client->object.handle, object->parent->handle, + object->handle); - rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_ALLOC, sizeof(*rpc) + argc); + nvkm_debug(&gsp->subdev, "cls:0x%08x params_size:%d\n", oclass, + params_size); + + rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_ALLOC, + sizeof(*rpc) + params_size); if (IS_ERR(rpc)) return rpc; @@ -690,7 +698,7 @@ r535_gsp_rpc_rm_alloc_get(struct nvkm_gsp_object *object, u32 oclass, u32 argc) rpc->hObject = object->handle; rpc->hClass = oclass; rpc->status = 0; - rpc->paramsSize = argc; + rpc->paramsSize = params_size; return rpc->params; } @@ -733,16 +741,17 @@ r535_gsp_rpc_rm_ctrl_push(struct nvkm_gsp_object *object, void **params, u32 rep } static void * -r535_gsp_rpc_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 argc) +r535_gsp_rpc_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 params_size) { struct nvkm_gsp_client *client = object->client; struct nvkm_gsp *gsp = client->gsp; rpc_gsp_rm_control_v03_00 *rpc; - nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x argc:%d\n", - client->object.handle, object->handle, cmd, argc); + nvkm_debug(&gsp->subdev, "cli:0x%08x obj:0x%08x ctrl cmd:0x%08x params_size:%d\n", + client->object.handle, object->handle, cmd, params_size); - rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL, sizeof(*rpc) + argc); + rpc = nvkm_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_GSP_RM_CONTROL, + sizeof(*rpc) + params_size); if (IS_ERR(rpc)) return rpc; @@ -750,7 +759,7 @@ r535_gsp_rpc_rm_ctrl_get(struct nvkm_gsp_object *object, u32 cmd, u32 argc) rpc->hObject = object->handle; rpc->cmd = cmd; rpc->status = 0; - rpc->paramsSize = argc; + rpc->paramsSize = params_size; return rpc->params; } @@ -763,11 +772,12 @@ r535_gsp_rpc_done(struct nvkm_gsp *gsp, void *repv) } static void * -r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 argc) +r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 payload_size) { struct nvfw_gsp_rpc *rpc; - rpc = r535_gsp_cmdq_get(gsp, ALIGN(sizeof(*rpc) + argc, sizeof(u64))); + rpc = r535_gsp_cmdq_get(gsp, ALIGN(sizeof(*rpc) + payload_size, + sizeof(u64))); if (IS_ERR(rpc)) return ERR_CAST(rpc); @@ -776,7 +786,7 @@ r535_gsp_rpc_get(struct nvkm_gsp *gsp, u32 fn, u32 argc) rpc->function = fn; rpc->rpc_result = 0xffffffff; rpc->rpc_result_private = 0xffffffff; - rpc->length = sizeof(*rpc) + argc; + rpc->length = sizeof(*rpc) + payload_size; return rpc->data; } From bbae6680cfe38b033250b483722e60ccd865976f Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:50 -0800 Subject: [PATCH 0193/1443] drm/nouveau: fix the broken marco GSP_MSG_MAX_SIZE The macro GSP_MSG_MAX_SIZE refers to another macro that doesn't exist. It represents the max GSP message element size. Fix the broken marco so it can be used to replace some magic numbers in the code. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-8-zhiw@nvidia.com --- drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 0753fe07e914..4e1136999678 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -61,7 +61,7 @@ extern struct dentry *nouveau_debugfs_root; #define GSP_MSG_MIN_SIZE GSP_PAGE_SIZE -#define GSP_MSG_MAX_SIZE GSP_PAGE_MIN_SIZE * 16 +#define GSP_MSG_MAX_SIZE (GSP_MSG_MIN_SIZE * 16) /** * DOC: GSP message queue element From 1bb9bb50a493ad349262edbd81e3f3c583e28122 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:51 -0800 Subject: [PATCH 0194/1443] drm/nouveau: remove the magic number in r535_gsp_rpc_push() There has been a GSP_MSG_MAX_SIZE which represents the max size of a GSP message element header. Use it instead of a magic number. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-9-zhiw@nvidia.com --- drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 4e1136999678..78b5dc66f9e0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -796,7 +796,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, { struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc); struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); - const u32 max_msg_size = (16 * 0x1000) - sizeof(struct r535_gsp_msg); + const u32 max_msg_size = GSP_MSG_MAX_SIZE - sizeof(*cmd); const u32 max_rpc_size = max_msg_size - sizeof(*rpc); u32 rpc_size = rpc->length - sizeof(*rpc); void *repv; From 0268040b9c570c8fc328d3e0ea60da2e068fb76b Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:52 -0800 Subject: [PATCH 0195/1443] drm/nouveau: refine the variable names in r535_gsp_rpc_push() The variable names in r535_gsp_rpc_push() are quite confusing and some of them are not representing what they really are. Update the names and explanations in the decoder section of the kernel doc. Refine the names to align with the terms in the kernel doc. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-10-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 27 ++++++++++--------- 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 78b5dc66f9e0..c28bb1b4f00f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -108,6 +108,9 @@ extern struct dentry *nouveau_debugfs_root; * * Terminology: * + * - gsp_msg(msg): GSP message element (element header + GSP RPC header + + * payload) + * - gsp_rpc(rpc): GSP RPC (RPC header + payload) * - gsp_rpc_len: size of (GSP RPC header + payload) * - params_size: size of params in the payload * - payload_size: size of (header if exists + params) in the payload @@ -795,30 +798,30 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, u32 gsp_rpc_len) { struct nvfw_gsp_rpc *rpc = to_gsp_hdr(payload, rpc); - struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); - const u32 max_msg_size = GSP_MSG_MAX_SIZE - sizeof(*cmd); - const u32 max_rpc_size = max_msg_size - sizeof(*rpc); - u32 rpc_size = rpc->length - sizeof(*rpc); + struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg); + const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*msg); + const u32 max_payload_size = max_rpc_size - sizeof(*rpc); + u32 payload_size = rpc->length - sizeof(*rpc); void *repv; mutex_lock(&gsp->cmdq.mutex); - if (rpc_size > max_rpc_size) { + if (payload_size > max_payload_size) { const u32 fn = rpc->function; /* Adjust length, and send initial RPC. */ - rpc->length = sizeof(*rpc) + max_rpc_size; - cmd->checksum = rpc->length; + rpc->length = sizeof(*rpc) + max_payload_size; + msg->checksum = rpc->length; repv = r535_gsp_rpc_send(gsp, payload, false, 0); if (IS_ERR(repv)) goto done; - payload += max_rpc_size; - rpc_size -= max_rpc_size; + payload += max_payload_size; + payload_size -= max_payload_size; /* Remaining chunks sent as CONTINUATION_RECORD RPCs. */ - while (rpc_size) { - u32 size = min(rpc_size, max_rpc_size); + while (payload_size) { + u32 size = min(payload_size, max_payload_size); void *next; next = r535_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD, size); @@ -834,7 +837,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, goto done; payload += size; - rpc_size -= size; + payload_size -= size; } /* Wait for reply. */ From 462445045269b2d31966b31ae5a73519d634cf16 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:53 -0800 Subject: [PATCH 0196/1443] drm/nouveau: refine the variable names in r535_gsp_msg_recv() The variable "msg" in r535_gsp_msg_recv() actually means the GSP RPC. Refine the names to align with the terms in the kernel doc. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-11-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 47 ++++++++++--------- 1 file changed, 24 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index c28bb1b4f00f..85cd9b6ca0d8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -336,59 +336,60 @@ static struct nvfw_gsp_rpc * r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len) { struct nvkm_subdev *subdev = &gsp->subdev; - struct nvfw_gsp_rpc *msg; + struct nvfw_gsp_rpc *rpc; int time = 4000000, i; u32 size; retry: - msg = r535_gsp_msgq_wait(gsp, sizeof(*msg), &size, &time); - if (IS_ERR_OR_NULL(msg)) - return msg; + rpc = r535_gsp_msgq_wait(gsp, sizeof(*rpc), &size, &time); + if (IS_ERR_OR_NULL(rpc)) + return rpc; - msg = r535_gsp_msgq_recv(gsp, msg->length, &time); - if (IS_ERR_OR_NULL(msg)) - return msg; + rpc = r535_gsp_msgq_recv(gsp, rpc->length, &time); + if (IS_ERR_OR_NULL(rpc)) + return rpc; - if (msg->rpc_result) { - r535_gsp_msg_dump(gsp, msg, NV_DBG_ERROR); - r535_gsp_msg_done(gsp, msg); + if (rpc->rpc_result) { + r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR); + r535_gsp_msg_done(gsp, rpc); return ERR_PTR(-EINVAL); } - r535_gsp_msg_dump(gsp, msg, NV_DBG_TRACE); + r535_gsp_msg_dump(gsp, rpc, NV_DBG_TRACE); - if (fn && msg->function == fn) { + if (fn && rpc->function == fn) { if (gsp_rpc_len) { - if (msg->length < sizeof(*msg) + gsp_rpc_len) { - nvkm_error(subdev, "msg len %d < %zd\n", - msg->length, sizeof(*msg) + + if (rpc->length < sizeof(*rpc) + gsp_rpc_len) { + nvkm_error(subdev, "rpc len %d < %zd\n", + rpc->length, sizeof(*rpc) + gsp_rpc_len); - r535_gsp_msg_dump(gsp, msg, NV_DBG_ERROR); - r535_gsp_msg_done(gsp, msg); + r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR); + r535_gsp_msg_done(gsp, rpc); return ERR_PTR(-EIO); } - return msg; + return rpc; } - r535_gsp_msg_done(gsp, msg); + r535_gsp_msg_done(gsp, rpc); return NULL; } for (i = 0; i < gsp->msgq.ntfy_nr; i++) { struct nvkm_gsp_msgq_ntfy *ntfy = &gsp->msgq.ntfy[i]; - if (ntfy->fn == msg->function) { + if (ntfy->fn == rpc->function) { if (ntfy->func) - ntfy->func(ntfy->priv, ntfy->fn, msg->data, msg->length - sizeof(*msg)); + ntfy->func(ntfy->priv, ntfy->fn, rpc->data, + rpc->length - sizeof(*rpc)); break; } } if (i == gsp->msgq.ntfy_nr) - r535_gsp_msg_dump(gsp, msg, NV_DBG_WARN); + r535_gsp_msg_dump(gsp, rpc, NV_DBG_WARN); - r535_gsp_msg_done(gsp, msg); + r535_gsp_msg_done(gsp, rpc); if (fn) goto retry; From 1829ee0b05177e4274cb169358f8bfea606334ea Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:54 -0800 Subject: [PATCH 0197/1443] drm/nouveau: rename the variable "cmd" to "msg" in r535_gsp_cmdq_{get, push}() Refine the name to align with the terms in the kernel doc. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-12-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 32 +++++++++---------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 85cd9b6ca0d8..ee5b471fd787 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -218,10 +218,10 @@ r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime) static int r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) { - struct r535_gsp_msg *cmd = to_gsp_hdr(rpc, cmd); + struct r535_gsp_msg *msg = to_gsp_hdr(rpc, msg); struct r535_gsp_msg *cqe; - u32 gsp_rpc_len = cmd->checksum; - u64 *ptr = (void *)cmd; + u32 gsp_rpc_len = msg->checksum; + u64 *ptr = (void *)msg; u64 *end; u64 csum = 0; int free, time = 1000000; @@ -231,15 +231,15 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) len = ALIGN(GSP_MSG_HDR_SIZE + gsp_rpc_len, GSP_PAGE_SIZE); end = (u64 *)((char *)ptr + len); - cmd->pad = 0; - cmd->checksum = 0; - cmd->sequence = gsp->cmdq.seq++; - cmd->elem_count = DIV_ROUND_UP(len, 0x1000); + msg->pad = 0; + msg->checksum = 0; + msg->sequence = gsp->cmdq.seq++; + msg->elem_count = DIV_ROUND_UP(len, 0x1000); while (ptr < end) csum ^= *ptr++; - cmd->checksum = upper_32_bits(csum) ^ lower_32_bits(csum); + msg->checksum = upper_32_bits(csum) ^ lower_32_bits(csum); wptr = *gsp->cmdq.wptr; do { @@ -254,7 +254,7 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) } while(--time); if (WARN_ON(!time)) { - kvfree(cmd); + kvfree(msg); return -ETIMEDOUT; } @@ -262,7 +262,7 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) step = min_t(u32, free, (gsp->cmdq.cnt - wptr)); size = min_t(u32, len, step * GSP_PAGE_SIZE); - memcpy(cqe, (u8 *)cmd + off, size); + memcpy(cqe, (u8 *)msg + off, size); wptr += DIV_ROUND_UP(size, 0x1000); if (wptr == gsp->cmdq.cnt) @@ -279,23 +279,23 @@ r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) nvkm_falcon_wr32(&gsp->falcon, 0xc00, 0x00000000); - kvfree(cmd); + kvfree(msg); return 0; } static void * r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 gsp_rpc_len) { - struct r535_gsp_msg *cmd; + struct r535_gsp_msg *msg; u32 size = GSP_MSG_HDR_SIZE + gsp_rpc_len; size = ALIGN(size, GSP_MSG_MIN_SIZE); - cmd = kvzalloc(size, GFP_KERNEL); - if (!cmd) + msg = kvzalloc(size, GFP_KERNEL); + if (!msg) return ERR_PTR(-ENOMEM); - cmd->checksum = gsp_rpc_len; - return cmd->data; + msg->checksum = gsp_rpc_len; + return msg->data; } struct nvfw_gsp_rpc { From c965e3598b4ee7863d64dd13a54258753f7b447f Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:55 -0800 Subject: [PATCH 0198/1443] drm/nouveau: factor out r535_gsp_msgq_peek() To receive a GSP message queue element from the GSP status queue, the driver needs to make sure there are available elements in the queue. The previous r535_gsp_msgq_wait() consists of three functions, which is a little too complicated for a single function: - wait for an available element. - peek the message element header in the queue. - recevice the element from the queue. Factor out r535_gsp_msgq_peek() and divide the functions in r535_gsp_msgq_wait() into three functions. No functional change is intended. Cc: Danilo Krummrich Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-13-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 89 ++++++++++++++----- 1 file changed, 66 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index ee5b471fd787..0d582ddc2c58 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -148,20 +148,16 @@ r535_rpc_status_to_errno(uint32_t rpc_status) } } -static void * -r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, u32 *prepc, - int *ptime) +static int +r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime) { - struct r535_gsp_msg *mqe; u32 size, rptr = *gsp->msgq.rptr; int used; - u8 *msg; - u32 len; size = DIV_ROUND_UP(GSP_MSG_HDR_SIZE + gsp_rpc_len, GSP_PAGE_SIZE); if (WARN_ON(!size || size >= gsp->msgq.cnt)) - return ERR_PTR(-EINVAL); + return -EINVAL; do { u32 wptr = *gsp->msgq.wptr; @@ -176,15 +172,69 @@ r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, u32 *prepc, } while (--(*ptime)); if (WARN_ON(!*ptime)) - return ERR_PTR(-ETIMEDOUT); + return -ETIMEDOUT; - mqe = (void *)((u8 *)gsp->shm.msgq.ptr + 0x1000 + rptr * 0x1000); + return used; +} - if (prepc) { - *prepc = (used * GSP_PAGE_SIZE) - sizeof(*mqe); - return mqe->data; - } +static struct r535_gsp_msg * +r535_gsp_msgq_get_entry(struct nvkm_gsp *gsp) +{ + u32 rptr = *gsp->msgq.rptr; + /* Skip the first page, which is the message queue info */ + return (void *)((u8 *)gsp->shm.msgq.ptr + GSP_PAGE_SIZE + + rptr * GSP_PAGE_SIZE); +} + +/** + * DOC: Receive a GSP message queue element + * + * Receiving a GSP message queue element from the message queue consists of + * the following steps: + * + * - Peek the element from the queue: r535_gsp_msgq_peek(). + * Peek the first page of the element to determine the total size of the + * message before allocating the proper memory. + * + * - Allocate memory and receive the message: r535_gsp_msgq_recv(). + * Once the total size of the message is determined from the GSP message + * queue element, allocate memory and copy the pages of the message + * into the allocated memory. + * + * - Free the allocated memory after processing the GSP message. + * The caller is responsible for freeing the memory allocated for the GSP + * message pages after they have been processed. + */ +static void * +r535_gsp_msgq_peek(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) +{ + struct r535_gsp_msg *mqe; + int ret; + + ret = r535_gsp_msgq_wait(gsp, gsp_rpc_len, retries); + if (ret < 0) + return ERR_PTR(ret); + + mqe = r535_gsp_msgq_get_entry(gsp); + + return mqe->data; +} + +static void * +r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) +{ + u32 rptr = *gsp->msgq.rptr; + struct r535_gsp_msg *mqe; + u32 size, len; + u8 *msg; + int ret; + + ret = r535_gsp_msgq_wait(gsp, gsp_rpc_len, retries); + if (ret < 0) + return ERR_PTR(ret); + + mqe = r535_gsp_msgq_get_entry(gsp); size = ALIGN(gsp_rpc_len + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); msg = kvmalloc(gsp_rpc_len, GFP_KERNEL); @@ -209,12 +259,6 @@ r535_gsp_msgq_wait(struct nvkm_gsp *gsp, u32 gsp_rpc_len, u32 *prepc, return msg; } -static void * -r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *ptime) -{ - return r535_gsp_msgq_wait(gsp, gsp_rpc_len, NULL, ptime); -} - static int r535_gsp_cmdq_push(struct nvkm_gsp *gsp, void *rpc) { @@ -337,15 +381,14 @@ r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len) { struct nvkm_subdev *subdev = &gsp->subdev; struct nvfw_gsp_rpc *rpc; - int time = 4000000, i; - u32 size; + int retries = 4000000, i; retry: - rpc = r535_gsp_msgq_wait(gsp, sizeof(*rpc), &size, &time); + rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), &retries); if (IS_ERR_OR_NULL(rpc)) return rpc; - rpc = r535_gsp_msgq_recv(gsp, rpc->length, &time); + rpc = r535_gsp_msgq_recv(gsp, rpc->length, &retries); if (IS_ERR_OR_NULL(rpc)) return rpc; From 3c48ecb38a736bf457233dc6869e305aee6d52f9 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:56 -0800 Subject: [PATCH 0199/1443] drm/nouveau: factor out r535_gsp_msgq_recv_one_elem() Prepare for supporting receive the large GSP RPC message. Factor out r535_gsp_msgq_recv_one_elem(). Fold its params into a data structure of params. Move the allocation of the GSP RPC message to its caller. Refine the variable names in the re-factor. No functional change is intended. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-14-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 72 +++++++++++++------ 1 file changed, 52 insertions(+), 20 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index 0d582ddc2c58..c074679605fc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -111,6 +111,7 @@ extern struct dentry *nouveau_debugfs_root; * - gsp_msg(msg): GSP message element (element header + GSP RPC header + * payload) * - gsp_rpc(rpc): GSP RPC (RPC header + payload) + * - gsp_rpc_buf: buffer for (GSP RPC header + payload) * - gsp_rpc_len: size of (GSP RPC header + payload) * - params_size: size of params in the payload * - payload_size: size of (header if exists + params) in the payload @@ -197,13 +198,16 @@ r535_gsp_msgq_get_entry(struct nvkm_gsp *gsp) * Peek the first page of the element to determine the total size of the * message before allocating the proper memory. * - * - Allocate memory and receive the message: r535_gsp_msgq_recv(). + * - Allocate memory for the message. * Once the total size of the message is determined from the GSP message - * queue element, allocate memory and copy the pages of the message - * into the allocated memory. + * queue element, the caller of r535_gsp_msgq_recv() allocates the + * required memory. * - * - Free the allocated memory after processing the GSP message. - * The caller is responsible for freeing the memory allocated for the GSP + * - Receive the message: r535_gsp_msgq_recv(). + * Copy the message into the allocated memory. Advance the read pointer. + * + * - Free the allocated memory: r535_gsp_msg_done(). + * The user is responsible for freeing the memory allocated for the GSP * message pages after they have been processed. */ static void * @@ -221,42 +225,70 @@ r535_gsp_msgq_peek(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) return mqe->data; } +struct r535_gsp_msg_info { + int *retries; + u32 gsp_rpc_len; + void *gsp_rpc_buf; +}; + static void * -r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) +r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp, + struct r535_gsp_msg_info *info) { + u8 *buf = info->gsp_rpc_buf; u32 rptr = *gsp->msgq.rptr; struct r535_gsp_msg *mqe; - u32 size, len; - u8 *msg; + u32 size, expected, len; int ret; - ret = r535_gsp_msgq_wait(gsp, gsp_rpc_len, retries); + expected = info->gsp_rpc_len; + + ret = r535_gsp_msgq_wait(gsp, expected, info->retries); if (ret < 0) return ERR_PTR(ret); mqe = r535_gsp_msgq_get_entry(gsp); - size = ALIGN(gsp_rpc_len + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); - - msg = kvmalloc(gsp_rpc_len, GFP_KERNEL); - if (!msg) - return ERR_PTR(-ENOMEM); + size = ALIGN(expected + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); len = ((gsp->msgq.cnt - rptr) * GSP_PAGE_SIZE) - sizeof(*mqe); - len = min_t(u32, gsp_rpc_len, len); - memcpy(msg, mqe->data, len); + len = min_t(u32, expected, len); + memcpy(buf, mqe->data, len); - gsp_rpc_len -= len; + expected -= len; - if (gsp_rpc_len) { + if (expected) { mqe = (void *)((u8 *)gsp->shm.msgq.ptr + 0x1000 + 0 * 0x1000); - memcpy(msg + len, mqe, gsp_rpc_len); + memcpy(buf + len, mqe, expected); } rptr = (rptr + DIV_ROUND_UP(size, GSP_PAGE_SIZE)) % gsp->msgq.cnt; mb(); (*gsp->msgq.rptr) = rptr; - return msg; + return buf; +} + +static void * +r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) +{ + struct r535_gsp_msg_info info = {0}; + void *buf; + + buf = kvmalloc(gsp_rpc_len, GFP_KERNEL); + if (!buf) + return ERR_PTR(-ENOMEM); + + info.gsp_rpc_buf = buf; + info.retries = retries; + info.gsp_rpc_len = gsp_rpc_len; + + buf = r535_gsp_msgq_recv_one_elem(gsp, &info); + if (IS_ERR(buf)) { + kvfree(info.gsp_rpc_buf); + info.gsp_rpc_buf = NULL; + } + + return buf; } static int From 50f290053d79e3b1d108f181c0ba6b8e30ca94c9 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:57 -0800 Subject: [PATCH 0200/1443] drm/nouveau: support handling the return of large GSP message The max GSP message element size is 16 pages (including the headers). To send a message larger than 16 pages, nvkm should split it into multiple and send them accordingly. The first element has the expected function number, while the rest are sent with function number as NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD. GSP consumes the elements from the cmdq and always writes the result back to the msgq. The result is also formed as split elements. However, nvkm is able to split the large GSP message and send them, but totally not aware of handling the return of the large GSP message, which are the split elements in the msgq. Thus, it keeps dumping the unknown RPC messages from msgq, which is actually CONTINUATION_RECORD message, discard them unexpectedly. Thus, the caller will not be able to consume the result from GSP. Introduce the handling of the return of large GSP message on the msgq path. Slightly re-factor the low-level part of msg receiving routines. Merge the split elements back into a large element before handling it to the upper level. Thus, the upper-level of GSP RPC APIs don't need to be heavily changed. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-15-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 116 +++++++++++++++--- 1 file changed, 97 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index c074679605fc..b3f8d33fdd60 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -106,6 +106,17 @@ extern struct dentry *nouveau_debugfs_root; * | params | * +------------------------+ * + * The max size of a message queue element is 16 pages (including the + * headers). When a GSP message to be sent is larger than 16 pages, the + * message should be split into multiple elements and sent accordingly. + * + * In the bunch of the split elements, the first element has the expected + * function number, while the rest of the elements are sent with the + * function number NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD. + * + * GSP consumes the elements from the cmdq and always writes the result + * back to the msgq. The result is also formed as split elements. + * * Terminology: * * - gsp_msg(msg): GSP message element (element header + GSP RPC header + @@ -127,6 +138,21 @@ struct r535_gsp_msg { u8 data[]; }; +struct nvfw_gsp_rpc { + u32 header_version; + u32 signature; + u32 length; + u32 function; + u32 rpc_result; + u32 rpc_result_private; + u32 sequence; + union { + u32 spare; + u32 cpuRmGfid; + }; + u8 data[]; +}; + #define GSP_MSG_HDR_SIZE offsetof(struct r535_gsp_msg, data) #define to_gsp_hdr(p, header) \ @@ -205,6 +231,11 @@ r535_gsp_msgq_get_entry(struct nvkm_gsp *gsp) * * - Receive the message: r535_gsp_msgq_recv(). * Copy the message into the allocated memory. Advance the read pointer. + * If the message is a large GSP message, r535_gsp_msgq_recv() calls + * r535_gsp_msgq_recv_one_elem() repeatedly to receive continuation parts + * until the complete message is received. + * r535_gsp_msgq_recv() assembles the payloads of cotinuation parts into + * the return of the large GSP message. * * - Free the allocated memory: r535_gsp_msg_done(). * The user is responsible for freeing the memory allocated for the GSP @@ -229,8 +260,12 @@ struct r535_gsp_msg_info { int *retries; u32 gsp_rpc_len; void *gsp_rpc_buf; + bool continuation; }; +static void +r535_gsp_msg_dump(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg, int lvl); + static void * r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp, struct r535_gsp_msg_info *info) @@ -248,11 +283,28 @@ r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp, return ERR_PTR(ret); mqe = r535_gsp_msgq_get_entry(gsp); + + if (info->continuation) { + struct nvfw_gsp_rpc *rpc = (struct nvfw_gsp_rpc *)mqe->data; + + if (rpc->function != NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD) { + nvkm_error(&gsp->subdev, + "Not a continuation of a large RPC\n"); + r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR); + return ERR_PTR(-EIO); + } + } + size = ALIGN(expected + GSP_MSG_HDR_SIZE, GSP_PAGE_SIZE); len = ((gsp->msgq.cnt - rptr) * GSP_PAGE_SIZE) - sizeof(*mqe); len = min_t(u32, expected, len); - memcpy(buf, mqe->data, len); + + if (info->continuation) + memcpy(buf, mqe->data + sizeof(struct nvfw_gsp_rpc), + len - sizeof(struct nvfw_gsp_rpc)); + else + memcpy(buf, mqe->data, len); expected -= len; @@ -271,16 +323,26 @@ r535_gsp_msgq_recv_one_elem(struct nvkm_gsp *gsp, static void * r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) { + struct r535_gsp_msg *mqe; + const u32 max_rpc_size = GSP_MSG_MAX_SIZE - sizeof(*mqe); + struct nvfw_gsp_rpc *rpc; struct r535_gsp_msg_info info = {0}; + u32 expected = gsp_rpc_len; void *buf; - buf = kvmalloc(gsp_rpc_len, GFP_KERNEL); + mqe = r535_gsp_msgq_get_entry(gsp); + rpc = (struct nvfw_gsp_rpc *)mqe->data; + + if (WARN_ON(rpc->length > max_rpc_size)) + return NULL; + + buf = kvmalloc(max_t(u32, rpc->length, expected), GFP_KERNEL); if (!buf) return ERR_PTR(-ENOMEM); info.gsp_rpc_buf = buf; info.retries = retries; - info.gsp_rpc_len = gsp_rpc_len; + info.gsp_rpc_len = rpc->length; buf = r535_gsp_msgq_recv_one_elem(gsp, &info); if (IS_ERR(buf)) { @@ -288,6 +350,37 @@ r535_gsp_msgq_recv(struct nvkm_gsp *gsp, u32 gsp_rpc_len, int *retries) info.gsp_rpc_buf = NULL; } + if (expected <= max_rpc_size) + return buf; + + info.gsp_rpc_buf += info.gsp_rpc_len; + expected -= info.gsp_rpc_len; + + while (expected) { + u32 size; + + rpc = r535_gsp_msgq_peek(gsp, sizeof(*rpc), info.retries); + if (IS_ERR_OR_NULL(rpc)) { + kfree(buf); + return rpc; + } + + info.gsp_rpc_len = rpc->length; + info.continuation = true; + + rpc = r535_gsp_msgq_recv_one_elem(gsp, &info); + if (IS_ERR_OR_NULL(rpc)) { + kfree(buf); + return rpc; + } + + size = info.gsp_rpc_len - sizeof(*rpc); + expected -= size; + info.gsp_rpc_buf += size; + } + + rpc = buf; + rpc->length = gsp_rpc_len; return buf; } @@ -374,21 +467,6 @@ r535_gsp_cmdq_get(struct nvkm_gsp *gsp, u32 gsp_rpc_len) return msg->data; } -struct nvfw_gsp_rpc { - u32 header_version; - u32 signature; - u32 length; - u32 function; - u32 rpc_result; - u32 rpc_result_private; - u32 sequence; - union { - u32 spare; - u32 cpuRmGfid; - }; - u8 data[]; -}; - static void r535_gsp_msg_done(struct nvkm_gsp *gsp, struct nvfw_gsp_rpc *msg) { @@ -420,7 +498,7 @@ r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len) if (IS_ERR_OR_NULL(rpc)) return rpc; - rpc = r535_gsp_msgq_recv(gsp, rpc->length, &retries); + rpc = r535_gsp_msgq_recv(gsp, gsp_rpc_len, &retries); if (IS_ERR_OR_NULL(rpc)) return rpc; From 24079ed2aad6567f643df0e31b631184578de4e7 Mon Sep 17 00:00:00 2001 From: Zhi Wang Date: Fri, 24 Jan 2025 10:29:58 -0800 Subject: [PATCH 0201/1443] drm/nouveau: consume the return of large GSP message As the GSP message recv path is able to handle the return of large GSP message, consume the return of large GSP message in the sending path. Signed-off-by: Zhi Wang Signed-off-by: Danilo Krummrich Link: https://patchwork.freedesktop.org/patch/msgid/20250124182958.2040494-16-zhiw@nvidia.com --- .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 32 +++++++++++-------- 1 file changed, 18 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c index b3f8d33fdd60..2075cad63805 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c @@ -512,10 +512,9 @@ r535_gsp_msg_recv(struct nvkm_gsp *gsp, int fn, u32 gsp_rpc_len) if (fn && rpc->function == fn) { if (gsp_rpc_len) { - if (rpc->length < sizeof(*rpc) + gsp_rpc_len) { - nvkm_error(subdev, "rpc len %d < %zd\n", - rpc->length, sizeof(*rpc) + - gsp_rpc_len); + if (rpc->length < gsp_rpc_len) { + nvkm_error(subdev, "rpc len %d < %d\n", + rpc->length, gsp_rpc_len); r535_gsp_msg_dump(gsp, rpc, NV_DBG_ERROR); r535_gsp_msg_done(gsp, rpc); return ERR_PTR(-EIO); @@ -961,6 +960,7 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, mutex_lock(&gsp->cmdq.mutex); if (payload_size > max_payload_size) { const u32 fn = rpc->function; + u32 remain_payload_size = payload_size; /* Adjust length, and send initial RPC. */ rpc->length = sizeof(*rpc) + max_payload_size; @@ -971,11 +971,12 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, goto done; payload += max_payload_size; - payload_size -= max_payload_size; + remain_payload_size -= max_payload_size; /* Remaining chunks sent as CONTINUATION_RECORD RPCs. */ - while (payload_size) { - u32 size = min(payload_size, max_payload_size); + while (remain_payload_size) { + u32 size = min(remain_payload_size, + max_payload_size); void *next; next = r535_gsp_rpc_get(gsp, NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD, size); @@ -991,18 +992,21 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload, bool wait, goto done; payload += size; - payload_size -= size; + remain_payload_size -= size; } /* Wait for reply. */ - if (wait) { - rpc = r535_gsp_msg_recv(gsp, fn, gsp_rpc_len); - if (!IS_ERR_OR_NULL(rpc)) + rpc = r535_gsp_msg_recv(gsp, fn, payload_size + + sizeof(*rpc)); + if (!IS_ERR_OR_NULL(rpc)) { + if (wait) { repv = rpc->data; - else - repv = rpc; + } else { + nvkm_gsp_rpc_done(gsp, rpc); + repv = NULL; + } } else { - repv = NULL; + repv = wait ? rpc : NULL; } } else { repv = r535_gsp_rpc_send(gsp, payload, wait, gsp_rpc_len); From e0f63bc68f59d281e2d06e596f6c1bd9382a15cd Mon Sep 17 00:00:00 2001 From: Gustavo Sousa Date: Tue, 21 Jan 2025 18:09:25 -0300 Subject: [PATCH 0202/1443] drm/print: Include drm_device.h The header drm_print.h uses members of struct drm_device pointers, as such, it should include drm_device.h to let the compiler know the full type definition. Without such include, users of drm_print.h that don't explicitly need drm_device.h would bump into build errors and be forced to include the latter. Signed-off-by: Gustavo Sousa Reviewed-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20250121210935.84357-1-gustavo.sousa@intel.com Signed-off-by: Lucas De Marchi --- include/drm/drm_print.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/drm/drm_print.h b/include/drm/drm_print.h index b3906dc04388..8d3e17d7554a 100644 --- a/include/drm/drm_print.h +++ b/include/drm/drm_print.h @@ -32,6 +32,7 @@ #include #include +#include struct debugfs_regset32; struct drm_device; From 41f198d58b6f2b36f9f8a4481d517369b324e773 Mon Sep 17 00:00:00 2001 From: Tanya Agarwal Date: Fri, 24 Jan 2025 01:18:00 +0530 Subject: [PATCH 0203/1443] tomoyo: fix spelling error Fix spelling error in security/tomoyo module comments that were identified using the codespell tool. No functional changes - documentation only. Signed-off-by: Tanya Agarwal Reviewed-by: Mimi Zohar Signed-off-by: Tetsuo Handa --- security/tomoyo/domain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/security/tomoyo/domain.c b/security/tomoyo/domain.c index 3a7b0874cf44..5f9ccab26e9a 100644 --- a/security/tomoyo/domain.c +++ b/security/tomoyo/domain.c @@ -920,7 +920,7 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos, #ifdef CONFIG_MMU /* * This is called at execve() time in order to dig around - * in the argv/environment of the new proceess + * in the argv/environment of the new process * (represented by bprm). */ mmap_read_lock(bprm->mm); From 691a1f3f180133965d01e0ab0f332248d0345554 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Sun, 26 Jan 2025 19:13:58 +0900 Subject: [PATCH 0204/1443] tomoyo: fix spelling errors No functional changes. Signed-off-by: Tetsuo Handa --- security/tomoyo/securityfs_if.c | 6 +++--- security/tomoyo/tomoyo.c | 5 +---- 2 files changed, 4 insertions(+), 7 deletions(-) diff --git a/security/tomoyo/securityfs_if.c b/security/tomoyo/securityfs_if.c index a2705798476f..7e69747b2f77 100644 --- a/security/tomoyo/securityfs_if.c +++ b/security/tomoyo/securityfs_if.c @@ -229,11 +229,11 @@ static void __init tomoyo_create_entry(const char *name, const umode_t mode, } /** - * tomoyo_initerface_init - Initialize /sys/kernel/security/tomoyo/ interface. + * tomoyo_interface_init - Initialize /sys/kernel/security/tomoyo/ interface. * * Returns 0. */ -static int __init tomoyo_initerface_init(void) +static int __init tomoyo_interface_init(void) { struct tomoyo_domain_info *domain; struct dentry *tomoyo_dir; @@ -270,4 +270,4 @@ static int __init tomoyo_initerface_init(void) return 0; } -fs_initcall(tomoyo_initerface_init); +fs_initcall(tomoyo_interface_init); diff --git a/security/tomoyo/tomoyo.c b/security/tomoyo/tomoyo.c index 04a92c3d65d4..d6ebcd9db80a 100644 --- a/security/tomoyo/tomoyo.c +++ b/security/tomoyo/tomoyo.c @@ -549,10 +549,7 @@ static const struct lsm_id tomoyo_lsmid = { .id = LSM_ID_TOMOYO, }; -/* - * tomoyo_security_ops is a "struct security_operations" which is used for - * registering TOMOYO. - */ +/* tomoyo_hooks is used for registering TOMOYO. */ static struct security_hook_list tomoyo_hooks[] __ro_after_init = { LSM_HOOK_INIT(cred_prepare, tomoyo_cred_prepare), LSM_HOOK_INIT(bprm_committed_creds, tomoyo_bprm_committed_creds), From c852646f12d4cd5b4f19eeec2976c5d98c0382f8 Mon Sep 17 00:00:00 2001 From: Olivier Moysan Date: Wed, 8 Jan 2025 18:03:54 +0100 Subject: [PATCH 0205/1443] drm: bridge: adv7511: fill stream capabilities Set no_i2s_capture and no_spdif_capture flags in hdmi_codec_pdata structure to report that the ADV7511 HDMI bridge does not support i2s or spdif audio capture. Signed-off-by: Olivier Moysan Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20250108170356.413063-2-olivier.moysan@foss.st.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/bridge/adv7511/adv7511_audio.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c index 51fb9a574b4e..ae853f6485b6 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c @@ -235,7 +235,9 @@ static const struct hdmi_codec_pdata codec_data = { .ops = &adv7511_codec_ops, .max_i2s_channels = 2, .i2s = 1, + .no_i2s_capture = 1, .spdif = 1, + .no_spdif_capture = 1, }; int adv7511_audio_init(struct device *dev, struct adv7511 *adv7511) From 038f79638e0676359e44c5db458d52994f9b5ac1 Mon Sep 17 00:00:00 2001 From: Olivier Moysan Date: Wed, 8 Jan 2025 18:03:55 +0100 Subject: [PATCH 0206/1443] ASoC: hdmi-codec: allow to refine formats actually supported Currently the hdmi-codec driver registers all the formats that are allowed on the I2S bus. Add i2s_formats field to codec data, to allow the hdmi codec client to refine the list of the audio I2S formats actually supported. Signed-off-by: Olivier Moysan Acked-by: Mark Brown Link: https://patchwork.freedesktop.org/patch/msgid/20250108170356.413063-3-olivier.moysan@foss.st.com Signed-off-by: Dmitry Baryshkov --- include/sound/hdmi-codec.h | 1 + sound/soc/codecs/hdmi-codec.c | 4 ++++ 2 files changed, 5 insertions(+) diff --git a/include/sound/hdmi-codec.h b/include/sound/hdmi-codec.h index b220072cfa1b..273f4c36fad9 100644 --- a/include/sound/hdmi-codec.h +++ b/include/sound/hdmi-codec.h @@ -120,6 +120,7 @@ struct hdmi_codec_ops { /* HDMI codec initalization data */ struct hdmi_codec_pdata { const struct hdmi_codec_ops *ops; + u64 i2s_formats; uint i2s:1; uint no_i2s_playback:1; uint no_i2s_capture:1; diff --git a/sound/soc/codecs/hdmi-codec.c b/sound/soc/codecs/hdmi-codec.c index 69f98975e14a..5c47aa0551c9 100644 --- a/sound/soc/codecs/hdmi-codec.c +++ b/sound/soc/codecs/hdmi-codec.c @@ -1077,6 +1077,10 @@ static int hdmi_codec_probe(struct platform_device *pdev) if (hcd->i2s) { daidrv[i] = hdmi_i2s_dai; daidrv[i].playback.channels_max = hcd->max_i2s_channels; + if (hcd->i2s_formats) { + daidrv[i].playback.formats = hcd->i2s_formats; + daidrv[i].capture.formats = hcd->i2s_formats; + } if (hcd->no_i2s_playback) memset(&daidrv[i].playback, 0, sizeof(daidrv[i].playback)); From 7c3127e4814c304c7d5a2fa796a42e75a9b8606f Mon Sep 17 00:00:00 2001 From: Olivier Moysan Date: Wed, 8 Jan 2025 18:03:56 +0100 Subject: [PATCH 0207/1443] drm: bridge: adv7511: remove s32 format from i2s capabilities The ADV7511 chip allows 24 bits samples max in I2S mode, excepted for direct AES3 mode (SNDRV_PCM_FORMAT_IEC958_SUBFRAME_LE format). However the HDMI codec exposes S32_LE format as supported. Adapt ADV7511 HDMI I2S format list to expose formats actually supported. Signed-off-by: Olivier Moysan Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20250108170356.413063-4-olivier.moysan@foss.st.com Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/bridge/adv7511/adv7511_audio.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c index ae853f6485b6..36a7b6f085f9 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c @@ -233,6 +233,9 @@ static const struct hdmi_codec_ops adv7511_codec_ops = { static const struct hdmi_codec_pdata codec_data = { .ops = &adv7511_codec_ops, + .i2s_formats = (SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S20_3LE | + SNDRV_PCM_FMTBIT_S24_3LE | SNDRV_PCM_FMTBIT_S24_LE | + SNDRV_PCM_FMTBIT_IEC958_SUBFRAME_LE), .max_i2s_channels = 2, .i2s = 1, .no_i2s_capture = 1, From 1f566840a82982141f94086061927a90e79440e5 Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Fri, 24 Jan 2025 20:54:41 -0500 Subject: [PATCH 0208/1443] clocksource: Use pr_info() for "Checking clocksource synchronization" message The "Checking clocksource synchronization" message is normally printed when clocksource_verify_percpu() is called for a given clocksource if both the CLOCK_SOURCE_UNSTABLE and CLOCK_SOURCE_VERIFY_PERCPU flags are set. It is an informational message and so pr_info() is the correct choice. Signed-off-by: Waiman Long Signed-off-by: Thomas Gleixner Reviewed-by: Paul E. McKenney Acked-by: John Stultz Link: https://lore.kernel.org/all/20250125015442.3740588-1-longman@redhat.com --- kernel/time/clocksource.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 7304d7cf47f2..77d9566d3aa6 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -382,7 +382,8 @@ void clocksource_verify_percpu(struct clocksource *cs) return; } testcpu = smp_processor_id(); - pr_warn("Checking clocksource %s synchronization from CPU %d to CPUs %*pbl.\n", cs->name, testcpu, cpumask_pr_args(&cpus_chosen)); + pr_info("Checking clocksource %s synchronization from CPU %d to CPUs %*pbl.\n", + cs->name, testcpu, cpumask_pr_args(&cpus_chosen)); for_each_cpu(cpu, &cpus_chosen) { if (cpu == testcpu) continue; From 825c78e6a60c309a59d18d5ac5968aa79cef0bd6 Mon Sep 17 00:00:00 2001 From: Xu Lu Date: Mon, 27 Jan 2025 17:38:46 +0800 Subject: [PATCH 0209/1443] irqchip/riscv: Ensure ordering of memory writes and IPI writes RISC-V distinguishes between memory accesses and device I/O and uses FENCE instruction to order them as viewed by other RISC-V harts and external devices or coprocessors. The FENCE instruction can order any combination of device input(I), device output(O), memory reads(R) and memory writes(W). For example, 'fence w, o' is used to ensure all memory writes from instructions preceding the FENCE instruction appear earlier in the global memory order than device output writes from instructions after the FENCE instruction. RISC-V issues IPIs by writing to the IMSIC/ACLINT MMIO registers, which is regarded as device output operation. However, the existing implementation of the IMSIC/ACLINT drivers issue the IPI via writel_relaxed(), which does not guarantee the order of device output operation and preceding memory writes. As a consequence the hart receiving the IPI might not observe the IPI related data. Fix this by replacing writel_relaxed() with writel() when issuing IPIs, which uses 'fence w, o' to ensure all previous writes made by the current hart are visible to other harts before they receive the IPI. Signed-off-by: Xu Lu Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/20250127093846.98625-1-luxu.kernel@bytedance.com --- drivers/irqchip/irq-riscv-imsic-early.c | 2 +- drivers/irqchip/irq-thead-c900-aclint-sswi.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/irqchip/irq-riscv-imsic-early.c b/drivers/irqchip/irq-riscv-imsic-early.c index c5c2e6929a2f..275df5005705 100644 --- a/drivers/irqchip/irq-riscv-imsic-early.c +++ b/drivers/irqchip/irq-riscv-imsic-early.c @@ -27,7 +27,7 @@ static void imsic_ipi_send(unsigned int cpu) { struct imsic_local_config *local = per_cpu_ptr(imsic->global.local, cpu); - writel_relaxed(IMSIC_IPI_ID, local->msi_va); + writel(IMSIC_IPI_ID, local->msi_va); } static void imsic_ipi_starting_cpu(void) diff --git a/drivers/irqchip/irq-thead-c900-aclint-sswi.c b/drivers/irqchip/irq-thead-c900-aclint-sswi.c index b0e366ade427..8ff6e7a1363b 100644 --- a/drivers/irqchip/irq-thead-c900-aclint-sswi.c +++ b/drivers/irqchip/irq-thead-c900-aclint-sswi.c @@ -31,7 +31,7 @@ static DEFINE_PER_CPU(void __iomem *, sswi_cpu_regs); static void thead_aclint_sswi_ipi_send(unsigned int cpu) { - writel_relaxed(0x1, per_cpu(sswi_cpu_regs, cpu)); + writel(0x1, per_cpu(sswi_cpu_regs, cpu)); } static void thead_aclint_sswi_ipi_clear(void) From 987f379b54091cc1b1db986bde71cee1081350b3 Mon Sep 17 00:00:00 2001 From: Stefan Eichenberger Date: Fri, 24 Jan 2025 09:50:39 +0100 Subject: [PATCH 0210/1443] irqchip/irq-mvebu-icu: Fix access to msi_data from irq_domain::host_data mvebu_icu_translate() incorrectly casts irq_domain::host_data directly to mvebu_icu_msi_data. However, host_data actually points to a structure of type msi_domain_info. This incorrect cast causes issues such as the thermal sensors of the CP110 platform malfunctioning. Specifically, the translation of the SEI interrupt to IRQ_TYPE_EDGE_RISING fails, preventing proper interrupt handling. The following error was observed: genirq: Setting trigger mode 4 for irq 85 failed (irq_chip_set_type_parent+0x0/0x34) armada_thermal f2400000.system-controller:thermal-sensor@70: Cannot request threaded IRQ 85 Resolve the issue by first casting host_data to msi_domain_info and then accessing mvebu_icu_msi_data through msi_domain_info::chip_data. Fixes: d929e4db22b6 ("irqchip/irq-mvebu-icu: Prepare for real per device MSI") Signed-off-by: Stefan Eichenberger Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250124085140.44792-1-eichest@gmail.com --- drivers/irqchip/irq-mvebu-icu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-mvebu-icu.c b/drivers/irqchip/irq-mvebu-icu.c index b337f6c05f18..4eebed39880a 100644 --- a/drivers/irqchip/irq-mvebu-icu.c +++ b/drivers/irqchip/irq-mvebu-icu.c @@ -68,7 +68,8 @@ static int mvebu_icu_translate(struct irq_domain *d, struct irq_fwspec *fwspec, unsigned long *hwirq, unsigned int *type) { unsigned int param_count = static_branch_unlikely(&legacy_bindings) ? 3 : 2; - struct mvebu_icu_msi_data *msi_data = d->host_data; + struct msi_domain_info *info = d->host_data; + struct mvebu_icu_msi_data *msi_data = info->chip_data; struct mvebu_icu *icu = msi_data->icu; /* Check the count of the parameters in dt */ From fb95897b8c60653805aa09daec575ca30983f768 Mon Sep 17 00:00:00 2001 From: Wentao Liang Date: Fri, 24 Jan 2025 11:22:28 +0800 Subject: [PATCH 0211/1443] xfs: Propagate errors from xfs_reflink_cancel_cow_range in xfs_dax_write_iomap_end In xfs_dax_write_iomap_end(), directly return the result of xfs_reflink_cancel_cow_range() when !written, ensuring proper error propagation and improving code robustness. Fixes: ea6c49b784f0 ("xfs: support CoW in fsdax mode") Cc: stable@vger.kernel.org # v6.0 Reviewed-by: Darrick J. Wong Signed-off-by: Wentao Liang Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_iomap.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 50fa3ef89f6c..d61460309a78 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -976,10 +976,8 @@ xfs_dax_write_iomap_end( if (!xfs_is_cow_inode(ip)) return 0; - if (!written) { - xfs_reflink_cancel_cow_range(ip, pos, length, true); - return 0; - } + if (!written) + return xfs_reflink_cancel_cow_range(ip, pos, length, true); return xfs_reflink_end_cow(ip, pos, written); } From 28aecef5b1015bf6023ddc12b1a67f6678271fcb Mon Sep 17 00:00:00 2001 From: Madhavan Srinivasan Date: Sun, 19 Jan 2025 22:02:38 +0530 Subject: [PATCH 0212/1443] selftests: livepatch: handle PRINTK_CALLER in check_result() Some arch configs (like ppc64) enable CONFIG_PRINTK_CALLER, which adds the caller id as part of the dmesg. With recent util-linux's update 467a5b3192f16 ('dmesg: add caller_id support') the standard "dmesg" has been enhanced to print PRINTK_CALLER fields. Due to this, even though the expected vs observed are same, end testcase results are failed. -% insmod test_modules/test_klp_livepatch.ko -livepatch: enabling patch 'test_klp_livepatch' -livepatch: 'test_klp_livepatch': initializing patching transition -livepatch: 'test_klp_livepatch': starting patching transition -livepatch: 'test_klp_livepatch': completing patching transition -livepatch: 'test_klp_livepatch': patching complete -% echo 0 > /sys/kernel/livepatch/test_klp_livepatch/enabled -livepatch: 'test_klp_livepatch': initializing unpatching transition -livepatch: 'test_klp_livepatch': starting unpatching transition -livepatch: 'test_klp_livepatch': completing unpatching transition -livepatch: 'test_klp_livepatch': unpatching complete -% rmmod test_klp_livepatch +[ T3659] % insmod test_modules/test_klp_livepatch.ko +[ T3682] livepatch: enabling patch 'test_klp_livepatch' +[ T3682] livepatch: 'test_klp_livepatch': initializing patching transition +[ T3682] livepatch: 'test_klp_livepatch': starting patching transition +[ T826] livepatch: 'test_klp_livepatch': completing patching transition +[ T826] livepatch: 'test_klp_livepatch': patching complete +[ T3659] % echo 0 > /sys/kernel/livepatch/test_klp_livepatch/enabled +[ T3659] livepatch: 'test_klp_livepatch': initializing unpatching transition +[ T3659] livepatch: 'test_klp_livepatch': starting unpatching transition +[ T789] livepatch: 'test_klp_livepatch': completing unpatching transition +[ T789] livepatch: 'test_klp_livepatch': unpatching complete +[ T3659] % rmmod test_klp_livepatch ERROR: livepatch kselftest(s) failed not ok 1 selftests: livepatch: test-livepatch.sh # exit=1 Currently the check_result() handles the "[time]" removal from the dmesg. Enhance the check to also handle removal of "[Thread Id]" or "[CPU Id]". Signed-off-by: Madhavan Srinivasan Acked-by: Miroslav Benes Reviewed-by: Petr Mladek Tested-by: Petr Mladek Link: https://lore.kernel.org/r/20250119163238.749847-1-maddy@linux.ibm.com Signed-off-by: Petr Mladek --- tools/testing/selftests/livepatch/functions.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/livepatch/functions.sh b/tools/testing/selftests/livepatch/functions.sh index e5d06fb40233..15601402dee6 100644 --- a/tools/testing/selftests/livepatch/functions.sh +++ b/tools/testing/selftests/livepatch/functions.sh @@ -306,7 +306,8 @@ function check_result { result=$(dmesg | awk -v last_dmesg="$LAST_DMESG" 'p; $0 == last_dmesg { p=1 }' | \ grep -e 'livepatch:' -e 'test_klp' | \ grep -v '\(tainting\|taints\) kernel' | \ - sed 's/^\[[ 0-9.]*\] //') + sed 's/^\[[ 0-9.]*\] //' | \ + sed 's/^\[[ ]*[CT][0-9]*\] //') if [[ "$expect" == "$result" ]] ; then echo "ok" From 26b63bee2f6e711c5a169997fd126fddcfb90848 Mon Sep 17 00:00:00 2001 From: Wentao Liang Date: Fri, 24 Jan 2025 11:45:09 +0800 Subject: [PATCH 0213/1443] xfs: Add error handling for xfs_reflink_cancel_cow_range In xfs_inactive(), xfs_reflink_cancel_cow_range() is called without error handling, risking unnoticed failures and inconsistent behavior compared to other parts of the code. Fix this issue by adding an error handling for the xfs_reflink_cancel_cow_range(), improving code robustness. Fixes: 6231848c3aa5 ("xfs: check for cow blocks before trying to clear them") Cc: stable@vger.kernel.org # v4.17 Reviewed-by: Darrick J. Wong Signed-off-by: Wentao Liang Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_inode.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index c95fe1b1de4e..b1f9f156ec88 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1404,8 +1404,11 @@ xfs_inactive( goto out; /* Try to clean out the cow blocks if there are any. */ - if (xfs_inode_has_cow_data(ip)) - xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true); + if (xfs_inode_has_cow_data(ip)) { + error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true); + if (error) + goto out; + } if (VFS_I(ip)->i_nlink != 0) { /* From e2a81c0cd7de6cb063058be304b18f200c64802b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Noralf=20Tr=C3=B8nnes?= Date: Thu, 23 Jan 2025 17:34:25 +0100 Subject: [PATCH 0214/1443] =?UTF-8?q?MAINTAINERS:=20Remove=20Noralf=20Tr?= =?UTF-8?q?=C3=B8nnes=20as=20driver=20maintainer?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove myself as maintainer for gud, mi0283qt, panel-mipi-dbi and repaper. My fatigue illness has finally closed the door on doing development of even moderate complexity so it's sad to let this go. Acked-by: Thomas Zimmremann Acked-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250123-remove-myself-as-maintainer-v1-1-cc3ab7cd98ae@tronnes.org Signed-off-by: Noralf Trønnes --- MAINTAINERS | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 4b038382481f..ed86d884ee0d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7173,8 +7173,7 @@ F: Documentation/devicetree/bindings/display/panel/panel-edp.yaml F: drivers/gpu/drm/panel/panel-edp.c DRM DRIVER FOR GENERIC USB DISPLAY -M: Noralf Trønnes -S: Maintained +S: Orphan W: https://github.com/notro/gud/wiki T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: drivers/gpu/drm/gud/ @@ -7279,15 +7278,13 @@ T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: drivers/gpu/drm/mgag200/ DRM DRIVER FOR MI0283QT -M: Noralf Trønnes -S: Maintained +S: Orphan T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/multi-inno,mi0283qt.txt F: drivers/gpu/drm/tiny/mi0283qt.c DRM DRIVER FOR MIPI DBI compatible panels -M: Noralf Trønnes -S: Maintained +S: Orphan W: https://github.com/notro/panel-mipi-dbi/wiki T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/panel/panel-mipi-dbi-spi.yaml @@ -7384,8 +7381,7 @@ F: Documentation/devicetree/bindings/display/bridge/ps8640.yaml F: drivers/gpu/drm/bridge/parade-ps8640.c DRM DRIVER FOR PERVASIVE DISPLAYS REPAPER PANELS -M: Noralf Trønnes -S: Maintained +S: Orphan T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/repaper.txt F: drivers/gpu/drm/tiny/repaper.c From 011c1e246a1d01c9d4b3b183e469bdb87b1f7bb8 Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Thu, 23 Jan 2025 21:04:06 -0800 Subject: [PATCH 0215/1443] drm/xe/pmu: Enable PMU interface Basic PMU enabling patch. Setup the basic framework for adding events. Based on previous versions by Bommu Krishnaiah, Aravind Iddamsetty and Riana Tauro, using i915 and rapl as reference implementations. Reviewed-by: Rodrigo Vivi Signed-off-by: Vinay Belgaumkar Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-1-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/Makefile | 2 + drivers/gpu/drm/xe/xe_device.c | 3 + drivers/gpu/drm/xe/xe_device_types.h | 4 + drivers/gpu/drm/xe/xe_pmu.c | 289 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_pmu.h | 18 ++ drivers/gpu/drm/xe/xe_pmu_types.h | 35 ++++ 6 files changed, 351 insertions(+) create mode 100644 drivers/gpu/drm/xe/xe_pmu.c create mode 100644 drivers/gpu/drm/xe/xe_pmu.h create mode 100644 drivers/gpu/drm/xe/xe_pmu_types.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 2e50a697f549..9122e4318773 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -122,6 +122,8 @@ xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o # graphics hardware monitoring (HWMON) support xe-$(CONFIG_HWMON) += xe_hwmon.o +xe-$(CONFIG_PERF_EVENTS) += xe_pmu.o + # graphics virtualization (SR-IOV) support xe-y += \ xe_gt_sriov_vf.o \ diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index af9b2bb62dee..90597a7ad0b0 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -49,6 +49,7 @@ #include "xe_pat.h" #include "xe_pcode.h" #include "xe_pm.h" +#include "xe_pmu.h" #include "xe_query.h" #include "xe_sriov.h" #include "xe_tile.h" @@ -872,6 +873,8 @@ int xe_device_probe(struct xe_device *xe) xe_oa_register(xe); + xe_pmu_register(&xe->pmu); + xe_debugfs_register(xe); xe_hwmon_register(xe); diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 16ebb2859877..58e79e19deaa 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -18,6 +18,7 @@ #include "xe_memirq_types.h" #include "xe_oa_types.h" #include "xe_platform_types.h" +#include "xe_pmu_types.h" #include "xe_pt_types.h" #include "xe_sriov_types.h" #include "xe_step_types.h" @@ -514,6 +515,9 @@ struct xe_device { int mode; } wedged; + /** @pmu: performance monitoring unit */ + struct xe_pmu pmu; + #ifdef TEST_VM_OPS_ERROR /** * @vm_inject_error_position: inject errors at different places in VM diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c new file mode 100644 index 000000000000..b3da3863928a --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -0,0 +1,289 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include +#include + +#include "xe_device.h" +#include "xe_pmu.h" + +/** + * DOC: Xe PMU (Performance Monitoring Unit) + * + * Expose events/counters like GT-C6 residency and GT frequency to user land via + * the perf interface. Events are per device. The GT can be selected with an + * extra config sub-field (bits 60-63). + * + * All events are listed in sysfs: + * + * $ ls -ld /sys/bus/event_source/devices/xe_* + * $ ls /sys/bus/event_source/devices/xe_0000_00_02.0/events/ + * $ ls /sys/bus/event_source/devices/xe_0000_00_02.0/format/ + * + * The format directory has info regarding the configs that can be used. + * The standard perf tool can be used to grep for a certain event as well. + * Example: + * + * $ perf list | grep gt-c6 + * + * To sample a specific event for a GT at regular intervals: + * + * $ perf stat -e -I + */ + +#define XE_PMU_EVENT_GT_MASK GENMASK_ULL(63, 60) +#define XE_PMU_EVENT_ID_MASK GENMASK_ULL(11, 0) + +static unsigned int config_to_event_id(u64 config) +{ + return FIELD_GET(XE_PMU_EVENT_ID_MASK, config); +} + +static unsigned int config_to_gt_id(u64 config) +{ + return FIELD_GET(XE_PMU_EVENT_GT_MASK, config); +} + +static struct xe_gt *event_to_gt(struct perf_event *event) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + u64 gt = config_to_gt_id(event->attr.config); + + return xe_device_get_gt(xe, gt); +} + +static bool event_supported(struct xe_pmu *pmu, unsigned int gt, + unsigned int id) +{ + if (gt >= XE_MAX_GT_PER_TILE) + return false; + + return false; +} + +static void xe_pmu_event_destroy(struct perf_event *event) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + + drm_WARN_ON(&xe->drm, event->parent); + drm_dev_put(&xe->drm); +} + +static int xe_pmu_event_init(struct perf_event *event) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct xe_pmu *pmu = &xe->pmu; + unsigned int id, gt; + + if (!pmu->registered) + return -ENODEV; + + if (event->attr.type != event->pmu->type) + return -ENOENT; + + /* unsupported modes and filters */ + if (event->attr.sample_period) /* no sampling */ + return -EINVAL; + + if (event->cpu < 0) + return -EINVAL; + + gt = config_to_gt_id(event->attr.config); + id = config_to_event_id(event->attr.config); + if (!event_supported(pmu, gt, id)) + return -ENOENT; + + if (has_branch_stack(event)) + return -EOPNOTSUPP; + + if (!event->parent) { + drm_dev_get(&xe->drm); + event->destroy = xe_pmu_event_destroy; + } + + return 0; +} + +static u64 __xe_pmu_event_read(struct perf_event *event) +{ + struct xe_gt *gt = event_to_gt(event); + u64 val = 0; + + if (!gt) + return 0; + + return val; +} + +static void xe_pmu_event_read(struct perf_event *event) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct hw_perf_event *hwc = &event->hw; + struct xe_pmu *pmu = &xe->pmu; + u64 prev, new; + + if (!pmu->registered) { + event->hw.state = PERF_HES_STOPPED; + return; + } + + prev = local64_read(&hwc->prev_count); + do { + new = __xe_pmu_event_read(event); + } while (!local64_try_cmpxchg(&hwc->prev_count, &prev, new)); + + local64_add(new - prev, &event->count); +} + +static void xe_pmu_enable(struct perf_event *event) +{ + /* + * Store the current counter value so we can report the correct delta + * for all listeners. Even when the event was already enabled and has + * an existing non-zero value. + */ + local64_set(&event->hw.prev_count, __xe_pmu_event_read(event)); +} + +static void xe_pmu_event_start(struct perf_event *event, int flags) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct xe_pmu *pmu = &xe->pmu; + + if (!pmu->registered) + return; + + xe_pmu_enable(event); + event->hw.state = 0; +} + +static void xe_pmu_event_stop(struct perf_event *event, int flags) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct xe_pmu *pmu = &xe->pmu; + + if (pmu->registered) + if (flags & PERF_EF_UPDATE) + xe_pmu_event_read(event); + + event->hw.state = PERF_HES_STOPPED; +} + +static int xe_pmu_event_add(struct perf_event *event, int flags) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct xe_pmu *pmu = &xe->pmu; + + if (!pmu->registered) + return -ENODEV; + + if (flags & PERF_EF_START) + xe_pmu_event_start(event, flags); + + return 0; +} + +static void xe_pmu_event_del(struct perf_event *event, int flags) +{ + xe_pmu_event_stop(event, PERF_EF_UPDATE); +} + +PMU_FORMAT_ATTR(gt, "config:60-63"); +PMU_FORMAT_ATTR(event, "config:0-11"); + +static struct attribute *pmu_format_attrs[] = { + &format_attr_event.attr, + &format_attr_gt.attr, + NULL, +}; + +static const struct attribute_group pmu_format_attr_group = { + .name = "format", + .attrs = pmu_format_attrs, +}; + +static struct attribute *pmu_event_attrs[] = { + /* No events yet */ + NULL, +}; + +static const struct attribute_group pmu_events_attr_group = { + .name = "events", + .attrs = pmu_event_attrs, +}; + +/** + * xe_pmu_unregister() - Remove/cleanup PMU registration + * @arg: Ptr to pmu + */ +static void xe_pmu_unregister(void *arg) +{ + struct xe_pmu *pmu = arg; + struct xe_device *xe = container_of(pmu, typeof(*xe), pmu); + + if (!pmu->registered) + return; + + pmu->registered = false; + + perf_pmu_unregister(&pmu->base); + kfree(pmu->name); +} + +/** + * xe_pmu_register() - Define basic PMU properties for Xe and add event callbacks. + * @pmu: the PMU object + * + * Returns 0 on success and an appropriate error code otherwise + */ +int xe_pmu_register(struct xe_pmu *pmu) +{ + struct xe_device *xe = container_of(pmu, typeof(*xe), pmu); + static const struct attribute_group *attr_groups[] = { + &pmu_format_attr_group, + &pmu_events_attr_group, + NULL + }; + int ret = -ENOMEM; + char *name; + + if (IS_SRIOV_VF(xe)) + return 0; + + name = kasprintf(GFP_KERNEL, "xe_%s", + dev_name(xe->drm.dev)); + if (!name) + goto err; + + /* tools/perf reserves colons as special. */ + strreplace(name, ':', '_'); + + pmu->name = name; + pmu->base.attr_groups = attr_groups; + pmu->base.scope = PERF_PMU_SCOPE_SYS_WIDE; + pmu->base.module = THIS_MODULE; + pmu->base.task_ctx_nr = perf_invalid_context; + pmu->base.event_init = xe_pmu_event_init; + pmu->base.add = xe_pmu_event_add; + pmu->base.del = xe_pmu_event_del; + pmu->base.start = xe_pmu_event_start; + pmu->base.stop = xe_pmu_event_stop; + pmu->base.read = xe_pmu_event_read; + + ret = perf_pmu_register(&pmu->base, pmu->name, -1); + if (ret) + goto err_name; + + pmu->registered = true; + + return devm_add_action_or_reset(xe->drm.dev, xe_pmu_unregister, pmu); + +err_name: + kfree(name); +err: + drm_err(&xe->drm, "Failed to register PMU (ret=%d)!\n", ret); + + return ret; +} diff --git a/drivers/gpu/drm/xe/xe_pmu.h b/drivers/gpu/drm/xe/xe_pmu.h new file mode 100644 index 000000000000..60c37126f87e --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pmu.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_PMU_H_ +#define _XE_PMU_H_ + +#include "xe_pmu_types.h" + +#if IS_ENABLED(CONFIG_PERF_EVENTS) +int xe_pmu_register(struct xe_pmu *pmu); +#else +static inline int xe_pmu_register(struct xe_pmu *pmu) { return 0; } +#endif + +#endif + diff --git a/drivers/gpu/drm/xe/xe_pmu_types.h b/drivers/gpu/drm/xe/xe_pmu_types.h new file mode 100644 index 000000000000..0e8faae6bc1b --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pmu_types.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_PMU_TYPES_H_ +#define _XE_PMU_TYPES_H_ + +#include +#include + +#define XE_PMU_MAX_GT 2 + +/** + * struct xe_pmu - PMU related data per Xe device + * + * Stores per device PMU info that includes event/perf attributes and sampling + * counters across all GTs for this device. + */ +struct xe_pmu { + /** + * @base: PMU base. + */ + struct pmu base; + /** + * @registered: PMU is registered and not in the unregistering process. + */ + bool registered; + /** + * @name: Name as registered with perf core. + */ + const char *name; +}; + +#endif From 257a10c18e18cbcec7b9621820c11a5c9ec613bd Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 21:04:07 -0800 Subject: [PATCH 0216/1443] drm/xe/pmu: Assert max gt XE_PMU_MAX_GT needs to be used due to a circular dependency, but we should make sure it doesn't go out of sync with XE_PMU_MAX_GT. Add a compile check for that. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-2-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pmu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index b3da3863928a..df93ba96bdc1 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -249,6 +249,8 @@ int xe_pmu_register(struct xe_pmu *pmu) int ret = -ENOMEM; char *name; + BUILD_BUG_ON(XE_MAX_GT_PER_TILE != XE_PMU_MAX_GT); + if (IS_SRIOV_VF(xe)) return 0; From ef7ce3938621e47c6aa2d517eab6f1d484ad365b Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 21:04:08 -0800 Subject: [PATCH 0217/1443] drm/xe/pmu: Extract xe_pmu_event_update() Like other pmu drivers, keep the update separate from the read so it can be called from other methods (like stop()) without side effects. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-3-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pmu.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index df93ba96bdc1..33598272db6a 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -117,18 +117,11 @@ static u64 __xe_pmu_event_read(struct perf_event *event) return val; } -static void xe_pmu_event_read(struct perf_event *event) +static void xe_pmu_event_update(struct perf_event *event) { - struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); struct hw_perf_event *hwc = &event->hw; - struct xe_pmu *pmu = &xe->pmu; u64 prev, new; - if (!pmu->registered) { - event->hw.state = PERF_HES_STOPPED; - return; - } - prev = local64_read(&hwc->prev_count); do { new = __xe_pmu_event_read(event); @@ -137,6 +130,19 @@ static void xe_pmu_event_read(struct perf_event *event) local64_add(new - prev, &event->count); } +static void xe_pmu_event_read(struct perf_event *event) +{ + struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); + struct xe_pmu *pmu = &xe->pmu; + + if (!pmu->registered) { + event->hw.state = PERF_HES_STOPPED; + return; + } + + xe_pmu_event_update(event); +} + static void xe_pmu_enable(struct perf_event *event) { /* @@ -166,7 +172,7 @@ static void xe_pmu_event_stop(struct perf_event *event, int flags) if (pmu->registered) if (flags & PERF_EF_UPDATE) - xe_pmu_event_read(event); + xe_pmu_event_update(event); event->hw.state = PERF_HES_STOPPED; } From 4ee64041bcca5289623c287336bfd94a42f42104 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 21:04:09 -0800 Subject: [PATCH 0218/1443] drm/xe/pmu: Get/put runtime pm on event init When the event is created, make sure runtime pm is taken and later put: in order to read an event counter the GPU needs to remain accessible and doing a get/put during perf's read is not possible it's holding a raw_spinlock. Suggested-by: Rodrigo Vivi Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-4-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pmu.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index 33598272db6a..6678340d6195 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -7,6 +7,7 @@ #include #include "xe_device.h" +#include "xe_pm.h" #include "xe_pmu.h" /** @@ -68,6 +69,7 @@ static void xe_pmu_event_destroy(struct perf_event *event) struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); drm_WARN_ON(&xe->drm, event->parent); + xe_pm_runtime_put(xe); drm_dev_put(&xe->drm); } @@ -100,6 +102,7 @@ static int xe_pmu_event_init(struct perf_event *event) if (!event->parent) { drm_dev_get(&xe->drm); + xe_pm_runtime_get(xe); event->destroy = xe_pmu_event_destroy; } From 6ea5bf169ad4b5b2da1e7753031b446c22edecf3 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 21:04:10 -0800 Subject: [PATCH 0219/1443] drm/xe/pmu: Add attribute skeleton Add the generic support for defining new attributes. This only adds the macros and common infra for the event counters, but no counters yet. This is going to be added as follow up changes. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-5-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pmu.c | 69 +++++++++++++++++++++++++++++-- drivers/gpu/drm/xe/xe_pmu_types.h | 4 ++ 2 files changed, 69 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index 6678340d6195..6e19e08dd617 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -61,7 +61,8 @@ static bool event_supported(struct xe_pmu *pmu, unsigned int gt, if (gt >= XE_MAX_GT_PER_TILE) return false; - return false; + return id < sizeof(pmu->supported_events) * BITS_PER_BYTE && + pmu->supported_events & BIT_ULL(id); } static void xe_pmu_event_destroy(struct perf_event *event) @@ -213,16 +214,73 @@ static const struct attribute_group pmu_format_attr_group = { .attrs = pmu_format_attrs, }; -static struct attribute *pmu_event_attrs[] = { - /* No events yet */ +__maybe_unused static ssize_t event_attr_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct perf_pmu_events_attr *pmu_attr = + container_of(attr, struct perf_pmu_events_attr, attr); + + return sprintf(buf, "event=%#04llx\n", pmu_attr->id); +} + +#define XE_EVENT_ATTR(name_, v_, id_) \ + PMU_EVENT_ATTR(name_, pmu_event_ ## v_, id_, event_attr_show) + +#define XE_EVENT_ATTR_UNIT(name_, v_, unit_) \ + PMU_EVENT_ATTR_STRING(name_.unit, pmu_event_unit_ ## v_, unit_) + +#define XE_EVENT_ATTR_GROUP(v_, id_, ...) \ + static struct attribute *pmu_attr_ ##v_[] = { \ + __VA_ARGS__, \ + NULL \ + }; \ + static umode_t is_visible_##v_(struct kobject *kobj, \ + struct attribute *attr, int idx) \ + { \ + struct perf_pmu_events_attr *pmu_attr; \ + struct xe_pmu *pmu; \ + \ + pmu_attr = container_of(attr, typeof(*pmu_attr), attr.attr); \ + pmu = container_of(dev_get_drvdata(kobj_to_dev(kobj)), \ + typeof(*pmu), base); \ + \ + return event_supported(pmu, 0, id_) ? attr->mode : 0; \ + } \ + static const struct attribute_group pmu_group_ ##v_ = { \ + .name = "events", \ + .attrs = pmu_attr_ ## v_, \ + .is_visible = is_visible_ ## v_, \ + } + +#define XE_EVENT_ATTR_SIMPLE(name_, v_, id_, unit_) \ + XE_EVENT_ATTR(name_, v_, id_) \ + XE_EVENT_ATTR_UNIT(name_, v_, unit_) \ + XE_EVENT_ATTR_GROUP(v_, id_, &pmu_event_ ##v_.attr.attr, \ + &pmu_event_unit_ ##v_.attr.attr) + +#define XE_EVENT_ATTR_NOUNIT(name_, v_, id_) \ + XE_EVENT_ATTR(name_, v_, id_) \ + XE_EVENT_ATTR_GROUP(v_, id_, &pmu_event_ ##v_.attr.attr) + +static struct attribute *pmu_empty_event_attrs[] = { + /* Empty - all events are added as groups with .attr_update() */ NULL, }; static const struct attribute_group pmu_events_attr_group = { .name = "events", - .attrs = pmu_event_attrs, + .attrs = pmu_empty_event_attrs, }; +static const struct attribute_group *pmu_events_attr_update[] = { + /* No events yet */ + NULL, +}; + +static void set_supported_events(struct xe_pmu *pmu) +{ +} + /** * xe_pmu_unregister() - Remove/cleanup PMU registration * @arg: Ptr to pmu @@ -273,6 +331,7 @@ int xe_pmu_register(struct xe_pmu *pmu) pmu->name = name; pmu->base.attr_groups = attr_groups; + pmu->base.attr_update = pmu_events_attr_update; pmu->base.scope = PERF_PMU_SCOPE_SYS_WIDE; pmu->base.module = THIS_MODULE; pmu->base.task_ctx_nr = perf_invalid_context; @@ -283,6 +342,8 @@ int xe_pmu_register(struct xe_pmu *pmu) pmu->base.stop = xe_pmu_event_stop; pmu->base.read = xe_pmu_event_read; + set_supported_events(pmu); + ret = perf_pmu_register(&pmu->base, pmu->name, -1); if (ret) goto err_name; diff --git a/drivers/gpu/drm/xe/xe_pmu_types.h b/drivers/gpu/drm/xe/xe_pmu_types.h index 0e8faae6bc1b..f5ba4d56622c 100644 --- a/drivers/gpu/drm/xe/xe_pmu_types.h +++ b/drivers/gpu/drm/xe/xe_pmu_types.h @@ -30,6 +30,10 @@ struct xe_pmu { * @name: Name as registered with perf core. */ const char *name; + /** + * @supported_events: Bitmap of supported events, indexed by event id + */ + u64 supported_events; }; #endif From 897286f2948cefe5f9d37fc3148fc19d0c7b160c Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Thu, 23 Jan 2025 21:04:11 -0800 Subject: [PATCH 0220/1443] drm/xe/pmu: Add GT C6 events Provide a PMU interface for GT C6 residency counters. The interface is similar to the one available for i915, but gt is passed in the config when creating the event. Sample usage and output: $ perf list | grep gt-c6 xe_0000_00_02.0/gt-c6-residency/ [Kernel PMU event] $ tail /sys/bus/event_source/devices/xe_0000_00_02.0/events/gt-c6-residency* ==> /sys/bus/event_source/devices/xe_0000_00_02.0/events/gt-c6-residency <== event=0x01 ==> /sys/bus/event_source/devices/xe_0000_00_02.0/events/gt-c6-residency.unit <== ms $ perf stat -e xe_0000_00_02.0/gt-c6-residency,gt=0/ -I1000 # time counts unit events 1.001196056 1,001 ms xe_0000_00_02.0/gt-c6-residency,gt=0/ 2.005216219 1,003 ms xe_0000_00_02.0/gt-c6-residency,gt=0/ Reviewed-by: Rodrigo Vivi Reviewed-by: Riana Tauro Signed-off-by: Vinay Belgaumkar Link: https://patchwork.freedesktop.org/patch/msgid/20250124050411.2189060-6-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pmu.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index 6e19e08dd617..3910a82328ee 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -7,6 +7,7 @@ #include #include "xe_device.h" +#include "xe_gt_idle.h" #include "xe_pm.h" #include "xe_pmu.h" @@ -47,6 +48,8 @@ static unsigned int config_to_gt_id(u64 config) return FIELD_GET(XE_PMU_EVENT_GT_MASK, config); } +#define XE_PMU_EVENT_GT_C6_RESIDENCY 0x01 + static struct xe_gt *event_to_gt(struct perf_event *event) { struct xe_device *xe = container_of(event->pmu, typeof(*xe), pmu.base); @@ -113,12 +116,16 @@ static int xe_pmu_event_init(struct perf_event *event) static u64 __xe_pmu_event_read(struct perf_event *event) { struct xe_gt *gt = event_to_gt(event); - u64 val = 0; if (!gt) return 0; - return val; + switch (config_to_event_id(event->attr.config)) { + case XE_PMU_EVENT_GT_C6_RESIDENCY: + return xe_gt_idle_residency_msec(>->gtidle); + } + + return 0; } static void xe_pmu_event_update(struct perf_event *event) @@ -214,8 +221,8 @@ static const struct attribute_group pmu_format_attr_group = { .attrs = pmu_format_attrs, }; -__maybe_unused static ssize_t event_attr_show(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t event_attr_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct perf_pmu_events_attr *pmu_attr = container_of(attr, struct perf_pmu_events_attr, attr); @@ -262,6 +269,8 @@ __maybe_unused static ssize_t event_attr_show(struct device *dev, XE_EVENT_ATTR(name_, v_, id_) \ XE_EVENT_ATTR_GROUP(v_, id_, &pmu_event_ ##v_.attr.attr) +XE_EVENT_ATTR_SIMPLE(gt-c6-residency, gt_c6_residency, XE_PMU_EVENT_GT_C6_RESIDENCY, "ms"); + static struct attribute *pmu_empty_event_attrs[] = { /* Empty - all events are added as groups with .attr_update() */ NULL, @@ -273,12 +282,16 @@ static const struct attribute_group pmu_events_attr_group = { }; static const struct attribute_group *pmu_events_attr_update[] = { - /* No events yet */ + &pmu_group_gt_c6_residency, NULL, }; static void set_supported_events(struct xe_pmu *pmu) { + struct xe_device *xe = container_of(pmu, typeof(*xe), pmu); + + if (!xe->info.skip_guc_pc) + pmu->supported_events |= BIT_ULL(XE_PMU_EVENT_GT_C6_RESIDENCY); } /** From 698244bbb3bfd32ddf9a0b70a12b1c7d69056497 Mon Sep 17 00:00:00 2001 From: Nick Chan Date: Sun, 19 Jan 2025 00:31:42 +0800 Subject: [PATCH 0221/1443] irqchip/apple-aic: Only handle PMC interrupt as FIQ when configured so The CPU PMU in Apple SoCs can be configured to fire its interrupt in one of several ways, and since Apple A11 one of the methods is FIQ, but the check of the configuration register fails to test explicitely for FIQ mode. It tests whether the IMODE bitfield is zero or not and the PMCRO_IACT bit is set. That results in false positives when the IMODE bitfield is not zero, but does not have the mode PMCR0_IMODE_FIQ. Only handle the PMC interrupt as a FIQ when the CPU PMU has been configured to fire FIQs, i.e. the IMODE bitfield value is PMCR0_IMODE_FIQ and PMCR0_IACT is set. Fixes: c7708816c944 ("irqchip/apple-aic: Wire PMU interrupts") Signed-off-by: Nick Chan Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250118163554.16733-1-towinchenmi@gmail.com --- drivers/irqchip/irq-apple-aic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-apple-aic.c b/drivers/irqchip/irq-apple-aic.c index da5250f0155c..2b1684c60e3c 100644 --- a/drivers/irqchip/irq-apple-aic.c +++ b/drivers/irqchip/irq-apple-aic.c @@ -577,7 +577,8 @@ static void __exception_irq_entry aic_handle_fiq(struct pt_regs *regs) AIC_FIQ_HWIRQ(AIC_TMR_EL02_VIRT)); } - if (read_sysreg_s(SYS_IMP_APL_PMCR0_EL1) & PMCR0_IACT) { + if ((read_sysreg_s(SYS_IMP_APL_PMCR0_EL1) & (PMCR0_IMODE | PMCR0_IACT)) == + (FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_FIQ) | PMCR0_IACT)) { int irq; if (cpumask_test_cpu(smp_processor_id(), &aic_irqc->fiq_aff[AIC_CPU_PMU_P]->aff)) From d6f3e7d564b2309e1f17e709a70eca78d7ca2bb8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 24 Jan 2025 12:22:12 -1000 Subject: [PATCH 0222/1443] sched_ext: Fix incorrect autogroup migration detection scx_move_task() is called from sched_move_task() and tells the BPF scheduler that cgroup migration is being committed. sched_move_task() is used by both cgroup and autogroup migrations and scx_move_task() tried to filter out autogroup migrations by testing the destination cgroup and PF_EXITING but this is not enough. In fact, without explicitly tagging the thread which is doing the cgroup migration, there is no good way to tell apart scx_move_task() invocations for racing migration to the root cgroup and an autogroup migration. This led to scx_move_task() incorrectly ignoring a migration from non-root cgroup to an autogroup of the root cgroup triggering the following warning: WARNING: CPU: 7 PID: 1 at kernel/sched/ext.c:3725 scx_cgroup_can_attach+0x196/0x340 ... Call Trace: cgroup_migrate_execute+0x5b1/0x700 cgroup_attach_task+0x296/0x400 __cgroup_procs_write+0x128/0x140 cgroup_procs_write+0x17/0x30 kernfs_fop_write_iter+0x141/0x1f0 vfs_write+0x31d/0x4a0 __x64_sys_write+0x72/0xf0 do_syscall_64+0x82/0x160 entry_SYSCALL_64_after_hwframe+0x76/0x7e Fix it by adding an argument to sched_move_task() that indicates whether the moving is for a cgroup or autogroup migration. After the change, scx_move_task() is called only for cgroup migrations and renamed to scx_cgroup_move_task(). Link: https://github.com/sched-ext/scx/issues/370 Fixes: 819513666966 ("sched_ext: Add cgroup support") Cc: stable@vger.kernel.org # v6.12+ Acked-by: Peter Zijlstra (Intel) Signed-off-by: Tejun Heo --- kernel/sched/autogroup.c | 4 ++-- kernel/sched/core.c | 7 ++++--- kernel/sched/ext.c | 15 +-------------- kernel/sched/ext.h | 4 ++-- kernel/sched/sched.h | 2 +- 5 files changed, 10 insertions(+), 22 deletions(-) diff --git a/kernel/sched/autogroup.c b/kernel/sched/autogroup.c index db68a964e34e..c4a3ccf6a8ac 100644 --- a/kernel/sched/autogroup.c +++ b/kernel/sched/autogroup.c @@ -150,7 +150,7 @@ void sched_autogroup_exit_task(struct task_struct *p) * see this thread after that: we can no longer use signal->autogroup. * See the PF_EXITING check in task_wants_autogroup(). */ - sched_move_task(p); + sched_move_task(p, true); } static void @@ -182,7 +182,7 @@ autogroup_move_group(struct task_struct *p, struct autogroup *ag) * sched_autogroup_exit_task(). */ for_each_thread(p, t) - sched_move_task(t); + sched_move_task(t, true); unlock_task_sighand(p, &flags); autogroup_kref_put(prev); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 901170708e2a..e77897a62442 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -9042,7 +9042,7 @@ static void sched_change_group(struct task_struct *tsk, struct task_group *group * now. This function just updates tsk->se.cfs_rq and tsk->se.parent to reflect * its new group. */ -void sched_move_task(struct task_struct *tsk) +void sched_move_task(struct task_struct *tsk, bool for_autogroup) { int queued, running, queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; @@ -9071,7 +9071,8 @@ void sched_move_task(struct task_struct *tsk) put_prev_task(rq, tsk); sched_change_group(tsk, group); - scx_move_task(tsk); + if (!for_autogroup) + scx_cgroup_move_task(tsk); if (queued) enqueue_task(rq, tsk, queue_flags); @@ -9172,7 +9173,7 @@ static void cpu_cgroup_attach(struct cgroup_taskset *tset) struct cgroup_subsys_state *css; cgroup_taskset_for_each(task, css, tset) - sched_move_task(task); + sched_move_task(task, false); scx_cgroup_finish_attach(); } diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 7081c7be5f62..c7b159f48834 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -4323,24 +4323,11 @@ int scx_cgroup_can_attach(struct cgroup_taskset *tset) return ops_sanitize_err("cgroup_prep_move", ret); } -void scx_move_task(struct task_struct *p) +void scx_cgroup_move_task(struct task_struct *p) { if (!scx_cgroup_enabled) return; - /* - * We're called from sched_move_task() which handles both cgroup and - * autogroup moves. Ignore the latter. - * - * Also ignore exiting tasks, because in the exit path tasks transition - * from the autogroup to the root group, so task_group_is_autogroup() - * alone isn't able to catch exiting autogroup tasks. This is safe for - * cgroup_move(), because cgroup migrations never happen for PF_EXITING - * tasks. - */ - if (task_group_is_autogroup(task_group(p)) || (p->flags & PF_EXITING)) - return; - /* * @p must have ops.cgroup_prep_move() called on it and thus * cgrp_moving_from set. diff --git a/kernel/sched/ext.h b/kernel/sched/ext.h index 4d022d17ac7d..1079b56b0f7a 100644 --- a/kernel/sched/ext.h +++ b/kernel/sched/ext.h @@ -73,7 +73,7 @@ static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) {} int scx_tg_online(struct task_group *tg); void scx_tg_offline(struct task_group *tg); int scx_cgroup_can_attach(struct cgroup_taskset *tset); -void scx_move_task(struct task_struct *p); +void scx_cgroup_move_task(struct task_struct *p); void scx_cgroup_finish_attach(void); void scx_cgroup_cancel_attach(struct cgroup_taskset *tset); void scx_group_set_weight(struct task_group *tg, unsigned long cgrp_weight); @@ -82,7 +82,7 @@ void scx_group_set_idle(struct task_group *tg, bool idle); static inline int scx_tg_online(struct task_group *tg) { return 0; } static inline void scx_tg_offline(struct task_group *tg) {} static inline int scx_cgroup_can_attach(struct cgroup_taskset *tset) { return 0; } -static inline void scx_move_task(struct task_struct *p) {} +static inline void scx_cgroup_move_task(struct task_struct *p) {} static inline void scx_cgroup_finish_attach(void) {} static inline void scx_cgroup_cancel_attach(struct cgroup_taskset *tset) {} static inline void scx_group_set_weight(struct task_group *tg, unsigned long cgrp_weight) {} diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 38e0e323dda2..b93c8c3dc05a 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -572,7 +572,7 @@ extern void sched_online_group(struct task_group *tg, extern void sched_destroy_group(struct task_group *tg); extern void sched_release_group(struct task_group *tg); -extern void sched_move_task(struct task_struct *tsk); +extern void sched_move_task(struct task_struct *tsk, bool for_autogroup); #ifdef CONFIG_FAIR_GROUP_SCHED extern int sched_group_set_shares(struct task_group *tg, unsigned long shares); From 14b66746088098f43db1e8732ff8902ddbdd4de0 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 24 Jan 2025 19:52:47 +0100 Subject: [PATCH 0223/1443] drm/xe/pf: Use GuC Buffer Cache during policy provisioning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Start using GuC buffer cache for the SRIOV policy configuration actions. This is a required step before we could declare SRIOV PF as being a reclaim safe. Signed-off-by: Michal Wajdeczko Cc: Thomas Hellström Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20250124185247.676-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf_policy.c | 50 +++++++++++----------- 1 file changed, 24 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_policy.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_policy.c index c00fb354705f..4445f660e6d1 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_policy.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_policy.c @@ -10,6 +10,7 @@ #include "xe_gt_sriov_pf_helpers.h" #include "xe_gt_sriov_pf_policy.h" #include "xe_gt_sriov_printk.h" +#include "xe_guc_buf.h" #include "xe_guc_ct.h" #include "xe_guc_klv_helpers.h" #include "xe_pm.h" @@ -34,48 +35,28 @@ static int guc_action_update_vgt_policy(struct xe_guc *guc, u64 addr, u32 size) * Return: number of KLVs that were successfully parsed and saved, * negative error code on failure. */ -static int pf_send_policy_klvs(struct xe_gt *gt, const u32 *klvs, u32 num_dwords) +static int pf_send_policy_klvs(struct xe_gt *gt, struct xe_guc_buf buf, u32 num_dwords) { - const u32 bytes = num_dwords * sizeof(u32); - struct xe_tile *tile = gt_to_tile(gt); - struct xe_device *xe = tile_to_xe(tile); struct xe_guc *guc = >->uc.guc; - struct xe_bo *bo; - int ret; - bo = xe_bo_create_pin_map(xe, tile, NULL, - ALIGN(bytes, PAGE_SIZE), - ttm_bo_type_kernel, - XE_BO_FLAG_VRAM_IF_DGFX(tile) | - XE_BO_FLAG_GGTT); - if (IS_ERR(bo)) - return PTR_ERR(bo); - - xe_map_memcpy_to(xe, &bo->vmap, 0, klvs, bytes); - - ret = guc_action_update_vgt_policy(guc, xe_bo_ggtt_addr(bo), num_dwords); - - xe_bo_unpin_map_no_vm(bo); - - return ret; + return guc_action_update_vgt_policy(guc, xe_guc_buf_flush(buf), num_dwords); } /* * Return: 0 on success, -ENOKEY if some KLVs were not updated, -EPROTO if reply was malformed, * negative error code on failure. */ -static int pf_push_policy_klvs(struct xe_gt *gt, u32 num_klvs, - const u32 *klvs, u32 num_dwords) +static int pf_push_policy_buf_klvs(struct xe_gt *gt, u32 num_klvs, + struct xe_guc_buf buf, u32 num_dwords) { int ret; - xe_gt_assert(gt, num_klvs == xe_guc_klv_count(klvs, num_dwords)); - - ret = pf_send_policy_klvs(gt, klvs, num_dwords); + ret = pf_send_policy_klvs(gt, buf, num_dwords); if (ret != num_klvs) { int err = ret < 0 ? ret : ret < num_klvs ? -ENOKEY : -EPROTO; struct drm_printer p = xe_gt_info_printer(gt); + void *klvs = xe_guc_buf_cpu_ptr(buf); xe_gt_sriov_notice(gt, "Failed to push %u policy KLV%s (%pe)\n", num_klvs, str_plural(num_klvs), ERR_PTR(err)); @@ -86,6 +67,23 @@ static int pf_push_policy_klvs(struct xe_gt *gt, u32 num_klvs, return 0; } +/* + * Return: 0 on success, -ENOBUFS if there is no free buffer for the indirect data, + * negative error code on failure. + */ +static int pf_push_policy_klvs(struct xe_gt *gt, u32 num_klvs, + const u32 *klvs, u32 num_dwords) +{ + CLASS(xe_guc_buf_from_data, buf)(>->uc.guc.buf, klvs, num_dwords * sizeof(u32)); + + xe_gt_assert(gt, num_klvs == xe_guc_klv_count(klvs, num_dwords)); + + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + + return pf_push_policy_buf_klvs(gt, num_klvs, buf, num_dwords); +} + static int pf_push_policy_u32(struct xe_gt *gt, u16 key, u32 value) { u32 klv[] = { From 5f52bbf2f6e0997394cf9c449d44e1c80ff4282c Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Sat, 25 Jan 2025 18:14:12 +0100 Subject: [PATCH 0224/1443] tools/sched_ext: Add helper to check task migration state Introduce a new helper for BPF schedulers to determine whether a task can migrate or not (supporting both SMP and UP systems). Fixes: e9fe182772dc ("sched_ext: selftests/dsp_local_on: Fix sporadic failures") Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- tools/sched_ext/include/scx/common.bpf.h | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/include/scx/common.bpf.h index f3e15e9efa76..f254a39b86a5 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -404,6 +404,17 @@ static __always_inline const struct cpumask *cast_mask(struct bpf_cpumask *mask) return (const struct cpumask *)mask; } +/* + * Return true if task @p cannot migrate to a different CPU, false + * otherwise. + */ +static inline bool is_migration_disabled(const struct task_struct *p) +{ + if (bpf_core_field_exists(p->migration_disabled)) + return p->migration_disabled; + return false; +} + /* rcu */ void bpf_rcu_read_lock(void) __ksym; void bpf_rcu_read_unlock(void) __ksym; From 3c7d51b0d29954c40ea3a097e0ec7884b4344331 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Sat, 25 Jan 2025 10:36:07 +0100 Subject: [PATCH 0225/1443] sched_ext: selftests/dsp_local_on: Fix selftest on UP systems In UP systems p->migration_disabled is not available. Fix this by using the portable helper is_migration_disabled(p). Fixes: e9fe182772dc ("sched_ext: selftests/dsp_local_on: Fix sporadic failures") Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- tools/testing/selftests/sched_ext/dsp_local_on.bpf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c index 758b479bd1ee..c02b2aa6fc64 100644 --- a/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c +++ b/tools/testing/selftests/sched_ext/dsp_local_on.bpf.c @@ -43,7 +43,7 @@ void BPF_STRUCT_OPS(dsp_local_on_dispatch, s32 cpu, struct task_struct *prev) if (!p) return; - if (p->nr_cpus_allowed == nr_cpus && !p->migration_disabled) + if (p->nr_cpus_allowed == nr_cpus && !is_migration_disabled(p)) target = bpf_get_prandom_u32() % nr_cpus; else target = scx_bpf_task_cpu(p); From a4d1c5d0b99b75263a5626d2e52d569db3844b33 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Sat, 25 Jan 2025 22:55:05 +0100 Subject: [PATCH 0226/1443] drm/xe/pf: Move VFs reprovisioning to worker MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since the GuC is reset during GT reset, we need to re-send the entire SR-IOV provisioning configuration to the GuC. But since this whole configuration is protected by the PF master mutex and we can't avoid making allocations under this mutex (like during LMEM provisioning), we can't do this reprovisioning from gt-reset path if we want to be reclaim-safe. Move VFs reprovisioning to a async worker that we will start from the gt-reset path. Signed-off-by: Michal Wajdeczko Cc: Thomas Hellström Cc: Matthew Brost Reviewed-by: Michał Winiarski Reviewed-by: Stuart Summers Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20250125215505.720-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf.c | 43 +++++++++++++++++++++-- drivers/gpu/drm/xe/xe_gt_sriov_pf_types.h | 10 ++++++ 2 files changed, 51 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c index 6f906c8e8108..d66478deab98 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c @@ -15,7 +15,11 @@ #include "xe_gt_sriov_pf_helpers.h" #include "xe_gt_sriov_pf_migration.h" #include "xe_gt_sriov_pf_service.h" +#include "xe_gt_sriov_printk.h" #include "xe_mmio.h" +#include "xe_pm.h" + +static void pf_worker_restart_func(struct work_struct *w); /* * VF's metadata is maintained in the flexible array where: @@ -41,6 +45,11 @@ static int pf_alloc_metadata(struct xe_gt *gt) return 0; } +static void pf_init_workers(struct xe_gt *gt) +{ + INIT_WORK(>->sriov.pf.workers.restart, pf_worker_restart_func); +} + /** * xe_gt_sriov_pf_init_early - Prepare SR-IOV PF data structures on PF. * @gt: the &xe_gt to initialize @@ -65,6 +74,8 @@ int xe_gt_sriov_pf_init_early(struct xe_gt *gt) if (err) return err; + pf_init_workers(gt); + return 0; } @@ -155,6 +166,35 @@ void xe_gt_sriov_pf_sanitize_hw(struct xe_gt *gt, unsigned int vfid) pf_clear_vf_scratch_regs(gt, vfid); } +static void pf_restart(struct xe_gt *gt) +{ + struct xe_device *xe = gt_to_xe(gt); + + xe_pm_runtime_get(xe); + xe_gt_sriov_pf_config_restart(gt); + xe_gt_sriov_pf_control_restart(gt); + xe_pm_runtime_put(xe); + + xe_gt_sriov_dbg(gt, "restart completed\n"); +} + +static void pf_worker_restart_func(struct work_struct *w) +{ + struct xe_gt *gt = container_of(w, typeof(*gt), sriov.pf.workers.restart); + + pf_restart(gt); +} + +static void pf_queue_restart(struct xe_gt *gt) +{ + struct xe_device *xe = gt_to_xe(gt); + + xe_gt_assert(gt, IS_SRIOV_PF(xe)); + + if (!queue_work(xe->sriov.wq, >->sriov.pf.workers.restart)) + xe_gt_sriov_dbg(gt, "restart already in queue!\n"); +} + /** * xe_gt_sriov_pf_restart - Restart SR-IOV support after a GT reset. * @gt: the &xe_gt @@ -163,6 +203,5 @@ void xe_gt_sriov_pf_sanitize_hw(struct xe_gt *gt, unsigned int vfid) */ void xe_gt_sriov_pf_restart(struct xe_gt *gt) { - xe_gt_sriov_pf_config_restart(gt); - xe_gt_sriov_pf_control_restart(gt); + pf_queue_restart(gt); } diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf_types.h index 0426b1a77069..a64a6835ad65 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_types.h +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_types.h @@ -35,8 +35,17 @@ struct xe_gt_sriov_metadata { struct xe_gt_sriov_state_snapshot snapshot; }; +/** + * struct xe_gt_sriov_pf_workers - GT level workers used by the PF. + */ +struct xe_gt_sriov_pf_workers { + /** @restart: worker that executes actions post GT reset */ + struct work_struct restart; +}; + /** * struct xe_gt_sriov_pf - GT level PF virtualization data. + * @workers: workers data. * @service: service data. * @control: control data. * @policy: policy data. @@ -45,6 +54,7 @@ struct xe_gt_sriov_metadata { * @vfs: metadata for all VFs. */ struct xe_gt_sriov_pf { + struct xe_gt_sriov_pf_workers workers; struct xe_gt_sriov_pf_service service; struct xe_gt_sriov_pf_control control; struct xe_gt_sriov_pf_policy policy; From ef34861098c081a6222dbbe84439b571fa35c8a0 Mon Sep 17 00:00:00 2001 From: John Harrison Date: Fri, 17 Jan 2025 16:54:03 -0800 Subject: [PATCH 0227/1443] drm/xe: Upgrade complaint about missing slice info The steering code needs to know slice/subslice counts and this information should be retrieved from the hwconfig table. However, earlier platforms don't have it, hence the KMD has a fallback path. Newer platforms really should have the entries and if they are missing that is a bug that needs to be fixed in the table. So update the complaint to be an error on newer platforms and remove it completely for older ones that we know are bad (but are not POR for the Xe driver anyway). Also, re-word the message a little to make it clearer what the issue is. Signed-off-by: John Harrison Reviewed-by: Matt Roper Reviewed-by: Stuart Summers Acked-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250118005403.2960807-1-John.C.Harrison@Intel.com --- drivers/gpu/drm/xe/xe_gt_mcr.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c index a1676b787fdc..605aad3554e7 100644 --- a/drivers/gpu/drm/xe/xe_gt_mcr.c +++ b/drivers/gpu/drm/xe/xe_gt_mcr.c @@ -341,7 +341,13 @@ static unsigned int dss_per_group(struct xe_gt *gt) return DIV_ROUND_UP(max_subslices, max_slices); fallback: - xe_gt_dbg(gt, "GuC hwconfig cannot provide dss/slice; using typical fallback values\n"); + /* + * Some older platforms don't have tables or don't have complete tables. + * Newer platforms should always have the required info. + */ + if (GRAPHICS_VERx100(gt_to_xe(gt)) >= 2000) + xe_gt_err(gt, "Slice/Subslice counts missing from hwconfig table; using typical fallback values\n"); + if (gt_to_xe(gt)->info.platform == XE_PVC) return 8; else if (GRAPHICS_VERx100(gt_to_xe(gt)) >= 1250) From 1626e5ef0b00386a4fd083fa7c46c8edbd75f9b4 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Mon, 27 Jan 2025 23:06:16 +0100 Subject: [PATCH 0228/1443] sched_ext: Fix lock imbalance in dispatch_to_local_dsq() While performing the rq locking dance in dispatch_to_local_dsq(), we may trigger the following lock imbalance condition, in particular when multiple tasks are rapidly changing CPU affinity (i.e., running a `stress-ng --race-sched 0`): [ 13.413579] ===================================== [ 13.413660] WARNING: bad unlock balance detected! [ 13.413729] 6.13.0-virtme #15 Not tainted [ 13.413792] ------------------------------------- [ 13.413859] kworker/1:1/80 is trying to release lock (&rq->__lock) at: [ 13.413954] [] dispatch_to_local_dsq+0x108/0x1a0 [ 13.414111] but there are no more locks to release! [ 13.414176] [ 13.414176] other info that might help us debug this: [ 13.414258] 1 lock held by kworker/1:1/80: [ 13.414318] #0: ffff8b66feb41698 (&rq->__lock){-.-.}-{2:2}, at: raw_spin_rq_lock_nested+0x20/0x90 [ 13.414612] [ 13.414612] stack backtrace: [ 13.415255] CPU: 1 UID: 0 PID: 80 Comm: kworker/1:1 Not tainted 6.13.0-virtme #15 [ 13.415505] Workqueue: 0x0 (events) [ 13.415567] Sched_ext: dsp_local_on (enabled+all), task: runnable_at=-2ms [ 13.415570] Call Trace: [ 13.415700] [ 13.415744] dump_stack_lvl+0x78/0xe0 [ 13.415806] ? dispatch_to_local_dsq+0x108/0x1a0 [ 13.415884] print_unlock_imbalance_bug+0x11b/0x130 [ 13.415965] ? dispatch_to_local_dsq+0x108/0x1a0 [ 13.416226] lock_release+0x231/0x2c0 [ 13.416326] _raw_spin_unlock+0x1b/0x40 [ 13.416422] dispatch_to_local_dsq+0x108/0x1a0 [ 13.416554] flush_dispatch_buf+0x199/0x1d0 [ 13.416652] balance_one+0x194/0x370 [ 13.416751] balance_scx+0x61/0x1e0 [ 13.416848] prev_balance+0x43/0xb0 [ 13.416947] __pick_next_task+0x6b/0x1b0 [ 13.417052] __schedule+0x20d/0x1740 This happens because dispatch_to_local_dsq() is racing with dispatch_dequeue() and, when the latter wins, we incorrectly assume that the task has been moved to dst_rq. Fix by properly tracking the currently locked rq. Fixes: 4d3ca89bdd31 ("sched_ext: Refactor consume_remote_task()") Signed-off-by: Andrea Righi Signed-off-by: Tejun Heo --- kernel/sched/ext.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index c7b159f48834..a6d6d6dadde5 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -2575,6 +2575,9 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, { struct rq *src_rq = task_rq(p); struct rq *dst_rq = container_of(dst_dsq, struct rq, scx.local_dsq); +#ifdef CONFIG_SMP + struct rq *locked_rq = rq; +#endif /* * We're synchronized against dequeue through DISPATCHING. As @p can't @@ -2611,8 +2614,9 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, atomic_long_set_release(&p->scx.ops_state, SCX_OPSS_NONE); /* switch to @src_rq lock */ - if (rq != src_rq) { - raw_spin_rq_unlock(rq); + if (locked_rq != src_rq) { + raw_spin_rq_unlock(locked_rq); + locked_rq = src_rq; raw_spin_rq_lock(src_rq); } @@ -2630,6 +2634,8 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, } else { move_remote_task_to_local_dsq(p, enq_flags, src_rq, dst_rq); + /* task has been moved to dst_rq, which is now locked */ + locked_rq = dst_rq; } /* if the destination CPU is idle, wake it up */ @@ -2638,8 +2644,8 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, } /* switch back to @rq lock */ - if (rq != dst_rq) { - raw_spin_rq_unlock(dst_rq); + if (locked_rq != rq) { + raw_spin_rq_unlock(locked_rq); raw_spin_rq_lock(rq); } #else /* CONFIG_SMP */ From a37934ea75d331fafa7fe80b6180642ba5193422 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Wed, 22 Jan 2025 21:11:11 -0800 Subject: [PATCH 0229/1443] drm/xe/devcoredump: Move exec queue snapshot to Contexts section MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Having the exec queue snapshot inside a "GuC CT" section was always wrong. Commit c28fd6c358db ("drm/xe/devcoredump: Improve section headings and add tile info") tried to fix that bug, but with that also broke the mesa tool that parses the devcoredump, hence it was reverted in commit 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool"). With the mesa tool also fixed, this can propagate as a fix on both kernel and userspace side to avoid unnecessary headache for a debug feature. Cc: John Harrison Cc: Julia Filipchuk Cc: José Roberto de Souza Cc: stable@vger.kernel.org Fixes: 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") Reviewed-by: José Roberto de Souza Link: https://patchwork.freedesktop.org/patch/msgid/20250123051112.1938193-2-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_devcoredump.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index 81dc7795c065..a7946a76777e 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -119,11 +119,7 @@ static ssize_t __xe_devcoredump_read(char *buffer, size_t count, drm_puts(&p, "\n**** GuC CT ****\n"); xe_guc_ct_snapshot_print(ss->guc.ct, &p); - /* - * Don't add a new section header here because the mesa debug decoder - * tool expects the context information to be in the 'GuC CT' section. - */ - /* drm_puts(&p, "\n**** Contexts ****\n"); */ + drm_puts(&p, "\n**** Contexts ****\n"); xe_guc_exec_queue_snapshot_print(ss->ge, &p); drm_puts(&p, "\n**** Job ****\n"); From 2c95bbf5002776117a69caed3b31c10bf7341bec Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 12:22:03 -0800 Subject: [PATCH 0230/1443] drm/xe: Fix and re-enable xe_print_blob_ascii85() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") partially reverted some changes to workaround breakage caused to mesa tools. However, in doing so it also broke fetching the GuC log via debugfs since xe_print_blob_ascii85() simply bails out. The fix is to avoid the extra newlines: the devcoredump interface is line-oriented and adding random newlines in the middle breaks it. If a tool is able to parse it by looking at the data and checking for chars that are out of the ascii85 space, it can still do so. A format change that breaks the line-oriented output on devcoredump however needs better coordination with existing tools. v2: Add suffix description comment v3: Reword explanation of xe_print_blob_ascii85() calling drm_puts() in a loop Reviewed-by: José Roberto de Souza Cc: John Harrison Cc: Julia Filipchuk Cc: José Roberto de Souza Cc: stable@vger.kernel.org Fixes: 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") Fixes: ec1455ce7e35 ("drm/xe/devcoredump: Add ASCII85 dump helper function") Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-2-jose.souza@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_devcoredump.c | 36 +++++++++++------------------ drivers/gpu/drm/xe/xe_devcoredump.h | 2 +- drivers/gpu/drm/xe/xe_guc_ct.c | 3 ++- drivers/gpu/drm/xe/xe_guc_log.c | 4 +++- 4 files changed, 20 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index a7946a76777e..39fe485d2085 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -391,42 +391,34 @@ int xe_devcoredump_init(struct xe_device *xe) /** * xe_print_blob_ascii85 - print a BLOB to some useful location in ASCII85 * - * The output is split to multiple lines because some print targets, e.g. dmesg - * cannot handle arbitrarily long lines. Note also that printing to dmesg in - * piece-meal fashion is not possible, each separate call to drm_puts() has a - * line-feed automatically added! Therefore, the entire output line must be - * constructed in a local buffer first, then printed in one atomic output call. + * The output is split into multiple calls to drm_puts() because some print + * targets, e.g. dmesg, cannot handle arbitrarily long lines. These targets may + * add newlines, as is the case with dmesg: each drm_puts() call creates a + * separate line. * * There is also a scheduler yield call to prevent the 'task has been stuck for * 120s' kernel hang check feature from firing when printing to a slow target * such as dmesg over a serial port. * - * TODO: Add compression prior to the ASCII85 encoding to shrink huge buffers down. - * * @p: the printer object to output to * @prefix: optional prefix to add to output string + * @suffix: optional suffix to add at the end. 0 disables it and is + * not added to the output, which is useful when using multiple calls + * to dump data to @p * @blob: the Binary Large OBject to dump out * @offset: offset in bytes to skip from the front of the BLOB, must be a multiple of sizeof(u32) * @size: the size in bytes of the BLOB, must be a multiple of sizeof(u32) */ -void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, +void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char suffix, const void *blob, size_t offset, size_t size) { const u32 *blob32 = (const u32 *)blob; char buff[ASCII85_BUFSZ], *line_buff; size_t line_pos = 0; - /* - * Splitting blobs across multiple lines is not compatible with the mesa - * debug decoder tool. Note that even dropping the explicit '\n' below - * doesn't help because the GuC log is so big some underlying implementation - * still splits the lines at 512K characters. So just bail completely for - * the moment. - */ - return; - #define DMESG_MAX_LINE_LEN 800 -#define MIN_SPACE (ASCII85_BUFSZ + 2) /* 85 + "\n\0" */ + /* Always leave space for the suffix char and the \0 */ +#define MIN_SPACE (ASCII85_BUFSZ + 2) /* 85 + "\0" */ if (size & 3) drm_printf(p, "Size not word aligned: %zu", size); @@ -458,7 +450,6 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, line_pos += strlen(line_buff + line_pos); if ((line_pos + MIN_SPACE) >= DMESG_MAX_LINE_LEN) { - line_buff[line_pos++] = '\n'; line_buff[line_pos++] = 0; drm_puts(p, line_buff); @@ -470,10 +461,11 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, } } - if (line_pos) { - line_buff[line_pos++] = '\n'; - line_buff[line_pos++] = 0; + if (suffix) + line_buff[line_pos++] = suffix; + if (line_pos) { + line_buff[line_pos++] = 0; drm_puts(p, line_buff); } diff --git a/drivers/gpu/drm/xe/xe_devcoredump.h b/drivers/gpu/drm/xe/xe_devcoredump.h index 6a17e6d60102..5391a80a4d1b 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.h +++ b/drivers/gpu/drm/xe/xe_devcoredump.h @@ -29,7 +29,7 @@ static inline int xe_devcoredump_init(struct xe_device *xe) } #endif -void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, +void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char suffix, const void *blob, size_t offset, size_t size); #endif diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 8b65c5e959cc..50c8076b5158 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -1724,7 +1724,8 @@ void xe_guc_ct_snapshot_print(struct xe_guc_ct_snapshot *snapshot, snapshot->g2h_outstanding); if (snapshot->ctb) - xe_print_blob_ascii85(p, "CTB data", snapshot->ctb, 0, snapshot->ctb_size); + xe_print_blob_ascii85(p, "CTB data", '\n', + snapshot->ctb, 0, snapshot->ctb_size); } else { drm_puts(p, "CT disabled\n"); } diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index 80151ff6a71f..44482ea91992 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -207,8 +207,10 @@ void xe_guc_log_snapshot_print(struct xe_guc_log_snapshot *snapshot, struct drm_ remain = snapshot->size; for (i = 0; i < snapshot->num_chunks; i++) { size_t size = min(GUC_LOG_CHUNK_SIZE, remain); + const char *prefix = i ? NULL : "Log data"; + char suffix = i == snapshot->num_chunks - 1 ? '\n' : 0; - xe_print_blob_ascii85(p, i ? NULL : "Log data", snapshot->copy[i], 0, size); + xe_print_blob_ascii85(p, prefix, suffix, snapshot->copy[i], 0, size); remain -= size; } } From cb1f868ca13756c0c18ba54d1591332476760d07 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jos=C3=A9=20Roberto=20de=20Souza?= Date: Thu, 23 Jan 2025 12:22:04 -0800 Subject: [PATCH 0231/1443] drm/xe: Make GUC binaries dump consistent with other binaries in devcoredump MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit All other(hwsp, hwctx and vmas) binaries follow this format: [name].length: 0x1000 [name].data: xxxxxxx [name].error: errno The error one is just in case by some reason it was not able to capture the binary. So this GuC binaries should follow the same patern. v2: - renamed GUC binary to LOG Cc: John Harrison Cc: Lucas De Marchi Reviewed-by: Lucas De Marchi Signed-off-by: José Roberto de Souza Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-3-jose.souza@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_guc_ct.c | 6 ++++-- drivers/gpu/drm/xe/xe_guc_log.c | 3 ++- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 50c8076b5158..497036675a38 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -1723,9 +1723,11 @@ void xe_guc_ct_snapshot_print(struct xe_guc_ct_snapshot *snapshot, drm_printf(p, "\tg2h outstanding: %d\n", snapshot->g2h_outstanding); - if (snapshot->ctb) - xe_print_blob_ascii85(p, "CTB data", '\n', + if (snapshot->ctb) { + drm_printf(p, "[CTB].length: 0x%lx\n", snapshot->ctb_size); + xe_print_blob_ascii85(p, "[CTB].data", '\n', snapshot->ctb, 0, snapshot->ctb_size); + } } else { drm_puts(p, "CT disabled\n"); } diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index 44482ea91992..ab97ac728d7a 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -204,10 +204,11 @@ void xe_guc_log_snapshot_print(struct xe_guc_log_snapshot *snapshot, struct drm_ drm_printf(p, "GuC timestamp: 0x%08llX [%llu]\n", snapshot->stamp, snapshot->stamp); drm_printf(p, "Log level: %u\n", snapshot->level); + drm_printf(p, "[LOG].length: 0x%lx\n", snapshot->size); remain = snapshot->size; for (i = 0; i < snapshot->num_chunks; i++) { size_t size = min(GUC_LOG_CHUNK_SIZE, remain); - const char *prefix = i ? NULL : "Log data"; + const char *prefix = i ? NULL : "[LOG].data"; char suffix = i == snapshot->num_chunks - 1 ? '\n' : 0; xe_print_blob_ascii85(p, prefix, suffix, snapshot->copy[i], 0, size); From a9ab28b3d21aec6d0f56fe722953e20ce470237b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 28 Jan 2025 06:22:58 +0100 Subject: [PATCH 0232/1443] xfs: remove xfs_buf_cache.bc_lock xfs_buf_cache.bc_lock serializes adding buffers to and removing them from the hashtable. But as the rhashtable code already uses fine grained internal locking for inserts and removals the extra protection isn't actually required. It also happens to fix a lock order inversion vs b_lock added by the recent lookup race fix. Fixes: ee10f6fcdb96 ("xfs: fix buffer lookup vs release race") Reported-by: Lai, Yi Signed-off-by: Christoph Hellwig Reviewed-by: Carlos Maiolino Reviewed-by: Dave Chinner Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_buf.c | 31 +++++++++++++++++-------------- fs/xfs/xfs_buf.h | 1 - 2 files changed, 17 insertions(+), 15 deletions(-) diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index f1252ed8bd0a..ef207784876c 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -41,8 +41,7 @@ struct kmem_cache *xfs_buf_cache; * * xfs_buf_rele: * b_lock - * pag_buf_lock - * lru_lock + * lru_lock * * xfs_buftarg_drain_rele * lru_lock @@ -220,14 +219,21 @@ _xfs_buf_alloc( */ flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD); - spin_lock_init(&bp->b_lock); + /* + * A new buffer is held and locked by the owner. This ensures that the + * buffer is owned by the caller and racing RCU lookups right after + * inserting into the hash table are safe (and will have to wait for + * the unlock to do anything non-trivial). + */ bp->b_hold = 1; + sema_init(&bp->b_sema, 0); /* held, no waiters */ + + spin_lock_init(&bp->b_lock); atomic_set(&bp->b_lru_ref, 1); init_completion(&bp->b_iowait); INIT_LIST_HEAD(&bp->b_lru); INIT_LIST_HEAD(&bp->b_list); INIT_LIST_HEAD(&bp->b_li_list); - sema_init(&bp->b_sema, 0); /* held, no waiters */ bp->b_target = target; bp->b_mount = target->bt_mount; bp->b_flags = flags; @@ -497,7 +503,6 @@ int xfs_buf_cache_init( struct xfs_buf_cache *bch) { - spin_lock_init(&bch->bc_lock); return rhashtable_init(&bch->bc_hash, &xfs_buf_hash_params); } @@ -647,17 +652,20 @@ xfs_buf_find_insert( if (error) goto out_free_buf; - spin_lock(&bch->bc_lock); + /* The new buffer keeps the perag reference until it is freed. */ + new_bp->b_pag = pag; + + rcu_read_lock(); bp = rhashtable_lookup_get_insert_fast(&bch->bc_hash, &new_bp->b_rhash_head, xfs_buf_hash_params); if (IS_ERR(bp)) { + rcu_read_unlock(); error = PTR_ERR(bp); - spin_unlock(&bch->bc_lock); goto out_free_buf; } if (bp && xfs_buf_try_hold(bp)) { /* found an existing buffer */ - spin_unlock(&bch->bc_lock); + rcu_read_unlock(); error = xfs_buf_find_lock(bp, flags); if (error) xfs_buf_rele(bp); @@ -665,10 +673,8 @@ xfs_buf_find_insert( *bpp = bp; goto out_free_buf; } + rcu_read_unlock(); - /* The new buffer keeps the perag reference until it is freed. */ - new_bp->b_pag = pag; - spin_unlock(&bch->bc_lock); *bpp = new_bp; return 0; @@ -1085,7 +1091,6 @@ xfs_buf_rele_cached( } /* we are asked to drop the last reference */ - spin_lock(&bch->bc_lock); __xfs_buf_ioacct_dec(bp); if (!(bp->b_flags & XBF_STALE) && atomic_read(&bp->b_lru_ref)) { /* @@ -1097,7 +1102,6 @@ xfs_buf_rele_cached( bp->b_state &= ~XFS_BSTATE_DISPOSE; else bp->b_hold--; - spin_unlock(&bch->bc_lock); } else { bp->b_hold--; /* @@ -1115,7 +1119,6 @@ xfs_buf_rele_cached( ASSERT(!(bp->b_flags & _XBF_DELWRI_Q)); rhashtable_remove_fast(&bch->bc_hash, &bp->b_rhash_head, xfs_buf_hash_params); - spin_unlock(&bch->bc_lock); if (pag) xfs_perag_put(pag); freebuf = true; diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index 7e73663c5d4a..3b4ed42e11c0 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -80,7 +80,6 @@ typedef unsigned int xfs_buf_flags_t; #define XFS_BSTATE_IN_FLIGHT (1 << 1) /* I/O in flight */ struct xfs_buf_cache { - spinlock_t bc_lock; struct rhashtable bc_hash; }; From fd39c41bcd82d5ebaaebadb944eab5598c668a90 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Mon, 27 Jan 2025 14:44:14 +0100 Subject: [PATCH 0233/1443] drm/ast: astdp: Fix timeout for enabling video signal The ASTDP transmitter sometimes takes up to 1 second for enabling the video signal, while the timeout is only 200 msec. This results in a kernel error message. Increase the timeout to 1 second. An example of the error message is shown below. [ 697.084433] ------------[ cut here ]------------ [ 697.091115] ast 0000:02:00.0: [drm] drm_WARN_ON(!__ast_dp_wait_enable(ast, enabled)) [ 697.091233] WARNING: CPU: 1 PID: 160 at drivers/gpu/drm/ast/ast_dp.c:232 ast_dp_set_enable+0x123/0x140 [ast] [...] [ 697.272469] RIP: 0010:ast_dp_set_enable+0x123/0x140 [ast] [...] [ 697.415283] Call Trace: [ 697.420727] [ 697.425908] ? show_trace_log_lvl+0x196/0x2c0 [ 697.433304] ? show_trace_log_lvl+0x196/0x2c0 [ 697.440693] ? drm_atomic_helper_commit_modeset_enables+0x30a/0x470 [ 697.450115] ? ast_dp_set_enable+0x123/0x140 [ast] [ 697.458059] ? __warn.cold+0xaf/0xca [ 697.464713] ? ast_dp_set_enable+0x123/0x140 [ast] [ 697.472633] ? report_bug+0x134/0x1d0 [ 697.479544] ? handle_bug+0x58/0x90 [ 697.486127] ? exc_invalid_op+0x13/0x40 [ 697.492975] ? asm_exc_invalid_op+0x16/0x20 [ 697.500224] ? preempt_count_sub+0x14/0xc0 [ 697.507473] ? ast_dp_set_enable+0x123/0x140 [ast] [ 697.515377] ? ast_dp_set_enable+0x123/0x140 [ast] [ 697.523227] drm_atomic_helper_commit_modeset_enables+0x30a/0x470 [ 697.532388] drm_atomic_helper_commit_tail+0x58/0x90 [ 697.540400] ast_mode_config_helper_atomic_commit_tail+0x30/0x40 [ast] [ 697.550009] commit_tail+0xfe/0x1d0 [ 697.556547] drm_atomic_helper_commit+0x198/0x1c0 This is a cosmetical problem. Enabling the video signal still works even with the error message. The problem has always been present, but only recent versions of the ast driver warn about missing the timeout. Signed-off-by: Thomas Zimmermann Fixes: 4e29cc7c5c67 ("drm/ast: astdp: Replace ast_dp_set_on_off()") Cc: Thomas Zimmermann Cc: Jocelyn Falempe Cc: Dave Airlie Cc: dri-devel@lists.freedesktop.org Cc: # v6.13+ Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250127134423.84266-1-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_dp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 0e282b7b167c..b9eb67e3fa90 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -195,7 +195,7 @@ static bool __ast_dp_wait_enable(struct ast_device *ast, bool enabled) if (enabled) vgacrdf_test |= AST_IO_VGACRDF_DP_VIDEO_ENABLE; - for (i = 0; i < 200; ++i) { + for (i = 0; i < 1000; ++i) { if (i) mdelay(1); vgacrdf = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xdf, From 5e940312a2ac64ba0d6239aff72135226818b238 Mon Sep 17 00:00:00 2001 From: Riana Tauro Date: Tue, 28 Jan 2025 15:26:30 +0530 Subject: [PATCH 0234/1443] drm/xe: Add functions and sysfs for boot survivability MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Boot Survivability is a software based workflow for recovering a system in a failed boot state. Here system recoverability is concerned with recovering the firmware responsible for boot. This is implemented by loading the driver with bare minimum (no drm card) to allow the firmware to be flashed through mei-gsc and collect telemetry. The driver's probe flow is modified such that it enters survivability mode when pcode initialization is incomplete and boot status denotes a failure. In this mode, drm card is not exposed and presence of survivability_mode entry in PCI sysfs is used to indicate survivability mode and provide additional information required for debug This patch adds initialization functions and exposes admin readable sysfs entries The new sysfs will have the below layout /sys/bus/.../bdf ├── survivability_mode v2: reorder headers fix doc remove survivability info and use mode to display information use separate function for logging survivability information for critical error (Rodrigo) v3: use for loop use dev logs instead of drm use helper function for aux history(Rodrigo) remove unnecessary error check of greater than max_scratch as we are reading only 3 bit v4: fix checkpatch warnings fix space (Rodrigo) rename register Signed-off-by: Riana Tauro Acked-by: Ashwin Kumar Kulkarni Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250128095632.1294722-2-riana.tauro@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/Makefile | 1 + drivers/gpu/drm/xe/xe_device_types.h | 4 + drivers/gpu/drm/xe/xe_pcode_api.h | 14 ++ drivers/gpu/drm/xe/xe_survivability_mode.c | 215 ++++++++++++++++++ drivers/gpu/drm/xe/xe_survivability_mode.h | 17 ++ .../gpu/drm/xe/xe_survivability_mode_types.h | 35 +++ 6 files changed, 286 insertions(+) create mode 100644 drivers/gpu/drm/xe/xe_survivability_mode.c create mode 100644 drivers/gpu/drm/xe/xe_survivability_mode.h create mode 100644 drivers/gpu/drm/xe/xe_survivability_mode_types.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 9122e4318773..ef900b9f0457 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -96,6 +96,7 @@ xe-y += xe_bb.o \ xe_sa.o \ xe_sched_job.o \ xe_step.o \ + xe_survivability_mode.o \ xe_sync.o \ xe_tile.o \ xe_tile_sysfs.o \ diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 58e79e19deaa..89f532b67bc4 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -22,6 +22,7 @@ #include "xe_pt_types.h" #include "xe_sriov_types.h" #include "xe_step_types.h" +#include "xe_survivability_mode_types.h" #if IS_ENABLED(CONFIG_DRM_XE_DEBUG) #define TEST_VM_OPS_ERROR @@ -331,6 +332,9 @@ struct xe_device { u8 skip_pcode:1; } info; + /** @survivability: survivability information for device */ + struct xe_survivability survivability; + /** @irq: device interrupt state */ struct { /** @irq.lock: lock for processing irq's on this device */ diff --git a/drivers/gpu/drm/xe/xe_pcode_api.h b/drivers/gpu/drm/xe/xe_pcode_api.h index f153ce96f69a..2bae9afdbd35 100644 --- a/drivers/gpu/drm/xe/xe_pcode_api.h +++ b/drivers/gpu/drm/xe/xe_pcode_api.h @@ -49,6 +49,20 @@ /* Domain IDs (param2) */ #define PCODE_MBOX_DOMAIN_HBM 0x2 +#define PCODE_SCRATCH(x) XE_REG(0x138320 + ((x) * 4)) +/* PCODE_SCRATCH0 */ +#define AUXINFO_REG_OFFSET REG_GENMASK(17, 15) +#define OVERFLOW_REG_OFFSET REG_GENMASK(14, 12) +#define HISTORY_TRACKING REG_BIT(11) +#define OVERFLOW_SUPPORT REG_BIT(10) +#define AUXINFO_SUPPORT REG_BIT(9) +#define BOOT_STATUS REG_GENMASK(3, 1) +#define CRITICAL_FAILURE 4 +#define NON_CRITICAL_FAILURE 7 + +/* Auxiliary info bits */ +#define AUXINFO_HISTORY_OFFSET REG_GENMASK(31, 29) + struct pcode_err_decode { int errno; const char *str; diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c new file mode 100644 index 000000000000..9911e9f6b99b --- /dev/null +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -0,0 +1,215 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2025 Intel Corporation + */ + +#include "xe_survivability_mode.h" +#include "xe_survivability_mode_types.h" + +#include +#include +#include + +#include "xe_device.h" +#include "xe_gt.h" +#include "xe_mmio.h" +#include "xe_pcode_api.h" + +#define MAX_SCRATCH_MMIO 8 + +/** + * DOC: Xe Boot Survivability + * + * Boot Survivability is a software based workflow for recovering a system in a failed boot state + * Here system recoverability is concerned with recovering the firmware responsible for boot. + * + * This is implemented by loading the driver with bare minimum (no drm card) to allow the firmware + * to be flashed through mei and collect telemetry. The driver's probe flow is modified + * such that it enters survivability mode when pcode initialization is incomplete and boot status + * denotes a failure. The driver then populates the survivability_mode PCI sysfs indicating + * survivability mode and provides additional information required for debug + * + * KMD exposes below admin-only readable sysfs in survivability mode + * + * device/survivability_mode: The presence of this file indicates that the card is in survivability + * mode. Also, provides additional information on why the driver entered + * survivability mode. + * + * Capability Information - Provides boot status + * Postcode Information - Provides information about the failure + * Overflow Information - Provides history of previous failures + * Auxiliary Information - Certain failures may have information in + * addition to postcode information + */ + +static u32 aux_history_offset(u32 reg_value) +{ + return REG_FIELD_GET(AUXINFO_HISTORY_OFFSET, reg_value); +} + +static void set_survivability_info(struct xe_mmio *mmio, struct xe_survivability_info *info, + int id, char *name) +{ + strscpy(info[id].name, name, sizeof(info[id].name)); + info[id].reg = PCODE_SCRATCH(id).raw; + info[id].value = xe_mmio_read32(mmio, PCODE_SCRATCH(id)); +} + +static void populate_survivability_info(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct xe_survivability_info *info = survivability->info; + struct xe_mmio *mmio; + u32 id = 0, reg_value; + char name[NAME_MAX]; + int index; + + mmio = xe_root_tile_mmio(xe); + set_survivability_info(mmio, info, id, "Capability Info"); + reg_value = info[id].value; + + if (reg_value & HISTORY_TRACKING) { + id++; + set_survivability_info(mmio, info, id, "Postcode Info"); + + if (reg_value & OVERFLOW_SUPPORT) { + id = REG_FIELD_GET(OVERFLOW_REG_OFFSET, reg_value); + set_survivability_info(mmio, info, id, "Overflow Info"); + } + } + + if (reg_value & AUXINFO_SUPPORT) { + id = REG_FIELD_GET(AUXINFO_REG_OFFSET, reg_value); + + for (index = 0; id && reg_value; index++, reg_value = info[id].value, + id = aux_history_offset(reg_value)) { + snprintf(name, NAME_MAX, "Auxiliary Info %d", index); + set_survivability_info(mmio, info, id, name); + } + } +} + +static void log_survivability_info(struct pci_dev *pdev) +{ + struct xe_device *xe = pdev_to_xe_device(pdev); + struct xe_survivability *survivability = &xe->survivability; + struct xe_survivability_info *info = survivability->info; + int id; + + dev_info(&pdev->dev, "Survivability Boot Status : Critical Failure (%d)\n", + survivability->boot_status); + for (id = 0; id < MAX_SCRATCH_MMIO; id++) { + if (info[id].reg) + dev_info(&pdev->dev, "%s: 0x%x - 0x%x\n", info[id].name, + info[id].reg, info[id].value); + } +} + +static ssize_t survivability_mode_show(struct device *dev, + struct device_attribute *attr, char *buff) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct xe_device *xe = pdev_to_xe_device(pdev); + struct xe_survivability *survivability = &xe->survivability; + struct xe_survivability_info *info = survivability->info; + int index = 0, count = 0; + + for (index = 0; index < MAX_SCRATCH_MMIO; index++) { + if (info[index].reg) + count += sysfs_emit_at(buff, count, "%s: 0x%x - 0x%x\n", info[index].name, + info[index].reg, info[index].value); + } + + return count; +} + +static DEVICE_ATTR_ADMIN_RO(survivability_mode); + +static void enable_survivability_mode(struct pci_dev *pdev) +{ + struct device *dev = &pdev->dev; + struct xe_device *xe = pdev_to_xe_device(pdev); + struct xe_survivability *survivability = &xe->survivability; + int ret = 0; + + /* set survivability mode */ + survivability->mode = true; + dev_info(dev, "In Survivability Mode\n"); + + /* create survivability mode sysfs */ + ret = sysfs_create_file(&dev->kobj, &dev_attr_survivability_mode.attr); + if (ret) { + dev_warn(dev, "Failed to create survivability sysfs files\n"); + return; + } +} + +/** + * xe_survivability_mode_required - checks if survivability mode is required + * @xe: xe device instance + * + * This function reads the boot status from Pcode + * + * Return: true if boot status indicates failure, false otherwise + */ +bool xe_survivability_mode_required(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct xe_mmio *mmio = xe_root_tile_mmio(xe); + u32 data; + + data = xe_mmio_read32(mmio, PCODE_SCRATCH(0)); + survivability->boot_status = REG_FIELD_GET(BOOT_STATUS, data); + + return (survivability->boot_status == NON_CRITICAL_FAILURE || + survivability->boot_status == CRITICAL_FAILURE); +} + +/** + * xe_survivability_mode_remove - remove survivability mode + * @xe: xe device instance + * + * clean up sysfs entries of survivability mode + */ +void xe_survivability_mode_remove(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + struct device *dev = &pdev->dev; + + sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr); + kfree(survivability->info); + pci_set_drvdata(pdev, NULL); +} + +/** + * xe_survivability_mode_init - Initialize the survivability mode + * @xe: xe device instance + * + * Initializes survivability information and enables survivability mode + */ +void xe_survivability_mode_init(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + struct xe_survivability_info *info; + struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + + survivability->size = MAX_SCRATCH_MMIO; + + info = kcalloc(survivability->size, sizeof(*info), GFP_KERNEL); + if (!info) + return; + + survivability->info = info; + + populate_survivability_info(xe); + + /* Only log debug information and exit if it is a critical failure */ + if (survivability->boot_status == CRITICAL_FAILURE) { + log_survivability_info(pdev); + kfree(survivability->info); + return; + } + + enable_survivability_mode(pdev); +} diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h new file mode 100644 index 000000000000..410e3ee5f5d1 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_survivability_mode.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_SURVIVABILITY_MODE_H_ +#define _XE_SURVIVABILITY_MODE_H_ + +#include + +struct xe_device; + +void xe_survivability_mode_init(struct xe_device *xe); +void xe_survivability_mode_remove(struct xe_device *xe); +bool xe_survivability_mode_required(struct xe_device *xe); + +#endif /* _XE_SURVIVABILITY_MODE_H_ */ diff --git a/drivers/gpu/drm/xe/xe_survivability_mode_types.h b/drivers/gpu/drm/xe/xe_survivability_mode_types.h new file mode 100644 index 000000000000..19d433e253df --- /dev/null +++ b/drivers/gpu/drm/xe/xe_survivability_mode_types.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2025 Intel Corporation + */ + +#ifndef _XE_SURVIVABILITY_MODE_TYPES_H_ +#define _XE_SURVIVABILITY_MODE_TYPES_H_ + +#include +#include + +struct xe_survivability_info { + char name[NAME_MAX]; + u32 reg; + u32 value; +}; + +/** + * struct xe_survivability: Contains survivability mode information + */ +struct xe_survivability { + /** @info: struct that holds survivability info from scratch registers */ + struct xe_survivability_info *info; + + /** @size: number of scratch registers */ + u32 size; + + /** @boot_status: indicates critical/non critical boot failure */ + u8 boot_status; + + /** @mode: boolean to indicate survivability mode */ + bool mode; +}; + +#endif /* _XE_SURVIVABILITY_MODE_TYPES_H_ */ From 256daa32c9e0dcf924b3237e2165d8163f4d89cc Mon Sep 17 00:00:00 2001 From: Riana Tauro Date: Tue, 28 Jan 2025 15:26:31 +0530 Subject: [PATCH 0235/1443] drm/xe: Enable Boot Survivability mode Enable boot survivability mode if pcode initialization fails and if boot status indicates a failure. In this mode, drm card is not exposed and driver probe returns success after loading the bare minimum to allow firmware to be flashed via mei. v2: abstract survivability mode variable add BMG check inside function (Jani, Rodrigo) v3: return -EBUSY during system suspend (Anshuman) check survivability mode in pci probe only on error Signed-off-by: Riana Tauro Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250128095632.1294722-3-riana.tauro@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_device.c | 7 ++++++- drivers/gpu/drm/xe/xe_pci.c | 23 ++++++++++++++++++++-- drivers/gpu/drm/xe/xe_survivability_mode.c | 16 +++++++++++++++ drivers/gpu/drm/xe/xe_survivability_mode.h | 1 + 4 files changed, 44 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 90597a7ad0b0..37ea1c099a67 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -52,6 +52,7 @@ #include "xe_pmu.h" #include "xe_query.h" #include "xe_sriov.h" +#include "xe_survivability_mode.h" #include "xe_tile.h" #include "xe_ttm_stolen_mgr.h" #include "xe_ttm_sys_mgr.h" @@ -694,8 +695,12 @@ int xe_device_probe_early(struct xe_device *xe) update_device_info(xe); err = xe_pcode_probe_early(xe); - if (err) + if (err) { + if (xe_survivability_mode_required(xe)) + xe_survivability_mode_init(xe); + return err; + } err = wait_for_lmem_ready(xe); if (err) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index bf35a18bf5e7..f05cf26a088c 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -30,6 +30,7 @@ #include "xe_pm.h" #include "xe_sriov.h" #include "xe_step.h" +#include "xe_survivability_mode.h" #include "xe_tile.h" enum toggle_d3cold { @@ -761,6 +762,9 @@ static void xe_pci_remove(struct pci_dev *pdev) if (IS_SRIOV_PF(xe)) xe_pci_sriov_configure(pdev, 0); + if (xe_survivability_mode_enabled(xe)) + return xe_survivability_mode_remove(xe); + xe_device_remove(xe); xe_pm_runtime_fini(xe); pci_set_drvdata(pdev, NULL); @@ -833,8 +837,19 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return err; err = xe_device_probe_early(xe); - if (err) + + /* + * In Boot Survivability mode, no drm card is exposed + * and driver is loaded with bare minimum to allow + * for firmware to be flashed through mei. Return + * success if survivability mode is enabled. + */ + if (err) { + if (xe_survivability_mode_enabled(xe)) + return 0; + return err; + } err = xe_info_init(xe, desc->graphics, desc->media); if (err) @@ -921,9 +936,13 @@ static void d3cold_toggle(struct pci_dev *pdev, enum toggle_d3cold toggle) static int xe_pci_suspend(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); + struct xe_device *xe = pdev_to_xe_device(pdev); int err; - err = xe_pm_suspend(pdev_to_xe_device(pdev)); + if (xe_survivability_mode_enabled(xe)) + return -EBUSY; + + err = xe_pm_suspend(xe); if (err) return err; diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c index 9911e9f6b99b..633f5effa349 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.c +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -144,6 +144,19 @@ static void enable_survivability_mode(struct pci_dev *pdev) } } +/** + * xe_survivability_mode_enabled - check if survivability mode is enabled + * @xe: xe device instance + * + * Returns true if in survivability mode, false otherwise + */ +bool xe_survivability_mode_enabled(struct xe_device *xe) +{ + struct xe_survivability *survivability = &xe->survivability; + + return survivability->mode; +} + /** * xe_survivability_mode_required - checks if survivability mode is required * @xe: xe device instance @@ -158,6 +171,9 @@ bool xe_survivability_mode_required(struct xe_device *xe) struct xe_mmio *mmio = xe_root_tile_mmio(xe); u32 data; + if (!IS_DGFX(xe) || xe->info.platform < XE_BATTLEMAGE) + return false; + data = xe_mmio_read32(mmio, PCODE_SCRATCH(0)); survivability->boot_status = REG_FIELD_GET(BOOT_STATUS, data); diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h index 410e3ee5f5d1..f530507a22c6 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.h +++ b/drivers/gpu/drm/xe/xe_survivability_mode.h @@ -12,6 +12,7 @@ struct xe_device; void xe_survivability_mode_init(struct xe_device *xe); void xe_survivability_mode_remove(struct xe_device *xe); +bool xe_survivability_mode_enabled(struct xe_device *xe); bool xe_survivability_mode_required(struct xe_device *xe); #endif /* _XE_SURVIVABILITY_MODE_H_ */ From 8b47c9cdb6a78364fe68f8af0abfd6f265577001 Mon Sep 17 00:00:00 2001 From: Riana Tauro Date: Tue, 28 Jan 2025 15:26:32 +0530 Subject: [PATCH 0236/1443] drm/xe: Initialize mei-gsc and vsec in survivability mode Initialize mei-gsc in survivability mode and disable HECI interrupts. Also initialize vsec in survivability mode Signed-off-by: Riana Tauro Reviewed-by: Rodrigo Vivi Reviewed-by: Alexander Usyskin Link: https://patchwork.freedesktop.org/patch/msgid/20250128095632.1294722-4-riana.tauro@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_heci_gsc.c | 3 ++- drivers/gpu/drm/xe/xe_survivability_mode.c | 7 +++++++ 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_heci_gsc.c b/drivers/gpu/drm/xe/xe_heci_gsc.c index d765bfd3636b..06dc78d3a812 100644 --- a/drivers/gpu/drm/xe/xe_heci_gsc.c +++ b/drivers/gpu/drm/xe/xe_heci_gsc.c @@ -12,6 +12,7 @@ #include "xe_drv.h" #include "xe_heci_gsc.h" #include "xe_platform_types.h" +#include "xe_survivability_mode.h" #define GSC_BAR_LENGTH 0x00000FFC @@ -200,7 +201,7 @@ void xe_heci_gsc_init(struct xe_device *xe) return; } - if (!def->use_polling) { + if (!def->use_polling && !xe_survivability_mode_enabled(xe)) { ret = heci_gsc_irq_setup(xe); if (ret) goto fail; diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c index 633f5effa349..c619560af74f 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.c +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -12,8 +12,10 @@ #include "xe_device.h" #include "xe_gt.h" +#include "xe_heci_gsc.h" #include "xe_mmio.h" #include "xe_pcode_api.h" +#include "xe_vsec.h" #define MAX_SCRATCH_MMIO 8 @@ -142,6 +144,10 @@ static void enable_survivability_mode(struct pci_dev *pdev) dev_warn(dev, "Failed to create survivability sysfs files\n"); return; } + + xe_heci_gsc_init(xe); + + xe_vsec_init(xe); } /** @@ -194,6 +200,7 @@ void xe_survivability_mode_remove(struct xe_device *xe) struct device *dev = &pdev->dev; sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr); + xe_heci_gsc_fini(xe); kfree(survivability->info); pci_set_drvdata(pdev, NULL); } From 7bf6b497a747b0e28a411beacdd62f1488d0781c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 28 Jan 2025 08:55:33 +0100 Subject: [PATCH 0237/1443] nvmet: the result field in nvmet_alloc_ctrl_args is little endian So use the __le32 type for it. Fixes: 6202783184bf ("nvmet: Improve nvmet_alloc_ctrl() interface and implementation") Signed-off-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Keith Busch --- drivers/nvme/target/nvmet.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h index f4df458df9db..6a9af4e4d732 100644 --- a/drivers/nvme/target/nvmet.h +++ b/drivers/nvme/target/nvmet.h @@ -582,7 +582,7 @@ struct nvmet_alloc_ctrl_args { const struct nvmet_fabrics_ops *ops; struct device *p2p_client; u32 kato; - u32 result; + __le32 result; u16 error_loc; u16 status; }; From cc3d4671a0db9499b201c43faba6c46e1a21274c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 28 Jan 2025 08:55:34 +0100 Subject: [PATCH 0238/1443] nvmet: add a missing endianess conversion in nvmet_execute_admin_connect The kato field is little endian on the wire, but native endian in the in-core structure, add the missing byte swap. Fixes: 6202783184bf ("nvmet: Improve nvmet_alloc_ctrl() interface and implementation") Signed-off-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Keith Busch --- drivers/nvme/target/fabrics-cmd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/target/fabrics-cmd.c b/drivers/nvme/target/fabrics-cmd.c index a7ff05b3be29..eb406c90c167 100644 --- a/drivers/nvme/target/fabrics-cmd.c +++ b/drivers/nvme/target/fabrics-cmd.c @@ -287,7 +287,7 @@ static void nvmet_execute_admin_connect(struct nvmet_req *req) args.subsysnqn = d->subsysnqn; args.hostnqn = d->hostnqn; args.hostid = &d->hostid; - args.kato = c->kato; + args.kato = le32_to_cpu(c->kato); ctrl = nvmet_alloc_ctrl(&args); if (!ctrl) From f8864e27e363b76d1bb4d5825d837d62d5b85bff Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 28 Jan 2025 14:14:43 +0200 Subject: [PATCH 0239/1443] MAINTAINERS: Update drm/rcar-du maintainers Update drm/rcar-du maintainer entries: * Add myself as drm/rcar-du maintainer. * Update Laurent's email to include +renesas. * Switch Kieran from a maintainer to reviewer. * Change rcar-du to be under drm-misc umbrella. Reviewed-by: Laurent Pinchart Acked-by: Kieran Bingham Acked-by: Krzysztof Kozlowski Acked-by: Maxime Ripard Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20250128-rcar-du-maintainers-v2-1-4a3860a3e1ef@ideasonboard.com --- MAINTAINERS | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index ed86d884ee0d..b256bb4f6d3f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7604,7 +7604,6 @@ X: drivers/gpu/drm/mediatek/ X: drivers/gpu/drm/msm/ X: drivers/gpu/drm/nouveau/ X: drivers/gpu/drm/radeon/ -X: drivers/gpu/drm/renesas/rcar-du/ X: drivers/gpu/drm/tegra/ DRM DRIVERS FOR ALLWINNER A10 @@ -7759,12 +7758,13 @@ F: include/linux/host1x.h F: include/uapi/drm/tegra_drm.h DRM DRIVERS FOR RENESAS R-CAR -M: Laurent Pinchart -M: Kieran Bingham +M: Laurent Pinchart +M: Tomi Valkeinen +R: Kieran Bingham L: dri-devel@lists.freedesktop.org L: linux-renesas-soc@vger.kernel.org S: Supported -T: git git://linuxtv.org/pinchartl/media drm/du/next +T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/bridge/renesas,dsi-csi2-tx.yaml F: Documentation/devicetree/bindings/display/bridge/renesas,dw-hdmi.yaml F: Documentation/devicetree/bindings/display/bridge/renesas,lvds.yaml From aa77a01603ccc94de82441d1425863e2f30821a5 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 28 Jan 2025 10:47:18 +0100 Subject: [PATCH 0240/1443] dt-bindings: display: renesas,du: narrow interrupts and resets per variants Each variable-length property like interrupts or resets must have fixed constraints on number of items for given variant in binding. The clauses in "if:then:" block should define both limits: upper and lower. Acked-by: Conor Dooley Signed-off-by: Krzysztof Kozlowski Reviewed-by: Tomi Valkeinen Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20250128094719.63776-1-krzysztof.kozlowski@linaro.org --- .../devicetree/bindings/display/renesas,du.yaml | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/Documentation/devicetree/bindings/display/renesas,du.yaml b/Documentation/devicetree/bindings/display/renesas,du.yaml index 3880b4c2ea9a..88ecabc4348d 100644 --- a/Documentation/devicetree/bindings/display/renesas,du.yaml +++ b/Documentation/devicetree/bindings/display/renesas,du.yaml @@ -174,6 +174,7 @@ allOf: - pattern: '^dclkin\.[01]$' interrupts: + minItems: 2 maxItems: 2 resets: @@ -229,6 +230,7 @@ allOf: - pattern: '^dclkin\.[01]$' interrupts: + minItems: 2 maxItems: 2 resets: @@ -282,6 +284,7 @@ allOf: - pattern: '^dclkin\.[01]$' interrupts: + minItems: 2 maxItems: 2 resets: @@ -336,6 +339,7 @@ allOf: - pattern: '^dclkin\.[01]$' interrupts: + minItems: 2 maxItems: 2 resets: @@ -397,6 +401,7 @@ allOf: - pattern: '^dclkin\.[012]$' interrupts: + minItems: 3 maxItems: 3 resets: @@ -461,9 +466,11 @@ allOf: - pattern: '^dclkin\.[0123]$' interrupts: + minItems: 4 maxItems: 4 resets: + minItems: 2 maxItems: 2 reset-names: @@ -534,9 +541,11 @@ allOf: - pattern: '^dclkin\.[012]$' interrupts: + minItems: 3 maxItems: 3 resets: + minItems: 2 maxItems: 2 reset-names: @@ -605,9 +614,11 @@ allOf: - pattern: '^dclkin\.[013]$' interrupts: + minItems: 3 maxItems: 3 resets: + minItems: 2 maxItems: 2 reset-names: @@ -726,6 +737,7 @@ allOf: - pattern: '^dclkin\.[01]$' interrupts: + minItems: 2 maxItems: 2 resets: From c51549cf16d39398499445f6b0004cf84d158265 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 28 Jan 2025 10:47:19 +0100 Subject: [PATCH 0241/1443] dt-bindings: display: renesas,du: add top-level constraints Properties with variable number of items per each device are expected to have widest constraints in top-level "properties:" block and further customized (narrowed) in "if:then:". Add missing top-level constraints for clocks, clock-names, interrupts, resets, reset-names, renesas,cmms and renesas,vsps. Reviewed-by: Laurent Pinchart Acked-by: Conor Dooley Signed-off-by: Krzysztof Kozlowski Reviewed-by: Tomi Valkeinen Signed-off-by: Tomi Valkeinen Link: https://patchwork.freedesktop.org/patch/msgid/20250128094719.63776-2-krzysztof.kozlowski@linaro.org --- .../bindings/display/renesas,du.yaml | 24 +++++++++++++++---- 1 file changed, 19 insertions(+), 5 deletions(-) diff --git a/Documentation/devicetree/bindings/display/renesas,du.yaml b/Documentation/devicetree/bindings/display/renesas,du.yaml index 88ecabc4348d..c27dfea7fc62 100644 --- a/Documentation/devicetree/bindings/display/renesas,du.yaml +++ b/Documentation/devicetree/bindings/display/renesas,du.yaml @@ -47,12 +47,26 @@ properties: maxItems: 1 # See compatible-specific constraints below. - clocks: true - clock-names: true + clocks: + minItems: 1 + maxItems: 8 + + clock-names: + minItems: 1 + maxItems: 8 + interrupts: + minItems: 1 + maxItems: 4 description: Interrupt specifiers, one per DU channel - resets: true - reset-names: true + + resets: + minItems: 1 + maxItems: 2 + + reset-names: + minItems: 1 + maxItems: 2 power-domains: maxItems: 1 @@ -74,7 +88,7 @@ properties: renesas,cmms: $ref: /schemas/types.yaml#/definitions/phandle-array - minItems: 1 + minItems: 2 maxItems: 4 items: maxItems: 1 From 0e9724d0f89e8d77fa683e3129cadaed7c6e609d Mon Sep 17 00:00:00 2001 From: Aditya Garg Date: Mon, 20 Jan 2025 16:50:47 +0000 Subject: [PATCH 0242/1443] wifi: brcmfmac: use random seed flag for BCM4355 and BCM4364 firmware Before 6.13, random seed to the firmware was given based on the logic whether the device had valid OTP or not, and such devices were found mainly on the T2 and Apple Silicon Macs. In 6.13, the logic was changed, and the device table was used for this purpose, so as to cover the special case of BCM43752 chip. During the transition, the device table for BCM4364 and BCM4355 Wi-Fi chips which had valid OTP was not modified, thus breaking Wi-Fi on these devices. This patch adds does the necessary changes, similar to the ones done for other chips. Fixes: ea11a89c3ac6 ("wifi: brcmfmac: add flag for random seed during firmware download") Cc: stable@vger.kernel.org Signed-off-by: Aditya Garg Acked-by: Arend van Spriel Signed-off-by: Kalle Valo Link: https://patch.msgid.link/47E43F07-E11D-478C-86D4-23627154AC7C@live.com --- drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c index e4395b1f8c11..d2caa80e9412 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c @@ -2712,7 +2712,7 @@ static const struct pci_device_id brcmf_pcie_devid_table[] = { BRCMF_PCIE_DEVICE(BRCM_PCIE_4350_DEVICE_ID, WCC), BRCMF_PCIE_DEVICE_SUB(0x4355, BRCM_PCIE_VENDOR_ID_BROADCOM, 0x4355, WCC), BRCMF_PCIE_DEVICE(BRCM_PCIE_4354_RAW_DEVICE_ID, WCC), - BRCMF_PCIE_DEVICE(BRCM_PCIE_4355_DEVICE_ID, WCC), + BRCMF_PCIE_DEVICE(BRCM_PCIE_4355_DEVICE_ID, WCC_SEED), BRCMF_PCIE_DEVICE(BRCM_PCIE_4356_DEVICE_ID, WCC), BRCMF_PCIE_DEVICE(BRCM_PCIE_43567_DEVICE_ID, WCC), BRCMF_PCIE_DEVICE(BRCM_PCIE_43570_DEVICE_ID, WCC), @@ -2723,7 +2723,7 @@ static const struct pci_device_id brcmf_pcie_devid_table[] = { BRCMF_PCIE_DEVICE(BRCM_PCIE_43602_2G_DEVICE_ID, WCC), BRCMF_PCIE_DEVICE(BRCM_PCIE_43602_5G_DEVICE_ID, WCC), BRCMF_PCIE_DEVICE(BRCM_PCIE_43602_RAW_DEVICE_ID, WCC), - BRCMF_PCIE_DEVICE(BRCM_PCIE_4364_DEVICE_ID, WCC), + BRCMF_PCIE_DEVICE(BRCM_PCIE_4364_DEVICE_ID, WCC_SEED), BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_DEVICE_ID, BCA), BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_2G_DEVICE_ID, BCA), BRCMF_PCIE_DEVICE(BRCM_PCIE_4365_5G_DEVICE_ID, BCA), From f4c9c2cc827d803159730b1da813a0c595969831 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Tue, 28 Jan 2025 16:11:06 +0100 Subject: [PATCH 0243/1443] batman-adv: Fix incorrect offset in batadv_tt_tvlv_ogm_handler_v1() Since commit 4436df478860 ("batman-adv: Add flex array to struct batadv_tvlv_tt_data"), the introduction of batadv_tvlv_tt_data's flex array member in batadv_tt_tvlv_ogm_handler_v1() put tt_changes at invalid offset. Those TT changes are supposed to be filled from the end of batadv_tvlv_tt_data structure (including vlan_data flexible array), but only the flex array size is taken into account missing completely the size of the fixed part of the structure itself. Fix the tt_change offset computation by using struct_size() instead of flex_array_size() so both flex array member and its container structure sizes are taken into account. Cc: stable@vger.kernel.org Fixes: 4436df478860 ("batman-adv: Add flex array to struct batadv_tvlv_tt_data") Signed-off-by: Remi Pommarel Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 760d51fdbdf6..7d5de4cbb814 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -3959,23 +3959,21 @@ static void batadv_tt_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv, struct batadv_tvlv_tt_change *tt_change; struct batadv_tvlv_tt_data *tt_data; u16 num_entries, num_vlan; - size_t flex_size; + size_t tt_data_sz; if (tvlv_value_len < sizeof(*tt_data)) return; tt_data = tvlv_value; - tvlv_value_len -= sizeof(*tt_data); - num_vlan = ntohs(tt_data->num_vlan); - flex_size = flex_array_size(tt_data, vlan_data, num_vlan); - if (tvlv_value_len < flex_size) + tt_data_sz = struct_size(tt_data, vlan_data, num_vlan); + if (tvlv_value_len < tt_data_sz) return; tt_change = (struct batadv_tvlv_tt_change *)((void *)tt_data - + flex_size); - tvlv_value_len -= flex_size; + + tt_data_sz); + tvlv_value_len -= tt_data_sz; num_entries = batadv_tt_entries(tvlv_value_len); From fa8ffaae1b15236b8afb0fbbc04117ff7c900a83 Mon Sep 17 00:00:00 2001 From: Shekhar Chauhan Date: Tue, 28 Jan 2025 21:50:15 +0530 Subject: [PATCH 0244/1443] drm/xe/bmg: Add new PCI IDs Add 3 new PCI IDs for BMG. v2: Fix typo -> Replace '.' with ',' Signed-off-by: Shekhar Chauhan Reviewed-by: Clint Taylor Link: https://patchwork.freedesktop.org/patch/msgid/20250128162015.3288675-1-shekhar.chauhan@intel.com Signed-off-by: Rodrigo Vivi --- include/drm/intel/pciids.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h index 7883384acd5e..e53d7d9bb64d 100644 --- a/include/drm/intel/pciids.h +++ b/include/drm/intel/pciids.h @@ -817,7 +817,10 @@ MACRO__(0xE20B, ## __VA_ARGS__), \ MACRO__(0xE20C, ## __VA_ARGS__), \ MACRO__(0xE20D, ## __VA_ARGS__), \ - MACRO__(0xE212, ## __VA_ARGS__) + MACRO__(0xE210, ## __VA_ARGS__), \ + MACRO__(0xE212, ## __VA_ARGS__), \ + MACRO__(0xE215, ## __VA_ARGS__), \ + MACRO__(0xE216, ## __VA_ARGS__) /* PTL */ #define INTEL_PTL_IDS(MACRO__, ...) \ From 16016ade13f691da315fac7b23ebf1ab7b28b7ab Mon Sep 17 00:00:00 2001 From: Matt Atwood Date: Tue, 28 Jan 2025 09:51:02 -0800 Subject: [PATCH 0245/1443] drm/xe/ptl: Update the PTL pci id table Update to current bspec table. Bspec: 72574 Signed-off-by: Matt Atwood Reviewed-by: Clint Taylor Link: https://patchwork.freedesktop.org/patch/msgid/20250128175102.45797-1-matthew.s.atwood@intel.com Signed-off-by: Rodrigo Vivi --- include/drm/intel/pciids.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h index e53d7d9bb64d..23f349f650af 100644 --- a/include/drm/intel/pciids.h +++ b/include/drm/intel/pciids.h @@ -827,12 +827,10 @@ MACRO__(0xB080, ## __VA_ARGS__), \ MACRO__(0xB081, ## __VA_ARGS__), \ MACRO__(0xB082, ## __VA_ARGS__), \ + MACRO__(0xB083, ## __VA_ARGS__), \ + MACRO__(0xB08F, ## __VA_ARGS__), \ MACRO__(0xB090, ## __VA_ARGS__), \ - MACRO__(0xB091, ## __VA_ARGS__), \ - MACRO__(0xB092, ## __VA_ARGS__), \ MACRO__(0xB0A0, ## __VA_ARGS__), \ - MACRO__(0xB0A1, ## __VA_ARGS__), \ - MACRO__(0xB0A2, ## __VA_ARGS__), \ MACRO__(0xB0B0, ## __VA_ARGS__) #endif /* __PCIIDS_H__ */ From c28f72c6ca98e039c2aa5aac6752c416bc31dbab Mon Sep 17 00:00:00 2001 From: Jocelyn Falempe Date: Fri, 24 Jan 2025 15:11:31 +0100 Subject: [PATCH 0246/1443] drm/ast: Fix ast_dp connection status ast_dp_is_connected() used to also check for link training success to report the DP connector as connected. Without this check, the physical_status is always connected. So if no monitor is present, it will fail to read the EDID and set the default resolution to 640x480 instead of 1024x768. Signed-off-by: Jocelyn Falempe Fixes: 2281475168d2 ("drm/ast: astdp: Perform link training during atomic_enable") Reported-by: Jerry Hoemann Tested-by: Jose Lopez Cc: Thomas Zimmermann Cc: Dave Airlie Cc: dri-devel@lists.freedesktop.org Cc: # v6.12+ Reviewed-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20250124141142.2434138-1-jfalempe@redhat.com --- drivers/gpu/drm/ast/ast_dp.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 0e282b7b167c..30aad5c0112a 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -17,6 +17,12 @@ static bool ast_astdp_is_connected(struct ast_device *ast) { if (!ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xDF, AST_IO_VGACRDF_HPD)) return false; + /* + * HPD might be set even if no monitor is connected, so also check that + * the link training was successful. + */ + if (!ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xDC, AST_IO_VGACRDC_LINK_SUCCESS)) + return false; return true; } From 55d4b69861e853ac987f6d593b44a3c18b468576 Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Tue, 28 Jan 2025 17:32:48 -0500 Subject: [PATCH 0247/1443] Revert "drm/xe/lnl: Enable GuC SLPC DCC task" This reverts commit 50554bf3e56dd0c78ef1eedb685d0ab36c9c9987. DCC in LNL should be disabled. It was a mistake to decide to go against GuC platform defaults in this case and this could lead to regressions in some TDP limited scenarios instead of helping. Cc: Vinay Belgaumkar Cc: Jonathan Cavitt Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20250128223248.660748-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_pc.c | 25 ------------------------- 1 file changed, 25 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 44b5211066ef..43f9617baba2 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -993,27 +993,6 @@ static int pc_init_freqs(struct xe_guc_pc *pc) return ret; } -static int slpc_enable_dcc(struct xe_guc_pc *pc) -{ - int ret; - - ret = pc_action_set_param(pc, SLPC_PARAM_TASK_ENABLE_DCC, 1); - if (ret) - return ret; - - return pc_action_set_param(pc, SLPC_PARAM_TASK_DISABLE_DCC, 0); -} - -static int slpc_set_policies(struct xe_guc_pc *pc) -{ - struct xe_device *xe = pc_to_xe(pc); - - if (xe->info.platform == XE_LUNARLAKE) - return slpc_enable_dcc(pc); - - return 0; -} - /** * xe_guc_pc_start - Start GuC's Power Conservation component * @pc: Xe_GuC_PC instance @@ -1058,10 +1037,6 @@ int xe_guc_pc_start(struct xe_guc_pc *pc) goto out; } - ret = slpc_set_policies(pc); - if (ret) - goto out; - ret = pc_init_freqs(pc); if (ret) goto out; From 7332537962956fab2c055b37e5e2e6a0d2a8d6bf Mon Sep 17 00:00:00 2001 From: Jared Kangas Date: Tue, 21 Jan 2025 06:25:04 -0800 Subject: [PATCH 0248/1443] bpf: Remove unnecessary BTF lookups in bpf_sk_storage_tracing_allowed When loading BPF programs, bpf_sk_storage_tracing_allowed() does a series of lookups to get a type name from the program's attach_btf_id, making the assumption that the type is present in the vmlinux BTF along the way. However, this results in btf_type_by_id() returning a null pointer if a non-vmlinux kernel BTF is attached to. Proof-of-concept on a kernel with CONFIG_IPV6=m: $ cat bpfcrash.c #include #include #include static int bpf(enum bpf_cmd cmd, union bpf_attr *attr) { return syscall(__NR_bpf, cmd, attr, sizeof(*attr)); } int main(void) { const int btf_fd = bpf(BPF_BTF_GET_FD_BY_ID, &(union bpf_attr) { .btf_id = BTF_ID, }); if (btf_fd < 0) return 1; const int bpf_sk_storage_get = 107; const struct bpf_insn insns[] = { { .code = BPF_JMP | BPF_CALL, .imm = bpf_sk_storage_get}, { .code = BPF_JMP | BPF_EXIT }, }; return bpf(BPF_PROG_LOAD, &(union bpf_attr) { .prog_type = BPF_PROG_TYPE_TRACING, .expected_attach_type = BPF_TRACE_FENTRY, .license = (unsigned long)"GPL", .insns = (unsigned long)&insns, .insn_cnt = sizeof(insns) / sizeof(insns[0]), .attach_btf_obj_fd = btf_fd, .attach_btf_id = TYPE_ID, }); } $ sudo bpftool btf list | grep ipv6 2: name [ipv6] size 928200B $ sudo bpftool btf dump id 2 | awk '$3 ~ /inet6_sock_destruct/' [130689] FUNC 'inet6_sock_destruct' type_id=130677 linkage=static $ gcc -D_DEFAULT_SOURCE -DBTF_ID=2 -DTYPE_ID=130689 \ bpfcrash.c -o bpfcrash $ sudo ./bpfcrash This causes a null pointer dereference: Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 Call trace: bpf_sk_storage_tracing_allowed+0x8c/0xb0 P check_helper_call.isra.0+0xa8/0x1730 do_check+0xa18/0xb40 do_check_common+0x140/0x640 bpf_check+0xb74/0xcb8 bpf_prog_load+0x598/0x9a8 __sys_bpf+0x580/0x980 __arm64_sys_bpf+0x28/0x40 invoke_syscall.constprop.0+0x54/0xe8 do_el0_svc+0xb4/0xd0 el0_svc+0x44/0x1f8 el0t_64_sync_handler+0x13c/0x160 el0t_64_sync+0x184/0x188 Resolve this by using prog->aux->attach_func_name and removing the lookups. Fixes: 8e4597c627fb ("bpf: Allow using bpf_sk_storage in FENTRY/FEXIT/RAW_TP") Suggested-by: Martin KaFai Lau Signed-off-by: Jared Kangas Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20250121142504.1369436-1-jkangas@redhat.com Signed-off-by: Alexei Starovoitov --- net/core/bpf_sk_storage.c | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-) diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 7d41cde1bcca..2e538399757f 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -355,11 +355,6 @@ const struct bpf_func_proto bpf_sk_storage_delete_proto = { static bool bpf_sk_storage_tracing_allowed(const struct bpf_prog *prog) { - const struct btf *btf_vmlinux; - const struct btf_type *t; - const char *tname; - u32 btf_id; - if (prog->aux->dst_prog) return false; @@ -374,13 +369,7 @@ static bool bpf_sk_storage_tracing_allowed(const struct bpf_prog *prog) return true; case BPF_TRACE_FENTRY: case BPF_TRACE_FEXIT: - btf_vmlinux = bpf_get_btf_vmlinux(); - if (IS_ERR_OR_NULL(btf_vmlinux)) - return false; - btf_id = prog->aux->attach_btf_id; - t = btf_type_by_id(btf_vmlinux, btf_id); - tname = btf_name_by_offset(btf_vmlinux, t->name_off); - return !!strncmp(tname, "bpf_sk_storage", + return !!strncmp(prog->aux->attach_func_name, "bpf_sk_storage", strlen("bpf_sk_storage")); default: return false; From 6b3d638ca897e099fa99bd6d02189d3176f80a47 Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Wed, 22 Jan 2025 00:06:42 +0900 Subject: [PATCH 0249/1443] bpf, test_run: Fix use-after-free issue in eth_skb_pkt_type() KMSAN reported a use-after-free issue in eth_skb_pkt_type()[1]. The cause of the issue was that eth_skb_pkt_type() accessed skb's data that didn't contain an Ethernet header. This occurs when bpf_prog_test_run_xdp() passes an invalid value as the user_data argument to bpf_test_init(). Fix this by returning an error when user_data is less than ETH_HLEN in bpf_test_init(). Additionally, remove the check for "if (user_size > size)" as it is unnecessary. [1] BUG: KMSAN: use-after-free in eth_skb_pkt_type include/linux/etherdevice.h:627 [inline] BUG: KMSAN: use-after-free in eth_type_trans+0x4ee/0x980 net/ethernet/eth.c:165 eth_skb_pkt_type include/linux/etherdevice.h:627 [inline] eth_type_trans+0x4ee/0x980 net/ethernet/eth.c:165 __xdp_build_skb_from_frame+0x5a8/0xa50 net/core/xdp.c:635 xdp_recv_frames net/bpf/test_run.c:272 [inline] xdp_test_run_batch net/bpf/test_run.c:361 [inline] bpf_test_run_xdp_live+0x2954/0x3330 net/bpf/test_run.c:390 bpf_prog_test_run_xdp+0x148e/0x1b10 net/bpf/test_run.c:1318 bpf_prog_test_run+0x5b7/0xa30 kernel/bpf/syscall.c:4371 __sys_bpf+0x6a6/0xe20 kernel/bpf/syscall.c:5777 __do_sys_bpf kernel/bpf/syscall.c:5866 [inline] __se_sys_bpf kernel/bpf/syscall.c:5864 [inline] __x64_sys_bpf+0xa4/0xf0 kernel/bpf/syscall.c:5864 x64_sys_call+0x2ea0/0x3d90 arch/x86/include/generated/asm/syscalls_64.h:322 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xd9/0x1d0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Uninit was created at: free_pages_prepare mm/page_alloc.c:1056 [inline] free_unref_page+0x156/0x1320 mm/page_alloc.c:2657 __free_pages+0xa3/0x1b0 mm/page_alloc.c:4838 bpf_ringbuf_free kernel/bpf/ringbuf.c:226 [inline] ringbuf_map_free+0xff/0x1e0 kernel/bpf/ringbuf.c:235 bpf_map_free kernel/bpf/syscall.c:838 [inline] bpf_map_free_deferred+0x17c/0x310 kernel/bpf/syscall.c:862 process_one_work kernel/workqueue.c:3229 [inline] process_scheduled_works+0xa2b/0x1b60 kernel/workqueue.c:3310 worker_thread+0xedf/0x1550 kernel/workqueue.c:3391 kthread+0x535/0x6b0 kernel/kthread.c:389 ret_from_fork+0x6e/0x90 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 CPU: 1 UID: 0 PID: 17276 Comm: syz.1.16450 Not tainted 6.12.0-05490-g9bb88c659673 #8 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-3.fc41 04/01/2014 Fixes: be3d72a2896c ("bpf: move user_size out of bpf_test_init") Reported-by: syzkaller Suggested-by: Martin KaFai Lau Signed-off-by: Shigeru Yoshida Signed-off-by: Martin KaFai Lau Acked-by: Stanislav Fomichev Acked-by: Daniel Borkmann Link: https://patch.msgid.link/20250121150643.671650-1-syoshida@redhat.com Signed-off-by: Alexei Starovoitov --- net/bpf/test_run.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 8f6f7db48d4e..7cb192cbd65f 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -660,12 +660,9 @@ static void *bpf_test_init(const union bpf_attr *kattr, u32 user_size, void __user *data_in = u64_to_user_ptr(kattr->test.data_in); void *data; - if (size < ETH_HLEN || size > PAGE_SIZE - headroom - tailroom) + if (user_size < ETH_HLEN || user_size > PAGE_SIZE - headroom - tailroom) return ERR_PTR(-EINVAL); - if (user_size > size) - return ERR_PTR(-EMSGSIZE); - size = SKB_DATA_ALIGN(size); data = kzalloc(size + headroom + tailroom, GFP_USER); if (!data) From c7f2188d68c114095660a950b7e880a1e5a71c8f Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Wed, 22 Jan 2025 00:06:43 +0900 Subject: [PATCH 0250/1443] selftests/bpf: Adjust data size to have ETH_HLEN The function bpf_test_init() now returns an error if user_size (.data_size_in) is less than ETH_HLEN, causing the tests to fail. Adjust the data size to ensure it meets the requirement of ETH_HLEN. Signed-off-by: Shigeru Yoshida Signed-off-by: Martin KaFai Lau Link: https://patch.msgid.link/20250121150643.671650-2-syoshida@redhat.com Signed-off-by: Alexei Starovoitov --- .../testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c | 4 ++-- .../testing/selftests/bpf/prog_tests/xdp_devmap_attach.c | 8 ++++---- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c index c7f74f068e78..df27535995af 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c @@ -52,10 +52,10 @@ static void test_xdp_with_cpumap_helpers(void) ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id"); /* send a packet to trigger any potential bugs in there */ - char data[10] = {}; + char data[ETH_HLEN] = {}; DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts, .data_in = &data, - .data_size_in = 10, + .data_size_in = sizeof(data), .flags = BPF_F_TEST_XDP_LIVE_FRAMES, .repeat = 1, ); diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c index 27ffed17d4be..461ab18705d5 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c @@ -23,7 +23,7 @@ static void test_xdp_with_devmap_helpers(void) __u32 len = sizeof(info); int err, dm_fd, dm_fd_redir, map_fd; struct nstoken *nstoken = NULL; - char data[10] = {}; + char data[ETH_HLEN] = {}; __u32 idx = 0; SYS(out_close, "ip netns add %s", TEST_NS); @@ -58,7 +58,7 @@ static void test_xdp_with_devmap_helpers(void) /* send a packet to trigger any potential bugs in there */ DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts, .data_in = &data, - .data_size_in = 10, + .data_size_in = sizeof(data), .flags = BPF_F_TEST_XDP_LIVE_FRAMES, .repeat = 1, ); @@ -158,7 +158,7 @@ static void test_xdp_with_devmap_helpers_veth(void) struct nstoken *nstoken = NULL; __u32 len = sizeof(info); int err, dm_fd, dm_fd_redir, map_fd, ifindex_dst; - char data[10] = {}; + char data[ETH_HLEN] = {}; __u32 idx = 0; SYS(out_close, "ip netns add %s", TEST_NS); @@ -208,7 +208,7 @@ static void test_xdp_with_devmap_helpers_veth(void) /* send a packet to trigger any potential bugs in there */ DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts, .data_in = &data, - .data_size_in = 10, + .data_size_in = sizeof(data), .flags = BPF_F_TEST_XDP_LIVE_FRAMES, .repeat = 1, ); From 98671a0fd1f14e4a518ee06b19037c20014900eb Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 28 Jan 2025 17:22:45 -0800 Subject: [PATCH 0251/1443] bpf: unify VM_WRITE vs VM_MAYWRITE use in BPF map mmaping logic For all BPF maps we ensure that VM_MAYWRITE is cleared when memory-mapping BPF map contents as initially read-only VMA. This is because in some cases BPF verifier relies on the underlying data to not be modified afterwards by user space, so once something is mapped read-only, it shouldn't be re-mmap'ed as read-write. As such, it's not necessary to check VM_MAYWRITE in bpf_map_mmap() and map->ops->map_mmap() callbacks: VM_WRITE should be consistently set for read-write mappings, and if VM_WRITE is not set, there is no way for user space to upgrade read-only mapping to read-write one. This patch cleans up this VM_WRITE vs VM_MAYWRITE handling within bpf_map_mmap(), which is an entry point for any BPF map mmap()-ing logic. We also drop unnecessary sanitization of VM_MAYWRITE in BPF ringbuf's map_mmap() callback implementation, as it is already performed by common code in bpf_map_mmap(). Note, though, that in bpf_map_mmap_{open,close}() callbacks we can't drop VM_MAYWRITE use, because it's possible (and is outside of subsystem's control) to have initially read-write memory mapping, which is subsequently dropped to read-only by user space through mprotect(). In such case, from BPF verifier POV it's read-write data throughout the lifetime of BPF map, and is counted as "active writer". But its VMAs will start out as VM_WRITE|VM_MAYWRITE, then mprotect() can change it to just VM_MAYWRITE (and no VM_WRITE), so when its finally munmap()'ed and bpf_map_mmap_close() is called, vm_flags will be just VM_MAYWRITE, but we still need to decrement active writer count with bpf_map_write_active_dec() as it's still considered to be a read-write mapping by the rest of BPF subsystem. Similar reasoning applies to bpf_map_mmap_open(), which is called whenever mmap(), munmap(), and/or mprotect() forces mm subsystem to split original VMA into multiple discontiguous VMAs. Memory-mapping handling is a bit tricky, yes. Cc: Jann Horn Cc: Suren Baghdasaryan Cc: Shakeel Butt Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20250129012246.1515826-1-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/ringbuf.c | 4 ---- kernel/bpf/syscall.c | 10 ++++++++-- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index e1cfe890e0be..1499d8caa9a3 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -268,8 +268,6 @@ static int ringbuf_map_mmap_kern(struct bpf_map *map, struct vm_area_struct *vma /* allow writable mapping for the consumer_pos only */ if (vma->vm_pgoff != 0 || vma->vm_end - vma->vm_start != PAGE_SIZE) return -EPERM; - } else { - vm_flags_clear(vma, VM_MAYWRITE); } /* remap_vmalloc_range() checks size and offset constraints */ return remap_vmalloc_range(vma, rb_map->rb, @@ -289,8 +287,6 @@ static int ringbuf_map_mmap_user(struct bpf_map *map, struct vm_area_struct *vma * position, and the ring buffer data itself. */ return -EPERM; - } else { - vm_flags_clear(vma, VM_MAYWRITE); } /* remap_vmalloc_range() checks size and offset constraints */ return remap_vmalloc_range(vma, rb_map->rb, vma->vm_pgoff + RINGBUF_PGOFF); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 0daf098e3207..9bec3dce421f 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1065,15 +1065,21 @@ static int bpf_map_mmap(struct file *filp, struct vm_area_struct *vma) vma->vm_ops = &bpf_map_default_vmops; vma->vm_private_data = map; vm_flags_clear(vma, VM_MAYEXEC); + /* If mapping is read-only, then disallow potentially re-mapping with + * PROT_WRITE by dropping VM_MAYWRITE flag. This VM_MAYWRITE clearing + * means that as far as BPF map's memory-mapped VMAs are concerned, + * VM_WRITE and VM_MAYWRITE and equivalent, if one of them is set, + * both should be set, so we can forget about VM_MAYWRITE and always + * check just VM_WRITE + */ if (!(vma->vm_flags & VM_WRITE)) - /* disallow re-mapping with PROT_WRITE */ vm_flags_clear(vma, VM_MAYWRITE); err = map->ops->map_mmap(map, vma); if (err) goto out; - if (vma->vm_flags & VM_MAYWRITE) + if (vma->vm_flags & VM_WRITE) bpf_map_write_active_inc(map); out: mutex_unlock(&map->freeze_mutex); From bc27c52eea189e8f7492d40739b7746d67b65beb Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 28 Jan 2025 17:22:46 -0800 Subject: [PATCH 0252/1443] bpf: avoid holding freeze_mutex during mmap operation We use map->freeze_mutex to prevent races between map_freeze() and memory mapping BPF map contents with writable permissions. The way we naively do this means we'll hold freeze_mutex for entire duration of all the mm and VMA manipulations, which is completely unnecessary. This can potentially also lead to deadlocks, as reported by syzbot in [0]. So, instead, hold freeze_mutex only during writeability checks, bump (proactively) "write active" count for the map, unlock the mutex and proceed with mmap logic. And only if something went wrong during mmap logic, then undo that "write active" counter increment. [0] https://lore.kernel.org/bpf/678dcbc9.050a0220.303755.0066.GAE@google.com/ Fixes: fc9702273e2e ("bpf: Add mmap() support for BPF_MAP_TYPE_ARRAY") Reported-by: syzbot+4dc041c686b7c816a71e@syzkaller.appspotmail.com Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20250129012246.1515826-2-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/syscall.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 9bec3dce421f..14d6e99459d3 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -1035,7 +1035,7 @@ static const struct vm_operations_struct bpf_map_default_vmops = { static int bpf_map_mmap(struct file *filp, struct vm_area_struct *vma) { struct bpf_map *map = filp->private_data; - int err; + int err = 0; if (!map->ops->map_mmap || !IS_ERR_OR_NULL(map->record)) return -ENOTSUPP; @@ -1059,7 +1059,12 @@ static int bpf_map_mmap(struct file *filp, struct vm_area_struct *vma) err = -EACCES; goto out; } + bpf_map_write_active_inc(map); } +out: + mutex_unlock(&map->freeze_mutex); + if (err) + return err; /* set default open/close callbacks */ vma->vm_ops = &bpf_map_default_vmops; @@ -1076,13 +1081,11 @@ static int bpf_map_mmap(struct file *filp, struct vm_area_struct *vma) vm_flags_clear(vma, VM_MAYWRITE); err = map->ops->map_mmap(map, vma); - if (err) - goto out; + if (err) { + if (vma->vm_flags & VM_WRITE) + bpf_map_write_active_dec(map); + } - if (vma->vm_flags & VM_WRITE) - bpf_map_write_active_inc(map); -out: - mutex_unlock(&map->freeze_mutex); return err; } From 7748289df510638ba61fed86b59ce7d2fb4a194c Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Tue, 28 Jan 2025 07:42:42 -0800 Subject: [PATCH 0253/1443] drm/xe/guc: Fix size_t print format MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use %zx format to print size_t to remove the following warning when building for i386: >> drivers/gpu/drm/xe/xe_guc_ct.c:1727:43: warning: format specifies type 'unsigned long' but the argument has type 'size_t' (aka 'unsigned int') [-Wformat] 1727 | drm_printf(p, "[CTB].length: 0x%lx\n", snapshot->ctb_size); | ~~~ ^~~~~~~~~~~~~~~~~~ | %zx Cc: José Roberto de Souza Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501281627.H6nj184e-lkp@intel.com/ Fixes: cb1f868ca137 ("drm/xe: Make GUC binaries dump consistent with other binaries in devcoredump") Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250128154242.3371687-1-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_guc_ct.c | 2 +- drivers/gpu/drm/xe/xe_guc_log.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 497036675a38..72ad576fc18e 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -1724,7 +1724,7 @@ void xe_guc_ct_snapshot_print(struct xe_guc_ct_snapshot *snapshot, snapshot->g2h_outstanding); if (snapshot->ctb) { - drm_printf(p, "[CTB].length: 0x%lx\n", snapshot->ctb_size); + drm_printf(p, "[CTB].length: 0x%zx\n", snapshot->ctb_size); xe_print_blob_ascii85(p, "[CTB].data", '\n', snapshot->ctb, 0, snapshot->ctb_size); } diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index ab97ac728d7a..80514a446ba2 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -204,7 +204,7 @@ void xe_guc_log_snapshot_print(struct xe_guc_log_snapshot *snapshot, struct drm_ drm_printf(p, "GuC timestamp: 0x%08llX [%llu]\n", snapshot->stamp, snapshot->stamp); drm_printf(p, "Log level: %u\n", snapshot->level); - drm_printf(p, "[LOG].length: 0x%lx\n", snapshot->size); + drm_printf(p, "[LOG].length: 0x%zx\n", snapshot->size); remain = snapshot->size; for (i = 0; i < snapshot->num_chunks; i++) { size_t size = min(GUC_LOG_CHUNK_SIZE, remain); From e1e17a1715982201034024863efbf238bee2bdf9 Mon Sep 17 00:00:00 2001 From: Prasad Pandit Date: Mon, 11 Mar 2024 16:21:22 +0530 Subject: [PATCH 0254/1443] firmware: iscsi_ibft: fix ISCSI_IBFT Kconfig entry Fix ISCSI_IBFT Kconfig entry, replace tab with a space character. Fixes: 138fe4e0697 ("Firmware: add iSCSI iBFT Support") Signed-off-by: Prasad Pandit Signed-off-by: Konrad Rzeszutek Wilk --- drivers/firmware/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig index 71d8b26c4103..9f35f69e0f9e 100644 --- a/drivers/firmware/Kconfig +++ b/drivers/firmware/Kconfig @@ -106,7 +106,7 @@ config ISCSI_IBFT select ISCSI_BOOT_SYSFS select ISCSI_IBFT_FIND if X86 depends on ACPI && SCSI && SCSI_LOWLEVEL - default n + default n help This option enables support for detection and exposing of iSCSI Boot Firmware Table (iBFT) via sysfs to userspace. If you wish to From 07e0d99a2f701123ad3104c0f1a1e66bce74d6e5 Mon Sep 17 00:00:00 2001 From: Chengen Du Date: Tue, 14 Jan 2025 12:12:34 +0800 Subject: [PATCH 0255/1443] iscsi_ibft: Fix UBSAN shift-out-of-bounds warning in ibft_attr_show_nic() When performing an iSCSI boot using IPv6, iscsistart still reads the /sys/firmware/ibft/ethernetX/subnet-mask entry. Since the IPv6 prefix length is 64, this causes the shift exponent to become negative, triggering a UBSAN warning. As the concept of a subnet mask does not apply to IPv6, the value is set to ~0 to suppress the warning message. Signed-off-by: Chengen Du Signed-off-by: Konrad Rzeszutek Wilk --- drivers/firmware/iscsi_ibft.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/firmware/iscsi_ibft.c b/drivers/firmware/iscsi_ibft.c index 6e9788324fea..371f24569b3b 100644 --- a/drivers/firmware/iscsi_ibft.c +++ b/drivers/firmware/iscsi_ibft.c @@ -310,7 +310,10 @@ static ssize_t ibft_attr_show_nic(void *data, int type, char *buf) str += sprintf_ipaddr(str, nic->ip_addr); break; case ISCSI_BOOT_ETH_SUBNET_MASK: - val = cpu_to_be32(~((1 << (32-nic->subnet_mask_prefix))-1)); + if (nic->subnet_mask_prefix > 32) + val = cpu_to_be32(~0); + else + val = cpu_to_be32(~((1 << (32-nic->subnet_mask_prefix))-1)); str += sprintf(str, "%pI4", &val); break; case ISCSI_BOOT_ETH_PREFIX_LEN: From 79fc672a092d93a7eac24fe20a571d4efd8fa5a4 Mon Sep 17 00:00:00 2001 From: Haoxiang Li Date: Thu, 19 Dec 2024 17:02:56 +0800 Subject: [PATCH 0256/1443] drm/komeda: Add check for komeda_get_layer_fourcc_list() Add check for the return value of komeda_get_layer_fourcc_list() to catch the potential exception. Fixes: 5d51f6c0da1b ("drm/komeda: Add writeback support") Cc: stable@vger.kernel.org Signed-off-by: Haoxiang Li Acked-by: Liviu Dudau Link: https://lore.kernel.org/r/20241219090256.146424-1-haoxiang_li2024@163.com Signed-off-by: Liviu Dudau --- drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c b/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c index ebccb74306a7..f30b3d5eeca5 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_wb_connector.c @@ -160,6 +160,10 @@ static int komeda_wb_connector_add(struct komeda_kms_dev *kms, formats = komeda_get_layer_fourcc_list(&mdev->fmt_tbl, kwb_conn->wb_layer->layer_type, &n_formats); + if (!formats) { + kfree(kwb_conn); + return -ENOMEM; + } err = drm_writeback_connector_init(&kms->base, wb_conn, &komeda_wb_connector_funcs, From 0532a79efd68a4d9686b0385e4993af4b130ff82 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 22 Jan 2025 18:09:13 +0800 Subject: [PATCH 0257/1443] strparser: Add read_sock callback Added a new read_sock handler, allowing users to customize read operations instead of relying on the native socket's read_sock. Signed-off-by: Jiayuan Chen Signed-off-by: Martin KaFai Lau Reviewed-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://patch.msgid.link/20250122100917.49845-2-mrpre@163.com --- Documentation/networking/strparser.rst | 9 ++++++++- include/net/strparser.h | 2 ++ net/strparser/strparser.c | 11 +++++++++-- 3 files changed, 19 insertions(+), 3 deletions(-) diff --git a/Documentation/networking/strparser.rst b/Documentation/networking/strparser.rst index 6cab1f74ae05..7f623d1db72a 100644 --- a/Documentation/networking/strparser.rst +++ b/Documentation/networking/strparser.rst @@ -112,7 +112,7 @@ Functions Callbacks ========= -There are six callbacks: +There are seven callbacks: :: @@ -182,6 +182,13 @@ There are six callbacks: the length of the message. skb->len - offset may be greater then full_len since strparser does not trim the skb. + :: + + int (*read_sock)(struct strparser *strp, read_descriptor_t *desc, + sk_read_actor_t recv_actor); + + The read_sock callback is used by strparser instead of + sock->ops->read_sock, if provided. :: int (*read_sock_done)(struct strparser *strp, int err); diff --git a/include/net/strparser.h b/include/net/strparser.h index 41e2ce9e9e10..0a83010b3a64 100644 --- a/include/net/strparser.h +++ b/include/net/strparser.h @@ -43,6 +43,8 @@ struct strparser; struct strp_callbacks { int (*parse_msg)(struct strparser *strp, struct sk_buff *skb); void (*rcv_msg)(struct strparser *strp, struct sk_buff *skb); + int (*read_sock)(struct strparser *strp, read_descriptor_t *desc, + sk_read_actor_t recv_actor); int (*read_sock_done)(struct strparser *strp, int err); void (*abort_parser)(struct strparser *strp, int err); void (*lock)(struct strparser *strp); diff --git a/net/strparser/strparser.c b/net/strparser/strparser.c index 8299ceb3e373..95696f42647e 100644 --- a/net/strparser/strparser.c +++ b/net/strparser/strparser.c @@ -347,7 +347,10 @@ static int strp_read_sock(struct strparser *strp) struct socket *sock = strp->sk->sk_socket; read_descriptor_t desc; - if (unlikely(!sock || !sock->ops || !sock->ops->read_sock)) + if (unlikely(!sock || !sock->ops)) + return -EBUSY; + + if (unlikely(!strp->cb.read_sock && !sock->ops->read_sock)) return -EBUSY; desc.arg.data = strp; @@ -355,7 +358,10 @@ static int strp_read_sock(struct strparser *strp) desc.count = 1; /* give more than one skb per call */ /* sk should be locked here, so okay to do read_sock */ - sock->ops->read_sock(strp->sk, &desc, strp_recv); + if (strp->cb.read_sock) + strp->cb.read_sock(strp, &desc, strp_recv); + else + sock->ops->read_sock(strp->sk, &desc, strp_recv); desc.error = strp->cb.read_sock_done(strp, desc.error); @@ -468,6 +474,7 @@ int strp_init(struct strparser *strp, struct sock *sk, strp->cb.unlock = cb->unlock ? : strp_sock_unlock; strp->cb.rcv_msg = cb->rcv_msg; strp->cb.parse_msg = cb->parse_msg; + strp->cb.read_sock = cb->read_sock; strp->cb.read_sock_done = cb->read_sock_done ? : default_read_sock_done; strp->cb.abort_parser = cb->abort_parser ? : strp_abort_strp; From 36b62df5683c315ba58c950f1a9c771c796c30ec Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 22 Jan 2025 18:09:14 +0800 Subject: [PATCH 0258/1443] bpf: Fix wrong copied_seq calculation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 'sk->copied_seq' was updated in the tcp_eat_skb() function when the action of a BPF program was SK_REDIRECT. For other actions, like SK_PASS, the update logic for 'sk->copied_seq' was moved to tcp_bpf_recvmsg_parser() to ensure the accuracy of the 'fionread' feature. It works for a single stream_verdict scenario, as it also modified sk_data_ready->sk_psock_verdict_data_ready->tcp_read_skb to remove updating 'sk->copied_seq'. However, for programs where both stream_parser and stream_verdict are active (strparser purpose), tcp_read_sock() was used instead of tcp_read_skb() (sk_data_ready->strp_data_ready->tcp_read_sock). tcp_read_sock() now still updates 'sk->copied_seq', leading to duplicate updates. In summary, for strparser + SK_PASS, copied_seq is redundantly calculated in both tcp_read_sock() and tcp_bpf_recvmsg_parser(). The issue causes incorrect copied_seq calculations, which prevent correct data reads from the recv() interface in user-land. We do not want to add new proto_ops to implement a new version of tcp_read_sock, as this would introduce code complexity [1]. We could have added noack and copied_seq to desc, and then called ops->read_sock. However, unfortunately, other modules didn’t fully initialize desc to zero. So, for now, we are directly calling tcp_read_sock_noack() in tcp_bpf.c. [1]: https://lore.kernel.org/bpf/20241218053408.437295-1-mrpre@163.com Fixes: e5c6de5fa025 ("bpf, sockmap: Incorrectly handling copied_seq") Suggested-by: Jakub Sitnicki Signed-off-by: Jiayuan Chen Signed-off-by: Martin KaFai Lau Reviewed-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://patch.msgid.link/20250122100917.49845-3-mrpre@163.com --- include/linux/skmsg.h | 2 ++ include/net/tcp.h | 8 ++++++++ net/core/skmsg.c | 7 +++++++ net/ipv4/tcp.c | 29 ++++++++++++++++++++++++----- net/ipv4/tcp_bpf.c | 36 ++++++++++++++++++++++++++++++++++++ 5 files changed, 77 insertions(+), 5 deletions(-) diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index 2cbe0c22a32f..0b9095a281b8 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -91,6 +91,8 @@ struct sk_psock { struct sk_psock_progs progs; #if IS_ENABLED(CONFIG_BPF_STREAM_PARSER) struct strparser strp; + u32 copied_seq; + u32 ingress_bytes; #endif struct sk_buff_head ingress_skb; struct list_head ingress_msg; diff --git a/include/net/tcp.h b/include/net/tcp.h index 5b2b04835688..9c044fb9ab26 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -729,6 +729,9 @@ void tcp_get_info(struct sock *, struct tcp_info *); /* Read 'sendfile()'-style from a TCP socket */ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, sk_read_actor_t recv_actor); +int tcp_read_sock_noack(struct sock *sk, read_descriptor_t *desc, + sk_read_actor_t recv_actor, bool noack, + u32 *copied_seq); int tcp_read_skb(struct sock *sk, skb_read_actor_t recv_actor); struct sk_buff *tcp_recv_skb(struct sock *sk, u32 seq, u32 *off); void tcp_read_done(struct sock *sk, size_t len); @@ -2599,6 +2602,11 @@ struct sk_psock; #ifdef CONFIG_BPF_SYSCALL int tcp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore); void tcp_bpf_clone(const struct sock *sk, struct sock *newsk); +#ifdef CONFIG_BPF_STREAM_PARSER +struct strparser; +int tcp_bpf_strp_read_sock(struct strparser *strp, read_descriptor_t *desc, + sk_read_actor_t recv_actor); +#endif /* CONFIG_BPF_STREAM_PARSER */ #endif /* CONFIG_BPF_SYSCALL */ #ifdef CONFIG_INET diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 61f3f3d4e528..0ddc4c718833 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -549,6 +549,9 @@ static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb, return num_sge; } +#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER) + psock->ingress_bytes += len; +#endif copied = len; msg->sg.start = 0; msg->sg.size = copied; @@ -1144,6 +1147,10 @@ int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock) if (!ret) sk_psock_set_state(psock, SK_PSOCK_RX_STRP_ENABLED); + if (sk_is_tcp(sk)) { + psock->strp.cb.read_sock = tcp_bpf_strp_read_sock; + psock->copied_seq = tcp_sk(sk)->copied_seq; + } return ret; } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0d704bda6c41..285678d8ce07 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1565,12 +1565,13 @@ EXPORT_SYMBOL(tcp_recv_skb); * or for 'peeking' the socket using this routine * (although both would be easy to implement). */ -int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, - sk_read_actor_t recv_actor) +static int __tcp_read_sock(struct sock *sk, read_descriptor_t *desc, + sk_read_actor_t recv_actor, bool noack, + u32 *copied_seq) { struct sk_buff *skb; struct tcp_sock *tp = tcp_sk(sk); - u32 seq = tp->copied_seq; + u32 seq = *copied_seq; u32 offset; int copied = 0; @@ -1624,9 +1625,12 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, tcp_eat_recv_skb(sk, skb); if (!desc->count) break; - WRITE_ONCE(tp->copied_seq, seq); + WRITE_ONCE(*copied_seq, seq); } - WRITE_ONCE(tp->copied_seq, seq); + WRITE_ONCE(*copied_seq, seq); + + if (noack) + goto out; tcp_rcv_space_adjust(sk); @@ -1635,10 +1639,25 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, tcp_recv_skb(sk, seq, &offset); tcp_cleanup_rbuf(sk, copied); } +out: return copied; } + +int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, + sk_read_actor_t recv_actor) +{ + return __tcp_read_sock(sk, desc, recv_actor, false, + &tcp_sk(sk)->copied_seq); +} EXPORT_SYMBOL(tcp_read_sock); +int tcp_read_sock_noack(struct sock *sk, read_descriptor_t *desc, + sk_read_actor_t recv_actor, bool noack, + u32 *copied_seq) +{ + return __tcp_read_sock(sk, desc, recv_actor, noack, copied_seq); +} + int tcp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) { struct sk_buff *skb; diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 47f65b1b70ca..ba581785adb4 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -646,6 +646,42 @@ static int tcp_bpf_assert_proto_ops(struct proto *ops) ops->sendmsg == tcp_sendmsg ? 0 : -ENOTSUPP; } +#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER) +int tcp_bpf_strp_read_sock(struct strparser *strp, read_descriptor_t *desc, + sk_read_actor_t recv_actor) +{ + struct sock *sk = strp->sk; + struct sk_psock *psock; + struct tcp_sock *tp; + int copied = 0; + + tp = tcp_sk(sk); + rcu_read_lock(); + psock = sk_psock(sk); + if (WARN_ON_ONCE(!psock)) { + desc->error = -EINVAL; + goto out; + } + + psock->ingress_bytes = 0; + copied = tcp_read_sock_noack(sk, desc, recv_actor, true, + &psock->copied_seq); + if (copied < 0) + goto out; + /* recv_actor may redirect skb to another socket (SK_REDIRECT) or + * just put skb into ingress queue of current socket (SK_PASS). + * For SK_REDIRECT, we need to ack the frame immediately but for + * SK_PASS, we want to delay the ack until tcp_bpf_recvmsg_parser(). + */ + tp->copied_seq = psock->copied_seq - psock->ingress_bytes; + tcp_rcv_space_adjust(sk); + __tcp_cleanup_rbuf(sk, copied - psock->ingress_bytes); +out: + rcu_read_unlock(); + return copied; +} +#endif /* CONFIG_BPF_STREAM_PARSER */ + int tcp_bpf_update_proto(struct sock *sk, struct sk_psock *psock, bool restore) { int family = sk->sk_family == AF_INET6 ? TCP_BPF_IPV6 : TCP_BPF_IPV4; From 5459cce6bf49e72ee29be21865869c2ac42419f5 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 22 Jan 2025 18:09:15 +0800 Subject: [PATCH 0259/1443] bpf: Disable non stream socket for strparser Currently, only TCP supports strparser, but sockmap doesn't intercept non-TCP connections to attach strparser. For example, with UDP, although the read/write handlers are replaced, strparser is not executed due to the lack of a read_sock operation. Furthermore, in udp_bpf_recvmsg(), it checks whether the psock has data, and if not, it falls back to the native UDP read interface, making UDP + strparser appear to read correctly. According to its commit history, this behavior is unexpected. Moreover, since UDP lacks the concept of streams, we intercept it directly. Fixes: 1fa1fe8ff161 ("bpf, sockmap: Test shutdown() correctly exits epoll and recv()=0") Signed-off-by: Jiayuan Chen Signed-off-by: Martin KaFai Lau Acked-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://patch.msgid.link/20250122100917.49845-4-mrpre@163.com --- net/core/sock_map.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/core/sock_map.c b/net/core/sock_map.c index f1b9b3958792..3b0f59d9b4db 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -303,7 +303,10 @@ static int sock_map_link(struct bpf_map *map, struct sock *sk) write_lock_bh(&sk->sk_callback_lock); if (stream_parser && stream_verdict && !psock->saved_data_ready) { - ret = sk_psock_init_strp(sk, psock); + if (sk_is_tcp(sk)) + ret = sk_psock_init_strp(sk, psock); + else + ret = -EOPNOTSUPP; if (ret) { write_unlock_bh(&sk->sk_callback_lock); sk_psock_put(sk, psock); From a0c11149509aa905aeec10cf9998091443472b0b Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 22 Jan 2025 18:09:16 +0800 Subject: [PATCH 0260/1443] selftests/bpf: Fix invalid flag of recv() SOCK_NONBLOCK flag is only effective during socket creation, not during recv. Use MSG_DONTWAIT instead. Signed-off-by: Jiayuan Chen Signed-off-by: Martin KaFai Lau Acked-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://patch.msgid.link/20250122100917.49845-5-mrpre@163.com --- tools/testing/selftests/bpf/prog_tests/sockmap_basic.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c index 884ad87783d5..0c51b7288978 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c @@ -522,8 +522,8 @@ static void test_sockmap_skb_verdict_shutdown(void) if (!ASSERT_EQ(err, 1, "epoll_wait(fd)")) goto out_close; - n = recv(c1, &b, 1, SOCK_NONBLOCK); - ASSERT_EQ(n, 0, "recv_timeout(fin)"); + n = recv(c1, &b, 1, MSG_DONTWAIT); + ASSERT_EQ(n, 0, "recv(fin)"); out_close: close(c1); close(p1); @@ -628,7 +628,7 @@ static void test_sockmap_skb_verdict_fionread(bool pass_prog) ASSERT_EQ(avail, expected, "ioctl(FIONREAD)"); /* On DROP test there will be no data to read */ if (pass_prog) { - recvd = recv_timeout(c1, &buf, sizeof(buf), SOCK_NONBLOCK, IO_TIMEOUT_SEC); + recvd = recv_timeout(c1, &buf, sizeof(buf), MSG_DONTWAIT, IO_TIMEOUT_SEC); ASSERT_EQ(recvd, sizeof(buf), "recv_timeout(c0)"); } From 6fcfe96e0f6e9bebe1b185f1548a9a8cb1b68dea Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 22 Jan 2025 18:09:17 +0800 Subject: [PATCH 0261/1443] selftests/bpf: Add strparser test for bpf Add test cases for bpf + strparser and separated them from sockmap_basic, as strparser has more encapsulation and parsing capabilities compared to standard sockmap. Signed-off-by: Jiayuan Chen Signed-off-by: Martin KaFai Lau Acked-by: Jakub Sitnicki Acked-by: John Fastabend Link: https://patch.msgid.link/20250122100917.49845-6-mrpre@163.com --- .../selftests/bpf/prog_tests/sockmap_basic.c | 53 -- .../selftests/bpf/prog_tests/sockmap_strp.c | 454 ++++++++++++++++++ .../selftests/bpf/progs/test_sockmap_strp.c | 53 ++ 3 files changed, 507 insertions(+), 53 deletions(-) create mode 100644 tools/testing/selftests/bpf/prog_tests/sockmap_strp.c create mode 100644 tools/testing/selftests/bpf/progs/test_sockmap_strp.c diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c index 0c51b7288978..f8953455db29 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_basic.c @@ -531,57 +531,6 @@ static void test_sockmap_skb_verdict_shutdown(void) test_sockmap_pass_prog__destroy(skel); } -static void test_sockmap_stream_pass(void) -{ - int zero = 0, sent, recvd; - int verdict, parser; - int err, map; - int c = -1, p = -1; - struct test_sockmap_pass_prog *pass = NULL; - char snd[256] = "0123456789"; - char rcv[256] = "0"; - - pass = test_sockmap_pass_prog__open_and_load(); - verdict = bpf_program__fd(pass->progs.prog_skb_verdict); - parser = bpf_program__fd(pass->progs.prog_skb_parser); - map = bpf_map__fd(pass->maps.sock_map_rx); - - err = bpf_prog_attach(parser, map, BPF_SK_SKB_STREAM_PARSER, 0); - if (!ASSERT_OK(err, "bpf_prog_attach stream parser")) - goto out; - - err = bpf_prog_attach(verdict, map, BPF_SK_SKB_STREAM_VERDICT, 0); - if (!ASSERT_OK(err, "bpf_prog_attach stream verdict")) - goto out; - - err = create_pair(AF_INET, SOCK_STREAM, &c, &p); - if (err) - goto out; - - /* sk_data_ready of 'p' will be replaced by strparser handler */ - err = bpf_map_update_elem(map, &zero, &p, BPF_NOEXIST); - if (!ASSERT_OK(err, "bpf_map_update_elem(p)")) - goto out_close; - - /* - * as 'prog_skb_parser' return the original skb len and - * 'prog_skb_verdict' return SK_PASS, the kernel will just - * pass it through to original socket 'p' - */ - sent = xsend(c, snd, sizeof(snd), 0); - ASSERT_EQ(sent, sizeof(snd), "xsend(c)"); - - recvd = recv_timeout(p, rcv, sizeof(rcv), SOCK_NONBLOCK, - IO_TIMEOUT_SEC); - ASSERT_EQ(recvd, sizeof(rcv), "recv_timeout(p)"); - -out_close: - close(c); - close(p); - -out: - test_sockmap_pass_prog__destroy(pass); -} static void test_sockmap_skb_verdict_fionread(bool pass_prog) { @@ -1101,8 +1050,6 @@ void test_sockmap_basic(void) test_sockmap_progs_query(BPF_SK_SKB_VERDICT); if (test__start_subtest("sockmap skb_verdict shutdown")) test_sockmap_skb_verdict_shutdown(); - if (test__start_subtest("sockmap stream parser and verdict pass")) - test_sockmap_stream_pass(); if (test__start_subtest("sockmap skb_verdict fionread")) test_sockmap_skb_verdict_fionread(true); if (test__start_subtest("sockmap skb_verdict fionread on drop")) diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_strp.c b/tools/testing/selftests/bpf/prog_tests/sockmap_strp.c new file mode 100644 index 000000000000..621b3b71888e --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/sockmap_strp.c @@ -0,0 +1,454 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include "sockmap_helpers.h" +#include "test_skmsg_load_helpers.skel.h" +#include "test_sockmap_strp.skel.h" + +#define STRP_PKT_HEAD_LEN 4 +#define STRP_PKT_BODY_LEN 6 +#define STRP_PKT_FULL_LEN (STRP_PKT_HEAD_LEN + STRP_PKT_BODY_LEN) + +static const char packet[STRP_PKT_FULL_LEN] = "head+body\0"; +static const int test_packet_num = 100; + +/* Current implementation of tcp_bpf_recvmsg_parser() invokes data_ready + * with sk held if an skb exists in sk_receive_queue. Then for the + * data_ready implementation of strparser, it will delay the read + * operation if sk is held and EAGAIN is returned. + */ +static int sockmap_strp_consume_pre_data(int p) +{ + int recvd; + bool retried = false; + char rcv[10]; + +retry: + errno = 0; + recvd = recv_timeout(p, rcv, sizeof(rcv), 0, 1); + if (recvd < 0 && errno == EAGAIN && retried == false) { + /* On the first call, EAGAIN will certainly be returned. + * A 1-second wait is enough for the workqueue to finish. + */ + sleep(1); + retried = true; + goto retry; + } + + if (!ASSERT_EQ(recvd, STRP_PKT_FULL_LEN, "recv error or truncated data") || + !ASSERT_OK(memcmp(packet, rcv, STRP_PKT_FULL_LEN), + "data mismatch")) + return -1; + return 0; +} + +static struct test_sockmap_strp *sockmap_strp_init(int *out_map, bool pass, + bool need_parser) +{ + struct test_sockmap_strp *strp = NULL; + int verdict, parser; + int err; + + strp = test_sockmap_strp__open_and_load(); + *out_map = bpf_map__fd(strp->maps.sock_map); + + if (need_parser) + parser = bpf_program__fd(strp->progs.prog_skb_parser_partial); + else + parser = bpf_program__fd(strp->progs.prog_skb_parser); + + if (pass) + verdict = bpf_program__fd(strp->progs.prog_skb_verdict_pass); + else + verdict = bpf_program__fd(strp->progs.prog_skb_verdict); + + err = bpf_prog_attach(parser, *out_map, BPF_SK_SKB_STREAM_PARSER, 0); + if (!ASSERT_OK(err, "bpf_prog_attach stream parser")) + goto err; + + err = bpf_prog_attach(verdict, *out_map, BPF_SK_SKB_STREAM_VERDICT, 0); + if (!ASSERT_OK(err, "bpf_prog_attach stream verdict")) + goto err; + + return strp; +err: + test_sockmap_strp__destroy(strp); + return NULL; +} + +/* Dispatch packets to different socket by packet size: + * + * ------ ------ + * | pkt4 || pkt1 |... > remote socket + * ------ ------ / ------ ------ + * | pkt8 | pkt7 |... + * ------ ------ \ ------ ------ + * | pkt3 || pkt2 |... > local socket + * ------ ------ + */ +static void test_sockmap_strp_dispatch_pkt(int family, int sotype) +{ + int i, j, zero = 0, one = 1, recvd; + int err, map; + int c0 = -1, p0 = -1, c1 = -1, p1 = -1; + struct test_sockmap_strp *strp = NULL; + int test_cnt = 6; + char rcv[10]; + struct { + char data[7]; + int data_len; + int send_cnt; + int *receiver; + } send_dir[2] = { + /* data expected to deliver to local */ + {"llllll", 6, 0, &p0}, + /* data expected to deliver to remote */ + {"rrrrr", 5, 0, &c1} + }; + + strp = sockmap_strp_init(&map, false, false); + if (!ASSERT_TRUE(strp, "sockmap_strp_init")) + return; + + err = create_socket_pairs(family, sotype, &c0, &c1, &p0, &p1); + if (!ASSERT_OK(err, "create_socket_pairs()")) + goto out; + + err = bpf_map_update_elem(map, &zero, &p0, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(p0)")) + goto out_close; + + err = bpf_map_update_elem(map, &one, &p1, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(p1)")) + goto out_close; + + err = setsockopt(c1, IPPROTO_TCP, TCP_NODELAY, &zero, sizeof(zero)); + if (!ASSERT_OK(err, "setsockopt(TCP_NODELAY)")) + goto out_close; + + /* deliver data with data size greater than 5 to local */ + strp->data->verdict_max_size = 5; + + for (i = 0; i < test_cnt; i++) { + int d = i % 2; + + xsend(c0, send_dir[d].data, send_dir[d].data_len, 0); + send_dir[d].send_cnt++; + } + + for (i = 0; i < 2; i++) { + for (j = 0; j < send_dir[i].send_cnt; j++) { + int expected = send_dir[i].data_len; + + recvd = recv_timeout(*send_dir[i].receiver, rcv, + expected, MSG_DONTWAIT, + IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, expected, "recv_timeout()")) + goto out_close; + if (!ASSERT_OK(memcmp(send_dir[i].data, rcv, recvd), + "data mismatch")) + goto out_close; + } + } +out_close: + close(c0); + close(c1); + close(p0); + close(p1); +out: + test_sockmap_strp__destroy(strp); +} + +/* We have multiple packets in one skb + * ------------ ------------ ------------ + * | packet1 | packet2 | ... + * ------------ ------------ ------------ + */ +static void test_sockmap_strp_multiple_pkt(int family, int sotype) +{ + int i, zero = 0; + int sent, recvd, total; + int err, map; + int c = -1, p = -1; + struct test_sockmap_strp *strp = NULL; + char *snd = NULL, *rcv = NULL; + + strp = sockmap_strp_init(&map, true, true); + if (!ASSERT_TRUE(strp, "sockmap_strp_init")) + return; + + err = create_pair(family, sotype, &c, &p); + if (err) + goto out; + + err = bpf_map_update_elem(map, &zero, &p, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(zero, p)")) + goto out_close; + + /* construct multiple packets in one buffer */ + total = test_packet_num * STRP_PKT_FULL_LEN; + snd = malloc(total); + rcv = malloc(total + 1); + if (!ASSERT_TRUE(snd, "malloc(snd)") || + !ASSERT_TRUE(rcv, "malloc(rcv)")) + goto out_close; + + for (i = 0; i < test_packet_num; i++) { + memcpy(snd + i * STRP_PKT_FULL_LEN, + packet, STRP_PKT_FULL_LEN); + } + + sent = xsend(c, snd, total, 0); + if (!ASSERT_EQ(sent, total, "xsend(c)")) + goto out_close; + + /* try to recv one more byte to avoid truncation check */ + recvd = recv_timeout(p, rcv, total + 1, MSG_DONTWAIT, IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, total, "recv(rcv)")) + goto out_close; + + /* we sent TCP segment with multiple encapsulation + * then check whether packets are handled correctly + */ + if (!ASSERT_OK(memcmp(snd, rcv, total), "data mismatch")) + goto out_close; + +out_close: + close(c); + close(p); + if (snd) + free(snd); + if (rcv) + free(rcv); +out: + test_sockmap_strp__destroy(strp); +} + +/* Test strparser with partial read */ +static void test_sockmap_strp_partial_read(int family, int sotype) +{ + int zero = 0, recvd, off; + int err, map; + int c = -1, p = -1; + struct test_sockmap_strp *strp = NULL; + char rcv[STRP_PKT_FULL_LEN + 1] = "0"; + + strp = sockmap_strp_init(&map, true, true); + if (!ASSERT_TRUE(strp, "sockmap_strp_init")) + return; + + err = create_pair(family, sotype, &c, &p); + if (err) + goto out; + + /* sk_data_ready of 'p' will be replaced by strparser handler */ + err = bpf_map_update_elem(map, &zero, &p, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(zero, p)")) + goto out_close; + + /* 1.1 send partial head, 1 byte header left */ + off = STRP_PKT_HEAD_LEN - 1; + xsend(c, packet, off, 0); + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, 1); + if (!ASSERT_EQ(-1, recvd, "partial head sent, expected no data")) + goto out_close; + + /* 1.2 send remaining head and body */ + xsend(c, packet + off, STRP_PKT_FULL_LEN - off, 0); + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, STRP_PKT_FULL_LEN, "expected full data")) + goto out_close; + + /* 2.1 send partial head, 1 byte header left */ + off = STRP_PKT_HEAD_LEN - 1; + xsend(c, packet, off, 0); + + /* 2.2 send remaining head and partial body, 1 byte body left */ + xsend(c, packet + off, STRP_PKT_FULL_LEN - off - 1, 0); + off = STRP_PKT_FULL_LEN - 1; + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, 1); + if (!ASSERT_EQ(-1, recvd, "partial body sent, expected no data")) + goto out_close; + + /* 2.3 send remaining body */ + xsend(c, packet + off, STRP_PKT_FULL_LEN - off, 0); + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, STRP_PKT_FULL_LEN, "expected full data")) + goto out_close; + +out_close: + close(c); + close(p); + +out: + test_sockmap_strp__destroy(strp); +} + +/* Test simple socket read/write with strparser + FIONREAD */ +static void test_sockmap_strp_pass(int family, int sotype, bool fionread) +{ + int zero = 0, pkt_size = STRP_PKT_FULL_LEN, sent, recvd, avail; + int err, map; + int c = -1, p = -1; + int test_cnt = 10, i; + struct test_sockmap_strp *strp = NULL; + char rcv[STRP_PKT_FULL_LEN + 1] = "0"; + + strp = sockmap_strp_init(&map, true, true); + if (!ASSERT_TRUE(strp, "sockmap_strp_init")) + return; + + err = create_pair(family, sotype, &c, &p); + if (err) + goto out; + + /* inject some data before bpf process, it should be read + * correctly because we check sk_receive_queue in + * tcp_bpf_recvmsg_parser(). + */ + sent = xsend(c, packet, pkt_size, 0); + if (!ASSERT_EQ(sent, pkt_size, "xsend(pre-data)")) + goto out_close; + + /* sk_data_ready of 'p' will be replaced by strparser handler */ + err = bpf_map_update_elem(map, &zero, &p, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(p)")) + goto out_close; + + /* consume previous data we injected */ + if (sockmap_strp_consume_pre_data(p)) + goto out_close; + + /* Previously, we encountered issues such as deadlocks and + * sequence errors that resulted in the inability to read + * continuously. Therefore, we perform multiple iterations + * of testing here. + */ + for (i = 0; i < test_cnt; i++) { + sent = xsend(c, packet, pkt_size, 0); + if (!ASSERT_EQ(sent, pkt_size, "xsend(c)")) + goto out_close; + + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, + IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, pkt_size, "recv_timeout(p)") || + !ASSERT_OK(memcmp(packet, rcv, pkt_size), + "memcmp, data mismatch")) + goto out_close; + } + + if (fionread) { + sent = xsend(c, packet, pkt_size, 0); + if (!ASSERT_EQ(sent, pkt_size, "second xsend(c)")) + goto out_close; + + err = ioctl(p, FIONREAD, &avail); + if (!ASSERT_OK(err, "ioctl(FIONREAD) error") || + !ASSERT_EQ(avail, pkt_size, "ioctl(FIONREAD)")) + goto out_close; + + recvd = recv_timeout(p, rcv, sizeof(rcv), MSG_DONTWAIT, + IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, pkt_size, "second recv_timeout(p)") || + !ASSERT_OK(memcmp(packet, rcv, pkt_size), + "second memcmp, data mismatch")) + goto out_close; + } + +out_close: + close(c); + close(p); + +out: + test_sockmap_strp__destroy(strp); +} + +/* Test strparser with verdict mode */ +static void test_sockmap_strp_verdict(int family, int sotype) +{ + int zero = 0, one = 1, sent, recvd, off; + int err, map; + int c0 = -1, p0 = -1, c1 = -1, p1 = -1; + struct test_sockmap_strp *strp = NULL; + char rcv[STRP_PKT_FULL_LEN + 1] = "0"; + + strp = sockmap_strp_init(&map, false, true); + if (!ASSERT_TRUE(strp, "sockmap_strp_init")) + return; + + /* We simulate a reverse proxy server. + * When p0 receives data from c0, we forward it to c1. + * From c1's perspective, it will consider this data + * as being sent by p1. + */ + err = create_socket_pairs(family, sotype, &c0, &c1, &p0, &p1); + if (!ASSERT_OK(err, "create_socket_pairs()")) + goto out; + + err = bpf_map_update_elem(map, &zero, &p0, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(p0)")) + goto out_close; + + err = bpf_map_update_elem(map, &one, &p1, BPF_NOEXIST); + if (!ASSERT_OK(err, "bpf_map_update_elem(p1)")) + goto out_close; + + sent = xsend(c0, packet, STRP_PKT_FULL_LEN, 0); + if (!ASSERT_EQ(sent, STRP_PKT_FULL_LEN, "xsend(c0)")) + goto out_close; + + recvd = recv_timeout(c1, rcv, sizeof(rcv), MSG_DONTWAIT, + IO_TIMEOUT_SEC); + if (!ASSERT_EQ(recvd, STRP_PKT_FULL_LEN, "recv_timeout(c1)") || + !ASSERT_OK(memcmp(packet, rcv, STRP_PKT_FULL_LEN), + "received data does not match the sent data")) + goto out_close; + + /* send again to ensure the stream is functioning correctly. */ + sent = xsend(c0, packet, STRP_PKT_FULL_LEN, 0); + if (!ASSERT_EQ(sent, STRP_PKT_FULL_LEN, "second xsend(c0)")) + goto out_close; + + /* partial read */ + off = STRP_PKT_FULL_LEN / 2; + recvd = recv_timeout(c1, rcv, off, MSG_DONTWAIT, + IO_TIMEOUT_SEC); + recvd += recv_timeout(c1, rcv + off, sizeof(rcv) - off, MSG_DONTWAIT, + IO_TIMEOUT_SEC); + + if (!ASSERT_EQ(recvd, STRP_PKT_FULL_LEN, "partial recv_timeout(c1)") || + !ASSERT_OK(memcmp(packet, rcv, STRP_PKT_FULL_LEN), + "partial received data does not match the sent data")) + goto out_close; + +out_close: + close(c0); + close(c1); + close(p0); + close(p1); +out: + test_sockmap_strp__destroy(strp); +} + +void test_sockmap_strp(void) +{ + if (test__start_subtest("sockmap strp tcp pass")) + test_sockmap_strp_pass(AF_INET, SOCK_STREAM, false); + if (test__start_subtest("sockmap strp tcp v6 pass")) + test_sockmap_strp_pass(AF_INET6, SOCK_STREAM, false); + if (test__start_subtest("sockmap strp tcp pass fionread")) + test_sockmap_strp_pass(AF_INET, SOCK_STREAM, true); + if (test__start_subtest("sockmap strp tcp v6 pass fionread")) + test_sockmap_strp_pass(AF_INET6, SOCK_STREAM, true); + if (test__start_subtest("sockmap strp tcp verdict")) + test_sockmap_strp_verdict(AF_INET, SOCK_STREAM); + if (test__start_subtest("sockmap strp tcp v6 verdict")) + test_sockmap_strp_verdict(AF_INET6, SOCK_STREAM); + if (test__start_subtest("sockmap strp tcp partial read")) + test_sockmap_strp_partial_read(AF_INET, SOCK_STREAM); + if (test__start_subtest("sockmap strp tcp multiple packets")) + test_sockmap_strp_multiple_pkt(AF_INET, SOCK_STREAM); + if (test__start_subtest("sockmap strp tcp dispatch")) + test_sockmap_strp_dispatch_pkt(AF_INET, SOCK_STREAM); +} diff --git a/tools/testing/selftests/bpf/progs/test_sockmap_strp.c b/tools/testing/selftests/bpf/progs/test_sockmap_strp.c new file mode 100644 index 000000000000..dde3d5bec515 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/test_sockmap_strp.c @@ -0,0 +1,53 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +int verdict_max_size = 10000; +struct { + __uint(type, BPF_MAP_TYPE_SOCKMAP); + __uint(max_entries, 20); + __type(key, int); + __type(value, int); +} sock_map SEC(".maps"); + +SEC("sk_skb/stream_verdict") +int prog_skb_verdict(struct __sk_buff *skb) +{ + __u32 one = 1; + + if (skb->len > verdict_max_size) + return SK_PASS; + + return bpf_sk_redirect_map(skb, &sock_map, one, 0); +} + +SEC("sk_skb/stream_verdict") +int prog_skb_verdict_pass(struct __sk_buff *skb) +{ + return SK_PASS; +} + +SEC("sk_skb/stream_parser") +int prog_skb_parser(struct __sk_buff *skb) +{ + return skb->len; +} + +SEC("sk_skb/stream_parser") +int prog_skb_parser_partial(struct __sk_buff *skb) +{ + /* agreement with the test program on a 4-byte size header + * and 6-byte body. + */ + if (skb->len < 4) { + /* need more header to determine full length */ + return 0; + } + /* return full length decoded from header. + * the return value may be larger than skb->len which + * means framework must wait body coming. + */ + return 10; +} + +char _license[] SEC("license") = "GPL"; From c78f4afbd962f43a3989f45f3ca04300252b19b5 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Sat, 21 Dec 2024 14:10:16 +0800 Subject: [PATCH 0262/1443] bpf: Fix deadlock when freeing cgroup storage The following commit bc235cdb423a ("bpf: Prevent deadlock from recursive bpf_task_storage_[get|delete]") first introduced deadlock prevention for fentry/fexit programs attaching on bpf_task_storage helpers. That commit also employed the logic in map free path in its v6 version. Later bpf_cgrp_storage was first introduced in c4bcfb38a95e ("bpf: Implement cgroup storage available to non-cgroup-attached bpf progs") which faces the same issue as bpf_task_storage, instead of its busy counter, NULL was passed to bpf_local_storage_map_free() which opened a window to cause deadlock: (acquiring local_storage->lock) _raw_spin_lock_irqsave+0x3d/0x50 bpf_local_storage_update+0xd1/0x460 bpf_cgrp_storage_get+0x109/0x130 bpf_prog_a4d4a370ba857314_cgrp_ptr+0x139/0x170 ? __bpf_prog_enter_recur+0x16/0x80 bpf_trampoline_6442485186+0x43/0xa4 cgroup_storage_ptr+0x9/0x20 (holding local_storage->lock) bpf_selem_unlink_storage_nolock.constprop.0+0x135/0x160 bpf_selem_unlink_storage+0x6f/0x110 bpf_local_storage_map_free+0xa2/0x110 bpf_map_free_deferred+0x5b/0x90 process_one_work+0x17c/0x390 worker_thread+0x251/0x360 kthread+0xd2/0x100 ret_from_fork+0x34/0x50 ret_from_fork_asm+0x1a/0x30 Progs: - A: SEC("fentry/cgroup_storage_ptr") - cgid (BPF_MAP_TYPE_HASH) Record the id of the cgroup the current task belonging to in this hash map, using the address of the cgroup as the map key. - cgrpa (BPF_MAP_TYPE_CGRP_STORAGE) If current task is a kworker, lookup the above hash map using function parameter @owner as the key to get its corresponding cgroup id which is then used to get a trusted pointer to the cgroup through bpf_cgroup_from_id(). This trusted pointer can then be passed to bpf_cgrp_storage_get() to finally trigger the deadlock issue. - B: SEC("tp_btf/sys_enter") - cgrpb (BPF_MAP_TYPE_CGRP_STORAGE) The only purpose of this prog is to fill Prog A's hash map by calling bpf_cgrp_storage_get() for as many userspace tasks as possible. Steps to reproduce: - Run A; - while (true) { Run B; Destroy B; } Fix this issue by passing its busy counter to the free procedure so it can be properly incremented before storage/smap locking. Fixes: c4bcfb38a95e ("bpf: Implement cgroup storage available to non-cgroup-attached bpf progs") Signed-off-by: Abel Wu Acked-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20241221061018.37717-1-wuyun.abel@bytedance.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/bpf_cgrp_storage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/bpf/bpf_cgrp_storage.c b/kernel/bpf/bpf_cgrp_storage.c index d5dc65bb1755..54ff2a85d4c0 100644 --- a/kernel/bpf/bpf_cgrp_storage.c +++ b/kernel/bpf/bpf_cgrp_storage.c @@ -153,7 +153,7 @@ static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr) static void cgroup_storage_map_free(struct bpf_map *map) { - bpf_local_storage_map_free(map, &cgroup_cache, NULL); + bpf_local_storage_map_free(map, &cgroup_cache, &bpf_cgrp_storage_busy); } /* *gfp_flags* is a hidden argument provided by the verifier */ From 8f6ddb4ab5db955bc826481d2f5c145aa5802ec1 Mon Sep 17 00:00:00 2001 From: Francois Dugast Date: Wed, 29 Jan 2025 18:52:41 +0100 Subject: [PATCH 0263/1443] drm/xe/gt_pagefault: Print engine class string The engine class index which is printed here is an internal representation for debugging. It is _not_ an index based on DRM_XE_ENGINE_CLASS_* values provided in the uAPI. Add the string representation of the engine class to the output in order to limit possible confusion by users when analyzing the logs. Reviewed-by: Matthew Brost Link: https://patchwork.freedesktop.org/patch/msgid/20250129175241.338043-1-francois.dugast@intel.com Signed-off-by: Francois Dugast --- drivers/gpu/drm/xe/xe_gt_pagefault.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c index 79c426dc2505..39344aeab112 100644 --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c @@ -264,12 +264,13 @@ static void print_pagefault(struct xe_device *xe, struct pagefault *pf) "\tFaultType: %d\n" "\tAccessType: %d\n" "\tFaultLevel: %d\n" - "\tEngineClass: %d\n" + "\tEngineClass: %d %s\n" "\tEngineInstance: %d\n", pf->asid, pf->vfid, pf->pdata, upper_32_bits(pf->page_addr), lower_32_bits(pf->page_addr), pf->fault_type, pf->access_type, pf->fault_level, - pf->engine_class, pf->engine_instance); + pf->engine_class, xe_hw_engine_class_to_str(pf->engine_class), + pf->engine_instance); } #define PF_MSG_LEN_DW 4 From bdc35f164b0f60480b2f5e098bb8f3c0cea05cd2 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Fri, 31 Jan 2025 00:27:44 +0900 Subject: [PATCH 0264/1443] tomoyo: use better patterns for procfs in learning mode Commit 08ae2487b202 ("tomoyo: automatically use patterns for several situations in learning mode") replaced only $PID part of procfs pathname with \$ pattern. But it turned out that we need to also replace $TID part and $FD part to make this functionality useful for e.g. /bin/lsof . Signed-off-by: Tetsuo Handa --- security/tomoyo/common.c | 145 ++++++++++++++++++++++++++++++--------- 1 file changed, 112 insertions(+), 33 deletions(-) diff --git a/security/tomoyo/common.c b/security/tomoyo/common.c index d9fa69632147..0f78898bce09 100644 --- a/security/tomoyo/common.c +++ b/security/tomoyo/common.c @@ -1980,6 +1980,114 @@ static int tomoyo_truncate(char *str) return strlen(start) + 1; } +/** + * tomoyo_numscan - sscanf() which stores the length of a decimal integer value. + * + * @str: String to scan. + * @head: Leading string that must start with. + * @width: Pointer to "int" for storing length of a decimal integer value after @head. + * @tail: Optional character that must match after a decimal integer value. + * + * Returns whether @str starts with @head and a decimal value follows @head. + */ +static bool tomoyo_numscan(const char *str, const char *head, int *width, const char tail) +{ + const char *cp; + const int n = strlen(head); + + if (!strncmp(str, head, n)) { + cp = str + n; + while (*cp && *cp >= '0' && *cp <= '9') + cp++; + if (*cp == tail || !tail) { + *width = cp - (str + n); + return *width != 0; + } + } + *width = 0; + return 0; +} + +/** + * tomoyo_patternize_path - Make patterns for file path. Used by learning mode. + * + * @buffer: Destination buffer. + * @len: Size of @buffer. + * @entry: Original line. + * + * Returns nothing. + */ +static void tomoyo_patternize_path(char *buffer, const int len, char *entry) +{ + int width; + char *cp = entry; + + /* Nothing to do if this line is not for "file" related entry. */ + if (strncmp(entry, "file ", 5)) + goto flush; + /* + * Nothing to do if there is no colon in this line, for this rewriting + * applies to only filesystems where numeric values in the path are volatile. + */ + cp = strchr(entry + 5, ':'); + if (!cp) { + cp = entry; + goto flush; + } + /* Flush e.g. "file ioctl" part. */ + while (*cp != ' ') + cp--; + *cp++ = '\0'; + tomoyo_addprintf(buffer, len, "%s ", entry); + /* e.g. file ioctl pipe:[$INO] $CMD */ + if (tomoyo_numscan(cp, "pipe:[", &width, ']')) { + cp += width + 7; + tomoyo_addprintf(buffer, len, "pipe:[\\$]"); + goto flush; + } + /* e.g. file ioctl socket:[$INO] $CMD */ + if (tomoyo_numscan(cp, "socket:[", &width, ']')) { + cp += width + 9; + tomoyo_addprintf(buffer, len, "socket:[\\$]"); + goto flush; + } + if (!strncmp(cp, "proc:/self", 10)) { + /* e.g. file read proc:/self/task/$TID/fdinfo/$FD */ + cp += 10; + tomoyo_addprintf(buffer, len, "proc:/self"); + } else if (tomoyo_numscan(cp, "proc:/", &width, 0)) { + /* e.g. file read proc:/$PID/task/$TID/fdinfo/$FD */ + /* + * Don't patternize $PID part if $PID == 1, for several + * programs access only files in /proc/1/ directory. + */ + cp += width + 6; + if (width == 1 && *(cp - 1) == '1') + tomoyo_addprintf(buffer, len, "proc:/1"); + else + tomoyo_addprintf(buffer, len, "proc:/\\$"); + } else { + goto flush; + } + /* Patternize $TID part if "/task/" follows. */ + if (tomoyo_numscan(cp, "/task/", &width, 0)) { + cp += width + 6; + tomoyo_addprintf(buffer, len, "/task/\\$"); + } + /* Patternize $FD part if "/fd/" or "/fdinfo/" follows. */ + if (tomoyo_numscan(cp, "/fd/", &width, 0)) { + cp += width + 4; + tomoyo_addprintf(buffer, len, "/fd/\\$"); + } else if (tomoyo_numscan(cp, "/fdinfo/", &width, 0)) { + cp += width + 8; + tomoyo_addprintf(buffer, len, "/fdinfo/\\$"); + } +flush: + /* Flush remaining part if any. */ + if (*cp) + tomoyo_addprintf(buffer, len, "%s", cp); +} + /** * tomoyo_add_entry - Add an ACL to current thread's domain. Used by learning mode. * @@ -2003,7 +2111,8 @@ static void tomoyo_add_entry(struct tomoyo_domain_info *domain, char *header) if (!cp) return; *cp++ = '\0'; - len = strlen(cp) + 1; + /* Reserve some space for potentially using patterns. */ + len = strlen(cp) + 16; /* strstr() will return NULL if ordering is wrong. */ if (*cp == 'f') { argv0 = strstr(header, " argv[]={ \""); @@ -2020,40 +2129,10 @@ static void tomoyo_add_entry(struct tomoyo_domain_info *domain, char *header) if (symlink) len += tomoyo_truncate(symlink + 1) + 1; } - buffer = kmalloc(len, GFP_NOFS); + buffer = kmalloc(len, GFP_NOFS | __GFP_ZERO); if (!buffer) return; - snprintf(buffer, len - 1, "%s", cp); - if (*cp == 'f' && strchr(buffer, ':')) { - /* Automatically replace 2 or more digits with \$ pattern. */ - char *cp2; - - /* e.g. file read proc:/$PID/stat */ - cp = strstr(buffer, " proc:/"); - if (cp && simple_strtoul(cp + 7, &cp2, 10) >= 10 && *cp2 == '/') { - *(cp + 7) = '\\'; - *(cp + 8) = '$'; - memmove(cp + 9, cp2, strlen(cp2) + 1); - goto ok; - } - /* e.g. file ioctl pipe:[$INO] $CMD */ - cp = strstr(buffer, " pipe:["); - if (cp && simple_strtoul(cp + 7, &cp2, 10) >= 10 && *cp2 == ']') { - *(cp + 7) = '\\'; - *(cp + 8) = '$'; - memmove(cp + 9, cp2, strlen(cp2) + 1); - goto ok; - } - /* e.g. file ioctl socket:[$INO] $CMD */ - cp = strstr(buffer, " socket:["); - if (cp && simple_strtoul(cp + 9, &cp2, 10) >= 10 && *cp2 == ']') { - *(cp + 9) = '\\'; - *(cp + 10) = '$'; - memmove(cp + 11, cp2, strlen(cp2) + 1); - goto ok; - } - } -ok: + tomoyo_patternize_path(buffer, len, cp); if (realpath) tomoyo_addprintf(buffer, len, " exec.%s", realpath); if (argv0) From 21ccac0e22aaf27b767f9de4bf573e7c47f619c8 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Wed, 29 Jan 2025 20:59:46 +0100 Subject: [PATCH 0265/1443] drm/xe/pf: Don't send BEGIN_ID if VF has no context/doorbells MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It turned out that GuC validates VF configuration immediately after receiving "some" set of configuration KLVs and complains if one of the critical, from GuC understanding, resource is left unprovisioned, even if PF should be still allowed to make late VF config adjustments, since VF was not yet started. This issue was discovered after we decided to asynchronously re-send configuration KLVs after GT reset/resume, as then fair VF auto-provisioning could already allocate some of the resources, which was a prerequiste for sending those config KLVs: # fair GGTT provisioning [] xe 0000:00:02.0: [drm] GT0: PF: pushed VF1 config with 2 KLVs: [] xe 0000:00:02.0: [drm] GT0: { key 0x0001 : 64b value 0x176a000 } # ggtt_start [] xe 0000:00:02.0: [drm] GT0: { key 0x0002 : 64b value 0xfd696000 } # ggtt_size [] xe 0000:00:02.0: [drm] GT0: PF: VF1 provisioned with 4251541504 (3.96 GiB) GGTT # re-provisioning worker [] xe 0000:00:02.0: [drm] *ERROR* GT0: H2G request 0x5503 failed: error 0x60 hint 0x0 [] xe 0000:00:02.0: [drm] GT0: PF: Failed to push VF1 14 config KLVs (-EIO) [] xe 0000:00:02.0: [drm] GT0: { key 0x0001 : 64b value 0x176a000 } # ggtt_start [] xe 0000:00:02.0: [drm] GT0: { key 0x0002 : 64b value 0xfd696000 } # ggtt_size [] xe 0000:00:02.0: [drm] GT0: { key 0x8a0b : 32b value 0 } # begin_ctx_id [] xe 0000:00:02.0: [drm] GT0: { key 0x0004 : 32b value 0 } # num_contexts [] xe 0000:00:02.0: [drm] GT0: { key 0x8a0a : 32b value 0 } # begin_db_id [] xe 0000:00:02.0: [drm] GT0: { key 0x0006 : 32b value 0 } # num_doorbells [] xe 0000:00:02.0: [drm] GT0: { key 0x8a01 : 32b value 0 } # exec_quantum [] xe 0000:00:02.0: [drm] GT0: { key 0x8a02 : 32b value 0 } # preempt_timeout [] xe 0000:00:02.0: [drm] GT0: { key 0x8a03 : 32b value 0 } # cat_error_count [] xe 0000:00:02.0: [drm] GT0: { key 0x8a04 : 32b value 0 } # engine_reset_count [] xe 0000:00:02.0: [drm] GT0: { key 0x8a05 : 32b value 0 } # page_fault_count [] xe 0000:00:02.0: [drm] GT0: { key 0x8a06 : 32b value 0 } # guc_time_us [] xe 0000:00:02.0: [drm] GT0: { key 0x8a07 : 32b value 0 } # irq_time_us [] xe 0000:00:02.0: [drm] GT0: { key 0x8a08 : 32b value 0 } # doorbell_time_us [] xe 0000:00:02.0: [drm] GT0: PF: Failed to push VF1 configuration (-EIO) To avoid such errors stop sending BEGIN_CONTEXT/DOORBELL_ID KLVs if no GuC context/doorbell IDs were provisioned to VF. Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4176 Signed-off-by: Michal Wajdeczko Reviewed-by: Michał Winiarski Link: https://patchwork.freedesktop.org/patch/msgid/20250129195947.764-2-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 9db1c920219d..c219c55f23ab 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -264,7 +264,7 @@ static u32 encode_config(u32 *cfg, const struct xe_gt_sriov_config *config, bool n += encode_config_ggtt(cfg, config, details); - if (details) { + if (details && config->num_ctxs) { cfg[n++] = PREP_GUC_KLV_TAG(VF_CFG_BEGIN_CONTEXT_ID); cfg[n++] = config->begin_ctx; } @@ -272,7 +272,7 @@ static u32 encode_config(u32 *cfg, const struct xe_gt_sriov_config *config, bool cfg[n++] = PREP_GUC_KLV_TAG(VF_CFG_NUM_CONTEXTS); cfg[n++] = config->num_ctxs; - if (details) { + if (details && config->num_dbs) { cfg[n++] = PREP_GUC_KLV_TAG(VF_CFG_BEGIN_DOORBELL_ID); cfg[n++] = config->begin_db; } From 33f17e2cbd930a2a00eb007d9b241b6db010a880 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Wed, 29 Jan 2025 20:59:47 +0100 Subject: [PATCH 0266/1443] drm/xe/pf: Reset GuC VF config when unprovisioning critical resource MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit GuC firmware counts received VF configuration KLVs and may start validation of the complete VF config even if some resources where unprovisioned in the meantime, leading to unexpected errors like: $ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/contexts_quota $ echo 0 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/contexts_quota $ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/doorbells_quota $ echo 0 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/doorbells_quota $ echo 1 | sudo tee /sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/ggtt_quota tee: '/sys/kernel/debug/dri/0000:00:02.0/gt0/vf1/ggtt_quota': Input/output error To mitigate this problem trigger explicit VF config reset after unprovisioning any of the critical resources (GGTT, context or doorbell IDs) that GuC is monitoring. Signed-off-by: Michal Wajdeczko Reviewed-by: Michał Winiarski Link: https://patchwork.freedesktop.org/patch/msgid/20250129195947.764-3-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 37 +++++++++++++++++++--- 1 file changed, 33 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index c219c55f23ab..b1d994d65589 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -338,6 +338,26 @@ static int pf_push_full_vf_config(struct xe_gt *gt, unsigned int vfid) return err; } +static int pf_push_vf_cfg(struct xe_gt *gt, unsigned int vfid, bool reset) +{ + int err = 0; + + xe_gt_assert(gt, vfid); + lockdep_assert_held(xe_gt_sriov_pf_master_mutex(gt)); + + if (reset) + err = pf_send_vf_cfg_reset(gt, vfid); + if (!err) + err = pf_push_full_vf_config(gt, vfid); + + return err; +} + +static int pf_refresh_vf_cfg(struct xe_gt *gt, unsigned int vfid) +{ + return pf_push_vf_cfg(gt, vfid, true); +} + static u64 pf_get_ggtt_alignment(struct xe_gt *gt) { struct xe_device *xe = gt_to_xe(gt); @@ -434,6 +454,10 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size) return err; pf_release_vf_config_ggtt(gt, config); + + err = pf_refresh_vf_cfg(gt, vfid); + if (unlikely(err)) + return err; } xe_gt_assert(gt, !xe_ggtt_node_allocated(config->ggtt_region)); @@ -759,6 +783,10 @@ static int pf_provision_vf_ctxs(struct xe_gt *gt, unsigned int vfid, u32 num_ctx return ret; pf_release_config_ctxs(gt, config); + + ret = pf_refresh_vf_cfg(gt, vfid); + if (unlikely(ret)) + return ret; } if (!num_ctxs) @@ -1056,6 +1084,10 @@ static int pf_provision_vf_dbs(struct xe_gt *gt, unsigned int vfid, u32 num_dbs) return ret; pf_release_config_dbs(gt, config); + + ret = pf_refresh_vf_cfg(gt, vfid); + if (unlikely(ret)) + return ret; } if (!num_dbs) @@ -2087,10 +2119,7 @@ int xe_gt_sriov_pf_config_push(struct xe_gt *gt, unsigned int vfid, bool refresh xe_gt_assert(gt, vfid); mutex_lock(xe_gt_sriov_pf_master_mutex(gt)); - if (refresh) - err = pf_send_vf_cfg_reset(gt, vfid); - if (!err) - err = pf_push_full_vf_config(gt, vfid); + err = pf_push_vf_cfg(gt, vfid, refresh); mutex_unlock(xe_gt_sriov_pf_master_mutex(gt)); if (unlikely(err)) { From c13a42f2107c768715fbac518072363b57258e36 Mon Sep 17 00:00:00 2001 From: Gustavo Sousa Date: Wed, 15 Jan 2025 11:08:04 -0300 Subject: [PATCH 0267/1443] drm/xe: Fix sort order of .o lists in Makefile The Makefile for xe asks us to keep the lists of object files sorted: # Please keep these build lists sorted! Reshuffle the lists into the correct sort order. That was done by filtering each unsorted list through 'LC_ALL=C sort'. Signed-off-by: Gustavo Sousa Reviewed-by: Lucas De Marchi Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250115140812.20799-1-gustavo.sousa@intel.com Signed-off-by: Matt Roper --- drivers/gpu/drm/xe/Makefile | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index ef900b9f0457..fdd3a0c9cb37 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -34,8 +34,8 @@ xe-y += xe_bb.o \ xe_dma_buf.o \ xe_drm_client.o \ xe_exec.o \ - xe_execlist.o \ xe_exec_queue.o \ + xe_execlist.o \ xe_force_wake.o \ xe_ggtt.o \ xe_gpu_scheduler.o \ @@ -67,11 +67,11 @@ xe-y += xe_bb.o \ xe_guc_pc.o \ xe_guc_submit.o \ xe_heci_gsc.o \ + xe_huc.o \ xe_hw_engine.o \ xe_hw_engine_class_sysfs.o \ xe_hw_engine_group.o \ xe_hw_fence.o \ - xe_huc.o \ xe_irq.o \ xe_lrc.o \ xe_migrate.o \ @@ -91,8 +91,8 @@ xe-y += xe_bb.o \ xe_range_fence.o \ xe_reg_sr.o \ xe_reg_whitelist.o \ - xe_rtp.o \ xe_ring_ops.o \ + xe_rtp.o \ xe_sa.o \ xe_sched_job.o \ xe_step.o \ @@ -104,8 +104,8 @@ xe-y += xe_bb.o \ xe_trace_bo.o \ xe_trace_guc.o \ xe_trace_lrc.o \ - xe_ttm_sys_mgr.o \ xe_ttm_stolen_mgr.o \ + xe_ttm_sys_mgr.o \ xe_ttm_vram_mgr.o \ xe_tuning.o \ xe_uc.o \ @@ -114,8 +114,8 @@ xe-y += xe_bb.o \ xe_vram.o \ xe_vram_freq.o \ xe_vsec.o \ - xe_wait_user_fence.o \ xe_wa.o \ + xe_wait_user_fence.o \ xe_wopcm.o xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o @@ -224,6 +224,7 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \ i915-display/intel_display_wa.o \ i915-display/intel_dkl_phy.o \ i915-display/intel_dmc.o \ + i915-display/intel_dmc_wl.o \ i915-display/intel_dp.o \ i915-display/intel_dp_aux.o \ i915-display/intel_dp_aux_backlight.o \ @@ -271,7 +272,6 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \ i915-display/intel_vdsc.o \ i915-display/intel_vga.o \ i915-display/intel_vrr.o \ - i915-display/intel_dmc_wl.o \ i915-display/intel_wm.o \ i915-display/skl_scaler.o \ i915-display/skl_universal_plane.o \ From ee2ab467bddfb2d7f68d996dbab94d7b88f8eaf7 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 21 Jan 2025 18:11:33 -0700 Subject: [PATCH 0268/1443] x86/boot: Use '-std=gnu11' to fix build with GCC 15 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit GCC 15 changed the default C standard version to C23, which should not have impacted the kernel because it requests the gnu11 standard via '-std=' in the main Makefile. However, the x86 compressed boot Makefile uses its own set of KBUILD_CFLAGS without a '-std=' value (i.e., using the default), resulting in errors from the kernel's definitions of bool, true, and false in stddef.h, which are reserved keywords under C23. ./include/linux/stddef.h:11:9: error: expected identifier before ‘false’ 11 | false = 0, ./include/linux/types.h:35:33: error: two or more data types in declaration specifiers 35 | typedef _Bool bool; Set '-std=gnu11' in the x86 compressed boot Makefile to resolve the error and consistently use the same C standard version for the entire kernel. Closes: https://lore.kernel.org/4OAhbllK7x4QJGpZjkYjtBYNLd_2whHx9oFiuZcGwtVR4hIzvduultkgfAIRZI3vQpZylu7Gl929HaYFRGeMEalWCpeMzCIIhLxxRhq4U-Y=@protonmail.com/ Closes: https://lore.kernel.org/Z4467umXR2PZ0M1H@tucnak/ Reported-by: Kostadin Shishmanov Reported-by: Jakub Jelinek Signed-off-by: Nathan Chancellor Signed-off-by: Dave Hansen Reviewed-by: Ard Biesheuvel Cc:stable@vger.kernel.org Link: https://lore.kernel.org/all/20250121-x86-use-std-consistently-gcc-15-v1-1-8ab0acf645cb%40kernel.org --- arch/x86/boot/compressed/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile index f2051644de94..606c74f27459 100644 --- a/arch/x86/boot/compressed/Makefile +++ b/arch/x86/boot/compressed/Makefile @@ -25,6 +25,7 @@ targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma \ # avoid errors with '-march=i386', and future flags may depend on the target to # be valid. KBUILD_CFLAGS := -m$(BITS) -O2 $(CLANG_FLAGS) +KBUILD_CFLAGS += -std=gnu11 KBUILD_CFLAGS += -fno-strict-aliasing -fPIE KBUILD_CFLAGS += -Wundef KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING From b73aebc7a1e0fd9d2a9d8ab7b88ada58bd80bb9f Mon Sep 17 00:00:00 2001 From: Jakub Kolakowski Date: Tue, 28 Jan 2025 11:03:00 +0000 Subject: [PATCH 0269/1443] drm/xe/pf: Add runtime registers for graphics gen >= 30 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add missing runtime registers for graphics versions of 3000 or higher. This is required for Xe3 where additionally we have MIRROR_L3BANK_ENABLE register. Signed-off-by: Jakub Kolakowski Suggested-by: Piotr Piórkowski Cc: Adam Miszczak Cc: Jakub Kolakowski Cc: Lukasz Laguna Cc: Marcin Bernatowicz Cc: Michal Wajdeczko Cc: Piotr Piorkowski Cc: Satyanarayana K V P Reviewed-by: Satyanarayana K V P Tested-by: Marcin Bernatowicz Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250128110300.2840596-2-jakub1.kolakowski@intel.com --- drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c | 23 ++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c index 924e75b94aec..6b5f849a0722 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c @@ -176,11 +176,32 @@ static const struct xe_reg ver_2000_runtime_regs[] = { TIMESTAMP_OVERRIDE, /* _MMIO(0x44074) */ }; +static const struct xe_reg ver_3000_runtime_regs[] = { + RPM_CONFIG0, /* _MMIO(0x0d00) */ + XEHP_FUSE4, /* _MMIO(0x9114) */ + MIRROR_FUSE3, /* _MMIO(0x9118) */ + MIRROR_FUSE1, /* _MMIO(0x911c) */ + MIRROR_L3BANK_ENABLE, /* _MMIO(0x9130) */ + XELP_EU_ENABLE, /* _MMIO(0x9134) */ + XELP_GT_GEOMETRY_DSS_ENABLE, /* _MMIO(0x913c) */ + GT_VEBOX_VDBOX_DISABLE, /* _MMIO(0x9140) */ + XEHP_GT_COMPUTE_DSS_ENABLE, /* _MMIO(0x9144) */ + XEHPC_GT_COMPUTE_DSS_ENABLE_EXT,/* _MMIO(0x9148) */ + XE2_GT_COMPUTE_DSS_2, /* _MMIO(0x914c) */ + XE2_GT_GEOMETRY_DSS_1, /* _MMIO(0x9150) */ + XE2_GT_GEOMETRY_DSS_2, /* _MMIO(0x9154) */ + CTC_MODE, /* _MMIO(0xa26c) */ + HUC_KERNEL_LOAD_INFO, /* _MMIO(0xc1dc) */ +}; + static const struct xe_reg *pick_runtime_regs(struct xe_device *xe, unsigned int *count) { const struct xe_reg *regs; - if (GRAPHICS_VERx100(xe) >= 2000) { + if (GRAPHICS_VERx100(xe) >= 3000) { + *count = ARRAY_SIZE(ver_3000_runtime_regs); + regs = ver_3000_runtime_regs; + } else if (GRAPHICS_VERx100(xe) >= 2000) { *count = ARRAY_SIZE(ver_2000_runtime_regs); regs = ver_2000_runtime_regs; } else if (GRAPHICS_VERx100(xe) >= 1270) { From bb2784d9ab49587ba4fbff37a319fff2924db289 Mon Sep 17 00:00:00 2001 From: Easwar Hariharan Date: Thu, 30 Jan 2025 19:26:58 +0000 Subject: [PATCH 0270/1443] jiffies: Cast to unsigned long in secs_to_jiffies() conversion While converting users of msecs_to_jiffies(), lkp reported that some range checks would always be true because of the mismatch between the implied int value of secs_to_jiffies() vs the unsigned long return value of the msecs_to_jiffies() calls it was replacing. Fix this by casting the secs_to_jiffies() input value to unsigned long. Fixes: b35108a51cf7ba ("jiffies: Define secs_to_jiffies()") Reported-by: kernel test robot Signed-off-by: Easwar Hariharan Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250130192701.99626-1-eahariha@linux.microsoft.com Closes: https://lore.kernel.org/oe-kbuild-all/202501301334.NB6NszQR-lkp@intel.com/ --- include/linux/jiffies.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index ed945f42e064..0ea8c9887429 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -537,7 +537,7 @@ static __always_inline unsigned long msecs_to_jiffies(const unsigned int m) * * Return: jiffies value */ -#define secs_to_jiffies(_secs) ((_secs) * HZ) +#define secs_to_jiffies(_secs) (unsigned long)((_secs) * HZ) extern unsigned long __usecs_to_jiffies(const unsigned int u); #if !(USEC_PER_SEC % HZ) From cf29a866a139519d6274756d65ebc29c9d749034 Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 21 Jan 2025 15:28:48 +0100 Subject: [PATCH 0271/1443] drm/xe/display: Add intel_plane_initial_vblank_wait We're changing the driver to have no interrupts during early init for Xe, so we poll the PIPE_FRMSTMSMP counter instead. Interrupts cannot be enabled during FB readout because memirq's requires an allocation. This would overwrite the FB we want to read out. While it might be possible to also run do the same in i915 and run it without interrupts, the platforms i915 supports had a less clear distinction between display and graphics. For this reason I choose only to touch Xe for now. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250121142850.4960-1-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/i915/display/intel_display.c | 6 +++--- .../drm/i915/display/intel_plane_initial.c | 7 ++++++- .../drm/i915/display/intel_plane_initial.h | 2 ++ drivers/gpu/drm/xe/display/xe_plane_initial.c | 19 ++++++++++++++++++- 4 files changed, 29 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 4271da219b41..6f6a535ea486 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -796,7 +796,7 @@ void intel_plane_disable_noatomic(struct intel_crtc *crtc, if ((crtc_state->active_planes & ~BIT(PLANE_CURSOR)) == 0 && hsw_ips_disable(crtc_state)) { crtc_state->ips_enabled = false; - intel_crtc_wait_for_next_vblank(crtc); + intel_plane_initial_vblank_wait(crtc); } /* @@ -810,7 +810,7 @@ void intel_plane_disable_noatomic(struct intel_crtc *crtc, */ if (HAS_GMCH(dev_priv) && intel_set_memory_cxsr(dev_priv, false)) - intel_crtc_wait_for_next_vblank(crtc); + intel_plane_initial_vblank_wait(crtc); /* * Gen2 reports pipe underruns whenever all planes are disabled. @@ -820,7 +820,7 @@ void intel_plane_disable_noatomic(struct intel_crtc *crtc, intel_set_cpu_fifo_underrun_reporting(dev_priv, crtc->pipe, false); intel_plane_disable_arm(NULL, plane, crtc_state); - intel_crtc_wait_for_next_vblank(crtc); + intel_plane_initial_vblank_wait(crtc); } unsigned int diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c index 6789b7f14095..b1675b46e06c 100644 --- a/drivers/gpu/drm/i915/display/intel_plane_initial.c +++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c @@ -14,6 +14,11 @@ #include "intel_frontbuffer.h" #include "intel_plane_initial.h" +void intel_plane_initial_vblank_wait(struct intel_crtc *crtc) +{ + intel_crtc_wait_for_next_vblank(crtc); +} + static bool intel_reuse_initial_plane_obj(struct intel_crtc *this, const struct intel_initial_plane_config plane_configs[], @@ -442,7 +447,7 @@ void intel_initial_plane_config(struct intel_display *display) intel_find_initial_plane_obj(crtc, plane_configs); if (display->funcs.display->fixup_initial_plane_config(crtc, plane_config)) - intel_crtc_wait_for_next_vblank(crtc); + intel_plane_initial_vblank_wait(crtc); plane_config_fini(plane_config); } diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.h b/drivers/gpu/drm/i915/display/intel_plane_initial.h index 6c6aa717ed21..5c315acda210 100644 --- a/drivers/gpu/drm/i915/display/intel_plane_initial.h +++ b/drivers/gpu/drm/i915/display/intel_plane_initial.h @@ -6,8 +6,10 @@ #ifndef __INTEL_PLANE_INITIAL_H__ #define __INTEL_PLANE_INITIAL_H__ +struct intel_crtc; struct intel_display; void intel_initial_plane_config(struct intel_display *display); +void intel_plane_initial_vblank_wait(struct intel_crtc *crtc); #endif diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c index 2eb9633f163a..a22e7adfb09d 100644 --- a/drivers/gpu/drm/xe/display/xe_plane_initial.c +++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c @@ -8,7 +8,9 @@ #include "regs/xe_gtt_defs.h" #include "xe_ggtt.h" +#include "xe_mmio.h" +#include "i915_reg.h" #include "intel_atomic_plane.h" #include "intel_crtc.h" #include "intel_display.h" @@ -22,6 +24,21 @@ #include +void intel_plane_initial_vblank_wait(struct intel_crtc *crtc) +{ + /* Early xe has no irq */ + struct xe_device *xe = to_xe_device(crtc->base.dev); + struct xe_reg pipe_frmtmstmp = XE_REG(i915_mmio_reg_offset(PIPE_FRMTMSTMP(crtc->pipe))); + u32 timestamp; + int ret; + + timestamp = xe_mmio_read32(xe_root_tile_mmio(xe), pipe_frmtmstmp); + + ret = xe_mmio_wait32_not(xe_root_tile_mmio(xe), pipe_frmtmstmp, ~0U, timestamp, 40000U, ×tamp, false); + if (ret < 0) + drm_warn(&xe->drm, "waiting for early vblank failed with %i\n", ret); +} + static bool intel_reuse_initial_plane_obj(struct intel_crtc *this, const struct intel_initial_plane_config plane_configs[], @@ -303,7 +320,7 @@ void intel_initial_plane_config(struct intel_display *display) intel_find_initial_plane_obj(crtc, plane_configs); if (display->funcs.display->fixup_initial_plane_config(crtc, plane_config)) - intel_crtc_wait_for_next_vblank(crtc); + intel_plane_initial_vblank_wait(crtc); plane_config_fini(plane_config); } From f595fe5f6ae7d74d7352e2577ca1577704b64b3b Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 21 Jan 2025 15:28:49 +0100 Subject: [PATCH 0272/1443] drm/xe: Defer irq init until after xe_display_init_noaccel As stated in previous commit, we have to move interrupt handling until after xe_display_init_noaccel, as using memirqs would require an allocation. A full solution will of course require memirq allocation to be moved, but the first part only focuses on the required changes to display. Reviewed-by: Ilia Levi Link: https://patchwork.freedesktop.org/patch/msgid/20250121142850.4960-2-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/xe/xe_device.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 8fedc72e9db4..4d401e4da70b 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -814,10 +814,6 @@ int xe_device_probe(struct xe_device *xe) if (err) return err; - err = xe_irq_install(xe); - if (err) - goto err; - err = probe_has_flat_ccs(xe); if (err) goto err; @@ -851,6 +847,10 @@ int xe_device_probe(struct xe_device *xe) goto err; } + err = xe_irq_install(xe); + if (err) + goto err; + for_each_gt(gt, xe, id) { last_gt = id; From 65e366ace5ee3637179658e8cf37f934c857f563 Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Tue, 21 Jan 2025 15:28:50 +0100 Subject: [PATCH 0273/1443] drm/xe/display: Use a single early init call for display Now that interrupts are disabled for xe_display_init_noaccel, both xe_display_init_noirq and xe_display_init_noaccel run in the same context. This means that we can get rid of the 3 different init calls. Without interrupts, nothing is touching display up to this point. Unify those 3 early display calls into a single xe_display_init_early(), this makes the init sequence cleaner, and display less tangled during init. Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250121142850.4960-3-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/xe/display/xe_display.c | 73 +++++++------------------ drivers/gpu/drm/xe/display/xe_display.h | 8 +-- drivers/gpu/drm/xe/xe_device.c | 10 +--- 3 files changed, 22 insertions(+), 69 deletions(-) diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c index b3921dbc52ff..7f0d8f00acff 100644 --- a/drivers/gpu/drm/xe/display/xe_display.c +++ b/drivers/gpu/drm/xe/display/xe_display.c @@ -101,19 +101,25 @@ int xe_display_create(struct xe_device *xe) return drmm_add_action_or_reset(&xe->drm, display_destroy, NULL); } -static void xe_display_fini_nommio(struct drm_device *dev, void *dummy) +static void xe_display_fini_early(void *arg) { - struct xe_device *xe = to_xe_device(dev); + struct xe_device *xe = arg; struct intel_display *display = &xe->display; if (!xe->info.probe_display) return; + intel_display_driver_remove_nogem(display); + intel_display_driver_remove_noirq(display); + intel_opregion_cleanup(display); intel_power_domains_cleanup(display); } -int xe_display_init_nommio(struct xe_device *xe) +int xe_display_init_early(struct xe_device *xe) { + struct intel_display *display = &xe->display; + int err; + if (!xe->info.probe_display) return 0; @@ -123,29 +129,6 @@ int xe_display_init_nommio(struct xe_device *xe) /* This must be called before any calls to HAS_PCH_* */ intel_detect_pch(xe); - return drmm_add_action_or_reset(&xe->drm, xe_display_fini_nommio, xe); -} - -static void xe_display_fini_noirq(void *arg) -{ - struct xe_device *xe = arg; - struct intel_display *display = &xe->display; - - if (!xe->info.probe_display) - return; - - intel_display_driver_remove_noirq(display); - intel_opregion_cleanup(display); -} - -int xe_display_init_noirq(struct xe_device *xe) -{ - struct intel_display *display = &xe->display; - int err; - - if (!xe->info.probe_display) - return 0; - intel_display_driver_early_probe(display); /* Early display init.. */ @@ -162,38 +145,20 @@ int xe_display_init_noirq(struct xe_device *xe) intel_display_device_info_runtime_init(display); err = intel_display_driver_probe_noirq(display); - if (err) { - intel_opregion_cleanup(display); - return err; - } - - return devm_add_action_or_reset(xe->drm.dev, xe_display_fini_noirq, xe); -} - -static void xe_display_fini_noaccel(void *arg) -{ - struct xe_device *xe = arg; - struct intel_display *display = &xe->display; - - if (!xe->info.probe_display) - return; - - intel_display_driver_remove_nogem(display); -} - -int xe_display_init_noaccel(struct xe_device *xe) -{ - struct intel_display *display = &xe->display; - int err; - - if (!xe->info.probe_display) - return 0; + if (err) + goto err_opregion; err = intel_display_driver_probe_nogem(display); if (err) - return err; + goto err_noirq; - return devm_add_action_or_reset(xe->drm.dev, xe_display_fini_noaccel, xe); + return devm_add_action_or_reset(xe->drm.dev, xe_display_fini_early, xe); +err_noirq: + intel_display_driver_remove_noirq(display); + intel_power_domains_cleanup(display); +err_opregion: + intel_opregion_cleanup(display); + return err; } int xe_display_init(struct xe_device *xe) diff --git a/drivers/gpu/drm/xe/display/xe_display.h b/drivers/gpu/drm/xe/display/xe_display.h index 233f81a26c25..e2a99624f706 100644 --- a/drivers/gpu/drm/xe/display/xe_display.h +++ b/drivers/gpu/drm/xe/display/xe_display.h @@ -20,9 +20,7 @@ int xe_display_create(struct xe_device *xe); int xe_display_probe(struct xe_device *xe); -int xe_display_init_nommio(struct xe_device *xe); -int xe_display_init_noirq(struct xe_device *xe); -int xe_display_init_noaccel(struct xe_device *xe); +int xe_display_init_early(struct xe_device *xe); int xe_display_init(struct xe_device *xe); void xe_display_fini(struct xe_device *xe); @@ -54,9 +52,7 @@ static inline int xe_display_create(struct xe_device *xe) { return 0; } static inline int xe_display_probe(struct xe_device *xe) { return 0; } -static inline int xe_display_init_nommio(struct xe_device *xe) { return 0; } -static inline int xe_display_init_noirq(struct xe_device *xe) { return 0; } -static inline int xe_display_init_noaccel(struct xe_device *xe) { return 0; } +static inline int xe_display_init_early(struct xe_device *xe) { return 0; } static inline int xe_display_init(struct xe_device *xe) { return 0; } static inline void xe_display_fini(struct xe_device *xe) {} diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 4d401e4da70b..8a2763e8a6e2 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -752,10 +752,6 @@ int xe_device_probe(struct xe_device *xe) return err; xe->info.mem_region_mask = 1; - err = xe_display_init_nommio(xe); - if (err) - return err; - err = xe_set_dma_info(xe); if (err) return err; @@ -810,10 +806,6 @@ int xe_device_probe(struct xe_device *xe) if (err) return err; - err = xe_display_init_noirq(xe); - if (err) - return err; - err = probe_has_flat_ccs(xe); if (err) goto err; @@ -837,7 +829,7 @@ int xe_device_probe(struct xe_device *xe) * This is the reason the first allocation needs to be done * inside display. */ - err = xe_display_init_noaccel(xe); + err = xe_display_init_early(xe); if (err) goto err; From 9065ce69754dece78606c8bbb3821449272e56bf Mon Sep 17 00:00:00 2001 From: Christian Loehle Date: Wed, 29 Jan 2025 17:59:44 +0000 Subject: [PATCH 0274/1443] sched/debug: Provide slice length for fair tasks Since commit: 857b158dc5e8 ("sched/eevdf: Use sched_attr::sched_runtime to set request/slice suggestion") ... we have the userspace per-task tunable slice length, which is a key parameter that is otherwise difficult to obtain, so provide it in /proc/$PID/sched. [ mingo: Clarified the changelog. ] Signed-off-by: Christian Loehle Signed-off-by: Ingo Molnar Cc: Peter Zijlstra Link: https://lore.kernel.org/r/453349b1-1637-42f5-a7b2-2385392b5956@arm.com --- kernel/sched/debug.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index a1be00a988bf..5b32d3cc393b 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -1265,6 +1265,8 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, if (task_has_dl_policy(p)) { P(dl.runtime); P(dl.deadline); + } else if (fair_policy(p->policy)) { + P(se.slice); } #ifdef CONFIG_SCHED_CLASS_EXT __PS("ext.enabled", task_on_scx(p)); From d9bc304437da6b74ac2b6644fe47702b8286eb8d Mon Sep 17 00:00:00 2001 From: Riana Tauro Date: Fri, 31 Jan 2025 13:35:27 +0530 Subject: [PATCH 0275/1443] drm/xe: Skip survivability mode for VF Follow the probe flow in case of VF and do not enter survivability mode in case of pcode init failure. Fixes: 5e940312a2ac ("drm/xe: Add functions and sysfs for boot survivability") Suggested-by: Satyanarayana K V P Signed-off-by: Riana Tauro Reviewed-by: Satyanarayana K V P Link: https://patchwork.freedesktop.org/patch/msgid/20250131080527.2256475-1-riana.tauro@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_survivability_mode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c index c619560af74f..02b4eadf8407 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.c +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -177,7 +177,7 @@ bool xe_survivability_mode_required(struct xe_device *xe) struct xe_mmio *mmio = xe_root_tile_mmio(xe); u32 data; - if (!IS_DGFX(xe) || xe->info.platform < XE_BATTLEMAGE) + if (!IS_DGFX(xe) || xe->info.platform < XE_BATTLEMAGE || IS_SRIOV_VF(xe)) return false; data = xe_mmio_read32(mmio, PCODE_SCRATCH(0)); From 5f230f41fdd9e799f43a699348dc572bca7159aa Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 7 Jan 2025 16:43:41 +0100 Subject: [PATCH 0276/1443] KVM: s390: vsie: fix some corner-cases when grabbing vsie pages We try to reuse the same vsie page when re-executing the vsie with a given SCB address. The result is that we use the same shadow SCB -- residing in the vsie page -- and can avoid flushing the TLB when re-running the vsie on a CPU. So, when we allocate a fresh vsie page, or when we reuse a vsie page for a different SCB address -- reusing the shadow SCB in different context -- we set ihcpu=0xffff to trigger the flush. However, after we looked up the SCB address in the radix tree, but before we grabbed the vsie page by raising the refcount to 2, someone could reuse the vsie page for a different SCB address, adjusting page->index and the radix tree. In that case, we would be reusing the vsie page with a wrong page->index. Another corner case is that we might set the SCB address for a vsie page, but fail the insertion into the radix tree. Whoever would reuse that page would remove the corresponding radix tree entry -- which might now be a valid entry pointing at another page, resulting in the wrong vsie page getting removed from the radix tree. Let's handle such races better, by validating that the SCB address of a vsie page didn't change after we grabbed it (not reuse for a different SCB; the alternative would be performing another tree lookup), and by setting the SCB address to invalid until the insertion in the tree succeeded (SCB addresses are aligned to 512, so ULONG_MAX is invalid). These scenarios are rare, the effects a bit unclear, and these issues were only found by code inspection. Let's CC stable to be safe. Fixes: a3508fbe9dc6 ("KVM: s390: vsie: initial support for nested virtualization") Cc: stable@vger.kernel.org Signed-off-by: David Hildenbrand Reviewed-by: Claudio Imbrenda Reviewed-by: Christoph Schlameuss Tested-by: Christoph Schlameuss Message-ID: <20250107154344.1003072-2-david@redhat.com> Signed-off-by: Claudio Imbrenda --- arch/s390/kvm/vsie.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index a687695d8f68..513e608567cc 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -1362,8 +1362,14 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) page = radix_tree_lookup(&kvm->arch.vsie.addr_to_page, addr >> 9); rcu_read_unlock(); if (page) { - if (page_ref_inc_return(page) == 2) - return page_to_virt(page); + if (page_ref_inc_return(page) == 2) { + if (page->index == addr) + return page_to_virt(page); + /* + * We raced with someone reusing + putting this vsie + * page before we grabbed it. + */ + } page_ref_dec(page); } @@ -1393,15 +1399,20 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) kvm->arch.vsie.next++; kvm->arch.vsie.next %= nr_vcpus; } - radix_tree_delete(&kvm->arch.vsie.addr_to_page, page->index >> 9); + if (page->index != ULONG_MAX) + radix_tree_delete(&kvm->arch.vsie.addr_to_page, + page->index >> 9); } - page->index = addr; - /* double use of the same address */ + /* Mark it as invalid until it resides in the tree. */ + page->index = ULONG_MAX; + + /* Double use of the same address or allocation failure. */ if (radix_tree_insert(&kvm->arch.vsie.addr_to_page, addr >> 9, page)) { page_ref_dec(page); mutex_unlock(&kvm->arch.vsie.mutex); return NULL; } + page->index = addr; mutex_unlock(&kvm->arch.vsie.mutex); vsie_page = page_to_virt(page); @@ -1496,7 +1507,9 @@ void kvm_s390_vsie_destroy(struct kvm *kvm) vsie_page = page_to_virt(page); release_gmap_shadow(vsie_page); /* free the radix tree entry */ - radix_tree_delete(&kvm->arch.vsie.addr_to_page, page->index >> 9); + if (page->index != ULONG_MAX) + radix_tree_delete(&kvm->arch.vsie.addr_to_page, + page->index >> 9); __free_page(page); } kvm->arch.vsie.page_count = 0; From c5f64c98a1f7e4ca0e55b441620473389b8c7a72 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 7 Jan 2025 16:43:42 +0100 Subject: [PATCH 0277/1443] KVM: s390: vsie: stop using page->index Let's stop using page->index, and instead use a field inside "struct vsie_page" to hold that value. We have plenty of space left in there. This is one part of stopping using "struct page" when working with vsie pages. We place the "page_to_virt(page)" strategically, so the next cleanups requires less churn. Signed-off-by: David Hildenbrand Reviewed-by: Claudio Imbrenda Reviewed-by: Christoph Schlameuss Tested-by: Christoph Schlameuss Message-ID: <20250107154344.1003072-3-david@redhat.com> Signed-off-by: Claudio Imbrenda --- arch/s390/kvm/vsie.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 513e608567cc..3874a1b49dd5 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -46,7 +46,13 @@ struct vsie_page { gpa_t gvrd_gpa; /* 0x0240 */ gpa_t riccbd_gpa; /* 0x0248 */ gpa_t sdnx_gpa; /* 0x0250 */ - __u8 reserved[0x0700 - 0x0258]; /* 0x0258 */ + /* + * guest address of the original SCB. Remains set for free vsie + * pages, so we can properly look them up in our addr_to_page + * radix tree. + */ + gpa_t scb_gpa; /* 0x0258 */ + __u8 reserved[0x0700 - 0x0260]; /* 0x0260 */ struct kvm_s390_crypto_cb crycb; /* 0x0700 */ __u8 fac[S390_ARCH_FAC_LIST_SIZE_BYTE]; /* 0x0800 */ }; @@ -1362,9 +1368,10 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) page = radix_tree_lookup(&kvm->arch.vsie.addr_to_page, addr >> 9); rcu_read_unlock(); if (page) { + vsie_page = page_to_virt(page); if (page_ref_inc_return(page) == 2) { - if (page->index == addr) - return page_to_virt(page); + if (vsie_page->scb_gpa == addr) + return vsie_page; /* * We raced with someone reusing + putting this vsie * page before we grabbed it. @@ -1386,6 +1393,7 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) mutex_unlock(&kvm->arch.vsie.mutex); return ERR_PTR(-ENOMEM); } + vsie_page = page_to_virt(page); page_ref_inc(page); kvm->arch.vsie.pages[kvm->arch.vsie.page_count] = page; kvm->arch.vsie.page_count++; @@ -1393,18 +1401,19 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) /* reuse an existing entry that belongs to nobody */ while (true) { page = kvm->arch.vsie.pages[kvm->arch.vsie.next]; + vsie_page = page_to_virt(page); if (page_ref_inc_return(page) == 2) break; page_ref_dec(page); kvm->arch.vsie.next++; kvm->arch.vsie.next %= nr_vcpus; } - if (page->index != ULONG_MAX) + if (vsie_page->scb_gpa != ULONG_MAX) radix_tree_delete(&kvm->arch.vsie.addr_to_page, - page->index >> 9); + vsie_page->scb_gpa >> 9); } /* Mark it as invalid until it resides in the tree. */ - page->index = ULONG_MAX; + vsie_page->scb_gpa = ULONG_MAX; /* Double use of the same address or allocation failure. */ if (radix_tree_insert(&kvm->arch.vsie.addr_to_page, addr >> 9, page)) { @@ -1412,10 +1421,9 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) mutex_unlock(&kvm->arch.vsie.mutex); return NULL; } - page->index = addr; + vsie_page->scb_gpa = addr; mutex_unlock(&kvm->arch.vsie.mutex); - vsie_page = page_to_virt(page); memset(&vsie_page->scb_s, 0, sizeof(struct kvm_s390_sie_block)); release_gmap_shadow(vsie_page); vsie_page->fault_addr = 0; @@ -1507,9 +1515,9 @@ void kvm_s390_vsie_destroy(struct kvm *kvm) vsie_page = page_to_virt(page); release_gmap_shadow(vsie_page); /* free the radix tree entry */ - if (page->index != ULONG_MAX) + if (vsie_page->scb_gpa != ULONG_MAX) radix_tree_delete(&kvm->arch.vsie.addr_to_page, - page->index >> 9); + vsie_page->scb_gpa >> 9); __free_page(page); } kvm->arch.vsie.page_count = 0; From 905f5ce0835c938c501237d9371cdbc91d8f7e02 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 7 Jan 2025 16:43:43 +0100 Subject: [PATCH 0278/1443] KVM: s390: vsie: stop messing with page refcount Let's stop messing with the page refcount, and use a flag that is set / cleared atomically to remember whether a vsie page is currently in use. Note that we could use a page flag, or a lower bit of the scb_gpa. Let's keep it simple for now, we have sufficient space. While at it, stop passing "struct kvm *" to put_vsie_page(), it's unused. Signed-off-by: David Hildenbrand Reviewed-by: Claudio Imbrenda Reviewed-by: Christoph Schlameuss Tested-by: Christoph Schlameuss Message-ID: <20250107154344.1003072-4-david@redhat.com> Signed-off-by: Claudio Imbrenda --- arch/s390/kvm/vsie.c | 46 +++++++++++++++++++++++++++++--------------- 1 file changed, 30 insertions(+), 16 deletions(-) diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 3874a1b49dd5..424f80f5f6b2 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -23,6 +23,10 @@ #include "kvm-s390.h" #include "gaccess.h" +enum vsie_page_flags { + VSIE_PAGE_IN_USE = 0, +}; + struct vsie_page { struct kvm_s390_sie_block scb_s; /* 0x0000 */ /* @@ -52,7 +56,12 @@ struct vsie_page { * radix tree. */ gpa_t scb_gpa; /* 0x0258 */ - __u8 reserved[0x0700 - 0x0260]; /* 0x0260 */ + /* + * Flags: must be set/cleared atomically after the vsie page can be + * looked up by other CPUs. + */ + unsigned long flags; /* 0x0260 */ + __u8 reserved[0x0700 - 0x0268]; /* 0x0268 */ struct kvm_s390_crypto_cb crycb; /* 0x0700 */ __u8 fac[S390_ARCH_FAC_LIST_SIZE_BYTE]; /* 0x0800 */ }; @@ -1351,6 +1360,20 @@ static int vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) return rc; } +/* Try getting a given vsie page, returning "true" on success. */ +static inline bool try_get_vsie_page(struct vsie_page *vsie_page) +{ + if (test_bit(VSIE_PAGE_IN_USE, &vsie_page->flags)) + return false; + return !test_and_set_bit(VSIE_PAGE_IN_USE, &vsie_page->flags); +} + +/* Put a vsie page acquired through get_vsie_page / try_get_vsie_page. */ +static void put_vsie_page(struct vsie_page *vsie_page) +{ + clear_bit(VSIE_PAGE_IN_USE, &vsie_page->flags); +} + /* * Get or create a vsie page for a scb address. * @@ -1369,15 +1392,15 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) rcu_read_unlock(); if (page) { vsie_page = page_to_virt(page); - if (page_ref_inc_return(page) == 2) { + if (try_get_vsie_page(vsie_page)) { if (vsie_page->scb_gpa == addr) return vsie_page; /* * We raced with someone reusing + putting this vsie * page before we grabbed it. */ + put_vsie_page(vsie_page); } - page_ref_dec(page); } /* @@ -1394,7 +1417,7 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) return ERR_PTR(-ENOMEM); } vsie_page = page_to_virt(page); - page_ref_inc(page); + __set_bit(VSIE_PAGE_IN_USE, &vsie_page->flags); kvm->arch.vsie.pages[kvm->arch.vsie.page_count] = page; kvm->arch.vsie.page_count++; } else { @@ -1402,9 +1425,8 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) while (true) { page = kvm->arch.vsie.pages[kvm->arch.vsie.next]; vsie_page = page_to_virt(page); - if (page_ref_inc_return(page) == 2) + if (try_get_vsie_page(vsie_page)) break; - page_ref_dec(page); kvm->arch.vsie.next++; kvm->arch.vsie.next %= nr_vcpus; } @@ -1417,7 +1439,7 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) /* Double use of the same address or allocation failure. */ if (radix_tree_insert(&kvm->arch.vsie.addr_to_page, addr >> 9, page)) { - page_ref_dec(page); + put_vsie_page(vsie_page); mutex_unlock(&kvm->arch.vsie.mutex); return NULL; } @@ -1431,14 +1453,6 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) return vsie_page; } -/* put a vsie page acquired via get_vsie_page */ -static void put_vsie_page(struct kvm *kvm, struct vsie_page *vsie_page) -{ - struct page *page = pfn_to_page(__pa(vsie_page) >> PAGE_SHIFT); - - page_ref_dec(page); -} - int kvm_s390_handle_vsie(struct kvm_vcpu *vcpu) { struct vsie_page *vsie_page; @@ -1489,7 +1503,7 @@ int kvm_s390_handle_vsie(struct kvm_vcpu *vcpu) out_unpin_scb: unpin_scb(vcpu, vsie_page, scb_addr); out_put: - put_vsie_page(vcpu->kvm, vsie_page); + put_vsie_page(vsie_page); return rc < 0 ? rc : 0; } From 4514eda4c1dbd0b7062e06c769d2ceafd25c9284 Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Tue, 7 Jan 2025 16:43:44 +0100 Subject: [PATCH 0279/1443] KVM: s390: vsie: stop using "struct page" for vsie page Now that we no longer use page->index and the page refcount explicitly, let's avoid messing with "struct page" completely. Signed-off-by: David Hildenbrand Reviewed-by: Claudio Imbrenda Reviewed-by: Christoph Schlameuss Tested-by: Christoph Schlameuss Message-ID: <20250107154344.1003072-5-david@redhat.com> Signed-off-by: Claudio Imbrenda --- arch/s390/include/asm/kvm_host.h | 4 +++- arch/s390/kvm/vsie.c | 31 ++++++++++++------------------- 2 files changed, 15 insertions(+), 20 deletions(-) diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 97c7c8127543..4581388411b7 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -931,12 +931,14 @@ struct sie_page2 { u8 reserved928[0x1000 - 0x928]; /* 0x0928 */ }; +struct vsie_page; + struct kvm_s390_vsie { struct mutex mutex; struct radix_tree_root addr_to_page; int page_count; int next; - struct page *pages[KVM_MAX_VCPUS]; + struct vsie_page *pages[KVM_MAX_VCPUS]; }; struct kvm_s390_gisa_iam { diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index 424f80f5f6b2..a0398ff85d00 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -599,7 +599,6 @@ void kvm_s390_vsie_gmap_notifier(struct gmap *gmap, unsigned long start, struct kvm *kvm = gmap->private; struct vsie_page *cur; unsigned long prefix; - struct page *page; int i; if (!gmap_is_shadow(gmap)) @@ -609,10 +608,9 @@ void kvm_s390_vsie_gmap_notifier(struct gmap *gmap, unsigned long start, * therefore we can safely reference them all the time. */ for (i = 0; i < kvm->arch.vsie.page_count; i++) { - page = READ_ONCE(kvm->arch.vsie.pages[i]); - if (!page) + cur = READ_ONCE(kvm->arch.vsie.pages[i]); + if (!cur) continue; - cur = page_to_virt(page); if (READ_ONCE(cur->gmap) != gmap) continue; prefix = cur->scb_s.prefix << GUEST_PREFIX_SHIFT; @@ -1384,14 +1382,12 @@ static void put_vsie_page(struct vsie_page *vsie_page) static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) { struct vsie_page *vsie_page; - struct page *page; int nr_vcpus; rcu_read_lock(); - page = radix_tree_lookup(&kvm->arch.vsie.addr_to_page, addr >> 9); + vsie_page = radix_tree_lookup(&kvm->arch.vsie.addr_to_page, addr >> 9); rcu_read_unlock(); - if (page) { - vsie_page = page_to_virt(page); + if (vsie_page) { if (try_get_vsie_page(vsie_page)) { if (vsie_page->scb_gpa == addr) return vsie_page; @@ -1411,20 +1407,18 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) mutex_lock(&kvm->arch.vsie.mutex); if (kvm->arch.vsie.page_count < nr_vcpus) { - page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO | GFP_DMA); - if (!page) { + vsie_page = (void *)__get_free_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO | GFP_DMA); + if (!vsie_page) { mutex_unlock(&kvm->arch.vsie.mutex); return ERR_PTR(-ENOMEM); } - vsie_page = page_to_virt(page); __set_bit(VSIE_PAGE_IN_USE, &vsie_page->flags); - kvm->arch.vsie.pages[kvm->arch.vsie.page_count] = page; + kvm->arch.vsie.pages[kvm->arch.vsie.page_count] = vsie_page; kvm->arch.vsie.page_count++; } else { /* reuse an existing entry that belongs to nobody */ while (true) { - page = kvm->arch.vsie.pages[kvm->arch.vsie.next]; - vsie_page = page_to_virt(page); + vsie_page = kvm->arch.vsie.pages[kvm->arch.vsie.next]; if (try_get_vsie_page(vsie_page)) break; kvm->arch.vsie.next++; @@ -1438,7 +1432,8 @@ static struct vsie_page *get_vsie_page(struct kvm *kvm, unsigned long addr) vsie_page->scb_gpa = ULONG_MAX; /* Double use of the same address or allocation failure. */ - if (radix_tree_insert(&kvm->arch.vsie.addr_to_page, addr >> 9, page)) { + if (radix_tree_insert(&kvm->arch.vsie.addr_to_page, addr >> 9, + vsie_page)) { put_vsie_page(vsie_page); mutex_unlock(&kvm->arch.vsie.mutex); return NULL; @@ -1519,20 +1514,18 @@ void kvm_s390_vsie_init(struct kvm *kvm) void kvm_s390_vsie_destroy(struct kvm *kvm) { struct vsie_page *vsie_page; - struct page *page; int i; mutex_lock(&kvm->arch.vsie.mutex); for (i = 0; i < kvm->arch.vsie.page_count; i++) { - page = kvm->arch.vsie.pages[i]; + vsie_page = kvm->arch.vsie.pages[i]; kvm->arch.vsie.pages[i] = NULL; - vsie_page = page_to_virt(page); release_gmap_shadow(vsie_page); /* free the radix tree entry */ if (vsie_page->scb_gpa != ULONG_MAX) radix_tree_delete(&kvm->arch.vsie.addr_to_page, vsie_page->scb_gpa >> 9); - __free_page(page); + free_page((unsigned long)vsie_page); } kvm->arch.vsie.page_count = 0; mutex_unlock(&kvm->arch.vsie.mutex); From 66119f8ce135de664cb2fb88d9aaa322d7451a1f Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Thu, 23 Jan 2025 15:46:13 +0100 Subject: [PATCH 0280/1443] KVM: Do not restrict the size of KVM-internal memory regions Exempt KVM-internal memslots from the KVM_MEM_MAX_NR_PAGES restriction, as the limit on the number of pages exists purely to play nice with dirty bitmap operations, which use 32-bit values to index the bitmaps, and dirty logging isn't supported for KVM-internal memslots. Link: https://lore.kernel.org/all/20240802205003.353672-6-seanjc@google.com Signed-off-by: Sean Christopherson Reviewed-by: Christoph Schlameuss Reviewed-by: David Hildenbrand Link: https://lore.kernel.org/r/20250123144627.312456-2-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-2-imbrenda@linux.ibm.com> --- virt/kvm/kvm_main.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index faf10671eed2..3f04cd5e3a8c 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1971,7 +1971,15 @@ static int kvm_set_memory_region(struct kvm *kvm, return -EINVAL; if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr) return -EINVAL; - if ((mem->memory_size >> PAGE_SHIFT) > KVM_MEM_MAX_NR_PAGES) + + /* + * The size of userspace-defined memory regions is restricted in order + * to play nice with dirty bitmap operations, which are indexed with an + * "unsigned int". KVM's internal memory regions don't support dirty + * logging, and so are exempt. + */ + if (id < KVM_USER_MEM_SLOTS && + (mem->memory_size >> PAGE_SHIFT) > KVM_MEM_MAX_NR_PAGES) return -EINVAL; slots = __kvm_memslots(kvm, as_id); From decff09adbeba4b75a1982b1dc3991761914e2df Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:14 +0100 Subject: [PATCH 0281/1443] KVM: s390: wrapper for KVM_BUG Wrap the call to KVM_BUG; this reduces code duplication and improves readability. Reviewed-by: Christian Borntraeger Reviewed-by: Christoph Schlameuss Reviewed-by: Steffen Eiden Reviewed-by: David Hildenbrand Link: https://lore.kernel.org/r/20250123144627.312456-3-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-3-imbrenda@linux.ibm.com> --- arch/s390/kvm/kvm-s390.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index d8080c27d45b..ecbdd7d41230 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -4766,6 +4766,13 @@ static int vcpu_post_run_addressing_exception(struct kvm_vcpu *vcpu) return kvm_s390_inject_prog_irq(vcpu, &pgm_info); } +static void kvm_s390_assert_primary_as(struct kvm_vcpu *vcpu) +{ + KVM_BUG(current->thread.gmap_teid.as != PSW_BITS_AS_PRIMARY, vcpu->kvm, + "Unexpected program interrupt 0x%x, TEID 0x%016lx", + current->thread.gmap_int_code, current->thread.gmap_teid.val); +} + static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) { unsigned int flags = 0; @@ -4781,9 +4788,7 @@ static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) vcpu->stat.exit_null++; break; case PGM_NON_SECURE_STORAGE_ACCESS: - KVM_BUG(current->thread.gmap_teid.as != PSW_BITS_AS_PRIMARY, vcpu->kvm, - "Unexpected program interrupt 0x%x, TEID 0x%016lx", - current->thread.gmap_int_code, current->thread.gmap_teid.val); + kvm_s390_assert_primary_as(vcpu); /* * This is normal operation; a page belonging to a protected * guest has not been imported yet. Try to import the page into @@ -4794,9 +4799,7 @@ static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) break; case PGM_SECURE_STORAGE_ACCESS: case PGM_SECURE_STORAGE_VIOLATION: - KVM_BUG(current->thread.gmap_teid.as != PSW_BITS_AS_PRIMARY, vcpu->kvm, - "Unexpected program interrupt 0x%x, TEID 0x%016lx", - current->thread.gmap_int_code, current->thread.gmap_teid.val); + kvm_s390_assert_primary_as(vcpu); /* * This can happen after a reboot with asynchronous teardown; * the new guest (normal or protected) will run on top of the @@ -4825,9 +4828,7 @@ static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) case PGM_REGION_FIRST_TRANS: case PGM_REGION_SECOND_TRANS: case PGM_REGION_THIRD_TRANS: - KVM_BUG(current->thread.gmap_teid.as != PSW_BITS_AS_PRIMARY, vcpu->kvm, - "Unexpected program interrupt 0x%x, TEID 0x%016lx", - current->thread.gmap_int_code, current->thread.gmap_teid.val); + kvm_s390_assert_primary_as(vcpu); if (vcpu->arch.gmap->pfault_enabled) { rc = gmap_fault(vcpu->arch.gmap, gaddr, flags | FAULT_FLAG_RETRY_NOWAIT); if (rc == -EFAULT) From 413c98f24c63b3b8aff202fce6f01e8950730511 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:15 +0100 Subject: [PATCH 0282/1443] KVM: s390: fake memslot for ucontrol VMs Create a fake memslot for ucontrol VMs. The fake memslot identity-maps userspace. Now memslots will always be present, and ucontrol is not a special case anymore. Suggested-by: Sean Christopherson Reviewed-by: David Hildenbrand Reviewed-by: Janosch Frank Link: https://lore.kernel.org/r/20250123144627.312456-4-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-4-imbrenda@linux.ibm.com> --- Documentation/virt/kvm/api.rst | 2 +- arch/s390/include/asm/kvm_host.h | 2 ++ arch/s390/kvm/kvm-s390.c | 17 ++++++++++++++++- arch/s390/kvm/kvm-s390.h | 2 ++ 4 files changed, 21 insertions(+), 2 deletions(-) diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst index 0d1c3a820ce6..2b52eb77e29c 100644 --- a/Documentation/virt/kvm/api.rst +++ b/Documentation/virt/kvm/api.rst @@ -1419,7 +1419,7 @@ fetch) is injected in the guest. S390: ^^^^^ -Returns -EINVAL if the VM has the KVM_VM_S390_UCONTROL flag set. +Returns -EINVAL or -EEXIST if the VM has the KVM_VM_S390_UCONTROL flag set. Returns -EINVAL if called on a protected VM. 4.36 KVM_SET_TSS_ADDR diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 4581388411b7..9a367866cab0 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -30,6 +30,8 @@ #define KVM_S390_ESCA_CPU_SLOTS 248 #define KVM_MAX_VCPUS 255 +#define KVM_INTERNAL_MEM_SLOTS 1 + /* * These seem to be used for allocating ->chip in the routing table, which we * don't use. 1 is as small as we can get to reduce the needed memory. If we diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index ecbdd7d41230..fc44002a7b04 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -3428,8 +3428,20 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) VM_EVENT(kvm, 3, "vm created with type %lu", type); if (type & KVM_VM_S390_UCONTROL) { + struct kvm_userspace_memory_region2 fake_memslot = { + .slot = KVM_S390_UCONTROL_MEMSLOT, + .guest_phys_addr = 0, + .userspace_addr = 0, + .memory_size = ALIGN_DOWN(TASK_SIZE, _SEGMENT_SIZE), + .flags = 0, + }; + kvm->arch.gmap = NULL; kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT; + /* one flat fake memslot covering the whole address-space */ + mutex_lock(&kvm->slots_lock); + KVM_BUG_ON(kvm_set_internal_memslot(kvm, &fake_memslot), kvm); + mutex_unlock(&kvm->slots_lock); } else { if (sclp.hamax == U64_MAX) kvm->arch.mem_limit = TASK_SIZE_MAX; @@ -5854,7 +5866,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm, { gpa_t size; - if (kvm_is_ucontrol(kvm)) + if (kvm_is_ucontrol(kvm) && new->id < KVM_USER_MEM_SLOTS) return -EINVAL; /* When we are protected, we should not change the memory slots */ @@ -5906,6 +5918,9 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, { int rc = 0; + if (kvm_is_ucontrol(kvm)) + return; + switch (change) { case KVM_MR_DELETE: rc = gmap_unmap_segment(kvm->arch.gmap, old->base_gfn * PAGE_SIZE, diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 597d7a71deeb..30736ac16f84 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -20,6 +20,8 @@ #include #include +#define KVM_S390_UCONTROL_MEMSLOT (KVM_USER_MEM_SLOTS + 0) + static inline void kvm_s390_fpu_store(struct kvm_run *run) { fpu_stfpc(&run->s.regs.fpc); From 63e71519891024b622d00c486c4d0348c44ca911 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:16 +0100 Subject: [PATCH 0283/1443] KVM: s390: selftests: fix ucontrol memory region test With the latest patch, attempting to create a memslot from userspace will result in an EEXIST error for UCONTROL VMs, instead of EINVAL, since the new memslot will collide with the internal memslot. There is no simple way to bring back the previous behaviour. This is not a problem, but the test needs to be fixed accordingly. Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-5-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-5-imbrenda@linux.ibm.com> --- tools/testing/selftests/kvm/s390/ucontrol_test.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/kvm/s390/ucontrol_test.c b/tools/testing/selftests/kvm/s390/ucontrol_test.c index 135ee22856cf..22ce9219620c 100644 --- a/tools/testing/selftests/kvm/s390/ucontrol_test.c +++ b/tools/testing/selftests/kvm/s390/ucontrol_test.c @@ -459,10 +459,14 @@ TEST_F(uc_kvm, uc_no_user_region) }; ASSERT_EQ(-1, ioctl(self->vm_fd, KVM_SET_USER_MEMORY_REGION, ®ion)); - ASSERT_EQ(EINVAL, errno); + ASSERT_TRUE(errno == EEXIST || errno == EINVAL) + TH_LOG("errno %s (%i) not expected for ioctl KVM_SET_USER_MEMORY_REGION", + strerror(errno), errno); ASSERT_EQ(-1, ioctl(self->vm_fd, KVM_SET_USER_MEMORY_REGION2, ®ion2)); - ASSERT_EQ(EINVAL, errno); + ASSERT_TRUE(errno == EEXIST || errno == EINVAL) + TH_LOG("errno %s (%i) not expected for ioctl KVM_SET_USER_MEMORY_REGION2", + strerror(errno), errno); } TEST_F(uc_kvm, uc_map_unmap) From 5cbe24350b7d8ef6d466a37d56b07ae643c622ca Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:17 +0100 Subject: [PATCH 0284/1443] KVM: s390: move pv gmap functions into kvm Move gmap related functions from kernel/uv into kvm. Create a new file to collect gmap-related functions. Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss [fixed unpack_one(), thanks mhartmay@linux.ibm.com] Link: https://lore.kernel.org/r/20250123144627.312456-6-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-6-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 1 + arch/s390/include/asm/uv.h | 6 +- arch/s390/kernel/uv.c | 292 ++++------------------------------- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/gmap.c | 212 +++++++++++++++++++++++++ arch/s390/kvm/gmap.h | 17 ++ arch/s390/kvm/intercept.c | 3 +- arch/s390/kvm/kvm-s390.c | 1 + arch/s390/kvm/pv.c | 21 +++ arch/s390/mm/gmap.c | 28 ++++ 10 files changed, 315 insertions(+), 268 deletions(-) create mode 100644 arch/s390/kvm/gmap.c create mode 100644 arch/s390/kvm/gmap.h diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index 13f51a6a5bb1..3e66f53fe3cc 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -149,6 +149,7 @@ int s390_replace_asce(struct gmap *gmap); void s390_uv_destroy_pfns(unsigned long count, unsigned long *pfns); int __s390_uv_destroy_range(struct mm_struct *mm, unsigned long start, unsigned long end, bool interruptible); +int kvm_s390_wiggle_split_folio(struct mm_struct *mm, struct folio *folio, bool split); /** * s390_uv_destroy_range - Destroy a range of pages in the given mm. diff --git a/arch/s390/include/asm/uv.h b/arch/s390/include/asm/uv.h index dc332609f2c3..b11f5b6d0bd1 100644 --- a/arch/s390/include/asm/uv.h +++ b/arch/s390/include/asm/uv.h @@ -628,12 +628,12 @@ static inline int is_prot_virt_host(void) } int uv_pin_shared(unsigned long paddr); -int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb); -int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr); int uv_destroy_folio(struct folio *folio); int uv_destroy_pte(pte_t pte); int uv_convert_from_secure_pte(pte_t pte); -int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr); +int make_folio_secure(struct folio *folio, struct uv_cb_header *uvcb); +int uv_convert_from_secure(unsigned long paddr); +int uv_convert_from_secure_folio(struct folio *folio); void setup_uv(void); diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index 6f9654a191ad..9f05df2da2f7 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -19,19 +19,6 @@ #include #include -#if !IS_ENABLED(CONFIG_KVM) -unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr) -{ - return 0; -} - -int gmap_fault(struct gmap *gmap, unsigned long gaddr, - unsigned int fault_flags) -{ - return 0; -} -#endif - /* the bootdata_preserved fields come from ones in arch/s390/boot/uv.c */ int __bootdata_preserved(prot_virt_guest); EXPORT_SYMBOL(prot_virt_guest); @@ -159,6 +146,7 @@ int uv_destroy_folio(struct folio *folio) folio_put(folio); return rc; } +EXPORT_SYMBOL(uv_destroy_folio); /* * The present PTE still indirectly holds a folio reference through the mapping. @@ -175,7 +163,7 @@ int uv_destroy_pte(pte_t pte) * * @paddr: Absolute host address of page to be exported */ -static int uv_convert_from_secure(unsigned long paddr) +int uv_convert_from_secure(unsigned long paddr) { struct uv_cb_cfs uvcb = { .header.cmd = UVC_CMD_CONV_FROM_SEC_STOR, @@ -187,11 +175,12 @@ static int uv_convert_from_secure(unsigned long paddr) return -EINVAL; return 0; } +EXPORT_SYMBOL_GPL(uv_convert_from_secure); /* * The caller must already hold a reference to the folio. */ -static int uv_convert_from_secure_folio(struct folio *folio) +int uv_convert_from_secure_folio(struct folio *folio) { int rc; @@ -206,6 +195,7 @@ static int uv_convert_from_secure_folio(struct folio *folio) folio_put(folio); return rc; } +EXPORT_SYMBOL_GPL(uv_convert_from_secure_folio); /* * The present PTE still indirectly holds a folio reference through the mapping. @@ -237,13 +227,33 @@ static int expected_folio_refs(struct folio *folio) return res; } -static int make_folio_secure(struct folio *folio, struct uv_cb_header *uvcb) +/** + * make_folio_secure() - make a folio secure + * @folio: the folio to make secure + * @uvcb: the uvcb that describes the UVC to be used + * + * The folio @folio will be made secure if possible, @uvcb will be passed + * as-is to the UVC. + * + * Return: 0 on success; + * -EBUSY if the folio is in writeback or has too many references; + * -E2BIG if the folio is large; + * -EAGAIN if the UVC needs to be attempted again; + * -ENXIO if the address is not mapped; + * -EINVAL if the UVC failed for other reasons. + * + * Context: The caller must hold exactly one extra reference on the folio + * (it's the same logic as split_folio()) + */ +int make_folio_secure(struct folio *folio, struct uv_cb_header *uvcb) { int expected, cc = 0; + if (folio_test_large(folio)) + return -E2BIG; if (folio_test_writeback(folio)) - return -EAGAIN; - expected = expected_folio_refs(folio); + return -EBUSY; + expected = expected_folio_refs(folio) + 1; if (!folio_ref_freeze(folio, expected)) return -EBUSY; set_bit(PG_arch_1, &folio->flags); @@ -267,251 +277,7 @@ static int make_folio_secure(struct folio *folio, struct uv_cb_header *uvcb) return -EAGAIN; return uvcb->rc == 0x10a ? -ENXIO : -EINVAL; } - -/** - * should_export_before_import - Determine whether an export is needed - * before an import-like operation - * @uvcb: the Ultravisor control block of the UVC to be performed - * @mm: the mm of the process - * - * Returns whether an export is needed before every import-like operation. - * This is needed for shared pages, which don't trigger a secure storage - * exception when accessed from a different guest. - * - * Although considered as one, the Unpin Page UVC is not an actual import, - * so it is not affected. - * - * No export is needed also when there is only one protected VM, because the - * page cannot belong to the wrong VM in that case (there is no "other VM" - * it can belong to). - * - * Return: true if an export is needed before every import, otherwise false. - */ -static bool should_export_before_import(struct uv_cb_header *uvcb, struct mm_struct *mm) -{ - /* - * The misc feature indicates, among other things, that importing a - * shared page from a different protected VM will automatically also - * transfer its ownership. - */ - if (uv_has_feature(BIT_UV_FEAT_MISC)) - return false; - if (uvcb->cmd == UVC_CMD_UNPIN_PAGE_SHARED) - return false; - return atomic_read(&mm->context.protected_count) > 1; -} - -/* - * Drain LRU caches: the local one on first invocation and the ones of all - * CPUs on successive invocations. Returns "true" on the first invocation. - */ -static bool drain_lru(bool *drain_lru_called) -{ - /* - * If we have tried a local drain and the folio refcount - * still does not match our expected safe value, try with a - * system wide drain. This is needed if the pagevecs holding - * the page are on a different CPU. - */ - if (*drain_lru_called) { - lru_add_drain_all(); - /* We give up here, don't retry immediately. */ - return false; - } - /* - * We are here if the folio refcount does not match the - * expected safe value. The main culprits are usually - * pagevecs. With lru_add_drain() we drain the pagevecs - * on the local CPU so that hopefully the refcount will - * reach the expected safe value. - */ - lru_add_drain(); - *drain_lru_called = true; - /* The caller should try again immediately */ - return true; -} - -/* - * Requests the Ultravisor to make a page accessible to a guest. - * If it's brought in the first time, it will be cleared. If - * it has been exported before, it will be decrypted and integrity - * checked. - */ -int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb) -{ - struct vm_area_struct *vma; - bool drain_lru_called = false; - spinlock_t *ptelock; - unsigned long uaddr; - struct folio *folio; - pte_t *ptep; - int rc; - -again: - rc = -EFAULT; - mmap_read_lock(gmap->mm); - - uaddr = __gmap_translate(gmap, gaddr); - if (IS_ERR_VALUE(uaddr)) - goto out; - vma = vma_lookup(gmap->mm, uaddr); - if (!vma) - goto out; - /* - * Secure pages cannot be huge and userspace should not combine both. - * In case userspace does it anyway this will result in an -EFAULT for - * the unpack. The guest is thus never reaching secure mode. If - * userspace is playing dirty tricky with mapping huge pages later - * on this will result in a segmentation fault. - */ - if (is_vm_hugetlb_page(vma)) - goto out; - - rc = -ENXIO; - ptep = get_locked_pte(gmap->mm, uaddr, &ptelock); - if (!ptep) - goto out; - if (pte_present(*ptep) && !(pte_val(*ptep) & _PAGE_INVALID) && pte_write(*ptep)) { - folio = page_folio(pte_page(*ptep)); - rc = -EAGAIN; - if (folio_test_large(folio)) { - rc = -E2BIG; - } else if (folio_trylock(folio)) { - if (should_export_before_import(uvcb, gmap->mm)) - uv_convert_from_secure(PFN_PHYS(folio_pfn(folio))); - rc = make_folio_secure(folio, uvcb); - folio_unlock(folio); - } - - /* - * Once we drop the PTL, the folio may get unmapped and - * freed immediately. We need a temporary reference. - */ - if (rc == -EAGAIN || rc == -E2BIG) - folio_get(folio); - } - pte_unmap_unlock(ptep, ptelock); -out: - mmap_read_unlock(gmap->mm); - - switch (rc) { - case -E2BIG: - folio_lock(folio); - rc = split_folio(folio); - folio_unlock(folio); - folio_put(folio); - - switch (rc) { - case 0: - /* Splitting succeeded, try again immediately. */ - goto again; - case -EAGAIN: - /* Additional folio references. */ - if (drain_lru(&drain_lru_called)) - goto again; - return -EAGAIN; - case -EBUSY: - /* Unexpected race. */ - return -EAGAIN; - } - WARN_ON_ONCE(1); - return -ENXIO; - case -EAGAIN: - /* - * If we are here because the UVC returned busy or partial - * completion, this is just a useless check, but it is safe. - */ - folio_wait_writeback(folio); - folio_put(folio); - return -EAGAIN; - case -EBUSY: - /* Additional folio references. */ - if (drain_lru(&drain_lru_called)) - goto again; - return -EAGAIN; - case -ENXIO: - if (gmap_fault(gmap, gaddr, FAULT_FLAG_WRITE)) - return -EFAULT; - return -EAGAIN; - } - return rc; -} -EXPORT_SYMBOL_GPL(gmap_make_secure); - -int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr) -{ - struct uv_cb_cts uvcb = { - .header.cmd = UVC_CMD_CONV_TO_SEC_STOR, - .header.len = sizeof(uvcb), - .guest_handle = gmap->guest_handle, - .gaddr = gaddr, - }; - - return gmap_make_secure(gmap, gaddr, &uvcb); -} -EXPORT_SYMBOL_GPL(gmap_convert_to_secure); - -/** - * gmap_destroy_page - Destroy a guest page. - * @gmap: the gmap of the guest - * @gaddr: the guest address to destroy - * - * An attempt will be made to destroy the given guest page. If the attempt - * fails, an attempt is made to export the page. If both attempts fail, an - * appropriate error is returned. - */ -int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr) -{ - struct vm_area_struct *vma; - struct folio_walk fw; - unsigned long uaddr; - struct folio *folio; - int rc; - - rc = -EFAULT; - mmap_read_lock(gmap->mm); - - uaddr = __gmap_translate(gmap, gaddr); - if (IS_ERR_VALUE(uaddr)) - goto out; - vma = vma_lookup(gmap->mm, uaddr); - if (!vma) - goto out; - /* - * Huge pages should not be able to become secure - */ - if (is_vm_hugetlb_page(vma)) - goto out; - - rc = 0; - folio = folio_walk_start(&fw, vma, uaddr, 0); - if (!folio) - goto out; - /* - * See gmap_make_secure(): large folios cannot be secure. Small - * folio implies FW_LEVEL_PTE. - */ - if (folio_test_large(folio) || !pte_write(fw.pte)) - goto out_walk_end; - rc = uv_destroy_folio(folio); - /* - * Fault handlers can race; it is possible that two CPUs will fault - * on the same secure page. One CPU can destroy the page, reboot, - * re-enter secure mode and import it, while the second CPU was - * stuck at the beginning of the handler. At some point the second - * CPU will be able to progress, and it will not be able to destroy - * the page. In that case we do not want to terminate the process, - * we instead try to export the page. - */ - if (rc) - rc = uv_convert_from_secure_folio(folio); -out_walk_end: - folio_walk_end(&fw, vma); -out: - mmap_read_unlock(gmap->mm); - return rc; -} -EXPORT_SYMBOL_GPL(gmap_destroy_page); +EXPORT_SYMBOL_GPL(make_folio_secure); /* * To be called with the folio locked or with an extra reference! This will diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index 02217fb4ae10..d972dea657fd 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -8,7 +8,7 @@ include $(srctree)/virt/kvm/Makefile.kvm ccflags-y := -Ivirt/kvm -Iarch/s390/kvm kvm-y += kvm-s390.o intercept.o interrupt.o priv.o sigp.o -kvm-y += diag.o gaccess.o guestdbg.o vsie.o pv.o +kvm-y += diag.o gaccess.o guestdbg.o vsie.o pv.o gmap.o kvm-$(CONFIG_VFIO_PCI_ZDEV_KVM) += pci.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/gmap.c b/arch/s390/kvm/gmap.c new file mode 100644 index 000000000000..02adf151d4de --- /dev/null +++ b/arch/s390/kvm/gmap.c @@ -0,0 +1,212 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Guest memory management for KVM/s390 + * + * Copyright IBM Corp. 2008, 2020, 2024 + * + * Author(s): Claudio Imbrenda + * Martin Schwidefsky + * David Hildenbrand + * Janosch Frank + */ + +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "gmap.h" + +/** + * should_export_before_import - Determine whether an export is needed + * before an import-like operation + * @uvcb: the Ultravisor control block of the UVC to be performed + * @mm: the mm of the process + * + * Returns whether an export is needed before every import-like operation. + * This is needed for shared pages, which don't trigger a secure storage + * exception when accessed from a different guest. + * + * Although considered as one, the Unpin Page UVC is not an actual import, + * so it is not affected. + * + * No export is needed also when there is only one protected VM, because the + * page cannot belong to the wrong VM in that case (there is no "other VM" + * it can belong to). + * + * Return: true if an export is needed before every import, otherwise false. + */ +static bool should_export_before_import(struct uv_cb_header *uvcb, struct mm_struct *mm) +{ + /* + * The misc feature indicates, among other things, that importing a + * shared page from a different protected VM will automatically also + * transfer its ownership. + */ + if (uv_has_feature(BIT_UV_FEAT_MISC)) + return false; + if (uvcb->cmd == UVC_CMD_UNPIN_PAGE_SHARED) + return false; + return atomic_read(&mm->context.protected_count) > 1; +} + +static int __gmap_make_secure(struct gmap *gmap, struct page *page, void *uvcb) +{ + struct folio *folio = page_folio(page); + int rc; + + /* + * Secure pages cannot be huge and userspace should not combine both. + * In case userspace does it anyway this will result in an -EFAULT for + * the unpack. The guest is thus never reaching secure mode. + * If userspace plays dirty tricks and decides to map huge pages at a + * later point in time, it will receive a segmentation fault or + * KVM_RUN will return -EFAULT. + */ + if (folio_test_hugetlb(folio)) + return -EFAULT; + if (folio_test_large(folio)) { + mmap_read_unlock(gmap->mm); + rc = kvm_s390_wiggle_split_folio(gmap->mm, folio, true); + mmap_read_lock(gmap->mm); + if (rc) + return rc; + folio = page_folio(page); + } + + if (!folio_trylock(folio)) + return -EAGAIN; + if (should_export_before_import(uvcb, gmap->mm)) + uv_convert_from_secure(folio_to_phys(folio)); + rc = make_folio_secure(folio, uvcb); + folio_unlock(folio); + + /* + * In theory a race is possible and the folio might have become + * large again before the folio_trylock() above. In that case, no + * action is performed and -EAGAIN is returned; the callers will + * have to try again later. + * In most cases this implies running the VM again, getting the same + * exception again, and make another attempt in this function. + * This is expected to happen extremely rarely. + */ + if (rc == -E2BIG) + return -EAGAIN; + /* The folio has too many references, try to shake some off */ + if (rc == -EBUSY) { + mmap_read_unlock(gmap->mm); + kvm_s390_wiggle_split_folio(gmap->mm, folio, false); + mmap_read_lock(gmap->mm); + return -EAGAIN; + } + + return rc; +} + +/** + * gmap_make_secure() - make one guest page secure + * @gmap: the guest gmap + * @gaddr: the guest address that needs to be made secure + * @uvcb: the UVCB specifying which operation needs to be performed + * + * Context: needs to be called with kvm->srcu held. + * Return: 0 on success, < 0 in case of error (see __gmap_make_secure()). + */ +int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb) +{ + struct kvm *kvm = gmap->private; + struct page *page; + int rc = 0; + + lockdep_assert_held(&kvm->srcu); + + page = gfn_to_page(kvm, gpa_to_gfn(gaddr)); + mmap_read_lock(gmap->mm); + if (page) + rc = __gmap_make_secure(gmap, page, uvcb); + kvm_release_page_clean(page); + mmap_read_unlock(gmap->mm); + + return rc; +} + +int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr) +{ + struct uv_cb_cts uvcb = { + .header.cmd = UVC_CMD_CONV_TO_SEC_STOR, + .header.len = sizeof(uvcb), + .guest_handle = gmap->guest_handle, + .gaddr = gaddr, + }; + + return gmap_make_secure(gmap, gaddr, &uvcb); +} + +/** + * __gmap_destroy_page() - Destroy a guest page. + * @gmap: the gmap of the guest + * @page: the page to destroy + * + * An attempt will be made to destroy the given guest page. If the attempt + * fails, an attempt is made to export the page. If both attempts fail, an + * appropriate error is returned. + * + * Context: must be called holding the mm lock for gmap->mm + */ +static int __gmap_destroy_page(struct gmap *gmap, struct page *page) +{ + struct folio *folio = page_folio(page); + int rc; + + /* + * See gmap_make_secure(): large folios cannot be secure. Small + * folio implies FW_LEVEL_PTE. + */ + if (folio_test_large(folio)) + return -EFAULT; + + rc = uv_destroy_folio(folio); + /* + * Fault handlers can race; it is possible that two CPUs will fault + * on the same secure page. One CPU can destroy the page, reboot, + * re-enter secure mode and import it, while the second CPU was + * stuck at the beginning of the handler. At some point the second + * CPU will be able to progress, and it will not be able to destroy + * the page. In that case we do not want to terminate the process, + * we instead try to export the page. + */ + if (rc) + rc = uv_convert_from_secure_folio(folio); + + return rc; +} + +/** + * gmap_destroy_page() - Destroy a guest page. + * @gmap: the gmap of the guest + * @gaddr: the guest address to destroy + * + * An attempt will be made to destroy the given guest page. If the attempt + * fails, an attempt is made to export the page. If both attempts fail, an + * appropriate error is returned. + * + * Context: may sleep. + */ +int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr) +{ + struct page *page; + int rc = 0; + + mmap_read_lock(gmap->mm); + page = gfn_to_page(gmap->private, gpa_to_gfn(gaddr)); + if (page) + rc = __gmap_destroy_page(gmap, page); + kvm_release_page_clean(page); + mmap_read_unlock(gmap->mm); + return rc; +} diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h new file mode 100644 index 000000000000..f2b52ce29be3 --- /dev/null +++ b/arch/s390/kvm/gmap.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * KVM guest address space mapping code + * + * Copyright IBM Corp. 2007, 2016, 2025 + * Author(s): Martin Schwidefsky + * Claudio Imbrenda + */ + +#ifndef ARCH_KVM_S390_GMAP_H +#define ARCH_KVM_S390_GMAP_H + +int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb); +int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr); +int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr); + +#endif diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 5bbaadf75dc6..acf10aefd08f 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -21,6 +21,7 @@ #include "gaccess.h" #include "trace.h" #include "trace-s390.h" +#include "gmap.h" u8 kvm_s390_get_ilen(struct kvm_vcpu *vcpu) { @@ -549,7 +550,7 @@ static int handle_pv_uvc(struct kvm_vcpu *vcpu) * If the unpin did not succeed, the guest will exit again for the UVC * and we will retry the unpin. */ - if (rc == -EINVAL) + if (rc == -EINVAL || rc == -ENXIO) return 0; /* * If we got -EAGAIN here, we simply return it. It will eventually diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index fc44002a7b04..a25ca440760f 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -50,6 +50,7 @@ #include "kvm-s390.h" #include "gaccess.h" #include "pci.h" +#include "gmap.h" #define CREATE_TRACE_POINTS #include "trace.h" diff --git a/arch/s390/kvm/pv.c b/arch/s390/kvm/pv.c index 75e81ba26d04..22c012aa5206 100644 --- a/arch/s390/kvm/pv.c +++ b/arch/s390/kvm/pv.c @@ -17,6 +17,7 @@ #include #include #include "kvm-s390.h" +#include "gmap.h" bool kvm_s390_pv_is_protected(struct kvm *kvm) { @@ -638,10 +639,28 @@ static int unpack_one(struct kvm *kvm, unsigned long addr, u64 tweak, .tweak[1] = offset, }; int ret = gmap_make_secure(kvm->arch.gmap, addr, &uvcb); + unsigned long vmaddr; + bool unlocked; *rc = uvcb.header.rc; *rrc = uvcb.header.rrc; + if (ret == -ENXIO) { + mmap_read_lock(kvm->mm); + vmaddr = gfn_to_hva(kvm, gpa_to_gfn(addr)); + if (kvm_is_error_hva(vmaddr)) { + ret = -EFAULT; + } else { + ret = fixup_user_fault(kvm->mm, vmaddr, FAULT_FLAG_WRITE, &unlocked); + if (!ret) + ret = __gmap_link(kvm->arch.gmap, addr, vmaddr); + } + mmap_read_unlock(kvm->mm); + if (!ret) + return -EAGAIN; + return ret; + } + if (ret && ret != -EAGAIN) KVM_UV_EVENT(kvm, 3, "PROTVIRT VM UNPACK: failed addr %llx with rc %x rrc %x", uvcb.gaddr, *rc, *rrc); @@ -660,6 +679,8 @@ int kvm_s390_pv_unpack(struct kvm *kvm, unsigned long addr, unsigned long size, KVM_UV_EVENT(kvm, 3, "PROTVIRT VM UNPACK: start addr %lx size %lx", addr, size); + guard(srcu)(&kvm->srcu); + while (offset < size) { ret = unpack_one(kvm, addr, tweak, offset, rc, rrc); if (ret == -EAGAIN) { diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 16b8a36c56de..3e6e25119a96 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -3035,3 +3035,31 @@ int s390_replace_asce(struct gmap *gmap) return 0; } EXPORT_SYMBOL_GPL(s390_replace_asce); + +/** + * kvm_s390_wiggle_split_folio() - try to drain extra references to a folio and optionally split + * @mm: the mm containing the folio to work on + * @folio: the folio + * @split: whether to split a large folio + * + * Context: Must be called while holding an extra reference to the folio; + * the mm lock should not be held. + */ +int kvm_s390_wiggle_split_folio(struct mm_struct *mm, struct folio *folio, bool split) +{ + int rc; + + lockdep_assert_not_held(&mm->mmap_lock); + folio_wait_writeback(folio); + lru_add_drain_all(); + if (split) { + folio_lock(folio); + rc = split_folio(folio); + folio_unlock(folio); + + if (rc != -EBUSY) + return rc; + } + return -EAGAIN; +} +EXPORT_SYMBOL_GPL(kvm_s390_wiggle_split_folio); From 3762e905ec2e498c96464e094b7d46be98151d3b Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:18 +0100 Subject: [PATCH 0285/1443] KVM: s390: use __kvm_faultin_pfn() Refactor the existing page fault handling code to use __kvm_faultin_pfn(). This possible now that memslots are always present. Acked-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-7-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-7-imbrenda@linux.ibm.com> --- arch/s390/kvm/kvm-s390.c | 126 ++++++++++++++++++++++++++++++--------- arch/s390/kvm/kvm-s390.h | 6 ++ arch/s390/mm/gmap.c | 1 + 3 files changed, 106 insertions(+), 27 deletions(-) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index a25ca440760f..70c98bf12754 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -4786,11 +4786,104 @@ static void kvm_s390_assert_primary_as(struct kvm_vcpu *vcpu) current->thread.gmap_int_code, current->thread.gmap_teid.val); } +/* + * __kvm_s390_handle_dat_fault() - handle a dat fault for the gmap of a vcpu + * @vcpu: the vCPU whose gmap is to be fixed up + * @gfn: the guest frame number used for memslots (including fake memslots) + * @gaddr: the gmap address, does not have to match @gfn for ucontrol gmaps + * @flags: FOLL_* flags + * + * Return: 0 on success, < 0 in case of error. + * Context: The mm lock must not be held before calling. May sleep. + */ +int __kvm_s390_handle_dat_fault(struct kvm_vcpu *vcpu, gfn_t gfn, gpa_t gaddr, unsigned int flags) +{ + struct kvm_memory_slot *slot; + unsigned int fault_flags; + bool writable, unlocked; + unsigned long vmaddr; + struct page *page; + kvm_pfn_t pfn; + int rc; + + slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn); + if (!slot || slot->flags & KVM_MEMSLOT_INVALID) + return vcpu_post_run_addressing_exception(vcpu); + + fault_flags = flags & FOLL_WRITE ? FAULT_FLAG_WRITE : 0; + if (vcpu->arch.gmap->pfault_enabled) + flags |= FOLL_NOWAIT; + vmaddr = __gfn_to_hva_memslot(slot, gfn); + +try_again: + pfn = __kvm_faultin_pfn(slot, gfn, flags, &writable, &page); + + /* Access outside memory, inject addressing exception */ + if (is_noslot_pfn(pfn)) + return vcpu_post_run_addressing_exception(vcpu); + /* Signal pending: try again */ + if (pfn == KVM_PFN_ERR_SIGPENDING) + return -EAGAIN; + + /* Needs I/O, try to setup async pfault (only possible with FOLL_NOWAIT) */ + if (pfn == KVM_PFN_ERR_NEEDS_IO) { + trace_kvm_s390_major_guest_pfault(vcpu); + if (kvm_arch_setup_async_pf(vcpu)) + return 0; + vcpu->stat.pfault_sync++; + /* Could not setup async pfault, try again synchronously */ + flags &= ~FOLL_NOWAIT; + goto try_again; + } + /* Any other error */ + if (is_error_pfn(pfn)) + return -EFAULT; + + /* Success */ + mmap_read_lock(vcpu->arch.gmap->mm); + /* Mark the userspace PTEs as young and/or dirty, to avoid page fault loops */ + rc = fixup_user_fault(vcpu->arch.gmap->mm, vmaddr, fault_flags, &unlocked); + if (!rc) + rc = __gmap_link(vcpu->arch.gmap, gaddr, vmaddr); + scoped_guard(spinlock, &vcpu->kvm->mmu_lock) { + kvm_release_faultin_page(vcpu->kvm, page, false, writable); + } + mmap_read_unlock(vcpu->arch.gmap->mm); + return rc; +} + +static int vcpu_dat_fault_handler(struct kvm_vcpu *vcpu, unsigned long gaddr, unsigned int flags) +{ + unsigned long gaddr_tmp; + gfn_t gfn; + + gfn = gpa_to_gfn(gaddr); + if (kvm_is_ucontrol(vcpu->kvm)) { + /* + * This translates the per-vCPU guest address into a + * fake guest address, which can then be used with the + * fake memslots that are identity mapping userspace. + * This allows ucontrol VMs to use the normal fault + * resolution path, like normal VMs. + */ + mmap_read_lock(vcpu->arch.gmap->mm); + gaddr_tmp = __gmap_translate(vcpu->arch.gmap, gaddr); + mmap_read_unlock(vcpu->arch.gmap->mm); + if (gaddr_tmp == -EFAULT) { + vcpu->run->exit_reason = KVM_EXIT_S390_UCONTROL; + vcpu->run->s390_ucontrol.trans_exc_code = gaddr; + vcpu->run->s390_ucontrol.pgm_code = PGM_SEGMENT_TRANSLATION; + return -EREMOTE; + } + gfn = gpa_to_gfn(gaddr_tmp); + } + return __kvm_s390_handle_dat_fault(vcpu, gfn, gaddr, flags); +} + static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) { unsigned int flags = 0; unsigned long gaddr; - int rc = 0; gaddr = current->thread.gmap_teid.addr * PAGE_SIZE; if (kvm_s390_cur_gmap_fault_is_write()) @@ -4842,37 +4935,14 @@ static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) case PGM_REGION_SECOND_TRANS: case PGM_REGION_THIRD_TRANS: kvm_s390_assert_primary_as(vcpu); - if (vcpu->arch.gmap->pfault_enabled) { - rc = gmap_fault(vcpu->arch.gmap, gaddr, flags | FAULT_FLAG_RETRY_NOWAIT); - if (rc == -EFAULT) - return vcpu_post_run_addressing_exception(vcpu); - if (rc == -EAGAIN) { - trace_kvm_s390_major_guest_pfault(vcpu); - if (kvm_arch_setup_async_pf(vcpu)) - return 0; - vcpu->stat.pfault_sync++; - } else { - return rc; - } - } - rc = gmap_fault(vcpu->arch.gmap, gaddr, flags); - if (rc == -EFAULT) { - if (kvm_is_ucontrol(vcpu->kvm)) { - vcpu->run->exit_reason = KVM_EXIT_S390_UCONTROL; - vcpu->run->s390_ucontrol.trans_exc_code = gaddr; - vcpu->run->s390_ucontrol.pgm_code = 0x10; - return -EREMOTE; - } - return vcpu_post_run_addressing_exception(vcpu); - } - break; + return vcpu_dat_fault_handler(vcpu, gaddr, flags); default: KVM_BUG(1, vcpu->kvm, "Unexpected program interrupt 0x%x, TEID 0x%016lx", current->thread.gmap_int_code, current->thread.gmap_teid.val); send_sig(SIGSEGV, current, 0); break; } - return rc; + return 0; } static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) @@ -5751,7 +5821,9 @@ long kvm_arch_vcpu_ioctl(struct file *filp, } #endif case KVM_S390_VCPU_FAULT: { - r = gmap_fault(vcpu->arch.gmap, arg, 0); + idx = srcu_read_lock(&vcpu->kvm->srcu); + r = vcpu_dat_fault_handler(vcpu, arg, 0); + srcu_read_unlock(&vcpu->kvm->srcu, idx); break; } case KVM_ENABLE_CAP: diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 30736ac16f84..3be5291723c8 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -410,6 +410,12 @@ void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu); void kvm_s390_set_cpu_timer(struct kvm_vcpu *vcpu, __u64 cputm); __u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu); int kvm_s390_cpus_from_pv(struct kvm *kvm, u16 *rc, u16 *rrc); +int __kvm_s390_handle_dat_fault(struct kvm_vcpu *vcpu, gfn_t gfn, gpa_t gaddr, unsigned int flags); + +static inline int kvm_s390_handle_dat_fault(struct kvm_vcpu *vcpu, gpa_t gaddr, unsigned int flags) +{ + return __kvm_s390_handle_dat_fault(vcpu, gpa_to_gfn(gaddr), gaddr, flags); +} /* implemented in diag.c */ int kvm_s390_handle_diag(struct kvm_vcpu *vcpu); diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 3e6e25119a96..bfaba7733306 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -605,6 +605,7 @@ int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr) radix_tree_preload_end(); return rc; } +EXPORT_SYMBOL(__gmap_link); /** * fixup_user_fault_nowait - manually resolve a user page fault without waiting From 6eb84e130075b9ea35a946dcf9a2476ac2c749a0 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:19 +0100 Subject: [PATCH 0286/1443] KVM: s390: get rid of gmap_fault() All gmap page faults are already handled in kvm by the function kvm_s390_handle_dat_fault(); only few users of gmap_fault remained, all within kvm. Convert those calls to use kvm_s390_handle_dat_fault() instead. Remove gmap_fault() entirely since it has no more users. Acked-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-8-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-8-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 1 - arch/s390/kvm/intercept.c | 4 +- arch/s390/mm/gmap.c | 124 ----------------------------------- 3 files changed, 2 insertions(+), 127 deletions(-) diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index 3e66f53fe3cc..d4572729269f 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -113,7 +113,6 @@ int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len); unsigned long __gmap_translate(struct gmap *, unsigned long gaddr); unsigned long gmap_translate(struct gmap *, unsigned long gaddr); int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr); -int gmap_fault(struct gmap *, unsigned long gaddr, unsigned int fault_flags); void gmap_discard(struct gmap *, unsigned long from, unsigned long to); void __gmap_zap(struct gmap *, unsigned long gaddr); void gmap_unlink(struct mm_struct *, unsigned long *table, unsigned long vmaddr); diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index acf10aefd08f..610dd44a948b 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -368,7 +368,7 @@ static int handle_mvpg_pei(struct kvm_vcpu *vcpu) reg2, &srcaddr, GACC_FETCH, 0); if (rc) return kvm_s390_inject_prog_cond(vcpu, rc); - rc = gmap_fault(vcpu->arch.gmap, srcaddr, 0); + rc = kvm_s390_handle_dat_fault(vcpu, srcaddr, 0); if (rc != 0) return rc; @@ -377,7 +377,7 @@ static int handle_mvpg_pei(struct kvm_vcpu *vcpu) reg1, &dstaddr, GACC_STORE, 0); if (rc) return kvm_s390_inject_prog_cond(vcpu, rc); - rc = gmap_fault(vcpu->arch.gmap, dstaddr, FAULT_FLAG_WRITE); + rc = kvm_s390_handle_dat_fault(vcpu, dstaddr, FOLL_WRITE); if (rc != 0) return rc; diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index bfaba7733306..e124fca14737 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -607,130 +607,6 @@ int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr) } EXPORT_SYMBOL(__gmap_link); -/** - * fixup_user_fault_nowait - manually resolve a user page fault without waiting - * @mm: mm_struct of target mm - * @address: user address - * @fault_flags:flags to pass down to handle_mm_fault() - * @unlocked: did we unlock the mmap_lock while retrying - * - * This function behaves similarly to fixup_user_fault(), but it guarantees - * that the fault will be resolved without waiting. The function might drop - * and re-acquire the mm lock, in which case @unlocked will be set to true. - * - * The guarantee is that the fault is handled without waiting, but the - * function itself might sleep, due to the lock. - * - * Context: Needs to be called with mm->mmap_lock held in read mode, and will - * return with the lock held in read mode; @unlocked will indicate whether - * the lock has been dropped and re-acquired. This is the same behaviour as - * fixup_user_fault(). - * - * Return: 0 on success, -EAGAIN if the fault cannot be resolved without - * waiting, -EFAULT if the fault cannot be resolved, -ENOMEM if out of - * memory. - */ -static int fixup_user_fault_nowait(struct mm_struct *mm, unsigned long address, - unsigned int fault_flags, bool *unlocked) -{ - struct vm_area_struct *vma; - unsigned int test_flags; - vm_fault_t fault; - int rc; - - fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT; - test_flags = fault_flags & FAULT_FLAG_WRITE ? VM_WRITE : VM_READ; - - vma = find_vma(mm, address); - if (unlikely(!vma || address < vma->vm_start)) - return -EFAULT; - if (unlikely(!(vma->vm_flags & test_flags))) - return -EFAULT; - - fault = handle_mm_fault(vma, address, fault_flags, NULL); - /* the mm lock has been dropped, take it again */ - if (fault & VM_FAULT_COMPLETED) { - *unlocked = true; - mmap_read_lock(mm); - return 0; - } - /* the mm lock has not been dropped */ - if (fault & VM_FAULT_ERROR) { - rc = vm_fault_to_errno(fault, 0); - BUG_ON(!rc); - return rc; - } - /* the mm lock has not been dropped because of FAULT_FLAG_RETRY_NOWAIT */ - if (fault & VM_FAULT_RETRY) - return -EAGAIN; - /* nothing needed to be done and the mm lock has not been dropped */ - return 0; -} - -/** - * __gmap_fault - resolve a fault on a guest address - * @gmap: pointer to guest mapping meta data structure - * @gaddr: guest address - * @fault_flags: flags to pass down to handle_mm_fault() - * - * Context: Needs to be called with mm->mmap_lock held in read mode. Might - * drop and re-acquire the lock. Will always return with the lock held. - */ -static int __gmap_fault(struct gmap *gmap, unsigned long gaddr, unsigned int fault_flags) -{ - unsigned long vmaddr; - bool unlocked; - int rc = 0; - -retry: - unlocked = false; - - vmaddr = __gmap_translate(gmap, gaddr); - if (IS_ERR_VALUE(vmaddr)) - return vmaddr; - - if (fault_flags & FAULT_FLAG_RETRY_NOWAIT) - rc = fixup_user_fault_nowait(gmap->mm, vmaddr, fault_flags, &unlocked); - else - rc = fixup_user_fault(gmap->mm, vmaddr, fault_flags, &unlocked); - if (rc) - return rc; - /* - * In the case that fixup_user_fault unlocked the mmap_lock during - * fault-in, redo __gmap_translate() to avoid racing with a - * map/unmap_segment. - * In particular, __gmap_translate(), fixup_user_fault{,_nowait}(), - * and __gmap_link() must all be called atomically in one go; if the - * lock had been dropped in between, a retry is needed. - */ - if (unlocked) - goto retry; - - return __gmap_link(gmap, gaddr, vmaddr); -} - -/** - * gmap_fault - resolve a fault on a guest address - * @gmap: pointer to guest mapping meta data structure - * @gaddr: guest address - * @fault_flags: flags to pass down to handle_mm_fault() - * - * Returns 0 on success, -ENOMEM for out of memory conditions, -EFAULT if the - * vm address is already mapped to a different guest segment, and -EAGAIN if - * FAULT_FLAG_RETRY_NOWAIT was specified and the fault could not be processed - * immediately. - */ -int gmap_fault(struct gmap *gmap, unsigned long gaddr, unsigned int fault_flags) -{ - int rc; - - mmap_read_lock(gmap->mm); - rc = __gmap_fault(gmap, gaddr, fault_flags); - mmap_read_unlock(gmap->mm); - return rc; -} -EXPORT_SYMBOL_GPL(gmap_fault); - /* * this function is assumed to be called with mmap_lock held */ From d41993f71385ce7e9661c203e02a588a93a59b24 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:20 +0100 Subject: [PATCH 0287/1443] KVM: s390: get rid of gmap_translate() Add gpa_to_hva(), which uses memslots, and use it to replace all uses of gmap_translate(). Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-9-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-9-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 1 - arch/s390/kvm/interrupt.c | 19 +++++++++++-------- arch/s390/kvm/kvm-s390.h | 9 +++++++++ arch/s390/mm/gmap.c | 20 -------------------- 4 files changed, 20 insertions(+), 29 deletions(-) diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index d4572729269f..74b48f2e608a 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -111,7 +111,6 @@ int gmap_map_segment(struct gmap *gmap, unsigned long from, unsigned long to, unsigned long len); int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len); unsigned long __gmap_translate(struct gmap *, unsigned long gaddr); -unsigned long gmap_translate(struct gmap *, unsigned long gaddr); int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr); void gmap_discard(struct gmap *, unsigned long from, unsigned long to); void __gmap_zap(struct gmap *, unsigned long gaddr); diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index d4f031e086fc..07ff0e10cb7f 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -2893,7 +2893,8 @@ int kvm_set_routing_entry(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, const struct kvm_irq_routing_entry *ue) { - u64 uaddr; + u64 uaddr_s, uaddr_i; + int idx; switch (ue->type) { /* we store the userspace addresses instead of the guest addresses */ @@ -2901,14 +2902,16 @@ int kvm_set_routing_entry(struct kvm *kvm, if (kvm_is_ucontrol(kvm)) return -EINVAL; e->set = set_adapter_int; - uaddr = gmap_translate(kvm->arch.gmap, ue->u.adapter.summary_addr); - if (uaddr == -EFAULT) + + idx = srcu_read_lock(&kvm->srcu); + uaddr_s = gpa_to_hva(kvm, ue->u.adapter.summary_addr); + uaddr_i = gpa_to_hva(kvm, ue->u.adapter.ind_addr); + srcu_read_unlock(&kvm->srcu, idx); + + if (kvm_is_error_hva(uaddr_s) || kvm_is_error_hva(uaddr_i)) return -EFAULT; - e->adapter.summary_addr = uaddr; - uaddr = gmap_translate(kvm->arch.gmap, ue->u.adapter.ind_addr); - if (uaddr == -EFAULT) - return -EFAULT; - e->adapter.ind_addr = uaddr; + e->adapter.summary_addr = uaddr_s; + e->adapter.ind_addr = uaddr_i; e->adapter.summary_offset = ue->u.adapter.summary_offset; e->adapter.ind_offset = ue->u.adapter.ind_offset; e->adapter.adapter_id = ue->u.adapter.adapter_id; diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 3be5291723c8..61e8544924b3 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -281,6 +281,15 @@ static inline u32 kvm_s390_get_gisa_desc(struct kvm *kvm) return gd; } +static inline hva_t gpa_to_hva(struct kvm *kvm, gpa_t gpa) +{ + hva_t hva = gfn_to_hva(kvm, gpa_to_gfn(gpa)); + + if (!kvm_is_error_hva(hva)) + hva |= offset_in_page(gpa); + return hva; +} + /* implemented in pv.c */ int kvm_s390_pv_destroy_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc); int kvm_s390_pv_create_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc); diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index e124fca14737..7fd298732d1e 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -463,26 +463,6 @@ unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr) } EXPORT_SYMBOL_GPL(__gmap_translate); -/** - * gmap_translate - translate a guest address to a user space address - * @gmap: pointer to guest mapping meta data structure - * @gaddr: guest address - * - * Returns user space address which corresponds to the guest address or - * -EFAULT if no such mapping exists. - * This function does not establish potentially missing page table entries. - */ -unsigned long gmap_translate(struct gmap *gmap, unsigned long gaddr) -{ - unsigned long rc; - - mmap_read_lock(gmap->mm); - rc = __gmap_translate(gmap, gaddr); - mmap_read_unlock(gmap->mm); - return rc; -} -EXPORT_SYMBOL_GPL(gmap_translate); - /** * gmap_unlink - disconnect a page table from the gmap shadow tables * @mm: pointer to the parent mm_struct From c9f721ed8ec6942dad951d2d8c4fca291170165e Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:21 +0100 Subject: [PATCH 0288/1443] KVM: s390: move some gmap shadowing functions away from mm/gmap.c Move some gmap shadowing functions from mm/gmap.c to kvm/kvm-s390.c and the newly created kvm/gmap-vsie.c This is a step toward removing gmap from mm. Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-10-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-10-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 9 +- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/gmap-vsie.c | 142 +++++++++++++++++++++ arch/s390/kvm/gmap.h | 20 +++ arch/s390/kvm/kvm-s390.c | 74 ++++++++++- arch/s390/kvm/kvm-s390.h | 2 + arch/s390/kvm/vsie.c | 2 + arch/s390/mm/gmap.c | 238 +++++------------------------------ 8 files changed, 271 insertions(+), 218 deletions(-) create mode 100644 arch/s390/kvm/gmap-vsie.c diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index 74b48f2e608a..dbf2329281d2 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -106,6 +106,8 @@ struct gmap *gmap_create(struct mm_struct *mm, unsigned long limit); void gmap_remove(struct gmap *gmap); struct gmap *gmap_get(struct gmap *gmap); void gmap_put(struct gmap *gmap); +void gmap_free(struct gmap *gmap); +struct gmap *gmap_alloc(unsigned long limit); int gmap_map_segment(struct gmap *gmap, unsigned long from, unsigned long to, unsigned long len); @@ -118,9 +120,7 @@ void gmap_unlink(struct mm_struct *, unsigned long *table, unsigned long vmaddr) int gmap_read_table(struct gmap *gmap, unsigned long gaddr, unsigned long *val); -struct gmap *gmap_shadow(struct gmap *parent, unsigned long asce, - int edat_level); -int gmap_shadow_valid(struct gmap *sg, unsigned long asce, int edat_level); +void gmap_unshadow(struct gmap *sg); int gmap_shadow_r2t(struct gmap *sg, unsigned long saddr, unsigned long r2t, int fake); int gmap_shadow_r3t(struct gmap *sg, unsigned long saddr, unsigned long r3t, @@ -136,8 +136,7 @@ int gmap_shadow_page(struct gmap *sg, unsigned long saddr, pte_t pte); void gmap_register_pte_notifier(struct gmap_notifier *); void gmap_unregister_pte_notifier(struct gmap_notifier *); -int gmap_mprotect_notify(struct gmap *, unsigned long start, - unsigned long len, int prot); +int gmap_protect_one(struct gmap *gmap, unsigned long gaddr, int prot, unsigned long bits); void gmap_sync_dirty_log_pmd(struct gmap *gmap, unsigned long dirty_bitmap[4], unsigned long gaddr, unsigned long vmaddr); diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index d972dea657fd..f0ffe874adc2 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -8,7 +8,7 @@ include $(srctree)/virt/kvm/Makefile.kvm ccflags-y := -Ivirt/kvm -Iarch/s390/kvm kvm-y += kvm-s390.o intercept.o interrupt.o priv.o sigp.o -kvm-y += diag.o gaccess.o guestdbg.o vsie.o pv.o gmap.o +kvm-y += diag.o gaccess.o guestdbg.o vsie.o pv.o gmap.o gmap-vsie.o kvm-$(CONFIG_VFIO_PCI_ZDEV_KVM) += pci.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/gmap-vsie.c b/arch/s390/kvm/gmap-vsie.c new file mode 100644 index 000000000000..a6d1dbb04c97 --- /dev/null +++ b/arch/s390/kvm/gmap-vsie.c @@ -0,0 +1,142 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Guest memory management for KVM/s390 nested VMs. + * + * Copyright IBM Corp. 2008, 2020, 2024 + * + * Author(s): Claudio Imbrenda + * Martin Schwidefsky + * David Hildenbrand + * Janosch Frank + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "kvm-s390.h" +#include "gmap.h" + +/** + * gmap_find_shadow - find a specific asce in the list of shadow tables + * @parent: pointer to the parent gmap + * @asce: ASCE for which the shadow table is created + * @edat_level: edat level to be used for the shadow translation + * + * Returns the pointer to a gmap if a shadow table with the given asce is + * already available, ERR_PTR(-EAGAIN) if another one is just being created, + * otherwise NULL + * + * Context: Called with parent->shadow_lock held + */ +static struct gmap *gmap_find_shadow(struct gmap *parent, unsigned long asce, int edat_level) +{ + struct gmap *sg; + + lockdep_assert_held(&parent->shadow_lock); + list_for_each_entry(sg, &parent->children, list) { + if (!gmap_shadow_valid(sg, asce, edat_level)) + continue; + if (!sg->initialized) + return ERR_PTR(-EAGAIN); + refcount_inc(&sg->ref_count); + return sg; + } + return NULL; +} + +/** + * gmap_shadow - create/find a shadow guest address space + * @parent: pointer to the parent gmap + * @asce: ASCE for which the shadow table is created + * @edat_level: edat level to be used for the shadow translation + * + * The pages of the top level page table referred by the asce parameter + * will be set to read-only and marked in the PGSTEs of the kvm process. + * The shadow table will be removed automatically on any change to the + * PTE mapping for the source table. + * + * Returns a guest address space structure, ERR_PTR(-ENOMEM) if out of memory, + * ERR_PTR(-EAGAIN) if the caller has to retry and ERR_PTR(-EFAULT) if the + * parent gmap table could not be protected. + */ +struct gmap *gmap_shadow(struct gmap *parent, unsigned long asce, int edat_level) +{ + struct gmap *sg, *new; + unsigned long limit; + int rc; + + if (KVM_BUG_ON(parent->mm->context.allow_gmap_hpage_1m, (struct kvm *)parent->private) || + KVM_BUG_ON(gmap_is_shadow(parent), (struct kvm *)parent->private)) + return ERR_PTR(-EFAULT); + spin_lock(&parent->shadow_lock); + sg = gmap_find_shadow(parent, asce, edat_level); + spin_unlock(&parent->shadow_lock); + if (sg) + return sg; + /* Create a new shadow gmap */ + limit = -1UL >> (33 - (((asce & _ASCE_TYPE_MASK) >> 2) * 11)); + if (asce & _ASCE_REAL_SPACE) + limit = -1UL; + new = gmap_alloc(limit); + if (!new) + return ERR_PTR(-ENOMEM); + new->mm = parent->mm; + new->parent = gmap_get(parent); + new->private = parent->private; + new->orig_asce = asce; + new->edat_level = edat_level; + new->initialized = false; + spin_lock(&parent->shadow_lock); + /* Recheck if another CPU created the same shadow */ + sg = gmap_find_shadow(parent, asce, edat_level); + if (sg) { + spin_unlock(&parent->shadow_lock); + gmap_free(new); + return sg; + } + if (asce & _ASCE_REAL_SPACE) { + /* only allow one real-space gmap shadow */ + list_for_each_entry(sg, &parent->children, list) { + if (sg->orig_asce & _ASCE_REAL_SPACE) { + spin_lock(&sg->guest_table_lock); + gmap_unshadow(sg); + spin_unlock(&sg->guest_table_lock); + list_del(&sg->list); + gmap_put(sg); + break; + } + } + } + refcount_set(&new->ref_count, 2); + list_add(&new->list, &parent->children); + if (asce & _ASCE_REAL_SPACE) { + /* nothing to protect, return right away */ + new->initialized = true; + spin_unlock(&parent->shadow_lock); + return new; + } + spin_unlock(&parent->shadow_lock); + /* protect after insertion, so it will get properly invalidated */ + mmap_read_lock(parent->mm); + rc = __kvm_s390_mprotect_many(parent, asce & _ASCE_ORIGIN, + ((asce & _ASCE_TABLE_LENGTH) + 1), + PROT_READ, GMAP_NOTIFY_SHADOW); + mmap_read_unlock(parent->mm); + spin_lock(&parent->shadow_lock); + new->initialized = true; + if (rc) { + list_del(&new->list); + gmap_free(new); + new = ERR_PTR(rc); + } + spin_unlock(&parent->shadow_lock); + return new; +} diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h index f2b52ce29be3..978f541059f0 100644 --- a/arch/s390/kvm/gmap.h +++ b/arch/s390/kvm/gmap.h @@ -13,5 +13,25 @@ int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb); int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr); int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr); +struct gmap *gmap_shadow(struct gmap *parent, unsigned long asce, int edat_level); + +/** + * gmap_shadow_valid - check if a shadow guest address space matches the + * given properties and is still valid + * @sg: pointer to the shadow guest address space structure + * @asce: ASCE for which the shadow table is requested + * @edat_level: edat level to be used for the shadow translation + * + * Returns 1 if the gmap shadow is still valid and matches the given + * properties, the caller can continue using it. Returns 0 otherwise, the + * caller has to request a new shadow gmap in this case. + * + */ +static inline int gmap_shadow_valid(struct gmap *sg, unsigned long asce, int edat_level) +{ + if (sg->removed) + return 0; + return sg->orig_asce == asce && sg->edat_level == edat_level; +} #endif diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 70c98bf12754..ebecb96bacce 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -4511,6 +4511,75 @@ static bool ibs_enabled(struct kvm_vcpu *vcpu) return kvm_s390_test_cpuflags(vcpu, CPUSTAT_IBS); } +static int __kvm_s390_fixup_fault_sync(struct gmap *gmap, gpa_t gaddr, unsigned int flags) +{ + struct kvm *kvm = gmap->private; + gfn_t gfn = gpa_to_gfn(gaddr); + bool unlocked; + hva_t vmaddr; + gpa_t tmp; + int rc; + + if (kvm_is_ucontrol(kvm)) { + tmp = __gmap_translate(gmap, gaddr); + gfn = gpa_to_gfn(tmp); + } + + vmaddr = gfn_to_hva(kvm, gfn); + rc = fixup_user_fault(gmap->mm, vmaddr, FAULT_FLAG_WRITE, &unlocked); + if (!rc) + rc = __gmap_link(gmap, gaddr, vmaddr); + return rc; +} + +/** + * __kvm_s390_mprotect_many() - Apply specified protection to guest pages + * @gmap: the gmap of the guest + * @gpa: the starting guest address + * @npages: how many pages to protect + * @prot: indicates access rights: PROT_NONE, PROT_READ or PROT_WRITE + * @bits: pgste notification bits to set + * + * Returns: 0 in case of success, < 0 in case of error - see gmap_protect_one() + * + * Context: kvm->srcu and gmap->mm need to be held in read mode + */ +int __kvm_s390_mprotect_many(struct gmap *gmap, gpa_t gpa, u8 npages, unsigned int prot, + unsigned long bits) +{ + unsigned int fault_flag = (prot & PROT_WRITE) ? FAULT_FLAG_WRITE : 0; + gpa_t end = gpa + npages * PAGE_SIZE; + int rc; + + for (; gpa < end; gpa = ALIGN(gpa + 1, rc)) { + rc = gmap_protect_one(gmap, gpa, prot, bits); + if (rc == -EAGAIN) { + __kvm_s390_fixup_fault_sync(gmap, gpa, fault_flag); + rc = gmap_protect_one(gmap, gpa, prot, bits); + } + if (rc < 0) + return rc; + } + + return 0; +} + +static int kvm_s390_mprotect_notify_prefix(struct kvm_vcpu *vcpu) +{ + gpa_t gaddr = kvm_s390_get_prefix(vcpu); + int idx, rc; + + idx = srcu_read_lock(&vcpu->kvm->srcu); + mmap_read_lock(vcpu->arch.gmap->mm); + + rc = __kvm_s390_mprotect_many(vcpu->arch.gmap, gaddr, 2, PROT_WRITE, GMAP_NOTIFY_MPROT); + + mmap_read_unlock(vcpu->arch.gmap->mm); + srcu_read_unlock(&vcpu->kvm->srcu, idx); + + return rc; +} + static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) { retry: @@ -4526,9 +4595,8 @@ static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) */ if (kvm_check_request(KVM_REQ_REFRESH_GUEST_PREFIX, vcpu)) { int rc; - rc = gmap_mprotect_notify(vcpu->arch.gmap, - kvm_s390_get_prefix(vcpu), - PAGE_SIZE * 2, PROT_WRITE); + + rc = kvm_s390_mprotect_notify_prefix(vcpu); if (rc) { kvm_make_request(KVM_REQ_REFRESH_GUEST_PREFIX, vcpu); return rc; diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 61e8544924b3..8d3bbb2dd8d2 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -420,6 +420,8 @@ void kvm_s390_set_cpu_timer(struct kvm_vcpu *vcpu, __u64 cputm); __u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu); int kvm_s390_cpus_from_pv(struct kvm *kvm, u16 *rc, u16 *rrc); int __kvm_s390_handle_dat_fault(struct kvm_vcpu *vcpu, gfn_t gfn, gpa_t gaddr, unsigned int flags); +int __kvm_s390_mprotect_many(struct gmap *gmap, gpa_t gpa, u8 npages, unsigned int prot, + unsigned long bits); static inline int kvm_s390_handle_dat_fault(struct kvm_vcpu *vcpu, gpa_t gaddr, unsigned int flags) { diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index a0398ff85d00..a78df3a4f353 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -22,6 +23,7 @@ #include #include "kvm-s390.h" #include "gaccess.h" +#include "gmap.h" enum vsie_page_flags { VSIE_PAGE_IN_USE = 0, diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 7fd298732d1e..ae71b401312b 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -43,7 +43,7 @@ static struct page *gmap_alloc_crst(void) * * Returns a guest address space structure. */ -static struct gmap *gmap_alloc(unsigned long limit) +struct gmap *gmap_alloc(unsigned long limit) { struct gmap *gmap; struct page *page; @@ -97,6 +97,7 @@ static struct gmap *gmap_alloc(unsigned long limit) out: return NULL; } +EXPORT_SYMBOL_GPL(gmap_alloc); /** * gmap_create - create a guest address space @@ -191,7 +192,7 @@ static void gmap_rmap_radix_tree_free(struct radix_tree_root *root) * * No locks required. There are no references to this gmap anymore. */ -static void gmap_free(struct gmap *gmap) +void gmap_free(struct gmap *gmap) { struct page *page, *next; @@ -218,6 +219,7 @@ static void gmap_free(struct gmap *gmap) kfree(gmap); } +EXPORT_SYMBOL_GPL(gmap_free); /** * gmap_get - increase reference counter for guest address space @@ -958,86 +960,40 @@ static int gmap_protect_pte(struct gmap *gmap, unsigned long gaddr, * @prot: indicates access rights: PROT_NONE, PROT_READ or PROT_WRITE * @bits: pgste notification bits to set * - * Returns 0 if successfully protected, -ENOMEM if out of memory and - * -EFAULT if gaddr is invalid (or mapping for shadows is missing). + * Returns: + * PAGE_SIZE if a small page was successfully protected; + * HPAGE_SIZE if a large page was successfully protected; + * -ENOMEM if out of memory; + * -EFAULT if gaddr is invalid (or mapping for shadows is missing); + * -EAGAIN if the guest mapping is missing and should be fixed by the caller. * - * Called with sg->mm->mmap_lock in read. + * Context: Called with sg->mm->mmap_lock in read. */ -static int gmap_protect_range(struct gmap *gmap, unsigned long gaddr, - unsigned long len, int prot, unsigned long bits) +int gmap_protect_one(struct gmap *gmap, unsigned long gaddr, int prot, unsigned long bits) { - unsigned long vmaddr, dist; pmd_t *pmdp; - int rc; + int rc = 0; BUG_ON(gmap_is_shadow(gmap)); - while (len) { - rc = -EAGAIN; - pmdp = gmap_pmd_op_walk(gmap, gaddr); - if (pmdp) { - if (!pmd_leaf(*pmdp)) { - rc = gmap_protect_pte(gmap, gaddr, pmdp, prot, - bits); - if (!rc) { - len -= PAGE_SIZE; - gaddr += PAGE_SIZE; - } - } else { - rc = gmap_protect_pmd(gmap, gaddr, pmdp, prot, - bits); - if (!rc) { - dist = HPAGE_SIZE - (gaddr & ~HPAGE_MASK); - len = len < dist ? 0 : len - dist; - gaddr = (gaddr & HPAGE_MASK) + HPAGE_SIZE; - } - } - gmap_pmd_op_end(gmap, pmdp); - } - if (rc) { - if (rc == -EINVAL) - return rc; - /* -EAGAIN, fixup of userspace mm and gmap */ - vmaddr = __gmap_translate(gmap, gaddr); - if (IS_ERR_VALUE(vmaddr)) - return vmaddr; - rc = gmap_pte_op_fixup(gmap, gaddr, vmaddr, prot); - if (rc) - return rc; - } + pmdp = gmap_pmd_op_walk(gmap, gaddr); + if (!pmdp) + return -EAGAIN; + + if (!pmd_leaf(*pmdp)) { + rc = gmap_protect_pte(gmap, gaddr, pmdp, prot, bits); + if (!rc) + rc = PAGE_SIZE; + } else { + rc = gmap_protect_pmd(gmap, gaddr, pmdp, prot, bits); + if (!rc) + rc = HPAGE_SIZE; } - return 0; -} + gmap_pmd_op_end(gmap, pmdp); -/** - * gmap_mprotect_notify - change access rights for a range of ptes and - * call the notifier if any pte changes again - * @gmap: pointer to guest mapping meta data structure - * @gaddr: virtual address in the guest address space - * @len: size of area - * @prot: indicates access rights: PROT_NONE, PROT_READ or PROT_WRITE - * - * Returns 0 if for each page in the given range a gmap mapping exists, - * the new access rights could be set and the notifier could be armed. - * If the gmap mapping is missing for one or more pages -EFAULT is - * returned. If no memory could be allocated -ENOMEM is returned. - * This function establishes missing page table entries. - */ -int gmap_mprotect_notify(struct gmap *gmap, unsigned long gaddr, - unsigned long len, int prot) -{ - int rc; - - if ((gaddr & ~PAGE_MASK) || (len & ~PAGE_MASK) || gmap_is_shadow(gmap)) - return -EINVAL; - if (!MACHINE_HAS_ESOP && prot == PROT_READ) - return -EINVAL; - mmap_read_lock(gmap->mm); - rc = gmap_protect_range(gmap, gaddr, len, prot, GMAP_NOTIFY_MPROT); - mmap_read_unlock(gmap->mm); return rc; } -EXPORT_SYMBOL_GPL(gmap_mprotect_notify); +EXPORT_SYMBOL_GPL(gmap_protect_one); /** * gmap_read_table - get an unsigned long value from a guest page table using @@ -1488,7 +1444,7 @@ static void __gmap_unshadow_r1t(struct gmap *sg, unsigned long raddr, * * Called with sg->guest_table_lock */ -static void gmap_unshadow(struct gmap *sg) +void gmap_unshadow(struct gmap *sg) { unsigned long *table; @@ -1514,143 +1470,7 @@ static void gmap_unshadow(struct gmap *sg) break; } } - -/** - * gmap_find_shadow - find a specific asce in the list of shadow tables - * @parent: pointer to the parent gmap - * @asce: ASCE for which the shadow table is created - * @edat_level: edat level to be used for the shadow translation - * - * Returns the pointer to a gmap if a shadow table with the given asce is - * already available, ERR_PTR(-EAGAIN) if another one is just being created, - * otherwise NULL - */ -static struct gmap *gmap_find_shadow(struct gmap *parent, unsigned long asce, - int edat_level) -{ - struct gmap *sg; - - list_for_each_entry(sg, &parent->children, list) { - if (sg->orig_asce != asce || sg->edat_level != edat_level || - sg->removed) - continue; - if (!sg->initialized) - return ERR_PTR(-EAGAIN); - refcount_inc(&sg->ref_count); - return sg; - } - return NULL; -} - -/** - * gmap_shadow_valid - check if a shadow guest address space matches the - * given properties and is still valid - * @sg: pointer to the shadow guest address space structure - * @asce: ASCE for which the shadow table is requested - * @edat_level: edat level to be used for the shadow translation - * - * Returns 1 if the gmap shadow is still valid and matches the given - * properties, the caller can continue using it. Returns 0 otherwise, the - * caller has to request a new shadow gmap in this case. - * - */ -int gmap_shadow_valid(struct gmap *sg, unsigned long asce, int edat_level) -{ - if (sg->removed) - return 0; - return sg->orig_asce == asce && sg->edat_level == edat_level; -} -EXPORT_SYMBOL_GPL(gmap_shadow_valid); - -/** - * gmap_shadow - create/find a shadow guest address space - * @parent: pointer to the parent gmap - * @asce: ASCE for which the shadow table is created - * @edat_level: edat level to be used for the shadow translation - * - * The pages of the top level page table referred by the asce parameter - * will be set to read-only and marked in the PGSTEs of the kvm process. - * The shadow table will be removed automatically on any change to the - * PTE mapping for the source table. - * - * Returns a guest address space structure, ERR_PTR(-ENOMEM) if out of memory, - * ERR_PTR(-EAGAIN) if the caller has to retry and ERR_PTR(-EFAULT) if the - * parent gmap table could not be protected. - */ -struct gmap *gmap_shadow(struct gmap *parent, unsigned long asce, - int edat_level) -{ - struct gmap *sg, *new; - unsigned long limit; - int rc; - - BUG_ON(parent->mm->context.allow_gmap_hpage_1m); - BUG_ON(gmap_is_shadow(parent)); - spin_lock(&parent->shadow_lock); - sg = gmap_find_shadow(parent, asce, edat_level); - spin_unlock(&parent->shadow_lock); - if (sg) - return sg; - /* Create a new shadow gmap */ - limit = -1UL >> (33 - (((asce & _ASCE_TYPE_MASK) >> 2) * 11)); - if (asce & _ASCE_REAL_SPACE) - limit = -1UL; - new = gmap_alloc(limit); - if (!new) - return ERR_PTR(-ENOMEM); - new->mm = parent->mm; - new->parent = gmap_get(parent); - new->private = parent->private; - new->orig_asce = asce; - new->edat_level = edat_level; - new->initialized = false; - spin_lock(&parent->shadow_lock); - /* Recheck if another CPU created the same shadow */ - sg = gmap_find_shadow(parent, asce, edat_level); - if (sg) { - spin_unlock(&parent->shadow_lock); - gmap_free(new); - return sg; - } - if (asce & _ASCE_REAL_SPACE) { - /* only allow one real-space gmap shadow */ - list_for_each_entry(sg, &parent->children, list) { - if (sg->orig_asce & _ASCE_REAL_SPACE) { - spin_lock(&sg->guest_table_lock); - gmap_unshadow(sg); - spin_unlock(&sg->guest_table_lock); - list_del(&sg->list); - gmap_put(sg); - break; - } - } - } - refcount_set(&new->ref_count, 2); - list_add(&new->list, &parent->children); - if (asce & _ASCE_REAL_SPACE) { - /* nothing to protect, return right away */ - new->initialized = true; - spin_unlock(&parent->shadow_lock); - return new; - } - spin_unlock(&parent->shadow_lock); - /* protect after insertion, so it will get properly invalidated */ - mmap_read_lock(parent->mm); - rc = gmap_protect_range(parent, asce & _ASCE_ORIGIN, - ((asce & _ASCE_TABLE_LENGTH) + 1) * PAGE_SIZE, - PROT_READ, GMAP_NOTIFY_SHADOW); - mmap_read_unlock(parent->mm); - spin_lock(&parent->shadow_lock); - new->initialized = true; - if (rc) { - list_del(&new->list); - gmap_free(new); - new = ERR_PTR(rc); - } - spin_unlock(&parent->shadow_lock); - return new; -} -EXPORT_SYMBOL_GPL(gmap_shadow); +EXPORT_SYMBOL(gmap_unshadow); /** * gmap_shadow_r2t - create an empty shadow region 2 table From 37d1b5d8d588a9761e47d9941005e2da7def8310 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:22 +0100 Subject: [PATCH 0289/1443] KVM: s390: stop using page->index for non-shadow gmaps The host_to_guest radix tree will now map userspace addresses to guest addresses, instead of userspace addresses to segment tables. When segment tables and page tables are needed, they are found using an additional gmap_table_walk(). This gets rid of all usage of page->index for non-shadow gmaps. Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-11-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-11-imbrenda@linux.ibm.com> --- arch/s390/mm/gmap.c | 105 +++++++++++++++++++++++--------------------- 1 file changed, 54 insertions(+), 51 deletions(-) diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index ae71b401312b..1f83262a5a55 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -24,8 +24,20 @@ #include #include +/* + * The address is saved in a radix tree directly; NULL would be ambiguous, + * since 0 is a valid address, and NULL is returned when nothing was found. + * The lower bits are ignored by all users of the macro, so it can be used + * to distinguish a valid address 0 from a NULL. + */ +#define VALID_GADDR_FLAG 1 +#define IS_GADDR_VALID(gaddr) ((gaddr) & VALID_GADDR_FLAG) +#define MAKE_VALID_GADDR(gaddr) (((gaddr) & HPAGE_MASK) | VALID_GADDR_FLAG) + #define GMAP_SHADOW_FAKE_TABLE 1ULL +static inline unsigned long *gmap_table_walk(struct gmap *gmap, unsigned long gaddr, int level); + static struct page *gmap_alloc_crst(void) { struct page *page; @@ -82,7 +94,6 @@ struct gmap *gmap_alloc(unsigned long limit) page = gmap_alloc_crst(); if (!page) goto out_free; - page->index = 0; list_add(&page->lru, &gmap->crst_list); table = page_to_virt(page); crst_table_init(table, etype); @@ -303,7 +314,6 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table, list_add(&page->lru, &gmap->crst_list); *table = __pa(new) | _REGION_ENTRY_LENGTH | (*table & _REGION_ENTRY_TYPE_MASK); - page->index = gaddr; page = NULL; } spin_unlock(&gmap->guest_table_lock); @@ -312,21 +322,23 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table, return 0; } -/** - * __gmap_segment_gaddr - find virtual address from segment pointer - * @entry: pointer to a segment table entry in the guest address space - * - * Returns the virtual address in the guest address space for the segment - */ -static unsigned long __gmap_segment_gaddr(unsigned long *entry) +static unsigned long host_to_guest_lookup(struct gmap *gmap, unsigned long vmaddr) { - struct page *page; - unsigned long offset; + return (unsigned long)radix_tree_lookup(&gmap->host_to_guest, vmaddr >> PMD_SHIFT); +} - offset = (unsigned long) entry / sizeof(unsigned long); - offset = (offset & (PTRS_PER_PMD - 1)) * PMD_SIZE; - page = pmd_pgtable_page((pmd_t *) entry); - return page->index + offset; +static unsigned long host_to_guest_delete(struct gmap *gmap, unsigned long vmaddr) +{ + return (unsigned long)radix_tree_delete(&gmap->host_to_guest, vmaddr >> PMD_SHIFT); +} + +static pmd_t *host_to_guest_pmd_delete(struct gmap *gmap, unsigned long vmaddr, + unsigned long *gaddr) +{ + *gaddr = host_to_guest_delete(gmap, vmaddr); + if (IS_GADDR_VALID(*gaddr)) + return (pmd_t *)gmap_table_walk(gmap, *gaddr, 1); + return NULL; } /** @@ -338,16 +350,19 @@ static unsigned long __gmap_segment_gaddr(unsigned long *entry) */ static int __gmap_unlink_by_vmaddr(struct gmap *gmap, unsigned long vmaddr) { - unsigned long *entry; + unsigned long gaddr; int flush = 0; + pmd_t *pmdp; BUG_ON(gmap_is_shadow(gmap)); spin_lock(&gmap->guest_table_lock); - entry = radix_tree_delete(&gmap->host_to_guest, vmaddr >> PMD_SHIFT); - if (entry) { - flush = (*entry != _SEGMENT_ENTRY_EMPTY); - *entry = _SEGMENT_ENTRY_EMPTY; + + pmdp = host_to_guest_pmd_delete(gmap, vmaddr, &gaddr); + if (pmdp) { + flush = (pmd_val(*pmdp) != _SEGMENT_ENTRY_EMPTY); + *pmdp = __pmd(_SEGMENT_ENTRY_EMPTY); } + spin_unlock(&gmap->guest_table_lock); return flush; } @@ -564,7 +579,8 @@ int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr) spin_lock(&gmap->guest_table_lock); if (*table == _SEGMENT_ENTRY_EMPTY) { rc = radix_tree_insert(&gmap->host_to_guest, - vmaddr >> PMD_SHIFT, table); + vmaddr >> PMD_SHIFT, + (void *)MAKE_VALID_GADDR(gaddr)); if (!rc) { if (pmd_leaf(*pmd)) { *table = (pmd_val(*pmd) & @@ -1995,7 +2011,6 @@ void ptep_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte, unsigned long bits) { unsigned long offset, gaddr = 0; - unsigned long *table; struct gmap *gmap, *sg, *next; offset = ((unsigned long) pte) & (255 * sizeof(pte_t)); @@ -2003,12 +2018,9 @@ void ptep_notify(struct mm_struct *mm, unsigned long vmaddr, rcu_read_lock(); list_for_each_entry_rcu(gmap, &mm->context.gmap_list, list) { spin_lock(&gmap->guest_table_lock); - table = radix_tree_lookup(&gmap->host_to_guest, - vmaddr >> PMD_SHIFT); - if (table) - gaddr = __gmap_segment_gaddr(table) + offset; + gaddr = host_to_guest_lookup(gmap, vmaddr) + offset; spin_unlock(&gmap->guest_table_lock); - if (!table) + if (!IS_GADDR_VALID(gaddr)) continue; if (!list_empty(&gmap->children) && (bits & PGSTE_VSIE_BIT)) { @@ -2068,10 +2080,8 @@ static void gmap_pmdp_clear(struct mm_struct *mm, unsigned long vmaddr, rcu_read_lock(); list_for_each_entry_rcu(gmap, &mm->context.gmap_list, list) { spin_lock(&gmap->guest_table_lock); - pmdp = (pmd_t *)radix_tree_delete(&gmap->host_to_guest, - vmaddr >> PMD_SHIFT); + pmdp = host_to_guest_pmd_delete(gmap, vmaddr, &gaddr); if (pmdp) { - gaddr = __gmap_segment_gaddr((unsigned long *)pmdp); pmdp_notify_gmap(gmap, pmdp, gaddr); WARN_ON(pmd_val(*pmdp) & ~(_SEGMENT_ENTRY_HARDWARE_BITS_LARGE | _SEGMENT_ENTRY_GMAP_UC | @@ -2115,28 +2125,25 @@ EXPORT_SYMBOL_GPL(gmap_pmdp_csp); */ void gmap_pmdp_idte_local(struct mm_struct *mm, unsigned long vmaddr) { - unsigned long *entry, gaddr; + unsigned long gaddr; struct gmap *gmap; pmd_t *pmdp; rcu_read_lock(); list_for_each_entry_rcu(gmap, &mm->context.gmap_list, list) { spin_lock(&gmap->guest_table_lock); - entry = radix_tree_delete(&gmap->host_to_guest, - vmaddr >> PMD_SHIFT); - if (entry) { - pmdp = (pmd_t *)entry; - gaddr = __gmap_segment_gaddr(entry); + pmdp = host_to_guest_pmd_delete(gmap, vmaddr, &gaddr); + if (pmdp) { pmdp_notify_gmap(gmap, pmdp, gaddr); - WARN_ON(*entry & ~(_SEGMENT_ENTRY_HARDWARE_BITS_LARGE | - _SEGMENT_ENTRY_GMAP_UC | - _SEGMENT_ENTRY)); + WARN_ON(pmd_val(*pmdp) & ~(_SEGMENT_ENTRY_HARDWARE_BITS_LARGE | + _SEGMENT_ENTRY_GMAP_UC | + _SEGMENT_ENTRY)); if (MACHINE_HAS_TLB_GUEST) __pmdp_idte(gaddr, pmdp, IDTE_GUEST_ASCE, gmap->asce, IDTE_LOCAL); else if (MACHINE_HAS_IDTE) __pmdp_idte(gaddr, pmdp, 0, 0, IDTE_LOCAL); - *entry = _SEGMENT_ENTRY_EMPTY; + *pmdp = __pmd(_SEGMENT_ENTRY_EMPTY); } spin_unlock(&gmap->guest_table_lock); } @@ -2151,22 +2158,19 @@ EXPORT_SYMBOL_GPL(gmap_pmdp_idte_local); */ void gmap_pmdp_idte_global(struct mm_struct *mm, unsigned long vmaddr) { - unsigned long *entry, gaddr; + unsigned long gaddr; struct gmap *gmap; pmd_t *pmdp; rcu_read_lock(); list_for_each_entry_rcu(gmap, &mm->context.gmap_list, list) { spin_lock(&gmap->guest_table_lock); - entry = radix_tree_delete(&gmap->host_to_guest, - vmaddr >> PMD_SHIFT); - if (entry) { - pmdp = (pmd_t *)entry; - gaddr = __gmap_segment_gaddr(entry); + pmdp = host_to_guest_pmd_delete(gmap, vmaddr, &gaddr); + if (pmdp) { pmdp_notify_gmap(gmap, pmdp, gaddr); - WARN_ON(*entry & ~(_SEGMENT_ENTRY_HARDWARE_BITS_LARGE | - _SEGMENT_ENTRY_GMAP_UC | - _SEGMENT_ENTRY)); + WARN_ON(pmd_val(*pmdp) & ~(_SEGMENT_ENTRY_HARDWARE_BITS_LARGE | + _SEGMENT_ENTRY_GMAP_UC | + _SEGMENT_ENTRY)); if (MACHINE_HAS_TLB_GUEST) __pmdp_idte(gaddr, pmdp, IDTE_GUEST_ASCE, gmap->asce, IDTE_GLOBAL); @@ -2174,7 +2178,7 @@ void gmap_pmdp_idte_global(struct mm_struct *mm, unsigned long vmaddr) __pmdp_idte(gaddr, pmdp, 0, 0, IDTE_GLOBAL); else __pmdp_csp(pmdp); - *entry = _SEGMENT_ENTRY_EMPTY; + *pmdp = __pmd(_SEGMENT_ENTRY_EMPTY); } spin_unlock(&gmap->guest_table_lock); } @@ -2690,7 +2694,6 @@ int s390_replace_asce(struct gmap *gmap) page = gmap_alloc_crst(); if (!page) return -ENOMEM; - page->index = 0; table = page_to_virt(page); memcpy(table, gmap->table, 1UL << (CRST_ALLOC_ORDER + PAGE_SHIFT)); From ef0c8ef8485d9629c6d042cea8f2082f159b467e Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:23 +0100 Subject: [PATCH 0290/1443] KVM: s390: stop using lists to keep track of used dat tables Until now, every dat table allocated to map a guest was put in a linked list. The page->lru field of struct page was used to keep track of which pages were being used, and when the gmap is torn down, the list was walked and all pages freed. This patch gets rid of the usage of page->lru. Page tables are now freed by recursively walking the dat table tree. Since s390_unlist_old_asce() becomes useless now, remove it. Acked-by: Steffen Eiden Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-12-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-12-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 5 -- arch/s390/mm/gmap.c | 102 ++++++++--------------------------- arch/s390/mm/pgalloc.c | 2 - 3 files changed, 23 insertions(+), 86 deletions(-) diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index dbf2329281d2..b489c4589618 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -23,7 +23,6 @@ /** * struct gmap_struct - guest address space * @list: list head for the mm->context gmap list - * @crst_list: list of all crst tables used in the guest address space * @mm: pointer to the parent mm_struct * @guest_to_host: radix tree with guest to host address translation * @host_to_guest: radix tree with pointer to segment table entries @@ -35,7 +34,6 @@ * @guest_handle: protected virtual machine handle for the ultravisor * @host_to_rmap: radix tree with gmap_rmap lists * @children: list of shadow gmap structures - * @pt_list: list of all page tables used in the shadow guest address space * @shadow_lock: spinlock to protect the shadow gmap list * @parent: pointer to the parent gmap for shadow guest address spaces * @orig_asce: ASCE for which the shadow page table has been created @@ -45,7 +43,6 @@ */ struct gmap { struct list_head list; - struct list_head crst_list; struct mm_struct *mm; struct radix_tree_root guest_to_host; struct radix_tree_root host_to_guest; @@ -61,7 +58,6 @@ struct gmap { /* Additional data for shadow guest address spaces */ struct radix_tree_root host_to_rmap; struct list_head children; - struct list_head pt_list; spinlock_t shadow_lock; struct gmap *parent; unsigned long orig_asce; @@ -141,7 +137,6 @@ int gmap_protect_one(struct gmap *gmap, unsigned long gaddr, int prot, unsigned void gmap_sync_dirty_log_pmd(struct gmap *gmap, unsigned long dirty_bitmap[4], unsigned long gaddr, unsigned long vmaddr); int s390_disable_cow_sharing(void); -void s390_unlist_old_asce(struct gmap *gmap); int s390_replace_asce(struct gmap *gmap); void s390_uv_destroy_pfns(unsigned long count, unsigned long *pfns); int __s390_uv_destroy_range(struct mm_struct *mm, unsigned long start, diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 1f83262a5a55..07df1a7b5ebe 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -82,9 +82,7 @@ struct gmap *gmap_alloc(unsigned long limit) gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL_ACCOUNT); if (!gmap) goto out; - INIT_LIST_HEAD(&gmap->crst_list); INIT_LIST_HEAD(&gmap->children); - INIT_LIST_HEAD(&gmap->pt_list); INIT_RADIX_TREE(&gmap->guest_to_host, GFP_KERNEL_ACCOUNT); INIT_RADIX_TREE(&gmap->host_to_guest, GFP_ATOMIC | __GFP_ACCOUNT); INIT_RADIX_TREE(&gmap->host_to_rmap, GFP_ATOMIC | __GFP_ACCOUNT); @@ -94,7 +92,6 @@ struct gmap *gmap_alloc(unsigned long limit) page = gmap_alloc_crst(); if (!page) goto out_free; - list_add(&page->lru, &gmap->crst_list); table = page_to_virt(page); crst_table_init(table, etype); gmap->table = table; @@ -197,6 +194,27 @@ static void gmap_rmap_radix_tree_free(struct radix_tree_root *root) } while (nr > 0); } +static void gmap_free_crst(unsigned long *table, bool free_ptes) +{ + bool is_segment = (table[0] & _SEGMENT_ENTRY_TYPE_MASK) == 0; + int i; + + if (is_segment) { + if (!free_ptes) + goto out; + for (i = 0; i < _CRST_ENTRIES; i++) + if (!(table[i] & _SEGMENT_ENTRY_INVALID)) + page_table_free_pgste(page_ptdesc(phys_to_page(table[i]))); + } else { + for (i = 0; i < _CRST_ENTRIES; i++) + if (!(table[i] & _REGION_ENTRY_INVALID)) + gmap_free_crst(__va(table[i] & PAGE_MASK), free_ptes); + } + +out: + free_pages((unsigned long)table, CRST_ALLOC_ORDER); +} + /** * gmap_free - free a guest address space * @gmap: pointer to the guest address space structure @@ -205,24 +223,17 @@ static void gmap_rmap_radix_tree_free(struct radix_tree_root *root) */ void gmap_free(struct gmap *gmap) { - struct page *page, *next; - /* Flush tlb of all gmaps (if not already done for shadows) */ if (!(gmap_is_shadow(gmap) && gmap->removed)) gmap_flush_tlb(gmap); /* Free all segment & region tables. */ - list_for_each_entry_safe(page, next, &gmap->crst_list, lru) - __free_pages(page, CRST_ALLOC_ORDER); + gmap_free_crst(gmap->table, gmap_is_shadow(gmap)); + gmap_radix_tree_free(&gmap->guest_to_host); gmap_radix_tree_free(&gmap->host_to_guest); /* Free additional data for a shadow gmap */ if (gmap_is_shadow(gmap)) { - struct ptdesc *ptdesc, *n; - - /* Free all page tables. */ - list_for_each_entry_safe(ptdesc, n, &gmap->pt_list, pt_list) - page_table_free_pgste(ptdesc); gmap_rmap_radix_tree_free(&gmap->host_to_rmap); /* Release reference to the parent */ gmap_put(gmap->parent); @@ -311,7 +322,6 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table, crst_table_init(new, init); spin_lock(&gmap->guest_table_lock); if (*table & _REGION_ENTRY_INVALID) { - list_add(&page->lru, &gmap->crst_list); *table = __pa(new) | _REGION_ENTRY_LENGTH | (*table & _REGION_ENTRY_TYPE_MASK); page = NULL; @@ -1243,7 +1253,6 @@ static void gmap_unshadow_pgt(struct gmap *sg, unsigned long raddr) __gmap_unshadow_pgt(sg, raddr, __va(pgt)); /* Free page table */ ptdesc = page_ptdesc(phys_to_page(pgt)); - list_del(&ptdesc->pt_list); page_table_free_pgste(ptdesc); } @@ -1271,7 +1280,6 @@ static void __gmap_unshadow_sgt(struct gmap *sg, unsigned long raddr, __gmap_unshadow_pgt(sg, raddr, __va(pgt)); /* Free page table */ ptdesc = page_ptdesc(phys_to_page(pgt)); - list_del(&ptdesc->pt_list); page_table_free_pgste(ptdesc); } } @@ -1301,7 +1309,6 @@ static void gmap_unshadow_sgt(struct gmap *sg, unsigned long raddr) __gmap_unshadow_sgt(sg, raddr, __va(sgt)); /* Free segment table */ page = phys_to_page(sgt); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } @@ -1329,7 +1336,6 @@ static void __gmap_unshadow_r3t(struct gmap *sg, unsigned long raddr, __gmap_unshadow_sgt(sg, raddr, __va(sgt)); /* Free segment table */ page = phys_to_page(sgt); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } } @@ -1359,7 +1365,6 @@ static void gmap_unshadow_r3t(struct gmap *sg, unsigned long raddr) __gmap_unshadow_r3t(sg, raddr, __va(r3t)); /* Free region 3 table */ page = phys_to_page(r3t); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } @@ -1387,7 +1392,6 @@ static void __gmap_unshadow_r2t(struct gmap *sg, unsigned long raddr, __gmap_unshadow_r3t(sg, raddr, __va(r3t)); /* Free region 3 table */ page = phys_to_page(r3t); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } } @@ -1417,7 +1421,6 @@ static void gmap_unshadow_r2t(struct gmap *sg, unsigned long raddr) __gmap_unshadow_r2t(sg, raddr, __va(r2t)); /* Free region 2 table */ page = phys_to_page(r2t); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } @@ -1449,7 +1452,6 @@ static void __gmap_unshadow_r1t(struct gmap *sg, unsigned long raddr, r1t[i] = _REGION1_ENTRY_EMPTY; /* Free region 2 table */ page = phys_to_page(r2t); - list_del(&page->lru); __free_pages(page, CRST_ALLOC_ORDER); } } @@ -1544,7 +1546,6 @@ int gmap_shadow_r2t(struct gmap *sg, unsigned long saddr, unsigned long r2t, _REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_INVALID; if (sg->edat_level >= 1) *table |= (r2t & _REGION_ENTRY_PROTECT); - list_add(&page->lru, &sg->crst_list); if (fake) { /* nothing to protect for fake tables */ *table &= ~_REGION_ENTRY_INVALID; @@ -1628,7 +1629,6 @@ int gmap_shadow_r3t(struct gmap *sg, unsigned long saddr, unsigned long r3t, _REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_INVALID; if (sg->edat_level >= 1) *table |= (r3t & _REGION_ENTRY_PROTECT); - list_add(&page->lru, &sg->crst_list); if (fake) { /* nothing to protect for fake tables */ *table &= ~_REGION_ENTRY_INVALID; @@ -1712,7 +1712,6 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt, _REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INVALID; if (sg->edat_level >= 1) *table |= sgt & _REGION_ENTRY_PROTECT; - list_add(&page->lru, &sg->crst_list); if (fake) { /* nothing to protect for fake tables */ *table &= ~_REGION_ENTRY_INVALID; @@ -1833,7 +1832,6 @@ int gmap_shadow_pgt(struct gmap *sg, unsigned long saddr, unsigned long pgt, /* mark as invalid as long as the parent table is not protected */ *table = (unsigned long) s_pgt | _SEGMENT_ENTRY | (pgt & _SEGMENT_ENTRY_PROTECT) | _SEGMENT_ENTRY_INVALID; - list_add(&ptdesc->pt_list, &sg->pt_list); if (fake) { /* nothing to protect for fake tables */ *table &= ~_SEGMENT_ENTRY_INVALID; @@ -2623,49 +2621,6 @@ int __s390_uv_destroy_range(struct mm_struct *mm, unsigned long start, } EXPORT_SYMBOL_GPL(__s390_uv_destroy_range); -/** - * s390_unlist_old_asce - Remove the topmost level of page tables from the - * list of page tables of the gmap. - * @gmap: the gmap whose table is to be removed - * - * On s390x, KVM keeps a list of all pages containing the page tables of the - * gmap (the CRST list). This list is used at tear down time to free all - * pages that are now not needed anymore. - * - * This function removes the topmost page of the tree (the one pointed to by - * the ASCE) from the CRST list. - * - * This means that it will not be freed when the VM is torn down, and needs - * to be handled separately by the caller, unless a leak is actually - * intended. Notice that this function will only remove the page from the - * list, the page will still be used as a top level page table (and ASCE). - */ -void s390_unlist_old_asce(struct gmap *gmap) -{ - struct page *old; - - old = virt_to_page(gmap->table); - spin_lock(&gmap->guest_table_lock); - list_del(&old->lru); - /* - * Sometimes the topmost page might need to be "removed" multiple - * times, for example if the VM is rebooted into secure mode several - * times concurrently, or if s390_replace_asce fails after calling - * s390_remove_old_asce and is attempted again later. In that case - * the old asce has been removed from the list, and therefore it - * will not be freed when the VM terminates, but the ASCE is still - * in use and still pointed to. - * A subsequent call to replace_asce will follow the pointer and try - * to remove the same page from the list again. - * Therefore it's necessary that the page of the ASCE has valid - * pointers, so list_del can work (and do nothing) without - * dereferencing stale or invalid pointers. - */ - INIT_LIST_HEAD(&old->lru); - spin_unlock(&gmap->guest_table_lock); -} -EXPORT_SYMBOL_GPL(s390_unlist_old_asce); - /** * s390_replace_asce - Try to replace the current ASCE of a gmap with a copy * @gmap: the gmap whose ASCE needs to be replaced @@ -2685,8 +2640,6 @@ int s390_replace_asce(struct gmap *gmap) struct page *page; void *table; - s390_unlist_old_asce(gmap); - /* Replacing segment type ASCEs would cause serious issues */ if ((gmap->asce & _ASCE_TYPE_MASK) == _ASCE_TYPE_SEGMENT) return -EINVAL; @@ -2697,15 +2650,6 @@ int s390_replace_asce(struct gmap *gmap) table = page_to_virt(page); memcpy(table, gmap->table, 1UL << (CRST_ALLOC_ORDER + PAGE_SHIFT)); - /* - * The caller has to deal with the old ASCE, but here we make sure - * the new one is properly added to the CRST list, so that - * it will be freed when the VM is torn down. - */ - spin_lock(&gmap->guest_table_lock); - list_add(&page->lru, &gmap->crst_list); - spin_unlock(&gmap->guest_table_lock); - /* Set new table origin while preserving existing ASCE control bits */ asce = (gmap->asce & ~_ASCE_ORIGIN) | __pa(table); WRITE_ONCE(gmap->asce, asce); diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c index cd2fef79ad2c..30387a6e98ff 100644 --- a/arch/s390/mm/pgalloc.c +++ b/arch/s390/mm/pgalloc.c @@ -176,8 +176,6 @@ unsigned long *page_table_alloc(struct mm_struct *mm) } table = ptdesc_to_virt(ptdesc); __arch_set_page_dat(table, 1); - /* pt_list is used by gmap only */ - INIT_LIST_HEAD(&ptdesc->pt_list); memset64((u64 *)table, _PAGE_INVALID, PTRS_PER_PTE); memset64((u64 *)table + PTRS_PER_PTE, 0, PTRS_PER_PTE); return table; From 43656f774a4b4a2841035947e89dcde8ee136caa Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:24 +0100 Subject: [PATCH 0291/1443] KVM: s390: move gmap_shadow_pgt_lookup() into kvm Move gmap_shadow_pgt_lookup() from mm/gmap.c into kvm/gaccess.c . Reviewed-by: Steffen Eiden Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-13-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-13-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap.h | 3 +-- arch/s390/kvm/gaccess.c | 42 +++++++++++++++++++++++++++++++- arch/s390/kvm/gmap.h | 2 ++ arch/s390/mm/gmap.c | 46 ++---------------------------------- 4 files changed, 46 insertions(+), 47 deletions(-) diff --git a/arch/s390/include/asm/gmap.h b/arch/s390/include/asm/gmap.h index b489c4589618..4e73ef46d4b2 100644 --- a/arch/s390/include/asm/gmap.h +++ b/arch/s390/include/asm/gmap.h @@ -125,8 +125,6 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt, int fake); int gmap_shadow_pgt(struct gmap *sg, unsigned long saddr, unsigned long pgt, int fake); -int gmap_shadow_pgt_lookup(struct gmap *sg, unsigned long saddr, - unsigned long *pgt, int *dat_protection, int *fake); int gmap_shadow_page(struct gmap *sg, unsigned long saddr, pte_t pte); void gmap_register_pte_notifier(struct gmap_notifier *); @@ -142,6 +140,7 @@ void s390_uv_destroy_pfns(unsigned long count, unsigned long *pfns); int __s390_uv_destroy_range(struct mm_struct *mm, unsigned long start, unsigned long end, bool interruptible); int kvm_s390_wiggle_split_folio(struct mm_struct *mm, struct folio *folio, bool split); +unsigned long *gmap_table_walk(struct gmap *gmap, unsigned long gaddr, int level); /** * s390_uv_destroy_range - Destroy a range of pages in the given mm. diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 9816b0060fbe..bb1340389369 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -16,6 +16,7 @@ #include #include #include "kvm-s390.h" +#include "gmap.h" #include "gaccess.h" /* @@ -1392,6 +1393,42 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr, return 0; } +/** + * shadow_pgt_lookup() - find a shadow page table + * @sg: pointer to the shadow guest address space structure + * @saddr: the address in the shadow aguest address space + * @pgt: parent gmap address of the page table to get shadowed + * @dat_protection: if the pgtable is marked as protected by dat + * @fake: pgt references contiguous guest memory block, not a pgtable + * + * Returns 0 if the shadow page table was found and -EAGAIN if the page + * table was not found. + * + * Called with sg->mm->mmap_lock in read. + */ +static int shadow_pgt_lookup(struct gmap *sg, unsigned long saddr, unsigned long *pgt, + int *dat_protection, int *fake) +{ + unsigned long *table; + struct page *page; + int rc; + + spin_lock(&sg->guest_table_lock); + table = gmap_table_walk(sg, saddr, 1); /* get segment pointer */ + if (table && !(*table & _SEGMENT_ENTRY_INVALID)) { + /* Shadow page tables are full pages (pte+pgste) */ + page = pfn_to_page(*table >> PAGE_SHIFT); + *pgt = page->index & ~GMAP_SHADOW_FAKE_TABLE; + *dat_protection = !!(*table & _SEGMENT_ENTRY_PROTECT); + *fake = !!(page->index & GMAP_SHADOW_FAKE_TABLE); + rc = 0; + } else { + rc = -EAGAIN; + } + spin_unlock(&sg->guest_table_lock); + return rc; +} + /** * kvm_s390_shadow_fault - handle fault on a shadow page table * @vcpu: virtual cpu @@ -1415,6 +1452,9 @@ int kvm_s390_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, int dat_protection, fake; int rc; + if (KVM_BUG_ON(!gmap_is_shadow(sg), vcpu->kvm)) + return -EFAULT; + mmap_read_lock(sg->mm); /* * We don't want any guest-2 tables to change - so the parent @@ -1423,7 +1463,7 @@ int kvm_s390_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, */ ipte_lock(vcpu->kvm); - rc = gmap_shadow_pgt_lookup(sg, saddr, &pgt, &dat_protection, &fake); + rc = shadow_pgt_lookup(sg, saddr, &pgt, &dat_protection, &fake); if (rc) rc = kvm_s390_shadow_tables(sg, saddr, &pgt, &dat_protection, &fake); diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h index 978f541059f0..c8f031c9ea5f 100644 --- a/arch/s390/kvm/gmap.h +++ b/arch/s390/kvm/gmap.h @@ -10,6 +10,8 @@ #ifndef ARCH_KVM_S390_GMAP_H #define ARCH_KVM_S390_GMAP_H +#define GMAP_SHADOW_FAKE_TABLE 1ULL + int gmap_make_secure(struct gmap *gmap, unsigned long gaddr, void *uvcb); int gmap_convert_to_secure(struct gmap *gmap, unsigned long gaddr); int gmap_destroy_page(struct gmap *gmap, unsigned long gaddr); diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 07df1a7b5ebe..918ea14515a1 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -36,8 +36,6 @@ #define GMAP_SHADOW_FAKE_TABLE 1ULL -static inline unsigned long *gmap_table_walk(struct gmap *gmap, unsigned long gaddr, int level); - static struct page *gmap_alloc_crst(void) { struct page *page; @@ -738,8 +736,7 @@ static void gmap_call_notifier(struct gmap *gmap, unsigned long start, * * Note: Can also be called for shadow gmaps. */ -static inline unsigned long *gmap_table_walk(struct gmap *gmap, - unsigned long gaddr, int level) +unsigned long *gmap_table_walk(struct gmap *gmap, unsigned long gaddr, int level) { const int asce_type = gmap->asce & _ASCE_TYPE_MASK; unsigned long *table = gmap->table; @@ -790,6 +787,7 @@ static inline unsigned long *gmap_table_walk(struct gmap *gmap, } return table; } +EXPORT_SYMBOL(gmap_table_walk); /** * gmap_pte_op_walk - walk the gmap page table, get the page table lock @@ -1744,46 +1742,6 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt, } EXPORT_SYMBOL_GPL(gmap_shadow_sgt); -/** - * gmap_shadow_pgt_lookup - find a shadow page table - * @sg: pointer to the shadow guest address space structure - * @saddr: the address in the shadow aguest address space - * @pgt: parent gmap address of the page table to get shadowed - * @dat_protection: if the pgtable is marked as protected by dat - * @fake: pgt references contiguous guest memory block, not a pgtable - * - * Returns 0 if the shadow page table was found and -EAGAIN if the page - * table was not found. - * - * Called with sg->mm->mmap_lock in read. - */ -int gmap_shadow_pgt_lookup(struct gmap *sg, unsigned long saddr, - unsigned long *pgt, int *dat_protection, - int *fake) -{ - unsigned long *table; - struct page *page; - int rc; - - BUG_ON(!gmap_is_shadow(sg)); - spin_lock(&sg->guest_table_lock); - table = gmap_table_walk(sg, saddr, 1); /* get segment pointer */ - if (table && !(*table & _SEGMENT_ENTRY_INVALID)) { - /* Shadow page tables are full pages (pte+pgste) */ - page = pfn_to_page(*table >> PAGE_SHIFT); - *pgt = page->index & ~GMAP_SHADOW_FAKE_TABLE; - *dat_protection = !!(*table & _SEGMENT_ENTRY_PROTECT); - *fake = !!(page->index & GMAP_SHADOW_FAKE_TABLE); - rc = 0; - } else { - rc = -EAGAIN; - } - spin_unlock(&sg->guest_table_lock); - return rc; - -} -EXPORT_SYMBOL_GPL(gmap_shadow_pgt_lookup); - /** * gmap_shadow_pgt - instantiate a shadow page table * @sg: pointer to the shadow guest address space structure From c27e002626b9fbd2729fa00ddda789319648e7ba Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:25 +0100 Subject: [PATCH 0292/1443] KVM: s390: remove useless page->index usage The page->index field for VSIE dat tables is only used for segment tables. Stop setting the field for all region tables. Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-14-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-14-imbrenda@linux.ibm.com> --- arch/s390/mm/gmap.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 918ea14515a1..38f044321704 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -1520,9 +1520,6 @@ int gmap_shadow_r2t(struct gmap *sg, unsigned long saddr, unsigned long r2t, page = gmap_alloc_crst(); if (!page) return -ENOMEM; - page->index = r2t & _REGION_ENTRY_ORIGIN; - if (fake) - page->index |= GMAP_SHADOW_FAKE_TABLE; s_r2t = page_to_phys(page); /* Install shadow region second table */ spin_lock(&sg->guest_table_lock); @@ -1603,9 +1600,6 @@ int gmap_shadow_r3t(struct gmap *sg, unsigned long saddr, unsigned long r3t, page = gmap_alloc_crst(); if (!page) return -ENOMEM; - page->index = r3t & _REGION_ENTRY_ORIGIN; - if (fake) - page->index |= GMAP_SHADOW_FAKE_TABLE; s_r3t = page_to_phys(page); /* Install shadow region second table */ spin_lock(&sg->guest_table_lock); @@ -1686,9 +1680,6 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt, page = gmap_alloc_crst(); if (!page) return -ENOMEM; - page->index = sgt & _REGION_ENTRY_ORIGIN; - if (fake) - page->index |= GMAP_SHADOW_FAKE_TABLE; s_sgt = page_to_phys(page); /* Install shadow region second table */ spin_lock(&sg->guest_table_lock); From 1f4389931e9fea7e8b3c1f189d505b040b25be8a Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:26 +0100 Subject: [PATCH 0293/1443] KVM: s390: move PGSTE softbits Move the softbits in the PGSTEs to the other usable area. This leaves the 16-bit block of usable bits free, which will be used in the next patch for something else. Reviewed-by: Steffen Eiden Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-15-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-15-imbrenda@linux.ibm.com> --- arch/s390/include/asm/pgtable.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index a3b51056a177..a96bde2e5f18 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -420,9 +420,9 @@ void setup_protection_map(void); #define PGSTE_HC_BIT 0x0020000000000000UL #define PGSTE_GR_BIT 0x0004000000000000UL #define PGSTE_GC_BIT 0x0002000000000000UL -#define PGSTE_UC_BIT 0x0000800000000000UL /* user dirty (migration) */ -#define PGSTE_IN_BIT 0x0000400000000000UL /* IPTE notify bit */ -#define PGSTE_VSIE_BIT 0x0000200000000000UL /* ref'd in a shadow table */ +#define PGSTE_UC_BIT 0x0000000000008000UL /* user dirty (migration) */ +#define PGSTE_IN_BIT 0x0000000000004000UL /* IPTE notify bit */ +#define PGSTE_VSIE_BIT 0x0000000000002000UL /* ref'd in a shadow table */ /* Guest Page State used for virtualization */ #define _PGSTE_GPS_ZERO 0x0000000080000000UL From 84b7387692a8c849bd8bddd0f5c5474d4923aa6e Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 23 Jan 2025 15:46:27 +0100 Subject: [PATCH 0294/1443] KVM: s390: remove the last user of page->index Shadow page tables use page->index to keep the g2 address of the guest page table being shadowed. Instead of keeping the information in page->index, split the address and smear it over the 16-bit softbits areas of 4 PGSTEs. This removes the last s390 user of page->index. Reviewed-by: Steffen Eiden Reviewed-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250123144627.312456-16-imbrenda@linux.ibm.com Signed-off-by: Claudio Imbrenda Message-ID: <20250123144627.312456-16-imbrenda@linux.ibm.com> --- arch/s390/include/asm/pgtable.h | 15 +++++++++++++++ arch/s390/kvm/gaccess.c | 6 ++++-- arch/s390/mm/gmap.c | 22 ++++++++++++++++++++-- 3 files changed, 39 insertions(+), 4 deletions(-) diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index a96bde2e5f18..3ca5af4cfe43 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -420,6 +420,7 @@ void setup_protection_map(void); #define PGSTE_HC_BIT 0x0020000000000000UL #define PGSTE_GR_BIT 0x0004000000000000UL #define PGSTE_GC_BIT 0x0002000000000000UL +#define PGSTE_ST2_MASK 0x0000ffff00000000UL #define PGSTE_UC_BIT 0x0000000000008000UL /* user dirty (migration) */ #define PGSTE_IN_BIT 0x0000000000004000UL /* IPTE notify bit */ #define PGSTE_VSIE_BIT 0x0000000000002000UL /* ref'd in a shadow table */ @@ -2007,4 +2008,18 @@ extern void s390_reset_cmma(struct mm_struct *mm); #define pmd_pgtable(pmd) \ ((pgtable_t)__va(pmd_val(pmd) & -sizeof(pte_t)*PTRS_PER_PTE)) +static inline unsigned long gmap_pgste_get_pgt_addr(unsigned long *pgt) +{ + unsigned long *pgstes, res; + + pgstes = pgt + _PAGE_ENTRIES; + + res = (pgstes[0] & PGSTE_ST2_MASK) << 16; + res |= pgstes[1] & PGSTE_ST2_MASK; + res |= (pgstes[2] & PGSTE_ST2_MASK) >> 16; + res |= (pgstes[3] & PGSTE_ST2_MASK) >> 32; + + return res; +} + #endif /* _S390_PAGE_H */ diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index bb1340389369..f6fded15633a 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -1409,6 +1409,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr, static int shadow_pgt_lookup(struct gmap *sg, unsigned long saddr, unsigned long *pgt, int *dat_protection, int *fake) { + unsigned long pt_index; unsigned long *table; struct page *page; int rc; @@ -1418,9 +1419,10 @@ static int shadow_pgt_lookup(struct gmap *sg, unsigned long saddr, unsigned long if (table && !(*table & _SEGMENT_ENTRY_INVALID)) { /* Shadow page tables are full pages (pte+pgste) */ page = pfn_to_page(*table >> PAGE_SHIFT); - *pgt = page->index & ~GMAP_SHADOW_FAKE_TABLE; + pt_index = gmap_pgste_get_pgt_addr(page_to_virt(page)); + *pgt = pt_index & ~GMAP_SHADOW_FAKE_TABLE; *dat_protection = !!(*table & _SEGMENT_ENTRY_PROTECT); - *fake = !!(page->index & GMAP_SHADOW_FAKE_TABLE); + *fake = !!(pt_index & GMAP_SHADOW_FAKE_TABLE); rc = 0; } else { rc = -EAGAIN; diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index 38f044321704..94d927785800 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -1733,6 +1733,23 @@ int gmap_shadow_sgt(struct gmap *sg, unsigned long saddr, unsigned long sgt, } EXPORT_SYMBOL_GPL(gmap_shadow_sgt); +static void gmap_pgste_set_pgt_addr(struct ptdesc *ptdesc, unsigned long pgt_addr) +{ + unsigned long *pgstes = page_to_virt(ptdesc_page(ptdesc)); + + pgstes += _PAGE_ENTRIES; + + pgstes[0] &= ~PGSTE_ST2_MASK; + pgstes[1] &= ~PGSTE_ST2_MASK; + pgstes[2] &= ~PGSTE_ST2_MASK; + pgstes[3] &= ~PGSTE_ST2_MASK; + + pgstes[0] |= (pgt_addr >> 16) & PGSTE_ST2_MASK; + pgstes[1] |= pgt_addr & PGSTE_ST2_MASK; + pgstes[2] |= (pgt_addr << 16) & PGSTE_ST2_MASK; + pgstes[3] |= (pgt_addr << 32) & PGSTE_ST2_MASK; +} + /** * gmap_shadow_pgt - instantiate a shadow page table * @sg: pointer to the shadow guest address space structure @@ -1760,9 +1777,10 @@ int gmap_shadow_pgt(struct gmap *sg, unsigned long saddr, unsigned long pgt, ptdesc = page_table_alloc_pgste(sg->mm); if (!ptdesc) return -ENOMEM; - ptdesc->pt_index = pgt & _SEGMENT_ENTRY_ORIGIN; + origin = pgt & _SEGMENT_ENTRY_ORIGIN; if (fake) - ptdesc->pt_index |= GMAP_SHADOW_FAKE_TABLE; + origin |= GMAP_SHADOW_FAKE_TABLE; + gmap_pgste_set_pgt_addr(ptdesc, origin); s_pgt = page_to_phys(ptdesc_page(ptdesc)); /* Install shadow page table */ spin_lock(&sg->guest_table_lock); From 32239066776a27287837a193b37c6e55259e5c10 Mon Sep 17 00:00:00 2001 From: Christoph Schlameuss Date: Tue, 28 Jan 2025 14:18:03 +0100 Subject: [PATCH 0295/1443] KVM: s390: selftests: Streamline uc_skey test to issue iske after sske In some rare situations a non default storage key is already set on the memory used by the test. Within normal VMs the key is reset / zapped when the memory is added to the VM. This is not the case for ucontrol VMs. With the initial iske check removed this test case can work in all situations. The function of the iske instruction is still validated by the remaining code. Fixes: 0185fbc6a2d3 ("KVM: s390: selftests: Add uc_skey VM test case") Signed-off-by: Christoph Schlameuss Link: https://lore.kernel.org/r/20250128131803.1047388-1-schlameuss@linux.ibm.com Message-ID: <20250128131803.1047388-1-schlameuss@linux.ibm.com> Signed-off-by: Claudio Imbrenda --- .../selftests/kvm/s390/ucontrol_test.c | 24 +++++-------------- 1 file changed, 6 insertions(+), 18 deletions(-) diff --git a/tools/testing/selftests/kvm/s390/ucontrol_test.c b/tools/testing/selftests/kvm/s390/ucontrol_test.c index 22ce9219620c..d265b34c54be 100644 --- a/tools/testing/selftests/kvm/s390/ucontrol_test.c +++ b/tools/testing/selftests/kvm/s390/ucontrol_test.c @@ -88,10 +88,6 @@ asm("test_skey_asm:\n" " ahi %r0,1\n" " st %r1,0(%r5,%r6)\n" - " iske %r1,%r6\n" - " ahi %r0,1\n" - " diag 0,0,0x44\n" - " sske %r1,%r6\n" " xgr %r1,%r1\n" " iske %r1,%r6\n" @@ -600,7 +596,9 @@ TEST_F(uc_kvm, uc_skey) ASSERT_EQ(true, uc_handle_exit(self)); ASSERT_EQ(1, sync_regs->gprs[0]); - /* ISKE */ + /* SSKE + ISKE */ + sync_regs->gprs[1] = skeyvalue; + run->kvm_dirty_regs |= KVM_SYNC_GPRS; ASSERT_EQ(0, uc_run_once(self)); /* @@ -612,21 +610,11 @@ TEST_F(uc_kvm, uc_skey) TEST_ASSERT_EQ(0, sie_block->ictl & (ICTL_ISKE | ICTL_SSKE | ICTL_RRBE)); TEST_ASSERT_EQ(KVM_EXIT_S390_SIEIC, self->run->exit_reason); TEST_ASSERT_EQ(ICPT_INST, sie_block->icptcode); - TEST_REQUIRE(sie_block->ipa != 0xb229); + TEST_REQUIRE(sie_block->ipa != 0xb22b); - /* ISKE contd. */ + /* SSKE + ISKE contd. */ ASSERT_EQ(false, uc_handle_exit(self)); ASSERT_EQ(2, sync_regs->gprs[0]); - /* assert initial skey (ACC = 0, R & C = 1) */ - ASSERT_EQ(0x06, sync_regs->gprs[1]); - uc_assert_diag44(self); - - /* SSKE + ISKE */ - sync_regs->gprs[1] = skeyvalue; - run->kvm_dirty_regs |= KVM_SYNC_GPRS; - ASSERT_EQ(0, uc_run_once(self)); - ASSERT_EQ(false, uc_handle_exit(self)); - ASSERT_EQ(3, sync_regs->gprs[0]); ASSERT_EQ(skeyvalue, sync_regs->gprs[1]); uc_assert_diag44(self); @@ -635,7 +623,7 @@ TEST_F(uc_kvm, uc_skey) run->kvm_dirty_regs |= KVM_SYNC_GPRS; ASSERT_EQ(0, uc_run_once(self)); ASSERT_EQ(false, uc_handle_exit(self)); - ASSERT_EQ(4, sync_regs->gprs[0]); + ASSERT_EQ(3, sync_regs->gprs[0]); /* assert R reset but rest of skey unchanged */ ASSERT_EQ(skeyvalue & 0xfa, sync_regs->gprs[1]); ASSERT_EQ(0, sync_regs->gprs[1] & 0x04); From 6daaae5ff7f3b23a2dacc9c387ff3d4f95b67cad Mon Sep 17 00:00:00 2001 From: Hans Verkuil Date: Wed, 29 Jan 2025 10:51:48 +0100 Subject: [PATCH 0296/1443] gpu: drm_dp_cec: fix broken CEC adapter properties check If the hotplug detect of a display is low for longer than one second (configurable through drm_dp_cec_unregister_delay), then the CEC adapter is unregistered since we assume the display was disconnected. If the HPD went low for less than one second, then we check if the properties of the CEC adapter have changed, since that indicates that we actually switch to new hardware and we have to unregister the old CEC device and register a new one. Unfortunately, the test for changed properties was written poorly, and after a new CEC capability was added to the CEC core code the test always returned true (i.e. the properties had changed). As a result the CEC device was unregistered and re-registered for every HPD toggle. If the CEC remote controller integration was also enabled (CONFIG_MEDIA_CEC_RC was set), then the corresponding input device was also unregistered and re-registered. As a result the input device in /sys would keep incrementing its number, e.g.: /sys/devices/pci0000:00/0000:00:08.1/0000:e7:00.0/rc/rc0/input20 Since short HPD toggles are common, the number could over time get into the thousands. While not a serious issue (i.e. nothing crashes), it is not intended to work that way. This patch changes the test so that it only checks for the single CEC capability that can actually change, and it ignores any other capabilities, so this is now safe as well if new caps are added in the future. With the changed test the bit under #ifndef CONFIG_MEDIA_CEC_RC can be dropped as well, so that's a nice cleanup. Signed-off-by: Hans Verkuil Reported-by: Farblos Reviewed-by: Dmitry Baryshkov Fixes: 2c6d1fffa1d9 ("drm: add support for DisplayPort CEC-Tunneling-over-AUX") Tested-by: Farblos Link: https://patchwork.freedesktop.org/patch/msgid/361bb03d-1691-4e23-84da-0861ead5dbdc@xs4all.nl Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/display/drm_dp_cec.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/display/drm_dp_cec.c b/drivers/gpu/drm/display/drm_dp_cec.c index 007ceb281d00..56a4965e518c 100644 --- a/drivers/gpu/drm/display/drm_dp_cec.c +++ b/drivers/gpu/drm/display/drm_dp_cec.c @@ -311,16 +311,6 @@ void drm_dp_cec_attach(struct drm_dp_aux *aux, u16 source_physical_address) if (!aux->transfer) return; -#ifndef CONFIG_MEDIA_CEC_RC - /* - * CEC_CAP_RC is part of CEC_CAP_DEFAULTS, but it is stripped by - * cec_allocate_adapter() if CONFIG_MEDIA_CEC_RC is undefined. - * - * Do this here as well to ensure the tests against cec_caps are - * correct. - */ - cec_caps &= ~CEC_CAP_RC; -#endif cancel_delayed_work_sync(&aux->cec.unregister_work); mutex_lock(&aux->cec.lock); @@ -337,7 +327,9 @@ void drm_dp_cec_attach(struct drm_dp_aux *aux, u16 source_physical_address) num_las = CEC_MAX_LOG_ADDRS; if (aux->cec.adap) { - if (aux->cec.adap->capabilities == cec_caps && + /* Check if the adapter properties have changed */ + if ((aux->cec.adap->capabilities & CEC_CAP_MONITOR_ALL) == + (cec_caps & CEC_CAP_MONITOR_ALL) && aux->cec.adap->available_log_addrs == num_las) { /* Unchanged, so just set the phys addr */ cec_s_phys_addr(aux->cec.adap, source_physical_address, false); From d07569ebcbd645e6afac7145b2daea11da5c6384 Mon Sep 17 00:00:00 2001 From: Matthew Leung Date: Fri, 17 Jan 2025 10:09:37 -0700 Subject: [PATCH 0297/1443] bus: mhi: host: Refactor BHI/BHIe based firmware loading Refactor the firmware loading code to have distinct helper functions for BHI and BHIe operations. This lays the foundation for separating the firmware loading protocol from the firmware being loaded and the EE it is loaded in. Signed-off-by: Matthew Leung Reviewed-by: Youssef Samir Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Manivannan Sadhasivam Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-2-quic_jhugo@quicinc.com --- drivers/bus/mhi/host/boot.c | 144 +++++++++++++++++++++++++----------- 1 file changed, 99 insertions(+), 45 deletions(-) diff --git a/drivers/bus/mhi/host/boot.c b/drivers/bus/mhi/host/boot.c index e8c92972f9df..9fe13d3f09f0 100644 --- a/drivers/bus/mhi/host/boot.c +++ b/drivers/bus/mhi/host/boot.c @@ -177,6 +177,36 @@ int mhi_download_rddm_image(struct mhi_controller *mhi_cntrl, bool in_panic) } EXPORT_SYMBOL_GPL(mhi_download_rddm_image); +static void mhi_fw_load_error_dump(struct mhi_controller *mhi_cntrl) +{ + struct device *dev = &mhi_cntrl->mhi_dev->dev; + rwlock_t *pm_lock = &mhi_cntrl->pm_lock; + void __iomem *base = mhi_cntrl->bhi; + int ret, i; + u32 val; + struct { + char *name; + u32 offset; + } error_reg[] = { + { "ERROR_CODE", BHI_ERRCODE }, + { "ERROR_DBG1", BHI_ERRDBG1 }, + { "ERROR_DBG2", BHI_ERRDBG2 }, + { "ERROR_DBG3", BHI_ERRDBG3 }, + { NULL }, + }; + + read_lock_bh(pm_lock); + if (MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) { + for (i = 0; error_reg[i].name; i++) { + ret = mhi_read_reg(mhi_cntrl, base, error_reg[i].offset, &val); + if (ret) + break; + dev_err(dev, "Reg: %s value: 0x%x\n", error_reg[i].name, val); + } + } + read_unlock_bh(pm_lock); +} + static int mhi_fw_load_bhie(struct mhi_controller *mhi_cntrl, const struct mhi_buf *mhi_buf) { @@ -226,24 +256,13 @@ static int mhi_fw_load_bhie(struct mhi_controller *mhi_cntrl, } static int mhi_fw_load_bhi(struct mhi_controller *mhi_cntrl, - dma_addr_t dma_addr, - size_t size) + const struct mhi_buf *mhi_buf) { - u32 tx_status, val, session_id; - int i, ret; - void __iomem *base = mhi_cntrl->bhi; - rwlock_t *pm_lock = &mhi_cntrl->pm_lock; struct device *dev = &mhi_cntrl->mhi_dev->dev; - struct { - char *name; - u32 offset; - } error_reg[] = { - { "ERROR_CODE", BHI_ERRCODE }, - { "ERROR_DBG1", BHI_ERRDBG1 }, - { "ERROR_DBG2", BHI_ERRDBG2 }, - { "ERROR_DBG3", BHI_ERRDBG3 }, - { NULL }, - }; + rwlock_t *pm_lock = &mhi_cntrl->pm_lock; + void __iomem *base = mhi_cntrl->bhi; + u32 tx_status, session_id; + int ret; read_lock_bh(pm_lock); if (!MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) { @@ -255,11 +274,9 @@ static int mhi_fw_load_bhi(struct mhi_controller *mhi_cntrl, dev_dbg(dev, "Starting image download via BHI. Session ID: %u\n", session_id); mhi_write_reg(mhi_cntrl, base, BHI_STATUS, 0); - mhi_write_reg(mhi_cntrl, base, BHI_IMGADDR_HIGH, - upper_32_bits(dma_addr)); - mhi_write_reg(mhi_cntrl, base, BHI_IMGADDR_LOW, - lower_32_bits(dma_addr)); - mhi_write_reg(mhi_cntrl, base, BHI_IMGSIZE, size); + mhi_write_reg(mhi_cntrl, base, BHI_IMGADDR_HIGH, upper_32_bits(mhi_buf->dma_addr)); + mhi_write_reg(mhi_cntrl, base, BHI_IMGADDR_LOW, lower_32_bits(mhi_buf->dma_addr)); + mhi_write_reg(mhi_cntrl, base, BHI_IMGSIZE, mhi_buf->len); mhi_write_reg(mhi_cntrl, base, BHI_IMGTXDB, session_id); read_unlock_bh(pm_lock); @@ -274,18 +291,7 @@ static int mhi_fw_load_bhi(struct mhi_controller *mhi_cntrl, if (tx_status == BHI_STATUS_ERROR) { dev_err(dev, "Image transfer failed\n"); - read_lock_bh(pm_lock); - if (MHI_REG_ACCESS_VALID(mhi_cntrl->pm_state)) { - for (i = 0; error_reg[i].name; i++) { - ret = mhi_read_reg(mhi_cntrl, base, - error_reg[i].offset, &val); - if (ret) - break; - dev_err(dev, "Reg: %s value: 0x%x\n", - error_reg[i].name, val); - } - } - read_unlock_bh(pm_lock); + mhi_fw_load_error_dump(mhi_cntrl); goto invalid_pm_state; } @@ -296,6 +302,16 @@ static int mhi_fw_load_bhi(struct mhi_controller *mhi_cntrl, return -EIO; } +static void mhi_free_bhi_buffer(struct mhi_controller *mhi_cntrl, + struct image_info *image_info) +{ + struct mhi_buf *mhi_buf = image_info->mhi_buf; + + dma_free_coherent(mhi_cntrl->cntrl_dev, mhi_buf->len, mhi_buf->buf, mhi_buf->dma_addr); + kfree(image_info->mhi_buf); + kfree(image_info); +} + void mhi_free_bhie_table(struct mhi_controller *mhi_cntrl, struct image_info *image_info) { @@ -310,6 +326,45 @@ void mhi_free_bhie_table(struct mhi_controller *mhi_cntrl, kfree(image_info); } +static int mhi_alloc_bhi_buffer(struct mhi_controller *mhi_cntrl, + struct image_info **image_info, + size_t alloc_size) +{ + struct image_info *img_info; + struct mhi_buf *mhi_buf; + + img_info = kzalloc(sizeof(*img_info), GFP_KERNEL); + if (!img_info) + return -ENOMEM; + + /* Allocate memory for entry */ + img_info->mhi_buf = kzalloc(sizeof(*img_info->mhi_buf), GFP_KERNEL); + if (!img_info->mhi_buf) + goto error_alloc_mhi_buf; + + /* Allocate and populate vector table */ + mhi_buf = img_info->mhi_buf; + + mhi_buf->len = alloc_size; + mhi_buf->buf = dma_alloc_coherent(mhi_cntrl->cntrl_dev, mhi_buf->len, + &mhi_buf->dma_addr, GFP_KERNEL); + if (!mhi_buf->buf) + goto error_alloc_segment; + + img_info->bhi_vec = NULL; + img_info->entries = 1; + *image_info = img_info; + + return 0; + +error_alloc_segment: + kfree(mhi_buf); +error_alloc_mhi_buf: + kfree(img_info); + + return -ENOMEM; +} + int mhi_alloc_bhie_table(struct mhi_controller *mhi_cntrl, struct image_info **image_info, size_t alloc_size) @@ -364,9 +419,9 @@ int mhi_alloc_bhie_table(struct mhi_controller *mhi_cntrl, return -ENOMEM; } -static void mhi_firmware_copy(struct mhi_controller *mhi_cntrl, - const u8 *buf, size_t remainder, - struct image_info *img_info) +static void mhi_firmware_copy_bhie(struct mhi_controller *mhi_cntrl, + const u8 *buf, size_t remainder, + struct image_info *img_info) { size_t to_cpy; struct mhi_buf *mhi_buf = img_info->mhi_buf; @@ -390,10 +445,9 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) const struct firmware *firmware = NULL; struct device *dev = &mhi_cntrl->mhi_dev->dev; enum mhi_pm_state new_state; + struct image_info *image; const char *fw_name; const u8 *fw_data; - void *buf; - dma_addr_t dma_addr; size_t size, fw_sz; int ret; @@ -452,17 +506,17 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) fw_sz = firmware->size; skip_req_fw: - buf = dma_alloc_coherent(mhi_cntrl->cntrl_dev, size, &dma_addr, - GFP_KERNEL); - if (!buf) { + ret = mhi_alloc_bhi_buffer(mhi_cntrl, &image, size); + if (ret) { release_firmware(firmware); goto error_fw_load; } + /* Load the firmware into BHI vec table */ + memcpy(image->mhi_buf->buf, fw_data, size); /* Download image using BHI */ - memcpy(buf, fw_data, size); - ret = mhi_fw_load_bhi(mhi_cntrl, dma_addr, size); - dma_free_coherent(mhi_cntrl->cntrl_dev, size, buf, dma_addr); + ret = mhi_fw_load_bhi(mhi_cntrl, image->mhi_buf); + mhi_free_bhi_buffer(mhi_cntrl, image); /* Error or in EDL mode, we're done */ if (ret) { @@ -493,7 +547,7 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) } /* Load the firmware into BHIE vec table */ - mhi_firmware_copy(mhi_cntrl, fw_data, fw_sz, mhi_cntrl->fbc_image); + mhi_firmware_copy_bhie(mhi_cntrl, fw_data, fw_sz, mhi_cntrl->fbc_image); } release_firmware(firmware); From f88f1d0998ea2e5779a44013dde2a3503322c98e Mon Sep 17 00:00:00 2001 From: Matthew Leung Date: Fri, 17 Jan 2025 10:09:38 -0700 Subject: [PATCH 0298/1443] bus: mhi: host: Add a policy to enable image transfer via BHIe in PBL Currently, MHI host only performs firmware transfer via BHI in PBL and BHIe from SBL. To support BHIe transfer directly from PBL, a policy needs to be added. With this policy, BHIe will be used to transfer firmware in PBL if the MHI controller has BHIe regs, sets seg_len, and does not set fbc_download. The intention is to transfer firmware using BHIe in PBL without further BHIe transfers in SBL. Signed-off-by: Matthew Leung Reviewed-by: Youssef Samir Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Jacek Lawrynowicz Reviewed-by: Manivannan Sadhasivam Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-3-quic_jhugo@quicinc.com --- drivers/bus/mhi/host/boot.c | 73 ++++++++++++++++++++++++++------- drivers/bus/mhi/host/init.c | 2 +- drivers/bus/mhi/host/internal.h | 7 ++++ 3 files changed, 66 insertions(+), 16 deletions(-) diff --git a/drivers/bus/mhi/host/boot.c b/drivers/bus/mhi/host/boot.c index 9fe13d3f09f0..a206c3f0fcfd 100644 --- a/drivers/bus/mhi/host/boot.c +++ b/drivers/bus/mhi/host/boot.c @@ -440,12 +440,59 @@ static void mhi_firmware_copy_bhie(struct mhi_controller *mhi_cntrl, } } +static enum mhi_fw_load_type mhi_fw_load_type_get(const struct mhi_controller *mhi_cntrl) +{ + if (mhi_cntrl->fbc_download) { + return MHI_FW_LOAD_FBC; + } else { + if (mhi_cntrl->seg_len) + return MHI_FW_LOAD_BHIE; + else + return MHI_FW_LOAD_BHI; + } +} + +static int mhi_load_image_bhi(struct mhi_controller *mhi_cntrl, const u8 *fw_data, size_t size) +{ + struct image_info *image; + int ret; + + ret = mhi_alloc_bhi_buffer(mhi_cntrl, &image, size); + if (ret) + return ret; + + /* Load the firmware into BHI vec table */ + memcpy(image->mhi_buf->buf, fw_data, size); + + ret = mhi_fw_load_bhi(mhi_cntrl, &image->mhi_buf[image->entries - 1]); + mhi_free_bhi_buffer(mhi_cntrl, image); + + return ret; +} + +static int mhi_load_image_bhie(struct mhi_controller *mhi_cntrl, const u8 *fw_data, size_t size) +{ + struct image_info *image; + int ret; + + ret = mhi_alloc_bhie_table(mhi_cntrl, &image, size); + if (ret) + return ret; + + mhi_firmware_copy_bhie(mhi_cntrl, fw_data, size, image); + + ret = mhi_fw_load_bhie(mhi_cntrl, &image->mhi_buf[image->entries - 1]); + mhi_free_bhie_table(mhi_cntrl, image); + + return ret; +} + void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) { const struct firmware *firmware = NULL; struct device *dev = &mhi_cntrl->mhi_dev->dev; + enum mhi_fw_load_type fw_load_type; enum mhi_pm_state new_state; - struct image_info *image; const char *fw_name; const u8 *fw_data; size_t size, fw_sz; @@ -506,21 +553,17 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) fw_sz = firmware->size; skip_req_fw: - ret = mhi_alloc_bhi_buffer(mhi_cntrl, &image, size); - if (ret) { - release_firmware(firmware); - goto error_fw_load; - } - /* Load the firmware into BHI vec table */ - memcpy(image->mhi_buf->buf, fw_data, size); - - /* Download image using BHI */ - ret = mhi_fw_load_bhi(mhi_cntrl, image->mhi_buf); - mhi_free_bhi_buffer(mhi_cntrl, image); + fw_load_type = mhi_fw_load_type_get(mhi_cntrl); + if (fw_load_type == MHI_FW_LOAD_BHIE) + ret = mhi_load_image_bhie(mhi_cntrl, fw_data, size); + else + ret = mhi_load_image_bhi(mhi_cntrl, fw_data, size); /* Error or in EDL mode, we're done */ if (ret) { - dev_err(dev, "MHI did not load image over BHI, ret: %d\n", ret); + dev_err(dev, "MHI did not load image over BHI%s, ret: %d\n", + fw_load_type == MHI_FW_LOAD_BHIE ? "e" : "", + ret); release_firmware(firmware); goto error_fw_load; } @@ -539,7 +582,7 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) * If we're doing fbc, populate vector tables while * device transitioning into MHI READY state */ - if (mhi_cntrl->fbc_download) { + if (fw_load_type == MHI_FW_LOAD_FBC) { ret = mhi_alloc_bhie_table(mhi_cntrl, &mhi_cntrl->fbc_image, fw_sz); if (ret) { release_firmware(firmware); @@ -564,7 +607,7 @@ void mhi_fw_load_handler(struct mhi_controller *mhi_cntrl) return; error_ready_state: - if (mhi_cntrl->fbc_download) { + if (fw_load_type == MHI_FW_LOAD_FBC) { mhi_free_bhie_table(mhi_cntrl, mhi_cntrl->fbc_image); mhi_cntrl->fbc_image = NULL; } diff --git a/drivers/bus/mhi/host/init.c b/drivers/bus/mhi/host/init.c index a9b1f8beee7b..13e7a55f54ff 100644 --- a/drivers/bus/mhi/host/init.c +++ b/drivers/bus/mhi/host/init.c @@ -1144,7 +1144,7 @@ int mhi_prepare_for_power_up(struct mhi_controller *mhi_cntrl) } mhi_cntrl->bhi = mhi_cntrl->regs + bhi_off; - if (mhi_cntrl->fbc_download || mhi_cntrl->rddm_size) { + if (mhi_cntrl->fbc_download || mhi_cntrl->rddm_size || mhi_cntrl->seg_len) { ret = mhi_read_reg(mhi_cntrl, mhi_cntrl->regs, BHIEOFF, &bhie_off); if (ret) { diff --git a/drivers/bus/mhi/host/internal.h b/drivers/bus/mhi/host/internal.h index 3134f111be35..ce566f7d2e92 100644 --- a/drivers/bus/mhi/host/internal.h +++ b/drivers/bus/mhi/host/internal.h @@ -29,6 +29,13 @@ struct bhi_vec_entry { u64 size; }; +enum mhi_fw_load_type { + MHI_FW_LOAD_BHI, /* BHI only in PBL */ + MHI_FW_LOAD_BHIE, /* BHIe only in PBL */ + MHI_FW_LOAD_FBC, /* BHI in PBL followed by BHIe in SBL */ + MHI_FW_LOAD_MAX, +}; + enum mhi_ch_state_type { MHI_CH_STATE_TYPE_RESET, MHI_CH_STATE_TYPE_STOP, From ab73d80836734bb76088be0da4d2dbf0f7914c2d Mon Sep 17 00:00:00 2001 From: Youssef Samir Date: Fri, 17 Jan 2025 10:09:39 -0700 Subject: [PATCH 0299/1443] accel/qaic: Allocate an exact number of MSIs Devices use 1 MSI vector for the MHI controller and as many vectors as the DMA bridge channels on the device. During the probing of the device, the driver allocates 32 MSI vectors, which is usually more than what is needed for AIC100 devices, which is wasting resources. Allocate only the needed number of MSI vectors per device. Signed-off-by: Youssef Samir Reviewed-by: Troy Hanson Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Lizhi Hou Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-4-quic_jhugo@quicinc.com --- drivers/accel/qaic/qaic_drv.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index 81819b9ef8d4..6e9bed17b3f1 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -465,12 +465,13 @@ static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev) static int init_msi(struct qaic_device *qdev, struct pci_dev *pdev) { + int irq_count = qdev->num_dbc + 1; int mhi_irq; int ret; int i; /* Managed release since we use pcim_enable_device */ - ret = pci_alloc_irq_vectors(pdev, 32, 32, PCI_IRQ_MSI); + ret = pci_alloc_irq_vectors(pdev, irq_count, irq_count, PCI_IRQ_MSI); if (ret == -ENOSPC) { ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI); if (ret < 0) @@ -485,7 +486,8 @@ static int init_msi(struct qaic_device *qdev, struct pci_dev *pdev) * interrupted, it shouldn't race with itself. */ qdev->single_msi = true; - pci_info(pdev, "Allocating 32 MSIs failed, operating in 1 MSI mode. Performance may be impacted.\n"); + pci_info(pdev, "Allocating %d MSIs failed, operating in 1 MSI mode. Performance may be impacted.\n", + irq_count); } else if (ret < 0) { return ret; } From 0600195ec70a57c3937275e6672927ee6b3c6306 Mon Sep 17 00:00:00 2001 From: Youssef Samir Date: Fri, 17 Jan 2025 10:09:40 -0700 Subject: [PATCH 0300/1443] accel/qaic: Add support for MSI-X AIC200 device will support MSI-X while AIC100 devices will keep using MSI. pci_alloc_irq_vectors() will try to allocate MSI-X vectors if it is supported by the target device, otherwise, it will fallback to MSI. Add support for MSI-X vectors allocation for AIC200 devices. Signed-off-by: Youssef Samir Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Lizhi Hou Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-5-quic_jhugo@quicinc.com --- drivers/accel/qaic/qaic_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index 6e9bed17b3f1..ce0428f6cb82 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -471,9 +471,9 @@ static int init_msi(struct qaic_device *qdev, struct pci_dev *pdev) int i; /* Managed release since we use pcim_enable_device */ - ret = pci_alloc_irq_vectors(pdev, irq_count, irq_count, PCI_IRQ_MSI); + ret = pci_alloc_irq_vectors(pdev, irq_count, irq_count, PCI_IRQ_MSI | PCI_IRQ_MSIX); if (ret == -ENOSPC) { - ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI); + ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI | PCI_IRQ_MSIX); if (ret < 0) return ret; From 8685520474bfc0fe4be83c3cbfe3fb3e1ca1514a Mon Sep 17 00:00:00 2001 From: Youssef Samir Date: Fri, 17 Jan 2025 10:09:41 -0700 Subject: [PATCH 0301/1443] accel/qaic: Mask out SR-IOV PCI resources During the initialization of the qaic device, pci_select_bars() is used to fetch a bitmask of the BARs exposed by the device. On devices that have Virtual Functions capabilities, the bitmask includes SR-IOV BARs. Use a mask to filter out SR-IOV BARs if they exist. Signed-off-by: Youssef Samir Reviewed-by: Jeffrey Hugo Signed-off-by: Jeffrey Hugo Reviewed-by: Lizhi Hou Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-6-quic_jhugo@quicinc.com --- drivers/accel/qaic/qaic_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index ce0428f6cb82..00fa07aebacd 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -431,7 +431,7 @@ static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev) int bars; int ret; - bars = pci_select_bars(pdev, IORESOURCE_MEM); + bars = pci_select_bars(pdev, IORESOURCE_MEM) & 0x3f; /* make sure the device has the expected BARs */ if (bars != (BIT(0) | BIT(2) | BIT(4))) { From 5bbe412bc7937b5cabfe38aa186a2c6890f70df0 Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Fri, 17 Jan 2025 10:09:42 -0700 Subject: [PATCH 0302/1443] accel/qaic: Add config structs for supported cards As the number of cards supported by the driver grows, their configurations will differ. The driver needs to become more dynamic to support these configurations. Currently, each card may differ in the exposed BARs, the regions they map to, and the family. Create config structs for each card, and let the driver configure the qaic_device struct based on the configurations passed to the driver. Co-developed-by: Youssef Samir Signed-off-by: Youssef Samir Signed-off-by: Jeffrey Hugo Reviewed-by: Lizhi Hou Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-7-quic_jhugo@quicinc.com --- drivers/accel/qaic/qaic.h | 13 +++-- drivers/accel/qaic/qaic_drv.c | 76 ++++++++++++++++++++---------- drivers/accel/qaic/qaic_timesync.c | 2 +- 3 files changed, 61 insertions(+), 30 deletions(-) diff --git a/drivers/accel/qaic/qaic.h b/drivers/accel/qaic/qaic.h index 02561b6cecc6..cf97fd9a7e70 100644 --- a/drivers/accel/qaic/qaic.h +++ b/drivers/accel/qaic/qaic.h @@ -32,6 +32,11 @@ #define to_accel_kdev(qddev) (to_drm(qddev)->accel->kdev) /* Return Linux device of accel node */ #define to_qaic_device(dev) (to_qaic_drm_device((dev))->qdev) +enum aic_families { + FAMILY_AIC100, + FAMILY_MAX, +}; + enum __packed dev_states { /* Device is offline or will be very soon */ QAIC_OFFLINE, @@ -113,10 +118,10 @@ struct qaic_device { struct pci_dev *pdev; /* Req. ID of request that will be queued next in MHI control device */ u32 next_seq_num; - /* Base address of bar 0 */ - void __iomem *bar_0; - /* Base address of bar 2 */ - void __iomem *bar_2; + /* Base address of the MHI bar */ + void __iomem *bar_mhi; + /* Base address of the DBCs bar */ + void __iomem *bar_dbc; /* Controller structure for MHI devices */ struct mhi_controller *mhi_cntrl; /* MHI control channel device */ diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index 00fa07aebacd..4e63e475b389 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -34,13 +34,38 @@ MODULE_IMPORT_NS("DMA_BUF"); -#define PCI_DEV_AIC080 0xa080 -#define PCI_DEV_AIC100 0xa100 +#define PCI_DEVICE_ID_QCOM_AIC080 0xa080 +#define PCI_DEVICE_ID_QCOM_AIC100 0xa100 #define QAIC_NAME "qaic" #define QAIC_DESC "Qualcomm Cloud AI Accelerators" #define CNTL_MAJOR 5 #define CNTL_MINOR 0 +struct qaic_device_config { + /* Indicates the AIC family the device belongs to */ + int family; + /* A bitmask representing the available BARs */ + int bar_mask; + /* An index value used to identify the MHI controller BAR */ + unsigned int mhi_bar_idx; + /* An index value used to identify the DBCs BAR */ + unsigned int dbc_bar_idx; +}; + +static const struct qaic_device_config aic080_config = { + .family = FAMILY_AIC100, + .bar_mask = BIT(0) | BIT(2) | BIT(4), + .mhi_bar_idx = 0, + .dbc_bar_idx = 2, +}; + +static const struct qaic_device_config aic100_config = { + .family = FAMILY_AIC100, + .bar_mask = BIT(0) | BIT(2) | BIT(4), + .mhi_bar_idx = 0, + .dbc_bar_idx = 2, +}; + bool datapath_polling; module_param(datapath_polling, bool, 0400); MODULE_PARM_DESC(datapath_polling, "Operate the datapath in polling mode"); @@ -352,7 +377,8 @@ void qaic_dev_reset_clean_local_state(struct qaic_device *qdev) release_dbc(qdev, i); } -static struct qaic_device *create_qdev(struct pci_dev *pdev, const struct pci_device_id *id) +static struct qaic_device *create_qdev(struct pci_dev *pdev, + const struct qaic_device_config *config) { struct device *dev = &pdev->dev; struct qaic_drm_device *qddev; @@ -365,12 +391,10 @@ static struct qaic_device *create_qdev(struct pci_dev *pdev, const struct pci_de return NULL; qdev->dev_state = QAIC_OFFLINE; - if (id->device == PCI_DEV_AIC080 || id->device == PCI_DEV_AIC100) { - qdev->num_dbc = 16; - qdev->dbc = devm_kcalloc(dev, qdev->num_dbc, sizeof(*qdev->dbc), GFP_KERNEL); - if (!qdev->dbc) - return NULL; - } + qdev->num_dbc = 16; + qdev->dbc = devm_kcalloc(dev, qdev->num_dbc, sizeof(*qdev->dbc), GFP_KERNEL); + if (!qdev->dbc) + return NULL; qddev = devm_drm_dev_alloc(&pdev->dev, &qaic_accel_driver, struct qaic_drm_device, drm); if (IS_ERR(qddev)) @@ -426,7 +450,8 @@ static struct qaic_device *create_qdev(struct pci_dev *pdev, const struct pci_de return qdev; } -static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev) +static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev, + const struct qaic_device_config *config) { int bars; int ret; @@ -434,9 +459,9 @@ static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev) bars = pci_select_bars(pdev, IORESOURCE_MEM) & 0x3f; /* make sure the device has the expected BARs */ - if (bars != (BIT(0) | BIT(2) | BIT(4))) { - pci_dbg(pdev, "%s: expected BARs 0, 2, and 4 not found in device. Found 0x%x\n", - __func__, bars); + if (bars != config->bar_mask) { + pci_dbg(pdev, "%s: expected BARs %#x not found in device. Found %#x\n", + __func__, config->bar_mask, bars); return -EINVAL; } @@ -449,13 +474,13 @@ static int init_pci(struct qaic_device *qdev, struct pci_dev *pdev) return ret; dma_set_max_seg_size(&pdev->dev, UINT_MAX); - qdev->bar_0 = devm_ioremap_resource(&pdev->dev, &pdev->resource[0]); - if (IS_ERR(qdev->bar_0)) - return PTR_ERR(qdev->bar_0); + qdev->bar_mhi = devm_ioremap_resource(&pdev->dev, &pdev->resource[config->mhi_bar_idx]); + if (IS_ERR(qdev->bar_mhi)) + return PTR_ERR(qdev->bar_mhi); - qdev->bar_2 = devm_ioremap_resource(&pdev->dev, &pdev->resource[2]); - if (IS_ERR(qdev->bar_2)) - return PTR_ERR(qdev->bar_2); + qdev->bar_dbc = devm_ioremap_resource(&pdev->dev, &pdev->resource[config->dbc_bar_idx]); + if (IS_ERR(qdev->bar_dbc)) + return PTR_ERR(qdev->bar_dbc); /* Managed release since we use pcim_enable_device above */ pci_set_master(pdev); @@ -517,21 +542,22 @@ static int init_msi(struct qaic_device *qdev, struct pci_dev *pdev) static int qaic_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) { + struct qaic_device_config *config = (struct qaic_device_config *)id->driver_data; struct qaic_device *qdev; int mhi_irq; int ret; int i; - qdev = create_qdev(pdev, id); + qdev = create_qdev(pdev, config); if (!qdev) return -ENOMEM; - ret = init_pci(qdev, pdev); + ret = init_pci(qdev, pdev, config); if (ret) return ret; for (i = 0; i < qdev->num_dbc; ++i) - qdev->dbc[i].dbc_base = qdev->bar_2 + QAIC_DBC_OFF(i); + qdev->dbc[i].dbc_base = qdev->bar_dbc + QAIC_DBC_OFF(i); mhi_irq = init_msi(qdev, pdev); if (mhi_irq < 0) @@ -541,7 +567,7 @@ static int qaic_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (ret) return ret; - qdev->mhi_cntrl = qaic_mhi_register_controller(pdev, qdev->bar_0, mhi_irq, + qdev->mhi_cntrl = qaic_mhi_register_controller(pdev, qdev->bar_mhi, mhi_irq, qdev->single_msi); if (IS_ERR(qdev->mhi_cntrl)) { ret = PTR_ERR(qdev->mhi_cntrl); @@ -609,8 +635,8 @@ static struct mhi_driver qaic_mhi_driver = { }; static const struct pci_device_id qaic_ids[] = { - { PCI_DEVICE(PCI_VENDOR_ID_QCOM, PCI_DEV_AIC080), }, - { PCI_DEVICE(PCI_VENDOR_ID_QCOM, PCI_DEV_AIC100), }, + { PCI_DEVICE_DATA(QCOM, AIC080, (kernel_ulong_t)&aic080_config), }, + { PCI_DEVICE_DATA(QCOM, AIC100, (kernel_ulong_t)&aic100_config), }, { } }; MODULE_DEVICE_TABLE(pci, qaic_ids); diff --git a/drivers/accel/qaic/qaic_timesync.c b/drivers/accel/qaic/qaic_timesync.c index 301f4462d51b..2473c66309d4 100644 --- a/drivers/accel/qaic/qaic_timesync.c +++ b/drivers/accel/qaic/qaic_timesync.c @@ -201,7 +201,7 @@ static int qaic_timesync_probe(struct mhi_device *mhi_dev, const struct mhi_devi goto free_sync_msg; /* Qtimer register pointer */ - mqtsdev->qtimer_addr = qdev->bar_0 + QTIMER_REG_OFFSET; + mqtsdev->qtimer_addr = qdev->bar_mhi + QTIMER_REG_OFFSET; timer_setup(timer, qaic_timesync_timer, 0); timer->expires = jiffies + msecs_to_jiffies(timesync_delay_ms); add_timer(timer); From 7271a88629857dc285d4291055b8e7811f9d3603 Mon Sep 17 00:00:00 2001 From: Jeffrey Hugo Date: Fri, 17 Jan 2025 10:09:43 -0700 Subject: [PATCH 0303/1443] accel/qaic: Add AIC200 support Add basic support for the new AIC200 product. The PCIe Device ID is 0xa110. With this, we can turn on the lights for AIC200 by leveraging much of the existing driver. Co-developed-by: Youssef Samir Signed-off-by: Youssef Samir Signed-off-by: Jeffrey Hugo Reviewed-by: Lizhi Hou Acked-by: Manivannan Sadhasivam Link: https://patchwork.freedesktop.org/patch/msgid/20250117170943.2643280-8-quic_jhugo@quicinc.com --- drivers/accel/qaic/mhi_controller.c | 360 ++++++++++++++++++++++++++-- drivers/accel/qaic/mhi_controller.h | 2 +- drivers/accel/qaic/qaic.h | 1 + drivers/accel/qaic/qaic_drv.c | 11 +- drivers/accel/qaic/sahara.c | 39 ++- 5 files changed, 395 insertions(+), 18 deletions(-) diff --git a/drivers/accel/qaic/mhi_controller.c b/drivers/accel/qaic/mhi_controller.c index 8ab82e78dd94..13a14c6c6168 100644 --- a/drivers/accel/qaic/mhi_controller.c +++ b/drivers/accel/qaic/mhi_controller.c @@ -20,6 +20,11 @@ static unsigned int mhi_timeout_ms = 2000; /* 2 sec default */ module_param(mhi_timeout_ms, uint, 0600); MODULE_PARM_DESC(mhi_timeout_ms, "MHI controller timeout value"); +static const char *fw_image_paths[FAMILY_MAX] = { + [FAMILY_AIC100] = "qcom/aic100/sbl.bin", + [FAMILY_AIC200] = "qcom/aic200/sbl.bin", +}; + static const struct mhi_channel_config aic100_channels[] = { { .name = "QAIC_LOOPBACK", @@ -439,6 +444,297 @@ static const struct mhi_channel_config aic100_channels[] = { }, }; +static const struct mhi_channel_config aic200_channels[] = { + { + .name = "QAIC_LOOPBACK", + .num = 0, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_LOOPBACK", + .num = 1, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_SAHARA", + .num = 2, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_SBL, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_SAHARA", + .num = 3, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_SBL, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_SSR", + .num = 6, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_SSR", + .num = 7, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_CONTROL", + .num = 10, + .num_elements = 128, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_CONTROL", + .num = 11, + .num_elements = 128, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_LOGGING", + .num = 12, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_SBL, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_LOGGING", + .num = 13, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_SBL, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_STATUS", + .num = 14, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_STATUS", + .num = 15, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_TELEMETRY", + .num = 16, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_TELEMETRY", + .num = 17, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_TIMESYNC_PERIODIC", + .num = 22, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "QAIC_TIMESYNC_PERIODIC", + .num = 23, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "IPCR", + .num = 24, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "IPCR", + .num = 25, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = true, + .wake_capable = false, + }, +}; + static struct mhi_event_config aic100_events[] = { { .num_elements = 32, @@ -454,16 +750,44 @@ static struct mhi_event_config aic100_events[] = { }, }; -static struct mhi_controller_config aic100_config = { - .max_channels = 128, - .timeout_ms = 0, /* controlled by mhi_timeout */ - .buf_len = 0, - .num_channels = ARRAY_SIZE(aic100_channels), - .ch_cfg = aic100_channels, - .num_events = ARRAY_SIZE(aic100_events), - .event_cfg = aic100_events, - .use_bounce_buf = false, - .m2_no_db = false, +static struct mhi_event_config aic200_events[] = { + { + .num_elements = 32, + .irq_moderation_ms = 0, + .irq = 0, + .channel = U32_MAX, + .priority = 1, + .mode = MHI_DB_BRST_DISABLE, + .data_type = MHI_ER_CTRL, + .hardware_event = false, + .client_managed = false, + .offload_channel = false, + }, +}; + +static struct mhi_controller_config mhi_cntrl_configs[] = { + [FAMILY_AIC100] = { + .max_channels = 128, + .timeout_ms = 0, /* controlled by mhi_timeout */ + .buf_len = 0, + .num_channels = ARRAY_SIZE(aic100_channels), + .ch_cfg = aic100_channels, + .num_events = ARRAY_SIZE(aic100_events), + .event_cfg = aic100_events, + .use_bounce_buf = false, + .m2_no_db = false, + }, + [FAMILY_AIC200] = { + .max_channels = 128, + .timeout_ms = 0, /* controlled by mhi_timeout */ + .buf_len = 0, + .num_channels = ARRAY_SIZE(aic200_channels), + .ch_cfg = aic200_channels, + .num_events = ARRAY_SIZE(aic200_events), + .event_cfg = aic200_events, + .use_bounce_buf = false, + .m2_no_db = false, + }, }; static int mhi_read_reg(struct mhi_controller *mhi_cntrl, void __iomem *addr, u32 *out) @@ -545,8 +869,9 @@ static int mhi_reset_and_async_power_up(struct mhi_controller *mhi_cntrl) } struct mhi_controller *qaic_mhi_register_controller(struct pci_dev *pci_dev, void __iomem *mhi_bar, - int mhi_irq, bool shared_msi) + int mhi_irq, bool shared_msi, int family) { + struct mhi_controller_config mhi_config = mhi_cntrl_configs[family]; struct mhi_controller *mhi_cntrl; int ret; @@ -581,11 +906,18 @@ struct mhi_controller *qaic_mhi_register_controller(struct pci_dev *pci_dev, voi if (shared_msi) /* MSI shared with data path, no IRQF_NO_SUSPEND */ mhi_cntrl->irq_flags = IRQF_SHARED; - mhi_cntrl->fw_image = "qcom/aic100/sbl.bin"; + mhi_cntrl->fw_image = fw_image_paths[family]; + + if (family == FAMILY_AIC200) { + mhi_cntrl->name = "AIC200"; + mhi_cntrl->seg_len = SZ_512K; + } else { + mhi_cntrl->name = "AIC100"; + } /* use latest configured timeout */ - aic100_config.timeout_ms = mhi_timeout_ms; - ret = mhi_register_controller(mhi_cntrl, &aic100_config); + mhi_config.timeout_ms = mhi_timeout_ms; + ret = mhi_register_controller(mhi_cntrl, &mhi_config); if (ret) { pci_err(pci_dev, "mhi_register_controller failed %d\n", ret); return ERR_PTR(ret); diff --git a/drivers/accel/qaic/mhi_controller.h b/drivers/accel/qaic/mhi_controller.h index 500e7f4af2af..8939f6ae185e 100644 --- a/drivers/accel/qaic/mhi_controller.h +++ b/drivers/accel/qaic/mhi_controller.h @@ -8,7 +8,7 @@ #define MHICONTROLLERQAIC_H_ struct mhi_controller *qaic_mhi_register_controller(struct pci_dev *pci_dev, void __iomem *mhi_bar, - int mhi_irq, bool shared_msi); + int mhi_irq, bool shared_msi, int family); void qaic_mhi_free_controller(struct mhi_controller *mhi_cntrl, bool link_up); void qaic_mhi_start_reset(struct mhi_controller *mhi_cntrl); void qaic_mhi_reset_done(struct mhi_controller *mhi_cntrl); diff --git a/drivers/accel/qaic/qaic.h b/drivers/accel/qaic/qaic.h index cf97fd9a7e70..0dbb8e32e4b9 100644 --- a/drivers/accel/qaic/qaic.h +++ b/drivers/accel/qaic/qaic.h @@ -34,6 +34,7 @@ enum aic_families { FAMILY_AIC100, + FAMILY_AIC200, FAMILY_MAX, }; diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index 4e63e475b389..3b415e2c9431 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -36,6 +36,7 @@ MODULE_IMPORT_NS("DMA_BUF"); #define PCI_DEVICE_ID_QCOM_AIC080 0xa080 #define PCI_DEVICE_ID_QCOM_AIC100 0xa100 +#define PCI_DEVICE_ID_QCOM_AIC200 0xa110 #define QAIC_NAME "qaic" #define QAIC_DESC "Qualcomm Cloud AI Accelerators" #define CNTL_MAJOR 5 @@ -66,6 +67,13 @@ static const struct qaic_device_config aic100_config = { .dbc_bar_idx = 2, }; +static const struct qaic_device_config aic200_config = { + .family = FAMILY_AIC200, + .bar_mask = BIT(0) | BIT(1) | BIT(2) | BIT(4), + .mhi_bar_idx = 1, + .dbc_bar_idx = 2, +}; + bool datapath_polling; module_param(datapath_polling, bool, 0400); MODULE_PARM_DESC(datapath_polling, "Operate the datapath in polling mode"); @@ -568,7 +576,7 @@ static int qaic_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) return ret; qdev->mhi_cntrl = qaic_mhi_register_controller(pdev, qdev->bar_mhi, mhi_irq, - qdev->single_msi); + qdev->single_msi, config->family); if (IS_ERR(qdev->mhi_cntrl)) { ret = PTR_ERR(qdev->mhi_cntrl); qaic_destroy_drm_device(qdev, QAIC_NO_PARTITION); @@ -637,6 +645,7 @@ static struct mhi_driver qaic_mhi_driver = { static const struct pci_device_id qaic_ids[] = { { PCI_DEVICE_DATA(QCOM, AIC080, (kernel_ulong_t)&aic080_config), }, { PCI_DEVICE_DATA(QCOM, AIC100, (kernel_ulong_t)&aic100_config), }, + { PCI_DEVICE_DATA(QCOM, AIC200, (kernel_ulong_t)&aic200_config), }, { } }; MODULE_DEVICE_TABLE(pci, qaic_ids); diff --git a/drivers/accel/qaic/sahara.c b/drivers/accel/qaic/sahara.c index 09c8b055aa81..3ebcc1f7ff58 100644 --- a/drivers/accel/qaic/sahara.c +++ b/drivers/accel/qaic/sahara.c @@ -188,6 +188,34 @@ static const char * const aic100_image_table[] = { [10] = "qcom/aic100/fw10.bin", }; +static const char * const aic200_image_table[] = { + [5] = "qcom/aic200/uefi.elf", + [12] = "qcom/aic200/aic200-nsp.bin", + [23] = "qcom/aic200/aop.mbn", + [32] = "qcom/aic200/tz.mbn", + [33] = "qcom/aic200/hypvm.mbn", + [39] = "qcom/aic200/aic200_abl.elf", + [40] = "qcom/aic200/apdp.mbn", + [41] = "qcom/aic200/devcfg.mbn", + [42] = "qcom/aic200/sec.elf", + [43] = "qcom/aic200/aic200-hlos.elf", + [49] = "qcom/aic200/shrm.elf", + [50] = "qcom/aic200/cpucp.elf", + [51] = "qcom/aic200/aop_devcfg.mbn", + [57] = "qcom/aic200/cpucp_dtbs.elf", + [62] = "qcom/aic200/uefi_dtbs.elf", + [63] = "qcom/aic200/xbl_ac_config.mbn", + [64] = "qcom/aic200/tz_ac_config.mbn", + [65] = "qcom/aic200/hyp_ac_config.mbn", + [66] = "qcom/aic200/pdp.elf", + [67] = "qcom/aic200/pdp_cdb.elf", + [68] = "qcom/aic200/sdi.mbn", + [69] = "qcom/aic200/dcd.mbn", + [73] = "qcom/aic200/gearvm.mbn", + [74] = "qcom/aic200/sti.bin", + [75] = "qcom/aic200/pvs.bin", +}; + static int sahara_find_image(struct sahara_context *context, u32 image_id) { int ret; @@ -748,8 +776,15 @@ static int sahara_mhi_probe(struct mhi_device *mhi_dev, const struct mhi_device_ context->mhi_dev = mhi_dev; INIT_WORK(&context->fw_work, sahara_processing); INIT_WORK(&context->dump_work, sahara_dump_processing); - context->image_table = aic100_image_table; - context->table_size = ARRAY_SIZE(aic100_image_table); + + if (!strcmp(mhi_dev->mhi_cntrl->name, "AIC200")) { + context->image_table = aic200_image_table; + context->table_size = ARRAY_SIZE(aic200_image_table); + } else { + context->image_table = aic100_image_table; + context->table_size = ARRAY_SIZE(aic100_image_table); + } + context->active_image_id = SAHARA_IMAGE_ID_NONE; dev_set_drvdata(&mhi_dev->dev, context); From 743bbd93cf29f653fae0e1416a31f03231689911 Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Thu, 23 Jan 2025 16:01:16 +0100 Subject: [PATCH 0304/1443] ice: put Rx buffers after being done with current frame Introduce a new helper ice_put_rx_mbuf() that will go through gathered frags from current frame and will call ice_put_rx_buf() on them. Current logic that was supposed to simplify and optimize the driver where we go through a batch of all buffers processed in current NAPI instance turned out to be broken for jumbo frames and very heavy load that was coming from both multi-thread iperf and nginx/wrk pair between server and client. The delay introduced by approach that we are dropping is simply too big and we need to take the decision regarding page recycling/releasing as quick as we can. While at it, address an error path of ice_add_xdp_frag() - we were missing buffer putting from day 1 there. As a nice side effect we get rid of annoying and repetitive three-liner: xdp->data = NULL; rx_ring->first_desc = ntc; rx_ring->nr_frags = 0; by embedding it within introduced routine. Fixes: 1dc1a7e7f410 ("ice: Centrallize Rx buffer recycling") Reported-and-tested-by: Xu Du Reviewed-by: Przemek Kitszel Reviewed-by: Simon Horman Co-developed-by: Jacob Keller Signed-off-by: Jacob Keller Signed-off-by: Maciej Fijalkowski Tested-by: Chandan Kumar Rout (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_txrx.c | 79 ++++++++++++++--------- 1 file changed, 50 insertions(+), 29 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 5d2d7736fd5f..e173d9c98988 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1103,6 +1103,49 @@ ice_put_rx_buf(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf) rx_buf->page = NULL; } +/** + * ice_put_rx_mbuf - ice_put_rx_buf() caller, for all frame frags + * @rx_ring: Rx ring with all the auxiliary data + * @xdp: XDP buffer carrying linear + frags part + * @xdp_xmit: XDP_TX/XDP_REDIRECT verdict storage + * @ntc: a current next_to_clean value to be stored at rx_ring + * + * Walk through gathered fragments and satisfy internal page + * recycle mechanism; we take here an action related to verdict + * returned by XDP program; + */ +static void ice_put_rx_mbuf(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, + u32 *xdp_xmit, u32 ntc) +{ + u32 nr_frags = rx_ring->nr_frags + 1; + u32 idx = rx_ring->first_desc; + u32 cnt = rx_ring->count; + struct ice_rx_buf *buf; + int i; + + for (i = 0; i < nr_frags; i++) { + buf = &rx_ring->rx_buf[idx]; + + if (buf->act & (ICE_XDP_TX | ICE_XDP_REDIR)) { + ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); + *xdp_xmit |= buf->act; + } else if (buf->act & ICE_XDP_CONSUMED) { + buf->pagecnt_bias++; + } else if (buf->act == ICE_XDP_PASS) { + ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); + } + + ice_put_rx_buf(rx_ring, buf); + + if (++idx == cnt) + idx = 0; + } + + xdp->data = NULL; + rx_ring->first_desc = ntc; + rx_ring->nr_frags = 0; +} + /** * ice_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf * @rx_ring: Rx descriptor ring to transact packets on @@ -1120,7 +1163,6 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) unsigned int total_rx_bytes = 0, total_rx_pkts = 0; unsigned int offset = rx_ring->rx_offset; struct xdp_buff *xdp = &rx_ring->xdp; - u32 cached_ntc = rx_ring->first_desc; struct ice_tx_ring *xdp_ring = NULL; struct bpf_prog *xdp_prog = NULL; u32 ntc = rx_ring->next_to_clean; @@ -1128,7 +1170,6 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) u32 xdp_xmit = 0; u32 cached_ntu; bool failure; - u32 first; xdp_prog = READ_ONCE(rx_ring->xdp_prog); if (xdp_prog) { @@ -1190,6 +1231,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) xdp_prepare_buff(xdp, hard_start, offset, size, !!offset); xdp_buff_clear_frags_flag(xdp); } else if (ice_add_xdp_frag(rx_ring, xdp, rx_buf, size)) { + ice_put_rx_mbuf(rx_ring, xdp, NULL, ntc); break; } if (++ntc == cnt) @@ -1205,9 +1247,8 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) total_rx_bytes += xdp_get_buff_len(xdp); total_rx_pkts++; - xdp->data = NULL; - rx_ring->first_desc = ntc; - rx_ring->nr_frags = 0; + ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc); + continue; construct_skb: if (likely(ice_ring_uses_build_skb(rx_ring))) @@ -1221,14 +1262,11 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) if (unlikely(xdp_buff_has_frags(xdp))) ice_set_rx_bufs_act(xdp, rx_ring, ICE_XDP_CONSUMED); - xdp->data = NULL; - rx_ring->first_desc = ntc; - rx_ring->nr_frags = 0; - break; } - xdp->data = NULL; - rx_ring->first_desc = ntc; - rx_ring->nr_frags = 0; + ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc); + + if (!skb) + break; stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_RXE_S); if (unlikely(ice_test_staterr(rx_desc->wb.status_error0, @@ -1257,23 +1295,6 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) total_rx_pkts++; } - first = rx_ring->first_desc; - while (cached_ntc != first) { - struct ice_rx_buf *buf = &rx_ring->rx_buf[cached_ntc]; - - if (buf->act & (ICE_XDP_TX | ICE_XDP_REDIR)) { - ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); - xdp_xmit |= buf->act; - } else if (buf->act & ICE_XDP_CONSUMED) { - buf->pagecnt_bias++; - } else if (buf->act == ICE_XDP_PASS) { - ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); - } - - ice_put_rx_buf(rx_ring, buf); - if (++cached_ntc >= cnt) - cached_ntc = 0; - } rx_ring->next_to_clean = ntc; /* return up to cleaned_count buffers to hardware */ failure = ice_alloc_rx_bufs(rx_ring, ICE_RX_DESC_UNUSED(rx_ring)); From 11c4aa074d547d825b19cd8d9f288254d89d805c Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Thu, 23 Jan 2025 16:01:17 +0100 Subject: [PATCH 0305/1443] ice: gather page_count()'s of each frag right before XDP prog call If we store the pgcnt on few fragments while being in the middle of gathering the whole frame and we stumbled upon DD bit not being set, we terminate the NAPI Rx processing loop and come back later on. Then on next NAPI execution we work on previously stored pgcnt. Imagine that second half of page was used actively by networking stack and by the time we came back, stack is not busy with this page anymore and decremented the refcnt. The page reuse algorithm in this case should be good to reuse the page but given the old refcnt it will not do so and attempt to release the page via page_frag_cache_drain() with pagecnt_bias used as an arg. This in turn will result in negative refcnt on struct page, which was initially observed by Xu Du. Therefore, move the page count storage from ice_get_rx_buf() to a place where we are sure that whole frame has been collected, but before calling XDP program as it internally can also change the page count of fragments belonging to xdp_buff. Fixes: ac0753391195 ("ice: Store page count inside ice_rx_buf") Reported-and-tested-by: Xu Du Reviewed-by: Przemek Kitszel Reviewed-by: Simon Horman Co-developed-by: Jacob Keller Signed-off-by: Jacob Keller Signed-off-by: Maciej Fijalkowski Tested-by: Chandan Kumar Rout (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_txrx.c | 27 ++++++++++++++++++++++- 1 file changed, 26 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index e173d9c98988..cf46bcf143b4 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -924,7 +924,6 @@ ice_get_rx_buf(struct ice_rx_ring *rx_ring, const unsigned int size, struct ice_rx_buf *rx_buf; rx_buf = &rx_ring->rx_buf[ntc]; - rx_buf->pgcnt = page_count(rx_buf->page); prefetchw(rx_buf->page); if (!size) @@ -940,6 +939,31 @@ ice_get_rx_buf(struct ice_rx_ring *rx_ring, const unsigned int size, return rx_buf; } +/** + * ice_get_pgcnts - grab page_count() for gathered fragments + * @rx_ring: Rx descriptor ring to store the page counts on + * + * This function is intended to be called right before running XDP + * program so that the page recycling mechanism will be able to take + * a correct decision regarding underlying pages; this is done in such + * way as XDP program can change the refcount of page + */ +static void ice_get_pgcnts(struct ice_rx_ring *rx_ring) +{ + u32 nr_frags = rx_ring->nr_frags + 1; + u32 idx = rx_ring->first_desc; + struct ice_rx_buf *rx_buf; + u32 cnt = rx_ring->count; + + for (int i = 0; i < nr_frags; i++) { + rx_buf = &rx_ring->rx_buf[idx]; + rx_buf->pgcnt = page_count(rx_buf->page); + + if (++idx == cnt) + idx = 0; + } +} + /** * ice_build_skb - Build skb around an existing buffer * @rx_ring: Rx descriptor ring to transact packets on @@ -1241,6 +1265,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) if (ice_is_non_eop(rx_ring, rx_desc)) continue; + ice_get_pgcnts(rx_ring); ice_run_xdp(rx_ring, xdp, xdp_prog, xdp_ring, rx_buf, rx_desc); if (rx_buf->act == ICE_XDP_PASS) goto construct_skb; From 468a1952df78f65c5991b7ac885c8b5b7dd87bab Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Thu, 23 Jan 2025 16:01:18 +0100 Subject: [PATCH 0306/1443] ice: stop storing XDP verdict within ice_rx_buf Idea behind having ice_rx_buf::act was to simplify and speed up the Rx data path by walking through buffers that were representing cleaned HW Rx descriptors. Since it caused us a major headache recently and we rolled back to old approach that 'puts' Rx buffers right after running XDP prog/creating skb, this is useless now and should be removed. Get rid of ice_rx_buf::act and related logic. We still need to take care of a corner case where XDP program releases a particular fragment. Make ice_run_xdp() to return its result and use it within ice_put_rx_mbuf(). Fixes: 2fba7dc5157b ("ice: Add support for XDP multi-buffer on Rx side") Reviewed-by: Przemek Kitszel Reviewed-by: Simon Horman Signed-off-by: Maciej Fijalkowski Tested-by: Chandan Kumar Rout (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_txrx.c | 62 +++++++++++-------- drivers/net/ethernet/intel/ice/ice_txrx.h | 1 - drivers/net/ethernet/intel/ice/ice_txrx_lib.h | 43 ------------- 3 files changed, 36 insertions(+), 70 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index cf46bcf143b4..9c9ea4c1b93b 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -527,15 +527,14 @@ int ice_setup_rx_ring(struct ice_rx_ring *rx_ring) * @xdp: xdp_buff used as input to the XDP program * @xdp_prog: XDP program to run * @xdp_ring: ring to be used for XDP_TX action - * @rx_buf: Rx buffer to store the XDP action * @eop_desc: Last descriptor in packet to read metadata from * * Returns any of ICE_XDP_{PASS, CONSUMED, TX, REDIR} */ -static void +static u32 ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, struct bpf_prog *xdp_prog, struct ice_tx_ring *xdp_ring, - struct ice_rx_buf *rx_buf, union ice_32b_rx_flex_desc *eop_desc) + union ice_32b_rx_flex_desc *eop_desc) { unsigned int ret = ICE_XDP_PASS; u32 act; @@ -574,7 +573,7 @@ ice_run_xdp(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, ret = ICE_XDP_CONSUMED; } exit: - ice_set_rx_bufs_act(xdp, rx_ring, ret); + return ret; } /** @@ -860,10 +859,8 @@ ice_add_xdp_frag(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, xdp_buff_set_frags_flag(xdp); } - if (unlikely(sinfo->nr_frags == MAX_SKB_FRAGS)) { - ice_set_rx_bufs_act(xdp, rx_ring, ICE_XDP_CONSUMED); + if (unlikely(sinfo->nr_frags == MAX_SKB_FRAGS)) return -ENOMEM; - } __skb_fill_page_desc_noacc(sinfo, sinfo->nr_frags++, rx_buf->page, rx_buf->page_offset, size); @@ -1075,12 +1072,12 @@ ice_construct_skb(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) rx_buf->page_offset + headlen, size, xdp->frame_sz); } else { - /* buffer is unused, change the act that should be taken later - * on; data was copied onto skb's linear part so there's no + /* buffer is unused, restore biased page count in Rx buffer; + * data was copied onto skb's linear part so there's no * need for adjusting page offset and we can reuse this buffer * as-is */ - rx_buf->act = ICE_SKB_CONSUMED; + rx_buf->pagecnt_bias++; } if (unlikely(xdp_buff_has_frags(xdp))) { @@ -1133,29 +1130,34 @@ ice_put_rx_buf(struct ice_rx_ring *rx_ring, struct ice_rx_buf *rx_buf) * @xdp: XDP buffer carrying linear + frags part * @xdp_xmit: XDP_TX/XDP_REDIRECT verdict storage * @ntc: a current next_to_clean value to be stored at rx_ring + * @verdict: return code from XDP program execution * * Walk through gathered fragments and satisfy internal page * recycle mechanism; we take here an action related to verdict * returned by XDP program; */ static void ice_put_rx_mbuf(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, - u32 *xdp_xmit, u32 ntc) + u32 *xdp_xmit, u32 ntc, u32 verdict) { u32 nr_frags = rx_ring->nr_frags + 1; u32 idx = rx_ring->first_desc; u32 cnt = rx_ring->count; + u32 post_xdp_frags = 1; struct ice_rx_buf *buf; int i; - for (i = 0; i < nr_frags; i++) { + if (unlikely(xdp_buff_has_frags(xdp))) + post_xdp_frags += xdp_get_shared_info_from_buff(xdp)->nr_frags; + + for (i = 0; i < post_xdp_frags; i++) { buf = &rx_ring->rx_buf[idx]; - if (buf->act & (ICE_XDP_TX | ICE_XDP_REDIR)) { + if (verdict & (ICE_XDP_TX | ICE_XDP_REDIR)) { ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); - *xdp_xmit |= buf->act; - } else if (buf->act & ICE_XDP_CONSUMED) { + *xdp_xmit |= verdict; + } else if (verdict & ICE_XDP_CONSUMED) { buf->pagecnt_bias++; - } else if (buf->act == ICE_XDP_PASS) { + } else if (verdict == ICE_XDP_PASS) { ice_rx_buf_adjust_pg_offset(buf, xdp->frame_sz); } @@ -1164,6 +1166,17 @@ static void ice_put_rx_mbuf(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp, if (++idx == cnt) idx = 0; } + /* handle buffers that represented frags released by XDP prog; + * for these we keep pagecnt_bias as-is; refcount from struct page + * has been decremented within XDP prog and we do not have to increase + * the biased refcnt + */ + for (; i < nr_frags; i++) { + buf = &rx_ring->rx_buf[idx]; + ice_put_rx_buf(rx_ring, buf); + if (++idx == cnt) + idx = 0; + } xdp->data = NULL; rx_ring->first_desc = ntc; @@ -1190,9 +1203,9 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) struct ice_tx_ring *xdp_ring = NULL; struct bpf_prog *xdp_prog = NULL; u32 ntc = rx_ring->next_to_clean; + u32 cached_ntu, xdp_verdict; u32 cnt = rx_ring->count; u32 xdp_xmit = 0; - u32 cached_ntu; bool failure; xdp_prog = READ_ONCE(rx_ring->xdp_prog); @@ -1255,7 +1268,7 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) xdp_prepare_buff(xdp, hard_start, offset, size, !!offset); xdp_buff_clear_frags_flag(xdp); } else if (ice_add_xdp_frag(rx_ring, xdp, rx_buf, size)) { - ice_put_rx_mbuf(rx_ring, xdp, NULL, ntc); + ice_put_rx_mbuf(rx_ring, xdp, NULL, ntc, ICE_XDP_CONSUMED); break; } if (++ntc == cnt) @@ -1266,13 +1279,13 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) continue; ice_get_pgcnts(rx_ring); - ice_run_xdp(rx_ring, xdp, xdp_prog, xdp_ring, rx_buf, rx_desc); - if (rx_buf->act == ICE_XDP_PASS) + xdp_verdict = ice_run_xdp(rx_ring, xdp, xdp_prog, xdp_ring, rx_desc); + if (xdp_verdict == ICE_XDP_PASS) goto construct_skb; total_rx_bytes += xdp_get_buff_len(xdp); total_rx_pkts++; - ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc); + ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc, xdp_verdict); continue; construct_skb: @@ -1283,12 +1296,9 @@ int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget) /* exit if we failed to retrieve a buffer */ if (!skb) { rx_ring->ring_stats->rx_stats.alloc_page_failed++; - rx_buf->act = ICE_XDP_CONSUMED; - if (unlikely(xdp_buff_has_frags(xdp))) - ice_set_rx_bufs_act(xdp, rx_ring, - ICE_XDP_CONSUMED); + xdp_verdict = ICE_XDP_CONSUMED; } - ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc); + ice_put_rx_mbuf(rx_ring, xdp, &xdp_xmit, ntc, xdp_verdict); if (!skb) break; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index cb347c852ba9..806bce701df3 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -201,7 +201,6 @@ struct ice_rx_buf { struct page *page; unsigned int page_offset; unsigned int pgcnt; - unsigned int act; unsigned int pagecnt_bias; }; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h index 79f960c6680d..6cf32b404127 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h @@ -5,49 +5,6 @@ #define _ICE_TXRX_LIB_H_ #include "ice.h" -/** - * ice_set_rx_bufs_act - propagate Rx buffer action to frags - * @xdp: XDP buffer representing frame (linear and frags part) - * @rx_ring: Rx ring struct - * act: action to store onto Rx buffers related to XDP buffer parts - * - * Set action that should be taken before putting Rx buffer from first frag - * to the last. - */ -static inline void -ice_set_rx_bufs_act(struct xdp_buff *xdp, const struct ice_rx_ring *rx_ring, - const unsigned int act) -{ - u32 sinfo_frags = xdp_get_shared_info_from_buff(xdp)->nr_frags; - u32 nr_frags = rx_ring->nr_frags + 1; - u32 idx = rx_ring->first_desc; - u32 cnt = rx_ring->count; - struct ice_rx_buf *buf; - - for (int i = 0; i < nr_frags; i++) { - buf = &rx_ring->rx_buf[idx]; - buf->act = act; - - if (++idx == cnt) - idx = 0; - } - - /* adjust pagecnt_bias on frags freed by XDP prog */ - if (sinfo_frags < rx_ring->nr_frags && act == ICE_XDP_CONSUMED) { - u32 delta = rx_ring->nr_frags - sinfo_frags; - - while (delta) { - if (idx == 0) - idx = cnt - 1; - else - idx--; - buf = &rx_ring->rx_buf[idx]; - buf->pagecnt_bias--; - delta--; - } - } -} - /** * ice_test_staterr - tests bits in Rx descriptor status and error fields * @status_err_n: Rx descriptor status_error0 or status_error1 bits From 2d1a2dab95cdc6f2e0c6af3c0514b0bea94af482 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 28 Jan 2025 07:22:31 -0800 Subject: [PATCH 0307/1443] nvme: make nvme_tls_attrs_group static To suppress the compiler "warning: symbol 'nvme_tls_attrs_group' was not declared. Should it be static?" Fixes: 1e48b34c9bc79a ("nvme: split off TLS sysfs attributes into a separate group") Reviewed-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Keith Busch --- drivers/nvme/host/sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/sysfs.c b/drivers/nvme/host/sysfs.c index b68a9e5f1ea3..3a41b9ab0f13 100644 --- a/drivers/nvme/host/sysfs.c +++ b/drivers/nvme/host/sysfs.c @@ -792,7 +792,7 @@ static umode_t nvme_tls_attrs_are_visible(struct kobject *kobj, return a->mode; } -const struct attribute_group nvme_tls_attrs_group = { +static const struct attribute_group nvme_tls_attrs_group = { .attrs = nvme_tls_attrs, .is_visible = nvme_tls_attrs_are_visible, }; From c8ed6cb5d37bc09c7e25e49a670e9fd1a3bd1dfa Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Tue, 28 Jan 2025 17:34:47 +0100 Subject: [PATCH 0308/1443] nvme-fc: use ctrl state getter Do not access the state variable directly, instead use proper synchronization so not stale data is read. Fixes: e6e7f7ac03e4 ("nvme: ensure reset state check ordering") Signed-off-by: Daniel Wagner Reviewed-by: Christoph Hellwig Reviewed-by: Sagi Grimberg Signed-off-by: Keith Busch --- drivers/nvme/host/fc.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 55884d3df6f2..f4f1866fbd5b 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -2087,7 +2087,8 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req) nvme_fc_complete_rq(rq); check_error: - if (terminate_assoc && ctrl->ctrl.state != NVME_CTRL_RESETTING) + if (terminate_assoc && + nvme_ctrl_state(&ctrl->ctrl) != NVME_CTRL_RESETTING) queue_work(nvme_reset_wq, &ctrl->ioerr_work); } @@ -2541,6 +2542,8 @@ __nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues) static void nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) { + enum nvme_ctrl_state state = nvme_ctrl_state(&ctrl->ctrl); + /* * if an error (io timeout, etc) while (re)connecting, the remote * port requested terminating of the association (disconnect_ls) @@ -2548,7 +2551,7 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) * the controller. Abort any ios on the association and let the * create_association error path resolve things. */ - if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { + if (state == NVME_CTRL_CONNECTING) { __nvme_fc_abort_outstanding_ios(ctrl, true); dev_warn(ctrl->ctrl.device, "NVME-FC{%d}: transport error during (re)connect\n", @@ -2557,7 +2560,7 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) } /* Otherwise, only proceed if in LIVE state - e.g. on first error */ - if (ctrl->ctrl.state != NVME_CTRL_LIVE) + if (state != NVME_CTRL_LIVE) return; dev_warn(ctrl->ctrl.device, From a572593ac80e51eb69ecede7e614289fcccdbf8d Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 29 Jan 2025 14:56:35 -0800 Subject: [PATCH 0309/1443] md: Fix linear_set_limits() queue_limits_cancel_update() must only be called if queue_limits_start_update() is called first. Remove the queue_limits_cancel_update() call from linear_set_limits() because there is no corresponding queue_limits_start_update() call. This bug was discovered by annotating all mutex operations with clang thread-safety attributes and by building the kernel with clang and -Wthread-safety. Cc: Yu Kuai Cc: Coly Li Cc: Mike Snitzer Cc: Christoph Hellwig Fixes: 127186cfb184 ("md: reintroduce md-linear") Signed-off-by: Bart Van Assche Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250129225636.2667932-1-bvanassche@acm.org Signed-off-by: Song Liu --- drivers/md/md-linear.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/md/md-linear.c b/drivers/md/md-linear.c index a382929ce7ba..369aed044b40 100644 --- a/drivers/md/md-linear.c +++ b/drivers/md/md-linear.c @@ -76,10 +76,8 @@ static int linear_set_limits(struct mddev *mddev) lim.max_write_zeroes_sectors = mddev->chunk_sectors; lim.io_min = mddev->chunk_sectors << 9; err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY); - if (err) { - queue_limits_cancel_update(mddev->gendisk->queue); + if (err) return err; - } return queue_limits_set(mddev->gendisk->queue, &lim); } From 0f1a6c5c9784eff7e31e4915e17285fb89ad3644 Mon Sep 17 00:00:00 2001 From: Oliver Upton Date: Fri, 31 Jan 2025 22:29:24 +0000 Subject: [PATCH 0310/1443] KVM: arm64: Flush/sync debug state in protected mode The recent changes to debug state management broke self-hosted debug for guests when running in protected mode, since both the debug owner and the debug state itself aren't shared with the hyp's view of the vcpu. Fix it by flushing/syncing the relevant bits with the hyp vcpu. Fixes: beb470d96cec ("KVM: arm64: Use debug_owner to track if debug regs need save/restore") Reported-by: Mark Brown Closes: https://lore.kernel.org/kvmarm/5f62740f-a065-42d9-9f56-8fb648b9c63f@sirena.org.uk/ Signed-off-by: Oliver Upton Link: https://lore.kernel.org/r/20250131222922.1548780-3-oliver.upton@linux.dev Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/hyp-main.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c index 5c134520e180..6e12c070832f 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c @@ -91,11 +91,34 @@ static void fpsimd_sve_sync(struct kvm_vcpu *vcpu) *host_data_ptr(fp_owner) = FP_STATE_HOST_OWNED; } +static void flush_debug_state(struct pkvm_hyp_vcpu *hyp_vcpu) +{ + struct kvm_vcpu *host_vcpu = hyp_vcpu->host_vcpu; + + hyp_vcpu->vcpu.arch.debug_owner = host_vcpu->arch.debug_owner; + + if (kvm_guest_owns_debug_regs(&hyp_vcpu->vcpu)) + hyp_vcpu->vcpu.arch.vcpu_debug_state = host_vcpu->arch.vcpu_debug_state; + else if (kvm_host_owns_debug_regs(&hyp_vcpu->vcpu)) + hyp_vcpu->vcpu.arch.external_debug_state = host_vcpu->arch.external_debug_state; +} + +static void sync_debug_state(struct pkvm_hyp_vcpu *hyp_vcpu) +{ + struct kvm_vcpu *host_vcpu = hyp_vcpu->host_vcpu; + + if (kvm_guest_owns_debug_regs(&hyp_vcpu->vcpu)) + host_vcpu->arch.vcpu_debug_state = hyp_vcpu->vcpu.arch.vcpu_debug_state; + else if (kvm_host_owns_debug_regs(&hyp_vcpu->vcpu)) + host_vcpu->arch.external_debug_state = hyp_vcpu->vcpu.arch.external_debug_state; +} + static void flush_hyp_vcpu(struct pkvm_hyp_vcpu *hyp_vcpu) { struct kvm_vcpu *host_vcpu = hyp_vcpu->host_vcpu; fpsimd_sve_flush(); + flush_debug_state(hyp_vcpu); hyp_vcpu->vcpu.arch.ctxt = host_vcpu->arch.ctxt; @@ -123,6 +146,7 @@ static void sync_hyp_vcpu(struct pkvm_hyp_vcpu *hyp_vcpu) unsigned int i; fpsimd_sve_sync(&hyp_vcpu->vcpu); + sync_debug_state(hyp_vcpu); host_vcpu->arch.ctxt = hyp_vcpu->vcpu.arch.ctxt; From 46ded709232344b5750a852747a8881763c721ab Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 29 Jan 2025 15:13:42 -0800 Subject: [PATCH 0311/1443] net: bcmgenet: Correct overlaying of PHY and MAC Wake-on-LAN Some Wake-on-LAN modes such as WAKE_FILTER may only be supported by the MAC, while others might be only supported by the PHY. Make sure that the .get_wol() returns the union of both rather than only that of the PHY if the PHY supports Wake-on-LAN. When disabling Wake-on-LAN, make sure that this is done at both the PHY and MAC level, rather than doing an early return from the PHY driver. Fixes: 7e400ff35cbe ("net: bcmgenet: Add support for PHY-based Wake-on-LAN") Fixes: 9ee09edc05f2 ("net: bcmgenet: Properly overlay PHY and MAC Wake-on-LAN capabilities") Signed-off-by: Florian Fainelli Link: https://patch.msgid.link/20250129231342.35013-1-florian.fainelli@broadcom.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/broadcom/genet/bcmgenet_wol.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c index 0715ea5bf13e..3b082114f2e5 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c @@ -41,9 +41,12 @@ void bcmgenet_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct bcmgenet_priv *priv = netdev_priv(dev); struct device *kdev = &priv->pdev->dev; + u32 phy_wolopts = 0; - if (dev->phydev) + if (dev->phydev) { phy_ethtool_get_wol(dev->phydev, wol); + phy_wolopts = wol->wolopts; + } /* MAC is not wake-up capable, return what the PHY does */ if (!device_can_wakeup(kdev)) @@ -51,9 +54,14 @@ void bcmgenet_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) /* Overlay MAC capabilities with that of the PHY queried before */ wol->supported |= WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER; - wol->wolopts = priv->wolopts; - memset(wol->sopass, 0, sizeof(wol->sopass)); + wol->wolopts |= priv->wolopts; + /* Return the PHY configured magic password */ + if (phy_wolopts & WAKE_MAGICSECURE) + return; + + /* Otherwise the MAC one */ + memset(wol->sopass, 0, sizeof(wol->sopass)); if (wol->wolopts & WAKE_MAGICSECURE) memcpy(wol->sopass, priv->sopass, sizeof(priv->sopass)); } @@ -70,7 +78,7 @@ int bcmgenet_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) /* Try Wake-on-LAN from the PHY first */ if (dev->phydev) { ret = phy_ethtool_set_wol(dev->phydev, wol); - if (ret != -EOPNOTSUPP) + if (ret != -EOPNOTSUPP && wol->wolopts) return ret; } From c71a192976ded2f2f416d03c4f595cdd4478b825 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Jan 2025 19:15:18 -0800 Subject: [PATCH 0312/1443] net: ipv6: fix dst refleaks in rpl, seg6 and ioam6 lwtunnels dst_cache_get() gives us a reference, we need to release it. Discovered by the ioam6.sh test, kmemleak was recently fixed to catch per-cpu memory leaks. Fixes: 985ec6f5e623 ("net: ipv6: rpl_iptunnel: mitigate 2-realloc issue") Fixes: 40475b63761a ("net: ipv6: seg6_iptunnel: mitigate 2-realloc issue") Fixes: dce525185bc9 ("net: ipv6: ioam6_iptunnel: mitigate 2-realloc issue") Reviewed-by: Justin Iurman Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250130031519.2716843-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ipv6/ioam6_iptunnel.c | 5 +++-- net/ipv6/rpl_iptunnel.c | 6 ++++-- net/ipv6/seg6_iptunnel.c | 6 ++++-- 3 files changed, 11 insertions(+), 6 deletions(-) diff --git a/net/ipv6/ioam6_iptunnel.c b/net/ipv6/ioam6_iptunnel.c index 28e5a89dc255..3936c137a572 100644 --- a/net/ipv6/ioam6_iptunnel.c +++ b/net/ipv6/ioam6_iptunnel.c @@ -336,7 +336,7 @@ static int ioam6_do_encap(struct net *net, struct sk_buff *skb, static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) { - struct dst_entry *dst = skb_dst(skb), *cache_dst; + struct dst_entry *dst = skb_dst(skb), *cache_dst = NULL; struct in6_addr orig_daddr; struct ioam6_lwt *ilwt; int err = -EINVAL; @@ -407,7 +407,6 @@ static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) cache_dst = ip6_route_output(net, NULL, &fl6); if (cache_dst->error) { err = cache_dst->error; - dst_release(cache_dst); goto drop; } @@ -426,8 +425,10 @@ static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) return dst_output(net, sk, skb); } out: + dst_release(cache_dst); return dst->lwtstate->orig_output(net, sk, skb); drop: + dst_release(cache_dst); kfree_skb(skb); return err; } diff --git a/net/ipv6/rpl_iptunnel.c b/net/ipv6/rpl_iptunnel.c index 7ba22d2f2bfe..9b7d03563115 100644 --- a/net/ipv6/rpl_iptunnel.c +++ b/net/ipv6/rpl_iptunnel.c @@ -232,7 +232,6 @@ static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) dst = ip6_route_output(net, NULL, &fl6); if (dst->error) { err = dst->error; - dst_release(dst); goto drop; } @@ -251,6 +250,7 @@ static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) return dst_output(net, sk, skb); drop: + dst_release(dst); kfree_skb(skb); return err; } @@ -269,8 +269,10 @@ static int rpl_input(struct sk_buff *skb) local_bh_enable(); err = rpl_do_srh(skb, rlwt, dst); - if (unlikely(err)) + if (unlikely(err)) { + dst_release(dst); goto drop; + } if (!dst) { ip6_route_input(skb); diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 4bf937bfc263..eacc4e91b48e 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -482,8 +482,10 @@ static int seg6_input_core(struct net *net, struct sock *sk, local_bh_enable(); err = seg6_do_srh(skb, dst); - if (unlikely(err)) + if (unlikely(err)) { + dst_release(dst); goto drop; + } if (!dst) { ip6_route_input(skb); @@ -571,7 +573,6 @@ static int seg6_output_core(struct net *net, struct sock *sk, dst = ip6_route_output(net, NULL, &fl6); if (dst->error) { err = dst->error; - dst_release(dst); goto drop; } @@ -593,6 +594,7 @@ static int seg6_output_core(struct net *net, struct sock *sk, return dst_output(net, sk, skb); drop: + dst_release(dst); kfree_skb(skb); return err; } From 92191dd1073088753821b862b791dcc83e558e07 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Jan 2025 19:15:19 -0800 Subject: [PATCH 0313/1443] net: ipv6: fix dst ref loops in rpl, seg6 and ioam6 lwtunnels Some lwtunnels have a dst cache for post-transformation dst. If the packet destination did not change we may end up recording a reference to the lwtunnel in its own cache, and the lwtunnel state will never be freed. Discovered by the ioam6.sh test, kmemleak was recently fixed to catch per-cpu memory leaks. I'm not sure if rpl and seg6 can actually hit this, but in principle I don't see why not. Fixes: 8cb3bf8bff3c ("ipv6: ioam: Add support for the ip6ip6 encapsulation") Fixes: 6c8702c60b88 ("ipv6: sr: add support for SRH encapsulation and injection with lwtunnels") Fixes: a7a29f9c361f ("net: ipv6: add rpl sr tunnel") Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250130031519.2716843-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ipv6/ioam6_iptunnel.c | 9 ++++++--- net/ipv6/rpl_iptunnel.c | 9 ++++++--- net/ipv6/seg6_iptunnel.c | 9 ++++++--- 3 files changed, 18 insertions(+), 9 deletions(-) diff --git a/net/ipv6/ioam6_iptunnel.c b/net/ipv6/ioam6_iptunnel.c index 3936c137a572..2c383c12a431 100644 --- a/net/ipv6/ioam6_iptunnel.c +++ b/net/ipv6/ioam6_iptunnel.c @@ -410,9 +410,12 @@ static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) goto drop; } - local_bh_disable(); - dst_cache_set_ip6(&ilwt->cache, cache_dst, &fl6.saddr); - local_bh_enable(); + /* cache only if we don't create a dst reference loop */ + if (dst->lwtstate != cache_dst->lwtstate) { + local_bh_disable(); + dst_cache_set_ip6(&ilwt->cache, cache_dst, &fl6.saddr); + local_bh_enable(); + } err = skb_cow_head(skb, LL_RESERVED_SPACE(cache_dst->dev)); if (unlikely(err)) diff --git a/net/ipv6/rpl_iptunnel.c b/net/ipv6/rpl_iptunnel.c index 9b7d03563115..0ac4283acdf2 100644 --- a/net/ipv6/rpl_iptunnel.c +++ b/net/ipv6/rpl_iptunnel.c @@ -235,9 +235,12 @@ static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) goto drop; } - local_bh_disable(); - dst_cache_set_ip6(&rlwt->cache, dst, &fl6.saddr); - local_bh_enable(); + /* cache only if we don't create a dst reference loop */ + if (orig_dst->lwtstate != dst->lwtstate) { + local_bh_disable(); + dst_cache_set_ip6(&rlwt->cache, dst, &fl6.saddr); + local_bh_enable(); + } err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); if (unlikely(err)) diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index eacc4e91b48e..33833b2064c0 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -576,9 +576,12 @@ static int seg6_output_core(struct net *net, struct sock *sk, goto drop; } - local_bh_disable(); - dst_cache_set_ip6(&slwt->cache, dst, &fl6.saddr); - local_bh_enable(); + /* cache only if we don't create a dst reference loop */ + if (orig_dst->lwtstate != dst->lwtstate) { + local_bh_disable(); + dst_cache_set_ip6(&slwt->cache, dst, &fl6.saddr); + local_bh_enable(); + } err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); if (unlikely(err)) From a8aa6a6ddce9b5585f2b74f27f3feea1427fb4e7 Mon Sep 17 00:00:00 2001 From: Jiasheng Jiang Date: Fri, 31 Jan 2025 01:38:32 +0000 Subject: [PATCH 0314/1443] ice: Add check for devm_kzalloc() Add check for the return value of devm_kzalloc() to guarantee the success of allocation. Fixes: 42c2eb6b1f43 ("ice: Implement devlink-rate API") Signed-off-by: Jiasheng Jiang Reviewed-by: Michal Swiatkowski Link: https://patch.msgid.link/20250131013832.24805-1-jiashengjiangcool@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/devlink/devlink.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c index d116e2b10bce..dbdb83567364 100644 --- a/drivers/net/ethernet/intel/ice/devlink/devlink.c +++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c @@ -981,6 +981,9 @@ static int ice_devlink_rate_node_new(struct devlink_rate *rate_node, void **priv /* preallocate memory for ice_sched_node */ node = devm_kzalloc(ice_hw_to_dev(pi->hw), sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + *priv = node; return 0; From 3f1baa91a1fdf3de9dbad4bd615b35fab347874b Mon Sep 17 00:00:00 2001 From: Sankararaman Jayaraman Date: Fri, 31 Jan 2025 09:53:41 +0530 Subject: [PATCH 0315/1443] vmxnet3: Fix tx queue race condition with XDP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If XDP traffic runs on a CPU which is greater than or equal to the number of the Tx queues of the NIC, then vmxnet3_xdp_get_tq() always picks up queue 0 for transmission as it uses reciprocal scale instead of simple modulo operation. vmxnet3_xdp_xmit() and vmxnet3_xdp_xmit_frame() use the above returned queue without any locking which can lead to race conditions when multiple XDP xmits run in parallel on different CPU's. This patch uses a simple module scheme when the current CPU equals or exceeds the number of Tx queues on the NIC. It also adds locking in vmxnet3_xdp_xmit() and vmxnet3_xdp_xmit_frame() functions. Fixes: 54f00cce1178 ("vmxnet3: Add XDP support.") Signed-off-by: Sankararaman Jayaraman Signed-off-by: Ronak Doshi Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250131042340.156547-1-sankararaman.jayaraman@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/vmxnet3/vmxnet3_xdp.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/net/vmxnet3/vmxnet3_xdp.c b/drivers/net/vmxnet3/vmxnet3_xdp.c index 1341374a4588..616ecc38d172 100644 --- a/drivers/net/vmxnet3/vmxnet3_xdp.c +++ b/drivers/net/vmxnet3/vmxnet3_xdp.c @@ -28,7 +28,7 @@ vmxnet3_xdp_get_tq(struct vmxnet3_adapter *adapter) if (likely(cpu < tq_number)) tq = &adapter->tx_queue[cpu]; else - tq = &adapter->tx_queue[reciprocal_scale(cpu, tq_number)]; + tq = &adapter->tx_queue[cpu % tq_number]; return tq; } @@ -124,6 +124,7 @@ vmxnet3_xdp_xmit_frame(struct vmxnet3_adapter *adapter, u32 buf_size; u32 dw2; + spin_lock_irq(&tq->tx_lock); dw2 = (tq->tx_ring.gen ^ 0x1) << VMXNET3_TXD_GEN_SHIFT; dw2 |= xdpf->len; ctx.sop_txd = tq->tx_ring.base + tq->tx_ring.next2fill; @@ -134,6 +135,7 @@ vmxnet3_xdp_xmit_frame(struct vmxnet3_adapter *adapter, if (vmxnet3_cmd_ring_desc_avail(&tq->tx_ring) == 0) { tq->stats.tx_ring_full++; + spin_unlock_irq(&tq->tx_lock); return -ENOSPC; } @@ -142,8 +144,10 @@ vmxnet3_xdp_xmit_frame(struct vmxnet3_adapter *adapter, tbi->dma_addr = dma_map_single(&adapter->pdev->dev, xdpf->data, buf_size, DMA_TO_DEVICE); - if (dma_mapping_error(&adapter->pdev->dev, tbi->dma_addr)) + if (dma_mapping_error(&adapter->pdev->dev, tbi->dma_addr)) { + spin_unlock_irq(&tq->tx_lock); return -EFAULT; + } tbi->map_type |= VMXNET3_MAP_SINGLE; } else { /* XDP buffer from page pool */ page = virt_to_page(xdpf->data); @@ -182,6 +186,7 @@ vmxnet3_xdp_xmit_frame(struct vmxnet3_adapter *adapter, dma_wmb(); gdesc->dword[2] = cpu_to_le32(le32_to_cpu(gdesc->dword[2]) ^ VMXNET3_TXD_GEN); + spin_unlock_irq(&tq->tx_lock); /* No need to handle the case when tx_num_deferred doesn't reach * threshold. Backend driver at hypervisor side will poll and reset @@ -225,6 +230,7 @@ vmxnet3_xdp_xmit(struct net_device *dev, { struct vmxnet3_adapter *adapter = netdev_priv(dev); struct vmxnet3_tx_queue *tq; + struct netdev_queue *nq; int i; if (unlikely(test_bit(VMXNET3_STATE_BIT_QUIESCED, &adapter->state))) @@ -236,6 +242,9 @@ vmxnet3_xdp_xmit(struct net_device *dev, if (tq->stopped) return -ENETDOWN; + nq = netdev_get_tx_queue(adapter->netdev, tq->qid); + + __netif_tx_lock(nq, smp_processor_id()); for (i = 0; i < n; i++) { if (vmxnet3_xdp_xmit_frame(adapter, frames[i], tq, true)) { tq->stats.xdp_xmit_err++; @@ -243,6 +252,7 @@ vmxnet3_xdp_xmit(struct net_device *dev, } } tq->stats.xdp_xmit += i; + __netif_tx_unlock(nq); return i; } From 7faf14a7b0366f153284db0ad3347c457ea70136 Mon Sep 17 00:00:00 2001 From: Li Lingfeng Date: Sun, 26 Jan 2025 17:47:22 +0800 Subject: [PATCH 0316/1443] nfsd: clear acl_access/acl_default after releasing them If getting acl_default fails, acl_access and acl_default will be released simultaneously. However, acl_access will still retain a pointer pointing to the released posix_acl, which will trigger a WARNING in nfs3svc_release_getacl like this: ------------[ cut here ]------------ refcount_t: underflow; use-after-free. WARNING: CPU: 26 PID: 3199 at lib/refcount.c:28 refcount_warn_saturate+0xb5/0x170 Modules linked in: CPU: 26 UID: 0 PID: 3199 Comm: nfsd Not tainted 6.12.0-rc6-00079-g04ae226af01f-dirty #8 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.1-2.fc37 04/01/2014 RIP: 0010:refcount_warn_saturate+0xb5/0x170 Code: cc cc 0f b6 1d b3 20 a5 03 80 fb 01 0f 87 65 48 d8 00 83 e3 01 75 e4 48 c7 c7 c0 3b 9b 85 c6 05 97 20 a5 03 01 e8 fb 3e 30 ff <0f> 0b eb cd 0f b6 1d 8a3 RSP: 0018:ffffc90008637cd8 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff83904fde RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffff88871ed36380 RBP: ffff888158beeb40 R08: 0000000000000001 R09: fffff520010c6f56 R10: ffffc90008637ab7 R11: 0000000000000001 R12: 0000000000000001 R13: ffff888140e77400 R14: ffff888140e77408 R15: ffffffff858b42c0 FS: 0000000000000000(0000) GS:ffff88871ed00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000562384d32158 CR3: 000000055cc6a000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ? refcount_warn_saturate+0xb5/0x170 ? __warn+0xa5/0x140 ? refcount_warn_saturate+0xb5/0x170 ? report_bug+0x1b1/0x1e0 ? handle_bug+0x53/0xa0 ? exc_invalid_op+0x17/0x40 ? asm_exc_invalid_op+0x1a/0x20 ? tick_nohz_tick_stopped+0x1e/0x40 ? refcount_warn_saturate+0xb5/0x170 ? refcount_warn_saturate+0xb5/0x170 nfs3svc_release_getacl+0xc9/0xe0 svc_process_common+0x5db/0xb60 ? __pfx_svc_process_common+0x10/0x10 ? __rcu_read_unlock+0x69/0xa0 ? __pfx_nfsd_dispatch+0x10/0x10 ? svc_xprt_received+0xa1/0x120 ? xdr_init_decode+0x11d/0x190 svc_process+0x2a7/0x330 svc_handle_xprt+0x69d/0x940 svc_recv+0x180/0x2d0 nfsd+0x168/0x200 ? __pfx_nfsd+0x10/0x10 kthread+0x1a2/0x1e0 ? kthread+0xf4/0x1e0 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x34/0x60 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 Kernel panic - not syncing: kernel: panic_on_warn set ... Clear acl_access/acl_default after posix_acl_release is called to prevent UAF from being triggered. Fixes: a257cdd0e217 ("[PATCH] NFSD: Add server support for NFSv3 ACLs.") Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20241107014705.2509463-1-lilingfeng@huaweicloud.com/ Signed-off-by: Li Lingfeng Reviewed-by: Rick Macklem Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfs2acl.c | 2 ++ fs/nfsd/nfs3acl.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 4e3be7201b1c..5fb202acb0fd 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -84,6 +84,8 @@ static __be32 nfsacld_proc_getacl(struct svc_rqst *rqstp) fail: posix_acl_release(resp->acl_access); posix_acl_release(resp->acl_default); + resp->acl_access = NULL; + resp->acl_default = NULL; goto out; } diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index 5e34e98db969..7b5433bd3019 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -76,6 +76,8 @@ static __be32 nfsd3_proc_getacl(struct svc_rqst *rqstp) fail: posix_acl_release(resp->acl_access); posix_acl_release(resp->acl_default); + resp->acl_access = NULL; + resp->acl_default = NULL; goto out; } From b9382e29ca538b879645899ce45d652a304e2ed2 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Sat, 25 Jan 2025 20:13:18 -0500 Subject: [PATCH 0317/1443] nfsd: validate the nfsd_serv pointer before calling svc_wake_up nfsd_file_dispose_list_delayed can be called from the filecache laundrette, which is shut down after the nfsd threads are shut down and the nfsd_serv pointer is cleared. If nn->nfsd_serv is NULL then there are no threads to wake. Ensure that the nn->nfsd_serv pointer is non-NULL before calling svc_wake_up in nfsd_file_dispose_list_delayed. This is safe since the svc_serv is not freed until after the filecache laundrette is cancelled. Reported-by: Salvatore Bonaccorso Closes: https://bugs.debian.org/1093734 Fixes: ffb402596147 ("nfsd: Don't leave work of closing files to a work queue") Cc: stable@vger.kernel.org Signed-off-by: Jeff Layton Reviewed-by: NeilBrown Signed-off-by: Chuck Lever --- fs/nfsd/filecache.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index a1cdba42c4fa..78f4b5573b90 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -445,11 +445,20 @@ nfsd_file_dispose_list_delayed(struct list_head *dispose) struct nfsd_file, nf_gc); struct nfsd_net *nn = net_generic(nf->nf_net, nfsd_net_id); struct nfsd_fcache_disposal *l = nn->fcache_disposal; + struct svc_serv *serv; spin_lock(&l->lock); list_move_tail(&nf->nf_gc, &l->freeme); spin_unlock(&l->lock); - svc_wake_up(nn->nfsd_serv); + + /* + * The filecache laundrette is shut down after the + * nn->nfsd_serv pointer is cleared, but before the + * svc_serv is freed. + */ + serv = nn->nfsd_serv; + if (serv) + svc_wake_up(serv); } } From b69bb476dee99d564d65d418e9a20acca6f32c3f Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Thu, 30 Jan 2025 16:05:42 -0800 Subject: [PATCH 0318/1443] cgroup: fix race between fork and cgroup.kill MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Tejun reported the following race between fork() and cgroup.kill at [1]. Tejun: I was looking at cgroup.kill implementation and wondering whether there could be a race window. So, __cgroup_kill() does the following: k1. Set CGRP_KILL. k2. Iterate tasks and deliver SIGKILL. k3. Clear CGRP_KILL. The copy_process() does the following: c1. Copy a bunch of stuff. c2. Grab siglock. c3. Check fatal_signal_pending(). c4. Commit to forking. c5. Release siglock. c6. Call cgroup_post_fork() which puts the task on the css_set and tests CGRP_KILL. The intention seems to be that either a forking task gets SIGKILL and terminates on c3 or it sees CGRP_KILL on c6 and kills the child. However, I don't see what guarantees that k3 can't happen before c6. ie. After a forking task passes c5, k2 can take place and then before the forking task reaches c6, k3 can happen. Then, nobody would send SIGKILL to the child. What am I missing? This is indeed a race. One way to fix this race is by taking cgroup_threadgroup_rwsem in write mode in __cgroup_kill() as the fork() side takes cgroup_threadgroup_rwsem in read mode from cgroup_can_fork() to cgroup_post_fork(). However that would be heavy handed as this adds one more potential stall scenario for cgroup.kill which is usually called under extreme situation like memory pressure. To fix this race, let's maintain a sequence number per cgroup which gets incremented on __cgroup_kill() call. On the fork() side, the cgroup_can_fork() will cache the sequence number locally and recheck it against the cgroup's sequence number at cgroup_post_fork() site. If the sequence numbers mismatch, it means __cgroup_kill() can been called and we should send SIGKILL to the newly created task. Reported-by: Tejun Heo Closes: https://lore.kernel.org/all/Z5QHE2Qn-QZ6M-KW@slm.duckdns.org/ [1] Fixes: 661ee6280931 ("cgroup: introduce cgroup.kill") Cc: stable@vger.kernel.org # v5.14+ Signed-off-by: Shakeel Butt Reviewed-by: Michal Koutný Signed-off-by: Tejun Heo --- include/linux/cgroup-defs.h | 6 +++--- include/linux/sched/task.h | 1 + kernel/cgroup/cgroup.c | 20 ++++++++++++-------- 3 files changed, 16 insertions(+), 11 deletions(-) diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 1b20d2d8ef7c..17960a1e858d 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -71,9 +71,6 @@ enum { /* Cgroup is frozen. */ CGRP_FROZEN, - - /* Control group has to be killed. */ - CGRP_KILL, }; /* cgroup_root->flags */ @@ -461,6 +458,9 @@ struct cgroup { int nr_threaded_children; /* # of live threaded child cgroups */ + /* sequence number for cgroup.kill, serialized by css_set_lock. */ + unsigned int kill_seq; + struct kernfs_node *kn; /* cgroup kernfs entry */ struct cgroup_file procs_file; /* handle for "cgroup.procs" */ struct cgroup_file events_file; /* handle for "cgroup.events" */ diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h index 0f2aeb37bbb0..ca1db4b92c32 100644 --- a/include/linux/sched/task.h +++ b/include/linux/sched/task.h @@ -43,6 +43,7 @@ struct kernel_clone_args { void *fn_arg; struct cgroup *cgrp; struct css_set *cset; + unsigned int kill_seq; }; /* diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index d9061bd55436..afc665b7b1fe 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -4013,7 +4013,7 @@ static void __cgroup_kill(struct cgroup *cgrp) lockdep_assert_held(&cgroup_mutex); spin_lock_irq(&css_set_lock); - set_bit(CGRP_KILL, &cgrp->flags); + cgrp->kill_seq++; spin_unlock_irq(&css_set_lock); css_task_iter_start(&cgrp->self, CSS_TASK_ITER_PROCS | CSS_TASK_ITER_THREADED, &it); @@ -4029,10 +4029,6 @@ static void __cgroup_kill(struct cgroup *cgrp) send_sig(SIGKILL, task, 0); } css_task_iter_end(&it); - - spin_lock_irq(&css_set_lock); - clear_bit(CGRP_KILL, &cgrp->flags); - spin_unlock_irq(&css_set_lock); } static void cgroup_kill(struct cgroup *cgrp) @@ -6488,6 +6484,10 @@ static int cgroup_css_set_fork(struct kernel_clone_args *kargs) spin_lock_irq(&css_set_lock); cset = task_css_set(current); get_css_set(cset); + if (kargs->cgrp) + kargs->kill_seq = kargs->cgrp->kill_seq; + else + kargs->kill_seq = cset->dfl_cgrp->kill_seq; spin_unlock_irq(&css_set_lock); if (!(kargs->flags & CLONE_INTO_CGROUP)) { @@ -6668,6 +6668,7 @@ void cgroup_post_fork(struct task_struct *child, struct kernel_clone_args *kargs) __releases(&cgroup_threadgroup_rwsem) __releases(&cgroup_mutex) { + unsigned int cgrp_kill_seq = 0; unsigned long cgrp_flags = 0; bool kill = false; struct cgroup_subsys *ss; @@ -6681,10 +6682,13 @@ void cgroup_post_fork(struct task_struct *child, /* init tasks are special, only link regular threads */ if (likely(child->pid)) { - if (kargs->cgrp) + if (kargs->cgrp) { cgrp_flags = kargs->cgrp->flags; - else + cgrp_kill_seq = kargs->cgrp->kill_seq; + } else { cgrp_flags = cset->dfl_cgrp->flags; + cgrp_kill_seq = cset->dfl_cgrp->kill_seq; + } WARN_ON_ONCE(!list_empty(&child->cg_list)); cset->nr_tasks++; @@ -6719,7 +6723,7 @@ void cgroup_post_fork(struct task_struct *child, * child down right after we finished preparing it for * userspace. */ - kill = test_bit(CGRP_KILL, &cgrp_flags); + kill = kargs->kill_seq != cgrp_kill_seq; } spin_unlock_irq(&css_set_lock); From 029b6ce733712a41421955194b113f283dcb1026 Mon Sep 17 00:00:00 2001 From: Changwoo Min Date: Sun, 2 Feb 2025 12:37:48 +0900 Subject: [PATCH 0319/1443] sched_ext: Fix incorrect time delta calculation in time_delta() When (s64)(after - before) > 0, the code returns the result of (s64)(after - before) > 0 while the intended result should be (s64)(after - before). That happens because the middle operand of the ternary operator was omitted incorrectly, returning the result of (s64)(after - before) > 0. Thus, add the middle operand -- (s64)(after - before) -- to return the correct time calculation. Fixes: d07be814fc71 ("sched_ext: Add time helpers for BPF schedulers") Signed-off-by: Changwoo Min Acked-by: Andrea Righi Signed-off-by: Tejun Heo --- tools/sched_ext/include/scx/common.bpf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/include/scx/common.bpf.h index f254a39b86a5..d72b60a0c582 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -432,7 +432,7 @@ void bpf_rcu_read_unlock(void) __ksym; */ static inline s64 time_delta(u64 after, u64 before) { - return (s64)(after - before) > 0 ? : 0; + return (s64)(after - before) > 0 ? (s64)(after - before) : 0; } /** From a8c9a453387640dbe45761970f41301a6985e7fa Mon Sep 17 00:00:00 2001 From: Nikita Zhandarovich Date: Thu, 16 Jan 2025 06:24:36 -0800 Subject: [PATCH 0320/1443] ASoC: fsl_micfil: Enable default case in micfil_set_quality() If 'micfil->quality' received from micfil_quality_set() somehow ends up with an unpredictable value, switch() operator will fail to initialize local variable qsel before regmap_update_bits() tries to utilize it. While it is unlikely, play it safe and enable a default case that returns -EINVAL error. Found by Linux Verification Center (linuxtesting.org) with static analysis tool SVACE. Fixes: bea1d61d5892 ("ASoC: fsl_micfil: rework quality setting") Cc: stable@vger.kernel.org Signed-off-by: Nikita Zhandarovich Link: https://patch.msgid.link/20250116142436.22389-1-n.zhandarovich@fintech.ru Signed-off-by: Mark Brown --- sound/soc/fsl/fsl_micfil.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/soc/fsl/fsl_micfil.c b/sound/soc/fsl/fsl_micfil.c index 1075598a6647..fa4136683392 100644 --- a/sound/soc/fsl/fsl_micfil.c +++ b/sound/soc/fsl/fsl_micfil.c @@ -183,6 +183,8 @@ static int micfil_set_quality(struct fsl_micfil *micfil) case QUALITY_VLOW2: qsel = MICFIL_QSEL_VLOW2_QUALITY; break; + default: + return -EINVAL; } return regmap_update_bits(micfil->regmap, REG_MICFIL_CTRL2, From 64b48ec36dbed561ab1cd99708c33d96f4b7b729 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Mon, 3 Feb 2025 12:47:17 +1100 Subject: [PATCH 0321/1443] drivers/block/sunvdc.c: update the correct AIP call My sparc64 defconfig build failed like this: drivers/block/sunvdc.c: In function 'vdc_queue_drain': drivers/block/sunvdc.c:1130:9: error: too many arguments to function 'blk_mq_unquiesce_queue' 1130 | blk_mq_unquiesce_queue(q, memflags); | ^~~~~~~~~~~~~~~~~~~~~~ In file included from drivers/block/sunvdc.c:10: include/linux/blk-mq.h:895:6: note: declared here 895 | void blk_mq_unquiesce_queue(struct request_queue *q); | ^~~~~~~~~~~~~~~~~~~~~~ drivers/block/sunvdc.c:1131:9: error: too few arguments to function 'blk_mq_unfreeze_queue' 1131 | blk_mq_unfreeze_queue(q); | ^~~~~~~~~~~~~~~~~~~~~ In file included from drivers/block/sunvdc.c:10: include/linux/blk-mq.h:914:1: note: declared here 914 | blk_mq_unfreeze_queue(struct request_queue *q, unsigned int memflags) | ^~~~~~~~~~~~~~~~~~~~~ Fixes: 1e1a9cecfab3 ("block: force noio scope in blk_mq_freeze_queue") Cc: Christoph Hellwig Cc: Jens Axboe Signed-off-by: Stephen Rothwell Signed-off-by: Jens Axboe --- drivers/block/sunvdc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/block/sunvdc.c b/drivers/block/sunvdc.c index 05c4aee7f262..654ed962a772 100644 --- a/drivers/block/sunvdc.c +++ b/drivers/block/sunvdc.c @@ -1127,8 +1127,8 @@ static void vdc_queue_drain(struct vdc_port *port) spin_lock_irq(&port->vio.lock); port->drain = 0; - blk_mq_unquiesce_queue(q, memflags); - blk_mq_unfreeze_queue(q); + blk_mq_unquiesce_queue(q); + blk_mq_unfreeze_queue(q, memflags); } static void vdc_ldc_reset_timer_work(struct work_struct *work) From f0ada00a9b3801b71d203b0033b7612b687b7d72 Mon Sep 17 00:00:00 2001 From: Imran Shaik Date: Thu, 9 Jan 2025 14:27:44 +0530 Subject: [PATCH 0322/1443] dt-bindings: clock: qcom: Add GPU clocks for QCS8300 The QCS8300 GPU clock controller is a derivative of SA8775P, but has few additional clocks and minor differences. Hence, reuse gpucc bindings of SA8775P and add additional clocks required for QCS8300. Acked-by: Krzysztof Kozlowski Signed-off-by: Imran Shaik Link: https://lore.kernel.org/r/20250109-qcs8300-mm-patches-new-v4-1-63e8ac268b02@quicinc.com Signed-off-by: Rob Herring (Arm) --- .../devicetree/bindings/clock/qcom,gpucc.yaml | 3 +++ include/dt-bindings/clock/qcom,qcs8300-gpucc.h | 17 +++++++++++++++++ 2 files changed, 20 insertions(+) create mode 100644 include/dt-bindings/clock/qcom,qcs8300-gpucc.h diff --git a/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml index 0858fd635282..4cdff6161bf0 100644 --- a/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml +++ b/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml @@ -8,6 +8,7 @@ title: Qualcomm Graphics Clock & Reset Controller maintainers: - Taniya Das + - Imran Shaik description: | Qualcomm graphics clock control module provides the clocks, resets and power @@ -23,10 +24,12 @@ description: | include/dt-bindings/clock/qcom,gpucc-sm8150.h include/dt-bindings/clock/qcom,gpucc-sm8250.h include/dt-bindings/clock/qcom,gpucc-sm8350.h + include/dt-bindings/clock/qcom,qcs8300-gpucc.h properties: compatible: enum: + - qcom,qcs8300-gpucc - qcom,sdm845-gpucc - qcom,sa8775p-gpucc - qcom,sc7180-gpucc diff --git a/include/dt-bindings/clock/qcom,qcs8300-gpucc.h b/include/dt-bindings/clock/qcom,qcs8300-gpucc.h new file mode 100644 index 000000000000..afa187467b4c --- /dev/null +++ b/include/dt-bindings/clock/qcom,qcs8300-gpucc.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* + * Copyright (c) 2024, Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DT_BINDINGS_CLK_QCOM_GPUCC_QCS8300_H +#define _DT_BINDINGS_CLK_QCOM_GPUCC_QCS8300_H + +#include "qcom,sa8775p-gpucc.h" + +/* QCS8300 introduces below new clocks compared to SA8775P */ + +/* GPU_CC clocks */ +#define GPU_CC_CX_ACCU_SHIFT_CLK 23 +#define GPU_CC_GX_ACCU_SHIFT_CLK 24 + +#endif From 0e193cc558e32a879c717bb2d53a1cf8628b5d20 Mon Sep 17 00:00:00 2001 From: Imran Shaik Date: Thu, 9 Jan 2025 14:27:46 +0530 Subject: [PATCH 0323/1443] dt-bindings: clock: qcom: Add CAMCC clocks for QCS8300 The QCS8300 camera clock controller is a derivative of SA8775P, but has an additional clock and minor differences. Hence, reuse the SA8775P camera bindings and add additional clock required for QCS8300. Reviewed-by: Vladimir Zapolskiy Acked-by: Krzysztof Kozlowski Signed-off-by: Imran Shaik Link: https://lore.kernel.org/r/20250109-qcs8300-mm-patches-new-v4-3-63e8ac268b02@quicinc.com Signed-off-by: Rob Herring (Arm) --- .../bindings/clock/qcom,sa8775p-camcc.yaml | 6 +++++- include/dt-bindings/clock/qcom,qcs8300-camcc.h | 16 ++++++++++++++++ 2 files changed, 21 insertions(+), 1 deletion(-) create mode 100644 include/dt-bindings/clock/qcom,qcs8300-camcc.h diff --git a/Documentation/devicetree/bindings/clock/qcom,sa8775p-camcc.yaml b/Documentation/devicetree/bindings/clock/qcom,sa8775p-camcc.yaml index 36a60d8f5ae3..81623f59d11d 100644 --- a/Documentation/devicetree/bindings/clock/qcom,sa8775p-camcc.yaml +++ b/Documentation/devicetree/bindings/clock/qcom,sa8775p-camcc.yaml @@ -8,16 +8,20 @@ title: Qualcomm Camera Clock & Reset Controller on SA8775P maintainers: - Taniya Das + - Imran Shaik description: | Qualcomm camera clock control module provides the clocks, resets and power domains on SA8775p. - See also: include/dt-bindings/clock/qcom,sa8775p-camcc.h + See also: + include/dt-bindings/clock/qcom,qcs8300-camcc.h + include/dt-bindings/clock/qcom,sa8775p-camcc.h properties: compatible: enum: + - qcom,qcs8300-camcc - qcom,sa8775p-camcc clocks: diff --git a/include/dt-bindings/clock/qcom,qcs8300-camcc.h b/include/dt-bindings/clock/qcom,qcs8300-camcc.h new file mode 100644 index 000000000000..fc535c847859 --- /dev/null +++ b/include/dt-bindings/clock/qcom,qcs8300-camcc.h @@ -0,0 +1,16 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* + * Copyright (c) 2024, Qualcomm Innovation Center, Inc. All rights reserved. + */ + +#ifndef _DT_BINDINGS_CLK_QCOM_QCS8300_CAM_CC_H +#define _DT_BINDINGS_CLK_QCOM_QCS8300_CAM_CC_H + +#include "qcom,sa8775p-camcc.h" + +/* QCS8300 introduces below new clocks compared to SA8775P */ + +/* CAM_CC clocks */ +#define CAM_CC_TITAN_TOP_ACCU_SHIFT_CLK 86 + +#endif From 3e86e57356f0e2284454d82c7200807c6fa9e65b Mon Sep 17 00:00:00 2001 From: Imran Shaik Date: Thu, 9 Jan 2025 14:27:48 +0530 Subject: [PATCH 0324/1443] dt-bindings: clock: qcom: Add QCS8300 video clock controller The QCS8300 video clock controller is a derivative of SA8775P, but QCS8300 has minor difference. Hence, reuse the SA8775P videocc bindings for QCS8300 platform. Acked-by: Krzysztof Kozlowski Reviewed-by: Dmitry Baryshkov Signed-off-by: Imran Shaik Link: https://lore.kernel.org/r/20250109-qcs8300-mm-patches-new-v4-5-63e8ac268b02@quicinc.com Signed-off-by: Rob Herring (Arm) --- .../devicetree/bindings/clock/qcom,sa8775p-videocc.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/clock/qcom,sa8775p-videocc.yaml b/Documentation/devicetree/bindings/clock/qcom,sa8775p-videocc.yaml index 928131bff4c1..07e5d811d816 100644 --- a/Documentation/devicetree/bindings/clock/qcom,sa8775p-videocc.yaml +++ b/Documentation/devicetree/bindings/clock/qcom,sa8775p-videocc.yaml @@ -18,6 +18,7 @@ description: | properties: compatible: enum: + - qcom,qcs8300-videocc - qcom,sa8775p-videocc clocks: From e6649328dc07bff6227367eda6f1b2263d6c10f0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Wed, 29 Jan 2025 14:35:27 +0100 Subject: [PATCH 0325/1443] of: address: Add kunit test for __of_address_resource_bounds() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The overflow checking has to deal with different datatypes and edgecases. Add a new kunit testcase to make sure it works correctly. Signed-off-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20250129-of-address-overflow-v3-1-95d1760ed791@linutronix.de Signed-off-by: Rob Herring (Arm) --- drivers/of/address.c | 5 +- drivers/of/of_private.h | 4 ++ drivers/of/of_test.c | 119 +++++++++++++++++++++++++++++++++++++++- 3 files changed, 126 insertions(+), 2 deletions(-) diff --git a/drivers/of/address.c b/drivers/of/address.c index 125833e5ce52..d177a2b9edaf 100644 --- a/drivers/of/address.c +++ b/drivers/of/address.c @@ -16,6 +16,8 @@ #include #include /* for bus_dma_region */ +#include + /* Uncomment me to enable of_dump_addr() debugging output */ // #define DEBUG @@ -183,7 +185,7 @@ static u64 of_bus_pci_map(__be32 *addr, const __be32 *range, int na, int ns, #endif /* CONFIG_PCI */ -static int __of_address_resource_bounds(struct resource *r, u64 start, u64 size) +VISIBLE_IF_KUNIT int __of_address_resource_bounds(struct resource *r, u64 start, u64 size) { if (overflows_type(start, r->start)) return -EOVERFLOW; @@ -197,6 +199,7 @@ static int __of_address_resource_bounds(struct resource *r, u64 start, u64 size) return 0; } +EXPORT_SYMBOL_IF_KUNIT(__of_address_resource_bounds); /* * of_pci_range_to_resource - Create a resource from an of_pci_range diff --git a/drivers/of/of_private.h b/drivers/of/of_private.h index f3e1193c8ded..1bdc7ceef3c5 100644 --- a/drivers/of/of_private.h +++ b/drivers/of/of_private.h @@ -208,4 +208,8 @@ static void __maybe_unused of_dump_addr(const char *s, const __be32 *addr, int n static void __maybe_unused of_dump_addr(const char *s, const __be32 *addr, int na) { } #endif +#if IS_ENABLED(CONFIG_KUNIT) +int __of_address_resource_bounds(struct resource *r, u64 start, u64 size); +#endif + #endif /* _LINUX_OF_PRIVATE_H */ diff --git a/drivers/of/of_test.c b/drivers/of/of_test.c index b0557ded838f..8bba5a72c9c7 100644 --- a/drivers/of/of_test.c +++ b/drivers/of/of_test.c @@ -2,6 +2,7 @@ /* * KUnit tests for OF APIs */ +#include #include #include @@ -54,8 +55,124 @@ static struct kunit_suite of_dtb_suite = { .init = of_dtb_test_init, }; +struct of_address_resource_bounds_case { + u64 start; + u64 size; + int ret; + + u64 res_start; + u64 res_end; +}; + +static void of_address_resource_bounds_case_desc(const struct of_address_resource_bounds_case *p, + char *name) +{ + snprintf(name, KUNIT_PARAM_DESC_SIZE, "start=0x%016llx,size=0x%016llx", p->start, p->size); +} + +static const struct of_address_resource_bounds_case of_address_resource_bounds_cases[] = { + { + .start = 0, + .size = 0, + .ret = 0, + .res_start = 0, + .res_end = -1, + }, + { + .start = 0, + .size = 0x1000, + .ret = 0, + .res_start = 0, + .res_end = 0xfff, + }, + { + .start = 0x1000, + .size = 0, + .ret = 0, + .res_start = 0x1000, + .res_end = 0xfff, + }, + { + .start = 0x1000, + .size = 0x1000, + .ret = 0, + .res_start = 0x1000, + .res_end = 0x1fff, + }, + { + .start = 1, + .size = RESOURCE_SIZE_MAX, + .ret = 0, + .res_start = 1, + .res_end = RESOURCE_SIZE_MAX, + }, + { + .start = RESOURCE_SIZE_MAX, + .size = 1, + .ret = 0, + .res_start = RESOURCE_SIZE_MAX, + .res_end = RESOURCE_SIZE_MAX, + }, + { + .start = 2, + .size = RESOURCE_SIZE_MAX, + .ret = -EOVERFLOW, + }, + { + .start = RESOURCE_SIZE_MAX, + .size = 2, + .ret = -EOVERFLOW, + }, + { + .start = ULL(0x100000000), + .size = 1, + .ret = sizeof(resource_size_t) > sizeof(u32) ? 0 : -EOVERFLOW, + .res_start = ULL(0x100000000), + .res_end = ULL(0x100000000), + }, + { + .start = 0x1000, + .size = 0xffffffff, + .ret = sizeof(resource_size_t) > sizeof(u32) ? 0 : -EOVERFLOW, + .res_start = 0x1000, + .res_end = ULL(0x100000ffe), + }, +}; + +KUNIT_ARRAY_PARAM(of_address_resource_bounds, + of_address_resource_bounds_cases, of_address_resource_bounds_case_desc); + +static void of_address_resource_bounds(struct kunit *test) +{ + const struct of_address_resource_bounds_case *param = test->param_value; + struct resource r; /* Intentionally uninitialized */ + int ret; + + if (!IS_ENABLED(CONFIG_OF_ADDRESS)) + kunit_skip(test, "CONFIG_OF_ADDRESS not enabled\n"); + + ret = __of_address_resource_bounds(&r, param->start, param->size); + KUNIT_EXPECT_EQ(test, param->ret, ret); + if (ret == 0) { + KUNIT_EXPECT_EQ(test, (resource_size_t)param->res_start, r.start); + KUNIT_EXPECT_EQ(test, (resource_size_t)param->res_end, r.end); + KUNIT_EXPECT_EQ(test, param->size, resource_size(&r)); + } +} + +static struct kunit_case of_address_test_cases[] = { + KUNIT_CASE_PARAM(of_address_resource_bounds, of_address_resource_bounds_gen_params), + {} +}; + +static struct kunit_suite of_address_suite = { + .name = "of_address", + .test_cases = of_address_test_cases, +}; + kunit_test_suites( - &of_dtb_suite, + &of_dtb_suite, &of_address_suite, ); MODULE_DESCRIPTION("KUnit tests for OF APIs"); +MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING"); MODULE_LICENSE("GPL"); From fa803513ab68ba07369643393f1754b845160030 Mon Sep 17 00:00:00 2001 From: Lifeng Zheng Date: Fri, 10 Jan 2025 17:19:49 +0800 Subject: [PATCH 0326/1443] cpufreq/amd-pstate: Fix per-policy boost flag incorrect when fail Commit c8c68c38b56f ("cpufreq: amd-pstate: initialize core precision boost state") sets per-policy boost flag to false when boost fail. However, this boost flag will be set to reverse value in store_local_boost() and cpufreq_boost_trigger_state() in cpufreq.c. This will cause the per-policy boost flag set to true when fail to set boost. Remove the extra assignment in amd_pstate_set_boost() and keep all operations on per-policy boost flag outside of set_boost() to fix this problem. Fixes: c8c68c38b56f ("cpufreq: amd-pstate: initialize core precision boost state") Signed-off-by: Lifeng Zheng Reviewed-by: Mario Limonciello Link: https://lore.kernel.org/r/20250110091949.3610770-1-zhenglifeng1@huawei.com Signed-off-by: Mario Limonciello --- drivers/cpufreq/amd-pstate.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index dd9b8d6993d6..7120f035c0be 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -747,7 +747,6 @@ static int amd_pstate_set_boost(struct cpufreq_policy *policy, int state) guard(mutex)(&amd_pstate_driver_lock); ret = amd_pstate_cpu_boost_update(policy, state); - policy->boost_enabled = !ret ? state : false; refresh_frequency_limits(policy); return ret; From e4d4648eac8b4ef39f412d07715eb26f1ccd7342 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 28 Jan 2025 00:02:01 +0300 Subject: [PATCH 0327/1443] platform/x86: ideapad-laptop: pass a correct pointer to the driver data MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit devm_platform_profile_register() expects a pointer to the private driver data but instead an address of the pointer variable is passed due to a typo. This leads to the crashes later: BUG: unable to handle page fault for address: 00000000fe0d0044 PGD 0 P4D 0 Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI CPU: 6 UID: 0 PID: 1284 Comm: tuned Tainted: G W 6.13.0+ #7 Tainted: [W]=WARN Hardware name: LENOVO 21D0/LNVNB161216, BIOS J6CN45WW 03/17/2023 RIP: 0010:__mutex_lock.constprop.0+0x6bf/0x7f0 Call Trace: dytc_profile_set+0x4a/0x140 [ideapad_laptop] _store_and_notify+0x13/0x40 [platform_profile] class_for_each_device+0x145/0x180 platform_profile_store+0xc0/0x130 [platform_profile] kernfs_fop_write_iter+0x13e/0x1f0 vfs_write+0x290/0x450 ksys_write+0x6c/0xe0 do_syscall_64+0x82/0x160 entry_SYSCALL_64_after_hwframe+0x76/0x7e Found by Linux Verification Center (linuxtesting.org). Fixes: 249c576f0f9d ("ACPI: platform_profile: Let drivers set drvdata to the class device") Signed-off-by: Fedor Pchelkin Reviewed-by: Kurt Borja Link: https://lore.kernel.org/r/20250127210202.568691-1-pchelkin@ispras.ru Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/ideapad-laptop.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index dfb5d4b8c046..30bd366d7b58 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -1121,7 +1121,7 @@ static int ideapad_dytc_profile_init(struct ideapad_private *priv) /* Create platform_profile structure and register */ priv->dytc->ppdev = devm_platform_profile_register(&priv->platform_device->dev, - "ideapad-laptop", &priv->dytc, + "ideapad-laptop", priv->dytc, &dytc_profile_ops); if (IS_ERR(priv->dytc->ppdev)) { err = PTR_ERR(priv->dytc->ppdev); From 583ef25bb2a094813351a727ddec38b35a15b9f8 Mon Sep 17 00:00:00 2001 From: Dmitry Kandybka Date: Fri, 24 Jan 2025 01:07:39 +0300 Subject: [PATCH 0328/1443] platform/x86/intel: pmc: fix ltr decode in pmc_core_ltr_show() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In pmc_core_ltr_show(), promote 'val' to 'u64' to avoid possible integer overflow. Values (10 bit) are multiplied by the scale, the result of expression is in a range from 1 to 34,326,183,936 which is bigger then UINT32_MAX. Compile tested only. Found by Linux Verification Center (linuxtesting.org) with SVACE. Signed-off-by: Dmitry Kandybka Reviewed-by: Rajneesh Bhardwaj Reviewed-by: Ilpo Järvinen Link: https://lore.kernel.org/r/20250123220739.68087-1-d.kandybka@gmail.com Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/intel/pmc/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/intel/pmc/core.c b/drivers/platform/x86/intel/pmc/core.c index 10f04b944117..1ee0fb5f8250 100644 --- a/drivers/platform/x86/intel/pmc/core.c +++ b/drivers/platform/x86/intel/pmc/core.c @@ -626,8 +626,8 @@ static u32 convert_ltr_scale(u32 val) static int pmc_core_ltr_show(struct seq_file *s, void *unused) { struct pmc_dev *pmcdev = s->private; - u64 decoded_snoop_ltr, decoded_non_snoop_ltr; - u32 ltr_raw_data, scale, val; + u64 decoded_snoop_ltr, decoded_non_snoop_ltr, val; + u32 ltr_raw_data, scale; u16 snoop_ltr, nonsnoop_ltr; unsigned int i, index, ltr_index = 0; From 5c8f9a05336cf5cadbd57ad461621b386aadb762 Mon Sep 17 00:00:00 2001 From: Alexander Shiyan Date: Thu, 30 Jan 2025 08:38:49 +0300 Subject: [PATCH 0329/1443] arm64: dts: rockchip: Fix broken tsadc pinctrl names for rk3588 The tsadc driver does not handle pinctrl "gpio" and "otpout". Let's use the correct pinctrl names "default" and "sleep". Additionally, Alexey Charkov's testing [1] has established that it is necessary for pinctrl state to reference the &tsadc_shut_org configuration rather than &tsadc_shut for the driver to function correctly. [1] https://lkml.org/lkml/2025/1/24/966 Fixes: 32641b8ab1a5 ("arm64: dts: rockchip: add rk3588 thermal sensor") Cc: stable@vger.kernel.org Reviewed-by: Dragan Simic Signed-off-by: Alexander Shiyan Link: https://lore.kernel.org/r/20250130053849.4902-1-eagle.alexander923@gmail.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3588-base.dtsi | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi b/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi index 8cfa30837ce7..978de506d434 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588-base.dtsi @@ -2668,9 +2668,9 @@ tsadc: tsadc@fec00000 { rockchip,hw-tshut-temp = <120000>; rockchip,hw-tshut-mode = <0>; /* tshut mode 0:CRU 1:GPIO */ rockchip,hw-tshut-polarity = <0>; /* tshut polarity 0:LOW 1:HIGH */ - pinctrl-0 = <&tsadc_gpio_func>; - pinctrl-1 = <&tsadc_shut>; - pinctrl-names = "gpio", "otpout"; + pinctrl-0 = <&tsadc_shut_org>; + pinctrl-1 = <&tsadc_gpio_func>; + pinctrl-names = "default", "sleep"; #thermal-sensor-cells = <1>; status = "disabled"; }; From a6a7cba17c544fb95d5a29ab9d9ed4503029cb29 Mon Sep 17 00:00:00 2001 From: Tianling Shen Date: Sun, 19 Jan 2025 17:11:54 +0800 Subject: [PATCH 0330/1443] arm64: dts: rockchip: change eth phy mode to rgmii-id for orangepi r1 plus lts In general the delay should be added by the PHY instead of the MAC, and this improves network stability on some boards which seem to need different delay. Fixes: 387b3bbac5ea ("arm64: dts: rockchip: Add Xunlong OrangePi R1 Plus LTS") Cc: stable@vger.kernel.org # 6.6+ Signed-off-by: Tianling Shen Link: https://lore.kernel.org/r/20250119091154.1110762-1-cnsztl@gmail.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus-lts.dts | 3 +-- arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dts | 1 + arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dtsi | 1 - 3 files changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus-lts.dts b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus-lts.dts index 67c246ad8b8c..ec2ce894da1f 100644 --- a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus-lts.dts +++ b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus-lts.dts @@ -17,8 +17,7 @@ / { &gmac2io { phy-handle = <&yt8531c>; - tx_delay = <0x19>; - rx_delay = <0x05>; + phy-mode = "rgmii-id"; status = "okay"; mdio { diff --git a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dts b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dts index 324a8e951f7e..846b931e16d2 100644 --- a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dts +++ b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dts @@ -15,6 +15,7 @@ / { &gmac2io { phy-handle = <&rtl8211e>; + phy-mode = "rgmii"; tx_delay = <0x24>; rx_delay = <0x18>; status = "okay"; diff --git a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dtsi b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dtsi index 4f193704e5dc..09508e324a28 100644 --- a/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3328-orangepi-r1-plus.dtsi @@ -109,7 +109,6 @@ &gmac2io { assigned-clocks = <&cru SCLK_MAC2IO>, <&cru SCLK_MAC2IO_EXT>; assigned-clock-parents = <&gmac_clk>, <&gmac_clk>; clock_in_out = "input"; - phy-mode = "rgmii"; phy-supply = <&vcc_io>; pinctrl-0 = <&rgmiim1_pins>; pinctrl-names = "default"; From 4eee627ea59304cdd66c5d4194ef13486a6c44fc Mon Sep 17 00:00:00 2001 From: Lukasz Czechowski Date: Tue, 21 Jan 2025 13:56:03 +0100 Subject: [PATCH 0331/1443] arm64: dts: rockchip: Move uart5 pin configuration to px30 ringneck SoM In the PX30-uQ7 (Ringneck) SoM, the hardware CTS and RTS pins for uart5 cannot be used for the UART CTS/RTS, because they are already allocated for different purposes. CTS pin is routed to SUS_S3# signal, while RTS pin is used internally and is not available on Q7 connector. Move definition of the pinctrl-0 property from px30-ringneck-haikou.dts to px30-ringneck.dtsi. This commit is a dependency to next commit in the patch series, that disables DMA for uart5. Cc: stable@vger.kernel.org Reviewed-by: Quentin Schulz Signed-off-by: Lukasz Czechowski Link: https://lore.kernel.org/r/20250121125604.3115235-2-lukasz.czechowski@thaumatec.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/px30-ringneck-haikou.dts | 1 - arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi | 4 ++++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/rockchip/px30-ringneck-haikou.dts b/arch/arm64/boot/dts/rockchip/px30-ringneck-haikou.dts index e4517f47d519..eb9470a00e54 100644 --- a/arch/arm64/boot/dts/rockchip/px30-ringneck-haikou.dts +++ b/arch/arm64/boot/dts/rockchip/px30-ringneck-haikou.dts @@ -226,7 +226,6 @@ &uart0 { }; &uart5 { - pinctrl-0 = <&uart5_xfer>; rts-gpios = <&gpio0 RK_PB5 GPIO_ACTIVE_HIGH>; status = "okay"; }; diff --git a/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi b/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi index ae050cc6cd05..2c87005c89bd 100644 --- a/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi +++ b/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi @@ -396,6 +396,10 @@ &u2phy_host { status = "okay"; }; +&uart5 { + pinctrl-0 = <&uart5_xfer>; +}; + /* Mule UCAN */ &usb_host0_ehci { status = "okay"; From f3be8a9b1afffbcc70f8e41063b151b1038d7813 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Wed, 29 Jan 2025 13:40:07 +0100 Subject: [PATCH 0332/1443] accel/ivpu: Fix error handling in ivpu_boot() Ensure IRQs and IPC are properly disabled if HW sched or DCT initialization fails. Fixes: cc3c72c7e610 ("accel/ivpu: Refactor failure diagnostics during boot") Cc: stable@vger.kernel.org # v6.13+ Reviewed-by: Karol Wachowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129124009.1039982-2-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index ca2bf47ce248..0c4a82271c26 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -397,15 +397,19 @@ int ivpu_boot(struct ivpu_device *vdev) if (ivpu_fw_is_cold_boot(vdev)) { ret = ivpu_pm_dct_init(vdev); if (ret) - goto err_diagnose_failure; + goto err_disable_ipc; ret = ivpu_hw_sched_init(vdev); if (ret) - goto err_diagnose_failure; + goto err_disable_ipc; } return 0; +err_disable_ipc: + ivpu_ipc_disable(vdev); + ivpu_hw_irq_disable(vdev); + disable_irq(vdev->irq); err_diagnose_failure: ivpu_hw_diagnose_failure(vdev); ivpu_mmu_evtq_dump(vdev); From f2bc2afe34c107a02ce829a4039e85514feafe55 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Wed, 29 Jan 2025 13:40:08 +0100 Subject: [PATCH 0333/1443] accel/ivpu: Clear runtime_error after pm_runtime_resume_and_get() fails pm_runtime_resume_and_get() sets dev->power.runtime_error that causes all subsequent pm_runtime_get_sync() calls to fail. Clear the runtime_error using pm_runtime_set_suspended(), so the driver doesn't have to be reloaded to recover when the NPU fails to boot during runtime resume. Fixes: 7d4b4c74432d ("accel/ivpu: Remove suspend_reschedule_counter") Cc: stable@vger.kernel.org # v6.11+ Reviewed-by: Maciej Falkowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129124009.1039982-3-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_pm.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index 949f4233946c..c3774d222132 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -309,7 +309,10 @@ int ivpu_rpm_get(struct ivpu_device *vdev) int ret; ret = pm_runtime_resume_and_get(vdev->drm.dev); - drm_WARN_ON(&vdev->drm, ret < 0); + if (ret < 0) { + ivpu_err(vdev, "Failed to resume NPU: %d\n", ret); + pm_runtime_set_suspended(vdev->drm.dev); + } return ret; } From 41a2d8286c905614f29007f1bc8e652d54654b82 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Wed, 29 Jan 2025 13:40:09 +0100 Subject: [PATCH 0334/1443] accel/ivpu: Fix error handling in recovery/reset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Disable runtime PM for the duration of reset/recovery so it is possible to set the correct runtime PM state depending on the outcome of the `ivpu_resume()`. Don’t suspend or reset the HW if the NPU is suspended when the reset/recovery is requested. Also, move common reset/recovery code to separate functions for better code readability. Fixes: 27d19268cf39 ("accel/ivpu: Improve recovery and reset support") Cc: stable@vger.kernel.org # v6.8+ Reviewed-by: Maciej Falkowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129124009.1039982-4-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_pm.c | 79 ++++++++++++++++++++---------------- 1 file changed, 43 insertions(+), 36 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index c3774d222132..8b2b050cc41a 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -115,41 +115,57 @@ static int ivpu_resume(struct ivpu_device *vdev) return ret; } -static void ivpu_pm_recovery_work(struct work_struct *work) +static void ivpu_pm_reset_begin(struct ivpu_device *vdev) { - struct ivpu_pm_info *pm = container_of(work, struct ivpu_pm_info, recovery_work); - struct ivpu_device *vdev = pm->vdev; - char *evt[2] = {"IVPU_PM_EVENT=IVPU_RECOVER", NULL}; - int ret; - - ivpu_err(vdev, "Recovering the NPU (reset #%d)\n", atomic_read(&vdev->pm->reset_counter)); - - ret = pm_runtime_resume_and_get(vdev->drm.dev); - if (ret) - ivpu_err(vdev, "Failed to resume NPU: %d\n", ret); - - ivpu_jsm_state_dump(vdev); - ivpu_dev_coredump(vdev); + pm_runtime_disable(vdev->drm.dev); atomic_inc(&vdev->pm->reset_counter); atomic_set(&vdev->pm->reset_pending, 1); down_write(&vdev->pm->reset_lock); +} + +static void ivpu_pm_reset_complete(struct ivpu_device *vdev) +{ + int ret; - ivpu_suspend(vdev); ivpu_pm_prepare_cold_boot(vdev); ivpu_jobs_abort_all(vdev); ivpu_ms_cleanup_all(vdev); ret = ivpu_resume(vdev); - if (ret) + if (ret) { ivpu_err(vdev, "Failed to resume NPU: %d\n", ret); + pm_runtime_set_suspended(vdev->drm.dev); + } else { + pm_runtime_set_active(vdev->drm.dev); + } up_write(&vdev->pm->reset_lock); atomic_set(&vdev->pm->reset_pending, 0); - kobject_uevent_env(&vdev->drm.dev->kobj, KOBJ_CHANGE, evt); pm_runtime_mark_last_busy(vdev->drm.dev); - pm_runtime_put_autosuspend(vdev->drm.dev); + pm_runtime_enable(vdev->drm.dev); +} + +static void ivpu_pm_recovery_work(struct work_struct *work) +{ + struct ivpu_pm_info *pm = container_of(work, struct ivpu_pm_info, recovery_work); + struct ivpu_device *vdev = pm->vdev; + char *evt[2] = {"IVPU_PM_EVENT=IVPU_RECOVER", NULL}; + + ivpu_err(vdev, "Recovering the NPU (reset #%d)\n", atomic_read(&vdev->pm->reset_counter)); + + ivpu_pm_reset_begin(vdev); + + if (!pm_runtime_status_suspended(vdev->drm.dev)) { + ivpu_jsm_state_dump(vdev); + ivpu_dev_coredump(vdev); + ivpu_suspend(vdev); + } + + ivpu_pm_reset_complete(vdev); + + kobject_uevent_env(&vdev->drm.dev->kobj, KOBJ_CHANGE, evt); } void ivpu_pm_trigger_recovery(struct ivpu_device *vdev, const char *reason) @@ -328,16 +344,13 @@ void ivpu_pm_reset_prepare_cb(struct pci_dev *pdev) struct ivpu_device *vdev = pci_get_drvdata(pdev); ivpu_dbg(vdev, PM, "Pre-reset..\n"); - atomic_inc(&vdev->pm->reset_counter); - atomic_set(&vdev->pm->reset_pending, 1); - pm_runtime_get_sync(vdev->drm.dev); - down_write(&vdev->pm->reset_lock); - ivpu_prepare_for_reset(vdev); - ivpu_hw_reset(vdev); - ivpu_pm_prepare_cold_boot(vdev); - ivpu_jobs_abort_all(vdev); - ivpu_ms_cleanup_all(vdev); + ivpu_pm_reset_begin(vdev); + + if (!pm_runtime_status_suspended(vdev->drm.dev)) { + ivpu_prepare_for_reset(vdev); + ivpu_hw_reset(vdev); + } ivpu_dbg(vdev, PM, "Pre-reset done.\n"); } @@ -345,18 +358,12 @@ void ivpu_pm_reset_prepare_cb(struct pci_dev *pdev) void ivpu_pm_reset_done_cb(struct pci_dev *pdev) { struct ivpu_device *vdev = pci_get_drvdata(pdev); - int ret; ivpu_dbg(vdev, PM, "Post-reset..\n"); - ret = ivpu_resume(vdev); - if (ret) - ivpu_err(vdev, "Failed to set RESUME state: %d\n", ret); - up_write(&vdev->pm->reset_lock); - atomic_set(&vdev->pm->reset_pending, 0); - ivpu_dbg(vdev, PM, "Post-reset done.\n"); - pm_runtime_mark_last_busy(vdev->drm.dev); - pm_runtime_put_autosuspend(vdev->drm.dev); + ivpu_pm_reset_complete(vdev); + + ivpu_dbg(vdev, PM, "Post-reset done.\n"); } void ivpu_pm_init(struct ivpu_device *vdev) From 5ae4dca718eacd0a56173a687a3736eb7e627c77 Mon Sep 17 00:00:00 2001 From: Lukasz Czechowski Date: Tue, 21 Jan 2025 13:56:04 +0100 Subject: [PATCH 0335/1443] arm64: dts: rockchip: Disable DMA for uart5 on px30-ringneck MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit UART controllers without flow control seem to behave unstable in case DMA is enabled. The issues were indicated in the message: https://lore.kernel.org/linux-arm-kernel/CAMdYzYpXtMocCtCpZLU_xuWmOp2Ja_v0Aj0e6YFNRA-yV7u14g@mail.gmail.com/ In case of PX30-uQ7 Ringneck SoM, it was noticed that after couple of hours of UART communication, the CPU stall was occurring, leading to the system becoming unresponsive. After disabling the DMA, extensive UART communication tests for up to two weeks were performed, and no issues were further observed. The flow control pins for uart5 are not available on PX30-uQ7 Ringneck, as configured by pinctrl-0, so the DMA nodes were removed on SoM dtsi. Cc: stable@vger.kernel.org Fixes: c484cf93f61b ("arm64: dts: rockchip: add PX30-µQ7 (Ringneck) SoM with Haikou baseboard") Reviewed-by: Quentin Schulz Signed-off-by: Lukasz Czechowski Link: https://lore.kernel.org/r/20250121125604.3115235-3-lukasz.czechowski@thaumatec.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi b/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi index 2c87005c89bd..e80412abec08 100644 --- a/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi +++ b/arch/arm64/boot/dts/rockchip/px30-ringneck.dtsi @@ -397,6 +397,8 @@ &u2phy_host { }; &uart5 { + /delete-property/ dmas; + /delete-property/ dma-names; pinctrl-0 = <&uart5_xfer>; }; From 2f9eb5262e63396a315c7da34a6c80c5d335df9f Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Thu, 16 Jan 2025 15:36:31 +0100 Subject: [PATCH 0336/1443] arm64: dts: rockchip: fix fixed-regulator renames on rk3399-gru devices rk3399-gru chromebooks have a regulator chains where one named regulator supplies multiple regulators pp900-usb pp900_pcie that supply the named peripherals. The dtsi used somewhat creative structure to describe that in creating the base node 3 times with different phandles and describing the EC dependency in a comment. This didn't register in the recent regulator-node renaming, as the additional nodes were empty, so adapt the missing node names for now. Fixes: 5c96e6330197 ("arm64: dts: rockchip: adapt regulator nodenames to preferred form") Tested-by: Vicente Bergas Signed-off-by: Heiko Stuebner Link: https://lore.kernel.org/r/20250116143631.3650469-1-heiko@sntech.de --- .../dts/rockchip/rk3399-gru-chromebook.dtsi | 8 +++---- .../boot/dts/rockchip/rk3399-gru-scarlet.dtsi | 6 ++--- arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi | 22 +++++++++---------- 3 files changed, 18 insertions(+), 18 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3399-gru-chromebook.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-gru-chromebook.dtsi index 988e6ca32fac..a9ea4b0daa04 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-gru-chromebook.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3399-gru-chromebook.dtsi @@ -22,11 +22,11 @@ pp900_ap: regulator-pp900-ap { }; /* EC turns on w/ pp900_usb_en */ - pp900_usb: pp900-ap { + pp900_usb: regulator-pp900-ap { }; /* EC turns on w/ pp900_pcie_en */ - pp900_pcie: pp900-ap { + pp900_pcie: regulator-pp900-ap { }; pp3000: regulator-pp3000 { @@ -126,7 +126,7 @@ pp1800_pcie: regulator-pp1800-pcie { }; /* Always on; plain and simple */ - pp3000_ap: pp3000_emmc: pp3000 { + pp3000_ap: pp3000_emmc: regulator-pp3000 { }; pp1500_ap_io: regulator-pp1500-ap-io { @@ -160,7 +160,7 @@ pp3300_disp: regulator-pp3300-disp { }; /* EC turns on w/ pp3300_usb_en_l */ - pp3300_usb: pp3300 { + pp3300_usb: regulator-pp3300 { }; /* gpio is shared with pp1800_pcie and pinctrl is set there */ diff --git a/arch/arm64/boot/dts/rockchip/rk3399-gru-scarlet.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-gru-scarlet.dtsi index 19b23b438965..5e068377a0a2 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-gru-scarlet.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3399-gru-scarlet.dtsi @@ -92,7 +92,7 @@ pp900_s3: regulator-pp900-s3 { }; /* EC turns on pp1800_s3_en */ - pp1800_s3: pp1800 { + pp1800_s3: regulator-pp1800 { }; /* pp3300 children, sorted by name */ @@ -109,11 +109,11 @@ pp2800_cam: regulator-pp2800-avdd { }; /* EC turns on pp3300_s0_en */ - pp3300_s0: pp3300 { + pp3300_s0: regulator-pp3300 { }; /* EC turns on pp3300_s3_en */ - pp3300_s3: pp3300 { + pp3300_s3: regulator-pp3300 { }; /* diff --git a/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi index 6d9e60b01225..7eca1da78cff 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3399-gru.dtsi @@ -189,39 +189,39 @@ ppvar_gpu: ppvar-gpu { }; /* EC turns on w/ pp900_ddrpll_en */ - pp900_ddrpll: pp900-ap { + pp900_ddrpll: regulator-pp900-ap { }; /* EC turns on w/ pp900_pll_en */ - pp900_pll: pp900-ap { + pp900_pll: regulator-pp900-ap { }; /* EC turns on w/ pp900_pmu_en */ - pp900_pmu: pp900-ap { + pp900_pmu: regulator-pp900-ap { }; /* EC turns on w/ pp1800_s0_en_l */ - pp1800_ap_io: pp1800_emmc: pp1800_nfc: pp1800_s0: pp1800 { + pp1800_ap_io: pp1800_emmc: pp1800_nfc: pp1800_s0: regulator-pp1800 { }; /* EC turns on w/ pp1800_avdd_en_l */ - pp1800_avdd: pp1800 { + pp1800_avdd: regulator-pp1800 { }; /* EC turns on w/ pp1800_lid_en_l */ - pp1800_lid: pp1800_mic: pp1800 { + pp1800_lid: pp1800_mic: regulator-pp1800 { }; /* EC turns on w/ lpddr_pwr_en */ - pp1800_lpddr: pp1800 { + pp1800_lpddr: regulator-pp1800 { }; /* EC turns on w/ pp1800_pmu_en_l */ - pp1800_pmu: pp1800 { + pp1800_pmu: regulator-pp1800 { }; /* EC turns on w/ pp1800_usb_en_l */ - pp1800_usb: pp1800 { + pp1800_usb: regulator-pp1800 { }; pp3000_sd_slot: regulator-pp3000-sd-slot { @@ -259,11 +259,11 @@ ppvar_sd_card_io: ppvar-sd-card-io { }; /* EC turns on w/ pp3300_trackpad_en_l */ - pp3300_trackpad: pp3300-trackpad { + pp3300_trackpad: regulator-pp3300-trackpad { }; /* EC turns on w/ usb_a_en */ - pp5000_usb_a_vbus: pp5000 { + pp5000_usb_a_vbus: regulator-pp5000 { }; ap_rtc_clk: ap-rtc-clk { From a1d939055a22be06d8c12bf53afb258b9d38575f Mon Sep 17 00:00:00 2001 From: Andy Yan Date: Mon, 13 Jan 2025 18:47:34 +0800 Subject: [PATCH 0337/1443] arm64: dts: rockchip: Fix lcdpwr_en pin for Cool Pi GenBook According to the schematic, the lcdpwr_en pin is GPIO0_C4, not GPIO1_C4. Fixes: 4a8c1161b843 ("arm64: dts: rockchip: Add support for rk3588 based Cool Pi CM5 GenBook") Signed-off-by: Andy Yan Link: https://lore.kernel.org/r/20250113104825.2390427-1-andyshrk@163.com Signed-off-by: Heiko Stuebner --- arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5-genbook.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5-genbook.dts b/arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5-genbook.dts index 92f0ed83c990..bc6b43a77153 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5-genbook.dts +++ b/arch/arm64/boot/dts/rockchip/rk3588-coolpi-cm5-genbook.dts @@ -113,7 +113,7 @@ vcc3v3_lcd: regulator-vcc3v3-lcd { compatible = "regulator-fixed"; regulator-name = "vcc3v3_lcd"; enable-active-high; - gpio = <&gpio1 RK_PC4 GPIO_ACTIVE_HIGH>; + gpio = <&gpio0 RK_PC4 GPIO_ACTIVE_HIGH>; pinctrl-names = "default"; pinctrl-0 = <&lcdpwr_en>; vin-supply = <&vcc3v3_sys>; @@ -241,7 +241,7 @@ &pcie3x4 { &pinctrl { lcd { lcdpwr_en: lcdpwr-en { - rockchip,pins = <1 RK_PC4 RK_FUNC_GPIO &pcfg_pull_down>; + rockchip,pins = <0 RK_PC4 RK_FUNC_GPIO &pcfg_pull_down>; }; bl_en: bl-en { From 34776963452040f199bf9985ad8a2261a82d8b99 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Wed, 29 Jan 2025 13:56:31 +0100 Subject: [PATCH 0338/1443] accel/ivpu: Add support for hardware fault injection Introduces the capability to simulate hardware faults for testing purposes. The new `fail_hw` fault can be injected in `ivpu_hw_reg_poll_fld()`, which is used in various parts of the driver to wait for the hardware to reach a specific state. This allows to test failures during NPU boot and shutdown, IPC message handling and more. Fault injection can be enabled using debugfs or a module parameter. Reviewed-by: Maciej Falkowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-2-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_debugfs.c | 5 +++ drivers/accel/ivpu/ivpu_hw.c | 14 +++++++ drivers/accel/ivpu/ivpu_hw_ip.c | 4 +- drivers/accel/ivpu/ivpu_hw_reg_io.h | 62 +++++++++++++++++------------ 4 files changed, 57 insertions(+), 28 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c index 8180b95ed69d..e79715c53f2a 100644 --- a/drivers/accel/ivpu/ivpu_debugfs.c +++ b/drivers/accel/ivpu/ivpu_debugfs.c @@ -4,6 +4,7 @@ */ #include +#include #include #include @@ -430,4 +431,8 @@ void ivpu_debugfs_init(struct ivpu_device *vdev) debugfs_root, vdev, &fw_profiling_freq_fops); debugfs_create_file("dct", 0644, debugfs_root, vdev, &ivpu_dct_fops); } + +#ifdef CONFIG_FAULT_INJECTION + fault_create_debugfs_attr("fail_hw", debugfs_root, &ivpu_hw_failure); +#endif } diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index e332f19ab51d..8099ab047bfe 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -9,6 +9,15 @@ #include "ivpu_hw_ip.h" #include +#include + +#ifdef CONFIG_FAULT_INJECTION +DECLARE_FAULT_ATTR(ivpu_hw_failure); + +static char *ivpu_fail_hw; +module_param_named_unsafe(fail_hw, ivpu_fail_hw, charp, 0444); +MODULE_PARM_DESC(fail_hw, ",,,"); +#endif static char *platform_to_str(u32 platform) { @@ -247,6 +256,11 @@ int ivpu_hw_init(struct ivpu_device *vdev) timeouts_init(vdev); atomic_set(&vdev->hw->firewall_irq_counter, 0); +#ifdef CONFIG_FAULT_INJECTION + if (ivpu_fail_hw) + setup_fault_attr(&ivpu_hw_failure, ivpu_fail_hw); +#endif + return 0; } diff --git a/drivers/accel/ivpu/ivpu_hw_ip.c b/drivers/accel/ivpu/ivpu_hw_ip.c index 029dd065614b..823f6a57dc54 100644 --- a/drivers/accel/ivpu/ivpu_hw_ip.c +++ b/drivers/accel/ivpu/ivpu_hw_ip.c @@ -968,14 +968,14 @@ void ivpu_hw_ip_wdt_disable(struct ivpu_device *vdev) static u32 ipc_rx_count_get_37xx(struct ivpu_device *vdev) { - u32 count = REGV_RD32_SILENT(VPU_37XX_HOST_SS_TIM_IPC_FIFO_STAT); + u32 count = readl(vdev->regv + VPU_37XX_HOST_SS_TIM_IPC_FIFO_STAT); return REG_GET_FLD(VPU_37XX_HOST_SS_TIM_IPC_FIFO_STAT, FILL_LEVEL, count); } static u32 ipc_rx_count_get_40xx(struct ivpu_device *vdev) { - u32 count = REGV_RD32_SILENT(VPU_40XX_HOST_SS_TIM_IPC_FIFO_STAT); + u32 count = readl(vdev->regv + VPU_40XX_HOST_SS_TIM_IPC_FIFO_STAT); return REG_GET_FLD(VPU_40XX_HOST_SS_TIM_IPC_FIFO_STAT, FILL_LEVEL, count); } diff --git a/drivers/accel/ivpu/ivpu_hw_reg_io.h b/drivers/accel/ivpu/ivpu_hw_reg_io.h index 79b3f441eac4..66259b0ead02 100644 --- a/drivers/accel/ivpu/ivpu_hw_reg_io.h +++ b/drivers/accel/ivpu/ivpu_hw_reg_io.h @@ -7,6 +7,7 @@ #define __IVPU_HW_REG_IO_H__ #include +#include #include #include @@ -16,13 +17,11 @@ #define REG_IO_ERROR 0xffffffff #define REGB_RD32(reg) ivpu_hw_reg_rd32(vdev, vdev->regb, (reg), #reg, __func__) -#define REGB_RD32_SILENT(reg) readl(vdev->regb + (reg)) #define REGB_RD64(reg) ivpu_hw_reg_rd64(vdev, vdev->regb, (reg), #reg, __func__) #define REGB_WR32(reg, val) ivpu_hw_reg_wr32(vdev, vdev->regb, (reg), (val), #reg, __func__) #define REGB_WR64(reg, val) ivpu_hw_reg_wr64(vdev, vdev->regb, (reg), (val), #reg, __func__) #define REGV_RD32(reg) ivpu_hw_reg_rd32(vdev, vdev->regv, (reg), #reg, __func__) -#define REGV_RD32_SILENT(reg) readl(vdev->regv + (reg)) #define REGV_RD64(reg) ivpu_hw_reg_rd64(vdev, vdev->regv, (reg), #reg, __func__) #define REGV_WR32(reg, val) ivpu_hw_reg_wr32(vdev, vdev->regv, (reg), (val), #reg, __func__) #define REGV_WR64(reg, val) ivpu_hw_reg_wr64(vdev, vdev->regv, (reg), (val), #reg, __func__) @@ -47,31 +46,42 @@ #define REG_TEST_FLD_NUM(REG, FLD, num, val) \ ((num) == FIELD_GET(REG##_##FLD##_MASK, val)) -#define REGB_POLL_FLD(reg, fld, val, timeout_us) \ -({ \ - u32 var; \ - int r; \ - ivpu_dbg(vdev, REG, "%s : %s (0x%08x) Polling field %s started (expected 0x%x)\n", \ - __func__, #reg, reg, #fld, val); \ - r = read_poll_timeout(REGB_RD32_SILENT, var, (FIELD_GET(reg##_##fld##_MASK, var) == (val)),\ - REG_POLL_SLEEP_US, timeout_us, false, (reg)); \ - ivpu_dbg(vdev, REG, "%s : %s (0x%08x) Polling field %s %s (reg val 0x%08x)\n", \ - __func__, #reg, reg, #fld, r ? "ETIMEDOUT" : "OK", var); \ - r; \ -}) +#define REGB_POLL_FLD(reg, fld, exp_fld_val, timeout_us) \ + ivpu_hw_reg_poll_fld(vdev, vdev->regb, reg, reg##_##fld##_MASK, \ + FIELD_PREP(reg##_##fld##_MASK, exp_fld_val), timeout_us, \ + __func__, #reg, #fld) -#define REGV_POLL_FLD(reg, fld, val, timeout_us) \ -({ \ - u32 var; \ - int r; \ - ivpu_dbg(vdev, REG, "%s : %s (0x%08x) Polling field %s started (expected 0x%x)\n", \ - __func__, #reg, reg, #fld, val); \ - r = read_poll_timeout(REGV_RD32_SILENT, var, (FIELD_GET(reg##_##fld##_MASK, var) == (val)),\ - REG_POLL_SLEEP_US, timeout_us, false, (reg)); \ - ivpu_dbg(vdev, REG, "%s : %s (0x%08x) Polling field %s %s (reg val 0x%08x)\n", \ - __func__, #reg, reg, #fld, r ? "ETIMEDOUT" : "OK", var); \ - r; \ -}) +#define REGV_POLL_FLD(reg, fld, exp_fld_val, timeout_us) \ + ivpu_hw_reg_poll_fld(vdev, vdev->regv, reg, reg##_##fld##_MASK, \ + FIELD_PREP(reg##_##fld##_MASK, exp_fld_val), timeout_us, \ + __func__, #reg, #fld) + +extern struct fault_attr ivpu_hw_failure; + +static inline int __must_check +ivpu_hw_reg_poll_fld(struct ivpu_device *vdev, void __iomem *base, + u32 reg_offset, u32 reg_mask, u32 exp_masked_val, u32 timeout_us, + const char *func_name, const char *reg_name, const char *fld_name) +{ + u32 reg_val; + int ret; + + ivpu_dbg(vdev, REG, "%s : %s (0x%08x) POLL %s started (exp_val 0x%x)\n", + func_name, reg_name, reg_offset, fld_name, exp_masked_val); + + ret = read_poll_timeout(readl, reg_val, (reg_val & reg_mask) == exp_masked_val, + REG_POLL_SLEEP_US, timeout_us, false, base + reg_offset); + +#ifdef CONFIG_FAULT_INJECTION + if (should_fail(&ivpu_hw_failure, 1)) + ret = -ETIMEDOUT; +#endif + + ivpu_dbg(vdev, REG, "%s : %s (0x%08x) POLL %s %s (reg_val 0x%08x)\n", + func_name, reg_name, reg_offset, fld_name, ret ? "ETIMEDOUT" : "OK", reg_val); + + return ret; +} static inline u32 ivpu_hw_reg_rd32(struct ivpu_device *vdev, void __iomem *base, u32 reg, From b8c00323ae65e1bbee9f87a1e66592b589f53b50 Mon Sep 17 00:00:00 2001 From: Jacek Lawrynowicz Date: Wed, 29 Jan 2025 13:56:32 +0100 Subject: [PATCH 0339/1443] accel/ivpu: Update last_busy in IRQ handler Call pm_runtime_mark_last_busy() in top half of IRQ handler to prevent device from being runtime suspended before bottom half is executed on a workqueue. Reviewed-by: Karol Wachowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-3-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_hw.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 8099ab047bfe..4199f0bbb435 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -10,6 +10,7 @@ #include #include +#include #ifdef CONFIG_FAULT_INJECTION DECLARE_FAULT_ATTR(ivpu_hw_failure); @@ -331,7 +332,9 @@ irqreturn_t ivpu_hw_irq_handler(int irq, void *ptr) /* Re-enable global interrupts to re-trigger MSI for pending interrupts */ ivpu_hw_btrs_global_int_enable(vdev); - if (ip_handled || btrs_handled) - return IRQ_HANDLED; - return IRQ_NONE; + if (!ip_handled && !btrs_handled) + return IRQ_NONE; + + pm_runtime_mark_last_busy(vdev->drm.dev); + return IRQ_HANDLED; } From 2f5bbea1807a064a1e4c1b385c8cea4f37bb4b17 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Wed, 29 Jan 2025 13:56:33 +0100 Subject: [PATCH 0340/1443] accel/ivpu: Fix missing MMU events if file_priv is unbound Move the ivpu_mmu_discard_events() function to the common portion of the abort work function. This ensures it is called only once, even if there are no faulty contexts in context_xa, to guarantee that MMU events are discarded and new events are not missed. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-4-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index 3c162ac41a1d..c1013f511efa 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -375,7 +375,6 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) ivpu_jsm_context_release(vdev, file_priv->ctx.id); ivpu_mmu_disable_ssid_events(vdev, file_priv->ctx.id); - ivpu_mmu_discard_events(vdev); file_priv->aborted = true; } @@ -980,6 +979,13 @@ void ivpu_context_abort_work_fn(struct work_struct *work) } mutex_unlock(&vdev->context_list_lock); + /* + * We will not receive new MMU event interrupts until existing events are discarded + * however, we want to discard these events only after aborting the faulty context + * to avoid generating new faults from that context + */ + ivpu_mmu_discard_events(vdev); + if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) return; From 3a40d4f4613e437cf78b9b6a28202c6549227459 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Wed, 29 Jan 2025 13:56:34 +0100 Subject: [PATCH 0341/1443] accel/ivpu: Turn on HWS by default on all platforms Hardware scheduling (HWS) is supposed to be supported on all existing platform with recent FW including pre-silicon ones. Turn on HWS by default. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-5-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_fw.c | 10 +--------- 1 file changed, 1 insertion(+), 9 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index 6cf1fb826d1b..cfe8f79ec5a6 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -148,15 +148,7 @@ ivpu_fw_sched_mode_select(struct ivpu_device *vdev, const struct vpu_firmware_he if (IVPU_FW_CHECK_API_VER_LT(vdev, fw_hdr, JSM, 3, 24)) return VPU_SCHEDULING_MODE_OS; - switch (ivpu_device_id(vdev)) { - case PCI_DEVICE_ID_MTL: - case PCI_DEVICE_ID_ARL: - case PCI_DEVICE_ID_LNL: - case PCI_DEVICE_ID_PTL_P: - return VPU_SCHEDULING_MODE_HW; - default: - return VPU_SCHEDULING_MODE_OS; - } + return VPU_SCHEDULING_MODE_HW; } static int ivpu_fw_parse(struct ivpu_device *vdev) From af80fe138bebea8db089c5062bca9833a2602667 Mon Sep 17 00:00:00 2001 From: Tomasz Rusinowicz Date: Wed, 29 Jan 2025 13:56:35 +0100 Subject: [PATCH 0342/1443] accel/ivpu: Enable recovery and adjust timeouts for fpga Recovery now works on fpga. JSM state dump timeout needs to be really long for the new fpga model releases. Enable punit on fpga. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Tomasz Rusinowicz Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-6-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_hw.c | 10 +++++----- drivers/accel/ivpu/ivpu_pm.c | 5 ----- 2 files changed, 5 insertions(+), 10 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 4199f0bbb435..2057907c5aa6 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -58,7 +58,7 @@ static void platform_init(struct ivpu_device *vdev) static void wa_init(struct ivpu_device *vdev) { - vdev->wa.punit_disabled = ivpu_is_fpga(vdev); + vdev->wa.punit_disabled = false; vdev->wa.clear_runtime_mem = false; if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) @@ -87,12 +87,12 @@ static void timeouts_init(struct ivpu_device *vdev) vdev->timeout.autosuspend = -1; vdev->timeout.d0i3_entry_msg = -1; } else if (ivpu_is_fpga(vdev)) { - vdev->timeout.boot = 100000; - vdev->timeout.jsm = 50000; - vdev->timeout.tdr = 2000000; + vdev->timeout.boot = 50; + vdev->timeout.jsm = 15000; + vdev->timeout.tdr = 30000; vdev->timeout.autosuspend = -1; vdev->timeout.d0i3_entry_msg = 500; - vdev->timeout.state_dump_msg = 10; + vdev->timeout.state_dump_msg = 10000; } else if (ivpu_is_simics(vdev)) { vdev->timeout.boot = 50; vdev->timeout.jsm = 500; diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index f41b3bfe40af..8f6222d15720 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -161,11 +161,6 @@ void ivpu_pm_trigger_recovery(struct ivpu_device *vdev, const char *reason) return; } - if (ivpu_is_fpga(vdev)) { - ivpu_err(vdev, "Recovery not available on FPGA\n"); - return; - } - /* Trigger recovery if it's not in progress */ if (atomic_cmpxchg(&vdev->pm->reset_pending, 0, 1) == 0) { ivpu_hw_diagnose_failure(vdev); From d24b85df791840089c2df6a25809aca814b1206d Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Wed, 29 Jan 2025 13:56:36 +0100 Subject: [PATCH 0343/1443] accel/ivpu: Move recovery work to system_unbound_wq Recovery work doesn't need to be bound to any specific CPU, so move it to unbound workqueue to improve execution time and system latency. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Reviewed-by: Jeffrey Hugo Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250129125636.1047413-7-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_pm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index 8f6222d15720..d3db944ad864 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -165,7 +165,7 @@ void ivpu_pm_trigger_recovery(struct ivpu_device *vdev, const char *reason) if (atomic_cmpxchg(&vdev->pm->reset_pending, 0, 1) == 0) { ivpu_hw_diagnose_failure(vdev); ivpu_hw_irq_disable(vdev); /* Disable IRQ early to protect from IRQ storm */ - queue_work(system_long_wq, &vdev->pm->recovery_work); + queue_work(system_unbound_wq, &vdev->pm->recovery_work); } } From 48e487b002891eb0aeaec704c9bed51f028deff1 Mon Sep 17 00:00:00 2001 From: Stuart Hayhurst Date: Tue, 21 Jan 2025 20:00:07 +0000 Subject: [PATCH 0344/1443] HID: corsair-void: Add missing delayed work cancel for headset status The cancel_delayed_work_sync() call was missed, causing a use-after-free in corsair_void_remove(). Reported-by: yan kang Reported-by: yue sun Closes: https://lore.kernel.org/all/SY8P300MB042106286A2536707D2FB736A1E42@SY8P300MB0421.AUSP300.PROD.OUTLOOK.COM/ Closes: https://lore.kernel.org/all/SY8P300MB0421872E0AE934C9616FA61EA1E42@SY8P300MB0421.AUSP300.PROD.OUTLOOK.COM/ Fixes: 6ea2a6fd3872 ("HID: corsair-void: Add Corsair Void headset family driver") Cc: stable@vger.kernel.org Signed-off-by: Stuart Hayhurst Signed-off-by: Jiri Kosina --- drivers/hid/hid-corsair-void.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/hid-corsair-void.c b/drivers/hid/hid-corsair-void.c index 6ece56b850fc..bd8f3d849b58 100644 --- a/drivers/hid/hid-corsair-void.c +++ b/drivers/hid/hid-corsair-void.c @@ -726,6 +726,7 @@ static void corsair_void_remove(struct hid_device *hid_dev) if (drvdata->battery) power_supply_unregister(drvdata->battery); + cancel_delayed_work_sync(&drvdata->delayed_status_work); cancel_delayed_work_sync(&drvdata->delayed_firmware_work); sysfs_remove_group(&hid_dev->dev.kobj, &corsair_void_attr_group); } From c098363828f7006ef5c5121b673bc5e26571e6c8 Mon Sep 17 00:00:00 2001 From: Stuart Hayhurst Date: Tue, 21 Jan 2025 20:00:08 +0000 Subject: [PATCH 0345/1443] HID: corsair-void: Initialise memory for psy_cfg power_supply_config psy_cfg was missing its initialiser, add it in. Fixes: 6ea2a6fd3872 ("HID: corsair-void: Add Corsair Void headset family driver") Cc: stable@vger.kernel.org Signed-off-by: Stuart Hayhurst Signed-off-by: Jiri Kosina --- drivers/hid/hid-corsair-void.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/hid-corsair-void.c b/drivers/hid/hid-corsair-void.c index bd8f3d849b58..56e858066c3c 100644 --- a/drivers/hid/hid-corsair-void.c +++ b/drivers/hid/hid-corsair-void.c @@ -553,7 +553,7 @@ static void corsair_void_battery_remove_work_handler(struct work_struct *work) static void corsair_void_battery_add_work_handler(struct work_struct *work) { struct corsair_void_drvdata *drvdata; - struct power_supply_config psy_cfg; + struct power_supply_config psy_cfg = {}; struct power_supply *new_supply; drvdata = container_of(work, struct corsair_void_drvdata, From 4b54ae69197b9f416baa0fceadff7e89075f8454 Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Wed, 22 Jan 2025 09:29:00 +0800 Subject: [PATCH 0346/1443] HID: intel-ish-hid: fix the length of MNG_SYNC_FW_CLOCK in doorbell The timestamps in the Firmware log and HID sensor samples are incorrect. They show 1970-01-01 because the current IPC driver only uses the first 8 bytes of bootup time when synchronizing time with the firmware. The firmware converts the bootup time to UTC time, which results in the display of 1970-01-01. In write_ipc_from_queue(), when sending the MNG_SYNC_FW_CLOCK message, the clock is updated according to the definition of ipc_time_update_msg. However, in _ish_sync_fw_clock(), the message length is specified as the size of uint64_t when building the doorbell. As a result, the firmware only receives the first 8 bytes of struct ipc_time_update_msg. This patch corrects the length in the doorbell to ensure the entire ipc_time_update_msg is sent, fixing the timestamp issue. Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Signed-off-by: Jiri Kosina --- drivers/hid/intel-ish-hid/ipc/ipc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/hid/intel-ish-hid/ipc/ipc.c b/drivers/hid/intel-ish-hid/ipc/ipc.c index 3cd53fc80634..cb956a8c386c 100644 --- a/drivers/hid/intel-ish-hid/ipc/ipc.c +++ b/drivers/hid/intel-ish-hid/ipc/ipc.c @@ -578,14 +578,14 @@ static void fw_reset_work_fn(struct work_struct *work) static void _ish_sync_fw_clock(struct ishtp_device *dev) { static unsigned long prev_sync; - uint64_t usec; + struct ipc_time_update_msg time = {}; if (prev_sync && time_before(jiffies, prev_sync + 20 * HZ)) return; prev_sync = jiffies; - usec = ktime_to_us(ktime_get_boottime()); - ipc_send_mng_msg(dev, MNG_SYNC_FW_CLOCK, &usec, sizeof(uint64_t)); + /* The fields of time would be updated while sending message */ + ipc_send_mng_msg(dev, MNG_SYNC_FW_CLOCK, &time, sizeof(time)); } /** From 7e0d1cff12b895f44f4ddc8cf50311bc1f775201 Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Wed, 22 Jan 2025 09:29:01 +0800 Subject: [PATCH 0347/1443] HID: intel-ish-hid: Send clock sync message immediately after reset The ISH driver performs a clock sync with the firmware once at system startup and then every 20 seconds. If a firmware reset occurs right after a clock sync, the driver would wait 20 seconds before performing another clock sync with the firmware. This is particularly problematic with the introduction of the "load firmware from host" feature, where the driver performs a clock sync with the bootloader and then has to wait 20 seconds before syncing with the main firmware. This patch clears prev_sync immediately upon receiving an IPC reset, so that the main firmware and driver will perform a clock sync immediately after completing the IPC handshake. Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Signed-off-by: Jiri Kosina --- drivers/hid/intel-ish-hid/ipc/ipc.c | 9 ++++++--- drivers/hid/intel-ish-hid/ishtp/ishtp-dev.h | 2 ++ 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/hid/intel-ish-hid/ipc/ipc.c b/drivers/hid/intel-ish-hid/ipc/ipc.c index cb956a8c386c..4c861119e97a 100644 --- a/drivers/hid/intel-ish-hid/ipc/ipc.c +++ b/drivers/hid/intel-ish-hid/ipc/ipc.c @@ -517,6 +517,10 @@ static int ish_fw_reset_handler(struct ishtp_device *dev) /* ISH FW is dead */ if (!ish_is_input_ready(dev)) return -EPIPE; + + /* Send clock sync at once after reset */ + ishtp_dev->prev_sync = 0; + /* * Set HOST2ISH.ILUP. Apparently we need this BEFORE sending * RESET_NOTIFY_ACK - FW will be checking for it @@ -577,13 +581,12 @@ static void fw_reset_work_fn(struct work_struct *work) */ static void _ish_sync_fw_clock(struct ishtp_device *dev) { - static unsigned long prev_sync; struct ipc_time_update_msg time = {}; - if (prev_sync && time_before(jiffies, prev_sync + 20 * HZ)) + if (dev->prev_sync && time_before(jiffies, dev->prev_sync + 20 * HZ)) return; - prev_sync = jiffies; + dev->prev_sync = jiffies; /* The fields of time would be updated while sending message */ ipc_send_mng_msg(dev, MNG_SYNC_FW_CLOCK, &time, sizeof(time)); } diff --git a/drivers/hid/intel-ish-hid/ishtp/ishtp-dev.h b/drivers/hid/intel-ish-hid/ishtp/ishtp-dev.h index 44eddc411e97..ec9f6e87aaf2 100644 --- a/drivers/hid/intel-ish-hid/ishtp/ishtp-dev.h +++ b/drivers/hid/intel-ish-hid/ishtp/ishtp-dev.h @@ -253,6 +253,8 @@ struct ishtp_device { unsigned int ipc_tx_cnt; unsigned long long ipc_tx_bytes_cnt; + /* Time of the last clock sync */ + unsigned long prev_sync; const struct ishtp_hw_ops *ops; size_t mtu; uint32_t ishtp_msg_hdr; From 52572cde8b4a44676557ccb67b035291833112c5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 22 Jan 2025 07:50:57 +0100 Subject: [PATCH 0348/1443] HID: lenovo: select CONFIG_ACPI_PLATFORM_PROFILE A previous patch tried to fix this link failure: x86_64-linux-ld: drivers/hid/hid-lenovo.o: in function `lenovo_raw_event': hid-lenovo.c:(.text+0x22c): undefined reference to `platform_profile_cycle' but got it wrong in three ways: - the link failure still exists with CONFIG_ACPI_PLATFORM_PROFILE=m when hid-lenovo is built-in - There is no way to manually enable CONFIG_ACPI_PLATFORM_PROFILE, as it is intended to be selected by its users. Remove the broken #if check again and instead select the symbol like the other users do. This requires adding a dependency on CONFIG_ACPI. Fixes: 52e7d1f7c2fd ("HID: lenovo: Fix undefined platform_profile_cycle in ThinkPad X12 keyboard patch") Signed-off-by: Arnd Bergmann Signed-off-by: Jiri Kosina --- drivers/hid/Kconfig | 2 ++ drivers/hid/hid-lenovo.c | 7 +------ 2 files changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig index b53eb569bd49..8adb745c5b28 100644 --- a/drivers/hid/Kconfig +++ b/drivers/hid/Kconfig @@ -570,6 +570,8 @@ config HID_LED config HID_LENOVO tristate "Lenovo / Thinkpad devices" + depends on ACPI + select ACPI_PLATFORM_PROFILE select NEW_LEDS select LEDS_CLASS help diff --git a/drivers/hid/hid-lenovo.c b/drivers/hid/hid-lenovo.c index 4d00bc4d656e..a7d9ca02779e 100644 --- a/drivers/hid/hid-lenovo.c +++ b/drivers/hid/hid-lenovo.c @@ -32,9 +32,7 @@ #include #include -#if IS_ENABLED(CONFIG_ACPI_PLATFORM_PROFILE) #include -#endif /* CONFIG_ACPI_PLATFORM_PROFILE */ #include "hid-ids.h" @@ -730,13 +728,10 @@ static int lenovo_raw_event_TP_X12_tab(struct hid_device *hdev, u32 raw_data) if (hdev->product == USB_DEVICE_ID_LENOVO_X12_TAB) { report_key_event(input, KEY_RFKILL); return 1; - } -#if IS_ENABLED(CONFIG_ACPI_PLATFORM_PROFILE) - else { + } else { platform_profile_cycle(); return 1; } -#endif /* CONFIG_ACPI_PLATFORM_PROFILE */ return 0; case TP_X12_RAW_HOTKEY_FN_F10: /* TAB1 has PICKUP Phone and TAB2 use Snipping tool*/ From a5a056c8d2ba60017dffb914bdf92c5562defc48 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 23 Jan 2025 14:48:12 +0100 Subject: [PATCH 0349/1443] HID: intel-thc: fix CONFIG_HID dependency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In drivers/hid/, most drivers depend on CONFIG_HID, while a couple of the drivers in subdirectories instead depend on CONFIG_HID_SUPPORT and use 'select HID'. With the newly added INTEL_THC_HID, this causes a build warning for a circular dependency: WARNING: unmet direct dependencies detected for HID Depends on [m]: HID_SUPPORT [=y] && INPUT [=m] Selected by [y]: - INTEL_THC_HID [=y] && HID_SUPPORT [=y] && X86_64 [=y] && PCI [=y] && ACPI [=y] WARNING: unmet direct dependencies detected for INPUT_FF_MEMLESS Depends on [m]: INPUT [=m] Selected by [y]: - HID_MICROSOFT [=y] && HID_SUPPORT [=y] && HID [=y] - GREENASIA_FF [=y] && HID_SUPPORT [=y] && HID [=y] && HID_GREENASIA [=y] - HID_WIIMOTE [=y] && HID_SUPPORT [=y] && HID [=y] && LEDS_CLASS [=y] - ZEROPLUS_FF [=y] && HID_SUPPORT [=y] && HID [=y] && HID_ZEROPLUS [=y] Selected by [m]: - HID_ACRUX_FF [=y] && HID_SUPPORT [=y] && HID [=y] && HID_ACRUX [=m] - HID_EMS_FF [=m] && HID_SUPPORT [=y] && HID [=y] - HID_GOOGLE_STADIA_FF [=m] && HID_SUPPORT [=y] && HID [=y] - PANTHERLORD_FF [=y] && HID_SUPPORT [=y] && HID [=y] && HID_PANTHERLORD [=m] It's better to be consistent and always use 'depends on HID' for HID drivers. The notable exception here is USB_KBD/USB_MOUSE, which are alternative implementations that do not depend on the HID subsystem. Do this by extending the "if HID" section below, which means that a few of the duplicate "depends on HID" and "depends on INPUT" statements can be removed in the process. Fixes: 1b2d05384c29 ("HID: intel-thc-hid: Add basic THC driver skeleton") Signed-off-by: Arnd Bergmann Reviewed-by: Ilpo Järvinen Reviewed-by: Maximilian Luz Reviewed-by: Even Xu Signed-off-by: Jiri Kosina --- drivers/hid/Kconfig | 10 ++++++---- drivers/hid/amd-sfh-hid/Kconfig | 1 - drivers/hid/i2c-hid/Kconfig | 2 +- drivers/hid/intel-ish-hid/Kconfig | 1 - drivers/hid/intel-thc-hid/Kconfig | 1 - drivers/hid/surface-hid/Kconfig | 2 -- drivers/hid/usbhid/Kconfig | 3 +-- net/bluetooth/hidp/Kconfig | 3 +-- 8 files changed, 9 insertions(+), 14 deletions(-) diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig index 8adb745c5b28..ed657ef7281c 100644 --- a/drivers/hid/Kconfig +++ b/drivers/hid/Kconfig @@ -1376,10 +1376,6 @@ endmenu source "drivers/hid/bpf/Kconfig" -endif # HID - -source "drivers/hid/usbhid/Kconfig" - source "drivers/hid/i2c-hid/Kconfig" source "drivers/hid/intel-ish-hid/Kconfig" @@ -1390,4 +1386,10 @@ source "drivers/hid/surface-hid/Kconfig" source "drivers/hid/intel-thc-hid/Kconfig" +endif # HID + +# USB support may be used with HID disabled + +source "drivers/hid/usbhid/Kconfig" + endif # HID_SUPPORT diff --git a/drivers/hid/amd-sfh-hid/Kconfig b/drivers/hid/amd-sfh-hid/Kconfig index 329de5e12c1a..3291786a5ee6 100644 --- a/drivers/hid/amd-sfh-hid/Kconfig +++ b/drivers/hid/amd-sfh-hid/Kconfig @@ -5,7 +5,6 @@ menu "AMD SFH HID Support" config AMD_SFH_HID tristate "AMD Sensor Fusion Hub" - depends on HID depends on X86 help If you say yes to this option, support will be included for the diff --git a/drivers/hid/i2c-hid/Kconfig b/drivers/hid/i2c-hid/Kconfig index ef7c595c9403..e8d51f410cc1 100644 --- a/drivers/hid/i2c-hid/Kconfig +++ b/drivers/hid/i2c-hid/Kconfig @@ -2,7 +2,7 @@ menuconfig I2C_HID tristate "I2C HID support" default y - depends on I2C && INPUT && HID + depends on I2C if I2C_HID diff --git a/drivers/hid/intel-ish-hid/Kconfig b/drivers/hid/intel-ish-hid/Kconfig index 253dc10d35ef..568c8688784e 100644 --- a/drivers/hid/intel-ish-hid/Kconfig +++ b/drivers/hid/intel-ish-hid/Kconfig @@ -6,7 +6,6 @@ config INTEL_ISH_HID tristate "Intel Integrated Sensor Hub" default n depends on X86 - depends on HID help The Integrated Sensor Hub (ISH) enables the ability to offload sensor polling and algorithm processing to a dedicated low power diff --git a/drivers/hid/intel-thc-hid/Kconfig b/drivers/hid/intel-thc-hid/Kconfig index 91ec84902db8..0351d1137607 100644 --- a/drivers/hid/intel-thc-hid/Kconfig +++ b/drivers/hid/intel-thc-hid/Kconfig @@ -7,7 +7,6 @@ menu "Intel THC HID Support" config INTEL_THC_HID tristate "Intel Touch Host Controller" depends on ACPI - select HID help THC (Touch Host Controller) is the name of the IP block in PCH that interfaces with Touch Devices (ex: touchscreen, touchpad etc.). It diff --git a/drivers/hid/surface-hid/Kconfig b/drivers/hid/surface-hid/Kconfig index 7ce9b5d641eb..d0cfd0d29926 100644 --- a/drivers/hid/surface-hid/Kconfig +++ b/drivers/hid/surface-hid/Kconfig @@ -1,7 +1,6 @@ # SPDX-License-Identifier: GPL-2.0+ menu "Surface System Aggregator Module HID support" depends on SURFACE_AGGREGATOR - depends on INPUT config SURFACE_HID tristate "HID transport driver for Surface System Aggregator Module" @@ -39,4 +38,3 @@ endmenu config SURFACE_HID_CORE tristate - select HID diff --git a/drivers/hid/usbhid/Kconfig b/drivers/hid/usbhid/Kconfig index 7c2032f7f44d..f3194767a45e 100644 --- a/drivers/hid/usbhid/Kconfig +++ b/drivers/hid/usbhid/Kconfig @@ -5,8 +5,7 @@ menu "USB HID support" config USB_HID tristate "USB HID transport layer" default y - depends on USB && INPUT - select HID + depends on HID help Say Y here if you want to connect USB keyboards, mice, joysticks, graphic tablets, or any other HID based devices diff --git a/net/bluetooth/hidp/Kconfig b/net/bluetooth/hidp/Kconfig index 6746be07e222..e08aae35351a 100644 --- a/net/bluetooth/hidp/Kconfig +++ b/net/bluetooth/hidp/Kconfig @@ -1,8 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only config BT_HIDP tristate "HIDP protocol support" - depends on BT_BREDR && INPUT && HID_SUPPORT - select HID + depends on BT_BREDR && HID help HIDP (Human Interface Device Protocol) is a transport layer for HID reports. HIDP is required for the Bluetooth Human From e0efe83ed325277bb70f9435d4d9fc70bebdcca8 Mon Sep 17 00:00:00 2001 From: Lenny Szubowicz Date: Thu, 30 Jan 2025 16:57:54 -0500 Subject: [PATCH 0350/1443] tg3: Disable tg3 PCIe AER on system reboot Disable PCIe AER on the tg3 device on system reboot on a limited list of Dell PowerEdge systems. This prevents a fatal PCIe AER event on the tg3 device during the ACPI _PTS (prepare to sleep) method for S5 on those systems. The _PTS is invoked by acpi_enter_sleep_state_prep() as part of the kernel's reboot sequence as a result of commit 38f34dba806a ("PM: ACPI: reboot: Reinstate S5 for reboot"). There was an earlier fix for this problem by commit 2ca1c94ce0b6 ("tg3: Disable tg3 device on system reboot to avoid triggering AER"). But it was discovered that this earlier fix caused a reboot hang when some Dell PowerEdge servers were booted via ipxe. To address this reboot hang, the earlier fix was essentially reverted by commit 9fc3bc764334 ("tg3: power down device only on SYSTEM_POWER_OFF"). This re-exposed the tg3 PCIe AER on reboot problem. This fix is not an ideal solution because the root cause of the AER is in system firmware. Instead, it's a targeted work-around in the tg3 driver. Note also that the PCIe AER must be disabled on the tg3 device even if the system is configured to use "firmware first" error handling. V3: - Fix sparse warning on improper comparison of pdev->current_state - Adhere to netdev comment style Fixes: 9fc3bc764334 ("tg3: power down device only on SYSTEM_POWER_OFF") Signed-off-by: Lenny Szubowicz Reviewed-by: Pavan Chebbi Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 58 +++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 1c94bf1db718..d9d675f1ebfe 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -55,6 +55,7 @@ #include #include #include +#include #include #include @@ -18212,6 +18213,50 @@ static int tg3_resume(struct device *device) static SIMPLE_DEV_PM_OPS(tg3_pm_ops, tg3_suspend, tg3_resume); +/* Systems where ACPI _PTS (Prepare To Sleep) S5 will result in a fatal + * PCIe AER event on the tg3 device if the tg3 device is not, or cannot + * be, powered down. + */ +static const struct dmi_system_id tg3_restart_aer_quirk_table[] = { + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R440"), + }, + }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R540"), + }, + }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R640"), + }, + }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R650"), + }, + }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R740"), + }, + }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "PowerEdge R750"), + }, + }, + {} +}; + static void tg3_shutdown(struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); @@ -18228,6 +18273,19 @@ static void tg3_shutdown(struct pci_dev *pdev) if (system_state == SYSTEM_POWER_OFF) tg3_power_down(tp); + else if (system_state == SYSTEM_RESTART && + dmi_first_match(tg3_restart_aer_quirk_table) && + pdev->current_state != PCI_D3cold && + pdev->current_state != PCI_UNKNOWN) { + /* Disable PCIe AER on the tg3 to avoid a fatal + * error during this system restart. + */ + pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL, + PCI_EXP_DEVCTL_CERE | + PCI_EXP_DEVCTL_NFERE | + PCI_EXP_DEVCTL_FERE | + PCI_EXP_DEVCTL_URRE); + } rtnl_unlock(); From 235174b2bed88501fda689c113c55737f99332d8 Mon Sep 17 00:00:00 2001 From: Yan Zhai Date: Fri, 31 Jan 2025 00:31:39 -0800 Subject: [PATCH 0351/1443] udp: gso: do not drop small packets when PMTU reduces Commit 4094871db1d6 ("udp: only do GSO if # of segs > 1") avoided GSO for small packets. But the kernel currently dismisses GSO requests only after checking MTU/PMTU on gso_size. This means any packets, regardless of their payload sizes, could be dropped when PMTU becomes smaller than requested gso_size. We encountered this issue in production and it caused a reliability problem that new QUIC connection cannot be established before PMTU cache expired, while non GSO sockets still worked fine at the same time. Ideally, do not check any GSO related constraints when payload size is smaller than requested gso_size, and return EMSGSIZE instead of EINVAL on MTU/PMTU check failure to be more specific on the error cause. Fixes: 4094871db1d6 ("udp: only do GSO if # of segs > 1") Signed-off-by: Yan Zhai Suggested-by: Willem de Bruijn Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/ipv4/udp.c | 4 ++-- net/ipv6/udp.c | 4 ++-- tools/testing/selftests/net/udpgso.c | 26 ++++++++++++++++++++++++++ 3 files changed, 30 insertions(+), 4 deletions(-) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index c472c9a57cf6..a9bb9ce5438e 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1141,9 +1141,9 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4, const int hlen = skb_network_header_len(skb) + sizeof(struct udphdr); - if (hlen + cork->gso_size > cork->fragsize) { + if (hlen + min(datalen, cork->gso_size) > cork->fragsize) { kfree_skb(skb); - return -EINVAL; + return -EMSGSIZE; } if (datalen > cork->gso_size * UDP_MAX_SEGMENTS) { kfree_skb(skb); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 6671daa67f4f..c6ea438b5c75 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1389,9 +1389,9 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6, const int hlen = skb_network_header_len(skb) + sizeof(struct udphdr); - if (hlen + cork->gso_size > cork->fragsize) { + if (hlen + min(datalen, cork->gso_size) > cork->fragsize) { kfree_skb(skb); - return -EINVAL; + return -EMSGSIZE; } if (datalen > cork->gso_size * UDP_MAX_SEGMENTS) { kfree_skb(skb); diff --git a/tools/testing/selftests/net/udpgso.c b/tools/testing/selftests/net/udpgso.c index 3f2fca02fec5..36ff28af4b19 100644 --- a/tools/testing/selftests/net/udpgso.c +++ b/tools/testing/selftests/net/udpgso.c @@ -102,6 +102,19 @@ struct testcase testcases_v4[] = { .gso_len = CONST_MSS_V4, .r_num_mss = 1, }, + { + /* datalen <= MSS < gso_len: will fall back to no GSO */ + .tlen = CONST_MSS_V4, + .gso_len = CONST_MSS_V4 + 1, + .r_num_mss = 0, + .r_len_last = CONST_MSS_V4, + }, + { + /* MSS < datalen < gso_len: fail */ + .tlen = CONST_MSS_V4 + 1, + .gso_len = CONST_MSS_V4 + 2, + .tfail = true, + }, { /* send a single MSS + 1B */ .tlen = CONST_MSS_V4 + 1, @@ -205,6 +218,19 @@ struct testcase testcases_v6[] = { .gso_len = CONST_MSS_V6, .r_num_mss = 1, }, + { + /* datalen <= MSS < gso_len: will fall back to no GSO */ + .tlen = CONST_MSS_V6, + .gso_len = CONST_MSS_V6 + 1, + .r_num_mss = 0, + .r_len_last = CONST_MSS_V6, + }, + { + /* MSS < datalen < gso_len: fail */ + .tlen = CONST_MSS_V6 + 1, + .gso_len = CONST_MSS_V6 + 2, + .tfail = true + }, { /* send a single MSS + 1B */ .tlen = CONST_MSS_V6 + 1, From 363236d709e75610b628c2a4337ccbe42e454b6d Mon Sep 17 00:00:00 2001 From: "Chia-Lin Kao (AceLan)" Date: Wed, 15 Jan 2025 15:00:20 +0800 Subject: [PATCH 0352/1443] HID: ignore non-functional sensor in HP 5MP Camera The HP 5MP Camera (USB ID 0408:5473) reports a HID sensor interface that is not actually implemented. Attempting to access this non-functional sensor via iio_info causes system hangs as runtime PM tries to wake up an unresponsive sensor. [453] hid-sensor-hub 0003:0408:5473.0003: Report latency attributes: ffffffff:ffffffff [453] hid-sensor-hub 0003:0408:5473.0003: common attributes: 5:1, 2:1, 3:1 ffffffff:ffffffff Add this device to the HID ignore list since the sensor interface is non-functional by design and should not be exposed to userspace. Signed-off-by: Chia-Lin Kao (AceLan) Acked-by: Srinivas Pandruvada Signed-off-by: Jiri Kosina --- drivers/hid/hid-ids.h | 1 + drivers/hid/hid-quirks.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index c448de53bf91..7debfe0c5cb9 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -1095,6 +1095,7 @@ #define USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3001 0x3001 #define USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3003 0x3003 #define USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3008 0x3008 +#define USB_DEVICE_ID_QUANTA_HP_5MP_CAMERA_5473 0x5473 #define I2C_VENDOR_ID_RAYDIUM 0x2386 #define I2C_PRODUCT_ID_RAYDIUM_4B33 0x4b33 diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c index e0bbf0c6345d..5d7a418ccdbe 100644 --- a/drivers/hid/hid-quirks.c +++ b/drivers/hid/hid-quirks.c @@ -891,6 +891,7 @@ static const struct hid_device_id hid_ignore_list[] = { { HID_USB_DEVICE(USB_VENDOR_ID_SYNAPTICS, USB_DEVICE_ID_SYNAPTICS_DPAD) }, #endif { HID_USB_DEVICE(USB_VENDOR_ID_YEALINK, USB_DEVICE_ID_YEALINK_P1K_P4K_B2K) }, + { HID_USB_DEVICE(USB_VENDOR_ID_QUANTA, USB_DEVICE_ID_QUANTA_HP_5MP_CAMERA_5473) }, { } }; From 05c4ede6951b5d8e083b6bb237950cac59bdeb92 Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Wed, 15 Jan 2025 17:28:16 -0800 Subject: [PATCH 0353/1443] HID: hid-steam: Fix issues with disabling both gamepad mode and lizard mode When lizard mode is disabled, there were two issues: 1. Switching between gamepad mode and desktop mode still functioned, even though desktop mode did not. This lead to the ability to "break" gamepad mode by holding down the Options key even while lizard mode is disabled 2. If you were in desktop mode when lizard mode is disabled, you would immediately enter this faulty mode. This patch properly disables the ability to switch between gamepad mode and the faulty desktop mode by holding the Options key, as well as effectively removing the faulty mode by bypassing the early returns if lizard mode is disabled. Reported-by: Eugeny Shcheglov Signed-off-by: Vicki Pfau Signed-off-by: Jiri Kosina --- drivers/hid/hid-steam.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/hid/hid-steam.c b/drivers/hid/hid-steam.c index af38fc8eb34f..b008fd0834b9 100644 --- a/drivers/hid/hid-steam.c +++ b/drivers/hid/hid-steam.c @@ -1050,10 +1050,10 @@ static void steam_mode_switch_cb(struct work_struct *work) struct steam_device, mode_switch); unsigned long flags; bool client_opened; - steam->gamepad_mode = !steam->gamepad_mode; if (!lizard_mode) return; + steam->gamepad_mode = !steam->gamepad_mode; if (steam->gamepad_mode) steam_set_lizard_mode(steam, false); else { @@ -1599,7 +1599,7 @@ static void steam_do_deck_input_event(struct steam_device *steam, schedule_delayed_work(&steam->mode_switch, 45 * HZ / 100); } - if (!steam->gamepad_mode) + if (!steam->gamepad_mode && lizard_mode) return; lpad_touched = b10 & BIT(3); @@ -1669,7 +1669,7 @@ static void steam_do_deck_sensors_event(struct steam_device *steam, */ steam->sensor_timestamp_us += 4000; - if (!steam->gamepad_mode) + if (!steam->gamepad_mode && lizard_mode) return; input_event(sensors, EV_MSC, MSC_TIMESTAMP, steam->sensor_timestamp_us); From 3fb3cb4350befc4f901c54e0cb4a2a47b1302e08 Mon Sep 17 00:00:00 2001 From: Andrey Vatoropin Date: Thu, 30 Jan 2025 09:00:34 +0000 Subject: [PATCH 0354/1443] power: supply: da9150-fg: fix potential overflow Size of variable sd_gain equals four bytes - DA9150_QIF_SD_GAIN_SIZE. Size of variable shunt_val equals two bytes - DA9150_QIF_SHUNT_VAL_SIZE. The expression sd_gain * shunt_val is currently being evaluated using 32-bit arithmetic. So during the multiplication an overflow may occur. As the value of type 'u64' is used as storage for the eventual result, put ULL variable at the first position of each expression in order to give the compiler complete information about the proper arithmetic to use. According to C99 the guaranteed width for a variable of type 'unsigned long long' >= 64 bits. Remove the explicit cast to u64 as it is meaningless. Just for the sake of consistency, perform the similar trick with another expression concerning 'iavg'. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: a419b4fd9138 ("power: Add support for DA9150 Fuel-Gauge") Signed-off-by: Andrey Vatoropin Link: https://lore.kernel.org/r/20250130090030.53422-1-a.vatoropin@crpt.ru Signed-off-by: Sebastian Reichel --- drivers/power/supply/da9150-fg.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/power/supply/da9150-fg.c b/drivers/power/supply/da9150-fg.c index 652c1f213af1..4f28ef1bba1a 100644 --- a/drivers/power/supply/da9150-fg.c +++ b/drivers/power/supply/da9150-fg.c @@ -247,9 +247,9 @@ static int da9150_fg_current_avg(struct da9150_fg *fg, DA9150_QIF_SD_GAIN_SIZE); da9150_fg_read_sync_end(fg); - div = (u64) (sd_gain * shunt_val * 65536ULL); + div = 65536ULL * sd_gain * shunt_val; do_div(div, 1000000); - res = (u64) (iavg * 1000000ULL); + res = 1000000ULL * iavg; do_div(res, div); val->intval = (int) res; From 64dd6edfc421479e416301c48b79cece8d0351fc Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Thu, 30 Jan 2025 15:00:35 +0100 Subject: [PATCH 0355/1443] power: supply: core: Fix extension related lockdep warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit 6037802bbae8 ("power: supply: core: implement extension API") there is the following ABBA deadlock (simplified) between the LED trigger code and the power-supply code: 1) When registering a power-supply class device, power_supply_register() calls led_trigger_register() from power_supply_create_triggers() in a scoped_guard(rwsem_read, &psy->extensions_sem) context. led_trigger_register() then in turn takes a LED subsystem lock. So here we have the following locking order: * Read-lock extensions_sem * Lock LED subsystem lock(s) 2) When registering a LED class device, with its default trigger set to a power-supply LED trigger (which has already been registered) The LED class code calls power_supply_led_trigger_activate() when setting up the default trigger. power_supply_led_trigger_activate() calls power_supply_get_property() to determine the initial value of to assign to the LED and that read-locks extensions_sem. So now we have the following locking order: * Lock LED subsystem lock(s) * Read-lock extensions_sem Fixing this is easy, there is no need to hold the extensions_sem when calling power_supply_create_triggers() since all triggers are always created rather then checking for the presence of certain attributes as power_supply_add_hwmon_sysfs() does. Move power_supply_create_triggers() out of the guard block to fix this. Here is the lockdep report fixed by this change: [ 31.249343] ====================================================== [ 31.249378] WARNING: possible circular locking dependency detected [ 31.249413] 6.13.0-rc6+ #251 Tainted: G C E [ 31.249440] ------------------------------------------------------ [ 31.249471] (udev-worker)/553 is trying to acquire lock: [ 31.249501] ffff892adbcaf660 (&psy->extensions_sem){.+.+}-{4:4}, at: power_supply_get_property.part.0+0x22/0x150 [ 31.249574] but task is already holding lock: [ 31.249603] ffff892adbc0bad0 (&led_cdev->trigger_lock){+.+.}-{4:4}, at: led_trigger_set_default+0x34/0xe0 [ 31.249657] which lock already depends on the new lock. [ 31.249696] the existing dependency chain (in reverse order) is: [ 31.249735] -> #2 (&led_cdev->trigger_lock){+.+.}-{4:4}: [ 31.249778] down_write+0x3b/0xd0 [ 31.249803] led_trigger_set_default+0x34/0xe0 [ 31.249833] led_classdev_register_ext+0x311/0x3a0 [ 31.249863] input_leds_connect+0x1dc/0x2a0 [ 31.249889] input_attach_handler.isra.0+0x75/0x90 [ 31.249921] input_register_device.cold+0xa1/0x150 [ 31.249955] hidinput_connect+0x8a2/0xb80 [ 31.249982] hid_connect+0x582/0x5c0 [ 31.250007] hid_hw_start+0x3f/0x60 [ 31.250030] hid_device_probe+0x122/0x1f0 [ 31.250053] really_probe+0xde/0x340 [ 31.250080] __driver_probe_device+0x78/0x110 [ 31.250105] driver_probe_device+0x1f/0xa0 [ 31.250132] __device_attach_driver+0x85/0x110 [ 31.250160] bus_for_each_drv+0x78/0xc0 [ 31.250184] __device_attach+0xb0/0x1b0 [ 31.250207] bus_probe_device+0x94/0xb0 [ 31.250230] device_add+0x64a/0x860 [ 31.250252] hid_add_device+0xe5/0x240 [ 31.250279] usbhid_probe+0x4dc/0x620 [ 31.250303] usb_probe_interface+0xe4/0x2a0 [ 31.250329] really_probe+0xde/0x340 [ 31.250353] __driver_probe_device+0x78/0x110 [ 31.250377] driver_probe_device+0x1f/0xa0 [ 31.250404] __device_attach_driver+0x85/0x110 [ 31.250431] bus_for_each_drv+0x78/0xc0 [ 31.250455] __device_attach+0xb0/0x1b0 [ 31.250478] bus_probe_device+0x94/0xb0 [ 31.250501] device_add+0x64a/0x860 [ 31.250523] usb_set_configuration+0x606/0x8a0 [ 31.250552] usb_generic_driver_probe+0x3e/0x60 [ 31.250579] usb_probe_device+0x3d/0x120 [ 31.250605] really_probe+0xde/0x340 [ 31.250629] __driver_probe_device+0x78/0x110 [ 31.250653] driver_probe_device+0x1f/0xa0 [ 31.250680] __device_attach_driver+0x85/0x110 [ 31.250707] bus_for_each_drv+0x78/0xc0 [ 31.250731] __device_attach+0xb0/0x1b0 [ 31.250753] bus_probe_device+0x94/0xb0 [ 31.250776] device_add+0x64a/0x860 [ 31.250798] usb_new_device.cold+0x141/0x38f [ 31.250828] hub_event+0x1166/0x1980 [ 31.250854] process_one_work+0x20f/0x580 [ 31.250879] worker_thread+0x1d1/0x3b0 [ 31.250904] kthread+0xee/0x120 [ 31.250926] ret_from_fork+0x30/0x50 [ 31.250954] ret_from_fork_asm+0x1a/0x30 [ 31.250982] -> #1 (triggers_list_lock){++++}-{4:4}: [ 31.251022] down_write+0x3b/0xd0 [ 31.251045] led_trigger_register+0x40/0x1b0 [ 31.251074] power_supply_register_led_trigger+0x88/0x150 [ 31.251107] power_supply_create_triggers+0x55/0xe0 [ 31.251135] __power_supply_register.part.0+0x34e/0x4a0 [ 31.251164] devm_power_supply_register+0x70/0xc0 [ 31.251190] bq27xxx_battery_setup+0x1a1/0x6d0 [bq27xxx_battery] [ 31.251235] bq27xxx_battery_i2c_probe+0xe5/0x17f [bq27xxx_battery_i2c] [ 31.251272] i2c_device_probe+0x125/0x2b0 [ 31.251299] really_probe+0xde/0x340 [ 31.251324] __driver_probe_device+0x78/0x110 [ 31.251348] driver_probe_device+0x1f/0xa0 [ 31.251375] __driver_attach+0xba/0x1c0 [ 31.251398] bus_for_each_dev+0x6b/0xb0 [ 31.251421] bus_add_driver+0x111/0x1f0 [ 31.251445] driver_register+0x6e/0xc0 [ 31.251470] i2c_register_driver+0x41/0xb0 [ 31.251498] do_one_initcall+0x5e/0x3a0 [ 31.251522] do_init_module+0x60/0x220 [ 31.251550] __do_sys_init_module+0x15f/0x190 [ 31.251575] do_syscall_64+0x93/0x180 [ 31.251598] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 31.251629] -> #0 (&psy->extensions_sem){.+.+}-{4:4}: [ 31.251668] __lock_acquire+0x13ce/0x21c0 [ 31.251694] lock_acquire+0xcf/0x2e0 [ 31.251719] down_read+0x3e/0x170 [ 31.251741] power_supply_get_property.part.0+0x22/0x150 [ 31.251774] power_supply_update_leds+0x8d/0x230 [ 31.251804] power_supply_led_trigger_activate+0x18/0x20 [ 31.251837] led_trigger_set+0x1fc/0x300 [ 31.251863] led_trigger_set_default+0x90/0xe0 [ 31.251892] led_classdev_register_ext+0x311/0x3a0 [ 31.251921] devm_led_classdev_multicolor_register_ext+0x6e/0xb80 [led_class_multicolor] [ 31.251969] ktd202x_probe+0x464/0x5c0 [leds_ktd202x] [ 31.252002] i2c_device_probe+0x125/0x2b0 [ 31.252027] really_probe+0xde/0x340 [ 31.252052] __driver_probe_device+0x78/0x110 [ 31.252076] driver_probe_device+0x1f/0xa0 [ 31.252103] __driver_attach+0xba/0x1c0 [ 31.252125] bus_for_each_dev+0x6b/0xb0 [ 31.252148] bus_add_driver+0x111/0x1f0 [ 31.252172] driver_register+0x6e/0xc0 [ 31.252197] i2c_register_driver+0x41/0xb0 [ 31.252225] do_one_initcall+0x5e/0x3a0 [ 31.252248] do_init_module+0x60/0x220 [ 31.252274] __do_sys_init_module+0x15f/0x190 [ 31.253986] do_syscall_64+0x93/0x180 [ 31.255826] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 31.257614] other info that might help us debug this: [ 31.257619] Chain exists of: &psy->extensions_sem --> triggers_list_lock --> &led_cdev->trigger_lock [ 31.257630] Possible unsafe locking scenario: [ 31.257632] CPU0 CPU1 [ 31.257633] ---- ---- [ 31.257634] lock(&led_cdev->trigger_lock); [ 31.257637] lock(triggers_list_lock); [ 31.257640] lock(&led_cdev->trigger_lock); [ 31.257643] rlock(&psy->extensions_sem); [ 31.257646] *** DEADLOCK *** [ 31.289433] 4 locks held by (udev-worker)/553: [ 31.289443] #0: ffff892ad9658108 (&dev->mutex){....}-{4:4}, at: __driver_attach+0xaf/0x1c0 [ 31.289463] #1: ffff892adbc0bbc8 (&led_cdev->led_access){+.+.}-{4:4}, at: led_classdev_register_ext+0x1c7/0x3a0 [ 31.289476] #2: ffffffffad0e30b0 (triggers_list_lock){++++}-{4:4}, at: led_trigger_set_default+0x2c/0xe0 [ 31.289487] #3: ffff892adbc0bad0 (&led_cdev->trigger_lock){+.+.}-{4:4}, at: led_trigger_set_default+0x34/0xe0 Fixes: 6037802bbae8 ("power: supply: core: implement extension API") Cc: Thomas Weißschuh Cc: Armin Wolf Signed-off-by: Hans de Goede Reviewed-by: Thomas Weißschuh Link: https://lore.kernel.org/r/20250130140035.20636-1-hdegoede@redhat.com Signed-off-by: Sebastian Reichel --- drivers/power/supply/power_supply_core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/power/supply/power_supply_core.c b/drivers/power/supply/power_supply_core.c index d0bb52a7a036..76c340b38015 100644 --- a/drivers/power/supply/power_supply_core.c +++ b/drivers/power/supply/power_supply_core.c @@ -1592,11 +1592,11 @@ __power_supply_register(struct device *parent, if (rc) goto register_thermal_failed; - scoped_guard(rwsem_read, &psy->extensions_sem) { - rc = power_supply_create_triggers(psy); - if (rc) - goto create_triggers_failed; + rc = power_supply_create_triggers(psy); + if (rc) + goto create_triggers_failed; + scoped_guard(rwsem_read, &psy->extensions_sem) { rc = power_supply_add_hwmon_sysfs(psy); if (rc) goto add_hwmon_sysfs_failed; From 98380110bd48fbfd6a798ee11fffff893d36062c Mon Sep 17 00:00:00 2001 From: Chris Morgan Date: Fri, 31 Jan 2025 17:14:51 -0600 Subject: [PATCH 0356/1443] power: supply: axp20x_battery: Fix fault handling for AXP717 Correct the fault handling for the AXP717 by changing the i2c write from regmap_update_bits() to regmap_write_bits(). The update bits function does not work properly on a RW1C register where we must write a 1 back to an existing register to clear it. Additionally, as part of this testing I confirmed the behavior of errors reappearing, so remove comment about assumptions. Fixes: 6625767049c2 ("power: supply: axp20x_battery: add support for AXP717") Signed-off-by: Chris Morgan Reviewed-by: Chen-Yu Tsai Link: https://lore.kernel.org/r/20250131231455.153447-2-macroalpha82@gmail.com Signed-off-by: Sebastian Reichel --- drivers/power/supply/axp20x_battery.c | 31 +++++++++++++-------------- 1 file changed, 15 insertions(+), 16 deletions(-) diff --git a/drivers/power/supply/axp20x_battery.c b/drivers/power/supply/axp20x_battery.c index fa27195f074e..3c3158f31a48 100644 --- a/drivers/power/supply/axp20x_battery.c +++ b/drivers/power/supply/axp20x_battery.c @@ -466,10 +466,9 @@ static int axp717_battery_get_prop(struct power_supply *psy, /* * If a fault is detected it must also be cleared; if the - * condition persists it should reappear (This is an - * assumption, it's actually not documented). A restart was - * not sufficient to clear the bit in testing despite the - * register listed as POR. + * condition persists it should reappear. A restart was not + * sufficient to clear the bit in testing despite the register + * listed as POR. */ case POWER_SUPPLY_PROP_HEALTH: ret = regmap_read(axp20x_batt->regmap, AXP717_PMU_FAULT, @@ -480,26 +479,26 @@ static int axp717_battery_get_prop(struct power_supply *psy, switch (reg & AXP717_BATT_PMU_FAULT_MASK) { case AXP717_BATT_UVLO_2_5V: val->intval = POWER_SUPPLY_HEALTH_DEAD; - regmap_update_bits(axp20x_batt->regmap, - AXP717_PMU_FAULT, - AXP717_BATT_UVLO_2_5V, - AXP717_BATT_UVLO_2_5V); + regmap_write_bits(axp20x_batt->regmap, + AXP717_PMU_FAULT, + AXP717_BATT_UVLO_2_5V, + AXP717_BATT_UVLO_2_5V); return 0; case AXP717_BATT_OVER_TEMP: val->intval = POWER_SUPPLY_HEALTH_HOT; - regmap_update_bits(axp20x_batt->regmap, - AXP717_PMU_FAULT, - AXP717_BATT_OVER_TEMP, - AXP717_BATT_OVER_TEMP); + regmap_write_bits(axp20x_batt->regmap, + AXP717_PMU_FAULT, + AXP717_BATT_OVER_TEMP, + AXP717_BATT_OVER_TEMP); return 0; case AXP717_BATT_UNDER_TEMP: val->intval = POWER_SUPPLY_HEALTH_COLD; - regmap_update_bits(axp20x_batt->regmap, - AXP717_PMU_FAULT, - AXP717_BATT_UNDER_TEMP, - AXP717_BATT_UNDER_TEMP); + regmap_write_bits(axp20x_batt->regmap, + AXP717_PMU_FAULT, + AXP717_BATT_UNDER_TEMP, + AXP717_BATT_UNDER_TEMP); return 0; default: From ac5a41b472b4ef8bb37d7550796d059b377b4646 Mon Sep 17 00:00:00 2001 From: Josua Mayer Date: Mon, 27 Jan 2025 21:12:02 +0100 Subject: [PATCH 0357/1443] Revert "mmc: sdhci_am654: Add sdhci_am654_start_signal_voltage_switch" This reverts commit 941a7abd4666912b84ab209396fdb54b0dae685d. This commit uses presence of device-tree properties vmmc-supply and vqmmc-supply for deciding whether to enable a quirk affecting timing of clock and data. The intention was to address issues observed with eMMC and SD on AM62 platforms. This new quirk is however also enabled for AM64 breaking microSD access on the SolidRun HimmingBoard-T which is supported in-tree since v6.11, causing a regression. During boot microSD initialization now fails with the error below: [ 2.008520] mmc1: SDHCI controller on fa00000.mmc [fa00000.mmc] using ADMA 64-bit [ 2.115348] mmc1: error -110 whilst initialising SD card The heuristics for enabling the quirk are clearly not correct as they break at least one but potentially many existing boards. Revert the change and restore original behaviour until a more appropriate method of selecting the quirk is derived. Fixes: 941a7abd4666 ("mmc: sdhci_am654: Add sdhci_am654_start_signal_voltage_switch") Closes: https://lore.kernel.org/linux-mmc/a70fc9fc-186f-4165-a652-3de50733763a@solid-run.com/ Cc: stable@vger.kernel.org Signed-off-by: Josua Mayer Acked-by: Adrian Hunter Link: https://lore.kernel.org/r/20250127-am654-mmc-regression-v2-1-9bb39fb12810@solid-run.com Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci_am654.c | 30 ------------------------------ 1 file changed, 30 deletions(-) diff --git a/drivers/mmc/host/sdhci_am654.c b/drivers/mmc/host/sdhci_am654.c index b73f673db92b..f75c31815ab0 100644 --- a/drivers/mmc/host/sdhci_am654.c +++ b/drivers/mmc/host/sdhci_am654.c @@ -155,7 +155,6 @@ struct sdhci_am654_data { u32 tuning_loop; #define SDHCI_AM654_QUIRK_FORCE_CDTEST BIT(0) -#define SDHCI_AM654_QUIRK_SUPPRESS_V1P8_ENA BIT(1) }; struct window { @@ -357,29 +356,6 @@ static void sdhci_j721e_4bit_set_clock(struct sdhci_host *host, sdhci_set_clock(host, clock); } -static int sdhci_am654_start_signal_voltage_switch(struct mmc_host *mmc, struct mmc_ios *ios) -{ - struct sdhci_host *host = mmc_priv(mmc); - struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); - struct sdhci_am654_data *sdhci_am654 = sdhci_pltfm_priv(pltfm_host); - int ret; - - if ((sdhci_am654->quirks & SDHCI_AM654_QUIRK_SUPPRESS_V1P8_ENA) && - ios->signal_voltage == MMC_SIGNAL_VOLTAGE_180) { - if (!IS_ERR(mmc->supply.vqmmc)) { - ret = mmc_regulator_set_vqmmc(mmc, ios); - if (ret < 0) { - pr_err("%s: Switching to 1.8V signalling voltage failed,\n", - mmc_hostname(mmc)); - return -EIO; - } - } - return 0; - } - - return sdhci_start_signal_voltage_switch(mmc, ios); -} - static u8 sdhci_am654_write_power_on(struct sdhci_host *host, u8 val, int reg) { writeb(val, host->ioaddr + reg); @@ -868,11 +844,6 @@ static int sdhci_am654_get_of_property(struct platform_device *pdev, if (device_property_read_bool(dev, "ti,fails-without-test-cd")) sdhci_am654->quirks |= SDHCI_AM654_QUIRK_FORCE_CDTEST; - /* Suppress v1p8 ena for eMMC and SD with vqmmc supply */ - if (!!of_parse_phandle(dev->of_node, "vmmc-supply", 0) == - !!of_parse_phandle(dev->of_node, "vqmmc-supply", 0)) - sdhci_am654->quirks |= SDHCI_AM654_QUIRK_SUPPRESS_V1P8_ENA; - sdhci_get_of_property(pdev); return 0; @@ -969,7 +940,6 @@ static int sdhci_am654_probe(struct platform_device *pdev) goto err_pltfm_free; } - host->mmc_host_ops.start_signal_voltage_switch = sdhci_am654_start_signal_voltage_switch; host->mmc_host_ops.execute_tuning = sdhci_am654_execute_tuning; pm_runtime_get_noresume(dev); From 3e68abf2b9cebe76c6cd4b1aca8e95cd671035a3 Mon Sep 17 00:00:00 2001 From: Andy-ld Lu Date: Thu, 23 Jan 2025 17:26:01 +0800 Subject: [PATCH 0358/1443] mmc: mtk-sd: Fix register settings for hs400(es) mode For hs400(es) mode, the 'hs400-ds-delay' is typically configured in the dts. However, some projects may only define 'mediatek,hs400-ds-dly3', which can lead to initialization failures in hs400es mode. CMD13 reported response crc error in the mmc_switch_status() just after switching to hs400es mode. [ 1.914038][ T82] mmc0: mmc_select_hs400es failed, error -84 [ 1.914954][ T82] mmc0: error -84 whilst initialising MMC card Currently, the hs400_ds_dly3 value is set within the tuning function. This means that the PAD_DS_DLY3 field is not configured before tuning process, which is the reason for the above-mentioned CMD13 response crc error. Move the PAD_DS_DLY3 field configuration into msdc_prepare_hs400_tuning(), and add a value check of hs400_ds_delay to prevent overwriting by zero when the 'hs400-ds-delay' is not set in the dts. In addition, since hs400(es) only tune the PAD_DS_DLY1, the PAD_DS_DLY2_SEL bit should be cleared to bypass it. Fixes: c4ac38c6539b ("mmc: mtk-sd: Add HS400 online tuning support") Signed-off-by: Andy-ld Lu Reviewed-by: AngeloGioacchino Del Regno Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250123092644.7359-1-andy-ld.lu@mediatek.com Signed-off-by: Ulf Hansson --- drivers/mmc/host/mtk-sd.c | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/drivers/mmc/host/mtk-sd.c b/drivers/mmc/host/mtk-sd.c index 4b6e91372526..345ea91629e0 100644 --- a/drivers/mmc/host/mtk-sd.c +++ b/drivers/mmc/host/mtk-sd.c @@ -273,6 +273,7 @@ #define MSDC_PAD_TUNE_CMD2_SEL BIT(21) /* RW */ #define PAD_DS_TUNE_DLY_SEL BIT(0) /* RW */ +#define PAD_DS_TUNE_DLY2_SEL BIT(1) /* RW */ #define PAD_DS_TUNE_DLY1 GENMASK(6, 2) /* RW */ #define PAD_DS_TUNE_DLY2 GENMASK(11, 7) /* RW */ #define PAD_DS_TUNE_DLY3 GENMASK(16, 12) /* RW */ @@ -318,6 +319,7 @@ /* EMMC50_PAD_DS_TUNE mask */ #define PAD_DS_DLY_SEL BIT(16) /* RW */ +#define PAD_DS_DLY2_SEL BIT(15) /* RW */ #define PAD_DS_DLY1 GENMASK(14, 10) /* RW */ #define PAD_DS_DLY3 GENMASK(4, 0) /* RW */ @@ -2504,13 +2506,23 @@ static int msdc_execute_tuning(struct mmc_host *mmc, u32 opcode) static int msdc_prepare_hs400_tuning(struct mmc_host *mmc, struct mmc_ios *ios) { struct msdc_host *host = mmc_priv(mmc); + host->hs400_mode = true; - if (host->top_base) - writel(host->hs400_ds_delay, - host->top_base + EMMC50_PAD_DS_TUNE); - else - writel(host->hs400_ds_delay, host->base + PAD_DS_TUNE); + if (host->top_base) { + if (host->hs400_ds_dly3) + sdr_set_field(host->top_base + EMMC50_PAD_DS_TUNE, + PAD_DS_DLY3, host->hs400_ds_dly3); + if (host->hs400_ds_delay) + writel(host->hs400_ds_delay, + host->top_base + EMMC50_PAD_DS_TUNE); + } else { + if (host->hs400_ds_dly3) + sdr_set_field(host->base + PAD_DS_TUNE, + PAD_DS_TUNE_DLY3, host->hs400_ds_dly3); + if (host->hs400_ds_delay) + writel(host->hs400_ds_delay, host->base + PAD_DS_TUNE); + } /* hs400 mode must set it to 0 */ sdr_clr_bits(host->base + MSDC_PATCH_BIT2, MSDC_PATCH_BIT2_CFGCRCSTS); /* to improve read performance, set outstanding to 2 */ @@ -2530,14 +2542,11 @@ static int msdc_execute_hs400_tuning(struct mmc_host *mmc, struct mmc_card *card if (host->top_base) { sdr_set_bits(host->top_base + EMMC50_PAD_DS_TUNE, PAD_DS_DLY_SEL); - if (host->hs400_ds_dly3) - sdr_set_field(host->top_base + EMMC50_PAD_DS_TUNE, - PAD_DS_DLY3, host->hs400_ds_dly3); + sdr_clr_bits(host->top_base + EMMC50_PAD_DS_TUNE, + PAD_DS_DLY2_SEL); } else { sdr_set_bits(host->base + PAD_DS_TUNE, PAD_DS_TUNE_DLY_SEL); - if (host->hs400_ds_dly3) - sdr_set_field(host->base + PAD_DS_TUNE, - PAD_DS_TUNE_DLY3, host->hs400_ds_dly3); + sdr_clr_bits(host->base + PAD_DS_TUNE, PAD_DS_TUNE_DLY2_SEL); } host->hs400_tuning = true; From 707e860119d5510e897da89492f8a69ae627762c Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:01 +0100 Subject: [PATCH 0359/1443] drm/ast: Remove 1152x864 from list of widescreen resolutions The resolution 1152x864 has a ratio of 4:3 and is already handled as such. Remove it from the list of widescreen resolutions. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-2-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 9d5321c81e68..b533eb6e233e 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1035,8 +1035,6 @@ ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode return MODE_OK; if ((mode->hdisplay == 1600) && (mode->vdisplay == 900)) return MODE_OK; - if ((mode->hdisplay == 1152) && (mode->vdisplay == 864)) - return MODE_OK; if ((ast->chip == AST2100) || // GEN2, but not AST1100 (?) (ast->chip == AST2200) || // GEN3, but not AST2150 (?) From fdebbb2cec9d53aa6b83847d3762c7783514273a Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:02 +0100 Subject: [PATCH 0360/1443] drm/ast: Align naming in widescreen detection code to manual Rename variables and register constants to align with the programming manual. Add new constants where necessary. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-3-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_drv.c | 2 +- drivers/gpu/drm/ast/ast_main.c | 8 ++++---- drivers/gpu/drm/ast/ast_post.c | 2 +- drivers/gpu/drm/ast/ast_reg.h | 11 ++++++----- 4 files changed, 12 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index cddd69972e89..6fbf62a99c48 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -170,7 +170,7 @@ static int ast_detect_chip(struct pci_dev *pdev, /* Patch AST2500/AST2510 */ if ((pdev->revision & 0xf0) == 0x40) { - if (!(vgacrd0 & AST_VRAM_INIT_STATUS_MASK)) + if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK)) ast_patch_ahb_2500(regs); } diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index ba69280b33e7..2a813f0128eb 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -38,7 +38,7 @@ static void ast_detect_widescreen(struct ast_device *ast) { - u8 jreg; + u8 vgacrd0; /* Check if we support wide screen */ switch (AST_GEN(ast)) { @@ -46,10 +46,10 @@ static void ast_detect_widescreen(struct ast_device *ast) ast->support_wide_screen = false; break; default: - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - if (!(jreg & 0x80)) + vgacrd0 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd0); + if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_BY_BMC)) ast->support_wide_screen = true; - else if (jreg & 0x01) + else if (vgacrd0 & AST_IO_VGACRD0_IKVM_WIDESCREEN) ast->support_wide_screen = true; else { ast->support_wide_screen = false; diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 0daa8e52a092..91e85e457bdf 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -2062,7 +2062,7 @@ void ast_post_chip_2500(struct ast_device *ast) u8 reg; reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - if ((reg & AST_VRAM_INIT_STATUS_MASK) == 0) {/* vga only */ + if ((reg & AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK) == 0) {/* vga only */ /* Clear bus lock condition */ ast_patch_ahb_2500(ast->regs); diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h index 0745d58e5b45..b4ff38949a56 100644 --- a/drivers/gpu/drm/ast/ast_reg.h +++ b/drivers/gpu/drm/ast/ast_reg.h @@ -38,6 +38,12 @@ #define AST_IO_VGACRCB_HWC_16BPP BIT(0) /* set: ARGB4444, cleared: 2bpp palette */ #define AST_IO_VGACRCB_HWC_ENABLED BIT(1) +/* mirrors SCU100[7:0] */ +#define AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK GENMASK(7, 6) +#define AST_IO_VGACRD0_VRAM_INIT_BY_BMC BIT(7) +#define AST_IO_VGACRD0_VRAM_INIT_READY BIT(6) +#define AST_IO_VGACRD0_IKVM_WIDESCREEN BIT(0) + #define AST_IO_VGACRD1_MCU_FW_EXECUTING BIT(5) /* Display Transmitter Type */ #define AST_IO_VGACRD1_TX_TYPE_MASK GENMASK(3, 1) @@ -61,11 +67,6 @@ #define AST_IO_VGAIR1_R (0x5A) #define AST_IO_VGAIR1_VREFRESH BIT(3) - -#define AST_VRAM_INIT_STATUS_MASK GENMASK(7, 6) -//#define AST_VRAM_INIT_BY_BMC BIT(7) -//#define AST_VRAM_INIT_READY BIT(6) - /* * AST DisplayPort */ From 219c6a4a6f6eada64773bc250cc02c60b5358278 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:03 +0100 Subject: [PATCH 0361/1443] drm/ast: Rename support_wide_screen to support_wsxga_p The flag support_wide_screen in struct ast_device selects devices that support resolutions of at least 1680x1050 (WSXGA+). Name it accordingly. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-4-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_drv.h | 2 +- drivers/gpu/drm/ast/ast_main.c | 16 ++++++++-------- drivers/gpu/drm/ast/ast_mode.c | 2 +- 3 files changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index cf9edef8fca6..ec5b204be1da 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -205,7 +205,7 @@ struct ast_device { } astdp; } output; - bool support_wide_screen; + bool support_wsxga_p; /* 1680x1050 */ u8 *dp501_fw_addr; const struct firmware *dp501_fw; /* dp501 fw */ diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 2a813f0128eb..1cfbe404e5a0 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -43,24 +43,24 @@ static void ast_detect_widescreen(struct ast_device *ast) /* Check if we support wide screen */ switch (AST_GEN(ast)) { case 1: - ast->support_wide_screen = false; + ast->support_wsxga_p = false; break; default: vgacrd0 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd0); if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_BY_BMC)) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; else if (vgacrd0 & AST_IO_VGACRD0_IKVM_WIDESCREEN) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; else { - ast->support_wide_screen = false; + ast->support_wsxga_p = false; if (ast->chip == AST1300) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; if (ast->chip == AST1400) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; if (ast->chip == AST2510) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; if (IS_AST_GEN7(ast)) - ast->support_wide_screen = true; + ast->support_wsxga_p = true; } break; } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index b533eb6e233e..d1b3136b3769 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1024,7 +1024,7 @@ ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode enum drm_mode_status status; uint32_t jtemp; - if (ast->support_wide_screen) { + if (ast->support_wsxga_p) { if ((mode->hdisplay == 1680) && (mode->vdisplay == 1050)) return MODE_OK; if ((mode->hdisplay == 1280) && (mode->vdisplay == 800)) From 1bb3f70c418f8ac51a9f19b1f0fe6ddd889794e3 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:04 +0100 Subject: [PATCH 0362/1443] drm/ast: Reorganize widescreen test around hardware Gens Testing for support of widescreen modes mixes up various hardware Gens. First branch by hardware Gen, then do specific tests for each Gen. By default, widesscreen support is disabled. Later patches will add more specific tests for each Gen. v2: - move shared detection code into helper (Jocelyn) Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-5-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_main.c | 54 ++++++++++++++++++++-------------- 1 file changed, 32 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 1cfbe404e5a0..93ae9a275c96 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -36,33 +36,43 @@ #include "ast_drv.h" +/* Try to detect WSXGA+ on Gen2+ */ +static bool __ast_2100_detect_wsxga_p(struct ast_device *ast) +{ + u8 vgacrd0 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd0); + + if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_BY_BMC)) + return true; + if (vgacrd0 & AST_IO_VGACRD0_IKVM_WIDESCREEN) + return true; + + return false; +} + static void ast_detect_widescreen(struct ast_device *ast) { - u8 vgacrd0; + ast->support_wsxga_p = false; - /* Check if we support wide screen */ - switch (AST_GEN(ast)) { - case 1: - ast->support_wsxga_p = false; - break; - default: - vgacrd0 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd0); - if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_BY_BMC)) + if (AST_GEN(ast) >= 7) { + ast->support_wsxga_p = true; + } else if (AST_GEN(ast) >= 6) { + if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; - else if (vgacrd0 & AST_IO_VGACRD0_IKVM_WIDESCREEN) + else if (ast->chip == AST2510) + ast->support_wsxga_p = true; + } else if (AST_GEN(ast) >= 5) { + if (__ast_2100_detect_wsxga_p(ast)) + ast->support_wsxga_p = true; + else if (ast->chip == AST1400) + ast->support_wsxga_p = true; + } else if (AST_GEN(ast) >= 4) { + if (__ast_2100_detect_wsxga_p(ast)) + ast->support_wsxga_p = true; + else if (ast->chip == AST1300) + ast->support_wsxga_p = true; + } else if (AST_GEN(ast) >= 2) { + if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; - else { - ast->support_wsxga_p = false; - if (ast->chip == AST1300) - ast->support_wsxga_p = true; - if (ast->chip == AST1400) - ast->support_wsxga_p = true; - if (ast->chip == AST2510) - ast->support_wsxga_p = true; - if (IS_AST_GEN7(ast)) - ast->support_wsxga_p = true; - } - break; } } From 7a74caabe1013b482b1db10751c79bc75bdaae8c Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:05 +0100 Subject: [PATCH 0363/1443] drm/ast: Add support_fullhd flag to struct ast_device Detect support for 1920x1080 (FullHD) in ast_detect_widescreen(). The flag is cleared by default. The test logic has been taken from existing code in ast_crtc_helper_mode_valid(). The code in that function is being replaced by the new flag. For Gen3, a new branch duplicates the Gen2 logic and adds a test for AST2200. Gen2 adds a test for AST2100. v2: - use fullhd flag for setting max width/height Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-6-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_drv.h | 1 + drivers/gpu/drm/ast/ast_main.c | 19 +++++++++++++++++++ drivers/gpu/drm/ast/ast_mode.c | 12 ++---------- 3 files changed, 22 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index ec5b204be1da..01ef0c0d5db7 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -206,6 +206,7 @@ struct ast_device { } output; bool support_wsxga_p; /* 1680x1050 */ + bool support_fullhd; /* 1920x1080 */ u8 *dp501_fw_addr; const struct firmware *dp501_fw; /* dp501 fw */ diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 93ae9a275c96..6e964a0714b4 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -52,27 +52,46 @@ static bool __ast_2100_detect_wsxga_p(struct ast_device *ast) static void ast_detect_widescreen(struct ast_device *ast) { ast->support_wsxga_p = false; + ast->support_fullhd = false; if (AST_GEN(ast) >= 7) { ast->support_wsxga_p = true; + ast->support_fullhd = true; } else if (AST_GEN(ast) >= 6) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; else if (ast->chip == AST2510) ast->support_wsxga_p = true; + if (ast->support_wsxga_p) + ast->support_fullhd = true; } else if (AST_GEN(ast) >= 5) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; else if (ast->chip == AST1400) ast->support_wsxga_p = true; + if (ast->support_wsxga_p) + ast->support_fullhd = true; } else if (AST_GEN(ast) >= 4) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; else if (ast->chip == AST1300) ast->support_wsxga_p = true; + if (ast->support_wsxga_p) + ast->support_fullhd = true; + } else if (AST_GEN(ast) >= 3) { + if (__ast_2100_detect_wsxga_p(ast)) + ast->support_wsxga_p = true; + if (ast->support_wsxga_p) { + if (ast->chip == AST2200) + ast->support_fullhd = true; + } } else if (AST_GEN(ast) >= 2) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; + if (ast->support_wsxga_p) { + if (ast->chip == AST2100) + ast->support_fullhd = true; + } } } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index d1b3136b3769..dda2c4fb0a48 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1036,10 +1036,7 @@ ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode if ((mode->hdisplay == 1600) && (mode->vdisplay == 900)) return MODE_OK; - if ((ast->chip == AST2100) || // GEN2, but not AST1100 (?) - (ast->chip == AST2200) || // GEN3, but not AST2150 (?) - IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || - IS_AST_GEN6(ast) || IS_AST_GEN7(ast)) { + if (ast->support_fullhd) { if ((mode->hdisplay == 1920) && (mode->vdisplay == 1080)) return MODE_OK; @@ -1371,12 +1368,7 @@ int ast_mode_config_init(struct ast_device *ast) dev->mode_config.min_height = 0; dev->mode_config.preferred_depth = 24; - if (ast->chip == AST2100 || // GEN2, but not AST1100 (?) - ast->chip == AST2200 || // GEN3, but not AST2150 (?) - IS_AST_GEN7(ast) || - IS_AST_GEN6(ast) || - IS_AST_GEN5(ast) || - IS_AST_GEN4(ast)) { + if (ast->support_fullhd) { dev->mode_config.max_width = 1920; dev->mode_config.max_height = 2048; } else { From 454bdacccac4a020f931bd39a8b8744f765b0ecd Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:06 +0100 Subject: [PATCH 0364/1443] drm/ast: Add support_wuxga flag to struct ast_device Detect support for 1920x1200 (WUXGA) in ast_detect_widescreen(). The flag is cleared by default. The test logic has been taken from existing code in ast_crtc_helper_mode_valid(). The code in that function is being replaced by the new flag. v2: - move shared detection code into helper (Jocelyn) Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-7-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_drv.h | 1 + drivers/gpu/drm/ast/ast_main.c | 27 +++++++++++++++++++++++++++ drivers/gpu/drm/ast/ast_mode.c | 8 +++----- drivers/gpu/drm/ast/ast_reg.h | 1 + 4 files changed, 32 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 01ef0c0d5db7..4e3a88f8a85c 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -207,6 +207,7 @@ struct ast_device { bool support_wsxga_p; /* 1680x1050 */ bool support_fullhd; /* 1920x1080 */ + bool support_wuxga; /* 1920x1200 */ u8 *dp501_fw_addr; const struct firmware *dp501_fw; /* dp501 fw */ diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 6e964a0714b4..44b9b5f659fc 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -49,14 +49,31 @@ static bool __ast_2100_detect_wsxga_p(struct ast_device *ast) return false; } +/* Try to detect WUXGA on Gen2+ */ +static bool __ast_2100_detect_wuxga(struct ast_device *ast) +{ + u8 vgacrd1; + + if (ast->support_fullhd) { + vgacrd1 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xd1); + if (!(vgacrd1 & AST_IO_VGACRD1_SUPPORTS_WUXGA)) + return true; + } + + return false; +} + static void ast_detect_widescreen(struct ast_device *ast) { ast->support_wsxga_p = false; ast->support_fullhd = false; + ast->support_wuxga = false; if (AST_GEN(ast) >= 7) { ast->support_wsxga_p = true; ast->support_fullhd = true; + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } else if (AST_GEN(ast) >= 6) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; @@ -64,6 +81,8 @@ static void ast_detect_widescreen(struct ast_device *ast) ast->support_wsxga_p = true; if (ast->support_wsxga_p) ast->support_fullhd = true; + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } else if (AST_GEN(ast) >= 5) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; @@ -71,6 +90,8 @@ static void ast_detect_widescreen(struct ast_device *ast) ast->support_wsxga_p = true; if (ast->support_wsxga_p) ast->support_fullhd = true; + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } else if (AST_GEN(ast) >= 4) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; @@ -78,6 +99,8 @@ static void ast_detect_widescreen(struct ast_device *ast) ast->support_wsxga_p = true; if (ast->support_wsxga_p) ast->support_fullhd = true; + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } else if (AST_GEN(ast) >= 3) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; @@ -85,6 +108,8 @@ static void ast_detect_widescreen(struct ast_device *ast) if (ast->chip == AST2200) ast->support_fullhd = true; } + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } else if (AST_GEN(ast) >= 2) { if (__ast_2100_detect_wsxga_p(ast)) ast->support_wsxga_p = true; @@ -92,6 +117,8 @@ static void ast_detect_widescreen(struct ast_device *ast) if (ast->chip == AST2100) ast->support_fullhd = true; } + if (__ast_2100_detect_wuxga(ast)) + ast->support_wuxga = true; } } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index dda2c4fb0a48..bc0c7db5ad46 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1022,7 +1022,6 @@ ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode { struct ast_device *ast = to_ast_device(crtc->dev); enum drm_mode_status status; - uint32_t jtemp; if (ast->support_wsxga_p) { if ((mode->hdisplay == 1680) && (mode->vdisplay == 1050)) @@ -1041,11 +1040,10 @@ ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode return MODE_OK; if ((mode->hdisplay == 1920) && (mode->vdisplay == 1200)) { - jtemp = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, 0xff); - if (jtemp & 0x01) - return MODE_NOMODE; - else + if (ast->support_wuxga) return MODE_OK; + else + return MODE_NOMODE; } } } diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h index b4ff38949a56..9db0d584652a 100644 --- a/drivers/gpu/drm/ast/ast_reg.h +++ b/drivers/gpu/drm/ast/ast_reg.h @@ -55,6 +55,7 @@ #define AST_IO_VGACRD1_TX_ANX9807_VBIOS 0x0a #define AST_IO_VGACRD1_TX_FW_EMBEDDED_FW 0x0c /* special case of DP501 */ #define AST_IO_VGACRD1_TX_ASTDP 0x0e +#define AST_IO_VGACRD1_SUPPORTS_WUXGA BIT(0) #define AST_IO_VGACRD7_EDID_VALID_FLAG BIT(0) #define AST_IO_VGACRDC_LINK_SUCCESS BIT(0) From 255b3ff65d8ffd70db8c29dda6c8765d58fea2b6 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:07 +0100 Subject: [PATCH 0365/1443] drm/ast: Always validate H/V sync flags The ast driver matches DRM display modes against an internal list of modes supported by the VBIOS. Matching H/V sync flags between modes is preferred, but optional. If sync flags are not matching, the driver would program the VBIOS settings to hardware and let the display handle the difference. DRM modes are generated from attached displays or standard mode lines. Therefore differences to the VBIOS modes are not just cosmetical, but signal possible incompatibility with the display hardware. Hence make matching H/V sync flags mandatory. If the VBIOS does not support a certain mode, we should report it as unsupported. Note that the VBIOS mode tables all appear to refer to standard modes. (If sync flags really make no difference to the VBIOS, the ast driver shouldn't match them in the first place.) Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-8-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 41 +++++++++++++--------------------- 1 file changed, 16 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index bc0c7db5ad46..a484292c9645 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -113,8 +113,8 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, { u32 refresh_rate_index = 0, refresh_rate; const struct ast_vbios_enhtable *best = NULL; + const struct ast_vbios_enhtable *loop; u32 hborder, vborder; - bool check_sync; switch (format->cpp[0] * 8) { case 8: @@ -176,36 +176,27 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, } refresh_rate = drm_mode_vrefresh(mode); - check_sync = vbios_mode->enh_table->flags & WideScreenMode; - while (1) { - const struct ast_vbios_enhtable *loop = vbios_mode->enh_table; + loop = vbios_mode->enh_table; - while (loop->refresh_rate != 0xff) { - if ((check_sync) && - (((mode->flags & DRM_MODE_FLAG_NVSYNC) && - (loop->flags & PVSync)) || - ((mode->flags & DRM_MODE_FLAG_PVSYNC) && - (loop->flags & NVSync)) || - ((mode->flags & DRM_MODE_FLAG_NHSYNC) && - (loop->flags & PHSync)) || - ((mode->flags & DRM_MODE_FLAG_PHSYNC) && - (loop->flags & NHSync)))) { - loop++; - continue; - } - if (loop->refresh_rate <= refresh_rate - && (!best || loop->refresh_rate > best->refresh_rate)) - best = loop; + while (loop->refresh_rate != 0xff) { + if (((mode->flags & DRM_MODE_FLAG_NVSYNC) && (loop->flags & PVSync)) || + ((mode->flags & DRM_MODE_FLAG_PVSYNC) && (loop->flags & NVSync)) || + ((mode->flags & DRM_MODE_FLAG_NHSYNC) && (loop->flags & PHSync)) || + ((mode->flags & DRM_MODE_FLAG_PHSYNC) && (loop->flags & NHSync))) { loop++; + continue; } - if (best || !check_sync) - break; - check_sync = 0; + if (loop->refresh_rate <= refresh_rate && + (!best || loop->refresh_rate > best->refresh_rate)) + best = loop; + loop++; } - if (best) - vbios_mode->enh_table = best; + if (!best) + return false; + + vbios_mode->enh_table = best; hborder = (vbios_mode->enh_table->flags & HBorder) ? 8 : 0; vborder = (vbios_mode->enh_table->flags & VBorder) ? 8 : 0; From c81202906b5cd56db403e95db3d29c9dfc8c74c1 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:08 +0100 Subject: [PATCH 0366/1443] drm/ast: Find VBIOS mode from regular display size The ast driver looks up supplied display modes from an internal list of display modes supported by the VBIOS. Do not use the crtc_-prefixed display values from struct drm_display_mode for looking up the VBIOS mode. The fields contain raw values that the driver programs to hardware. They are affected by display settings like double-scan or interlace. Instead use the regular vdisplay and hdisplay fields for lookup. As the programmed values can now differ from the values used for lookup, set struct drm_display_mode.crtc_vdisplay and .crtc_hdisplay from the VBIOS mode. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-9-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index a484292c9645..961bc2efbc02 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -131,7 +131,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, return false; } - switch (mode->crtc_hdisplay) { + switch (mode->hdisplay) { case 640: vbios_mode->enh_table = &res_640x480[refresh_rate_index]; break; @@ -145,7 +145,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, vbios_mode->enh_table = &res_1152x864[refresh_rate_index]; break; case 1280: - if (mode->crtc_vdisplay == 800) + if (mode->vdisplay == 800) vbios_mode->enh_table = &res_1280x800[refresh_rate_index]; else vbios_mode->enh_table = &res_1280x1024[refresh_rate_index]; @@ -157,7 +157,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, vbios_mode->enh_table = &res_1440x900[refresh_rate_index]; break; case 1600: - if (mode->crtc_vdisplay == 900) + if (mode->vdisplay == 900) vbios_mode->enh_table = &res_1600x900[refresh_rate_index]; else vbios_mode->enh_table = &res_1600x1200[refresh_rate_index]; @@ -166,7 +166,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, vbios_mode->enh_table = &res_1680x1050[refresh_rate_index]; break; case 1920: - if (mode->crtc_vdisplay == 1080) + if (mode->vdisplay == 1080) vbios_mode->enh_table = &res_1920x1080[refresh_rate_index]; else vbios_mode->enh_table = &res_1920x1200[refresh_rate_index]; @@ -201,6 +201,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, hborder = (vbios_mode->enh_table->flags & HBorder) ? 8 : 0; vborder = (vbios_mode->enh_table->flags & VBorder) ? 8 : 0; + adjusted_mode->crtc_hdisplay = vbios_mode->enh_table->hde; adjusted_mode->crtc_htotal = vbios_mode->enh_table->ht; adjusted_mode->crtc_hblank_start = vbios_mode->enh_table->hde + hborder; adjusted_mode->crtc_hblank_end = vbios_mode->enh_table->ht - hborder; @@ -210,6 +211,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, vbios_mode->enh_table->hfp + vbios_mode->enh_table->hsync); + adjusted_mode->crtc_vdisplay = vbios_mode->enh_table->vde; adjusted_mode->crtc_vtotal = vbios_mode->enh_table->vt; adjusted_mode->crtc_vblank_start = vbios_mode->enh_table->vde + vborder; adjusted_mode->crtc_vblank_end = vbios_mode->enh_table->vt - vborder; From 192cc9f2b74a8e3b7d7cc7f4d041698d7ba63efa Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:09 +0100 Subject: [PATCH 0367/1443] drm/ast: Add empty initializer for VBIOS modes VBIOS mode tables are terminated by an entry with a refresh rate of 0xff. The code is hard to read and fragile to use. Therefore create an empty entry with AST_VBIOS_MODE_INVALID to terminate each mode list. Stop at the invalid entry when searching for modes in the tables. Instead of testing for refresh == 0xff, test with a helper function if the mode's size and refresh have meaningful values. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-10-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 2 +- drivers/gpu/drm/ast/ast_tables.h | 52 ++++++++++++++------------------ 2 files changed, 24 insertions(+), 30 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 961bc2efbc02..3f437f871a35 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -179,7 +179,7 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, loop = vbios_mode->enh_table; - while (loop->refresh_rate != 0xff) { + while (ast_vbios_mode_is_valid(loop)) { if (((mode->flags & DRM_MODE_FLAG_NVSYNC) && (loop->flags & PVSync)) || ((mode->flags & DRM_MODE_FLAG_PVSYNC) && (loop->flags & NVSync)) || ((mode->flags & DRM_MODE_FLAG_NHSYNC) && (loop->flags & PHSync)) || diff --git a/drivers/gpu/drm/ast/ast_tables.h b/drivers/gpu/drm/ast/ast_tables.h index 0378c9bc079b..4367817b2f80 100644 --- a/drivers/gpu/drm/ast/ast_tables.h +++ b/drivers/gpu/drm/ast/ast_tables.h @@ -24,6 +24,8 @@ #ifndef AST_TABLES_H #define AST_TABLES_H +#include "ast_drv.h" + /* Std. Table Index Definition */ #define TextModeIndex 0 #define EGAModeIndex 1 @@ -212,6 +214,14 @@ static const struct ast_vbios_stdtable vbios_stdtable[] = { }, }; +#define AST_VBIOS_INVALID_MODE \ + {0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u} + +static inline bool ast_vbios_mode_is_valid(const struct ast_vbios_enhtable *vmode) +{ + return vmode->ht && vmode->vt && vmode->refresh_rate; +} + static const struct ast_vbios_enhtable res_640x480[] = { { 800, 640, 8, 96, 525, 480, 2, 2, VCLK25_175, /* 60Hz */ (SyncNN | HBorder | VBorder | Charx8Dot), 60, 1, 0x2E }, @@ -221,8 +231,7 @@ static const struct ast_vbios_enhtable res_640x480[] = { (SyncNN | Charx8Dot) , 75, 3, 0x2E }, { 832, 640, 56, 56, 509, 480, 1, 3, VCLK36, /* 85Hz */ (SyncNN | Charx8Dot) , 85, 4, 0x2E }, - { 832, 640, 56, 56, 509, 480, 1, 3, VCLK36, /* end */ - (SyncNN | Charx8Dot) , 0xFF, 4, 0x2E }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_800x600[] = { @@ -236,8 +245,7 @@ static const struct ast_vbios_enhtable res_800x600[] = { (SyncPP | Charx8Dot), 75, 4, 0x30 }, {1048, 800, 32, 64, 631, 600, 1, 3, VCLK56_25, /* 85Hz */ (SyncPP | Charx8Dot), 84, 5, 0x30 }, - {1048, 800, 32, 64, 631, 600, 1, 3, VCLK56_25, /* end */ - (SyncPP | Charx8Dot), 0xFF, 5, 0x30 }, + AST_VBIOS_INVALID_MODE, /* end */ }; @@ -250,8 +258,7 @@ static const struct ast_vbios_enhtable res_1024x768[] = { (SyncPP | Charx8Dot), 75, 3, 0x31 }, {1376, 1024, 48, 96, 808, 768, 1, 3, VCLK94_5, /* 85Hz */ (SyncPP | Charx8Dot), 84, 4, 0x31 }, - {1376, 1024, 48, 96, 808, 768, 1, 3, VCLK94_5, /* end */ - (SyncPP | Charx8Dot), 0xFF, 4, 0x31 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1280x1024[] = { @@ -261,31 +268,26 @@ static const struct ast_vbios_enhtable res_1280x1024[] = { (SyncPP | Charx8Dot), 75, 2, 0x32 }, {1728, 1280, 64, 160, 1072, 1024, 1, 3, VCLK157_5, /* 85Hz */ (SyncPP | Charx8Dot), 85, 3, 0x32 }, - {1728, 1280, 64, 160, 1072, 1024, 1, 3, VCLK157_5, /* end */ - (SyncPP | Charx8Dot), 0xFF, 3, 0x32 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1600x1200[] = { {2160, 1600, 64, 192, 1250, 1200, 1, 3, VCLK162, /* 60Hz */ (SyncPP | Charx8Dot), 60, 1, 0x33 }, - {2160, 1600, 64, 192, 1250, 1200, 1, 3, VCLK162, /* end */ - (SyncPP | Charx8Dot), 0xFF, 1, 0x33 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1152x864[] = { {1600, 1152, 64, 128, 900, 864, 1, 3, VCLK108, /* 75Hz */ (SyncPP | Charx8Dot | NewModeInfo), 75, 1, 0x3B }, - {1600, 1152, 64, 128, 900, 864, 1, 3, VCLK108, /* end */ - (SyncPP | Charx8Dot | NewModeInfo), 0xFF, 1, 0x3B }, + AST_VBIOS_INVALID_MODE, /* end */ }; /* 16:9 */ static const struct ast_vbios_enhtable res_1360x768[] = { {1792, 1360, 64, 112, 795, 768, 3, 6, VCLK85_5, /* 60Hz */ (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x39 }, - {1792, 1360, 64, 112, 795, 768, 3, 6, VCLK85_5, /* end */ - (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 0xFF, 1, 0x39 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1600x900[] = { @@ -294,17 +296,14 @@ static const struct ast_vbios_enhtable res_1600x900[] = { AST2500PreCatchCRT), 60, 1, 0x3A }, {2112, 1600, 88, 168, 934, 900, 3, 5, VCLK118_25, /* 60Hz CVT */ (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x3A }, - {2112, 1600, 88, 168, 934, 900, 3, 5, VCLK118_25, /* 60Hz CVT */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 2, 0x3A }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1920x1080[] = { {2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5, /* 60Hz */ (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | AST2500PreCatchCRT), 60, 1, 0x38 }, - {2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5, /* 60Hz */ - (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 0xFF, 1, 0x38 }, + AST_VBIOS_INVALID_MODE, /* end */ }; @@ -315,8 +314,7 @@ static const struct ast_vbios_enhtable res_1280x800[] = { AST2500PreCatchCRT), 60, 1, 0x35 }, {1680, 1280, 72,128, 831, 800, 3, 6, VCLK83_5, /* 60Hz */ (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x35 }, - {1680, 1280, 72,128, 831, 800, 3, 6, VCLK83_5, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 2, 0x35 }, + AST_VBIOS_INVALID_MODE, /* end */ }; @@ -326,8 +324,7 @@ static const struct ast_vbios_enhtable res_1440x900[] = { AST2500PreCatchCRT), 60, 1, 0x36 }, {1904, 1440, 80,152, 934, 900, 3, 6, VCLK106_5, /* 60Hz */ (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x36 }, - {1904, 1440, 80,152, 934, 900, 3, 6, VCLK106_5, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 2, 0x36 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1680x1050[] = { @@ -336,17 +333,14 @@ static const struct ast_vbios_enhtable res_1680x1050[] = { AST2500PreCatchCRT), 60, 1, 0x37 }, {2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25, /* 60Hz */ (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x37 }, - {2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 2, 0x37 }, + AST_VBIOS_INVALID_MODE, /* end */ }; static const struct ast_vbios_enhtable res_1920x1200[] = { {2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154, /* 60Hz RB*/ (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | AST2500PreCatchCRT), 60, 1, 0x34 }, - {2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154, /* 60Hz RB */ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 0xFF, 1, 0x34 }, + AST_VBIOS_INVALID_MODE, /* end */ }; #endif From 517e28d40c9fab2e96b7a88787007954efd6cf50 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:10 +0100 Subject: [PATCH 0368/1443] drm/ast: Add helpers for VBIOS mode lookup Mode lines are independent from hardware Gen or TX chip, so hide all VBIOS mode tables in ast_vbios.c. Move the look-up code for VBIOS modes from ast_vbios_get_mode_info() to ast_vbios_find_mode(). The new look-up function respects the supported-mode flags in struct ast_device. For example, if a device does not have struct ast_device.support_fullhd set, the helper does not return a valid mode for 1920x1080. Taking the supported-mode flags into account allows for making the VBIOS tables the single reference for validating and setting display modes against hardware capabilities. v2: - replace mode switch with look-up table (Jocelyn) Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-11-tzimmermann@suse.de --- drivers/gpu/drm/ast/Makefile | 1 + drivers/gpu/drm/ast/ast_dp.c | 1 + drivers/gpu/drm/ast/ast_drv.h | 18 +-- drivers/gpu/drm/ast/ast_mode.c | 77 +--------- drivers/gpu/drm/ast/ast_tables.h | 177 ----------------------- drivers/gpu/drm/ast/ast_vbios.c | 241 +++++++++++++++++++++++++++++++ drivers/gpu/drm/ast/ast_vbios.h | 108 ++++++++++++++ 7 files changed, 360 insertions(+), 263 deletions(-) create mode 100644 drivers/gpu/drm/ast/ast_vbios.c create mode 100644 drivers/gpu/drm/ast/ast_vbios.h diff --git a/drivers/gpu/drm/ast/Makefile b/drivers/gpu/drm/ast/Makefile index 47da848fa3fc..3107ea9c7bf5 100644 --- a/drivers/gpu/drm/ast/Makefile +++ b/drivers/gpu/drm/ast/Makefile @@ -13,6 +13,7 @@ ast-y := \ ast_mode.o \ ast_post.o \ ast_sil164.o \ + ast_vbios.o \ ast_vga.o obj-$(CONFIG_DRM_AST) := ast.o diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 30aad5c0112a..eb65c0881fa7 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -12,6 +12,7 @@ #include #include "ast_drv.h" +#include "ast_vbios.h" static bool ast_astdp_is_connected(struct ast_device *ast) { diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 4e3a88f8a85c..9309cc5e81b0 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -39,6 +39,8 @@ #include "ast_reg.h" +struct ast_vbios_enhtable; + #define DRIVER_AUTHOR "Dave Airlie" #define DRIVER_NAME "ast" @@ -350,22 +352,6 @@ struct ast_vbios_stdtable { u8 gr[9]; }; -struct ast_vbios_enhtable { - u32 ht; - u32 hde; - u32 hfp; - u32 hsync; - u32 vt; - u32 vde; - u32 vfp; - u32 vsync; - u32 dclk_index; - u32 flags; - u32 refresh_rate; - u32 refresh_rate_index; - u32 mode_id; -}; - struct ast_vbios_dclk_info { u8 param1; u8 param2; diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 3f437f871a35..14c0c281a683 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -47,6 +47,7 @@ #include "ast_drv.h" #include "ast_tables.h" +#include "ast_vbios.h" #define AST_LUT_SIZE 256 @@ -106,14 +107,12 @@ static void ast_crtc_set_gamma(struct ast_device *ast, } } -static bool ast_get_vbios_mode_info(const struct drm_format_info *format, +static bool ast_get_vbios_mode_info(struct ast_device *ast, + const struct drm_format_info *format, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode, struct ast_vbios_mode_info *vbios_mode) { - u32 refresh_rate_index = 0, refresh_rate; - const struct ast_vbios_enhtable *best = NULL; - const struct ast_vbios_enhtable *loop; u32 hborder, vborder; switch (format->cpp[0] * 8) { @@ -131,72 +130,9 @@ static bool ast_get_vbios_mode_info(const struct drm_format_info *format, return false; } - switch (mode->hdisplay) { - case 640: - vbios_mode->enh_table = &res_640x480[refresh_rate_index]; - break; - case 800: - vbios_mode->enh_table = &res_800x600[refresh_rate_index]; - break; - case 1024: - vbios_mode->enh_table = &res_1024x768[refresh_rate_index]; - break; - case 1152: - vbios_mode->enh_table = &res_1152x864[refresh_rate_index]; - break; - case 1280: - if (mode->vdisplay == 800) - vbios_mode->enh_table = &res_1280x800[refresh_rate_index]; - else - vbios_mode->enh_table = &res_1280x1024[refresh_rate_index]; - break; - case 1360: - vbios_mode->enh_table = &res_1360x768[refresh_rate_index]; - break; - case 1440: - vbios_mode->enh_table = &res_1440x900[refresh_rate_index]; - break; - case 1600: - if (mode->vdisplay == 900) - vbios_mode->enh_table = &res_1600x900[refresh_rate_index]; - else - vbios_mode->enh_table = &res_1600x1200[refresh_rate_index]; - break; - case 1680: - vbios_mode->enh_table = &res_1680x1050[refresh_rate_index]; - break; - case 1920: - if (mode->vdisplay == 1080) - vbios_mode->enh_table = &res_1920x1080[refresh_rate_index]; - else - vbios_mode->enh_table = &res_1920x1200[refresh_rate_index]; - break; - default: + vbios_mode->enh_table = ast_vbios_find_mode(ast, mode); + if (!vbios_mode->enh_table) return false; - } - - refresh_rate = drm_mode_vrefresh(mode); - - loop = vbios_mode->enh_table; - - while (ast_vbios_mode_is_valid(loop)) { - if (((mode->flags & DRM_MODE_FLAG_NVSYNC) && (loop->flags & PVSync)) || - ((mode->flags & DRM_MODE_FLAG_PVSYNC) && (loop->flags & NVSync)) || - ((mode->flags & DRM_MODE_FLAG_NHSYNC) && (loop->flags & PHSync)) || - ((mode->flags & DRM_MODE_FLAG_PHSYNC) && (loop->flags & NHSync))) { - loop++; - continue; - } - if (loop->refresh_rate <= refresh_rate && - (!best || loop->refresh_rate > best->refresh_rate)) - best = loop; - loop++; - } - - if (!best) - return false; - - vbios_mode->enh_table = best; hborder = (vbios_mode->enh_table->flags & HBorder) ? 8 : 0; vborder = (vbios_mode->enh_table->flags & VBorder) ? 8 : 0; @@ -1109,6 +1045,7 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc); struct ast_crtc_state *old_ast_crtc_state = to_ast_crtc_state(old_crtc_state); struct drm_device *dev = crtc->dev; + struct ast_device *ast = to_ast_device(dev); struct ast_crtc_state *ast_state; const struct drm_format_info *format; bool succ; @@ -1143,7 +1080,7 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, } } - succ = ast_get_vbios_mode_info(format, &crtc_state->mode, + succ = ast_get_vbios_mode_info(ast, format, &crtc_state->mode, &crtc_state->adjusted_mode, &ast_state->vbios_mode_info); if (!succ) diff --git a/drivers/gpu/drm/ast/ast_tables.h b/drivers/gpu/drm/ast/ast_tables.h index 4367817b2f80..f1c9f7e1f1fc 100644 --- a/drivers/gpu/drm/ast/ast_tables.h +++ b/drivers/gpu/drm/ast/ast_tables.h @@ -33,54 +33,6 @@ #define HiCModeIndex 3 #define TrueCModeIndex 4 -#define Charx8Dot 0x00000001 -#define HalfDCLK 0x00000002 -#define DoubleScanMode 0x00000004 -#define LineCompareOff 0x00000008 -#define HBorder 0x00000020 -#define VBorder 0x00000010 -#define WideScreenMode 0x00000100 -#define NewModeInfo 0x00000200 -#define NHSync 0x00000400 -#define PHSync 0x00000800 -#define NVSync 0x00001000 -#define PVSync 0x00002000 -#define SyncPP (PVSync | PHSync) -#define SyncPN (PVSync | NHSync) -#define SyncNP (NVSync | PHSync) -#define SyncNN (NVSync | NHSync) -#define AST2500PreCatchCRT 0x00004000 - -/* DCLK Index */ -#define VCLK25_175 0x00 -#define VCLK28_322 0x01 -#define VCLK31_5 0x02 -#define VCLK36 0x03 -#define VCLK40 0x04 -#define VCLK49_5 0x05 -#define VCLK50 0x06 -#define VCLK56_25 0x07 -#define VCLK65 0x08 -#define VCLK75 0x09 -#define VCLK78_75 0x0A -#define VCLK94_5 0x0B -#define VCLK108 0x0C -#define VCLK135 0x0D -#define VCLK157_5 0x0E -#define VCLK162 0x0F -/* #define VCLK193_25 0x10 */ -#define VCLK154 0x10 -#define VCLK83_5 0x11 -#define VCLK106_5 0x12 -#define VCLK146_25 0x13 -#define VCLK148_5 0x14 -#define VCLK71 0x15 -#define VCLK88_75 0x16 -#define VCLK119 0x17 -#define VCLK85_5 0x18 -#define VCLK97_75 0x19 -#define VCLK118_25 0x1A - static const struct ast_vbios_dclk_info dclk_table[] = { {0x2C, 0xE7, 0x03}, /* 00: VCLK25_175 */ {0x95, 0x62, 0x03}, /* 01: VCLK28_322 */ @@ -214,133 +166,4 @@ static const struct ast_vbios_stdtable vbios_stdtable[] = { }, }; -#define AST_VBIOS_INVALID_MODE \ - {0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u} - -static inline bool ast_vbios_mode_is_valid(const struct ast_vbios_enhtable *vmode) -{ - return vmode->ht && vmode->vt && vmode->refresh_rate; -} - -static const struct ast_vbios_enhtable res_640x480[] = { - { 800, 640, 8, 96, 525, 480, 2, 2, VCLK25_175, /* 60Hz */ - (SyncNN | HBorder | VBorder | Charx8Dot), 60, 1, 0x2E }, - { 832, 640, 16, 40, 520, 480, 1, 3, VCLK31_5, /* 72Hz */ - (SyncNN | HBorder | VBorder | Charx8Dot), 72, 2, 0x2E }, - { 840, 640, 16, 64, 500, 480, 1, 3, VCLK31_5, /* 75Hz */ - (SyncNN | Charx8Dot) , 75, 3, 0x2E }, - { 832, 640, 56, 56, 509, 480, 1, 3, VCLK36, /* 85Hz */ - (SyncNN | Charx8Dot) , 85, 4, 0x2E }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_800x600[] = { - {1024, 800, 24, 72, 625, 600, 1, 2, VCLK36, /* 56Hz */ - (SyncPP | Charx8Dot), 56, 1, 0x30 }, - {1056, 800, 40, 128, 628, 600, 1, 4, VCLK40, /* 60Hz */ - (SyncPP | Charx8Dot), 60, 2, 0x30 }, - {1040, 800, 56, 120, 666, 600, 37, 6, VCLK50, /* 72Hz */ - (SyncPP | Charx8Dot), 72, 3, 0x30 }, - {1056, 800, 16, 80, 625, 600, 1, 3, VCLK49_5, /* 75Hz */ - (SyncPP | Charx8Dot), 75, 4, 0x30 }, - {1048, 800, 32, 64, 631, 600, 1, 3, VCLK56_25, /* 85Hz */ - (SyncPP | Charx8Dot), 84, 5, 0x30 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - - -static const struct ast_vbios_enhtable res_1024x768[] = { - {1344, 1024, 24, 136, 806, 768, 3, 6, VCLK65, /* 60Hz */ - (SyncNN | Charx8Dot), 60, 1, 0x31 }, - {1328, 1024, 24, 136, 806, 768, 3, 6, VCLK75, /* 70Hz */ - (SyncNN | Charx8Dot), 70, 2, 0x31 }, - {1312, 1024, 16, 96, 800, 768, 1, 3, VCLK78_75, /* 75Hz */ - (SyncPP | Charx8Dot), 75, 3, 0x31 }, - {1376, 1024, 48, 96, 808, 768, 1, 3, VCLK94_5, /* 85Hz */ - (SyncPP | Charx8Dot), 84, 4, 0x31 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1280x1024[] = { - {1688, 1280, 48, 112, 1066, 1024, 1, 3, VCLK108, /* 60Hz */ - (SyncPP | Charx8Dot), 60, 1, 0x32 }, - {1688, 1280, 16, 144, 1066, 1024, 1, 3, VCLK135, /* 75Hz */ - (SyncPP | Charx8Dot), 75, 2, 0x32 }, - {1728, 1280, 64, 160, 1072, 1024, 1, 3, VCLK157_5, /* 85Hz */ - (SyncPP | Charx8Dot), 85, 3, 0x32 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1600x1200[] = { - {2160, 1600, 64, 192, 1250, 1200, 1, 3, VCLK162, /* 60Hz */ - (SyncPP | Charx8Dot), 60, 1, 0x33 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1152x864[] = { - {1600, 1152, 64, 128, 900, 864, 1, 3, VCLK108, /* 75Hz */ - (SyncPP | Charx8Dot | NewModeInfo), 75, 1, 0x3B }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -/* 16:9 */ -static const struct ast_vbios_enhtable res_1360x768[] = { - {1792, 1360, 64, 112, 795, 768, 3, 6, VCLK85_5, /* 60Hz */ - (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x39 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1600x900[] = { - {1760, 1600, 48, 32, 926, 900, 3, 5, VCLK97_75, /* 60Hz CVT RB */ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x3A }, - {2112, 1600, 88, 168, 934, 900, 3, 5, VCLK118_25, /* 60Hz CVT */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x3A }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1920x1080[] = { - {2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5, /* 60Hz */ - (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x38 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - - -/* 16:10 */ -static const struct ast_vbios_enhtable res_1280x800[] = { - {1440, 1280, 48, 32, 823, 800, 3, 6, VCLK71, /* 60Hz RB */ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x35 }, - {1680, 1280, 72,128, 831, 800, 3, 6, VCLK83_5, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x35 }, - AST_VBIOS_INVALID_MODE, /* end */ - -}; - -static const struct ast_vbios_enhtable res_1440x900[] = { - {1600, 1440, 48, 32, 926, 900, 3, 6, VCLK88_75, /* 60Hz RB */ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x36 }, - {1904, 1440, 80,152, 934, 900, 3, 6, VCLK106_5, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x36 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1680x1050[] = { - {1840, 1680, 48, 32, 1080, 1050, 3, 6, VCLK119, /* 60Hz RB */ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x37 }, - {2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25, /* 60Hz */ - (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x37 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - -static const struct ast_vbios_enhtable res_1920x1200[] = { - {2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154, /* 60Hz RB*/ - (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | - AST2500PreCatchCRT), 60, 1, 0x34 }, - AST_VBIOS_INVALID_MODE, /* end */ -}; - #endif diff --git a/drivers/gpu/drm/ast/ast_vbios.c b/drivers/gpu/drm/ast/ast_vbios.c new file mode 100644 index 000000000000..0953e6dd3976 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_vbios.c @@ -0,0 +1,241 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright (c) 2005 ASPEED Technology Inc. + * + * Permission to use, copy, modify, distribute, and sell this software and its + * documentation for any purpose is hereby granted without fee, provided that + * the above copyright notice appear in all copies and that both that + * copyright notice and this permission notice appear in supporting + * documentation, and that the name of the authors not be used in + * advertising or publicity pertaining to distribution of the software without + * specific, written prior permission. The authors makes no representations + * about the suitability of this software for any purpose. It is provided + * "as is" without express or implied warranty. + * + * THE AUTHORS DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, + * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO + * EVENT SHALL THE AUTHORS BE LIABLE FOR ANY SPECIAL, INDIRECT OR + * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, + * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER + * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR + * PERFORMANCE OF THIS SOFTWARE. + */ + +#include "ast_drv.h" +#include "ast_vbios.h" + +/* 4:3 */ + +static const struct ast_vbios_enhtable res_640x480[] = { + { 800, 640, 8, 96, 525, 480, 2, 2, VCLK25_175, /* 60 Hz */ + (SyncNN | HBorder | VBorder | Charx8Dot), 60, 1, 0x2e }, + { 832, 640, 16, 40, 520, 480, 1, 3, VCLK31_5, /* 72 Hz */ + (SyncNN | HBorder | VBorder | Charx8Dot), 72, 2, 0x2e }, + { 840, 640, 16, 64, 500, 480, 1, 3, VCLK31_5, /* 75 Hz */ + (SyncNN | Charx8Dot), 75, 3, 0x2e }, + { 832, 640, 56, 56, 509, 480, 1, 3, VCLK36, /* 85 Hz */ + (SyncNN | Charx8Dot), 85, 4, 0x2e }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_800x600[] = { + { 1024, 800, 24, 72, 625, 600, 1, 2, VCLK36, /* 56 Hz */ + (SyncPP | Charx8Dot), 56, 1, 0x30 }, + { 1056, 800, 40, 128, 628, 600, 1, 4, VCLK40, /* 60 Hz */ + (SyncPP | Charx8Dot), 60, 2, 0x30 }, + { 1040, 800, 56, 120, 666, 600, 37, 6, VCLK50, /* 72 Hz */ + (SyncPP | Charx8Dot), 72, 3, 0x30 }, + { 1056, 800, 16, 80, 625, 600, 1, 3, VCLK49_5, /* 75 Hz */ + (SyncPP | Charx8Dot), 75, 4, 0x30 }, + { 1048, 800, 32, 64, 631, 600, 1, 3, VCLK56_25, /* 85 Hz */ + (SyncPP | Charx8Dot), 84, 5, 0x30 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1024x768[] = { + { 1344, 1024, 24, 136, 806, 768, 3, 6, VCLK65, /* 60 Hz */ + (SyncNN | Charx8Dot), 60, 1, 0x31 }, + { 1328, 1024, 24, 136, 806, 768, 3, 6, VCLK75, /* 70 Hz */ + (SyncNN | Charx8Dot), 70, 2, 0x31 }, + { 1312, 1024, 16, 96, 800, 768, 1, 3, VCLK78_75, /* 75 Hz */ + (SyncPP | Charx8Dot), 75, 3, 0x31 }, + { 1376, 1024, 48, 96, 808, 768, 1, 3, VCLK94_5, /* 85 Hz */ + (SyncPP | Charx8Dot), 84, 4, 0x31 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1152x864[] = { + { 1600, 1152, 64, 128, 900, 864, 1, 3, VCLK108, /* 75 Hz */ + (SyncPP | Charx8Dot | NewModeInfo), 75, 1, 0x3b }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1280x1024[] = { + { 1688, 1280, 48, 112, 1066, 1024, 1, 3, VCLK108, /* 60 Hz */ + (SyncPP | Charx8Dot), 60, 1, 0x32 }, + { 1688, 1280, 16, 144, 1066, 1024, 1, 3, VCLK135, /* 75 Hz */ + (SyncPP | Charx8Dot), 75, 2, 0x32 }, + { 1728, 1280, 64, 160, 1072, 1024, 1, 3, VCLK157_5, /* 85 Hz */ + (SyncPP | Charx8Dot), 85, 3, 0x32 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1600x1200[] = { + { 2160, 1600, 64, 192, 1250, 1200, 1, 3, VCLK162, /* 60 Hz */ + (SyncPP | Charx8Dot), 60, 1, 0x33 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +/* 16:9 */ + +static const struct ast_vbios_enhtable res_1360x768[] = { + { 1792, 1360, 64, 112, 795, 768, 3, 6, VCLK85_5, /* 60 Hz */ + (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x39 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1600x900[] = { + { 1760, 1600, 48, 32, 926, 900, 3, 5, VCLK97_75, /* 60 Hz CVT RB */ + (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x3a }, + { 2112, 1600, 88, 168, 934, 900, 3, 5, VCLK118_25, /* 60 Hz CVT */ + (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x3a }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1920x1080[] = { + { 2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5, /* 60 Hz */ + (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x38 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +/* 16:10 */ + +static const struct ast_vbios_enhtable res_1280x800[] = { + { 1440, 1280, 48, 32, 823, 800, 3, 6, VCLK71, /* 60 Hz RB */ + (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x35 }, + { 1680, 1280, 72, 128, 831, 800, 3, 6, VCLK83_5, /* 60 Hz */ + (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x35 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1440x900[] = { + { 1600, 1440, 48, 32, 926, 900, 3, 6, VCLK88_75, /* 60 Hz RB */ + (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x36 }, + { 1904, 1440, 80, 152, 934, 900, 3, 6, VCLK106_5, /* 60 Hz */ + (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x36 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1680x1050[] = { + { 1840, 1680, 48, 32, 1080, 1050, 3, 6, VCLK119, /* 60 Hz RB */ + (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x37 }, + { 2240, 1680, 104, 176, 1089, 1050, 3, 6, VCLK146_25, /* 60 Hz */ + (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 2, 0x37 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +static const struct ast_vbios_enhtable res_1920x1200[] = { + { 2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154, /* 60 Hz RB*/ + (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo | + AST2500PreCatchCRT), 60, 1, 0x34 }, + AST_VBIOS_INVALID_MODE, /* end */ +}; + +/* + * VBIOS mode tables + */ + +static const struct ast_vbios_enhtable *res_table_wuxga[] = { + &res_1920x1200[0], + NULL, +}; + +static const struct ast_vbios_enhtable *res_table_fullhd[] = { + &res_1920x1080[0], + NULL, +}; + +static const struct ast_vbios_enhtable *res_table_wsxga_p[] = { + &res_1280x800[0], + &res_1360x768[0], + &res_1440x900[0], + &res_1600x900[0], + &res_1680x1050[0], + NULL, +}; + +static const struct ast_vbios_enhtable *res_table[] = { + &res_640x480[0], + &res_800x600[0], + &res_1024x768[0], + &res_1152x864[0], + &res_1280x1024[0], + &res_1600x1200[0], + NULL, +}; + +static const struct ast_vbios_enhtable * +__ast_vbios_find_mode_table(const struct ast_vbios_enhtable **vmode_tables, + unsigned int hdisplay, + unsigned int vdisplay) +{ + while (*vmode_tables) { + if ((*vmode_tables)->hde == hdisplay && (*vmode_tables)->vde == vdisplay) + return *vmode_tables; + ++vmode_tables; + } + + return NULL; +} + +static const struct ast_vbios_enhtable *ast_vbios_find_mode_table(const struct ast_device *ast, + unsigned int hdisplay, + unsigned int vdisplay) +{ + const struct ast_vbios_enhtable *vmode_table = NULL; + + if (ast->support_wuxga) + vmode_table = __ast_vbios_find_mode_table(res_table_wuxga, hdisplay, vdisplay); + if (!vmode_table && ast->support_fullhd) + vmode_table = __ast_vbios_find_mode_table(res_table_fullhd, hdisplay, vdisplay); + if (!vmode_table && ast->support_wsxga_p) + vmode_table = __ast_vbios_find_mode_table(res_table_wsxga_p, hdisplay, vdisplay); + if (!vmode_table) + vmode_table = __ast_vbios_find_mode_table(res_table, hdisplay, vdisplay); + + return vmode_table; +} + +const struct ast_vbios_enhtable *ast_vbios_find_mode(const struct ast_device *ast, + const struct drm_display_mode *mode) +{ + const struct ast_vbios_enhtable *best_vmode = NULL; + const struct ast_vbios_enhtable *vmode_table; + const struct ast_vbios_enhtable *vmode; + u32 refresh_rate; + + vmode_table = ast_vbios_find_mode_table(ast, mode->hdisplay, mode->vdisplay); + if (!vmode_table) + return NULL; + + refresh_rate = drm_mode_vrefresh(mode); + + for (vmode = vmode_table; ast_vbios_mode_is_valid(vmode); ++vmode) { + if (((mode->flags & DRM_MODE_FLAG_NVSYNC) && (vmode->flags & PVSync)) || + ((mode->flags & DRM_MODE_FLAG_PVSYNC) && (vmode->flags & NVSync)) || + ((mode->flags & DRM_MODE_FLAG_NHSYNC) && (vmode->flags & PHSync)) || + ((mode->flags & DRM_MODE_FLAG_PHSYNC) && (vmode->flags & NHSync))) { + continue; + } + if (vmode->refresh_rate <= refresh_rate && + (!best_vmode || vmode->refresh_rate > best_vmode->refresh_rate)) + best_vmode = vmode; + } + + return best_vmode; +} diff --git a/drivers/gpu/drm/ast/ast_vbios.h b/drivers/gpu/drm/ast/ast_vbios.h new file mode 100644 index 000000000000..8cf025010594 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_vbios.h @@ -0,0 +1,108 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright (c) 2005 ASPEED Technology Inc. + * + * Permission to use, copy, modify, distribute, and sell this software and its + * documentation for any purpose is hereby granted without fee, provided that + * the above copyright notice appear in all copies and that both that + * copyright notice and this permission notice appear in supporting + * documentation, and that the name of the authors not be used in + * advertising or publicity pertaining to distribution of the software without + * specific, written prior permission. The authors makes no representations + * about the suitability of this software for any purpose. It is provided + * "as is" without express or implied warranty. + * + * THE AUTHORS DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, + * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO + * EVENT SHALL THE AUTHORS BE LIABLE FOR ANY SPECIAL, INDIRECT OR + * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, + * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER + * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR + * PERFORMANCE OF THIS SOFTWARE. + */ +/* Ported from xf86-video-ast driver */ + +#ifndef AST_VBIOS_H +#define AST_VBIOS_H + +#include + +struct ast_device; +struct drm_display_mode; + +#define Charx8Dot 0x00000001 +#define HalfDCLK 0x00000002 +#define DoubleScanMode 0x00000004 +#define LineCompareOff 0x00000008 +#define HBorder 0x00000020 +#define VBorder 0x00000010 +#define WideScreenMode 0x00000100 +#define NewModeInfo 0x00000200 +#define NHSync 0x00000400 +#define PHSync 0x00000800 +#define NVSync 0x00001000 +#define PVSync 0x00002000 +#define SyncPP (PVSync | PHSync) +#define SyncPN (PVSync | NHSync) +#define SyncNP (NVSync | PHSync) +#define SyncNN (NVSync | NHSync) +#define AST2500PreCatchCRT 0x00004000 + +/* DCLK Index */ +#define VCLK25_175 0x00 +#define VCLK28_322 0x01 +#define VCLK31_5 0x02 +#define VCLK36 0x03 +#define VCLK40 0x04 +#define VCLK49_5 0x05 +#define VCLK50 0x06 +#define VCLK56_25 0x07 +#define VCLK65 0x08 +#define VCLK75 0x09 +#define VCLK78_75 0x0a +#define VCLK94_5 0x0b +#define VCLK108 0x0c +#define VCLK135 0x0d +#define VCLK157_5 0x0e +#define VCLK162 0x0f +/* #define VCLK193_25 0x10 */ +#define VCLK154 0x10 +#define VCLK83_5 0x11 +#define VCLK106_5 0x12 +#define VCLK146_25 0x13 +#define VCLK148_5 0x14 +#define VCLK71 0x15 +#define VCLK88_75 0x16 +#define VCLK119 0x17 +#define VCLK85_5 0x18 +#define VCLK97_75 0x19 +#define VCLK118_25 0x1a + +struct ast_vbios_enhtable { + u32 ht; + u32 hde; + u32 hfp; + u32 hsync; + u32 vt; + u32 vde; + u32 vfp; + u32 vsync; + u32 dclk_index; + u32 flags; + u32 refresh_rate; + u32 refresh_rate_index; + u32 mode_id; +}; + +#define AST_VBIOS_INVALID_MODE \ + {0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u, 0u} + +static inline bool ast_vbios_mode_is_valid(const struct ast_vbios_enhtable *vmode) +{ + return vmode->ht && vmode->vt && vmode->refresh_rate; +} + +const struct ast_vbios_enhtable *ast_vbios_find_mode(const struct ast_device *ast, + const struct drm_display_mode *mode); + +#endif From 2e90349bf44b30478351be4805a640b3bf5accb9 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:11 +0100 Subject: [PATCH 0369/1443] drm/ast: Validate DRM display modes against VBIOS modes Test DRM display modes against the list of modes supported by the VBIOS. The helper will respect the supported-modes flags in struct ast_device. Hence only DRM display modes supported by the VBIOS will be reported; without the current duplication of this information. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-12-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 62 +++------------------------------- 1 file changed, 5 insertions(+), 57 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 14c0c281a683..f58ec35cb3a3 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -950,65 +950,13 @@ static enum drm_mode_status ast_crtc_helper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *mode) { struct ast_device *ast = to_ast_device(crtc->dev); - enum drm_mode_status status; + const struct ast_vbios_enhtable *vmode; - if (ast->support_wsxga_p) { - if ((mode->hdisplay == 1680) && (mode->vdisplay == 1050)) - return MODE_OK; - if ((mode->hdisplay == 1280) && (mode->vdisplay == 800)) - return MODE_OK; - if ((mode->hdisplay == 1440) && (mode->vdisplay == 900)) - return MODE_OK; - if ((mode->hdisplay == 1360) && (mode->vdisplay == 768)) - return MODE_OK; - if ((mode->hdisplay == 1600) && (mode->vdisplay == 900)) - return MODE_OK; + vmode = ast_vbios_find_mode(ast, mode); + if (!vmode) + return MODE_NOMODE; - if (ast->support_fullhd) { - if ((mode->hdisplay == 1920) && (mode->vdisplay == 1080)) - return MODE_OK; - - if ((mode->hdisplay == 1920) && (mode->vdisplay == 1200)) { - if (ast->support_wuxga) - return MODE_OK; - else - return MODE_NOMODE; - } - } - } - - status = MODE_NOMODE; - - switch (mode->hdisplay) { - case 640: - if (mode->vdisplay == 480) - status = MODE_OK; - break; - case 800: - if (mode->vdisplay == 600) - status = MODE_OK; - break; - case 1024: - if (mode->vdisplay == 768) - status = MODE_OK; - break; - case 1152: - if (mode->vdisplay == 864) - status = MODE_OK; - break; - case 1280: - if (mode->vdisplay == 1024) - status = MODE_OK; - break; - case 1600: - if (mode->vdisplay == 1200) - status = MODE_OK; - break; - default: - break; - } - - return status; + return MODE_OK; } static void ast_crtc_helper_mode_set_nofb(struct drm_crtc *crtc) From c337d4c1de88bda6da88477a390c2f7ba87466b6 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:12 +0100 Subject: [PATCH 0370/1443] drm/ast: Inline ast_get_vbios_mode_info() The helper ast_get_vbios_mode_info() retrieves hardware-specific format and display-mode information. Inline the function into its only caller. While at it, also replace a use of struct drm_format_info.cpp with the correct DRM 4CC codes. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-13-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 112 ++++++++++++++++----------------- 1 file changed, 54 insertions(+), 58 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index f58ec35cb3a3..1676958e9768 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -107,59 +107,6 @@ static void ast_crtc_set_gamma(struct ast_device *ast, } } -static bool ast_get_vbios_mode_info(struct ast_device *ast, - const struct drm_format_info *format, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode, - struct ast_vbios_mode_info *vbios_mode) -{ - u32 hborder, vborder; - - switch (format->cpp[0] * 8) { - case 8: - vbios_mode->std_table = &vbios_stdtable[VGAModeIndex]; - break; - case 16: - vbios_mode->std_table = &vbios_stdtable[HiCModeIndex]; - break; - case 24: - case 32: - vbios_mode->std_table = &vbios_stdtable[TrueCModeIndex]; - break; - default: - return false; - } - - vbios_mode->enh_table = ast_vbios_find_mode(ast, mode); - if (!vbios_mode->enh_table) - return false; - - hborder = (vbios_mode->enh_table->flags & HBorder) ? 8 : 0; - vborder = (vbios_mode->enh_table->flags & VBorder) ? 8 : 0; - - adjusted_mode->crtc_hdisplay = vbios_mode->enh_table->hde; - adjusted_mode->crtc_htotal = vbios_mode->enh_table->ht; - adjusted_mode->crtc_hblank_start = vbios_mode->enh_table->hde + hborder; - adjusted_mode->crtc_hblank_end = vbios_mode->enh_table->ht - hborder; - adjusted_mode->crtc_hsync_start = vbios_mode->enh_table->hde + hborder + - vbios_mode->enh_table->hfp; - adjusted_mode->crtc_hsync_end = (vbios_mode->enh_table->hde + hborder + - vbios_mode->enh_table->hfp + - vbios_mode->enh_table->hsync); - - adjusted_mode->crtc_vdisplay = vbios_mode->enh_table->vde; - adjusted_mode->crtc_vtotal = vbios_mode->enh_table->vt; - adjusted_mode->crtc_vblank_start = vbios_mode->enh_table->vde + vborder; - adjusted_mode->crtc_vblank_end = vbios_mode->enh_table->vt - vborder; - adjusted_mode->crtc_vsync_start = vbios_mode->enh_table->vde + vborder + - vbios_mode->enh_table->vfp; - adjusted_mode->crtc_vsync_end = (vbios_mode->enh_table->vde + vborder + - vbios_mode->enh_table->vfp + - vbios_mode->enh_table->vsync); - - return true; -} - static void ast_set_vbios_color_reg(struct ast_device *ast, const struct drm_format_info *format, const struct ast_vbios_mode_info *vbios_mode) @@ -990,13 +937,17 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, struct drm_atomic_state *state) { struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode; struct drm_crtc_state *old_crtc_state = drm_atomic_get_old_crtc_state(state, crtc); struct ast_crtc_state *old_ast_crtc_state = to_ast_crtc_state(old_crtc_state); struct drm_device *dev = crtc->dev; struct ast_device *ast = to_ast_device(dev); struct ast_crtc_state *ast_state; const struct drm_format_info *format; - bool succ; + struct ast_vbios_mode_info *vbios_mode; + const struct ast_vbios_enhtable *vmode; + unsigned int hborder = 0; + unsigned int vborder = 0; int ret; if (!crtc_state->enable) @@ -1028,11 +979,56 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, } } - succ = ast_get_vbios_mode_info(ast, format, &crtc_state->mode, - &crtc_state->adjusted_mode, - &ast_state->vbios_mode_info); - if (!succ) + vbios_mode = &ast_state->vbios_mode_info; + + /* + * Set register tables. + * + * TODO: These tables mix all kinds of fields and should + * probably be resolved into various helper functions. + */ + switch (format->format) { + case DRM_FORMAT_C8: + vbios_mode->std_table = &vbios_stdtable[VGAModeIndex]; + break; + case DRM_FORMAT_RGB565: + vbios_mode->std_table = &vbios_stdtable[HiCModeIndex]; + break; + case DRM_FORMAT_RGB888: + case DRM_FORMAT_XRGB8888: + vbios_mode->std_table = &vbios_stdtable[TrueCModeIndex]; + break; + default: return -EINVAL; + } + + /* + * Find the VBIOS mode and adjust the DRM display mode accordingly. + */ + + vmode = ast_vbios_find_mode(ast, &crtc_state->mode); + if (!vmode) + return -EINVAL; + ast_state->vbios_mode_info.enh_table = vmode; + + if (vmode->flags & HBorder) + hborder = 8; + if (vmode->flags & VBorder) + vborder = 8; + + adjusted_mode->crtc_hdisplay = vmode->hde; + adjusted_mode->crtc_hblank_start = vmode->hde + hborder; + adjusted_mode->crtc_hblank_end = vmode->ht - hborder; + adjusted_mode->crtc_hsync_start = vmode->hde + hborder + vmode->hfp; + adjusted_mode->crtc_hsync_end = vmode->hde + hborder + vmode->hfp + vmode->hsync; + adjusted_mode->crtc_htotal = vmode->ht; + + adjusted_mode->crtc_vdisplay = vmode->vde; + adjusted_mode->crtc_vblank_start = vmode->vde + vborder; + adjusted_mode->crtc_vblank_end = vmode->vt - vborder; + adjusted_mode->crtc_vsync_start = vmode->vde + vborder + vmode->vfp; + adjusted_mode->crtc_vsync_end = vmode->vde + vborder + vmode->vfp + vmode->vsync; + adjusted_mode->crtc_vtotal = vmode->vt; return 0; } From d86f6beed067309eeea8bf7c91d967f310db98df Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:13 +0100 Subject: [PATCH 0371/1443] drm/ast: astdp: Rework display-mode setting ASTDP requires a mode index, depending on the resolution. Move the look-up code from ast_dp_set_mode() into a separate helper. Inline the rest of the function into its only caller. Rename the variable names and register constants to match the programming manual. As before, the mode-index lookup still happens during the update's atomic commit. Right now, there's no way of doing it during the atomic check. The lookup requires the VBIOS mode, which is not available at the atomic check's invocation. At least warn now if the mode index could not be found. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-14-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_dp.c | 172 +++++++++++++++++++--------------- drivers/gpu/drm/ast/ast_reg.h | 18 +--- 2 files changed, 100 insertions(+), 90 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index eb65c0881fa7..3f5e586325a9 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -14,6 +14,82 @@ #include "ast_drv.h" #include "ast_vbios.h" +static int ast_astdp_get_mode_index(const struct ast_vbios_enhtable *vmode) +{ + u8 refresh_rate_index; + + if (vmode->refresh_rate_index < 1 || vmode->refresh_rate_index > 255) + return -EINVAL; + + refresh_rate_index = vmode->refresh_rate_index - 1; + + switch (vmode->hde) { + case 320: + if (vmode->vde == 240) + return ASTDP_320x240_60; + break; + case 400: + if (vmode->vde == 300) + return ASTDP_400x300_60; + break; + case 512: + if (vmode->vde == 384) + return ASTDP_512x384_60; + break; + case 640: + if (vmode->vde == 480) + return (u8)(ASTDP_640x480_60 + (u8)refresh_rate_index); + break; + case 800: + if (vmode->vde == 600) + return (u8)(ASTDP_800x600_56 + (u8)refresh_rate_index); + break; + case 1024: + if (vmode->vde == 768) + return (u8)(ASTDP_1024x768_60 + (u8)refresh_rate_index); + break; + case 1152: + if (vmode->vde == 864) + return ASTDP_1152x864_75; + break; + case 1280: + if (vmode->vde == 800) + return (u8)(ASTDP_1280x800_60_RB - (u8)refresh_rate_index); + if (vmode->vde == 1024) + return (u8)(ASTDP_1280x1024_60 + (u8)refresh_rate_index); + break; + case 1360: + case 1366: + if (vmode->vde == 768) + return ASTDP_1366x768_60; + break; + case 1440: + if (vmode->vde == 900) + return (u8)(ASTDP_1440x900_60_RB - (u8)refresh_rate_index); + break; + case 1600: + if (vmode->vde == 900) + return (u8)(ASTDP_1600x900_60_RB - (u8)refresh_rate_index); + if (vmode->vde == 1200) + return ASTDP_1600x1200_60; + break; + case 1680: + if (vmode->vde == 1050) + return (u8)(ASTDP_1680x1050_60_RB - (u8)refresh_rate_index); + break; + case 1920: + if (vmode->vde == 1080) + return ASTDP_1920x1080_60; + if (vmode->vde == 1200) + return ASTDP_1920x1200_60; + break; + default: + break; + } + + return -EINVAL; +} + static bool ast_astdp_is_connected(struct ast_device *ast) { if (!ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xDF, AST_IO_VGACRDF_HPD)) @@ -228,80 +304,6 @@ static void ast_dp_set_enable(struct ast_device *ast, bool enabled) drm_WARN_ON(dev, !__ast_dp_wait_enable(ast, enabled)); } -static void ast_dp_set_mode(struct drm_crtc *crtc, struct ast_vbios_mode_info *vbios_mode) -{ - struct ast_device *ast = to_ast_device(crtc->dev); - - u32 ulRefreshRateIndex; - u8 ModeIdx; - - ulRefreshRateIndex = vbios_mode->enh_table->refresh_rate_index - 1; - - switch (crtc->mode.crtc_hdisplay) { - case 320: - ModeIdx = ASTDP_320x240_60; - break; - case 400: - ModeIdx = ASTDP_400x300_60; - break; - case 512: - ModeIdx = ASTDP_512x384_60; - break; - case 640: - ModeIdx = (ASTDP_640x480_60 + (u8) ulRefreshRateIndex); - break; - case 800: - ModeIdx = (ASTDP_800x600_56 + (u8) ulRefreshRateIndex); - break; - case 1024: - ModeIdx = (ASTDP_1024x768_60 + (u8) ulRefreshRateIndex); - break; - case 1152: - ModeIdx = ASTDP_1152x864_75; - break; - case 1280: - if (crtc->mode.crtc_vdisplay == 800) - ModeIdx = (ASTDP_1280x800_60_RB - (u8) ulRefreshRateIndex); - else // 1024 - ModeIdx = (ASTDP_1280x1024_60 + (u8) ulRefreshRateIndex); - break; - case 1360: - case 1366: - ModeIdx = ASTDP_1366x768_60; - break; - case 1440: - ModeIdx = (ASTDP_1440x900_60_RB - (u8) ulRefreshRateIndex); - break; - case 1600: - if (crtc->mode.crtc_vdisplay == 900) - ModeIdx = (ASTDP_1600x900_60_RB - (u8) ulRefreshRateIndex); - else //1200 - ModeIdx = ASTDP_1600x1200_60; - break; - case 1680: - ModeIdx = (ASTDP_1680x1050_60_RB - (u8) ulRefreshRateIndex); - break; - case 1920: - if (crtc->mode.crtc_vdisplay == 1080) - ModeIdx = ASTDP_1920x1080_60; - else //1200 - ModeIdx = ASTDP_1920x1200_60; - break; - default: - return; - } - - /* - * CRE0[7:0]: MISC0 ((0x00: 18-bpp) or (0x20: 24-bpp) - * CRE1[7:0]: MISC1 (default: 0x00) - * CRE2[7:0]: video format index (0x00 ~ 0x20 or 0x40 ~ 0x50) - */ - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xE0, ASTDP_AND_CLEAR_MASK, - ASTDP_MISC0_24bpp); - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xE1, ASTDP_AND_CLEAR_MASK, ASTDP_MISC1); - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xE2, ASTDP_AND_CLEAR_MASK, ModeIdx); -} - static void ast_wait_for_vretrace(struct ast_device *ast) { unsigned long timeout = jiffies + HZ; @@ -324,11 +326,29 @@ static void ast_astdp_encoder_helper_atomic_mode_set(struct drm_encoder *encoder struct drm_crtc_state *crtc_state, struct drm_connector_state *conn_state) { - struct drm_crtc *crtc = crtc_state->crtc; + struct drm_device *dev = encoder->dev; + struct ast_device *ast = to_ast_device(dev); struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state); struct ast_vbios_mode_info *vbios_mode_info = &ast_crtc_state->vbios_mode_info; + int mode_index; + u8 vgacre0, vgacre1, vgacre2; - ast_dp_set_mode(crtc, vbios_mode_info); + mode_index = ast_astdp_get_mode_index(vbios_mode_info->enh_table); + if (drm_WARN_ON(dev, mode_index < 0)) + return; + + /* + * CRE0[7:0]: MISC0 ((0x00: 18-bpp) or (0x20: 24-bpp) + * CRE1[7:0]: MISC1 (default: 0x00) + * CRE2[7:0]: video format index (0x00 ~ 0x20 or 0x40 ~ 0x50) + */ + vgacre0 = AST_IO_VGACRE0_24BPP; + vgacre1 = 0x00; + vgacre2 = mode_index & 0xff; + + ast_set_index_reg(ast, AST_IO_VGACRI, 0xe0, vgacre0); + ast_set_index_reg(ast, AST_IO_VGACRI, 0xe1, vgacre1); + ast_set_index_reg(ast, AST_IO_VGACRI, 0xe2, vgacre2); } static void ast_astdp_encoder_helper_atomic_enable(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h index 9db0d584652a..bb2cc1d8b84e 100644 --- a/drivers/gpu/drm/ast/ast_reg.h +++ b/drivers/gpu/drm/ast/ast_reg.h @@ -57,10 +57,14 @@ #define AST_IO_VGACRD1_TX_ASTDP 0x0e #define AST_IO_VGACRD1_SUPPORTS_WUXGA BIT(0) +/* + * AST DisplayPort + */ #define AST_IO_VGACRD7_EDID_VALID_FLAG BIT(0) #define AST_IO_VGACRDC_LINK_SUCCESS BIT(0) #define AST_IO_VGACRDF_HPD BIT(0) #define AST_IO_VGACRDF_DP_VIDEO_ENABLE BIT(4) /* mirrors AST_IO_VGACRE3_DP_VIDEO_ENABLE */ +#define AST_IO_VGACRE0_24BPP BIT(5) /* 18 bpp, if unset */ #define AST_IO_VGACRE3_DP_VIDEO_ENABLE BIT(0) #define AST_IO_VGACRE3_DP_PHY_SLEEP BIT(4) #define AST_IO_VGACRE5_EDID_READ_DONE BIT(0) @@ -68,18 +72,4 @@ #define AST_IO_VGAIR1_R (0x5A) #define AST_IO_VGAIR1_VREFRESH BIT(3) -/* - * AST DisplayPort - */ - -/* - * ASTDP setmode registers: - * CRE0[7:0]: MISC0 ((0x00: 18-bpp) or (0x20: 24-bpp) - * CRE1[7:0]: MISC1 (default: 0x00) - * CRE2[7:0]: video format index (0x00 ~ 0x20 or 0x40 ~ 0x50) - */ -#define ASTDP_MISC0_24bpp BIT(5) -#define ASTDP_MISC1 0 -#define ASTDP_AND_CLEAR_MASK 0x00 - #endif From 6af7e16c4d818c0d019c0cdb4eb12da6dc720b57 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:14 +0100 Subject: [PATCH 0372/1443] drm/ast: astdp: Look up mode index from table Replace the large switch statement with a look-up table when selecting the mode index. Makes the code easier to read. The table is sorted by resolutions; if run-time overhead from traversal becomes significant, binary search would be a possible optimization. The mode index requires a refresh-rate index to be added or subtracted, which still requires a minimal switch. In the original code, some of the indices did not contain this computation. Those cases would have been equivalent to adding 0, so they are now all subsumed in the switch's default branch. Signed-off-by: Thomas Zimmermann Suggested-by: Jocelyn Falempe Reviewed-by: Jocelyn Falempe v3: - explain the semantics of the new switch statement (Jocelyn) Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-15-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_dp.c | 132 +++++++++++++++++------------------ 1 file changed, 63 insertions(+), 69 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 3f5e586325a9..54b5a350a988 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -14,82 +14,76 @@ #include "ast_drv.h" #include "ast_vbios.h" -static int ast_astdp_get_mode_index(const struct ast_vbios_enhtable *vmode) +struct ast_astdp_mode_index_table_entry { + unsigned int hdisplay; + unsigned int vdisplay; + unsigned int mode_index; +}; + +/* FIXME: Do refresh rate and flags actually matter? */ +static const struct ast_astdp_mode_index_table_entry ast_astdp_mode_index_table[] = { + { 320, 240, ASTDP_320x240_60 }, + { 400, 300, ASTDP_400x300_60 }, + { 512, 384, ASTDP_512x384_60 }, + { 640, 480, ASTDP_640x480_60 }, + { 800, 600, ASTDP_800x600_56 }, + { 1024, 768, ASTDP_1024x768_60 }, + { 1152, 864, ASTDP_1152x864_75 }, + { 1280, 800, ASTDP_1280x800_60_RB }, + { 1280, 1024, ASTDP_1280x1024_60 }, + { 1360, 768, ASTDP_1366x768_60 }, // same as 1366x786 + { 1366, 768, ASTDP_1366x768_60 }, + { 1440, 900, ASTDP_1440x900_60_RB }, + { 1600, 900, ASTDP_1600x900_60_RB }, + { 1600, 1200, ASTDP_1600x1200_60 }, + { 1680, 1050, ASTDP_1680x1050_60_RB }, + { 1920, 1080, ASTDP_1920x1080_60 }, + { 1920, 1200, ASTDP_1920x1200_60 }, + { 0 } +}; + +static int __ast_astdp_get_mode_index(unsigned int hdisplay, unsigned int vdisplay) { - u8 refresh_rate_index; + const struct ast_astdp_mode_index_table_entry *entry = ast_astdp_mode_index_table; - if (vmode->refresh_rate_index < 1 || vmode->refresh_rate_index > 255) - return -EINVAL; - - refresh_rate_index = vmode->refresh_rate_index - 1; - - switch (vmode->hde) { - case 320: - if (vmode->vde == 240) - return ASTDP_320x240_60; - break; - case 400: - if (vmode->vde == 300) - return ASTDP_400x300_60; - break; - case 512: - if (vmode->vde == 384) - return ASTDP_512x384_60; - break; - case 640: - if (vmode->vde == 480) - return (u8)(ASTDP_640x480_60 + (u8)refresh_rate_index); - break; - case 800: - if (vmode->vde == 600) - return (u8)(ASTDP_800x600_56 + (u8)refresh_rate_index); - break; - case 1024: - if (vmode->vde == 768) - return (u8)(ASTDP_1024x768_60 + (u8)refresh_rate_index); - break; - case 1152: - if (vmode->vde == 864) - return ASTDP_1152x864_75; - break; - case 1280: - if (vmode->vde == 800) - return (u8)(ASTDP_1280x800_60_RB - (u8)refresh_rate_index); - if (vmode->vde == 1024) - return (u8)(ASTDP_1280x1024_60 + (u8)refresh_rate_index); - break; - case 1360: - case 1366: - if (vmode->vde == 768) - return ASTDP_1366x768_60; - break; - case 1440: - if (vmode->vde == 900) - return (u8)(ASTDP_1440x900_60_RB - (u8)refresh_rate_index); - break; - case 1600: - if (vmode->vde == 900) - return (u8)(ASTDP_1600x900_60_RB - (u8)refresh_rate_index); - if (vmode->vde == 1200) - return ASTDP_1600x1200_60; - break; - case 1680: - if (vmode->vde == 1050) - return (u8)(ASTDP_1680x1050_60_RB - (u8)refresh_rate_index); - break; - case 1920: - if (vmode->vde == 1080) - return ASTDP_1920x1080_60; - if (vmode->vde == 1200) - return ASTDP_1920x1200_60; - break; - default: - break; + while (entry->hdisplay && entry->vdisplay) { + if (entry->hdisplay == hdisplay && entry->vdisplay == vdisplay) + return entry->mode_index; + ++entry; } return -EINVAL; } +static int ast_astdp_get_mode_index(const struct ast_vbios_enhtable *vmode) +{ + int mode_index; + u8 refresh_rate_index; + + mode_index = __ast_astdp_get_mode_index(vmode->hde, vmode->vde); + if (mode_index < 0) + return mode_index; + + if (vmode->refresh_rate_index < 1 || vmode->refresh_rate_index > 255) + return -EINVAL; + refresh_rate_index = vmode->refresh_rate_index - 1; + + /* FIXME: Why are we doing this? */ + switch (mode_index) { + case ASTDP_1280x800_60_RB: + case ASTDP_1440x900_60_RB: + case ASTDP_1600x900_60_RB: + case ASTDP_1680x1050_60_RB: + mode_index = (u8)(mode_index - (u8)refresh_rate_index); + break; + default: + mode_index = (u8)(mode_index + (u8)refresh_rate_index); + break; + } + + return mode_index; +} + static bool ast_astdp_is_connected(struct ast_device *ast) { if (!ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xDF, AST_IO_VGACRDF_HPD)) From ec1d13fa90d6b760d01f6054faa9a0884af6c4d3 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:15 +0100 Subject: [PATCH 0373/1443] drm/ast: Remove struct ast_vbios_mode_info The type struct ast_vbios_mode_info used to store information about the color format and display mode. It has outlived its purpose. Inline its fields into struct ast_crtc_state and replace all instances. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-16-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_dp.c | 3 +- drivers/gpu/drm/ast/ast_drv.h | 8 ++--- drivers/gpu/drm/ast/ast_mode.c | 65 +++++++++++++++------------------- 3 files changed, 32 insertions(+), 44 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 54b5a350a988..192e640626c3 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -323,11 +323,10 @@ static void ast_astdp_encoder_helper_atomic_mode_set(struct drm_encoder *encoder struct drm_device *dev = encoder->dev; struct ast_device *ast = to_ast_device(dev); struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state); - struct ast_vbios_mode_info *vbios_mode_info = &ast_crtc_state->vbios_mode_info; int mode_index; u8 vgacre0, vgacre1, vgacre2; - mode_index = ast_astdp_get_mode_index(vbios_mode_info->enh_table); + mode_index = ast_astdp_get_mode_index(ast_crtc_state->vmode); if (drm_WARN_ON(dev, mode_index < 0)) return; diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 9309cc5e81b0..022a8c070c1b 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -358,18 +358,14 @@ struct ast_vbios_dclk_info { u8 param3; }; -struct ast_vbios_mode_info { - const struct ast_vbios_stdtable *std_table; - const struct ast_vbios_enhtable *enh_table; -}; - struct ast_crtc_state { struct drm_crtc_state base; /* Last known format of primary plane */ const struct drm_format_info *format; - struct ast_vbios_mode_info vbios_mode_info; + const struct ast_vbios_stdtable *std_table; + const struct ast_vbios_enhtable *vmode; }; #define to_ast_crtc_state(state) container_of(state, struct ast_crtc_state, base) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 1676958e9768..a93994b00f40 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -109,7 +109,7 @@ static void ast_crtc_set_gamma(struct ast_device *ast, static void ast_set_vbios_color_reg(struct ast_device *ast, const struct drm_format_info *format, - const struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_enhtable *vmode) { u32 color_index; @@ -132,7 +132,7 @@ static void ast_set_vbios_color_reg(struct ast_device *ast, ast_set_index_reg(ast, AST_IO_VGACRI, 0x91, 0x00); - if (vbios_mode->enh_table->flags & NewModeInfo) { + if (vmode->flags & NewModeInfo) { ast_set_index_reg(ast, AST_IO_VGACRI, 0x91, 0xa8); ast_set_index_reg(ast, AST_IO_VGACRI, 0x92, format->cpp[0] * 8); } @@ -140,19 +140,19 @@ static void ast_set_vbios_color_reg(struct ast_device *ast, static void ast_set_vbios_mode_reg(struct ast_device *ast, const struct drm_display_mode *adjusted_mode, - const struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_enhtable *vmode) { u32 refresh_rate_index, mode_id; - refresh_rate_index = vbios_mode->enh_table->refresh_rate_index; - mode_id = vbios_mode->enh_table->mode_id; + refresh_rate_index = vmode->refresh_rate_index; + mode_id = vmode->mode_id; ast_set_index_reg(ast, AST_IO_VGACRI, 0x8d, refresh_rate_index & 0xff); ast_set_index_reg(ast, AST_IO_VGACRI, 0x8e, mode_id & 0xff); ast_set_index_reg(ast, AST_IO_VGACRI, 0x91, 0x00); - if (vbios_mode->enh_table->flags & NewModeInfo) { + if (vmode->flags & NewModeInfo) { ast_set_index_reg(ast, AST_IO_VGACRI, 0x91, 0xa8); ast_set_index_reg(ast, AST_IO_VGACRI, 0x93, adjusted_mode->clock / 1000); ast_set_index_reg(ast, AST_IO_VGACRI, 0x94, adjusted_mode->crtc_hdisplay); @@ -164,14 +164,11 @@ static void ast_set_vbios_mode_reg(struct ast_device *ast, static void ast_set_std_reg(struct ast_device *ast, struct drm_display_mode *mode, - struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_stdtable *stdtable) { - const struct ast_vbios_stdtable *stdtable; u32 i; u8 jreg; - stdtable = vbios_mode->std_table; - jreg = stdtable->misc; ast_io_write8(ast, AST_IO_VGAMR_W, jreg); @@ -212,13 +209,13 @@ static void ast_set_std_reg(struct ast_device *ast, static void ast_set_crtc_reg(struct ast_device *ast, struct drm_display_mode *mode, - struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_enhtable *vmode) { u8 jreg05 = 0, jreg07 = 0, jreg09 = 0, jregAC = 0, jregAD = 0, jregAE = 0; u16 temp, precache = 0; if ((IS_AST_GEN6(ast) || IS_AST_GEN7(ast)) && - (vbios_mode->enh_table->flags & AST2500PreCatchCRT)) + (vmode->flags & AST2500PreCatchCRT)) precache = 40; ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x11, 0x7f, 0x00); @@ -337,14 +334,14 @@ static void ast_set_offset_reg(struct ast_device *ast, static void ast_set_dclk_reg(struct ast_device *ast, struct drm_display_mode *mode, - struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_enhtable *vmode) { const struct ast_vbios_dclk_info *clk_info; if (IS_AST_GEN6(ast) || IS_AST_GEN7(ast)) - clk_info = &dclk_table_ast2500[vbios_mode->enh_table->dclk_index]; + clk_info = &dclk_table_ast2500[vmode->dclk_index]; else - clk_info = &dclk_table[vbios_mode->enh_table->dclk_index]; + clk_info = &dclk_table[vmode->dclk_index]; ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xc0, 0x00, clk_info->param1); ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xc1, 0x00, clk_info->param2); @@ -402,15 +399,15 @@ static void ast_set_crtthd_reg(struct ast_device *ast) static void ast_set_sync_reg(struct ast_device *ast, struct drm_display_mode *mode, - struct ast_vbios_mode_info *vbios_mode) + const struct ast_vbios_enhtable *vmode) { u8 jreg; jreg = ast_io_read8(ast, AST_IO_VGAMR_R); jreg &= ~0xC0; - if (vbios_mode->enh_table->flags & NVSync) + if (vmode->flags & NVSync) jreg |= 0x80; - if (vbios_mode->enh_table->flags & NHSync) + if (vmode->flags & NHSync) jreg |= 0x40; ast_io_write8(ast, AST_IO_VGAMR_W, jreg); } @@ -530,10 +527,9 @@ static void ast_primary_plane_helper_atomic_update(struct drm_plane *plane, if (!old_fb || (fb->format != old_fb->format) || crtc_state->mode_changed) { struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state); - struct ast_vbios_mode_info *vbios_mode_info = &ast_crtc_state->vbios_mode_info; ast_set_color_reg(ast, fb->format); - ast_set_vbios_color_reg(ast, fb->format, vbios_mode_info); + ast_set_vbios_color_reg(ast, fb->format, ast_crtc_state->vmode); } drm_atomic_helper_damage_iter_init(&iter, old_plane_state, plane_state); @@ -912,8 +908,8 @@ static void ast_crtc_helper_mode_set_nofb(struct drm_crtc *crtc) struct ast_device *ast = to_ast_device(dev); struct drm_crtc_state *crtc_state = crtc->state; struct ast_crtc_state *ast_crtc_state = to_ast_crtc_state(crtc_state); - struct ast_vbios_mode_info *vbios_mode_info = - &ast_crtc_state->vbios_mode_info; + const struct ast_vbios_stdtable *std_table = ast_crtc_state->std_table; + const struct ast_vbios_enhtable *vmode = ast_crtc_state->vmode; struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode; /* @@ -924,13 +920,13 @@ static void ast_crtc_helper_mode_set_nofb(struct drm_crtc *crtc) */ ast_wait_for_vretrace(ast); - ast_set_vbios_mode_reg(ast, adjusted_mode, vbios_mode_info); + ast_set_vbios_mode_reg(ast, adjusted_mode, vmode); ast_set_index_reg(ast, AST_IO_VGACRI, 0xa1, 0x06); - ast_set_std_reg(ast, adjusted_mode, vbios_mode_info); - ast_set_crtc_reg(ast, adjusted_mode, vbios_mode_info); - ast_set_dclk_reg(ast, adjusted_mode, vbios_mode_info); + ast_set_std_reg(ast, adjusted_mode, std_table); + ast_set_crtc_reg(ast, adjusted_mode, vmode); + ast_set_dclk_reg(ast, adjusted_mode, vmode); ast_set_crtthd_reg(ast); - ast_set_sync_reg(ast, adjusted_mode, vbios_mode_info); + ast_set_sync_reg(ast, adjusted_mode, vmode); } static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, @@ -944,7 +940,6 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, struct ast_device *ast = to_ast_device(dev); struct ast_crtc_state *ast_state; const struct drm_format_info *format; - struct ast_vbios_mode_info *vbios_mode; const struct ast_vbios_enhtable *vmode; unsigned int hborder = 0; unsigned int vborder = 0; @@ -979,8 +974,6 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, } } - vbios_mode = &ast_state->vbios_mode_info; - /* * Set register tables. * @@ -989,14 +982,14 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, */ switch (format->format) { case DRM_FORMAT_C8: - vbios_mode->std_table = &vbios_stdtable[VGAModeIndex]; + ast_state->std_table = &vbios_stdtable[VGAModeIndex]; break; case DRM_FORMAT_RGB565: - vbios_mode->std_table = &vbios_stdtable[HiCModeIndex]; + ast_state->std_table = &vbios_stdtable[HiCModeIndex]; break; case DRM_FORMAT_RGB888: case DRM_FORMAT_XRGB8888: - vbios_mode->std_table = &vbios_stdtable[TrueCModeIndex]; + ast_state->std_table = &vbios_stdtable[TrueCModeIndex]; break; default: return -EINVAL; @@ -1009,7 +1002,7 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, vmode = ast_vbios_find_mode(ast, &crtc_state->mode); if (!vmode) return -EINVAL; - ast_state->vbios_mode_info.enh_table = vmode; + ast_state->vmode = vmode; if (vmode->flags & HBorder) hborder = 8; @@ -1130,8 +1123,8 @@ ast_crtc_atomic_duplicate_state(struct drm_crtc *crtc) ast_state = to_ast_crtc_state(crtc->state); new_ast_state->format = ast_state->format; - memcpy(&new_ast_state->vbios_mode_info, &ast_state->vbios_mode_info, - sizeof(new_ast_state->vbios_mode_info)); + new_ast_state->std_table = ast_state->std_table; + new_ast_state->vmode = ast_state->vmode; return &new_ast_state->base; } From 0c5f9cb67897fc48aef8883d21a8d74d11a2f8e3 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Fri, 31 Jan 2025 10:21:16 +0100 Subject: [PATCH 0374/1443] drm/ast: Only look up VBIOS mode on full modesets Only look up the VBIOS mode during atomic_check if the display mode changes. For page flips, the previous settings still apply. Avoids the runtime overhead of looking up the VBIOS mode on each page flip. Signed-off-by: Thomas Zimmermann Reviewed-by: Jocelyn Falempe Link: https://patchwork.freedesktop.org/patch/msgid/20250131092257.115596-17-tzimmermann@suse.de --- drivers/gpu/drm/ast/ast_mode.c | 46 ++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index a93994b00f40..bd781293b6d9 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -996,32 +996,34 @@ static int ast_crtc_helper_atomic_check(struct drm_crtc *crtc, } /* - * Find the VBIOS mode and adjust the DRM display mode accordingly. + * Find the VBIOS mode and adjust the DRM display mode accordingly + * if a full modeset is required. Otherwise keep the existing values. */ + if (drm_atomic_crtc_needs_modeset(crtc_state)) { + vmode = ast_vbios_find_mode(ast, &crtc_state->mode); + if (!vmode) + return -EINVAL; + ast_state->vmode = vmode; - vmode = ast_vbios_find_mode(ast, &crtc_state->mode); - if (!vmode) - return -EINVAL; - ast_state->vmode = vmode; + if (vmode->flags & HBorder) + hborder = 8; + if (vmode->flags & VBorder) + vborder = 8; - if (vmode->flags & HBorder) - hborder = 8; - if (vmode->flags & VBorder) - vborder = 8; + adjusted_mode->crtc_hdisplay = vmode->hde; + adjusted_mode->crtc_hblank_start = vmode->hde + hborder; + adjusted_mode->crtc_hblank_end = vmode->ht - hborder; + adjusted_mode->crtc_hsync_start = vmode->hde + hborder + vmode->hfp; + adjusted_mode->crtc_hsync_end = vmode->hde + hborder + vmode->hfp + vmode->hsync; + adjusted_mode->crtc_htotal = vmode->ht; - adjusted_mode->crtc_hdisplay = vmode->hde; - adjusted_mode->crtc_hblank_start = vmode->hde + hborder; - adjusted_mode->crtc_hblank_end = vmode->ht - hborder; - adjusted_mode->crtc_hsync_start = vmode->hde + hborder + vmode->hfp; - adjusted_mode->crtc_hsync_end = vmode->hde + hborder + vmode->hfp + vmode->hsync; - adjusted_mode->crtc_htotal = vmode->ht; - - adjusted_mode->crtc_vdisplay = vmode->vde; - adjusted_mode->crtc_vblank_start = vmode->vde + vborder; - adjusted_mode->crtc_vblank_end = vmode->vt - vborder; - adjusted_mode->crtc_vsync_start = vmode->vde + vborder + vmode->vfp; - adjusted_mode->crtc_vsync_end = vmode->vde + vborder + vmode->vfp + vmode->vsync; - adjusted_mode->crtc_vtotal = vmode->vt; + adjusted_mode->crtc_vdisplay = vmode->vde; + adjusted_mode->crtc_vblank_start = vmode->vde + vborder; + adjusted_mode->crtc_vblank_end = vmode->vt - vborder; + adjusted_mode->crtc_vsync_start = vmode->vde + vborder + vmode->vfp; + adjusted_mode->crtc_vsync_end = vmode->vde + vborder + vmode->vfp + vmode->vsync; + adjusted_mode->crtc_vtotal = vmode->vt; + } return 0; } From 6f36f103cff1737094f2187b1f9a7b312820d377 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 3 Feb 2025 15:10:27 +0200 Subject: [PATCH 0375/1443] pinctrl: cy8c95x0: Fix off-by-one in the regmap range settings The range_max is inclusive, so we need to use the number of the last accessible register address. Fixes: 8670de9fae49 ("pinctrl: cy8c95x0: Use regmap ranges") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/20250203131506.3318201-2-andriy.shevchenko@linux.intel.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-cy8c95x0.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/pinctrl/pinctrl-cy8c95x0.c b/drivers/pinctrl/pinctrl-cy8c95x0.c index 0d6c2027d4c1..5c6bcbf6c337 100644 --- a/drivers/pinctrl/pinctrl-cy8c95x0.c +++ b/drivers/pinctrl/pinctrl-cy8c95x0.c @@ -1438,15 +1438,15 @@ static int cy8c95x0_probe(struct i2c_client *client) switch (chip->tpin) { case 20: strscpy(chip->name, cy8c95x0_id[0].name); - regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 3 * MUXED_STRIDE; + regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 3 * MUXED_STRIDE - 1; break; case 40: strscpy(chip->name, cy8c95x0_id[1].name); - regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 6 * MUXED_STRIDE; + regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 6 * MUXED_STRIDE - 1; break; case 60: strscpy(chip->name, cy8c95x0_id[2].name); - regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 8 * MUXED_STRIDE; + regmap_range_conf.range_max = CY8C95X0_VIRTUAL + 8 * MUXED_STRIDE - 1; break; default: return -ENODEV; From 3fbe3fe28764455e4fc3578afb9765f46f9ce93d Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 3 Feb 2025 15:10:28 +0200 Subject: [PATCH 0376/1443] pinctrl: cy8c95x0: Avoid accessing reserved registers The checks for vrtual registers in the cy8c95x0_readable_register() and cy8c95x0_writeable_register() are not aligned and broken. Fix that by explicitly avoiding reserved registers to be accessed. Fixes: 71e4001a0455 ("pinctrl: pinctrl-cy8c95x0: Fix regcache") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/20250203131506.3318201-3-andriy.shevchenko@linux.intel.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-cy8c95x0.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/pinctrl/pinctrl-cy8c95x0.c b/drivers/pinctrl/pinctrl-cy8c95x0.c index 5c6bcbf6c337..c787a9aadfdf 100644 --- a/drivers/pinctrl/pinctrl-cy8c95x0.c +++ b/drivers/pinctrl/pinctrl-cy8c95x0.c @@ -328,14 +328,14 @@ static int cypress_get_pin_mask(struct cy8c95x0_pinctrl *chip, unsigned int pin) static bool cy8c95x0_readable_register(struct device *dev, unsigned int reg) { /* - * Only 12 registers are present per port (see Table 6 in the - * datasheet). + * Only 12 registers are present per port (see Table 6 in the datasheet). */ - if (reg >= CY8C95X0_VIRTUAL && (reg % MUXED_STRIDE) < 12) - return true; + if (reg >= CY8C95X0_VIRTUAL && (reg % MUXED_STRIDE) >= 12) + return false; switch (reg) { case 0x24 ... 0x27: + case 0x31 ... 0x3f: return false; default: return true; @@ -344,8 +344,11 @@ static bool cy8c95x0_readable_register(struct device *dev, unsigned int reg) static bool cy8c95x0_writeable_register(struct device *dev, unsigned int reg) { - if (reg >= CY8C95X0_VIRTUAL) - return true; + /* + * Only 12 registers are present per port (see Table 6 in the datasheet). + */ + if (reg >= CY8C95X0_VIRTUAL && (reg % MUXED_STRIDE) >= 12) + return false; switch (reg) { case CY8C95X0_INPUT_(0) ... CY8C95X0_INPUT_(7): @@ -353,6 +356,7 @@ static bool cy8c95x0_writeable_register(struct device *dev, unsigned int reg) case CY8C95X0_DEVID: return false; case 0x24 ... 0x27: + case 0x31 ... 0x3f: return false; default: return true; From aac4470fa6e695e4d6ac94cc77d4690b57f1d2bc Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 3 Feb 2025 15:10:29 +0200 Subject: [PATCH 0377/1443] pinctrl: cy8c95x0: Enable regmap locking for debug When regmap locking is disabled, debugfs is also disabled. Enable locking for debug when CONFIG_DEBUG_PINCTRL is set. Fixes: f71aba339a66 ("pinctrl: cy8c95x0: Use single I2C lock") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/20250203131506.3318201-4-andriy.shevchenko@linux.intel.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-cy8c95x0.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/pinctrl/pinctrl-cy8c95x0.c b/drivers/pinctrl/pinctrl-cy8c95x0.c index c787a9aadfdf..bfa16f70e29c 100644 --- a/drivers/pinctrl/pinctrl-cy8c95x0.c +++ b/drivers/pinctrl/pinctrl-cy8c95x0.c @@ -470,7 +470,11 @@ static const struct regmap_config cy8c9520_i2c_regmap = { .max_register = 0, /* Updated at runtime */ .num_reg_defaults_raw = 0, /* Updated at runtime */ .use_single_read = true, /* Workaround for regcache bug */ +#if IS_ENABLED(CONFIG_DEBUG_PINCTRL) + .disable_locking = false, +#else .disable_locking = true, +#endif }; static inline int cy8c95x0_regmap_update_bits_base(struct cy8c95x0_pinctrl *chip, From 0a7404fc5399e1100b14e7e2a4af2e4fd5e3b602 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 3 Feb 2025 15:10:30 +0200 Subject: [PATCH 0378/1443] pinctrl: cy8c95x0: Rename PWMSEL to SELPWM There are two registers in the hardware, one, "Select PWM", is per-port configuration enabling PWM function instead of GPIO. The other one is "PWM Select" is per-PWM selector to configure PWM itself. Original code uses abbreviation of the latter to describe the former. Rename it to follow the datasheet. Fixes: e6cbbe42944d ("pinctrl: Add Cypress cy8c95x0 support") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/20250203131506.3318201-5-andriy.shevchenko@linux.intel.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-cy8c95x0.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/pinctrl/pinctrl-cy8c95x0.c b/drivers/pinctrl/pinctrl-cy8c95x0.c index bfa16f70e29c..75100a9fb8e4 100644 --- a/drivers/pinctrl/pinctrl-cy8c95x0.c +++ b/drivers/pinctrl/pinctrl-cy8c95x0.c @@ -42,7 +42,7 @@ #define CY8C95X0_PORTSEL 0x18 /* Port settings, write PORTSEL first */ #define CY8C95X0_INTMASK 0x19 -#define CY8C95X0_PWMSEL 0x1A +#define CY8C95X0_SELPWM 0x1A #define CY8C95X0_INVERT 0x1B #define CY8C95X0_DIRECTION 0x1C /* Drive mode register change state on writing '1' */ @@ -369,8 +369,8 @@ static bool cy8c95x0_volatile_register(struct device *dev, unsigned int reg) case CY8C95X0_INPUT_(0) ... CY8C95X0_INPUT_(7): case CY8C95X0_INTSTATUS_(0) ... CY8C95X0_INTSTATUS_(7): case CY8C95X0_INTMASK: + case CY8C95X0_SELPWM: case CY8C95X0_INVERT: - case CY8C95X0_PWMSEL: case CY8C95X0_DIRECTION: case CY8C95X0_DRV_PU: case CY8C95X0_DRV_PD: @@ -399,7 +399,7 @@ static bool cy8c95x0_muxed_register(unsigned int reg) { switch (reg) { case CY8C95X0_INTMASK: - case CY8C95X0_PWMSEL: + case CY8C95X0_SELPWM: case CY8C95X0_INVERT: case CY8C95X0_DIRECTION: case CY8C95X0_DRV_PU: @@ -797,7 +797,7 @@ static int cy8c95x0_gpio_get_pincfg(struct cy8c95x0_pinctrl *chip, reg = CY8C95X0_DIRECTION; break; case PIN_CONFIG_MODE_PWM: - reg = CY8C95X0_PWMSEL; + reg = CY8C95X0_SELPWM; break; case PIN_CONFIG_OUTPUT: reg = CY8C95X0_OUTPUT; @@ -876,7 +876,7 @@ static int cy8c95x0_gpio_set_pincfg(struct cy8c95x0_pinctrl *chip, reg = CY8C95X0_DRV_PP_FAST; break; case PIN_CONFIG_MODE_PWM: - reg = CY8C95X0_PWMSEL; + reg = CY8C95X0_SELPWM; break; case PIN_CONFIG_OUTPUT_ENABLE: return cy8c95x0_pinmux_direction(chip, off, !arg); @@ -1161,7 +1161,7 @@ static void cy8c95x0_pin_dbg_show(struct pinctrl_dev *pctldev, struct seq_file * bitmap_zero(mask, MAX_LINE); __set_bit(pin, mask); - if (cy8c95x0_read_regs_mask(chip, CY8C95X0_PWMSEL, pwm, mask)) { + if (cy8c95x0_read_regs_mask(chip, CY8C95X0_SELPWM, pwm, mask)) { seq_puts(s, "not available"); return; } @@ -1206,7 +1206,7 @@ static int cy8c95x0_set_mode(struct cy8c95x0_pinctrl *chip, unsigned int off, bo u8 port = cypress_get_port(chip, off); u8 bit = cypress_get_pin_mask(chip, off); - return cy8c95x0_regmap_write_bits(chip, CY8C95X0_PWMSEL, port, bit, mode ? bit : 0); + return cy8c95x0_regmap_write_bits(chip, CY8C95X0_SELPWM, port, bit, mode ? bit : 0); } static int cy8c95x0_pinmux_mode(struct cy8c95x0_pinctrl *chip, From 448060463198924c0a485e7e1622fa8a9c03cf3e Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Tue, 17 Dec 2024 14:07:23 +0530 Subject: [PATCH 0379/1443] drm/i915/hdcp: Fix Repeater authentication during topology change When topology changes, before beginning a new HDCP authentication by sending AKE_init message we need to first authenticate only the repeater. Only after repeater authentication failure, it makes sense to start a new HDCP authentication. Even though it made sense to not enable HDCP directly from check_link and schedule it for later, repeater authentication needs to be done immediately. --v2 -Fix comment grammatical errors [Ankit] Fixes: 47ef55a8b784 ("drm/i915/hdcp: Don't enable HDCP2.2 directly from check_link") Signed-off-by: Suraj Kandpal Reviewed-by: Ankit Nautiyal Link: https://patchwork.freedesktop.org/patch/msgid/20241217083723.2883317-1-suraj.kandpal@intel.com (cherry picked from commit 605a33e765890e4f1345315afc25268d4ae0fb7c) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_hdcp.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 7464b44c8bb3..c60b22aaa819 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -2188,6 +2188,19 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) drm_dbg_kms(display->drm, "HDCP2.2 Downstream topology change\n"); + + ret = hdcp2_authenticate_repeater_topology(connector); + if (!ret) { + intel_hdcp_update_value(connector, + DRM_MODE_CONTENT_PROTECTION_ENABLED, + true); + goto out; + } + + drm_dbg_kms(display->drm, + "[CONNECTOR:%d:%s] Repeater topology auth failed.(%d)\n", + connector->base.base.id, connector->base.name, + ret); } else { drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s] HDCP2.2 link failed, retrying auth\n", From 8dd5a5eb6a209e3bdb4e536e36698400445c6c2e Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Fri, 17 Jan 2025 09:42:48 +0530 Subject: [PATCH 0380/1443] drm/i915/hdcp: Use correct function to check if encoder is HDMI Use intel_encoder_is_hdmi function which was recently introduced to see if encoder is HDMI or not. --v2 -Add Fixes tag [Jani] Fixes: 6a3691ca4799 ("drm/i915/hdcp: Disable HDCP Line Rekeying for HDCP2.2 on HDMI") Signed-off-by: Suraj Kandpal Reviewed-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20250117041247.1084381-1-suraj.kandpal@intel.com (cherry picked from commit 2499212e21601740ed7d5563563f39cf7e7d833a) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_hdcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index c60b22aaa819..1bab7c34a794 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -41,7 +41,7 @@ intel_hdcp_adjust_hdcp_line_rekeying(struct intel_encoder *encoder, u32 rekey_bit = 0; /* Here we assume HDMI is in TMDS mode of operation */ - if (encoder->type != INTEL_OUTPUT_HDMI) + if (!intel_encoder_is_hdmi(encoder)) return; if (DISPLAY_VER(display) >= 30) { From cb5fab2afd906307876d79537ef0329033c40dd3 Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Thu, 23 Jan 2025 11:38:39 -0800 Subject: [PATCH 0381/1443] drm/i915/pmu: Fix zero delta busyness issue When running igt@gem_exec_balancer@individual for multiple iterations, it is seen that the delta busyness returned by PMU is 0. The issue stems from a combination of 2 implementation specific details: 1) gt_park is throttling __update_guc_busyness_stats() so that it does not hog PCI bandwidth for some use cases. (Ref: 59bcdb564b3ba) 2) busyness implementation always returns monotonically increasing counters. (Ref: cf907f6d29421) If an application queried an engine while it was active, engine->stats.guc.running is set to true. Following that, if all PM wakeref's are released, then gt is parked. At this time the throttling of __update_guc_busyness_stats() may result in a missed update to the running state of the engine (due to (1) above). This means subsequent calls to guc_engine_busyness() will think that the engine is still running and they will keep updating the cached counter (stats->total). This results in an inflated cached counter. Later when the application runs a workload and queries for busyness, we return the cached value since it is larger than the actual value (due to (2) above) All subsequent queries will return the same large (inflated) value, so the application sees a delta busyness of zero. Fix the issue by resetting the running state of engines each time intel_guc_busyness_park() is called. v2: (Rodrigo) - Use the correct tag in commit message - Drop the redundant wakeref check in guc_engine_busyness() and update commit message Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13366 Fixes: cf907f6d2942 ("i915/guc: Ensure busyness counter increases motonically") Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250123193839.2394694-1-umesh.nerlige.ramappa@intel.com (cherry picked from commit 431b742e2bfc9f6dd713f261629741980996d001) Signed-off-by: Rodrigo Vivi --- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 12f1ba7ca9c1..bd4b3d2470e4 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -1469,6 +1469,19 @@ static void __reset_guc_busyness_stats(struct intel_guc *guc) spin_unlock_irqrestore(&guc->timestamp.lock, flags); } +static void __update_guc_busyness_running_state(struct intel_guc *guc) +{ + struct intel_gt *gt = guc_to_gt(guc); + struct intel_engine_cs *engine; + enum intel_engine_id id; + unsigned long flags; + + spin_lock_irqsave(&guc->timestamp.lock, flags); + for_each_engine(engine, gt, id) + engine->stats.guc.running = false; + spin_unlock_irqrestore(&guc->timestamp.lock, flags); +} + static void __update_guc_busyness_stats(struct intel_guc *guc) { struct intel_gt *gt = guc_to_gt(guc); @@ -1619,6 +1632,9 @@ void intel_guc_busyness_park(struct intel_gt *gt) if (!guc_submission_initialized(guc)) return; + /* Assume no engines are running and set running state to false */ + __update_guc_busyness_running_state(guc); + /* * There is a race with suspend flow where the worker runs after suspend * and causes an unclaimed register access warning. Cancel the worker From fa6182c8b13ebfdc70ebdc09161a70dd8131f3b1 Mon Sep 17 00:00:00 2001 From: Brian Geffon Date: Mon, 27 Jan 2025 15:43:32 -0500 Subject: [PATCH 0382/1443] drm/i915: Fix page cleanup on DMA remap failure MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When converting to folios the cleanup path of shmem_get_pages() was missed. When a DMA remap fails and the max segment size is greater than PAGE_SIZE it will attempt to retry the remap with a PAGE_SIZEd segment size. The cleanup code isn't properly using the folio apis and as a result isn't handling compound pages correctly. v2 -> v3: (Ville) Just use shmem_sg_free_table() as-is in the failure path of shmem_get_pages(). shmem_sg_free_table() will clear mapping unevictable but it will be reset when it retries in shmem_sg_alloc_table(). v1 -> v2: (Ville) Fixed locations where we were not clearing mapping unevictable. Cc: stable@vger.kernel.org Cc: Ville Syrjala Cc: Vidya Srinivas Link: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/13487 Link: https://lore.kernel.org/lkml/20250116135636.410164-1-bgeffon@google.com/ Fixes: 0b62af28f249 ("i915: convert shmem_sg_free_table() to use a folio_batch") Signed-off-by: Brian Geffon Suggested-by: Tomasz Figa Link: https://patchwork.freedesktop.org/patch/msgid/20250127204332.336665-1-bgeffon@google.com Reviewed-by: Jonathan Cavitt Tested-by: Vidya Srinivas Signed-off-by: Ville Syrjälä (cherry picked from commit 9e304a18630875352636ad52a3d2af47c3bde824) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index fe69f2c8527d..ae3343c81a64 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -209,8 +209,6 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj) struct address_space *mapping = obj->base.filp->f_mapping; unsigned int max_segment = i915_sg_segment_size(i915->drm.dev); struct sg_table *st; - struct sgt_iter sgt_iter; - struct page *page; int ret; /* @@ -239,9 +237,7 @@ static int shmem_get_pages(struct drm_i915_gem_object *obj) * for PAGE_SIZE chunks instead may be helpful. */ if (max_segment > PAGE_SIZE) { - for_each_sgt_page(page, sgt_iter, st) - put_page(page); - sg_free_table(st); + shmem_sg_free_table(st, mapping, false, false); kfree(st); max_segment = PAGE_SIZE; From c7b49506b3ba7a62335e6f666a43f67d5cd9fd1e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= Date: Wed, 18 Dec 2024 19:36:47 +0200 Subject: [PATCH 0383/1443] drm/i915: Drop 64bpp YUV formats from ICL+ SDR planes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I'm seeing underruns with these 64bpp YUV formats on TGL. The weird details: - only happens on pipe B/C/D SDR planes, pipe A SDR planes seem fine, as do all HDR planes - somehow CDCLK related, higher CDCLK allows for bigger plane with these formats without underruns. With 300MHz CDCLK I can only go up to 1200 pixels wide or so, with 650MHz even a 3840 pixel wide plane was OK - ICL and ADL so far appear unaffected So not really sure what's the deal with this, but bspec does state "64-bit formats supported only on the HDR planes" so let's just drop these formats from the SDR planes. We already disallow 64bpp RGB formats. Cc: stable@vger.kernel.org Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20241218173650.19782-2-ville.syrjala@linux.intel.com Reviewed-by: Juha-Pekka Heikkila (cherry picked from commit 35e1aacfe536d6e8d8d440cd7155366da2541ad4) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/skl_universal_plane.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index ff9764cac1e7..80e558042d97 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -106,8 +106,6 @@ static const u32 icl_sdr_y_plane_formats[] = { DRM_FORMAT_Y216, DRM_FORMAT_XYUV8888, DRM_FORMAT_XVYU2101010, - DRM_FORMAT_XVYU12_16161616, - DRM_FORMAT_XVYU16161616, }; static const u32 icl_sdr_uv_plane_formats[] = { @@ -134,8 +132,6 @@ static const u32 icl_sdr_uv_plane_formats[] = { DRM_FORMAT_Y216, DRM_FORMAT_XYUV8888, DRM_FORMAT_XVYU2101010, - DRM_FORMAT_XVYU12_16161616, - DRM_FORMAT_XVYU16161616, }; static const u32 icl_hdr_plane_formats[] = { From 57965269896313e1629a518d3971ad55f599b792 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Tue, 14 Jan 2025 16:13:34 -0800 Subject: [PATCH 0384/1443] drm/i915/guc: Debug print LRC state entries only if the context is pinned After the context is unpinned the backing memory can also be unpinned, so any accesses via the lrc_reg_state pointer can end up in unmapped memory. To avoid that, make sure to only access that memory if the context is pinned when printing its info. v2: fix newline alignment Fixes: 28ff6520a34d ("drm/i915/guc: Update GuC debugfs to support new GuC") Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Cc: Matthew Brost Cc: # v5.15+ Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250115001334.3875347-1-daniele.ceraolospurio@intel.com (cherry picked from commit 5bea40687c5cf2a33bf04e9110eb2e2b80222ef5) Signed-off-by: Rodrigo Vivi --- .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 20 +++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index bd4b3d2470e4..cc05bd9e43b4 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -5535,12 +5535,20 @@ static inline void guc_log_context(struct drm_printer *p, { drm_printf(p, "GuC lrc descriptor %u:\n", ce->guc_id.id); drm_printf(p, "\tHW Context Desc: 0x%08x\n", ce->lrc.lrca); - drm_printf(p, "\t\tLRC Head: Internal %u, Memory %u\n", - ce->ring->head, - ce->lrc_reg_state[CTX_RING_HEAD]); - drm_printf(p, "\t\tLRC Tail: Internal %u, Memory %u\n", - ce->ring->tail, - ce->lrc_reg_state[CTX_RING_TAIL]); + if (intel_context_pin_if_active(ce)) { + drm_printf(p, "\t\tLRC Head: Internal %u, Memory %u\n", + ce->ring->head, + ce->lrc_reg_state[CTX_RING_HEAD]); + drm_printf(p, "\t\tLRC Tail: Internal %u, Memory %u\n", + ce->ring->tail, + ce->lrc_reg_state[CTX_RING_TAIL]); + intel_context_unpin(ce); + } else { + drm_printf(p, "\t\tLRC Head: Internal %u, Memory not pinned\n", + ce->ring->head); + drm_printf(p, "\t\tLRC Tail: Internal %u, Memory not pinned\n", + ce->ring->tail); + } drm_printf(p, "\t\tContext Pin Count: %u\n", atomic_read(&ce->pin_count)); drm_printf(p, "\t\tGuC ID Ref Count: %u\n", From 4466302262b38f5e6c65325035b4036a42efc934 Mon Sep 17 00:00:00 2001 From: Ankit Nautiyal Date: Thu, 30 Jan 2025 10:46:06 +0530 Subject: [PATCH 0385/1443] drm/i915/dp: fix the Adaptive sync Operation mode for SDP Currently we support Adaptive sync operation mode with dynamic frame rate, but instead the operation mode with fixed rate is set. This was initially set correctly in the earlier version of changes but later got changed, while defining a macro for the same. Fixes: a5bd5991cb8a ("drm/i915/display: Compute AS SDP parameters") Cc: Mitul Golani Cc: Ankit Nautiyal Cc: Jani Nikula Reviewed-by: Mitul Golani Signed-off-by: Ankit Nautiyal Link: https://patchwork.freedesktop.org/patch/msgid/20250130051609.1796524-4-mitulkumar.ajitkumar.golani@intel.com (cherry picked from commit c5806862543ff6c2ad242409fcdf0667eac26dae) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_dp.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index f1f3b1bb1e89..bfc16fd25d22 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -2829,7 +2829,6 @@ static void intel_dp_compute_as_sdp(struct intel_dp *intel_dp, crtc_state->infoframes.enable |= intel_hdmi_infoframe_enable(DP_SDP_ADAPTIVE_SYNC); - /* Currently only DP_AS_SDP_AVT_FIXED_VTOTAL mode supported */ as_sdp->sdp_type = DP_SDP_ADAPTIVE_SYNC; as_sdp->length = 0x9; as_sdp->duration_incr_ms = 0; @@ -2840,7 +2839,7 @@ static void intel_dp_compute_as_sdp(struct intel_dp *intel_dp, as_sdp->target_rr = drm_mode_vrefresh(adjusted_mode); as_sdp->target_rr_divider = true; } else { - as_sdp->mode = DP_AS_SDP_AVT_FIXED_VTOTAL; + as_sdp->mode = DP_AS_SDP_AVT_DYNAMIC_VTOTAL; as_sdp->vtotal = adjusted_mode->vtotal; as_sdp->target_rr = 0; } From 985a44b02484a47f2c6ecbe971a5f0c47830120b Mon Sep 17 00:00:00 2001 From: Ankit Nautiyal Date: Fri, 31 Jan 2025 09:43:42 +0530 Subject: [PATCH 0386/1443] drm/i915/dp: Return min bpc supported by source instead of 0 Currently, intel_dp_dsc_max_src_input_bpc can return 0 for platforms not supporting DSC, which could theoretically cause issues in clamp() due to a low limit being greater than the high limit. Instead, return the minimum bpc supported by the source to prevent such issues. Reported-by: Linux Kernel Functional Testing Closes: https://lore.kernel.org/all/CA+G9fYtNfM399_=_ff81zeRJv=0+z7oFJfPGmJgTp6yrJmU+1w@mail.gmail.com/ Fixes: 160672b86b0d ("drm/i915/dp: Use clamp for pipe_bpp limits with DSC") Cc: Suraj Kandpal Cc: Jani Nikula Cc: Rodrigo Vivi Signed-off-by: Ankit Nautiyal Reviewed-by: Suraj Kandpal Tested-by: Chaitanya Kumar Borah Link: https://patchwork.freedesktop.org/patch/msgid/20250131041342.3086716-1-ankit.k.nautiyal@intel.com Signed-off-by: Rodrigo Vivi (cherry picked from commit a67221b5eb8d59fb7e1f0df3ef9945b6a0f32cca) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_dp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index bfc16fd25d22..be07034bfcc6 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -1791,7 +1791,7 @@ int intel_dp_dsc_max_src_input_bpc(struct intel_display *display) if (DISPLAY_VER(display) == 11) return 10; - return 0; + return intel_dp_dsc_min_src_input_bpc(); } int intel_dp_dsc_compute_max_bpp(const struct intel_connector *connector, From 3cf3ec911d70ee7774978f639fd3364c98d42b2c Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Tue, 21 Jan 2025 06:52:03 -0800 Subject: [PATCH 0387/1443] drm/i915/backlight: Return immediately when scale() finds invalid parameters The scale() functions detects invalid parameters, but continues its calculations anyway. This causes bad results if negative values are used for unsigned operations. Worst case, a division by 0 error will be seen if source_min == source_max. On top of that, after v6.13, the sequence of WARN_ON() followed by clamp() may result in a build error with gcc 13.x. drivers/gpu/drm/i915/display/intel_backlight.c: In function 'scale': include/linux/compiler_types.h:542:45: error: call to '__compiletime_assert_415' declared with attribute error: clamp() low limit source_min greater than high limit source_max This happens if the compiler decides to rearrange the code as follows. if (source_min > source_max) { WARN(..); /* Do the clamp() knowing that source_min > source_max */ source_val = clamp(source_val, source_min, source_max); } else { /* Do the clamp knowing that source_min <= source_max */ source_val = clamp(source_val, source_min, source_max); } Fix the problem by evaluating the return values from WARN_ON and returning immediately after a warning. While at it, fix divide by zero error seen if source_min == source_max. Analyzed-by: Linus Torvalds Suggested-by: Linus Torvalds Suggested-by: David Laight Cc: David Laight Cc: Jani Nikula Cc: Andy Shevchenko Signed-off-by: Guenter Roeck Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250121145203.2851237-1-linux@roeck-us.net Signed-off-by: Rodrigo Vivi (cherry picked from commit 6f71507415841d1a6d38118e5fa0eaf0caab9c17) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_backlight.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index fc1e517e074a..7e6ce905bdaf 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -41,8 +41,9 @@ static u32 scale(u32 source_val, { u64 target_val; - WARN_ON(source_min > source_max); - WARN_ON(target_min > target_max); + if (WARN_ON(source_min >= source_max) || + WARN_ON(target_min > target_max)) + return target_min; /* defensive */ source_val = clamp(source_val, source_min, source_max); From 6bb05a33337b2c842373857b63de5c9bf1ae2a09 Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Fri, 31 Jan 2025 12:33:23 -0500 Subject: [PATCH 0388/1443] clocksource: Use migrate_disable() to avoid calling get_random_u32() in atomic context The following bug report happened with a PREEMPT_RT kernel: BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 2012, name: kwatchdog preempt_count: 1, expected: 0 RCU nest depth: 0, expected: 0 get_random_u32+0x4f/0x110 clocksource_verify_choose_cpus+0xab/0x1a0 clocksource_verify_percpu.part.0+0x6b/0x330 clocksource_watchdog_kthread+0x193/0x1a0 It is due to the fact that clocksource_verify_choose_cpus() is invoked with preemption disabled. This function invokes get_random_u32() to obtain random numbers for choosing CPUs. The batched_entropy_32 local lock and/or the base_crng.lock spinlock in driver/char/random.c will be acquired during the call. In PREEMPT_RT kernel, they are both sleeping locks and so cannot be acquired in atomic context. Fix this problem by using migrate_disable() to allow smp_processor_id() to be reliably used without introducing atomic context. preempt_disable() is then called after clocksource_verify_choose_cpus() but before the clocksource measurement is being run to avoid introducing unexpected latency. Fixes: 7560c02bdffb ("clocksource: Check per-CPU clock synchronization when marked unstable") Suggested-by: Sebastian Andrzej Siewior Signed-off-by: Waiman Long Signed-off-by: Thomas Gleixner Reviewed-by: Paul E. McKenney Reviewed-by: Sebastian Andrzej Siewior Link: https://lore.kernel.org/all/20250131173323.891943-2-longman@redhat.com --- kernel/time/clocksource.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 77d9566d3aa6..2a7802ec480c 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -373,10 +373,10 @@ void clocksource_verify_percpu(struct clocksource *cs) cpumask_clear(&cpus_ahead); cpumask_clear(&cpus_behind); cpus_read_lock(); - preempt_disable(); + migrate_disable(); clocksource_verify_choose_cpus(); if (cpumask_empty(&cpus_chosen)) { - preempt_enable(); + migrate_enable(); cpus_read_unlock(); pr_warn("Not enough CPUs to check clocksource '%s'.\n", cs->name); return; @@ -384,6 +384,7 @@ void clocksource_verify_percpu(struct clocksource *cs) testcpu = smp_processor_id(); pr_info("Checking clocksource %s synchronization from CPU %d to CPUs %*pbl.\n", cs->name, testcpu, cpumask_pr_args(&cpus_chosen)); + preempt_disable(); for_each_cpu(cpu, &cpus_chosen) { if (cpu == testcpu) continue; @@ -403,6 +404,7 @@ void clocksource_verify_percpu(struct clocksource *cs) cs_nsec_min = cs_nsec; } preempt_enable(); + migrate_enable(); cpus_read_unlock(); if (!cpumask_empty(&cpus_ahead)) pr_warn(" CPUs %*pbl ahead of CPU %d for clocksource %s.\n", From 4fd2707e3e71bfd5d4df4f4c9656a009f09dfc7e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bence=20Cs=C3=B3k=C3=A1s?= Date: Mon, 3 Feb 2025 16:12:49 +0100 Subject: [PATCH 0389/1443] spi: atmel-quadspi: Fix warning in doc-comment MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The doc-comment for `struct atmel_qspi_pcal` had a typo in one of the struct members' name, causing a warning with the `W=1` option. Fixes: 5af42209a4d2 ("spi: atmel-quadspi: Add support for sama7g5 QSPI") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501311707.Ltj0qXse-lkp@intel.com/ Signed-off-by: Bence Csókás Link: https://patch.msgid.link/20250203151249.79876-2-csokas.bence@prolan.hu Signed-off-by: Mark Brown --- drivers/spi/atmel-quadspi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c index abdc49d9d940..d8c9be64d006 100644 --- a/drivers/spi/atmel-quadspi.c +++ b/drivers/spi/atmel-quadspi.c @@ -235,8 +235,8 @@ /** * struct atmel_qspi_pcal - Pad Calibration Clock Division * @pclk_rate: peripheral clock rate. - * @pclkdiv: calibration clock division. The clock applied to the calibration - * cell is divided by pclkdiv + 1. + * @pclk_div: calibration clock division. The clock applied to the calibration + * cell is divided by pclk_div + 1. */ struct atmel_qspi_pcal { u32 pclk_rate; From 9e8b21410f310c50733f6e1730bae5a8e30d3570 Mon Sep 17 00:00:00 2001 From: Selvarasu Ganesan Date: Sat, 18 Jan 2025 11:31:33 +0530 Subject: [PATCH 0390/1443] usb: gadget: f_midi: Fixing wMaxPacketSize exceeded issue during MIDI bind retries The current implementation sets the wMaxPacketSize of bulk in/out endpoints to 1024 bytes at the end of the f_midi_bind function. However, in cases where there is a failure in the first midi bind attempt, consider rebinding. This scenario may encounter an f_midi_bind issue due to the previous bind setting the bulk endpoint's wMaxPacketSize to 1024 bytes, which exceeds the ep->maxpacket_limit where configured dwc3 TX/RX FIFO's maxpacket size of 512 bytes for IN/OUT endpoints in support HS speed only. Here the term "rebind" in this context refers to attempting to bind the MIDI function a second time in certain scenarios. The situations where rebinding is considered include: * When there is a failure in the first UDC write attempt, which may be caused by other functions bind along with MIDI. * Runtime composition change : Example : MIDI,ADB to MIDI. Or MIDI to MIDI,ADB. This commit addresses this issue by resetting the wMaxPacketSize before endpoint claim. And here there is no need to reset all values in the usb endpoint descriptor structure, as all members except wMaxPacketSize and bEndpointAddress have predefined values. This ensures that restores the endpoint to its expected configuration, and preventing conflicts with value of ep->maxpacket_limit. It also aligns with the approach used in other function drivers, which treat endpoint descriptors as if they were full speed before endpoint claim. Fixes: 46decc82ffd5 ("usb: gadget: unconditionally allocate hs/ss descriptor in bind operation") Cc: stable@vger.kernel.org Signed-off-by: Selvarasu Ganesan Link: https://lore.kernel.org/r/20250118060134.927-1-selvarasu.g@samsung.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_midi.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/usb/gadget/function/f_midi.c b/drivers/usb/gadget/function/f_midi.c index 837fcdfa3840..9b991cf5b0f8 100644 --- a/drivers/usb/gadget/function/f_midi.c +++ b/drivers/usb/gadget/function/f_midi.c @@ -907,6 +907,15 @@ static int f_midi_bind(struct usb_configuration *c, struct usb_function *f) status = -ENODEV; + /* + * Reset wMaxPacketSize with maximum packet size of FS bulk transfer before + * endpoint claim. This ensures that the wMaxPacketSize does not exceed the + * limit during bind retries where configured dwc3 TX/RX FIFO's maxpacket + * size of 512 bytes for IN/OUT endpoints in support HS speed only. + */ + bulk_in_desc.wMaxPacketSize = cpu_to_le16(64); + bulk_out_desc.wMaxPacketSize = cpu_to_le16(64); + /* allocate instance-specific endpoints */ midi->in_ep = usb_ep_autoconfig(cdev->gadget, &bulk_in_desc); if (!midi->in_ep) From 309005e448c1f3e4b81e4416406991b7c3339c1d Mon Sep 17 00:00:00 2001 From: Alexander Stein Date: Mon, 20 Jan 2025 15:42:51 +0100 Subject: [PATCH 0391/1443] usb: phy: generic: Use proper helper for property detection Since commit c141ecc3cecd7 ("of: Warn when of_property_read_bool() is used on non-boolean properties") a warning is raised if this function is used for property detection. of_property_present() is the correct helper for this. Signed-off-by: Alexander Stein Link: https://lore.kernel.org/r/20250120144251.580981-1-alexander.stein@ew.tq-group.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/phy/phy-generic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/phy/phy-generic.c b/drivers/usb/phy/phy-generic.c index 6c3ececf9137..8423be59ec0f 100644 --- a/drivers/usb/phy/phy-generic.c +++ b/drivers/usb/phy/phy-generic.c @@ -212,7 +212,7 @@ int usb_phy_gen_create_phy(struct device *dev, struct usb_phy_generic *nop) if (of_property_read_u32(node, "clock-frequency", &clk_rate)) clk_rate = 0; - needs_clk = of_property_read_bool(node, "clocks"); + needs_clk = of_property_present(node, "clocks"); } nop->gpiod_reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); From 1ed3af5a2aaefd0ecd887ecabdc8da07220e31fe Mon Sep 17 00:00:00 2001 From: Thinh Nguyen Date: Tue, 21 Jan 2025 23:11:23 +0000 Subject: [PATCH 0392/1443] usb: dwc3: Document nostream_work Add missing description to the nostream_work of dwc3_ep. The work is used by bulk multi-stream endpoints for a NoStream event to reinitiate the stream if needed. Reported-by: Stephen Rothwell Closes: https://lore.kernel.org/linux-next/20250120182219.30dcb3c6@canb.auug.org.au/ Signed-off-by: Thinh Nguyen Link: https://lore.kernel.org/r/7cdeaa346d24907712aac533c1c5f90a03151189.1737500936.git.Thinh.Nguyen@synopsys.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc3/core.h | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index ac7c730f81ac..c955039bb4f6 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -717,6 +717,7 @@ struct dwc3_event_buffer { /** * struct dwc3_ep - device side endpoint representation * @endpoint: usb endpoint + * @nostream_work: work for handling bulk NoStream * @cancelled_list: list of cancelled requests for this endpoint * @pending_list: list of pending requests for this endpoint * @started_list: list of started requests on this endpoint From 335a1fc1193481f8027f176649c72868172f6f8b Mon Sep 17 00:00:00 2001 From: Guo Ren Date: Wed, 22 Jan 2025 03:12:31 -0500 Subject: [PATCH 0393/1443] usb: gadget: udc: renesas_usb3: Fix compiler warning drivers/usb/gadget/udc/renesas_usb3.c: In function 'renesas_usb3_probe': drivers/usb/gadget/udc/renesas_usb3.c:2638:73: warning: '%d' directive output may be truncated writing between 1 and 11 bytes into a region of size 6 [-Wformat-truncation=] 2638 | snprintf(usb3_ep->ep_name, sizeof(usb3_ep->ep_name), "ep%d", i); ^~~~~~~~~~~~~~~~~~~~~~~~ ^~ ^ Fixes: 746bfe63bba3 ("usb: gadget: renesas_usb3: add support for Renesas USB3.0 peripheral controller") Cc: stable@vger.kernel.org Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501201409.BIQPtkeB-lkp@intel.com/ Signed-off-by: Guo Ren Link: https://lore.kernel.org/r/20250122081231.47594-1-guoren@kernel.org Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc/renesas_usb3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/udc/renesas_usb3.c b/drivers/usb/gadget/udc/renesas_usb3.c index fce5c41d9f29..89b304cf6d03 100644 --- a/drivers/usb/gadget/udc/renesas_usb3.c +++ b/drivers/usb/gadget/udc/renesas_usb3.c @@ -310,7 +310,7 @@ struct renesas_usb3_request { struct list_head queue; }; -#define USB3_EP_NAME_SIZE 8 +#define USB3_EP_NAME_SIZE 16 struct renesas_usb3_ep { struct usb_ep ep; struct renesas_usb3 *usb3; From 2240fed37afbcdb5e8b627bc7ad986891100e05d Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 22 Jan 2025 14:26:17 -0500 Subject: [PATCH 0394/1443] USB: hub: Ignore non-compliant devices with too many configs or interfaces Robert Morris created a test program which can cause usb_hub_to_struct_hub() to dereference a NULL or inappropriate pointer: Oops: general protection fault, probably for non-canonical address 0xcccccccccccccccc: 0000 [#1] SMP DEBUG_PAGEALLOC PTI CPU: 7 UID: 0 PID: 117 Comm: kworker/7:1 Not tainted 6.13.0-rc3-00017-gf44d154d6e3d #14 Hardware name: FreeBSD BHYVE/BHYVE, BIOS 14.0 10/17/2021 Workqueue: usb_hub_wq hub_event RIP: 0010:usb_hub_adjust_deviceremovable+0x78/0x110 ... Call Trace: ? die_addr+0x31/0x80 ? exc_general_protection+0x1b4/0x3c0 ? asm_exc_general_protection+0x26/0x30 ? usb_hub_adjust_deviceremovable+0x78/0x110 hub_probe+0x7c7/0xab0 usb_probe_interface+0x14b/0x350 really_probe+0xd0/0x2d0 ? __pfx___device_attach_driver+0x10/0x10 __driver_probe_device+0x6e/0x110 driver_probe_device+0x1a/0x90 __device_attach_driver+0x7e/0xc0 bus_for_each_drv+0x7f/0xd0 __device_attach+0xaa/0x1a0 bus_probe_device+0x8b/0xa0 device_add+0x62e/0x810 usb_set_configuration+0x65d/0x990 usb_generic_driver_probe+0x4b/0x70 usb_probe_device+0x36/0xd0 The cause of this error is that the device has two interfaces, and the hub driver binds to interface 1 instead of interface 0, which is where usb_hub_to_struct_hub() looks. We can prevent the problem from occurring by refusing to accept hub devices that violate the USB spec by having more than one configuration or interface. Reported-and-tested-by: Robert Morris Cc: stable Closes: https://lore.kernel.org/linux-usb/95564.1737394039@localhost/ Signed-off-by: Alan Stern Link: https://lore.kernel.org/r/c27f3bf4-63d8-4fb5-ac82-09e3cd19f61c@rowland.harvard.edu Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hub.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index c3f839637cb5..0cd44f1fd56d 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1848,6 +1848,17 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id) desc = intf->cur_altsetting; hdev = interface_to_usbdev(intf); + /* + * The USB 2.0 spec prohibits hubs from having more than one + * configuration or interface, and we rely on this prohibition. + * Refuse to accept a device that violates it. + */ + if (hdev->descriptor.bNumConfigurations > 1 || + hdev->actconfig->desc.bNumInterfaces > 1) { + dev_err(&intf->dev, "Invalid hub with more than one config or interface\n"); + return -EINVAL; + } + /* * Set default autosuspend delay as 0 to speedup bus suspend, * based on the below considerations: From 58cd423820d5b5610977e55e4acdd06628829ede Mon Sep 17 00:00:00 2001 From: Fabrice Gasnier Date: Fri, 24 Jan 2025 18:33:25 +0100 Subject: [PATCH 0395/1443] usb: dwc2: gadget: remove of_node reference upon udc_stop In dwc2_hsotg_udc_start(), e.g. when binding composite driver, "of_node" is set to hsotg->dev->of_node. It causes errors when binding the gadget driver several times, on stm32mp157c-ev1 board. Below error is seen: "pin PA10 already requested by 49000000.usb-otg; cannot claim for gadget.0" The first time, no issue is seen as when registering the driver, of_node isn't NULL: -> gadget_dev_desc_UDC_store -> usb_gadget_register_driver_owner -> driver_register ... -> really_probe -> pinctrl_bind_pins (no effect) Then dwc2_hsotg_udc_start() sets of_node. The second time (stop the gadget, reconfigure it, then start it again), of_node has been set, so the probing code tries to acquire pins for the gadget. These pins are hold by the controller, hence the error. So clear gadget.dev.of_node in udc_stop() routine to avoid the issue. Fixes: 7d7b22928b90 ("usb: gadget: s3c-hsotg: Propagate devicetree to gadget drivers") Cc: stable Signed-off-by: Fabrice Gasnier Link: https://lore.kernel.org/r/20250124173325.2747710-1-fabrice.gasnier@foss.st.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc2/gadget.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/dwc2/gadget.c b/drivers/usb/dwc2/gadget.c index e7bf9cc635be..bd4c788f03bc 100644 --- a/drivers/usb/dwc2/gadget.c +++ b/drivers/usb/dwc2/gadget.c @@ -4615,6 +4615,7 @@ static int dwc2_hsotg_udc_stop(struct usb_gadget *gadget) spin_lock_irqsave(&hsotg->lock, flags); hsotg->driver = NULL; + hsotg->gadget.dev.of_node = NULL; hsotg->gadget.speed = USB_SPEED_UNKNOWN; hsotg->enabled = 0; From da1668997052ed1cb00322e1f3b63702615c9429 Mon Sep 17 00:00:00 2001 From: John Keeping Date: Thu, 30 Jan 2025 19:50:34 +0000 Subject: [PATCH 0396/1443] usb: gadget: f_midi: fix MIDI Streaming descriptor lengths While the MIDI jacks are configured correctly, and the MIDIStreaming endpoint descriptors are filled with the correct information, bNumEmbMIDIJack and bLength are set incorrectly in these descriptors. This does not matter when the numbers of in and out ports are equal, but when they differ the host will receive broken descriptors with uninitialized stack memory leaking into the descriptor for whichever value is smaller. The precise meaning of "in" and "out" in the port counts is not clearly defined and can be confusing. But elsewhere the driver consistently uses this to match the USB meaning of IN and OUT viewed from the host, so that "in" ports send data to the host and "out" ports receive data from it. Cc: stable Fixes: c8933c3f79568 ("USB: gadget: f_midi: allow a dynamic number of input and output ports") Signed-off-by: John Keeping Reviewed-by: Takashi Iwai Link: https://lore.kernel.org/r/20250130195035.3883857-1-jkeeping@inmusicbrands.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_midi.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/usb/gadget/function/f_midi.c b/drivers/usb/gadget/function/f_midi.c index 9b991cf5b0f8..47260d65066a 100644 --- a/drivers/usb/gadget/function/f_midi.c +++ b/drivers/usb/gadget/function/f_midi.c @@ -1009,11 +1009,11 @@ static int f_midi_bind(struct usb_configuration *c, struct usb_function *f) } /* configure the endpoint descriptors ... */ - ms_out_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->in_ports); - ms_out_desc.bNumEmbMIDIJack = midi->in_ports; + ms_out_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->out_ports); + ms_out_desc.bNumEmbMIDIJack = midi->out_ports; - ms_in_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->out_ports); - ms_in_desc.bNumEmbMIDIJack = midi->out_ports; + ms_in_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->in_ports); + ms_in_desc.bNumEmbMIDIJack = midi->in_ports; /* ... and add them to the list */ endpoint_descriptor_index = i; From ae5d9cde9b762fd4b7259e1f93a94e0c7f04681c Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Fri, 31 Jan 2025 14:39:08 -0800 Subject: [PATCH 0397/1443] drm/xe: Remove xe_dummy_exit() Since commit 014125c64d09 ("drm/xe: Support 'nomodeset' kernel command-line option") the dummy exit is not needed anymore since the caller check for a NULL pointer. Drop it. Reviewed-by: Raag Jadav Link: https://patchwork.freedesktop.org/patch/msgid/20250131223908.4147195-1-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_module.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c index 07b27114be9a..7185a2cdf6e3 100644 --- a/drivers/gpu/drm/xe/xe_module.c +++ b/drivers/gpu/drm/xe/xe_module.c @@ -77,10 +77,6 @@ struct init_funcs { void (*exit)(void); }; -static void xe_dummy_exit(void) -{ -} - static const struct init_funcs init_funcs[] = { { .init = xe_check_nomodeset, @@ -103,7 +99,6 @@ static const struct init_funcs init_funcs[] = { }, { .init = xe_pm_module_init, - .exit = xe_dummy_exit, }, }; From 2255b40cacc2e5ef1b127770fc1808c60de4a2fc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Marek=20Ol=C5=A1=C3=A1k?= Date: Fri, 24 Jan 2025 09:43:45 -0500 Subject: [PATCH 0398/1443] drm/amdgpu: add a BO metadata flag to disable write compression for Vulkan MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Vulkan can't support DCC and Z/S compression on GFX12 without WRITE_COMPRESS_DISABLE in this commit or a completely different DCC interface. AMDGPU_TILING_GFX12_SCANOUT is added because it's already used by userspace. Cc: stable@vger.kernel.org # 6.12.x Signed-off-by: Marek Olšák Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 8 ++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 ++ drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 5 +++-- include/uapi/drm/amdgpu_drm.h | 9 ++++++++- 5 files changed, 21 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 817116e53d44..dce9323fb410 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -119,9 +119,10 @@ * - 3.57.0 - Compute tunneling on GFX10+ * - 3.58.0 - Add GFX12 DCC support * - 3.59.0 - Cleared VRAM + * - 3.60.0 - Add AMDGPU_TILING_GFX12_DCC_WRITE_COMPRESS_DISABLE (Vulkan requirement) */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 59 +#define KMS_DRIVER_MINOR 60 #define KMS_DRIVER_PATCHLEVEL 0 /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index ff286940ab43..01ae2f88dec8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -309,7 +309,7 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, mutex_lock(&adev->mman.gtt_window_lock); while (src_mm.remaining) { uint64_t from, to, cur_size, tiling_flags; - uint32_t num_type, data_format, max_com; + uint32_t num_type, data_format, max_com, write_compress_disable; struct dma_fence *next; /* Never copy more than 256MiB at once to avoid a timeout */ @@ -340,9 +340,13 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, max_com = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_MAX_COMPRESSED_BLOCK); num_type = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_NUMBER_TYPE); data_format = AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_DATA_FORMAT); + write_compress_disable = + AMDGPU_TILING_GET(tiling_flags, GFX12_DCC_WRITE_COMPRESS_DISABLE); copy_flags |= (AMDGPU_COPY_FLAGS_SET(MAX_COMPRESSED, max_com) | AMDGPU_COPY_FLAGS_SET(NUMBER_TYPE, num_type) | - AMDGPU_COPY_FLAGS_SET(DATA_FORMAT, data_format)); + AMDGPU_COPY_FLAGS_SET(DATA_FORMAT, data_format) | + AMDGPU_COPY_FLAGS_SET(WRITE_COMPRESS_DISABLE, + write_compress_disable)); } r = amdgpu_copy_buffer(ring, from, to, cur_size, resv, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 461fb8090ae0..208b7d1d8a27 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -119,6 +119,8 @@ struct amdgpu_copy_mem { #define AMDGPU_COPY_FLAGS_NUMBER_TYPE_MASK 0x07 #define AMDGPU_COPY_FLAGS_DATA_FORMAT_SHIFT 8 #define AMDGPU_COPY_FLAGS_DATA_FORMAT_MASK 0x3f +#define AMDGPU_COPY_FLAGS_WRITE_COMPRESS_DISABLE_SHIFT 14 +#define AMDGPU_COPY_FLAGS_WRITE_COMPRESS_DISABLE_MASK 0x1 #define AMDGPU_COPY_FLAGS_SET(field, value) \ (((__u32)(value) & AMDGPU_COPY_FLAGS_##field##_MASK) << AMDGPU_COPY_FLAGS_##field##_SHIFT) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c index 9c17df2cf37b..7e10e94624e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c @@ -1741,11 +1741,12 @@ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib, uint32_t byte_count, uint32_t copy_flags) { - uint32_t num_type, data_format, max_com; + uint32_t num_type, data_format, max_com, write_cm; max_com = AMDGPU_COPY_FLAGS_GET(copy_flags, MAX_COMPRESSED); data_format = AMDGPU_COPY_FLAGS_GET(copy_flags, DATA_FORMAT); num_type = AMDGPU_COPY_FLAGS_GET(copy_flags, NUMBER_TYPE); + write_cm = AMDGPU_COPY_FLAGS_GET(copy_flags, WRITE_COMPRESS_DISABLE) ? 2 : 1; ib->ptr[ib->length_dw++] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_COPY) | SDMA_PKT_COPY_LINEAR_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR) | @@ -1762,7 +1763,7 @@ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib, if ((copy_flags & (AMDGPU_COPY_FLAGS_READ_DECOMPRESSED | AMDGPU_COPY_FLAGS_WRITE_COMPRESSED))) ib->ptr[ib->length_dw++] = SDMA_DCC_DATA_FORMAT(data_format) | SDMA_DCC_NUM_TYPE(num_type) | ((copy_flags & AMDGPU_COPY_FLAGS_READ_DECOMPRESSED) ? SDMA_DCC_READ_CM(2) : 0) | - ((copy_flags & AMDGPU_COPY_FLAGS_WRITE_COMPRESSED) ? SDMA_DCC_WRITE_CM(1) : 0) | + ((copy_flags & AMDGPU_COPY_FLAGS_WRITE_COMPRESSED) ? SDMA_DCC_WRITE_CM(write_cm) : 0) | SDMA_DCC_MAX_COM(max_com) | SDMA_DCC_MAX_UCOM(1); else ib->ptr[ib->length_dw++] = 0; diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index efe5de6ce208..aaa4f3bc688b 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -411,13 +411,20 @@ struct drm_amdgpu_gem_userptr { /* GFX12 and later: */ #define AMDGPU_TILING_GFX12_SWIZZLE_MODE_SHIFT 0 #define AMDGPU_TILING_GFX12_SWIZZLE_MODE_MASK 0x7 -/* These are DCC recompression setting for memory management: */ +/* These are DCC recompression settings for memory management: */ #define AMDGPU_TILING_GFX12_DCC_MAX_COMPRESSED_BLOCK_SHIFT 3 #define AMDGPU_TILING_GFX12_DCC_MAX_COMPRESSED_BLOCK_MASK 0x3 /* 0:64B, 1:128B, 2:256B */ #define AMDGPU_TILING_GFX12_DCC_NUMBER_TYPE_SHIFT 5 #define AMDGPU_TILING_GFX12_DCC_NUMBER_TYPE_MASK 0x7 /* CB_COLOR0_INFO.NUMBER_TYPE */ #define AMDGPU_TILING_GFX12_DCC_DATA_FORMAT_SHIFT 8 #define AMDGPU_TILING_GFX12_DCC_DATA_FORMAT_MASK 0x3f /* [0:4]:CB_COLOR0_INFO.FORMAT, [5]:MM */ +/* When clearing the buffer or moving it from VRAM to GTT, don't compress and set DCC metadata + * to uncompressed. Set when parts of an allocation bypass DCC and read raw data. */ +#define AMDGPU_TILING_GFX12_DCC_WRITE_COMPRESS_DISABLE_SHIFT 14 +#define AMDGPU_TILING_GFX12_DCC_WRITE_COMPRESS_DISABLE_MASK 0x1 +/* bit gap */ +#define AMDGPU_TILING_GFX12_SCANOUT_SHIFT 63 +#define AMDGPU_TILING_GFX12_SCANOUT_MASK 0x1 /* Set/Get helpers for tiling flags. */ #define AMDGPU_TILING_SET(field, value) \ From 8adbb2a98b00926315fd513b5fe2596b5716b82d Mon Sep 17 00:00:00 2001 From: Alex Hung Date: Fri, 17 Jan 2025 12:37:11 -0700 Subject: [PATCH 0399/1443] drm/amd/display: Fix out-of-bound accesses [WHAT & HOW] hpo_stream_to_link_encoder_mapping has size MAX_HPO_DP2_ENCODERS(=4), but location can have size up to 6. As a result, it is necessary to check location against MAX_HPO_DP2_ENCODERS. Similiarly, disp_cfg_stream_location can be used as an array index which should be 0..5, so the ASSERT's conditions should be less without equal. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3904 Reviewed-by: Austin Zheng Reviewed-by: Rodrigo Siqueira Signed-off-by: Alex Hung Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- .../amd/display/dc/dml2/dml21/dml21_translation_helper.c | 4 ++-- .../gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index b9c6b45f6872..0c8ec30ea672 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -1017,7 +1017,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s if (disp_cfg_stream_location < 0) disp_cfg_stream_location = dml_dispcfg->num_streams++; - ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location <= __DML2_WRAPPER_MAX_STREAMS_PLANES__); + ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location < __DML2_WRAPPER_MAX_STREAMS_PLANES__); populate_dml21_timing_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, context->streams[stream_index], dml_ctx); adjust_dml21_hblank_timing_config_from_pipe_ctx(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, &context->res_ctx.pipe_ctx[stream_index]); populate_dml21_output_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].output, context->streams[stream_index], &context->res_ctx.pipe_ctx[stream_index]); @@ -1042,7 +1042,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s if (disp_cfg_plane_location < 0) disp_cfg_plane_location = dml_dispcfg->num_planes++; - ASSERT(disp_cfg_plane_location >= 0 && disp_cfg_plane_location <= __DML2_WRAPPER_MAX_STREAMS_PLANES__); + ASSERT(disp_cfg_plane_location >= 0 && disp_cfg_plane_location < __DML2_WRAPPER_MAX_STREAMS_PLANES__); populate_dml21_surface_config_from_plane_state(in_dc, &dml_dispcfg->plane_descriptors[disp_cfg_plane_location].surface, context->stream_status[stream_index].plane_states[plane_index]); populate_dml21_plane_config_from_plane_state(dml_ctx, &dml_dispcfg->plane_descriptors[disp_cfg_plane_location], context->stream_status[stream_index].plane_states[plane_index], context, stream_index); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index b416320873e1..b8a34abaf519 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -786,7 +786,7 @@ static void populate_dml_output_cfg_from_stream_state(struct dml_output_cfg_st * case SIGNAL_TYPE_DISPLAY_PORT_MST: case SIGNAL_TYPE_DISPLAY_PORT: out->OutputEncoder[location] = dml_dp; - if (dml2->v20.scratch.hpo_stream_to_link_encoder_mapping[location] != -1) + if (location < MAX_HPO_DP2_ENCODERS && dml2->v20.scratch.hpo_stream_to_link_encoder_mapping[location] != -1) out->OutputEncoder[dml2->v20.scratch.hpo_stream_to_link_encoder_mapping[location]] = dml_dp2p0; break; case SIGNAL_TYPE_EDP: @@ -1343,7 +1343,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat if (disp_cfg_stream_location < 0) disp_cfg_stream_location = dml_dispcfg->num_timings++; - ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location <= __DML2_WRAPPER_MAX_STREAMS_PLANES__); + ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location < __DML2_WRAPPER_MAX_STREAMS_PLANES__); populate_dml_timing_cfg_from_stream_state(&dml_dispcfg->timing, disp_cfg_stream_location, context->streams[i]); populate_dml_output_cfg_from_stream_state(&dml_dispcfg->output, disp_cfg_stream_location, context->streams[i], current_pipe_context, dml2); @@ -1383,7 +1383,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat if (disp_cfg_plane_location < 0) disp_cfg_plane_location = dml_dispcfg->num_surfaces++; - ASSERT(disp_cfg_plane_location >= 0 && disp_cfg_plane_location <= __DML2_WRAPPER_MAX_STREAMS_PLANES__); + ASSERT(disp_cfg_plane_location >= 0 && disp_cfg_plane_location < __DML2_WRAPPER_MAX_STREAMS_PLANES__); populate_dml_surface_cfg_from_plane_state(dml2->v20.dml_core_ctx.project, &dml_dispcfg->surface, disp_cfg_plane_location, context->stream_status[i].plane_states[j]); populate_dml_plane_cfg_from_plane_state( From e01f07cb92513ca4b9b219ab9caa34d607bc1e2d Mon Sep 17 00:00:00 2001 From: Lo-an Chen Date: Fri, 17 Jan 2025 17:56:25 +0800 Subject: [PATCH 0400/1443] drm/amd/display: Fix seamless boot sequence [WHY] When the system powers up eDP with external monitors in seamless boot sequence, stutter get enabled before TTU and HUBP registers being programmed, which resulting in underflow. [HOW] Enable TTU in hubp_init. Change the sequence that do not perpare_bandwidth and optimize_bandwidth while having seamless boot streams. Cc: Mario Limonciello Cc: Alex Deucher Cc: stable@vger.kernel.org Reviewed-by: Nicholas Kazlauskas Signed-off-by: Lo-an Chen Signed-off-by: Paul Hsieh Signed-off-by: Alex Hung Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +- drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c | 3 ++- drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c | 3 ++- drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c | 3 ++- drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c | 3 ++- drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c | 2 ++ drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c | 2 ++ drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c | 3 ++- 8 files changed, 15 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index cecaadf741ad..f84e795e35f5 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -2133,7 +2133,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c dc_enable_stereo(dc, context, dc_streams, context->stream_count); - if (context->stream_count > get_seamless_boot_stream_count(context) || + if (get_seamless_boot_stream_count(context) == 0 || context->stream_count == 0) { /* Must wait for no flips to be pending before doing optimize bw */ hwss_wait_for_no_pipes_pending(dc, context); diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c index fe741100c0f8..d347bb06577a 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c @@ -129,7 +129,8 @@ bool hubbub3_program_watermarks( REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF); - hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter) + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); return wm_pending; } diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c index 7fb5523f9722..b98505b240a7 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c @@ -750,7 +750,8 @@ static bool hubbub31_program_watermarks( REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF);*/ - hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter) + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); return wm_pending; } diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c index 5264dc26cce1..32a6be543105 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c @@ -786,7 +786,8 @@ static bool hubbub32_program_watermarks( REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND, DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF);*/ - hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter) + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); hubbub32_force_usr_retraining_allow(hubbub, hubbub->ctx->dc->debug.force_usr_allow); diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c index 5eb3da8d5206..dce7269959ce 100644 --- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c @@ -326,7 +326,8 @@ static bool hubbub35_program_watermarks( DCHUBBUB_ARB_MIN_REQ_OUTSTAND_COMMIT_THRESHOLD, 0xA);/*hw delta*/ REG_UPDATE(DCHUBBUB_ARB_HOSTVM_CNTL, DCHUBBUB_ARB_MAX_QOS_COMMIT_THRESHOLD, 0xF); - hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); + if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter) + hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter); hubbub32_force_usr_retraining_allow(hubbub, hubbub->ctx->dc->debug.force_usr_allow); diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c index be0ac613675a..0da70b50e86d 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c @@ -500,6 +500,8 @@ void hubp3_init(struct hubp *hubp) //hubp[i].HUBPREQ_DEBUG.HUBPREQ_DEBUG[26] = 1; REG_WRITE(HUBPREQ_DEBUG, 1 << 26); + REG_UPDATE(DCHUBP_CNTL, HUBP_TTU_DISABLE, 0); + hubp_reset(hubp); } diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c index edd37898d550..f3a21c623f44 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c @@ -168,6 +168,8 @@ void hubp32_init(struct hubp *hubp) { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); REG_WRITE(HUBPREQ_DEBUG_DB, 1 << 8); + + REG_UPDATE(DCHUBP_CNTL, HUBP_TTU_DISABLE, 0); } static struct hubp_funcs dcn32_hubp_funcs = { .hubp_enable_tripleBuffer = hubp2_enable_triplebuffer, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index 623cde76debf..b907ad1acedd 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -236,7 +236,8 @@ void dcn35_init_hw(struct dc *dc) } hws->funcs.init_pipes(dc, dc->current_state); - if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) + if (dc->res_pool->hubbub->funcs->allow_self_refresh_control && + !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter) dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, !dc->res_pool->hubbub->ctx->dc->debug.disable_stutter); } From 588c20079e17dae9e1f49ba42981a05de1c9136e Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Thu, 16 Jan 2025 19:21:55 -0800 Subject: [PATCH 0401/1443] drm/xe/oa: Preserve oa_ctrl unused bits UMD's have interest in setting unused bits of the oa_ctrl register "out of band" for certain experiments. To facilitate this, don't clobber previous oa_ctrl unused bits, i.e. rmw the values rather than simply write them. Fixes: e936f885f1e9 ("drm/xe/oa/uapi: Expose OA stream fd") Signed-off-by: Ashutosh Dixit Reviewed-by: Umesh Nerlige Ramappa Link: https://patchwork.freedesktop.org/patch/msgid/20250117032155.3048063-1-ashutosh.dixit@intel.com (cherry picked from commit cfa9d40db8c30d894171010fe765d96e9bc6a47e) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/regs/xe_oa_regs.h | 6 ++++++ drivers/gpu/drm/xe/xe_oa.c | 12 ++++++++++-- 2 files changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_oa_regs.h b/drivers/gpu/drm/xe/regs/xe_oa_regs.h index a49561e9f3c3..a79ad2da070c 100644 --- a/drivers/gpu/drm/xe/regs/xe_oa_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_oa_regs.h @@ -51,6 +51,10 @@ /* Common to all OA units */ #define OA_OACONTROL_REPORT_BC_MASK REG_GENMASK(9, 9) #define OA_OACONTROL_COUNTER_SIZE_MASK REG_GENMASK(8, 8) +#define OAG_OACONTROL_USED_BITS \ + (OAG_OACONTROL_OA_PES_DISAG_EN | OAG_OACONTROL_OA_CCS_SELECT_MASK | \ + OAG_OACONTROL_OA_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE | \ + OA_OACONTROL_REPORT_BC_MASK | OA_OACONTROL_COUNTER_SIZE_MASK) #define OAG_OA_DEBUG XE_REG(0xdaf8, XE_REG_OPTION_MASKED) #define OAG_OA_DEBUG_DISABLE_MMIO_TRG REG_BIT(14) @@ -78,6 +82,8 @@ #define OAM_CONTEXT_CONTROL_OFFSET (0x1bc) #define OAM_CONTROL_OFFSET (0x194) #define OAM_CONTROL_COUNTER_SEL_MASK REG_GENMASK(3, 1) +#define OAM_OACONTROL_USED_BITS \ + (OAM_CONTROL_COUNTER_SEL_MASK | OAG_OACONTROL_OA_COUNTER_ENABLE) #define OAM_DEBUG_OFFSET (0x198) #define OAM_STATUS_OFFSET (0x19c) #define OAM_MMIO_TRG_OFFSET (0x1d0) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index eeb96b5f49e2..6a08e6c92835 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -452,6 +452,12 @@ static u32 __oa_ccs_select(struct xe_oa_stream *stream) return val; } +static u32 __oactrl_used_bits(struct xe_oa_stream *stream) +{ + return stream->hwe->oa_unit->type == DRM_XE_OA_UNIT_TYPE_OAG ? + OAG_OACONTROL_USED_BITS : OAM_OACONTROL_USED_BITS; +} + static void xe_oa_enable(struct xe_oa_stream *stream) { const struct xe_oa_format *format = stream->oa_buffer.format; @@ -472,14 +478,14 @@ static void xe_oa_enable(struct xe_oa_stream *stream) stream->hwe->oa_unit->type == DRM_XE_OA_UNIT_TYPE_OAG) val |= OAG_OACONTROL_OA_PES_DISAG_EN; - xe_mmio_write32(&stream->gt->mmio, regs->oa_ctrl, val); + xe_mmio_rmw32(&stream->gt->mmio, regs->oa_ctrl, __oactrl_used_bits(stream), val); } static void xe_oa_disable(struct xe_oa_stream *stream) { struct xe_mmio *mmio = &stream->gt->mmio; - xe_mmio_write32(mmio, __oa_regs(stream)->oa_ctrl, 0); + xe_mmio_rmw32(mmio, __oa_regs(stream)->oa_ctrl, __oactrl_used_bits(stream), 0); if (xe_mmio_wait32(mmio, __oa_regs(stream)->oa_ctrl, OAG_OACONTROL_OA_COUNTER_ENABLE, 0, 50000, NULL, false)) drm_err(&stream->oa->xe->drm, @@ -2534,6 +2540,8 @@ static void __xe_oa_init_oa_units(struct xe_gt *gt) u->type = DRM_XE_OA_UNIT_TYPE_OAM; } + xe_mmio_write32(>->mmio, u->regs.oa_ctrl, 0); + /* Ensure MMIO trigger remains disabled till there is a stream */ xe_mmio_write32(>->mmio, u->regs.oa_debug, oag_configure_mmio_trigger(NULL, false)); From 9f706fd8024208b0686bb8ec68589d758f765672 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Tue, 21 Jan 2025 00:24:43 +0100 Subject: [PATCH 0402/1443] drm/xe/pf: Fix migration initialization MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The migration support only needs to be initialized once, but it was incorrectly called from the xe_gt_sriov_pf_init_hw(), which is part of the reset flow and may be called multiple times. Fixes: d86e3737c7ab ("drm/xe/pf: Add functions to save and restore VF GuC state") Signed-off-by: Michal Wajdeczko Cc: Michał Winiarski Reviewed-by: Rodrigo Vivi Link: https://patchwork.freedesktop.org/patch/msgid/20250120232443.544-1-michal.wajdeczko@intel.com (cherry picked from commit 9ebb5846e1a3b1705f8a7cbc528888a1aa0b163e) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_gt.c | 4 +++- drivers/gpu/drm/xe/xe_gt_sriov_pf.c | 14 +++++++++++++- drivers/gpu/drm/xe/xe_gt_sriov_pf.h | 6 ++++++ 3 files changed, 22 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 26e64530ada2..5d6fb79957b6 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -532,8 +532,10 @@ static int all_fw_domain_init(struct xe_gt *gt) if (IS_SRIOV_PF(gt_to_xe(gt)) && !xe_gt_is_media_type(gt)) xe_lmtt_init_hw(>_to_tile(gt)->sriov.pf.lmtt); - if (IS_SRIOV_PF(gt_to_xe(gt))) + if (IS_SRIOV_PF(gt_to_xe(gt))) { + xe_gt_sriov_pf_init(gt); xe_gt_sriov_pf_init_hw(gt); + } xe_force_wake_put(gt_to_fw(gt), fw_ref); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c index e71fc3d2bda2..6f906c8e8108 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c @@ -68,6 +68,19 @@ int xe_gt_sriov_pf_init_early(struct xe_gt *gt) return 0; } +/** + * xe_gt_sriov_pf_init - Prepare SR-IOV PF data structures on PF. + * @gt: the &xe_gt to initialize + * + * Late one-time initialization of the PF data. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_gt_sriov_pf_init(struct xe_gt *gt) +{ + return xe_gt_sriov_pf_migration_init(gt); +} + static bool pf_needs_enable_ggtt_guest_update(struct xe_device *xe) { return GRAPHICS_VERx100(xe) == 1200; @@ -90,7 +103,6 @@ void xe_gt_sriov_pf_init_hw(struct xe_gt *gt) pf_enable_ggtt_guest_update(gt); xe_gt_sriov_pf_service_update(gt); - xe_gt_sriov_pf_migration_init(gt); } static u32 pf_get_vf_regs_stride(struct xe_device *xe) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h index 96fab779a906..f474509411c0 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h @@ -10,6 +10,7 @@ struct xe_gt; #ifdef CONFIG_PCI_IOV int xe_gt_sriov_pf_init_early(struct xe_gt *gt); +int xe_gt_sriov_pf_init(struct xe_gt *gt); void xe_gt_sriov_pf_init_hw(struct xe_gt *gt); void xe_gt_sriov_pf_sanitize_hw(struct xe_gt *gt, unsigned int vfid); void xe_gt_sriov_pf_restart(struct xe_gt *gt); @@ -19,6 +20,11 @@ static inline int xe_gt_sriov_pf_init_early(struct xe_gt *gt) return 0; } +static inline int xe_gt_sriov_pf_init(struct xe_gt *gt) +{ + return 0; +} + static inline void xe_gt_sriov_pf_init_hw(struct xe_gt *gt) { } From 990d35edc5d333ca6cd3acfdfc13683dc5bb105f Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Wed, 15 Jan 2025 14:20:29 -0800 Subject: [PATCH 0403/1443] drm/xe/oa: Set stream->pollin in xe_oa_buffer_check_unlocked We rely on stream->pollin to decide whether or not to block during poll/read calls. However, currently there are blocking read code paths which don't even set stream->pollin. The best place to consistently set stream->pollin for all code paths is therefore to set it in xe_oa_buffer_check_unlocked. Fixes: e936f885f1e9 ("drm/xe/oa/uapi: Expose OA stream fd") Signed-off-by: Ashutosh Dixit Acked-by: Rodrigo Vivi Reviewed-by: Jonathan Cavitt Reviewed-by: Umesh Nerlige Ramappa Link: https://patchwork.freedesktop.org/patch/msgid/20250115222029.3002103-1-ashutosh.dixit@intel.com (cherry picked from commit d3fedff828bb7e4a422c42caeafd5d974e24ee43) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_oa.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 6a08e6c92835..fa873f3d0a9d 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -237,7 +237,6 @@ static bool xe_oa_buffer_check_unlocked(struct xe_oa_stream *stream) u32 tail, hw_tail, partial_report_size, available; int report_size = stream->oa_buffer.format->size; unsigned long flags; - bool pollin; spin_lock_irqsave(&stream->oa_buffer.ptr_lock, flags); @@ -282,11 +281,11 @@ static bool xe_oa_buffer_check_unlocked(struct xe_oa_stream *stream) stream->oa_buffer.tail = tail; available = xe_oa_circ_diff(stream, stream->oa_buffer.tail, stream->oa_buffer.head); - pollin = available >= stream->wait_num_reports * report_size; + stream->pollin = available >= stream->wait_num_reports * report_size; spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags); - return pollin; + return stream->pollin; } static enum hrtimer_restart xe_oa_poll_check_timer_cb(struct hrtimer *hrtimer) @@ -294,10 +293,8 @@ static enum hrtimer_restart xe_oa_poll_check_timer_cb(struct hrtimer *hrtimer) struct xe_oa_stream *stream = container_of(hrtimer, typeof(*stream), poll_check_timer); - if (xe_oa_buffer_check_unlocked(stream)) { - stream->pollin = true; + if (xe_oa_buffer_check_unlocked(stream)) wake_up(&stream->poll_wq); - } hrtimer_forward_now(hrtimer, ns_to_ktime(stream->poll_period_ns)); From 042c48b73699c47d84b6ace73036e5a31a0d4cfc Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Wed, 22 Jan 2025 21:11:11 -0800 Subject: [PATCH 0404/1443] drm/xe/devcoredump: Move exec queue snapshot to Contexts section MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Having the exec queue snapshot inside a "GuC CT" section was always wrong. Commit c28fd6c358db ("drm/xe/devcoredump: Improve section headings and add tile info") tried to fix that bug, but with that also broke the mesa tool that parses the devcoredump, hence it was reverted in commit a53da2fb25a3 ("drm/xe: Revert some changes that break a mesa debug tool"). With the mesa tool also fixed, this can propagate as a fix on both kernel and userspace side to avoid unnecessary headache for a debug feature. Cc: John Harrison Cc: Julia Filipchuk Cc: José Roberto de Souza Cc: stable@vger.kernel.org Fixes: a53da2fb25a3 ("drm/xe: Revert some changes that break a mesa debug tool") Reviewed-by: José Roberto de Souza Link: https://patchwork.freedesktop.org/patch/msgid/20250123051112.1938193-2-lucas.demarchi@intel.com Signed-off-by: Lucas De Marchi (cherry picked from commit a37934ea75d331fafa7fe80b6180642ba5193422) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_devcoredump.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index 81dc7795c065..a7946a76777e 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -119,11 +119,7 @@ static ssize_t __xe_devcoredump_read(char *buffer, size_t count, drm_puts(&p, "\n**** GuC CT ****\n"); xe_guc_ct_snapshot_print(ss->guc.ct, &p); - /* - * Don't add a new section header here because the mesa debug decoder - * tool expects the context information to be in the 'GuC CT' section. - */ - /* drm_puts(&p, "\n**** Contexts ****\n"); */ + drm_puts(&p, "\n**** Contexts ****\n"); xe_guc_exec_queue_snapshot_print(ss->ge, &p); drm_puts(&p, "\n**** Job ****\n"); From a9ab6591b45258b79af1cb66112fd9f83c8855da Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 23 Jan 2025 12:22:03 -0800 Subject: [PATCH 0405/1443] drm/xe: Fix and re-enable xe_print_blob_ascii85() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") partially reverted some changes to workaround breakage caused to mesa tools. However, in doing so it also broke fetching the GuC log via debugfs since xe_print_blob_ascii85() simply bails out. The fix is to avoid the extra newlines: the devcoredump interface is line-oriented and adding random newlines in the middle breaks it. If a tool is able to parse it by looking at the data and checking for chars that are out of the ascii85 space, it can still do so. A format change that breaks the line-oriented output on devcoredump however needs better coordination with existing tools. v2: Add suffix description comment v3: Reword explanation of xe_print_blob_ascii85() calling drm_puts() in a loop Reviewed-by: José Roberto de Souza Cc: John Harrison Cc: Julia Filipchuk Cc: José Roberto de Souza Cc: stable@vger.kernel.org Fixes: 70fb86a85dc9 ("drm/xe: Revert some changes that break a mesa debug tool") Fixes: ec1455ce7e35 ("drm/xe/devcoredump: Add ASCII85 dump helper function") Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-2-jose.souza@intel.com Signed-off-by: Lucas De Marchi (cherry picked from commit 2c95bbf5002776117a69caed3b31c10bf7341bec) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_devcoredump.c | 36 +++++++++++------------------ drivers/gpu/drm/xe/xe_devcoredump.h | 2 +- drivers/gpu/drm/xe/xe_guc_ct.c | 3 ++- drivers/gpu/drm/xe/xe_guc_log.c | 4 +++- 4 files changed, 20 insertions(+), 25 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index a7946a76777e..39fe485d2085 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -391,42 +391,34 @@ int xe_devcoredump_init(struct xe_device *xe) /** * xe_print_blob_ascii85 - print a BLOB to some useful location in ASCII85 * - * The output is split to multiple lines because some print targets, e.g. dmesg - * cannot handle arbitrarily long lines. Note also that printing to dmesg in - * piece-meal fashion is not possible, each separate call to drm_puts() has a - * line-feed automatically added! Therefore, the entire output line must be - * constructed in a local buffer first, then printed in one atomic output call. + * The output is split into multiple calls to drm_puts() because some print + * targets, e.g. dmesg, cannot handle arbitrarily long lines. These targets may + * add newlines, as is the case with dmesg: each drm_puts() call creates a + * separate line. * * There is also a scheduler yield call to prevent the 'task has been stuck for * 120s' kernel hang check feature from firing when printing to a slow target * such as dmesg over a serial port. * - * TODO: Add compression prior to the ASCII85 encoding to shrink huge buffers down. - * * @p: the printer object to output to * @prefix: optional prefix to add to output string + * @suffix: optional suffix to add at the end. 0 disables it and is + * not added to the output, which is useful when using multiple calls + * to dump data to @p * @blob: the Binary Large OBject to dump out * @offset: offset in bytes to skip from the front of the BLOB, must be a multiple of sizeof(u32) * @size: the size in bytes of the BLOB, must be a multiple of sizeof(u32) */ -void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, +void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char suffix, const void *blob, size_t offset, size_t size) { const u32 *blob32 = (const u32 *)blob; char buff[ASCII85_BUFSZ], *line_buff; size_t line_pos = 0; - /* - * Splitting blobs across multiple lines is not compatible with the mesa - * debug decoder tool. Note that even dropping the explicit '\n' below - * doesn't help because the GuC log is so big some underlying implementation - * still splits the lines at 512K characters. So just bail completely for - * the moment. - */ - return; - #define DMESG_MAX_LINE_LEN 800 -#define MIN_SPACE (ASCII85_BUFSZ + 2) /* 85 + "\n\0" */ + /* Always leave space for the suffix char and the \0 */ +#define MIN_SPACE (ASCII85_BUFSZ + 2) /* 85 + "\0" */ if (size & 3) drm_printf(p, "Size not word aligned: %zu", size); @@ -458,7 +450,6 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, line_pos += strlen(line_buff + line_pos); if ((line_pos + MIN_SPACE) >= DMESG_MAX_LINE_LEN) { - line_buff[line_pos++] = '\n'; line_buff[line_pos++] = 0; drm_puts(p, line_buff); @@ -470,10 +461,11 @@ void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, } } - if (line_pos) { - line_buff[line_pos++] = '\n'; - line_buff[line_pos++] = 0; + if (suffix) + line_buff[line_pos++] = suffix; + if (line_pos) { + line_buff[line_pos++] = 0; drm_puts(p, line_buff); } diff --git a/drivers/gpu/drm/xe/xe_devcoredump.h b/drivers/gpu/drm/xe/xe_devcoredump.h index 6a17e6d60102..5391a80a4d1b 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.h +++ b/drivers/gpu/drm/xe/xe_devcoredump.h @@ -29,7 +29,7 @@ static inline int xe_devcoredump_init(struct xe_device *xe) } #endif -void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, +void xe_print_blob_ascii85(struct drm_printer *p, const char *prefix, char suffix, const void *blob, size_t offset, size_t size); #endif diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 8b65c5e959cc..50c8076b5158 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -1724,7 +1724,8 @@ void xe_guc_ct_snapshot_print(struct xe_guc_ct_snapshot *snapshot, snapshot->g2h_outstanding); if (snapshot->ctb) - xe_print_blob_ascii85(p, "CTB data", snapshot->ctb, 0, snapshot->ctb_size); + xe_print_blob_ascii85(p, "CTB data", '\n', + snapshot->ctb, 0, snapshot->ctb_size); } else { drm_puts(p, "CT disabled\n"); } diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index df4cfb698cdb..2baa4d95571f 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -211,8 +211,10 @@ void xe_guc_log_snapshot_print(struct xe_guc_log_snapshot *snapshot, struct drm_ remain = snapshot->size; for (i = 0; i < snapshot->num_chunks; i++) { size_t size = min(GUC_LOG_CHUNK_SIZE, remain); + const char *prefix = i ? NULL : "Log data"; + char suffix = i == snapshot->num_chunks - 1 ? '\n' : 0; - xe_print_blob_ascii85(p, i ? NULL : "Log data", snapshot->copy[i], 0, size); + xe_print_blob_ascii85(p, prefix, suffix, snapshot->copy[i], 0, size); remain -= size; } } From d275a5e0c5f528b4b877ec683b8cd8bfced96af5 Mon Sep 17 00:00:00 2001 From: Devarsh Thakkar Date: Mon, 3 Feb 2025 21:24:31 +0530 Subject: [PATCH 0406/1443] dt-bindings: display: ti: Fix compatible for am62a7 dss Fix incorrect format of compatible string (comma instead of hyphen) for TI's AM62A7 SoC. s/ti,am62a7,dss/ti,am62a7-dss Fixes: 7959ceb767e4 ("dt-bindings: display: ti: Add support for am62a7 dss") Reviewed-by: Krzysztof Kozlowski Signed-off-by: Devarsh Thakkar Link: https://lore.kernel.org/r/20250203155431.2174170-1-devarsht@ti.com Signed-off-by: Rob Herring (Arm) --- Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml b/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml index 55e3e490d0e6..31c4ffcb599c 100644 --- a/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml +++ b/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml @@ -23,7 +23,7 @@ properties: compatible: enum: - ti,am625-dss - - ti,am62a7,dss + - ti,am62a7-dss - ti,am65x-dss reg: From ff48e05d8d1eefbdeb4504c0275c78654b858046 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:25 -0800 Subject: [PATCH 0407/1443] drm/xe/pxp: Initialize PXP structure and KCR reg As the first step towards adding PXP support, hook in the PXP init function, allocate the PXP structure and initialize the KCR register to allow PXP HWDRM sessions. v2: remove unneeded includes, free PXP memory on error (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-2-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/Makefile | 1 + .../xe/compat-i915-headers/pxp/intel_pxp.h | 4 +- drivers/gpu/drm/xe/regs/xe_pxp_regs.h | 17 +++ drivers/gpu/drm/xe/xe_device.c | 6 + drivers/gpu/drm/xe/xe_device_types.h | 8 +- drivers/gpu/drm/xe/xe_pci.c | 2 + drivers/gpu/drm/xe/xe_pxp.c | 107 ++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp.h | 13 +++ drivers/gpu/drm/xe/xe_pxp_types.h | 26 +++++ 9 files changed, 180 insertions(+), 4 deletions(-) create mode 100644 drivers/gpu/drm/xe/regs/xe_pxp_regs.h create mode 100644 drivers/gpu/drm/xe/xe_pxp.c create mode 100644 drivers/gpu/drm/xe/xe_pxp.h create mode 100644 drivers/gpu/drm/xe/xe_pxp_types.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index e6960db45b73..708e9f0108dc 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -87,6 +87,7 @@ xe-y += xe_bb.o \ xe_preempt_fence.o \ xe_pt.o \ xe_pt_walk.o \ + xe_pxp.o \ xe_query.o \ xe_range_fence.o \ xe_reg_sr.o \ diff --git a/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h b/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h index 5dfc587c8237..419e8e926f00 100644 --- a/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h +++ b/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h @@ -10,9 +10,9 @@ #include struct drm_gem_object; -struct intel_pxp; +struct xe_pxp; -static inline int intel_pxp_key_check(struct intel_pxp *pxp, +static inline int intel_pxp_key_check(struct xe_pxp *pxp, struct drm_gem_object *obj, bool assign) { diff --git a/drivers/gpu/drm/xe/regs/xe_pxp_regs.h b/drivers/gpu/drm/xe/regs/xe_pxp_regs.h new file mode 100644 index 000000000000..d67cf210d23d --- /dev/null +++ b/drivers/gpu/drm/xe/regs/xe_pxp_regs.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2024, Intel Corporation. All rights reserved. + */ + +#ifndef __XE_PXP_REGS_H__ +#define __XE_PXP_REGS_H__ + +#include "regs/xe_regs.h" + +/* The following registers are only valid on platforms with a media GT */ + +/* KCR enable/disable control */ +#define KCR_INIT XE_REG(0x3860f0) +#define KCR_INIT_ALLOW_DISPLAY_ME_WRITES REG_BIT(14) + +#endif /* __XE_PXP_REGS_H__ */ diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 8a2763e8a6e2..f30f8f668dee 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -50,6 +50,7 @@ #include "xe_pcode.h" #include "xe_pm.h" #include "xe_pmu.h" +#include "xe_pxp.h" #include "xe_query.h" #include "xe_sriov.h" #include "xe_survivability_mode.h" @@ -861,6 +862,11 @@ int xe_device_probe(struct xe_device *xe) if (err) goto err_fini_oa; + /* A PXP init failure is not fatal */ + err = xe_pxp_init(xe); + if (err && err != -EOPNOTSUPP) + drm_err(&xe->drm, "PXP initialization failed: %pe\n", ERR_PTR(err)); + err = drm_dev_register(&xe->drm, 0); if (err) goto err_fini_display; diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 89f532b67bc4..c0e886bac183 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -36,6 +36,7 @@ struct xe_ggtt; struct xe_pat_ops; +struct xe_pxp; #define XE_BO_INVALID_OFFSET LONG_MAX @@ -307,6 +308,8 @@ struct xe_device { u8 has_heci_gscfi:1; /** @info.has_llc: Device has a shared CPU+GPU last level cache */ u8 has_llc:1; + /** @info.has_pxp: Device has PXP support */ + u8 has_pxp:1; /** @info.has_range_tlb_invalidation: Has range based TLB invalidations */ u8 has_range_tlb_invalidation:1; /** @info.has_sriov: Supports SR-IOV */ @@ -508,6 +511,9 @@ struct xe_device { /** @oa: oa observation subsystem */ struct xe_oa oa; + /** @pxp: Encapsulate Protected Xe Path support */ + struct xe_pxp *pxp; + /** @needs_flr_on_fini: requests function-reset on fini */ bool needs_flr_on_fini; @@ -583,8 +589,6 @@ struct xe_device { unsigned int czclk_freq; unsigned int fsb_freq, mem_freq, is_ddr3; }; - - void *pxp; #endif }; diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index f05cf26a088c..0a6e58d55682 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -62,6 +62,7 @@ struct xe_device_desc { u8 has_heci_gscfi:1; u8 has_heci_cscfi:1; u8 has_llc:1; + u8 has_pxp:1; u8 has_sriov:1; u8 skip_guc_pc:1; u8 skip_mtcfg:1; @@ -618,6 +619,7 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.has_heci_gscfi = desc->has_heci_gscfi; xe->info.has_heci_cscfi = desc->has_heci_cscfi; xe->info.has_llc = desc->has_llc; + xe->info.has_pxp = desc->has_pxp; xe->info.has_sriov = desc->has_sriov; xe->info.skip_guc_pc = desc->skip_guc_pc; xe->info.skip_mtcfg = desc->skip_mtcfg; diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c new file mode 100644 index 000000000000..1a4d12d37d96 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -0,0 +1,107 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright(c) 2024 Intel Corporation. + */ + +#include "xe_pxp.h" + +#include + +#include "xe_device_types.h" +#include "xe_force_wake.h" +#include "xe_gt.h" +#include "xe_gt_types.h" +#include "xe_mmio.h" +#include "xe_pxp_types.h" +#include "xe_uc_fw.h" +#include "regs/xe_pxp_regs.h" + +/** + * DOC: PXP + * + * PXP (Protected Xe Path) allows execution and flip to display of protected + * (i.e. encrypted) objects. This feature is currently only supported in + * integrated parts. + */ + +static bool pxp_is_supported(const struct xe_device *xe) +{ + return xe->info.has_pxp && IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY); +} + +static int kcr_pxp_set_status(const struct xe_pxp *pxp, bool enable) +{ + u32 val = enable ? _MASKED_BIT_ENABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES) : + _MASKED_BIT_DISABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES); + unsigned int fw_ref; + + fw_ref = xe_force_wake_get(gt_to_fw(pxp->gt), XE_FW_GT); + if (!xe_force_wake_ref_has_domain(fw_ref, XE_FW_GT)) + return -EIO; + + xe_mmio_write32(&pxp->gt->mmio, KCR_INIT, val); + xe_force_wake_put(gt_to_fw(pxp->gt), fw_ref); + + return 0; +} + +static int kcr_pxp_enable(const struct xe_pxp *pxp) +{ + return kcr_pxp_set_status(pxp, true); +} + +/** + * xe_pxp_init - initialize PXP support + * @xe: the xe_device structure + * + * Initialize the HW state and allocate the objects required for PXP support. + * Note that some of the requirement for PXP support (GSC proxy init, HuC auth) + * are performed asynchronously as part of the GSC init. PXP can only be used + * after both this function and the async worker have completed. + * + * Returns -EOPNOTSUPP if PXP is not supported, 0 if PXP initialization is + * successful, other errno value if there is an error during the init. + */ +int xe_pxp_init(struct xe_device *xe) +{ + struct xe_gt *gt = xe->tiles[0].media_gt; + struct xe_pxp *pxp; + int err; + + if (!pxp_is_supported(xe)) + return -EOPNOTSUPP; + + /* we only support PXP on single tile devices with a media GT */ + if (xe->info.tile_count > 1 || !gt) + return -EOPNOTSUPP; + + /* The GSCCS is required for submissions to the GSC FW */ + if (!(gt->info.engine_mask & BIT(XE_HW_ENGINE_GSCCS0))) + return -EOPNOTSUPP; + + /* PXP requires both GSC and HuC firmwares to be available */ + if (!xe_uc_fw_is_loadable(>->uc.gsc.fw) || + !xe_uc_fw_is_loadable(>->uc.huc.fw)) { + drm_info(&xe->drm, "skipping PXP init due to missing FW dependencies"); + return -EOPNOTSUPP; + } + + pxp = drmm_kzalloc(&xe->drm, sizeof(struct xe_pxp), GFP_KERNEL); + if (!pxp) + return -ENOMEM; + + pxp->xe = xe; + pxp->gt = gt; + + err = kcr_pxp_enable(pxp); + if (err) + goto out_free; + + xe->pxp = pxp; + + return 0; + +out_free: + drmm_kfree(&xe->drm, pxp); + return err; +} diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h new file mode 100644 index 000000000000..00f5e688c0d9 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2024, Intel Corporation. All rights reserved. + */ + +#ifndef __XE_PXP_H__ +#define __XE_PXP_H__ + +struct xe_device; + +int xe_pxp_init(struct xe_device *xe); + +#endif /* __XE_PXP_H__ */ diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h new file mode 100644 index 000000000000..4639cf49d267 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -0,0 +1,26 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2024, Intel Corporation. All rights reserved. + */ + +#ifndef __XE_PXP_TYPES_H__ +#define __XE_PXP_TYPES_H__ + +struct xe_device; +struct xe_gt; + +/** + * struct xe_pxp - pxp state + */ +struct xe_pxp { + /** @xe: Backpoiner to the xe_device struct */ + struct xe_device *xe; + + /** + * @gt: pointer to the gt that owns the submission-side of PXP + * (VDBOX, KCR and GSC) + */ + struct xe_gt *gt; +}; + +#endif /* __XE_PXP_TYPES_H__ */ From dcdd6b84d9acaa0794c29de7024cfdb20cfd7b92 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:26 -0800 Subject: [PATCH 0408/1443] drm/xe/pxp: Allocate PXP execution resources MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit PXP requires submissions to the HW for the following operations 1) Key invalidation, done via the VCS engine 2) Communication with the GSC FW for session management, done via the GSCCS. Key invalidation submissions are serialized (only 1 termination can be serviced at a given time) and done via GGTT, so we can allocate a simple BO and a kernel queue for it. Submissions for session management are tied to a PXP client (identified by a unique host_session_id); from the GSC POV this is a user-accessible construct, so all related submission must be done via PPGTT. The driver does not currently support PPGTT submission from within the kernel, so to add this support, the following changes have been included: - a new type of kernel-owned VM (marked as GSC), required to ensure we don't use fault mode on the engine and to mark the different lock usage with lockdep. - a new function to map a BO into a VM from within the kernel. v2: improve comments and function name, remove unneeded include (John) v3: fix variable/function names in documentation Signed-off-by: Daniele Ceraolo Spurio Cc: Matthew Brost Cc: Thomas Hellström Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-3-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/Makefile | 1 + drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h | 7 + drivers/gpu/drm/xe/xe_exec_queue.c | 3 + drivers/gpu/drm/xe/xe_pxp.c | 23 +- drivers/gpu/drm/xe/xe_pxp_submit.c | 199 ++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp_submit.h | 14 ++ drivers/gpu/drm/xe/xe_pxp_types.h | 48 +++++ drivers/gpu/drm/xe/xe_vm.c | 123 +++++++++-- drivers/gpu/drm/xe/xe_vm.h | 6 + drivers/gpu/drm/xe/xe_vm_types.h | 1 + 10 files changed, 412 insertions(+), 13 deletions(-) create mode 100644 drivers/gpu/drm/xe/xe_pxp_submit.c create mode 100644 drivers/gpu/drm/xe/xe_pxp_submit.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 708e9f0108dc..720d13c16fa5 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -88,6 +88,7 @@ xe-y += xe_bb.o \ xe_pt.o \ xe_pt_walk.o \ xe_pxp.o \ + xe_pxp_submit.o \ xe_query.o \ xe_range_fence.o \ xe_reg_sr.o \ diff --git a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h index 57520809e48d..f3c4cf10ba20 100644 --- a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h +++ b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h @@ -6,6 +6,7 @@ #ifndef _ABI_GSC_PXP_COMMANDS_ABI_H #define _ABI_GSC_PXP_COMMANDS_ABI_H +#include #include /* Heci client ID for PXP commands */ @@ -13,6 +14,12 @@ #define PXP_APIVER(x, y) (((x) & 0xFFFF) << 16 | ((y) & 0xFFFF)) +/* + * A PXP sub-section in an HECI packet can be up to 64K big in each direction. + * This does not include the top-level GSC header. + */ +#define PXP_MAX_PACKET_SIZE SZ_64K + /* * there are a lot of status codes for PXP, but we only define the cross-API * common ones that we actually can handle in the kernel driver. Other failure diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 7e1abbbfba12..252bfa11cae9 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -153,6 +153,9 @@ struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *v struct xe_exec_queue *q; int err; + /* VMs for GSCCS queues (and only those) must have the XE_VM_FLAG_GSC flag */ + xe_assert(xe, !vm || (!!(vm->flags & XE_VM_FLAG_GSC) == !!(hwe->engine_id == XE_HW_ENGINE_GSCCS0))); + q = __xe_exec_queue_alloc(xe, vm, logical_mask, width, hwe, flags, extensions); if (IS_ERR(q)) diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 1a4d12d37d96..89d71a69cdff 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -12,6 +12,7 @@ #include "xe_gt.h" #include "xe_gt_types.h" #include "xe_mmio.h" +#include "xe_pxp_submit.h" #include "xe_pxp_types.h" #include "xe_uc_fw.h" #include "regs/xe_pxp_regs.h" @@ -50,6 +51,20 @@ static int kcr_pxp_enable(const struct xe_pxp *pxp) return kcr_pxp_set_status(pxp, true); } +static int kcr_pxp_disable(const struct xe_pxp *pxp) +{ + return kcr_pxp_set_status(pxp, false); +} + +static void pxp_fini(void *arg) +{ + struct xe_pxp *pxp = arg; + + xe_pxp_destroy_execution_resources(pxp); + + /* no need to explicitly disable KCR since we're going to do an FLR */ +} + /** * xe_pxp_init - initialize PXP support * @xe: the xe_device structure @@ -97,10 +112,16 @@ int xe_pxp_init(struct xe_device *xe) if (err) goto out_free; + err = xe_pxp_allocate_execution_resources(pxp); + if (err) + goto out_kcr_disable; + xe->pxp = pxp; - return 0; + return devm_add_action_or_reset(xe->drm.dev, pxp_fini, pxp); +out_kcr_disable: + kcr_pxp_disable(pxp); out_free: drmm_kfree(&xe->drm, pxp); return err; diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c new file mode 100644 index 000000000000..52a6143b5ebe --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp_submit.c @@ -0,0 +1,199 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright(c) 2024 Intel Corporation. + */ + +#include "xe_pxp_submit.h" + +#include + +#include "xe_device_types.h" +#include "xe_bo.h" +#include "xe_exec_queue.h" +#include "xe_gsc_submit.h" +#include "xe_gt.h" +#include "xe_pxp_types.h" +#include "xe_vm.h" + +/* + * The VCS is used for kernel-owned GGTT submissions to issue key termination. + * Terminations are serialized, so we only need a single queue and a single + * batch. + */ +static int allocate_vcs_execution_resources(struct xe_pxp *pxp) +{ + struct xe_gt *gt = pxp->gt; + struct xe_device *xe = pxp->xe; + struct xe_tile *tile = gt_to_tile(gt); + struct xe_hw_engine *hwe; + struct xe_exec_queue *q; + struct xe_bo *bo; + int err; + + hwe = xe_gt_hw_engine(gt, XE_ENGINE_CLASS_VIDEO_DECODE, 0, true); + if (!hwe) + return -ENODEV; + + q = xe_exec_queue_create(xe, NULL, BIT(hwe->logical_instance), 1, hwe, + EXEC_QUEUE_FLAG_KERNEL | EXEC_QUEUE_FLAG_PERMANENT, 0); + if (IS_ERR(q)) + return PTR_ERR(q); + + /* + * Each termination is 16 DWORDS, so 4K is enough to contain a + * termination for each sessions. + */ + bo = xe_bo_create_pin_map(xe, tile, 0, SZ_4K, ttm_bo_type_kernel, + XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED | XE_BO_FLAG_GGTT); + if (IS_ERR(bo)) { + err = PTR_ERR(bo); + goto out_queue; + } + + pxp->vcs_exec.q = q; + pxp->vcs_exec.bo = bo; + + return 0; + +out_queue: + xe_exec_queue_put(q); + return err; +} + +static void destroy_vcs_execution_resources(struct xe_pxp *pxp) +{ + if (pxp->vcs_exec.bo) + xe_bo_unpin_map_no_vm(pxp->vcs_exec.bo); + + if (pxp->vcs_exec.q) + xe_exec_queue_put(pxp->vcs_exec.q); +} + +#define PXP_BB_SIZE XE_PAGE_SIZE +static int allocate_gsc_client_resources(struct xe_gt *gt, + struct xe_pxp_gsc_client_resources *gsc_res, + size_t inout_size) +{ + struct xe_tile *tile = gt_to_tile(gt); + struct xe_device *xe = tile_to_xe(tile); + struct xe_hw_engine *hwe; + struct xe_vm *vm; + struct xe_bo *bo; + struct xe_exec_queue *q; + struct dma_fence *fence; + long timeout; + int err = 0; + + hwe = xe_gt_hw_engine(gt, XE_ENGINE_CLASS_OTHER, 0, true); + + /* we shouldn't reach here if the GSC engine is not available */ + xe_assert(xe, hwe); + + /* PXP instructions must be issued from PPGTT */ + vm = xe_vm_create(xe, XE_VM_FLAG_GSC); + if (IS_ERR(vm)) + return PTR_ERR(vm); + + /* We allocate a single object for the batch and the in/out memory */ + xe_vm_lock(vm, false); + bo = xe_bo_create_pin_map(xe, tile, vm, PXP_BB_SIZE + inout_size * 2, + ttm_bo_type_kernel, + XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED | XE_BO_FLAG_NEEDS_UC); + xe_vm_unlock(vm); + if (IS_ERR(bo)) { + err = PTR_ERR(bo); + goto vm_out; + } + + fence = xe_vm_bind_kernel_bo(vm, bo, NULL, 0, XE_CACHE_WB); + if (IS_ERR(fence)) { + err = PTR_ERR(fence); + goto bo_out; + } + + timeout = dma_fence_wait_timeout(fence, false, HZ); + dma_fence_put(fence); + if (timeout <= 0) { + err = timeout ?: -ETIME; + goto bo_out; + } + + q = xe_exec_queue_create(xe, vm, BIT(hwe->logical_instance), 1, hwe, + EXEC_QUEUE_FLAG_KERNEL | + EXEC_QUEUE_FLAG_PERMANENT, 0); + if (IS_ERR(q)) { + err = PTR_ERR(q); + goto bo_out; + } + + gsc_res->vm = vm; + gsc_res->bo = bo; + gsc_res->inout_size = inout_size; + gsc_res->batch = IOSYS_MAP_INIT_OFFSET(&bo->vmap, 0); + gsc_res->msg_in = IOSYS_MAP_INIT_OFFSET(&bo->vmap, PXP_BB_SIZE); + gsc_res->msg_out = IOSYS_MAP_INIT_OFFSET(&bo->vmap, PXP_BB_SIZE + inout_size); + gsc_res->q = q; + + /* initialize host-session-handle (for all Xe-to-gsc-firmware PXP cmds) */ + gsc_res->host_session_handle = xe_gsc_create_host_session_id(); + + return 0; + +bo_out: + xe_bo_unpin_map_no_vm(bo); +vm_out: + xe_vm_close_and_put(vm); + + return err; +} + +static void destroy_gsc_client_resources(struct xe_pxp_gsc_client_resources *gsc_res) +{ + if (!gsc_res->q) + return; + + xe_exec_queue_put(gsc_res->q); + xe_bo_unpin_map_no_vm(gsc_res->bo); + xe_vm_close_and_put(gsc_res->vm); +} + +/** + * xe_pxp_allocate_execution_resources - Allocate PXP submission objects + * @pxp: the xe_pxp structure + * + * Allocates exec_queues objects for VCS and GSCCS submission. The GSCCS + * submissions are done via PPGTT, so this function allocates a VM for it and + * maps the object into it. + * + * Returns 0 if the allocation and mapping is successful, an errno value + * otherwise. + */ +int xe_pxp_allocate_execution_resources(struct xe_pxp *pxp) +{ + int err; + + err = allocate_vcs_execution_resources(pxp); + if (err) + return err; + + /* + * PXP commands can require a lot of BO space (see PXP_MAX_PACKET_SIZE), + * but we currently only support a subset of commands that are small + * (< 20 dwords), so a single page is enough for now. + */ + err = allocate_gsc_client_resources(pxp->gt, &pxp->gsc_res, XE_PAGE_SIZE); + if (err) + goto destroy_vcs_context; + + return 0; + +destroy_vcs_context: + destroy_vcs_execution_resources(pxp); + return err; +} + +void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp) +{ + destroy_gsc_client_resources(&pxp->gsc_res); + destroy_vcs_execution_resources(pxp); +} diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.h b/drivers/gpu/drm/xe/xe_pxp_submit.h new file mode 100644 index 000000000000..fd21ac935be1 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp_submit.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2024, Intel Corporation. All rights reserved. + */ + +#ifndef __XE_PXP_SUBMIT_H__ +#define __XE_PXP_SUBMIT_H__ + +struct xe_pxp; + +int xe_pxp_allocate_execution_resources(struct xe_pxp *pxp); +void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp); + +#endif /* __XE_PXP_SUBMIT_H__ */ diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h index 4639cf49d267..0f86b50756cc 100644 --- a/drivers/gpu/drm/xe/xe_pxp_types.h +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -6,8 +6,45 @@ #ifndef __XE_PXP_TYPES_H__ #define __XE_PXP_TYPES_H__ +#include +#include + +struct xe_bo; +struct xe_exec_queue; struct xe_device; struct xe_gt; +struct xe_vm; + +/** + * struct xe_pxp_gsc_client_resources - resources for GSC submission by a PXP + * client. The GSC FW supports multiple GSC client active at the same time. + */ +struct xe_pxp_gsc_client_resources { + /** + * @host_session_handle: handle used to identify the client in messages + * sent to the GSC firmware. + */ + u64 host_session_handle; + /** @vm: VM used for PXP submissions to the GSCCS */ + struct xe_vm *vm; + /** @q: GSCCS exec queue for PXP submissions */ + struct xe_exec_queue *q; + + /** + * @bo: BO used for submissions to the GSCCS and GSC FW. It includes + * space for the GSCCS batch and the input/output buffers read/written + * by the FW + */ + struct xe_bo *bo; + /** @inout_size: size of each of the msg_in/out sections individually */ + u32 inout_size; + /** @batch: iosys_map to the batch memory within the BO */ + struct iosys_map batch; + /** @msg_in: iosys_map to the input memory within the BO */ + struct iosys_map msg_in; + /** @msg_out: iosys_map to the output memory within the BO */ + struct iosys_map msg_out; +}; /** * struct xe_pxp - pxp state @@ -21,6 +58,17 @@ struct xe_pxp { * (VDBOX, KCR and GSC) */ struct xe_gt *gt; + + /** @vcs_exec: kernel-owned objects for PXP submissions to the VCS */ + struct { + /** @vcs_exec.q: kernel-owned VCS exec queue used for PXP terminations */ + struct xe_exec_queue *q; + /** @vcs_exec.bo: BO used for submissions to the VCS */ + struct xe_bo *bo; + } vcs_exec; + + /** @gsc_res: kernel-owned objects for PXP submissions to the GSCCS */ + struct xe_pxp_gsc_client_resources gsc_res; }; #endif /* __XE_PXP_TYPES_H__ */ diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 690330352d4c..b9270d059e18 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -1382,6 +1382,12 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags) struct xe_tile *tile; u8 id; + /* + * Since the GSCCS is not user-accessible, we don't expect a GSC VM to + * ever be in faulting mode. + */ + xe_assert(xe, !((flags & XE_VM_FLAG_GSC) && (flags & XE_VM_FLAG_FAULT_MODE))); + vm = kzalloc(sizeof(*vm), GFP_KERNEL); if (!vm) return ERR_PTR(-ENOMEM); @@ -1392,7 +1398,21 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags) vm->flags = flags; - init_rwsem(&vm->lock); + /** + * GSC VMs are kernel-owned, only used for PXP ops and can sometimes be + * manipulated under the PXP mutex. However, the PXP mutex can be taken + * under a user-VM lock when the PXP session is started at exec_queue + * creation time. Those are different VMs and therefore there is no risk + * of deadlock, but we need to tell lockdep that this is the case or it + * will print a warning. + */ + if (flags & XE_VM_FLAG_GSC) { + static struct lock_class_key gsc_vm_key; + + __init_rwsem(&vm->lock, "gsc_vm", &gsc_vm_key); + } else { + init_rwsem(&vm->lock); + } mutex_init(&vm->snap_mutex); INIT_LIST_HEAD(&vm->rebind_list); @@ -2668,11 +2688,10 @@ static void vm_bind_ioctl_ops_fini(struct xe_vm *vm, struct xe_vma_ops *vops, for (i = 0; i < vops->num_syncs; i++) xe_sync_entry_signal(vops->syncs + i, fence); xe_exec_queue_last_fence_set(wait_exec_queue, vm, fence); - dma_fence_put(fence); } -static int vm_bind_ioctl_ops_execute(struct xe_vm *vm, - struct xe_vma_ops *vops) +static struct dma_fence *vm_bind_ioctl_ops_execute(struct xe_vm *vm, + struct xe_vma_ops *vops) { struct drm_exec exec; struct dma_fence *fence; @@ -2685,21 +2704,21 @@ static int vm_bind_ioctl_ops_execute(struct xe_vm *vm, drm_exec_until_all_locked(&exec) { err = vm_bind_ioctl_ops_lock_and_prep(&exec, vm, vops); drm_exec_retry_on_contention(&exec); - if (err) - goto unlock; - - fence = ops_execute(vm, vops); - if (IS_ERR(fence)) { - err = PTR_ERR(fence); + if (err) { + fence = ERR_PTR(err); goto unlock; } + fence = ops_execute(vm, vops); + if (IS_ERR(fence)) + goto unlock; + vm_bind_ioctl_ops_fini(vm, vops, fence); } unlock: drm_exec_fini(&exec); - return err; + return fence; } ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO); @@ -2931,6 +2950,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) struct xe_sync_entry *syncs = NULL; struct drm_xe_vm_bind_op *bind_ops; struct xe_vma_ops vops; + struct dma_fence *fence; int err; int i; @@ -3095,7 +3115,11 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) if (err) goto unwind_ops; - err = vm_bind_ioctl_ops_execute(vm, &vops); + fence = vm_bind_ioctl_ops_execute(vm, &vops); + if (IS_ERR(fence)) + err = PTR_ERR(fence); + else + dma_fence_put(fence); unwind_ops: if (err && err != -ENODATA) @@ -3129,6 +3153,81 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) return err; } +/** + * xe_vm_bind_kernel_bo - bind a kernel BO to a VM + * @vm: VM to bind the BO to + * @bo: BO to bind + * @q: exec queue to use for the bind (optional) + * @addr: address at which to bind the BO + * @cache_lvl: PAT cache level to use + * + * Execute a VM bind map operation on a kernel-owned BO to bind it into a + * kernel-owned VM. + * + * Returns a dma_fence to track the binding completion if the job to do so was + * successfully submitted, an error pointer otherwise. + */ +struct dma_fence *xe_vm_bind_kernel_bo(struct xe_vm *vm, struct xe_bo *bo, + struct xe_exec_queue *q, u64 addr, + enum xe_cache_level cache_lvl) +{ + struct xe_vma_ops vops; + struct drm_gpuva_ops *ops = NULL; + struct dma_fence *fence; + int err; + + xe_bo_get(bo); + xe_vm_get(vm); + if (q) + xe_exec_queue_get(q); + + down_write(&vm->lock); + + xe_vma_ops_init(&vops, vm, q, NULL, 0); + + ops = vm_bind_ioctl_ops_create(vm, bo, 0, addr, bo->size, + DRM_XE_VM_BIND_OP_MAP, 0, 0, + vm->xe->pat.idx[cache_lvl]); + if (IS_ERR(ops)) { + err = PTR_ERR(ops); + goto release_vm_lock; + } + + err = vm_bind_ioctl_ops_parse(vm, ops, &vops); + if (err) + goto release_vm_lock; + + xe_assert(vm->xe, !list_empty(&vops.list)); + + err = xe_vma_ops_alloc(&vops, false); + if (err) + goto unwind_ops; + + fence = vm_bind_ioctl_ops_execute(vm, &vops); + if (IS_ERR(fence)) + err = PTR_ERR(fence); + +unwind_ops: + if (err && err != -ENODATA) + vm_bind_ioctl_ops_unwind(vm, &ops, 1); + + xe_vma_ops_fini(&vops); + drm_gpuva_ops_free(&vm->gpuvm, ops); + +release_vm_lock: + up_write(&vm->lock); + + if (q) + xe_exec_queue_put(q); + xe_vm_put(vm); + xe_bo_put(bo); + + if (err) + fence = ERR_PTR(err); + + return fence; +} + /** * xe_vm_lock() - Lock the vm's dma_resv object * @vm: The struct xe_vm whose lock is to be locked diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h index 23adb7442881..0a2fa6c0815b 100644 --- a/drivers/gpu/drm/xe/xe_vm.h +++ b/drivers/gpu/drm/xe/xe_vm.h @@ -18,6 +18,8 @@ struct drm_file; struct ttm_buffer_object; +struct dma_fence; + struct xe_exec_queue; struct xe_file; struct xe_sync_entry; @@ -247,6 +249,10 @@ int xe_vm_lock_vma(struct drm_exec *exec, struct xe_vma *vma); int xe_vm_validate_rebind(struct xe_vm *vm, struct drm_exec *exec, unsigned int num_fences); +struct dma_fence *xe_vm_bind_kernel_bo(struct xe_vm *vm, struct xe_bo *bo, + struct xe_exec_queue *q, u64 addr, + enum xe_cache_level cache_lvl); + /** * xe_vm_resv() - Return's the vm's reservation object * @vm: The vm diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h index 7f9a303e51d8..52467b9b5348 100644 --- a/drivers/gpu/drm/xe/xe_vm_types.h +++ b/drivers/gpu/drm/xe/xe_vm_types.h @@ -164,6 +164,7 @@ struct xe_vm { #define XE_VM_FLAG_BANNED BIT(5) #define XE_VM_FLAG_TILE_ID(flags) FIELD_GET(GENMASK(7, 6), flags) #define XE_VM_FLAG_SET_TILE_ID(tile) FIELD_PREP(GENMASK(7, 6), (tile)->id) +#define XE_VM_FLAG_GSC BIT(8) unsigned long flags; /** @composite_fence_ctx: context composite fence */ From f0c06677d1105a599370ffd0c94679996c3656f2 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:27 -0800 Subject: [PATCH 0409/1443] drm/xe/pxp: Add VCS inline termination support The key termination is done with a specific submission to the VCS engine. This flow will be triggered in response to a termination interrupt, whose handling is coming in a follow-up patch in the series. v2: clean up defines and command emission code. (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-4-daniele.ceraolospurio@intel.com --- .../gpu/drm/xe/instructions/xe_instr_defs.h | 1 + .../gpu/drm/xe/instructions/xe_mfx_commands.h | 28 +++++ .../gpu/drm/xe/instructions/xe_mi_commands.h | 5 + drivers/gpu/drm/xe/xe_lrc.h | 3 +- drivers/gpu/drm/xe/xe_pxp_submit.c | 114 ++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp_submit.h | 4 + drivers/gpu/drm/xe/xe_ring_ops.c | 4 +- 7 files changed, 156 insertions(+), 3 deletions(-) create mode 100644 drivers/gpu/drm/xe/instructions/xe_mfx_commands.h diff --git a/drivers/gpu/drm/xe/instructions/xe_instr_defs.h b/drivers/gpu/drm/xe/instructions/xe_instr_defs.h index fd2ce7ace510..e559969468c4 100644 --- a/drivers/gpu/drm/xe/instructions/xe_instr_defs.h +++ b/drivers/gpu/drm/xe/instructions/xe_instr_defs.h @@ -16,6 +16,7 @@ #define XE_INSTR_CMD_TYPE GENMASK(31, 29) #define XE_INSTR_MI REG_FIELD_PREP(XE_INSTR_CMD_TYPE, 0x0) #define XE_INSTR_GSC REG_FIELD_PREP(XE_INSTR_CMD_TYPE, 0x2) +#define XE_INSTR_VIDEOPIPE REG_FIELD_PREP(XE_INSTR_CMD_TYPE, 0x3) #define XE_INSTR_GFXPIPE REG_FIELD_PREP(XE_INSTR_CMD_TYPE, 0x3) #define XE_INSTR_GFX_STATE REG_FIELD_PREP(XE_INSTR_CMD_TYPE, 0x4) diff --git a/drivers/gpu/drm/xe/instructions/xe_mfx_commands.h b/drivers/gpu/drm/xe/instructions/xe_mfx_commands.h new file mode 100644 index 000000000000..3c0c97f78e90 --- /dev/null +++ b/drivers/gpu/drm/xe/instructions/xe_mfx_commands.h @@ -0,0 +1,28 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2024 Intel Corporation + */ + +#ifndef _XE_MFX_COMMANDS_H_ +#define _XE_MFX_COMMANDS_H_ + +#include "instructions/xe_instr_defs.h" + +#define MFX_CMD_SUBTYPE REG_GENMASK(28, 27) /* A.K.A cmd pipe */ +#define MFX_CMD_OPCODE REG_GENMASK(26, 24) +#define MFX_CMD_SUB_OPCODE REG_GENMASK(23, 16) +#define MFX_FLAGS_AND_LEN REG_GENMASK(15, 0) + +#define XE_MFX_INSTR(subtype, op, sub_op) \ + (XE_INSTR_VIDEOPIPE | \ + REG_FIELD_PREP(MFX_CMD_SUBTYPE, subtype) | \ + REG_FIELD_PREP(MFX_CMD_OPCODE, op) | \ + REG_FIELD_PREP(MFX_CMD_SUB_OPCODE, sub_op)) + +#define MFX_WAIT XE_MFX_INSTR(1, 0, 0) +#define MFX_WAIT_DW0_PXP_SYNC_CONTROL_FLAG REG_BIT(9) +#define MFX_WAIT_DW0_MFX_SYNC_CONTROL_FLAG REG_BIT(8) + +#define CRYPTO_KEY_EXCHANGE XE_MFX_INSTR(2, 6, 9) + +#endif diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h index 10ec2920d31b..167fb0f742de 100644 --- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h +++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h @@ -48,6 +48,7 @@ #define MI_LRI_LEN(x) (((x) & 0xff) + 1) #define MI_FLUSH_DW __MI_INSTR(0x26) +#define MI_FLUSH_DW_PROTECTED_MEM_EN REG_BIT(22) #define MI_FLUSH_DW_STORE_INDEX REG_BIT(21) #define MI_INVALIDATE_TLB REG_BIT(18) #define MI_FLUSH_DW_CCS REG_BIT(16) @@ -66,4 +67,8 @@ #define MI_BATCH_BUFFER_START __MI_INSTR(0x31) +#define MI_SET_APPID __MI_INSTR(0x0e) +#define MI_SET_APPID_SESSION_ID_MASK REG_GENMASK(6, 0) +#define MI_SET_APPID_SESSION_ID(x) REG_FIELD_PREP(MI_SET_APPID_SESSION_ID_MASK, x) + #endif diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h index 4206e6a8b50a..b27e80cd842a 100644 --- a/drivers/gpu/drm/xe/xe_lrc.h +++ b/drivers/gpu/drm/xe/xe_lrc.h @@ -39,7 +39,8 @@ struct xe_lrc_snapshot { u32 ctx_job_timestamp; }; -#define LRC_PPHWSP_SCRATCH_ADDR (0x34 * 4) +#define LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR (0x34 * 4) +#define LRC_PPHWSP_PXP_INVAL_SCRATCH_ADDR (0x40 * 4) struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm, u32 ring_size, u16 msix_vec); diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c index 52a6143b5ebe..326baea679a3 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.c +++ b/drivers/gpu/drm/xe/xe_pxp_submit.c @@ -5,15 +5,21 @@ #include "xe_pxp_submit.h" +#include #include #include "xe_device_types.h" +#include "xe_bb.h" #include "xe_bo.h" #include "xe_exec_queue.h" #include "xe_gsc_submit.h" #include "xe_gt.h" +#include "xe_lrc.h" #include "xe_pxp_types.h" +#include "xe_sched_job.h" #include "xe_vm.h" +#include "instructions/xe_mfx_commands.h" +#include "instructions/xe_mi_commands.h" /* * The VCS is used for kernel-owned GGTT submissions to issue key termination. @@ -197,3 +203,111 @@ void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp) destroy_gsc_client_resources(&pxp->gsc_res); destroy_vcs_execution_resources(pxp); } + +#define emit_cmd(xe_, map_, offset_, val_) \ + xe_map_wr(xe_, map_, (offset_) * sizeof(u32), u32, val_) + +/* stall until prior PXP and MFX/HCP/HUC objects are completed */ +#define MFX_WAIT_PXP (MFX_WAIT | \ + MFX_WAIT_DW0_PXP_SYNC_CONTROL_FLAG | \ + MFX_WAIT_DW0_MFX_SYNC_CONTROL_FLAG) +static u32 pxp_emit_wait(struct xe_device *xe, struct iosys_map *batch, u32 offset) +{ + /* wait for cmds to go through */ + emit_cmd(xe, batch, offset++, MFX_WAIT_PXP); + emit_cmd(xe, batch, offset++, 0); + + return offset; +} + +static u32 pxp_emit_session_selection(struct xe_device *xe, struct iosys_map *batch, + u32 offset, u32 idx) +{ + offset = pxp_emit_wait(xe, batch, offset); + + /* pxp off */ + emit_cmd(xe, batch, offset++, MI_FLUSH_DW | MI_FLUSH_IMM_DW); + emit_cmd(xe, batch, offset++, 0); + emit_cmd(xe, batch, offset++, 0); + emit_cmd(xe, batch, offset++, 0); + + /* select session */ + emit_cmd(xe, batch, offset++, MI_SET_APPID | MI_SET_APPID_SESSION_ID(idx)); + emit_cmd(xe, batch, offset++, 0); + + offset = pxp_emit_wait(xe, batch, offset); + + /* pxp on */ + emit_cmd(xe, batch, offset++, MI_FLUSH_DW | + MI_FLUSH_DW_PROTECTED_MEM_EN | + MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_DW_STORE_INDEX | + MI_FLUSH_IMM_DW); + emit_cmd(xe, batch, offset++, LRC_PPHWSP_PXP_INVAL_SCRATCH_ADDR | + MI_FLUSH_DW_USE_GTT); + emit_cmd(xe, batch, offset++, 0); + emit_cmd(xe, batch, offset++, 0); + + offset = pxp_emit_wait(xe, batch, offset); + + return offset; +} + +static u32 pxp_emit_inline_termination(struct xe_device *xe, + struct iosys_map *batch, u32 offset) +{ + /* session inline termination */ + emit_cmd(xe, batch, offset++, CRYPTO_KEY_EXCHANGE); + emit_cmd(xe, batch, offset++, 0); + + return offset; +} + +static u32 pxp_emit_session_termination(struct xe_device *xe, struct iosys_map *batch, + u32 offset, u32 idx) +{ + offset = pxp_emit_session_selection(xe, batch, offset, idx); + offset = pxp_emit_inline_termination(xe, batch, offset); + + return offset; +} + +/** + * xe_pxp_submit_session_termination - submits a PXP inline termination + * @pxp: the xe_pxp structure + * @id: the session to terminate + * + * Emit an inline termination via the VCS engine to terminate a session. + * + * Returns 0 if the submission is successful, an errno value otherwise. + */ +int xe_pxp_submit_session_termination(struct xe_pxp *pxp, u32 id) +{ + struct xe_sched_job *job; + struct dma_fence *fence; + long timeout; + u32 offset = 0; + u64 addr = xe_bo_ggtt_addr(pxp->vcs_exec.bo); + + offset = pxp_emit_session_termination(pxp->xe, &pxp->vcs_exec.bo->vmap, offset, id); + offset = pxp_emit_wait(pxp->xe, &pxp->vcs_exec.bo->vmap, offset); + emit_cmd(pxp->xe, &pxp->vcs_exec.bo->vmap, offset, MI_BATCH_BUFFER_END); + + job = xe_sched_job_create(pxp->vcs_exec.q, &addr); + if (IS_ERR(job)) + return PTR_ERR(job); + + xe_sched_job_arm(job); + fence = dma_fence_get(&job->drm.s_fence->finished); + xe_sched_job_push(job); + + timeout = dma_fence_wait_timeout(fence, false, HZ); + + dma_fence_put(fence); + + if (!timeout) + return -ETIMEDOUT; + else if (timeout < 0) + return timeout; + + return 0; +} diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.h b/drivers/gpu/drm/xe/xe_pxp_submit.h index fd21ac935be1..4ee8c0acfed9 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.h +++ b/drivers/gpu/drm/xe/xe_pxp_submit.h @@ -6,9 +6,13 @@ #ifndef __XE_PXP_SUBMIT_H__ #define __XE_PXP_SUBMIT_H__ +#include + struct xe_pxp; int xe_pxp_allocate_execution_resources(struct xe_pxp *pxp); void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp); +int xe_pxp_submit_session_termination(struct xe_pxp *pxp, u32 id); + #endif /* __XE_PXP_SUBMIT_H__ */ diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 9f327f27c072..0c230ee53bba 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -118,7 +118,7 @@ static int emit_flush_invalidate(u32 flag, u32 *dw, int i) dw[i++] |= MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_IMM_DW | MI_FLUSH_DW_STORE_INDEX; - dw[i++] = LRC_PPHWSP_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT; + dw[i++] = LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR | MI_FLUSH_DW_USE_GTT; dw[i++] = 0; dw[i++] = ~0U; @@ -156,7 +156,7 @@ static int emit_pipe_invalidate(u32 mask_flags, bool invalidate_tlb, u32 *dw, flags &= ~mask_flags; - return emit_pipe_control(dw, i, 0, flags, LRC_PPHWSP_SCRATCH_ADDR, 0); + return emit_pipe_control(dw, i, 0, flags, LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0); } static int emit_store_imm_ppgtt_posted(u64 addr, u64 value, From 96e84a2f5a5ba0efaaefb0dd5072e4b2e7f31f0e Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:28 -0800 Subject: [PATCH 0410/1443] drm/xe/pxp: Add GSC session invalidation support After a session is terminated, we need to inform the GSC so that it can clean up its side of the allocation. This is done by sending an invalidation command with the session ID. The invalidation will be triggered in response to a termination, interrupt, whose handling is coming in the next patch in the series. v2: Better comment and error messages (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-5-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h | 13 + drivers/gpu/drm/xe/xe_pxp_submit.c | 228 ++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp_submit.h | 3 + 3 files changed, 244 insertions(+) diff --git a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h index f3c4cf10ba20..f0da65ccdda2 100644 --- a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h +++ b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h @@ -31,6 +31,7 @@ enum pxp_status { PXP_STATUS_NOT_READY = 0x100e, PXP_STATUS_PLATFCONFIG_KF1_NOVERIF = 0x101a, PXP_STATUS_PLATFCONFIG_KF1_BAD = 0x101f, + PXP_STATUS_PLATFCONFIG_FIXED_KF1_NOT_SUPPORTED = 0x1037, PXP_STATUS_OP_NOT_PERMITTED = 0x4013 }; @@ -49,6 +50,7 @@ struct pxp_cmd_header { u32 buffer_len; } __packed; +#define PXP43_CMDID_INVALIDATE_STREAM_KEY 0x00000007 #define PXP43_CMDID_NEW_HUC_AUTH 0x0000003F /* MTL+ */ /* PXP-Input-Packet: HUC Auth-only */ @@ -63,4 +65,15 @@ struct pxp43_huc_auth_out { struct pxp_cmd_header header; } __packed; +/* PXP-Input-Packet: Invalidate Stream Key */ +struct pxp43_inv_stream_key_in { + struct pxp_cmd_header header; + u32 rsvd[3]; +} __packed; + +/* PXP-Output-Packet: Invalidate Stream Key */ +struct pxp43_inv_stream_key_out { + struct pxp_cmd_header header; + u32 rsvd; +} __packed; #endif diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c index 326baea679a3..f5f6b18c1054 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.c +++ b/drivers/gpu/drm/xe/xe_pxp_submit.c @@ -15,9 +15,13 @@ #include "xe_gsc_submit.h" #include "xe_gt.h" #include "xe_lrc.h" +#include "xe_map.h" #include "xe_pxp_types.h" #include "xe_sched_job.h" #include "xe_vm.h" +#include "abi/gsc_command_header_abi.h" +#include "abi/gsc_pxp_commands_abi.h" +#include "instructions/xe_gsc_commands.h" #include "instructions/xe_mfx_commands.h" #include "instructions/xe_mi_commands.h" @@ -311,3 +315,227 @@ int xe_pxp_submit_session_termination(struct xe_pxp *pxp, u32 id) return 0; } + +static bool +is_fw_err_platform_config(u32 type) +{ + switch (type) { + case PXP_STATUS_ERROR_API_VERSION: + case PXP_STATUS_PLATFCONFIG_KF1_NOVERIF: + case PXP_STATUS_PLATFCONFIG_KF1_BAD: + case PXP_STATUS_PLATFCONFIG_FIXED_KF1_NOT_SUPPORTED: + return true; + default: + break; + } + return false; +} + +static const char * +fw_err_to_string(u32 type) +{ + switch (type) { + case PXP_STATUS_ERROR_API_VERSION: + return "ERR_API_VERSION"; + case PXP_STATUS_NOT_READY: + return "ERR_NOT_READY"; + case PXP_STATUS_PLATFCONFIG_KF1_NOVERIF: + case PXP_STATUS_PLATFCONFIG_KF1_BAD: + case PXP_STATUS_PLATFCONFIG_FIXED_KF1_NOT_SUPPORTED: + return "ERR_PLATFORM_CONFIG"; + default: + break; + } + return NULL; +} + +static int pxp_pkt_submit(struct xe_exec_queue *q, u64 batch_addr) +{ + struct xe_gt *gt = q->gt; + struct xe_device *xe = gt_to_xe(gt); + struct xe_sched_job *job; + struct dma_fence *fence; + long timeout; + + xe_assert(xe, q->hwe->engine_id == XE_HW_ENGINE_GSCCS0); + + job = xe_sched_job_create(q, &batch_addr); + if (IS_ERR(job)) + return PTR_ERR(job); + + xe_sched_job_arm(job); + fence = dma_fence_get(&job->drm.s_fence->finished); + xe_sched_job_push(job); + + timeout = dma_fence_wait_timeout(fence, false, HZ); + dma_fence_put(fence); + if (timeout < 0) + return timeout; + else if (!timeout) + return -ETIME; + + return 0; +} + +static void emit_pxp_heci_cmd(struct xe_device *xe, struct iosys_map *batch, + u64 addr_in, u32 size_in, u64 addr_out, u32 size_out) +{ + u32 len = 0; + + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, GSC_HECI_CMD_PKT); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, lower_32_bits(addr_in)); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, upper_32_bits(addr_in)); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, size_in); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, lower_32_bits(addr_out)); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, upper_32_bits(addr_out)); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, size_out); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, 0); + xe_map_wr(xe, batch, len++ * sizeof(u32), u32, MI_BATCH_BUFFER_END); +} + +#define GSC_PENDING_RETRY_MAXCOUNT 40 +#define GSC_PENDING_RETRY_PAUSE_MS 50 +static int gsccs_send_message(struct xe_pxp_gsc_client_resources *gsc_res, + void *msg_in, size_t msg_in_size, + void *msg_out, size_t msg_out_size_max) +{ + struct xe_device *xe = gsc_res->vm->xe; + const size_t max_msg_size = gsc_res->inout_size - sizeof(struct intel_gsc_mtl_header); + u32 wr_offset; + u32 rd_offset; + u32 reply_size; + u32 min_reply_size = 0; + int ret; + int retry = GSC_PENDING_RETRY_MAXCOUNT; + + if (msg_in_size > max_msg_size || msg_out_size_max > max_msg_size) + return -ENOSPC; + + wr_offset = xe_gsc_emit_header(xe, &gsc_res->msg_in, 0, + HECI_MEADDRESS_PXP, + gsc_res->host_session_handle, + msg_in_size); + + /* NOTE: zero size packets are used for session-cleanups */ + if (msg_in && msg_in_size) { + xe_map_memcpy_to(xe, &gsc_res->msg_in, wr_offset, + msg_in, msg_in_size); + min_reply_size = sizeof(struct pxp_cmd_header); + } + + /* Make sure the reply header does not contain stale data */ + xe_gsc_poison_header(xe, &gsc_res->msg_out, 0); + + /* + * The BO is mapped at address 0 of the PPGTT, so no need to add its + * base offset when calculating the in/out addresses. + */ + emit_pxp_heci_cmd(xe, &gsc_res->batch, PXP_BB_SIZE, + wr_offset + msg_in_size, PXP_BB_SIZE + gsc_res->inout_size, + wr_offset + msg_out_size_max); + + xe_device_wmb(xe); + + /* + * If the GSC needs to communicate with CSME to complete our request, + * it'll set the "pending" flag in the return header. In this scenario + * we're expected to wait 50ms to give some time to the proxy code to + * handle the GSC<->CSME communication and then try again. Note that, + * although in most case the 50ms window is enough, the proxy flow is + * not actually guaranteed to complete within that time period, so we + * might have to try multiple times, up to a worst case of 2 seconds, + * after which the request is considered aborted. + */ + do { + ret = pxp_pkt_submit(gsc_res->q, 0); + if (ret) + break; + + if (xe_gsc_check_and_update_pending(xe, &gsc_res->msg_in, 0, + &gsc_res->msg_out, 0)) { + ret = -EAGAIN; + msleep(GSC_PENDING_RETRY_PAUSE_MS); + } + } while (--retry && ret == -EAGAIN); + + if (ret) { + drm_err(&xe->drm, "failed to submit GSC PXP message (%pe)\n", ERR_PTR(ret)); + return ret; + } + + ret = xe_gsc_read_out_header(xe, &gsc_res->msg_out, 0, + min_reply_size, &rd_offset); + if (ret) { + drm_err(&xe->drm, "invalid GSC reply for PXP (%pe)\n", ERR_PTR(ret)); + return ret; + } + + if (msg_out && min_reply_size) { + reply_size = xe_map_rd_field(xe, &gsc_res->msg_out, rd_offset, + struct pxp_cmd_header, buffer_len); + reply_size += sizeof(struct pxp_cmd_header); + + if (reply_size > msg_out_size_max) { + drm_warn(&xe->drm, "PXP reply size overflow: %u (%zu)\n", + reply_size, msg_out_size_max); + reply_size = msg_out_size_max; + } + + xe_map_memcpy_from(xe, msg_out, &gsc_res->msg_out, + rd_offset, reply_size); + } + + xe_gsc_poison_header(xe, &gsc_res->msg_in, 0); + + return ret; +} + +/** + * xe_pxp_submit_session_invalidation - submits a PXP GSC invalidation + * @gsc_res: the pxp client resources + * @id: the session to invalidate + * + * Submit a message to the GSC FW to notify it that a session has been + * terminated and is therefore invalid. + * + * Returns 0 if the submission is successful, an errno value otherwise. + */ +int xe_pxp_submit_session_invalidation(struct xe_pxp_gsc_client_resources *gsc_res, u32 id) +{ + struct xe_device *xe = gsc_res->vm->xe; + struct pxp43_inv_stream_key_in msg_in = {0}; + struct pxp43_inv_stream_key_out msg_out = {0}; + int ret = 0; + + /* + * Stream key invalidation reuses the same version 4.2 input/output + * command format but firmware requires 4.3 API interaction + */ + msg_in.header.api_version = PXP_APIVER(4, 3); + msg_in.header.command_id = PXP43_CMDID_INVALIDATE_STREAM_KEY; + msg_in.header.buffer_len = sizeof(msg_in) - sizeof(msg_in.header); + + msg_in.header.stream_id = FIELD_PREP(PXP_CMDHDR_EXTDATA_SESSION_VALID, 1); + msg_in.header.stream_id |= FIELD_PREP(PXP_CMDHDR_EXTDATA_APP_TYPE, 0); + msg_in.header.stream_id |= FIELD_PREP(PXP_CMDHDR_EXTDATA_SESSION_ID, id); + + ret = gsccs_send_message(gsc_res, &msg_in, sizeof(msg_in), + &msg_out, sizeof(msg_out)); + if (ret) { + drm_err(&xe->drm, "Failed to invalidate PXP stream-key %u (%pe)\n", + id, ERR_PTR(ret)); + } else if (msg_out.header.status != 0) { + ret = -EIO; + + if (is_fw_err_platform_config(msg_out.header.status)) + drm_info_once(&xe->drm, + "Failed to invalidate PXP stream-key %u: BIOS/SOC 0x%08x(%s)\n", + id, msg_out.header.status, + fw_err_to_string(msg_out.header.status)); + else + drm_dbg(&xe->drm, "Failed to invalidate stream-key %u, s=0x%08x\n", + id, msg_out.header.status); + } + + return ret; +} diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.h b/drivers/gpu/drm/xe/xe_pxp_submit.h index 4ee8c0acfed9..48fdc9b09116 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.h +++ b/drivers/gpu/drm/xe/xe_pxp_submit.h @@ -9,10 +9,13 @@ #include struct xe_pxp; +struct xe_pxp_gsc_client_resources; int xe_pxp_allocate_execution_resources(struct xe_pxp *pxp); void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp); int xe_pxp_submit_session_termination(struct xe_pxp *pxp, u32 id); +int xe_pxp_submit_session_invalidation(struct xe_pxp_gsc_client_resources *gsc_res, + u32 id); #endif /* __XE_PXP_SUBMIT_H__ */ From 3b506d73ec14977f3107ade94346cfb169c6f3b9 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:29 -0800 Subject: [PATCH 0411/1443] drm/xe/pxp: Handle the PXP termination interrupt When something happen to the session, the HW generates a termination interrupt. In reply to this, the driver is required to submit an inline session termination via the VCS, trigger the global termination and notify the GSC FW that the session is now invalid. v2: rename ARB define to make it cleaner to move it to uapi (John) v3: fix parameter name in documentation Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-6-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/regs/xe_irq_regs.h | 8 ++ drivers/gpu/drm/xe/regs/xe_pxp_regs.h | 6 ++ drivers/gpu/drm/xe/xe_irq.c | 20 +++- drivers/gpu/drm/xe/xe_pxp.c | 139 +++++++++++++++++++++++++- drivers/gpu/drm/xe/xe_pxp.h | 7 ++ drivers/gpu/drm/xe/xe_pxp_types.h | 13 +++ 6 files changed, 189 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_irq_regs.h b/drivers/gpu/drm/xe/regs/xe_irq_regs.h index 1776b3f78ccb..f0ecfcac4003 100644 --- a/drivers/gpu/drm/xe/regs/xe_irq_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_irq_regs.h @@ -44,6 +44,7 @@ #define ENGINE1_MASK REG_GENMASK(31, 16) #define ENGINE0_MASK REG_GENMASK(15, 0) #define GPM_WGBOXPERF_INTR_ENABLE XE_REG(0x19003c, XE_REG_OPTION_VF) +#define CRYPTO_RSVD_INTR_ENABLE XE_REG(0x190040) #define GUNIT_GSC_INTR_ENABLE XE_REG(0x190044, XE_REG_OPTION_VF) #define CCS_RSVD_INTR_ENABLE XE_REG(0x190048, XE_REG_OPTION_VF) @@ -54,6 +55,7 @@ #define INTR_ENGINE_INTR(x) REG_FIELD_GET(GENMASK(15, 0), x) #define OTHER_GUC_INSTANCE 0 #define OTHER_GSC_HECI2_INSTANCE 3 +#define OTHER_KCR_INSTANCE 4 #define OTHER_GSC_INSTANCE 6 #define IIR_REG_SELECTOR(x) XE_REG(0x190070 + ((x) * 4), XE_REG_OPTION_VF) @@ -65,6 +67,7 @@ #define HECI2_RSVD_INTR_MASK XE_REG(0x1900e4) #define GUC_SG_INTR_MASK XE_REG(0x1900e8, XE_REG_OPTION_VF) #define GPM_WGBOXPERF_INTR_MASK XE_REG(0x1900ec, XE_REG_OPTION_VF) +#define CRYPTO_RSVD_INTR_MASK XE_REG(0x1900f0) #define GUNIT_GSC_INTR_MASK XE_REG(0x1900f4, XE_REG_OPTION_VF) #define CCS0_CCS1_INTR_MASK XE_REG(0x190100) #define CCS2_CCS3_INTR_MASK XE_REG(0x190104) @@ -79,4 +82,9 @@ #define GT_CS_MASTER_ERROR_INTERRUPT REG_BIT(3) #define GT_RENDER_USER_INTERRUPT REG_BIT(0) +/* irqs for OTHER_KCR_INSTANCE */ +#define KCR_PXP_STATE_TERMINATED_INTERRUPT REG_BIT(1) +#define KCR_APP_TERMINATED_PER_FW_REQ_INTERRUPT REG_BIT(2) +#define KCR_PXP_STATE_RESET_COMPLETE_INTERRUPT REG_BIT(3) + #endif diff --git a/drivers/gpu/drm/xe/regs/xe_pxp_regs.h b/drivers/gpu/drm/xe/regs/xe_pxp_regs.h index d67cf210d23d..aa158938b42e 100644 --- a/drivers/gpu/drm/xe/regs/xe_pxp_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_pxp_regs.h @@ -14,4 +14,10 @@ #define KCR_INIT XE_REG(0x3860f0) #define KCR_INIT_ALLOW_DISPLAY_ME_WRITES REG_BIT(14) +/* KCR hwdrm session in play status 0-31 */ +#define KCR_SIP XE_REG(0x386260) + +/* PXP global terminate register for session termination */ +#define KCR_GLOBAL_TERMINATE XE_REG(0x3860f8) + #endif /* __XE_PXP_REGS_H__ */ diff --git a/drivers/gpu/drm/xe/xe_irq.c b/drivers/gpu/drm/xe/xe_irq.c index 32f5a67a917b..bf092e6391c7 100644 --- a/drivers/gpu/drm/xe/xe_irq.c +++ b/drivers/gpu/drm/xe/xe_irq.c @@ -20,6 +20,7 @@ #include "xe_hw_engine.h" #include "xe_memirq.h" #include "xe_mmio.h" +#include "xe_pxp.h" #include "xe_sriov.h" /* @@ -208,6 +209,15 @@ void xe_irq_enable_hwe(struct xe_gt *gt) } if (heci_mask) xe_mmio_write32(mmio, HECI2_RSVD_INTR_MASK, ~(heci_mask << 16)); + + if (xe_pxp_is_supported(xe)) { + u32 kcr_mask = KCR_PXP_STATE_TERMINATED_INTERRUPT | + KCR_APP_TERMINATED_PER_FW_REQ_INTERRUPT | + KCR_PXP_STATE_RESET_COMPLETE_INTERRUPT; + + xe_mmio_write32(mmio, CRYPTO_RSVD_INTR_ENABLE, kcr_mask << 16); + xe_mmio_write32(mmio, CRYPTO_RSVD_INTR_MASK, ~(kcr_mask << 16)); + } } } @@ -330,9 +340,15 @@ static void gt_irq_handler(struct xe_tile *tile, } if (class == XE_ENGINE_CLASS_OTHER) { - /* HECI GSCFI interrupts come from outside of GT */ + /* + * HECI GSCFI interrupts come from outside of GT. + * KCR irqs come from inside GT but are handled + * by the global PXP subsystem. + */ if (xe->info.has_heci_gscfi && instance == OTHER_GSC_INSTANCE) xe_heci_gsc_irq_handler(xe, intr_vec); + else if (instance == OTHER_KCR_INSTANCE) + xe_pxp_irq_handler(xe, intr_vec); else gt_other_irq_handler(engine_gt, instance, intr_vec); } @@ -510,6 +526,8 @@ static void gt_irq_reset(struct xe_tile *tile) xe_mmio_write32(mmio, GUNIT_GSC_INTR_ENABLE, 0); xe_mmio_write32(mmio, GUNIT_GSC_INTR_MASK, ~0); xe_mmio_write32(mmio, HECI2_RSVD_INTR_MASK, ~0); + xe_mmio_write32(mmio, CRYPTO_RSVD_INTR_ENABLE, 0); + xe_mmio_write32(mmio, CRYPTO_RSVD_INTR_MASK, ~0); } xe_mmio_write32(mmio, GPM_WGBOXPERF_INTR_ENABLE, 0); diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 89d71a69cdff..1452a4763ac2 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -12,9 +12,11 @@ #include "xe_gt.h" #include "xe_gt_types.h" #include "xe_mmio.h" +#include "xe_pm.h" #include "xe_pxp_submit.h" #include "xe_pxp_types.h" #include "xe_uc_fw.h" +#include "regs/xe_irq_regs.h" #include "regs/xe_pxp_regs.h" /** @@ -25,11 +27,132 @@ * integrated parts. */ -static bool pxp_is_supported(const struct xe_device *xe) +#define ARB_SESSION DRM_XE_PXP_HWDRM_DEFAULT_SESSION /* shorter define */ + +bool xe_pxp_is_supported(const struct xe_device *xe) { return xe->info.has_pxp && IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY); } +static bool pxp_is_enabled(const struct xe_pxp *pxp) +{ + return pxp; +} + +static int pxp_wait_for_session_state(struct xe_pxp *pxp, u32 id, bool in_play) +{ + struct xe_gt *gt = pxp->gt; + u32 mask = BIT(id); + + return xe_mmio_wait32(>->mmio, KCR_SIP, mask, in_play ? mask : 0, + 250, NULL, false); +} + +static void pxp_terminate(struct xe_pxp *pxp) +{ + int ret = 0; + struct xe_device *xe = pxp->xe; + struct xe_gt *gt = pxp->gt; + unsigned int fw_ref; + + drm_dbg(&xe->drm, "Terminating PXP\n"); + + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); + if (!xe_force_wake_ref_has_domain(fw_ref, XE_FW_GT)) { + ret = -EIO; + goto out; + } + + /* terminate the hw session */ + ret = xe_pxp_submit_session_termination(pxp, ARB_SESSION); + if (ret) + goto out; + + ret = pxp_wait_for_session_state(pxp, ARB_SESSION, false); + if (ret) + goto out; + + /* Trigger full HW cleanup */ + xe_mmio_write32(>->mmio, KCR_GLOBAL_TERMINATE, 1); + + /* now we can tell the GSC to clean up its own state */ + ret = xe_pxp_submit_session_invalidation(&pxp->gsc_res, ARB_SESSION); + +out: + xe_force_wake_put(gt_to_fw(gt), fw_ref); + + if (ret) + drm_err(&xe->drm, "PXP termination failed: %pe\n", ERR_PTR(ret)); +} + +static void pxp_terminate_complete(struct xe_pxp *pxp) +{ + /* TODO mark the session as ready to start */ +} + +static void pxp_irq_work(struct work_struct *work) +{ + struct xe_pxp *pxp = container_of(work, typeof(*pxp), irq.work); + struct xe_device *xe = pxp->xe; + u32 events = 0; + + spin_lock_irq(&xe->irq.lock); + events = pxp->irq.events; + pxp->irq.events = 0; + spin_unlock_irq(&xe->irq.lock); + + if (!events) + return; + + /* + * If we're processing a termination irq while suspending then don't + * bother, we're going to re-init everything on resume anyway. + */ + if ((events & PXP_TERMINATION_REQUEST) && !xe_pm_runtime_get_if_active(xe)) + return; + + if (events & PXP_TERMINATION_REQUEST) { + events &= ~PXP_TERMINATION_COMPLETE; + pxp_terminate(pxp); + } + + if (events & PXP_TERMINATION_COMPLETE) + pxp_terminate_complete(pxp); + + if (events & PXP_TERMINATION_REQUEST) + xe_pm_runtime_put(xe); +} + +/** + * xe_pxp_irq_handler - Handles PXP interrupts. + * @xe: the xe_device structure + * @iir: interrupt vector + */ +void xe_pxp_irq_handler(struct xe_device *xe, u16 iir) +{ + struct xe_pxp *pxp = xe->pxp; + + if (!pxp_is_enabled(pxp)) { + drm_err(&xe->drm, "PXP irq 0x%x received with PXP disabled!\n", iir); + return; + } + + lockdep_assert_held(&xe->irq.lock); + + if (unlikely(!iir)) + return; + + if (iir & (KCR_PXP_STATE_TERMINATED_INTERRUPT | + KCR_APP_TERMINATED_PER_FW_REQ_INTERRUPT)) + pxp->irq.events |= PXP_TERMINATION_REQUEST; + + if (iir & KCR_PXP_STATE_RESET_COMPLETE_INTERRUPT) + pxp->irq.events |= PXP_TERMINATION_COMPLETE; + + if (pxp->irq.events) + queue_work(pxp->irq.wq, &pxp->irq.work); +} + static int kcr_pxp_set_status(const struct xe_pxp *pxp, bool enable) { u32 val = enable ? _MASKED_BIT_ENABLE(KCR_INIT_ALLOW_DISPLAY_ME_WRITES) : @@ -60,6 +183,7 @@ static void pxp_fini(void *arg) { struct xe_pxp *pxp = arg; + destroy_workqueue(pxp->irq.wq); xe_pxp_destroy_execution_resources(pxp); /* no need to explicitly disable KCR since we're going to do an FLR */ @@ -83,7 +207,7 @@ int xe_pxp_init(struct xe_device *xe) struct xe_pxp *pxp; int err; - if (!pxp_is_supported(xe)) + if (!xe_pxp_is_supported(xe)) return -EOPNOTSUPP; /* we only support PXP on single tile devices with a media GT */ @@ -105,12 +229,19 @@ int xe_pxp_init(struct xe_device *xe) if (!pxp) return -ENOMEM; + INIT_WORK(&pxp->irq.work, pxp_irq_work); pxp->xe = xe; pxp->gt = gt; + pxp->irq.wq = alloc_ordered_workqueue("pxp-wq", 0); + if (!pxp->irq.wq) { + err = -ENOMEM; + goto out_free; + } + err = kcr_pxp_enable(pxp); if (err) - goto out_free; + goto out_wq; err = xe_pxp_allocate_execution_resources(pxp); if (err) @@ -122,6 +253,8 @@ int xe_pxp_init(struct xe_device *xe) out_kcr_disable: kcr_pxp_disable(pxp); +out_wq: + destroy_workqueue(pxp->irq.wq); out_free: drmm_kfree(&xe->drm, pxp); return err; diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index 00f5e688c0d9..39435c644dcd 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -6,8 +6,15 @@ #ifndef __XE_PXP_H__ #define __XE_PXP_H__ +#include + struct xe_device; +#define DRM_XE_PXP_HWDRM_DEFAULT_SESSION 0xF /* TODO: move to uapi */ + +bool xe_pxp_is_supported(const struct xe_device *xe); + int xe_pxp_init(struct xe_device *xe); +void xe_pxp_irq_handler(struct xe_device *xe, u16 iir); #endif /* __XE_PXP_H__ */ diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h index 0f86b50756cc..311d08111b5f 100644 --- a/drivers/gpu/drm/xe/xe_pxp_types.h +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -8,6 +8,7 @@ #include #include +#include struct xe_bo; struct xe_exec_queue; @@ -69,6 +70,18 @@ struct xe_pxp { /** @gsc_res: kernel-owned objects for PXP submissions to the GSCCS */ struct xe_pxp_gsc_client_resources gsc_res; + + /** @irq: wrapper for the worker and queue used for PXP irq support */ + struct { + /** @irq.work: worker that manages irq events. */ + struct work_struct work; + /** @irq.wq: workqueue on which to queue the irq work. */ + struct workqueue_struct *wq; + /** @irq.events: pending events, protected with xe->irq.lock. */ + u32 events; +#define PXP_TERMINATION_REQUEST BIT(0) +#define PXP_TERMINATION_COMPLETE BIT(1) + } irq; }; #endif /* __XE_PXP_TYPES_H__ */ From 0387d46ea7fd0496375b8668385db16ce6172ece Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:30 -0800 Subject: [PATCH 0412/1443] drm/xe/pxp: Add GSC session initialization support A session is initialized (i.e. started) by sending a message to the GSC. The initialization will be triggered when a user opts-in to using PXP; the interface for that is coming in a follow-up patch in the series. v2: clean up error messages, use new ARB define (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-7-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h | 21 +++++++++ drivers/gpu/drm/xe/xe_pxp_submit.c | 47 +++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp_submit.h | 1 + 3 files changed, 69 insertions(+) diff --git a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h index f0da65ccdda2..290e431cf10d 100644 --- a/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h +++ b/drivers/gpu/drm/xe/abi/gsc_pxp_commands_abi.h @@ -51,6 +51,7 @@ struct pxp_cmd_header { } __packed; #define PXP43_CMDID_INVALIDATE_STREAM_KEY 0x00000007 +#define PXP43_CMDID_INIT_SESSION 0x00000036 #define PXP43_CMDID_NEW_HUC_AUTH 0x0000003F /* MTL+ */ /* PXP-Input-Packet: HUC Auth-only */ @@ -65,6 +66,26 @@ struct pxp43_huc_auth_out { struct pxp_cmd_header header; } __packed; +/* PXP-Input-Packet: Init PXP session */ +struct pxp43_create_arb_in { + struct pxp_cmd_header header; + /* header.stream_id fields for vesion 4.3 of Init PXP session: */ + #define PXP43_INIT_SESSION_VALID BIT(0) + #define PXP43_INIT_SESSION_APPTYPE BIT(1) + #define PXP43_INIT_SESSION_APPID GENMASK(17, 2) + u32 protection_mode; + #define PXP43_INIT_SESSION_PROTECTION_ARB 0x2 + u32 sub_session_id; + u32 init_flags; + u32 rsvd[12]; +} __packed; + +/* PXP-Input-Packet: Init PXP session */ +struct pxp43_create_arb_out { + struct pxp_cmd_header header; + u32 rsvd[8]; +} __packed; + /* PXP-Input-Packet: Invalidate Stream Key */ struct pxp43_inv_stream_key_in { struct pxp_cmd_header header; diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c index f5f6b18c1054..b50fe037c74b 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.c +++ b/drivers/gpu/drm/xe/xe_pxp_submit.c @@ -16,6 +16,7 @@ #include "xe_gt.h" #include "xe_lrc.h" #include "xe_map.h" +#include "xe_pxp.h" #include "xe_pxp_types.h" #include "xe_sched_job.h" #include "xe_vm.h" @@ -490,6 +491,52 @@ static int gsccs_send_message(struct xe_pxp_gsc_client_resources *gsc_res, return ret; } +/** + * xe_pxp_submit_session_init - submits a PXP GSC session initialization + * @gsc_res: the pxp client resources + * @id: the session to initialize + * + * Submit a message to the GSC FW to initialize (i.e. start) a PXP session. + * + * Returns 0 if the submission is successful, an errno value otherwise. + */ +int xe_pxp_submit_session_init(struct xe_pxp_gsc_client_resources *gsc_res, u32 id) +{ + struct xe_device *xe = gsc_res->vm->xe; + struct pxp43_create_arb_in msg_in = {0}; + struct pxp43_create_arb_out msg_out = {0}; + int ret; + + msg_in.header.api_version = PXP_APIVER(4, 3); + msg_in.header.command_id = PXP43_CMDID_INIT_SESSION; + msg_in.header.stream_id = (FIELD_PREP(PXP43_INIT_SESSION_APPID, id) | + FIELD_PREP(PXP43_INIT_SESSION_VALID, 1) | + FIELD_PREP(PXP43_INIT_SESSION_APPTYPE, 0)); + msg_in.header.buffer_len = sizeof(msg_in) - sizeof(msg_in.header); + + if (id == DRM_XE_PXP_HWDRM_DEFAULT_SESSION) + msg_in.protection_mode = PXP43_INIT_SESSION_PROTECTION_ARB; + + ret = gsccs_send_message(gsc_res, &msg_in, sizeof(msg_in), + &msg_out, sizeof(msg_out)); + if (ret) { + drm_err(&xe->drm, "Failed to init PXP session %u (%pe)\n", id, ERR_PTR(ret)); + } else if (msg_out.header.status != 0) { + ret = -EIO; + + if (is_fw_err_platform_config(msg_out.header.status)) + drm_info_once(&xe->drm, + "Failed to init PXP session %u due to BIOS/SOC, s=0x%x(%s)\n", + id, msg_out.header.status, + fw_err_to_string(msg_out.header.status)); + else + drm_dbg(&xe->drm, "Failed to init PXP session %u, s=0x%x\n", + id, msg_out.header.status); + } + + return ret; +} + /** * xe_pxp_submit_session_invalidation - submits a PXP GSC invalidation * @gsc_res: the pxp client resources diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.h b/drivers/gpu/drm/xe/xe_pxp_submit.h index 48fdc9b09116..c9efda02f4b0 100644 --- a/drivers/gpu/drm/xe/xe_pxp_submit.h +++ b/drivers/gpu/drm/xe/xe_pxp_submit.h @@ -14,6 +14,7 @@ struct xe_pxp_gsc_client_resources; int xe_pxp_allocate_execution_resources(struct xe_pxp *pxp); void xe_pxp_destroy_execution_resources(struct xe_pxp *pxp); +int xe_pxp_submit_session_init(struct xe_pxp_gsc_client_resources *gsc_res, u32 id); int xe_pxp_submit_session_termination(struct xe_pxp *pxp, u32 id); int xe_pxp_submit_session_invalidation(struct xe_pxp_gsc_client_resources *gsc_res, u32 id); From f8caa80154c4b5481476d1aad8bef335b5f90e6d Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:31 -0800 Subject: [PATCH 0413/1443] drm/xe/pxp: Add PXP queue tracking and session start We expect every queue that uses PXP to be marked as doing so, to allow the driver to correctly manage the encryption status. The API for doing this from userspace is coming in the next patch, while this patch implement the management side of things. When a PXP queue is created, the driver will do the following: - Start the default PXP session if it is not already running; - assign an rpm ref to the queue to keep for its lifetime (this is required because PXP HWDRM sessions are killed by the HW suspend flow). Since PXP start and termination can race each other, this patch also introduces locking and a state machine to keep track of the pending operations. Note that since we'll need to take the lock from the suspend/resume paths as well, we can't do submissions while holding it, which means we need a slightly more complicated state machine to keep track of intermediate steps. v4: new patch in the series, split from the following interface patch to keep review manageable. Lock and status rework to not do submissions under lock. v5: Improve comments and error logs (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-8-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/xe_exec_queue.c | 1 + drivers/gpu/drm/xe/xe_exec_queue_types.h | 6 + drivers/gpu/drm/xe/xe_pxp.c | 383 ++++++++++++++++++++++- drivers/gpu/drm/xe/xe_pxp.h | 5 + drivers/gpu/drm/xe/xe_pxp_types.h | 30 ++ 5 files changed, 419 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 252bfa11cae9..2ec4e2eb6f2a 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -78,6 +78,7 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe, INIT_LIST_HEAD(&q->lr.link); INIT_LIST_HEAD(&q->multi_gt_link); INIT_LIST_HEAD(&q->hw_engine_group_link); + INIT_LIST_HEAD(&q->pxp.link); q->sched_props.timeslice_us = hwe->eclass->sched_props.timeslice_us; q->sched_props.preempt_timeout_us = diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h index 5af5419cec7a..6d85a069947f 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue_types.h +++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h @@ -130,6 +130,12 @@ struct xe_exec_queue { struct list_head link; } lr; + /** @pxp: PXP info tracking */ + struct { + /** @pxp.link: link into the list of PXP exec queues */ + struct list_head link; + } pxp; + /** @ops: submission backend exec queue operations */ const struct xe_exec_queue_ops *ops; diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 1452a4763ac2..69d49e34e34d 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -8,9 +8,13 @@ #include #include "xe_device_types.h" +#include "xe_exec_queue.h" #include "xe_force_wake.h" +#include "xe_guc_submit.h" +#include "xe_gsc_proxy.h" #include "xe_gt.h" #include "xe_gt_types.h" +#include "xe_huc.h" #include "xe_mmio.h" #include "xe_pm.h" #include "xe_pxp_submit.h" @@ -29,6 +33,15 @@ #define ARB_SESSION DRM_XE_PXP_HWDRM_DEFAULT_SESSION /* shorter define */ +/* + * A submission to GSC can take up to 250ms to complete, so use a 300ms + * timeout for activation where only one of those is involved. Termination + * additionally requires a submission to VCS and an interaction with KCR, so + * bump the timeout to 500ms for that. + */ +#define PXP_ACTIVATION_TIMEOUT_MS 300 +#define PXP_TERMINATION_TIMEOUT_MS 500 + bool xe_pxp_is_supported(const struct xe_device *xe) { return xe->info.has_pxp && IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY); @@ -39,6 +52,40 @@ static bool pxp_is_enabled(const struct xe_pxp *pxp) return pxp; } +static bool pxp_prerequisites_done(const struct xe_pxp *pxp) +{ + struct xe_gt *gt = pxp->gt; + unsigned int fw_ref; + bool ready; + + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); + + /* + * If force_wake fails we could falsely report the prerequisites as not + * done even if they are; the consequence of this would be that the + * callers won't go ahead with using PXP, but if force_wake doesn't work + * the GT is very likely in a bad state so not really a problem to abort + * PXP. Therefore, we can just log the force_wake error and not escalate + * it. + */ + XE_WARN_ON(!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)); + + /* PXP requires both HuC authentication via GSC and GSC proxy initialized */ + ready = xe_huc_is_authenticated(>->uc.huc, XE_HUC_AUTH_VIA_GSC) && + xe_gsc_proxy_init_done(>->uc.gsc); + + xe_force_wake_put(gt_to_fw(gt), fw_ref); + + return ready; +} + +static bool pxp_session_is_in_play(struct xe_pxp *pxp, u32 id) +{ + struct xe_gt *gt = pxp->gt; + + return xe_mmio_read32(>->mmio, KCR_SIP) & BIT(id); +} + static int pxp_wait_for_session_state(struct xe_pxp *pxp, u32 id, bool in_play) { struct xe_gt *gt = pxp->gt; @@ -48,14 +95,15 @@ static int pxp_wait_for_session_state(struct xe_pxp *pxp, u32 id, bool in_play) 250, NULL, false); } -static void pxp_terminate(struct xe_pxp *pxp) +static void pxp_invalidate_queues(struct xe_pxp *pxp); + +static int pxp_terminate_hw(struct xe_pxp *pxp) { - int ret = 0; - struct xe_device *xe = pxp->xe; struct xe_gt *gt = pxp->gt; unsigned int fw_ref; + int ret = 0; - drm_dbg(&xe->drm, "Terminating PXP\n"); + drm_dbg(&pxp->xe->drm, "Terminating PXP\n"); fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); if (!xe_force_wake_ref_has_domain(fw_ref, XE_FW_GT)) { @@ -80,14 +128,83 @@ static void pxp_terminate(struct xe_pxp *pxp) out: xe_force_wake_put(gt_to_fw(gt), fw_ref); + return ret; +} - if (ret) +static void mark_termination_in_progress(struct xe_pxp *pxp) +{ + lockdep_assert_held(&pxp->mutex); + + reinit_completion(&pxp->termination); + pxp->status = XE_PXP_TERMINATION_IN_PROGRESS; +} + +static void pxp_terminate(struct xe_pxp *pxp) +{ + int ret = 0; + struct xe_device *xe = pxp->xe; + + if (!wait_for_completion_timeout(&pxp->activation, + msecs_to_jiffies(PXP_ACTIVATION_TIMEOUT_MS))) + drm_err(&xe->drm, "failed to wait for PXP start before termination\n"); + + mutex_lock(&pxp->mutex); + + pxp_invalidate_queues(pxp); + + /* + * If we have a termination already in progress, we need to wait for + * it to complete before queueing another one. Once the first + * termination is completed we'll set the state back to + * NEEDS_TERMINATION and leave it to the pxp start code to issue it. + */ + if (pxp->status == XE_PXP_TERMINATION_IN_PROGRESS) { + pxp->status = XE_PXP_NEEDS_ADDITIONAL_TERMINATION; + mutex_unlock(&pxp->mutex); + return; + } + + mark_termination_in_progress(pxp); + + mutex_unlock(&pxp->mutex); + + ret = pxp_terminate_hw(pxp); + if (ret) { drm_err(&xe->drm, "PXP termination failed: %pe\n", ERR_PTR(ret)); + mutex_lock(&pxp->mutex); + pxp->status = XE_PXP_ERROR; + complete_all(&pxp->termination); + mutex_unlock(&pxp->mutex); + } } static void pxp_terminate_complete(struct xe_pxp *pxp) { - /* TODO mark the session as ready to start */ + /* + * We expect PXP to be in one of 2 states when we get here: + * - XE_PXP_TERMINATION_IN_PROGRESS: a single termination event was + * requested and it is now completing, so we're ready to start. + * - XE_PXP_NEEDS_ADDITIONAL_TERMINATION: a second termination was + * requested while the first one was still being processed. + */ + mutex_lock(&pxp->mutex); + + switch (pxp->status) { + case XE_PXP_TERMINATION_IN_PROGRESS: + pxp->status = XE_PXP_READY_TO_START; + break; + case XE_PXP_NEEDS_ADDITIONAL_TERMINATION: + pxp->status = XE_PXP_NEEDS_TERMINATION; + break; + default: + drm_err(&pxp->xe->drm, + "PXP termination complete while status was %u\n", + pxp->status); + } + + complete_all(&pxp->termination); + + mutex_unlock(&pxp->mutex); } static void pxp_irq_work(struct work_struct *work) @@ -229,10 +346,24 @@ int xe_pxp_init(struct xe_device *xe) if (!pxp) return -ENOMEM; + INIT_LIST_HEAD(&pxp->queues.list); + spin_lock_init(&pxp->queues.lock); INIT_WORK(&pxp->irq.work, pxp_irq_work); pxp->xe = xe; pxp->gt = gt; + /* + * we'll use the completions to check if there is an action pending, + * so we start them as completed and we reinit it when an action is + * triggered. + */ + init_completion(&pxp->activation); + init_completion(&pxp->termination); + complete_all(&pxp->termination); + complete_all(&pxp->activation); + + mutex_init(&pxp->mutex); + pxp->irq.wq = alloc_ordered_workqueue("pxp-wq", 0); if (!pxp->irq.wq) { err = -ENOMEM; @@ -259,3 +390,243 @@ int xe_pxp_init(struct xe_device *xe) drmm_kfree(&xe->drm, pxp); return err; } + +static int __pxp_start_arb_session(struct xe_pxp *pxp) +{ + int ret; + unsigned int fw_ref; + + fw_ref = xe_force_wake_get(gt_to_fw(pxp->gt), XE_FW_GT); + if (!xe_force_wake_ref_has_domain(fw_ref, XE_FW_GT)) + return -EIO; + + if (pxp_session_is_in_play(pxp, ARB_SESSION)) { + ret = -EEXIST; + goto out_force_wake; + } + + ret = xe_pxp_submit_session_init(&pxp->gsc_res, ARB_SESSION); + if (ret) { + drm_err(&pxp->xe->drm, "Failed to init PXP arb session: %pe\n", ERR_PTR(ret)); + goto out_force_wake; + } + + ret = pxp_wait_for_session_state(pxp, ARB_SESSION, true); + if (ret) { + drm_err(&pxp->xe->drm, "PXP ARB session failed to go in play%pe\n", ERR_PTR(ret)); + goto out_force_wake; + } + + drm_dbg(&pxp->xe->drm, "PXP ARB session is active\n"); + +out_force_wake: + xe_force_wake_put(gt_to_fw(pxp->gt), fw_ref); + return ret; +} + +static void __exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q) +{ + spin_lock_irq(&pxp->queues.lock); + list_add_tail(&q->pxp.link, &pxp->queues.list); + spin_unlock_irq(&pxp->queues.lock); +} + +/** + * xe_pxp_exec_queue_add - add a queue to the PXP list + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @q: the queue to add to the list + * + * If PXP is enabled and the prerequisites are done, start the PXP ARB + * session (if not already running) and add the queue to the PXP list. Note + * that the queue must have previously been marked as using PXP with + * xe_pxp_exec_queue_set_type. + * + * Returns 0 if the PXP ARB session is running and the queue is in the list, + * -ENODEV if PXP is disabled, -EBUSY if the PXP prerequisites are not done, + * other errno value if something goes wrong during the session start. + */ +int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q) +{ + int ret = 0; + + if (!pxp_is_enabled(pxp)) + return -ENODEV; + + /* + * Runtime suspend kills PXP, so we take a reference to prevent it from + * happening while we have active queues that use PXP + */ + xe_pm_runtime_get(pxp->xe); + + if (!pxp_prerequisites_done(pxp)) { + ret = -EBUSY; + goto out; + } + +wait_for_idle: + /* + * if there is an action in progress, wait for it. We need to wait + * outside the lock because the completion is done from within the lock. + * Note that the two action should never be pending at the same time. + */ + if (!wait_for_completion_timeout(&pxp->termination, + msecs_to_jiffies(PXP_TERMINATION_TIMEOUT_MS))) { + ret = -ETIMEDOUT; + goto out; + } + + if (!wait_for_completion_timeout(&pxp->activation, + msecs_to_jiffies(PXP_ACTIVATION_TIMEOUT_MS))) { + ret = -ETIMEDOUT; + goto out; + } + + mutex_lock(&pxp->mutex); + + /* If PXP is not already active, turn it on */ + switch (pxp->status) { + case XE_PXP_ERROR: + ret = -EIO; + break; + case XE_PXP_ACTIVE: + __exec_queue_add(pxp, q); + mutex_unlock(&pxp->mutex); + goto out; + case XE_PXP_READY_TO_START: + pxp->status = XE_PXP_START_IN_PROGRESS; + reinit_completion(&pxp->activation); + break; + case XE_PXP_START_IN_PROGRESS: + /* If a start is in progress then the completion must not be done */ + XE_WARN_ON(completion_done(&pxp->activation)); + mutex_unlock(&pxp->mutex); + goto wait_for_idle; + case XE_PXP_NEEDS_TERMINATION: + mark_termination_in_progress(pxp); + break; + case XE_PXP_TERMINATION_IN_PROGRESS: + case XE_PXP_NEEDS_ADDITIONAL_TERMINATION: + /* If a termination is in progress then the completion must not be done */ + XE_WARN_ON(completion_done(&pxp->termination)); + mutex_unlock(&pxp->mutex); + goto wait_for_idle; + default: + drm_err(&pxp->xe->drm, "unexpected state during PXP start: %u\n", pxp->status); + ret = -EIO; + break; + } + + mutex_unlock(&pxp->mutex); + + if (ret) + goto out; + + if (!completion_done(&pxp->termination)) { + ret = pxp_terminate_hw(pxp); + if (ret) { + drm_err(&pxp->xe->drm, "PXP termination failed before start\n"); + mutex_lock(&pxp->mutex); + pxp->status = XE_PXP_ERROR; + mutex_unlock(&pxp->mutex); + + goto out; + } + + goto wait_for_idle; + } + + /* All the cases except for start should have exited earlier */ + XE_WARN_ON(completion_done(&pxp->activation)); + ret = __pxp_start_arb_session(pxp); + + mutex_lock(&pxp->mutex); + + complete_all(&pxp->activation); + + /* + * Any other process should wait until the state goes away from + * XE_PXP_START_IN_PROGRESS, so if the state is not that something went + * wrong. Mark the status as needing termination and try again. + */ + if (pxp->status != XE_PXP_START_IN_PROGRESS) { + drm_err(&pxp->xe->drm, "unexpected state after PXP start: %u\n", pxp->status); + pxp->status = XE_PXP_NEEDS_TERMINATION; + mutex_unlock(&pxp->mutex); + goto wait_for_idle; + } + + /* If everything went ok, update the status and add the queue to the list */ + if (!ret) { + pxp->status = XE_PXP_ACTIVE; + __exec_queue_add(pxp, q); + } else { + pxp->status = XE_PXP_ERROR; + } + + mutex_unlock(&pxp->mutex); + +out: + /* + * in the successful case the PM ref is released from + * xe_pxp_exec_queue_remove + */ + if (ret) + xe_pm_runtime_put(pxp->xe); + + return ret; +} + +/** + * xe_pxp_exec_queue_remove - remove a queue from the PXP list + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @q: the queue to remove from the list + * + * If PXP is enabled and the exec_queue is in the list, the queue will be + * removed from the list and its PM reference will be released. It is safe to + * call this function multiple times for the same queue. + */ +void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q) +{ + bool need_pm_put = false; + + if (!pxp_is_enabled(pxp)) + return; + + spin_lock_irq(&pxp->queues.lock); + + if (!list_empty(&q->pxp.link)) { + list_del_init(&q->pxp.link); + need_pm_put = true; + } + + spin_unlock_irq(&pxp->queues.lock); + + if (need_pm_put) + xe_pm_runtime_put(pxp->xe); +} + +static void pxp_invalidate_queues(struct xe_pxp *pxp) +{ + struct xe_exec_queue *tmp, *q; + + spin_lock_irq(&pxp->queues.lock); + + /* + * Removing a queue from the PXP list requires a put of the RPM ref that + * the queue holds to keep the PXP session alive, which can't be done + * under spinlock. Since it is safe to kill a queue multiple times, we + * can leave the invalid queue in the list for now and postpone the + * removal and associated RPM put to when the queue is destroyed. + */ + list_for_each_entry(tmp, &pxp->queues.list, pxp.link) { + q = xe_exec_queue_get_unless_zero(tmp); + + if (!q) + continue; + + xe_exec_queue_kill(q); + xe_exec_queue_put(q); + } + + spin_unlock_irq(&pxp->queues.lock); +} diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index 39435c644dcd..f482567c27b5 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -9,6 +9,8 @@ #include struct xe_device; +struct xe_exec_queue; +struct xe_pxp; #define DRM_XE_PXP_HWDRM_DEFAULT_SESSION 0xF /* TODO: move to uapi */ @@ -17,4 +19,7 @@ bool xe_pxp_is_supported(const struct xe_device *xe); int xe_pxp_init(struct xe_device *xe); void xe_pxp_irq_handler(struct xe_device *xe, u16 iir); +int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q); +void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q); + #endif /* __XE_PXP_H__ */ diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h index 311d08111b5f..bd741720f67d 100644 --- a/drivers/gpu/drm/xe/xe_pxp_types.h +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -6,7 +6,10 @@ #ifndef __XE_PXP_TYPES_H__ #define __XE_PXP_TYPES_H__ +#include #include +#include +#include #include #include @@ -16,6 +19,16 @@ struct xe_device; struct xe_gt; struct xe_vm; +enum xe_pxp_status { + XE_PXP_ERROR = -1, + XE_PXP_NEEDS_TERMINATION = 0, /* starting status */ + XE_PXP_NEEDS_ADDITIONAL_TERMINATION, + XE_PXP_TERMINATION_IN_PROGRESS, + XE_PXP_READY_TO_START, + XE_PXP_START_IN_PROGRESS, + XE_PXP_ACTIVE, +}; + /** * struct xe_pxp_gsc_client_resources - resources for GSC submission by a PXP * client. The GSC FW supports multiple GSC client active at the same time. @@ -82,6 +95,23 @@ struct xe_pxp { #define PXP_TERMINATION_REQUEST BIT(0) #define PXP_TERMINATION_COMPLETE BIT(1) } irq; + + /** @mutex: protects the pxp status and the queue list */ + struct mutex mutex; + /** @status: the current pxp status */ + enum xe_pxp_status status; + /** @activation: completion struct that tracks pxp start */ + struct completion activation; + /** @termination: completion struct that tracks terminations */ + struct completion termination; + + /** @queues: management of exec_queues that use PXP */ + struct { + /** @queues.lock: spinlock protecting the queue management */ + spinlock_t lock; + /** @queues.list: list of exec_queues that use PXP */ + struct list_head list; + } queues; }; #endif /* __XE_PXP_TYPES_H__ */ From 72d479601d67026c4fafaad21762a777cf41f906 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:32 -0800 Subject: [PATCH 0414/1443] drm/xe/pxp/uapi: Add userspace and LRC support for PXP-using queues Userspace is required to mark a queue as using PXP to guarantee that the PXP instructions will work. In addition to managing the PXP sessions, when a PXP queue is created the driver will set the relevant bits in its context control register. On submission of a valid PXP queue, the driver will validate all encrypted objects mapped to the VM to ensured they were encrypted with the current key. v2: Remove pxp_types include outside of PXP code (Jani), better comments and code cleanup (John) v3: split the internal PXP management to a separate patch for ease of review. re-order ioctl checks to always return -EINVAL if parameters are invalid, rebase on msix changes. Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-9-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/regs/xe_engine_regs.h | 1 + drivers/gpu/drm/xe/xe_exec_queue.c | 56 +++++++++++++++++++++++- drivers/gpu/drm/xe/xe_exec_queue.h | 5 +++ drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 + drivers/gpu/drm/xe/xe_execlist.c | 2 +- drivers/gpu/drm/xe/xe_lrc.c | 18 ++++++-- drivers/gpu/drm/xe/xe_lrc.h | 4 +- drivers/gpu/drm/xe/xe_pxp.c | 35 +++++++++++++-- drivers/gpu/drm/xe/xe_pxp.h | 4 +- include/uapi/drm/xe_drm.h | 40 ++++++++++++++++- 10 files changed, 153 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h index d86219dedde2..c8fd3d5ca502 100644 --- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h @@ -132,6 +132,7 @@ #define RING_EXECLIST_STATUS_HI(base) XE_REG((base) + 0x234 + 4) #define RING_CONTEXT_CONTROL(base) XE_REG((base) + 0x244, XE_REG_OPTION_MASKED) +#define CTX_CTRL_PXP_ENABLE REG_BIT(10) #define CTX_CTRL_OAC_CONTEXT_ENABLE REG_BIT(8) #define CTX_CTRL_RUN_ALONE REG_BIT(7) #define CTX_CTRL_INDIRECT_RING_STATE_ENABLE REG_BIT(4) diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 2ec4e2eb6f2a..6051db78d706 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -25,6 +25,7 @@ #include "xe_ring_ops_types.h" #include "xe_trace.h" #include "xe_vm.h" +#include "xe_pxp.h" enum xe_exec_queue_sched_prop { XE_EXEC_QUEUE_JOB_TIMEOUT = 0, @@ -38,6 +39,8 @@ static int exec_queue_user_extensions(struct xe_device *xe, struct xe_exec_queue static void __xe_exec_queue_free(struct xe_exec_queue *q) { + if (xe_exec_queue_uses_pxp(q)) + xe_pxp_exec_queue_remove(gt_to_xe(q->gt)->pxp, q); if (q->vm) xe_vm_put(q->vm); @@ -113,6 +116,21 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q) { struct xe_vm *vm = q->vm; int i, err; + u32 flags = 0; + + /* + * PXP workloads executing on RCS or CCS must run in isolation (i.e. no + * other workload can use the EUs at the same time). On MTL this is done + * by setting the RUNALONE bit in the LRC, while starting on Xe2 there + * is a dedicated bit for it. + */ + if (xe_exec_queue_uses_pxp(q) && + (q->class == XE_ENGINE_CLASS_RENDER || q->class == XE_ENGINE_CLASS_COMPUTE)) { + if (GRAPHICS_VER(gt_to_xe(q->gt)) >= 20) + flags |= XE_LRC_CREATE_PXP; + else + flags |= XE_LRC_CREATE_RUNALONE; + } if (vm) { err = xe_vm_lock(vm, true); @@ -121,7 +139,7 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q) } for (i = 0; i < q->width; ++i) { - q->lrc[i] = xe_lrc_create(q->hwe, q->vm, SZ_16K, q->msix_vec); + q->lrc[i] = xe_lrc_create(q->hwe, q->vm, SZ_16K, q->msix_vec, flags); if (IS_ERR(q->lrc[i])) { err = PTR_ERR(q->lrc[i]); goto err_unlock; @@ -166,6 +184,19 @@ struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *v if (err) goto err_post_alloc; + /* + * We can only add the queue to the PXP list after the init is complete, + * because the PXP termination can call exec_queue_kill and that will + * go bad if the queue is only half-initialized. This means that we + * can't do it when we handle the PXP extension in __xe_exec_queue_alloc + * and we need to do it here instead. + */ + if (xe_exec_queue_uses_pxp(q)) { + err = xe_pxp_exec_queue_add(xe->pxp, q); + if (err) + goto err_post_alloc; + } + return q; err_post_alloc: @@ -254,6 +285,9 @@ void xe_exec_queue_destroy(struct kref *ref) struct xe_exec_queue *q = container_of(ref, struct xe_exec_queue, refcount); struct xe_exec_queue *eq, *next; + if (xe_exec_queue_uses_pxp(q)) + xe_pxp_exec_queue_remove(gt_to_xe(q->gt)->pxp, q); + xe_exec_queue_last_fence_put_unlocked(q); if (!(q->flags & EXEC_QUEUE_FLAG_BIND_ENGINE_CHILD)) { list_for_each_entry_safe(eq, next, &q->multi_gt_list, @@ -409,6 +443,22 @@ static int exec_queue_set_timeslice(struct xe_device *xe, struct xe_exec_queue * return 0; } +static int +exec_queue_set_pxp_type(struct xe_device *xe, struct xe_exec_queue *q, u64 value) +{ + if (value == DRM_XE_PXP_TYPE_NONE) + return 0; + + /* we only support HWDRM sessions right now */ + if (XE_IOCTL_DBG(xe, value != DRM_XE_PXP_TYPE_HWDRM)) + return -EINVAL; + + if (!xe_pxp_is_enabled(xe->pxp)) + return -ENODEV; + + return xe_pxp_exec_queue_set_type(xe->pxp, q, DRM_XE_PXP_TYPE_HWDRM); +} + typedef int (*xe_exec_queue_set_property_fn)(struct xe_device *xe, struct xe_exec_queue *q, u64 value); @@ -416,6 +466,7 @@ typedef int (*xe_exec_queue_set_property_fn)(struct xe_device *xe, static const xe_exec_queue_set_property_fn exec_queue_set_property_funcs[] = { [DRM_XE_EXEC_QUEUE_SET_PROPERTY_PRIORITY] = exec_queue_set_priority, [DRM_XE_EXEC_QUEUE_SET_PROPERTY_TIMESLICE] = exec_queue_set_timeslice, + [DRM_XE_EXEC_QUEUE_SET_PROPERTY_PXP_TYPE] = exec_queue_set_pxp_type, }; static int exec_queue_user_ext_set_property(struct xe_device *xe, @@ -435,7 +486,8 @@ static int exec_queue_user_ext_set_property(struct xe_device *xe, ARRAY_SIZE(exec_queue_set_property_funcs)) || XE_IOCTL_DBG(xe, ext.pad) || XE_IOCTL_DBG(xe, ext.property != DRM_XE_EXEC_QUEUE_SET_PROPERTY_PRIORITY && - ext.property != DRM_XE_EXEC_QUEUE_SET_PROPERTY_TIMESLICE)) + ext.property != DRM_XE_EXEC_QUEUE_SET_PROPERTY_TIMESLICE && + ext.property != DRM_XE_EXEC_QUEUE_SET_PROPERTY_PXP_TYPE)) return -EINVAL; idx = array_index_nospec(ext.property, ARRAY_SIZE(exec_queue_set_property_funcs)); diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h index 90c7f73eab88..17bc50a7f05a 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.h +++ b/drivers/gpu/drm/xe/xe_exec_queue.h @@ -57,6 +57,11 @@ static inline bool xe_exec_queue_is_parallel(struct xe_exec_queue *q) return q->width > 1; } +static inline bool xe_exec_queue_uses_pxp(struct xe_exec_queue *q) +{ + return q->pxp.type; +} + bool xe_exec_queue_is_lr(struct xe_exec_queue *q); bool xe_exec_queue_ring_full(struct xe_exec_queue *q); diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h index 6d85a069947f..6eb7ff091534 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue_types.h +++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h @@ -132,6 +132,8 @@ struct xe_exec_queue { /** @pxp: PXP info tracking */ struct { + /** @pxp.type: PXP session type used by this queue */ + u8 type; /** @pxp.link: link into the list of PXP exec queues */ struct list_head link; } pxp; diff --git a/drivers/gpu/drm/xe/xe_execlist.c b/drivers/gpu/drm/xe/xe_execlist.c index 5ef96deaa881..779a52daf3d7 100644 --- a/drivers/gpu/drm/xe/xe_execlist.c +++ b/drivers/gpu/drm/xe/xe_execlist.c @@ -269,7 +269,7 @@ struct xe_execlist_port *xe_execlist_port_create(struct xe_device *xe, port->hwe = hwe; - port->lrc = xe_lrc_create(hwe, NULL, SZ_16K, XE_IRQ_DEFAULT_MSIX); + port->lrc = xe_lrc_create(hwe, NULL, SZ_16K, XE_IRQ_DEFAULT_MSIX, 0); if (IS_ERR(port->lrc)) { err = PTR_ERR(port->lrc); goto err; diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index bbb9ffbf6367..df3ceddede07 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -883,7 +883,8 @@ static void xe_lrc_finish(struct xe_lrc *lrc) #define PVC_CTX_ACC_CTR_THOLD (0x2a + 1) static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, - struct xe_vm *vm, u32 ring_size, u16 msix_vec) + struct xe_vm *vm, u32 ring_size, u16 msix_vec, + u32 init_flags) { struct xe_gt *gt = hwe->gt; struct xe_tile *tile = gt_to_tile(gt); @@ -979,6 +980,16 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, RING_CTL_SIZE(lrc->ring.size) | RING_VALID); } + if (init_flags & XE_LRC_CREATE_RUNALONE) + xe_lrc_write_ctx_reg(lrc, CTX_CONTEXT_CONTROL, + xe_lrc_read_ctx_reg(lrc, CTX_CONTEXT_CONTROL) | + _MASKED_BIT_ENABLE(CTX_CTRL_RUN_ALONE)); + + if (init_flags & XE_LRC_CREATE_PXP) + xe_lrc_write_ctx_reg(lrc, CTX_CONTEXT_CONTROL, + xe_lrc_read_ctx_reg(lrc, CTX_CONTEXT_CONTROL) | + _MASKED_BIT_ENABLE(CTX_CTRL_PXP_ENABLE)); + xe_lrc_write_ctx_reg(lrc, CTX_TIMESTAMP, 0); if (xe->info.has_asid && vm) @@ -1021,6 +1032,7 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, * @vm: The VM (address space) * @ring_size: LRC ring size * @msix_vec: MSI-X interrupt vector (for platforms that support it) + * @flags: LRC initialization flags * * Allocate and initialize the Logical Ring Context (LRC). * @@ -1028,7 +1040,7 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, * upon failure. */ struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm, - u32 ring_size, u16 msix_vec) + u32 ring_size, u16 msix_vec, u32 flags) { struct xe_lrc *lrc; int err; @@ -1037,7 +1049,7 @@ struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm, if (!lrc) return ERR_PTR(-ENOMEM); - err = xe_lrc_init(lrc, hwe, vm, ring_size, msix_vec); + err = xe_lrc_init(lrc, hwe, vm, ring_size, msix_vec, flags); if (err) { kfree(lrc); return ERR_PTR(err); diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h index b27e80cd842a..0b40f349ab95 100644 --- a/drivers/gpu/drm/xe/xe_lrc.h +++ b/drivers/gpu/drm/xe/xe_lrc.h @@ -42,8 +42,10 @@ struct xe_lrc_snapshot { #define LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR (0x34 * 4) #define LRC_PPHWSP_PXP_INVAL_SCRATCH_ADDR (0x40 * 4) +#define XE_LRC_CREATE_RUNALONE 0x1 +#define XE_LRC_CREATE_PXP 0x2 struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm, - u32 ring_size, u16 msix_vec); + u32 ring_size, u16 msix_vec, u32 flags); void xe_lrc_destroy(struct kref *ref); /** diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 69d49e34e34d..b32121273e0d 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -6,6 +6,7 @@ #include "xe_pxp.h" #include +#include #include "xe_device_types.h" #include "xe_exec_queue.h" @@ -47,7 +48,7 @@ bool xe_pxp_is_supported(const struct xe_device *xe) return xe->info.has_pxp && IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY); } -static bool pxp_is_enabled(const struct xe_pxp *pxp) +bool xe_pxp_is_enabled(const struct xe_pxp *pxp) { return pxp; } @@ -249,7 +250,7 @@ void xe_pxp_irq_handler(struct xe_device *xe, u16 iir) { struct xe_pxp *pxp = xe->pxp; - if (!pxp_is_enabled(pxp)) { + if (!xe_pxp_is_enabled(pxp)) { drm_err(&xe->drm, "PXP irq 0x%x received with PXP disabled!\n", iir); return; } @@ -424,6 +425,27 @@ static int __pxp_start_arb_session(struct xe_pxp *pxp) return ret; } +/** + * xe_pxp_exec_queue_set_type - Mark a queue as using PXP + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @q: the queue to mark as using PXP + * @type: the type of PXP session this queue will use + * + * Returns 0 if the selected PXP type is supported, -ENODEV otherwise. + */ +int xe_pxp_exec_queue_set_type(struct xe_pxp *pxp, struct xe_exec_queue *q, u8 type) +{ + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + /* we only support HWDRM sessions right now */ + xe_assert(pxp->xe, type == DRM_XE_PXP_TYPE_HWDRM); + + q->pxp.type = type; + + return 0; +} + static void __exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q) { spin_lock_irq(&pxp->queues.lock); @@ -449,9 +471,12 @@ int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q) { int ret = 0; - if (!pxp_is_enabled(pxp)) + if (!xe_pxp_is_enabled(pxp)) return -ENODEV; + /* we only support HWDRM sessions right now */ + xe_assert(pxp->xe, q->pxp.type == DRM_XE_PXP_TYPE_HWDRM); + /* * Runtime suspend kills PXP, so we take a reference to prevent it from * happening while we have active queues that use PXP @@ -589,7 +614,7 @@ void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q) { bool need_pm_put = false; - if (!pxp_is_enabled(pxp)) + if (!xe_pxp_is_enabled(pxp)) return; spin_lock_irq(&pxp->queues.lock); @@ -599,6 +624,8 @@ void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q) need_pm_put = true; } + q->pxp.type = DRM_XE_PXP_TYPE_NONE; + spin_unlock_irq(&pxp->queues.lock); if (need_pm_put) diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index f482567c27b5..2e0ab186072a 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -12,13 +12,13 @@ struct xe_device; struct xe_exec_queue; struct xe_pxp; -#define DRM_XE_PXP_HWDRM_DEFAULT_SESSION 0xF /* TODO: move to uapi */ - bool xe_pxp_is_supported(const struct xe_device *xe); +bool xe_pxp_is_enabled(const struct xe_pxp *pxp); int xe_pxp_init(struct xe_device *xe); void xe_pxp_irq_handler(struct xe_device *xe, u16 iir); +int xe_pxp_exec_queue_set_type(struct xe_pxp *pxp, struct xe_exec_queue *q, u8 type); int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q); void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q); diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index e2160330ad01..9d53834c4c0a 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -1114,6 +1114,24 @@ struct drm_xe_vm_bind { /** * struct drm_xe_exec_queue_create - Input of &DRM_IOCTL_XE_EXEC_QUEUE_CREATE * + * This ioctl supports setting the following properties via the + * %DRM_XE_EXEC_QUEUE_EXTENSION_SET_PROPERTY extension, which uses the + * generic @drm_xe_ext_set_property struct: + * + * - %DRM_XE_EXEC_QUEUE_SET_PROPERTY_PRIORITY - set the queue priority. + * CAP_SYS_NICE is required to set a value above normal. + * - %DRM_XE_EXEC_QUEUE_SET_PROPERTY_TIMESLICE - set the queue timeslice + * duration in microseconds. + * - %DRM_XE_EXEC_QUEUE_SET_PROPERTY_PXP_TYPE - set the type of PXP session + * this queue will be used with. Valid values are listed in enum + * drm_xe_pxp_session_type. %DRM_XE_PXP_TYPE_NONE is the default behavior, so + * there is no need to explicitly set that. When a queue of type + * %DRM_XE_PXP_TYPE_HWDRM is created, the PXP default HWDRM session + * (%XE_PXP_HWDRM_DEFAULT_SESSION) will be started, if isn't already running. + * Given that going into a power-saving state kills PXP HWDRM sessions, + * runtime PM will be blocked while queues of this type are alive. + * All PXP queues will be killed if a PXP invalidation event occurs. + * * The example below shows how to use @drm_xe_exec_queue_create to create * a simple exec_queue (no parallel submission) of class * &DRM_XE_ENGINE_CLASS_RENDER. @@ -1137,7 +1155,7 @@ struct drm_xe_exec_queue_create { #define DRM_XE_EXEC_QUEUE_EXTENSION_SET_PROPERTY 0 #define DRM_XE_EXEC_QUEUE_SET_PROPERTY_PRIORITY 0 #define DRM_XE_EXEC_QUEUE_SET_PROPERTY_TIMESLICE 1 - +#define DRM_XE_EXEC_QUEUE_SET_PROPERTY_PXP_TYPE 2 /** @extensions: Pointer to the first extension struct, if any */ __u64 extensions; @@ -1756,6 +1774,26 @@ struct drm_xe_oa_stream_info { __u64 reserved[3]; }; +/** + * enum drm_xe_pxp_session_type - Supported PXP session types. + * + * We currently only support HWDRM sessions, which are used for protected + * content that ends up being displayed, but the HW supports multiple types, so + * we might extend support in the future. + */ +enum drm_xe_pxp_session_type { + /** @DRM_XE_PXP_TYPE_NONE: PXP not used */ + DRM_XE_PXP_TYPE_NONE = 0, + /** + * @DRM_XE_PXP_TYPE_HWDRM: HWDRM sessions are used for content that ends + * up on the display. + */ + DRM_XE_PXP_TYPE_HWDRM = 1, +}; + +/* ID of the protected content session managed by Xe when PXP is active */ +#define DRM_XE_PXP_HWDRM_DEFAULT_SESSION 0xf + #if defined(__cplusplus) } #endif From bd98ac2e05855ea781c9b7ad30b5e1a234aefe95 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:33 -0800 Subject: [PATCH 0415/1443] drm/xe/pxp/uapi: Add a query for PXP status MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit PXP prerequisites (SW proxy and HuC auth via GSC) are completed asynchronously from driver load, which means that userspace can start submitting before we're ready to start a PXP session. Therefore, we need a query that userspace can use to check not only if PXP is supported but also to wait until the prerequisites are done. v2: Improve doc, do not report TYPE_NONE as supported (José) v3: Better comments, remove unneeded copy_from_user (John) Signed-off-by: Daniele Ceraolo Spurio Cc: José Roberto de Souza Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-10-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/xe_pxp.c | 32 ++++++++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp.h | 1 + drivers/gpu/drm/xe/xe_query.c | 29 +++++++++++++++++++++++++++++ include/uapi/drm/xe_drm.h | 35 +++++++++++++++++++++++++++++++++++ 4 files changed, 97 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index b32121273e0d..24aef5c0f04a 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -80,6 +80,38 @@ static bool pxp_prerequisites_done(const struct xe_pxp *pxp) return ready; } +/** + * xe_pxp_get_readiness_status - check whether PXP is ready for userspace use + * @pxp: the xe_pxp pointer (can be NULL if PXP is disabled) + * + * Returns: 0 if PXP is not ready yet, 1 if it is ready, a negative errno value + * if PXP is not supported/enabled or if something went wrong in the + * initialization of the prerequisites. Note that the return values of this + * function follow the uapi (see drm_xe_query_pxp_status), so they can be used + * directly in the query ioctl. + */ +int xe_pxp_get_readiness_status(struct xe_pxp *pxp) +{ + int ret = 0; + + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + /* if the GSC or HuC FW are in an error state, PXP will never work */ + if (xe_uc_fw_status_to_error(pxp->gt->uc.huc.fw.status) || + xe_uc_fw_status_to_error(pxp->gt->uc.gsc.fw.status)) + return -EIO; + + xe_pm_runtime_get(pxp->xe); + + /* PXP requires both HuC loaded and GSC proxy initialized */ + if (pxp_prerequisites_done(pxp)) + ret = 1; + + xe_pm_runtime_put(pxp->xe); + return ret; +} + static bool pxp_session_is_in_play(struct xe_pxp *pxp, u32 id) { struct xe_gt *gt = pxp->gt; diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index 2e0ab186072a..868813cc84b9 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -14,6 +14,7 @@ struct xe_pxp; bool xe_pxp_is_supported(const struct xe_device *xe); bool xe_pxp_is_enabled(const struct xe_pxp *pxp); +int xe_pxp_get_readiness_status(struct xe_pxp *pxp); int xe_pxp_init(struct xe_device *xe); void xe_pxp_irq_handler(struct xe_device *xe, u16 iir); diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c index c059639613f7..042f87a688e7 100644 --- a/drivers/gpu/drm/xe/xe_query.c +++ b/drivers/gpu/drm/xe/xe_query.c @@ -24,6 +24,7 @@ #include "xe_macros.h" #include "xe_mmio.h" #include "xe_oa.h" +#include "xe_pxp.h" #include "xe_ttm_vram_mgr.h" #include "xe_wa.h" @@ -698,6 +699,33 @@ static int query_oa_units(struct xe_device *xe, return ret ? -EFAULT : 0; } +static int query_pxp_status(struct xe_device *xe, struct drm_xe_device_query *query) +{ + struct drm_xe_query_pxp_status __user *query_ptr = u64_to_user_ptr(query->data); + size_t size = sizeof(struct drm_xe_query_pxp_status); + struct drm_xe_query_pxp_status resp = { 0 }; + int ret; + + if (query->size == 0) { + query->size = size; + return 0; + } else if (XE_IOCTL_DBG(xe, query->size != size)) { + return -EINVAL; + } + + ret = xe_pxp_get_readiness_status(xe->pxp); + if (ret < 0) + return ret; + + resp.status = ret; + resp.supported_session_types = BIT(DRM_XE_PXP_TYPE_HWDRM); + + if (copy_to_user(query_ptr, &resp, size)) + return -EFAULT; + + return 0; +} + static int (* const xe_query_funcs[])(struct xe_device *xe, struct drm_xe_device_query *query) = { query_engines, @@ -709,6 +737,7 @@ static int (* const xe_query_funcs[])(struct xe_device *xe, query_engine_cycles, query_uc_fw_version, query_oa_units, + query_pxp_status, }; int xe_query_ioctl(struct drm_device *dev, void *data, struct drm_file *file) diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index 9d53834c4c0a..112fd27f3c75 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -629,6 +629,39 @@ struct drm_xe_query_uc_fw_version { __u64 reserved; }; +/** + * struct drm_xe_query_pxp_status - query if PXP is ready + * + * If PXP is enabled and no fatal error has occurred, the status will be set to + * one of the following values: + * 0: PXP init still in progress + * 1: PXP init complete + * + * If PXP is not enabled or something has gone wrong, the query will be failed + * with one of the following error codes: + * -ENODEV: PXP not supported or disabled; + * -EIO: fatal error occurred during init, so PXP will never be enabled; + * -EINVAL: incorrect value provided as part of the query; + * -EFAULT: error copying the memory between kernel and userspace. + * + * The status can only be 0 in the first few seconds after driver load. If + * everything works as expected, the status will transition to init complete in + * less than 1 second, while in case of errors the driver might take longer to + * start returning an error code, but it should still take less than 10 seconds. + * + * The supported session type bitmask is based on the values in + * enum drm_xe_pxp_session_type. TYPE_NONE is always supported and therefore + * is not reported in the bitmask. + * + */ +struct drm_xe_query_pxp_status { + /** @status: current PXP status */ + __u32 status; + + /** @supported_session_types: bitmask of supported PXP session types */ + __u32 supported_session_types; +}; + /** * struct drm_xe_device_query - Input of &DRM_IOCTL_XE_DEVICE_QUERY - main * structure to query device information @@ -648,6 +681,7 @@ struct drm_xe_query_uc_fw_version { * attributes. * - %DRM_XE_DEVICE_QUERY_GT_TOPOLOGY * - %DRM_XE_DEVICE_QUERY_ENGINE_CYCLES + * - %DRM_XE_DEVICE_QUERY_PXP_STATUS * * If size is set to 0, the driver fills it with the required size for * the requested type of data to query. If size is equal to the required @@ -700,6 +734,7 @@ struct drm_xe_device_query { #define DRM_XE_DEVICE_QUERY_ENGINE_CYCLES 6 #define DRM_XE_DEVICE_QUERY_UC_FW_VERSION 7 #define DRM_XE_DEVICE_QUERY_OA_UNITS 8 +#define DRM_XE_DEVICE_QUERY_PXP_STATUS 9 /** @query: The type of data to query */ __u32 query; From 41a97c4a12947c2786a1680d6839bb72d1c57cec Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:34 -0800 Subject: [PATCH 0416/1443] drm/xe/pxp/uapi: Add API to mark a BO as using PXP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The driver needs to know if a BO is encrypted with PXP to enable the display decryption at flip time. Furthermore, we want to keep track of the status of the encryption and reject any operation that involves a BO that is encrypted using an old key. There are two points in time where such checks can kick in: 1 - at VM bind time, all operations except for unmapping will be rejected if the key used to encrypt the BO is no longer valid. This check is opt-in via a new VM_BIND flag, to avoid a scenario where a malicious app purposely shares an invalid BO with a non-PXP aware app (such as a compositor). If the VM_BIND was failed, the compositor would be unable to display anything at all. Allowing the bind to go through means that output still works, it just displays garbage data within the bounds of the illegal BO. 2 - at job submission time, if the queue is marked as using PXP, all objects bound to the VM will be checked and the submission will be rejected if any of them was encrypted with a key that is no longer valid. Note that there is no risk of leaking the encrypted data if a user does not opt-in to those checks; the only consequence is that the user will not realize that the encryption key is changed and that the data is no longer valid. v2: Better commnnts and descriptions (John), rebase v3: Properly return the result of key_assign up the stack, do not use xe_bo in display headers (Jani) v4: improve key_instance variable documentation (John) Signed-off-by: Daniele Ceraolo Spurio Cc: Matthew Brost Cc: Thomas Hellström Cc: John Harrison Cc: Jani Nikula Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-11-daniele.ceraolospurio@intel.com --- .../xe/compat-i915-headers/pxp/intel_pxp.h | 12 ++- drivers/gpu/drm/xe/display/intel_bo.c | 2 +- drivers/gpu/drm/xe/xe_bo.c | 98 ++++++++++++++++++- drivers/gpu/drm/xe/xe_bo.h | 5 + drivers/gpu/drm/xe/xe_bo_types.h | 6 ++ drivers/gpu/drm/xe/xe_exec.c | 6 ++ drivers/gpu/drm/xe/xe_pxp.c | 90 +++++++++++++++++ drivers/gpu/drm/xe/xe_pxp.h | 6 ++ drivers/gpu/drm/xe/xe_pxp_types.h | 11 +++ drivers/gpu/drm/xe/xe_vm.c | 46 ++++++++- drivers/gpu/drm/xe/xe_vm.h | 2 + include/uapi/drm/xe_drm.h | 19 ++++ 12 files changed, 296 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h b/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h index 419e8e926f00..d2eb8e1f6c4b 100644 --- a/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h +++ b/drivers/gpu/drm/xe/compat-i915-headers/pxp/intel_pxp.h @@ -9,6 +9,8 @@ #include #include +#include "xe_pxp.h" + struct drm_gem_object; struct xe_pxp; @@ -16,7 +18,15 @@ static inline int intel_pxp_key_check(struct xe_pxp *pxp, struct drm_gem_object *obj, bool assign) { - return -ENODEV; + /* + * The assign variable is used in i915 to assign the key to the BO at + * first submission time. In Xe the key is instead assigned at BO + * creation time, so the assign variable must always be false. + */ + if (assign) + return -EINVAL; + + return xe_pxp_obj_key_check(pxp, obj); } #endif diff --git a/drivers/gpu/drm/xe/display/intel_bo.c b/drivers/gpu/drm/xe/display/intel_bo.c index b463f5bd4eed..27437c22bd70 100644 --- a/drivers/gpu/drm/xe/display/intel_bo.c +++ b/drivers/gpu/drm/xe/display/intel_bo.c @@ -25,7 +25,7 @@ bool intel_bo_is_shmem(struct drm_gem_object *obj) bool intel_bo_is_protected(struct drm_gem_object *obj) { - return false; + return xe_bo_is_protected(gem_to_xe_bo(obj)); } void intel_bo_flush_if_display(struct drm_gem_object *obj) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index c32201123d44..6812164e1470 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -6,6 +6,7 @@ #include "xe_bo.h" #include +#include #include #include @@ -26,6 +27,7 @@ #include "xe_migrate.h" #include "xe_pm.h" #include "xe_preempt_fence.h" +#include "xe_pxp.h" #include "xe_res_cursor.h" #include "xe_trace_bo.h" #include "xe_ttm_stolen_mgr.h" @@ -2155,6 +2157,93 @@ void xe_bo_vunmap(struct xe_bo *bo) __xe_bo_vunmap(bo); } +static int gem_create_set_pxp_type(struct xe_device *xe, struct xe_bo *bo, u64 value) +{ + if (value == DRM_XE_PXP_TYPE_NONE) + return 0; + + /* we only support DRM_XE_PXP_TYPE_HWDRM for now */ + if (XE_IOCTL_DBG(xe, value != DRM_XE_PXP_TYPE_HWDRM)) + return -EINVAL; + + return xe_pxp_key_assign(xe->pxp, bo); +} + +typedef int (*xe_gem_create_set_property_fn)(struct xe_device *xe, + struct xe_bo *bo, + u64 value); + +static const xe_gem_create_set_property_fn gem_create_set_property_funcs[] = { + [DRM_XE_GEM_CREATE_EXTENSION_SET_PROPERTY] = gem_create_set_pxp_type, +}; + +static int gem_create_user_ext_set_property(struct xe_device *xe, + struct xe_bo *bo, + u64 extension) +{ + u64 __user *address = u64_to_user_ptr(extension); + struct drm_xe_ext_set_property ext; + int err; + u32 idx; + + err = __copy_from_user(&ext, address, sizeof(ext)); + if (XE_IOCTL_DBG(xe, err)) + return -EFAULT; + + if (XE_IOCTL_DBG(xe, ext.property >= + ARRAY_SIZE(gem_create_set_property_funcs)) || + XE_IOCTL_DBG(xe, ext.pad) || + XE_IOCTL_DBG(xe, ext.property != DRM_XE_GEM_CREATE_EXTENSION_SET_PROPERTY)) + return -EINVAL; + + idx = array_index_nospec(ext.property, ARRAY_SIZE(gem_create_set_property_funcs)); + if (!gem_create_set_property_funcs[idx]) + return -EINVAL; + + return gem_create_set_property_funcs[idx](xe, bo, ext.value); +} + +typedef int (*xe_gem_create_user_extension_fn)(struct xe_device *xe, + struct xe_bo *bo, + u64 extension); + +static const xe_gem_create_user_extension_fn gem_create_user_extension_funcs[] = { + [DRM_XE_GEM_CREATE_EXTENSION_SET_PROPERTY] = gem_create_user_ext_set_property, +}; + +#define MAX_USER_EXTENSIONS 16 +static int gem_create_user_extensions(struct xe_device *xe, struct xe_bo *bo, + u64 extensions, int ext_number) +{ + u64 __user *address = u64_to_user_ptr(extensions); + struct drm_xe_user_extension ext; + int err; + u32 idx; + + if (XE_IOCTL_DBG(xe, ext_number >= MAX_USER_EXTENSIONS)) + return -E2BIG; + + err = __copy_from_user(&ext, address, sizeof(ext)); + if (XE_IOCTL_DBG(xe, err)) + return -EFAULT; + + if (XE_IOCTL_DBG(xe, ext.pad) || + XE_IOCTL_DBG(xe, ext.name >= ARRAY_SIZE(gem_create_user_extension_funcs))) + return -EINVAL; + + idx = array_index_nospec(ext.name, + ARRAY_SIZE(gem_create_user_extension_funcs)); + err = gem_create_user_extension_funcs[idx](xe, bo, extensions); + if (XE_IOCTL_DBG(xe, err)) + return err; + + if (ext.next_extension) + return gem_create_user_extensions(xe, bo, ext.next_extension, + ++ext_number); + + return 0; +} + int xe_gem_create_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { @@ -2167,8 +2256,7 @@ int xe_gem_create_ioctl(struct drm_device *dev, void *data, u32 handle; int err; - if (XE_IOCTL_DBG(xe, args->extensions) || - XE_IOCTL_DBG(xe, args->pad[0] || args->pad[1] || args->pad[2]) || + if (XE_IOCTL_DBG(xe, args->pad[0] || args->pad[1] || args->pad[2]) || XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1])) return -EINVAL; @@ -2250,6 +2338,12 @@ int xe_gem_create_ioctl(struct drm_device *dev, void *data, goto out_vm; } + if (args->extensions) { + err = gem_create_user_extensions(xe, bo, args->extensions, 0); + if (err) + goto out_bulk; + } + err = drm_gem_handle_create(file, &bo->ttm.base, &handle); if (err) goto out_bulk; diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h index 04995c5ced32..f09b9315721b 100644 --- a/drivers/gpu/drm/xe/xe_bo.h +++ b/drivers/gpu/drm/xe/xe_bo.h @@ -186,6 +186,11 @@ static inline bool xe_bo_is_pinned(struct xe_bo *bo) return bo->ttm.pin_count; } +static inline bool xe_bo_is_protected(const struct xe_bo *bo) +{ + return bo->pxp_key_instance; +} + static inline void xe_bo_unpin_map_no_vm(struct xe_bo *bo) { if (likely(bo)) { diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h index 46dc9e4e3e46..60c522866500 100644 --- a/drivers/gpu/drm/xe/xe_bo_types.h +++ b/drivers/gpu/drm/xe/xe_bo_types.h @@ -57,6 +57,12 @@ struct xe_bo { */ struct list_head client_link; #endif + /** + * @pxp_key_instance: PXP key instance this BO was created against. A + * 0 in this variable indicates that the BO does not use PXP encryption. + */ + u32 pxp_key_instance; + /** @freed: List node for delayed put. */ struct llist_node freed; /** @update_index: Update index if PT BO */ diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c index df8ce550deb4..b75adfc99fb7 100644 --- a/drivers/gpu/drm/xe/xe_exec.c +++ b/drivers/gpu/drm/xe/xe_exec.c @@ -262,6 +262,12 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file) goto err_exec; } + if (xe_exec_queue_uses_pxp(q)) { + err = xe_vm_validate_protected(q->vm); + if (err) + goto err_exec; + } + job = xe_sched_job_create(q, xe_exec_queue_is_parallel(q) ? addresses : &args->address); if (IS_ERR(job)) { diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 24aef5c0f04a..8060b4050be8 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -8,6 +8,8 @@ #include #include +#include "xe_bo.h" +#include "xe_bo_types.h" #include "xe_device_types.h" #include "xe_exec_queue.h" #include "xe_force_wake.h" @@ -185,6 +187,9 @@ static void pxp_terminate(struct xe_pxp *pxp) pxp_invalidate_queues(pxp); + if (pxp->status == XE_PXP_ACTIVE) + pxp->key_instance++; + /* * If we have a termination already in progress, we need to wait for * it to complete before queueing another one. Once the first @@ -385,6 +390,8 @@ int xe_pxp_init(struct xe_device *xe) pxp->xe = xe; pxp->gt = gt; + pxp->key_instance = 1; + /* * we'll use the completions to check if there is an action pending, * so we start them as completed and we reinit it when an action is @@ -689,3 +696,86 @@ static void pxp_invalidate_queues(struct xe_pxp *pxp) spin_unlock_irq(&pxp->queues.lock); } + +/** + * xe_pxp_key_assign - mark a BO as using the current PXP key iteration + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @bo: the BO to mark + * + * Returns: -ENODEV if PXP is disabled, 0 otherwise. + */ +int xe_pxp_key_assign(struct xe_pxp *pxp, struct xe_bo *bo) +{ + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + xe_assert(pxp->xe, !bo->pxp_key_instance); + + /* + * Note that the PXP key handling is inherently racey, because the key + * can theoretically change at any time (although it's unlikely to do + * so without triggers), even right after we copy it. Taking a lock + * wouldn't help because the value might still change as soon as we + * release the lock. + * Userspace needs to handle the fact that their BOs can go invalid at + * any point. + */ + bo->pxp_key_instance = pxp->key_instance; + + return 0; +} + +/** + * xe_pxp_bo_key_check - check if the key used by a xe_bo is valid + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @bo: the BO we want to check + * + * Checks whether a BO was encrypted with the current key or an obsolete one. + * + * Returns: 0 if the key is valid, -ENODEV if PXP is disabled, -EINVAL if the + * BO is not using PXP, -ENOEXEC if the key is not valid. + */ +int xe_pxp_bo_key_check(struct xe_pxp *pxp, struct xe_bo *bo) +{ + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + if (!xe_bo_is_protected(bo)) + return -EINVAL; + + xe_assert(pxp->xe, bo->pxp_key_instance); + + /* + * Note that the PXP key handling is inherently racey, because the key + * can theoretically change at any time (although it's unlikely to do + * so without triggers), even right after we check it. Taking a lock + * wouldn't help because the value might still change as soon as we + * release the lock. + * We mitigate the risk by checking the key at multiple points (on each + * submission involving the BO and right before flipping it on the + * display), but there is still a very small chance that we could + * operate on an invalid BO for a single submission or a single frame + * flip. This is a compromise made to protect the encrypted data (which + * is what the key termination is for). + */ + if (bo->pxp_key_instance != pxp->key_instance) + return -ENOEXEC; + + return 0; +} + +/** + * xe_pxp_obj_key_check - check if the key used by a drm_gem_obj is valid + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * @obj: the drm_gem_obj we want to check + * + * Checks whether a drm_gem_obj was encrypted with the current key or an + * obsolete one. + * + * Returns: 0 if the key is valid, -ENODEV if PXP is disabled, -EINVAL if the + * obj is not using PXP, -ENOEXEC if the key is not valid. + */ +int xe_pxp_obj_key_check(struct xe_pxp *pxp, struct drm_gem_object *obj) +{ + return xe_pxp_bo_key_check(pxp, gem_to_xe_bo(obj)); +} diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index 868813cc84b9..3dd70eac9da6 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -8,6 +8,8 @@ #include +struct drm_gem_object; +struct xe_bo; struct xe_device; struct xe_exec_queue; struct xe_pxp; @@ -23,4 +25,8 @@ int xe_pxp_exec_queue_set_type(struct xe_pxp *pxp, struct xe_exec_queue *q, u8 t int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q); void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q); +int xe_pxp_key_assign(struct xe_pxp *pxp, struct xe_bo *bo); +int xe_pxp_bo_key_check(struct xe_pxp *pxp, struct xe_bo *bo); +int xe_pxp_obj_key_check(struct xe_pxp *pxp, struct drm_gem_object *obj); + #endif /* __XE_PXP_H__ */ diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h index bd741720f67d..8e4569f0173d 100644 --- a/drivers/gpu/drm/xe/xe_pxp_types.h +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -112,6 +112,17 @@ struct xe_pxp { /** @queues.list: list of exec_queues that use PXP */ struct list_head list; } queues; + + /** + * @key_instance: keep track of the current iteration of the PXP key. + * Note that, due to the time needed for PXP termination and re-start + * to complete, the minimum time between 2 subsequent increases of this + * variable is 50ms, and even that only if there is a continuous attack; + * normal behavior is for this to increase much much slower than that. + * This means that we don't expect this to ever wrap and don't implement + * that case in the code. + */ + u32 key_instance; }; #endif /* __XE_PXP_TYPES_H__ */ diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index b9270d059e18..d664f2e418b2 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -33,6 +33,7 @@ #include "xe_pm.h" #include "xe_preempt_fence.h" #include "xe_pt.h" +#include "xe_pxp.h" #include "xe_res_cursor.h" #include "xe_sync.h" #include "xe_trace_bo.h" @@ -2726,7 +2727,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO); (DRM_XE_VM_BIND_FLAG_READONLY | \ DRM_XE_VM_BIND_FLAG_IMMEDIATE | \ DRM_XE_VM_BIND_FLAG_NULL | \ - DRM_XE_VM_BIND_FLAG_DUMPABLE) + DRM_XE_VM_BIND_FLAG_DUMPABLE | \ + DRM_XE_VM_BIND_FLAG_CHECK_PXP) #ifdef TEST_VM_OPS_ERROR #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR) @@ -2889,7 +2891,7 @@ static void xe_vma_ops_init(struct xe_vma_ops *vops, struct xe_vm *vm, static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo, u64 addr, u64 range, u64 obj_offset, - u16 pat_index) + u16 pat_index, u32 op, u32 bind_flags) { u16 coh_mode; @@ -2933,6 +2935,12 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo, return -EINVAL; } + /* If a BO is protected it can only be mapped if the key is still valid */ + if ((bind_flags & DRM_XE_VM_BIND_FLAG_CHECK_PXP) && xe_bo_is_protected(bo) && + op != DRM_XE_VM_BIND_OP_UNMAP && op != DRM_XE_VM_BIND_OP_UNMAP_ALL) + if (XE_IOCTL_DBG(xe, xe_pxp_bo_key_check(xe->pxp, bo) != 0)) + return -ENOEXEC; + return 0; } @@ -3022,6 +3030,8 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) u32 obj = bind_ops[i].obj; u64 obj_offset = bind_ops[i].obj_offset; u16 pat_index = bind_ops[i].pat_index; + u32 op = bind_ops[i].op; + u32 bind_flags = bind_ops[i].flags; if (!obj) continue; @@ -3034,7 +3044,8 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) bos[i] = gem_to_xe_bo(gem_obj); err = xe_vm_bind_ioctl_validate_bo(xe, bos[i], addr, range, - obj_offset, pat_index); + obj_offset, pat_index, op, + bind_flags); if (err) goto put_obj; } @@ -3334,6 +3345,35 @@ int xe_vm_invalidate_vma(struct xe_vma *vma) return ret; } +int xe_vm_validate_protected(struct xe_vm *vm) +{ + struct drm_gpuva *gpuva; + int err = 0; + + if (!vm) + return -ENODEV; + + mutex_lock(&vm->snap_mutex); + + drm_gpuvm_for_each_va(gpuva, &vm->gpuvm) { + struct xe_vma *vma = gpuva_to_vma(gpuva); + struct xe_bo *bo = vma->gpuva.gem.obj ? + gem_to_xe_bo(vma->gpuva.gem.obj) : NULL; + + if (!bo) + continue; + + if (xe_bo_is_protected(bo)) { + err = xe_pxp_bo_key_check(vm->xe->pxp, bo); + if (err) + break; + } + } + + mutex_unlock(&vm->snap_mutex); + return err; +} + struct xe_vm_snapshot { unsigned long num_snaps; struct { diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h index 0a2fa6c0815b..f66075f8a6fe 100644 --- a/drivers/gpu/drm/xe/xe_vm.h +++ b/drivers/gpu/drm/xe/xe_vm.h @@ -215,6 +215,8 @@ struct dma_fence *xe_vma_rebind(struct xe_vm *vm, struct xe_vma *vma, int xe_vm_invalidate_vma(struct xe_vma *vma); +int xe_vm_validate_protected(struct xe_vm *vm); + static inline void xe_vm_queue_rebind_worker(struct xe_vm *vm) { xe_assert(vm->xe, xe_vm_in_preempt_fence_mode(vm)); diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index 112fd27f3c75..892f54d3aa09 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -778,8 +778,23 @@ struct drm_xe_device_query { * - %DRM_XE_GEM_CPU_CACHING_WC - Allocate the pages as write-combined. This * is uncached. Scanout surfaces should likely use this. All objects * that can be placed in VRAM must use this. + * + * This ioctl supports setting the following properties via the + * %DRM_XE_GEM_CREATE_EXTENSION_SET_PROPERTY extension, which uses the + * generic @drm_xe_ext_set_property struct: + * + * - %DRM_XE_GEM_CREATE_SET_PROPERTY_PXP_TYPE - set the type of PXP session + * this object will be used with. Valid values are listed in enum + * drm_xe_pxp_session_type. %DRM_XE_PXP_TYPE_NONE is the default behavior, so + * there is no need to explicitly set that. Objects used with session of type + * %DRM_XE_PXP_TYPE_HWDRM will be marked as invalid if a PXP invalidation + * event occurs after their creation. Attempting to flip an invalid object + * will cause a black frame to be displayed instead. Submissions with invalid + * objects mapped in the VM will be rejected. */ struct drm_xe_gem_create { +#define DRM_XE_GEM_CREATE_EXTENSION_SET_PROPERTY 0 +#define DRM_XE_GEM_CREATE_SET_PROPERTY_PXP_TYPE 0 /** @extensions: Pointer to the first extension struct, if any */ __u64 extensions; @@ -968,6 +983,9 @@ struct drm_xe_vm_destroy { * will only be valid for DRM_XE_VM_BIND_OP_MAP operations, the BO * handle MBZ, and the BO offset MBZ. This flag is intended to * implement VK sparse bindings. + * - %DRM_XE_VM_BIND_FLAG_CHECK_PXP - If the object is encrypted via PXP, + * reject the binding if the encryption key is no longer valid. This + * flag has no effect on BOs that are not marked as using PXP. */ struct drm_xe_vm_bind_op { /** @extensions: Pointer to the first extension struct, if any */ @@ -1058,6 +1076,7 @@ struct drm_xe_vm_bind_op { #define DRM_XE_VM_BIND_FLAG_IMMEDIATE (1 << 1) #define DRM_XE_VM_BIND_FLAG_NULL (1 << 2) #define DRM_XE_VM_BIND_FLAG_DUMPABLE (1 << 3) +#define DRM_XE_VM_BIND_FLAG_CHECK_PXP (1 << 4) /** @flags: Bind flags */ __u32 flags; From 51462211f4a920687ae51823f6a8b0f499546bcc Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:35 -0800 Subject: [PATCH 0417/1443] drm/xe/pxp: add PXP PM support The HW suspend flow kills all PXP HWDRM sessions, so we need to mark all the queues and BOs as invalid and do a full termination when PXP is next used. v2: rebase v3: rebase on new status flow, defer termination to next PXP use as it makes things much easier and allows us to use the same function for all types of suspend. v4: fix the documentation of the suspend function (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-12-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/xe_pm.c | 39 ++++++--- drivers/gpu/drm/xe/xe_pxp.c | 129 +++++++++++++++++++++++++++++- drivers/gpu/drm/xe/xe_pxp.h | 3 + drivers/gpu/drm/xe/xe_pxp_types.h | 7 ++ 4 files changed, 165 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c index c9cc0c091dfd..35dd1757b40b 100644 --- a/drivers/gpu/drm/xe/xe_pm.c +++ b/drivers/gpu/drm/xe/xe_pm.c @@ -22,6 +22,7 @@ #include "xe_guc.h" #include "xe_irq.h" #include "xe_pcode.h" +#include "xe_pxp.h" #include "xe_trace.h" #include "xe_wa.h" @@ -122,6 +123,10 @@ int xe_pm_suspend(struct xe_device *xe) drm_dbg(&xe->drm, "Suspending device\n"); trace_xe_pm_suspend(xe, __builtin_return_address(0)); + err = xe_pxp_pm_suspend(xe->pxp); + if (err) + goto err; + for_each_gt(gt, xe, id) xe_gt_suspend_prepare(gt); @@ -130,14 +135,12 @@ int xe_pm_suspend(struct xe_device *xe) /* FIXME: Super racey... */ err = xe_bo_evict_all(xe); if (err) - goto err; + goto err_pxp; for_each_gt(gt, xe, id) { err = xe_gt_suspend(gt); - if (err) { - xe_display_pm_resume(xe); - goto err; - } + if (err) + goto err_display; } xe_irq_suspend(xe); @@ -146,6 +149,11 @@ int xe_pm_suspend(struct xe_device *xe) drm_dbg(&xe->drm, "Device suspended\n"); return 0; + +err_display: + xe_display_pm_resume(xe); +err_pxp: + xe_pxp_pm_resume(xe->pxp); err: drm_dbg(&xe->drm, "Device suspend failed %d\n", err); return err; @@ -195,6 +203,8 @@ int xe_pm_resume(struct xe_device *xe) if (err) goto err; + xe_pxp_pm_resume(xe->pxp); + drm_dbg(&xe->drm, "Device resumed\n"); return 0; err: @@ -389,6 +399,10 @@ int xe_pm_runtime_suspend(struct xe_device *xe) */ xe_rpm_lockmap_acquire(xe); + err = xe_pxp_pm_suspend(xe->pxp); + if (err) + goto out; + /* * Applying lock for entire list op as xe_ttm_bo_destroy and xe_bo_move_notify * also checks and deletes bo entry from user fault list. @@ -404,22 +418,27 @@ int xe_pm_runtime_suspend(struct xe_device *xe) if (xe->d3cold.allowed) { err = xe_bo_evict_all(xe); if (err) - goto out; + goto out_resume; } for_each_gt(gt, xe, id) { err = xe_gt_suspend(gt); if (err) - goto out; + goto out_resume; } xe_irq_suspend(xe); xe_display_pm_runtime_suspend_late(xe); + xe_rpm_lockmap_release(xe); + xe_pm_write_callback_task(xe, NULL); + return 0; + +out_resume: + xe_display_pm_runtime_resume(xe); + xe_pxp_pm_resume(xe->pxp); out: - if (err) - xe_display_pm_runtime_resume(xe); xe_rpm_lockmap_release(xe); xe_pm_write_callback_task(xe, NULL); return err; @@ -472,6 +491,8 @@ int xe_pm_runtime_resume(struct xe_device *xe) goto out; } + xe_pxp_pm_resume(xe->pxp); + out: xe_rpm_lockmap_release(xe); xe_pm_write_callback_task(xe, NULL); diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 8060b4050be8..5e7a1688a771 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -132,6 +132,14 @@ static int pxp_wait_for_session_state(struct xe_pxp *pxp, u32 id, bool in_play) static void pxp_invalidate_queues(struct xe_pxp *pxp); +static void pxp_invalidate_state(struct xe_pxp *pxp) +{ + pxp_invalidate_queues(pxp); + + if (pxp->status == XE_PXP_ACTIVE) + pxp->key_instance++; +} + static int pxp_terminate_hw(struct xe_pxp *pxp) { struct xe_gt *gt = pxp->gt; @@ -185,10 +193,16 @@ static void pxp_terminate(struct xe_pxp *pxp) mutex_lock(&pxp->mutex); - pxp_invalidate_queues(pxp); + pxp_invalidate_state(pxp); - if (pxp->status == XE_PXP_ACTIVE) - pxp->key_instance++; + /* + * we'll mark the status as needing termination on resume, so no need to + * emit a termination now. + */ + if (pxp->status == XE_PXP_SUSPENDED) { + mutex_unlock(&pxp->mutex); + return; + } /* * If we have a termination already in progress, we need to wait for @@ -219,11 +233,13 @@ static void pxp_terminate(struct xe_pxp *pxp) static void pxp_terminate_complete(struct xe_pxp *pxp) { /* - * We expect PXP to be in one of 2 states when we get here: + * We expect PXP to be in one of 3 states when we get here: * - XE_PXP_TERMINATION_IN_PROGRESS: a single termination event was * requested and it is now completing, so we're ready to start. * - XE_PXP_NEEDS_ADDITIONAL_TERMINATION: a second termination was * requested while the first one was still being processed. + * - XE_PXP_SUSPENDED: PXP is now suspended, so we defer everything to + * when we come back on resume. */ mutex_lock(&pxp->mutex); @@ -234,6 +250,9 @@ static void pxp_terminate_complete(struct xe_pxp *pxp) case XE_PXP_NEEDS_ADDITIONAL_TERMINATION: pxp->status = XE_PXP_NEEDS_TERMINATION; break; + case XE_PXP_SUSPENDED: + /* Nothing to do */ + break; default: drm_err(&pxp->xe->drm, "PXP termination complete while status was %u\n", @@ -391,6 +410,7 @@ int xe_pxp_init(struct xe_device *xe) pxp->gt = gt; pxp->key_instance = 1; + pxp->last_suspend_key_instance = 1; /* * we'll use the completions to check if there is an action pending, @@ -574,6 +594,7 @@ int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q) XE_WARN_ON(completion_done(&pxp->termination)); mutex_unlock(&pxp->mutex); goto wait_for_idle; + case XE_PXP_SUSPENDED: default: drm_err(&pxp->xe->drm, "unexpected state during PXP start: %u\n", pxp->status); ret = -EIO; @@ -779,3 +800,103 @@ int xe_pxp_obj_key_check(struct xe_pxp *pxp, struct drm_gem_object *obj) { return xe_pxp_bo_key_check(pxp, gem_to_xe_bo(obj)); } + +/** + * xe_pxp_pm_suspend - prepare PXP for HW suspend + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + * + * Makes sure all PXP actions have completed and invalidates all PXP queues + * and objects before we go into a suspend state. + * + * Returns: 0 if successful, a negative errno value otherwise. + */ +int xe_pxp_pm_suspend(struct xe_pxp *pxp) +{ + int ret = 0; + + if (!xe_pxp_is_enabled(pxp)) + return 0; + +wait_for_activation: + if (!wait_for_completion_timeout(&pxp->activation, + msecs_to_jiffies(PXP_ACTIVATION_TIMEOUT_MS))) + ret = -ETIMEDOUT; + + mutex_lock(&pxp->mutex); + + switch (pxp->status) { + case XE_PXP_ERROR: + case XE_PXP_READY_TO_START: + case XE_PXP_SUSPENDED: + case XE_PXP_TERMINATION_IN_PROGRESS: + case XE_PXP_NEEDS_ADDITIONAL_TERMINATION: + /* + * If PXP is not running there is nothing to cleanup. If there + * is a termination pending then no need to issue another one. + */ + break; + case XE_PXP_START_IN_PROGRESS: + mutex_unlock(&pxp->mutex); + goto wait_for_activation; + case XE_PXP_NEEDS_TERMINATION: + /* If PXP was never used we can skip the cleanup */ + if (pxp->key_instance == pxp->last_suspend_key_instance) + break; + fallthrough; + case XE_PXP_ACTIVE: + pxp_invalidate_state(pxp); + break; + default: + drm_err(&pxp->xe->drm, "unexpected state during PXP suspend: %u", + pxp->status); + ret = -EIO; + goto out; + } + + /* + * We set this even if we were in error state, hoping the suspend clears + * the error. Worse case we fail again and go in error state again. + */ + pxp->status = XE_PXP_SUSPENDED; + + mutex_unlock(&pxp->mutex); + + /* + * if there is a termination in progress, wait for it. + * We need to wait outside the lock because the completion is done from + * within the lock + */ + if (!wait_for_completion_timeout(&pxp->termination, + msecs_to_jiffies(PXP_TERMINATION_TIMEOUT_MS))) + ret = -ETIMEDOUT; + + pxp->last_suspend_key_instance = pxp->key_instance; + +out: + return ret; +} + +/** + * xe_pxp_pm_resume - re-init PXP after HW suspend + * @pxp: the xe->pxp pointer (it will be NULL if PXP is disabled) + */ +void xe_pxp_pm_resume(struct xe_pxp *pxp) +{ + int err; + + if (!xe_pxp_is_enabled(pxp)) + return; + + err = kcr_pxp_enable(pxp); + + mutex_lock(&pxp->mutex); + + xe_assert(pxp->xe, pxp->status == XE_PXP_SUSPENDED); + + if (err) + pxp->status = XE_PXP_ERROR; + else + pxp->status = XE_PXP_NEEDS_TERMINATION; + + mutex_unlock(&pxp->mutex); +} diff --git a/drivers/gpu/drm/xe/xe_pxp.h b/drivers/gpu/drm/xe/xe_pxp.h index 3dd70eac9da6..546b156d63aa 100644 --- a/drivers/gpu/drm/xe/xe_pxp.h +++ b/drivers/gpu/drm/xe/xe_pxp.h @@ -21,6 +21,9 @@ int xe_pxp_get_readiness_status(struct xe_pxp *pxp); int xe_pxp_init(struct xe_device *xe); void xe_pxp_irq_handler(struct xe_device *xe, u16 iir); +int xe_pxp_pm_suspend(struct xe_pxp *pxp); +void xe_pxp_pm_resume(struct xe_pxp *pxp); + int xe_pxp_exec_queue_set_type(struct xe_pxp *pxp, struct xe_exec_queue *q, u8 type); int xe_pxp_exec_queue_add(struct xe_pxp *pxp, struct xe_exec_queue *q); void xe_pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q); diff --git a/drivers/gpu/drm/xe/xe_pxp_types.h b/drivers/gpu/drm/xe/xe_pxp_types.h index 8e4569f0173d..53e9d48d10fb 100644 --- a/drivers/gpu/drm/xe/xe_pxp_types.h +++ b/drivers/gpu/drm/xe/xe_pxp_types.h @@ -27,6 +27,7 @@ enum xe_pxp_status { XE_PXP_READY_TO_START, XE_PXP_START_IN_PROGRESS, XE_PXP_ACTIVE, + XE_PXP_SUSPENDED, }; /** @@ -123,6 +124,12 @@ struct xe_pxp { * that case in the code. */ u32 key_instance; + /** + * @last_suspend_key_instance: value of key_instance at the last + * suspend. Used to check if any PXP session has been created between + * suspend cycles. + */ + u32 last_suspend_key_instance; }; #endif /* __XE_PXP_TYPES_H__ */ From 385a8015b21469f249cdf30453f2c93d32405aa9 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:36 -0800 Subject: [PATCH 0418/1443] drm/xe/pxp: Add PXP debugfs support This patch introduces 2 PXP debugfs entries: - info: prints the current PXP status and key instance - terminate: simulate a termination interrupt The first one is useful for debug, while the second one can be used for testing the termination flow. v2: move the info prints inside the lock (John) Signed-off-by: Daniele Ceraolo Spurio Cc: John Harrison Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-13-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/Makefile | 1 + drivers/gpu/drm/xe/xe_debugfs.c | 3 + drivers/gpu/drm/xe/xe_pxp_debugfs.c | 120 ++++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_pxp_debugfs.h | 13 +++ 4 files changed, 137 insertions(+) create mode 100644 drivers/gpu/drm/xe/xe_pxp_debugfs.c create mode 100644 drivers/gpu/drm/xe/xe_pxp_debugfs.h diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index 720d13c16fa5..be73362ef334 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -88,6 +88,7 @@ xe-y += xe_bb.o \ xe_pt.o \ xe_pt_walk.o \ xe_pxp.o \ + xe_pxp_debugfs.o \ xe_pxp_submit.o \ xe_query.o \ xe_range_fence.o \ diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c index 492b4877433f..56cb3788e752 100644 --- a/drivers/gpu/drm/xe/xe_debugfs.c +++ b/drivers/gpu/drm/xe/xe_debugfs.c @@ -18,6 +18,7 @@ #include "xe_gt_printk.h" #include "xe_guc_ads.h" #include "xe_pm.h" +#include "xe_pxp_debugfs.h" #include "xe_sriov.h" #include "xe_step.h" @@ -230,5 +231,7 @@ void xe_debugfs_register(struct xe_device *xe) for_each_gt(gt, xe, id) xe_gt_debugfs_register(gt); + xe_pxp_debugfs_register(xe->pxp); + fault_create_debugfs_attr("fail_gt_reset", root, >_reset_failure); } diff --git a/drivers/gpu/drm/xe/xe_pxp_debugfs.c b/drivers/gpu/drm/xe/xe_pxp_debugfs.c new file mode 100644 index 000000000000..ccfbacf08efc --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp_debugfs.c @@ -0,0 +1,120 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2024 Intel Corporation + */ + +#include "xe_pxp_debugfs.h" + +#include + +#include +#include +#include + +#include "xe_device.h" +#include "xe_pxp.h" +#include "xe_pxp_types.h" +#include "regs/xe_irq_regs.h" + +static struct xe_pxp *node_to_pxp(struct drm_info_node *node) +{ + return node->info_ent->data; +} + +static const char *pxp_status_to_str(struct xe_pxp *pxp) +{ + lockdep_assert_held(&pxp->mutex); + + switch (pxp->status) { + case XE_PXP_ERROR: + return "error"; + case XE_PXP_NEEDS_TERMINATION: + return "needs termination"; + case XE_PXP_TERMINATION_IN_PROGRESS: + return "termination in progress"; + case XE_PXP_READY_TO_START: + return "ready to start"; + case XE_PXP_ACTIVE: + return "active"; + case XE_PXP_SUSPENDED: + return "suspended"; + default: + return "unknown"; + } +}; + +static int pxp_info(struct seq_file *m, void *data) +{ + struct xe_pxp *pxp = node_to_pxp(m->private); + struct drm_printer p = drm_seq_file_printer(m); + const char *status; + + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + mutex_lock(&pxp->mutex); + status = pxp_status_to_str(pxp); + + drm_printf(&p, "status: %s\n", status); + drm_printf(&p, "instance counter: %u\n", pxp->key_instance); + mutex_unlock(&pxp->mutex); + + return 0; +} + +static int pxp_terminate(struct seq_file *m, void *data) +{ + struct xe_pxp *pxp = node_to_pxp(m->private); + struct drm_printer p = drm_seq_file_printer(m); + + if (!xe_pxp_is_enabled(pxp)) + return -ENODEV; + + /* simulate a termination interrupt */ + spin_lock_irq(&pxp->xe->irq.lock); + xe_pxp_irq_handler(pxp->xe, KCR_PXP_STATE_TERMINATED_INTERRUPT); + spin_unlock_irq(&pxp->xe->irq.lock); + + drm_printf(&p, "PXP termination queued\n"); + + return 0; +} + +static const struct drm_info_list debugfs_list[] = { + {"info", pxp_info, 0}, + {"terminate", pxp_terminate, 0}, +}; + +void xe_pxp_debugfs_register(struct xe_pxp *pxp) +{ + struct drm_minor *minor; + struct drm_info_list *local; + struct dentry *root; + int i; + + if (!xe_pxp_is_enabled(pxp)) + return; + + minor = pxp->xe->drm.primary; + if (!minor->debugfs_root) + return; + +#define DEBUGFS_SIZE (ARRAY_SIZE(debugfs_list) * sizeof(struct drm_info_list)) + local = drmm_kmalloc(&pxp->xe->drm, DEBUGFS_SIZE, GFP_KERNEL); + if (!local) + return; + + memcpy(local, debugfs_list, DEBUGFS_SIZE); +#undef DEBUGFS_SIZE + + for (i = 0; i < ARRAY_SIZE(debugfs_list); ++i) + local[i].data = pxp; + + root = debugfs_create_dir("pxp", minor->debugfs_root); + if (IS_ERR(root)) + return; + + drm_debugfs_create_files(local, + ARRAY_SIZE(debugfs_list), + root, minor); +} diff --git a/drivers/gpu/drm/xe/xe_pxp_debugfs.h b/drivers/gpu/drm/xe/xe_pxp_debugfs.h new file mode 100644 index 000000000000..988466aad50b --- /dev/null +++ b/drivers/gpu/drm/xe/xe_pxp_debugfs.h @@ -0,0 +1,13 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2024 Intel Corporation + */ + +#ifndef __XE_PXP_DEBUGFS_H__ +#define __XE_PXP_DEBUGFS_H__ + +struct xe_pxp; + +void xe_pxp_debugfs_register(struct xe_pxp *pxp); + +#endif /* __XE_PXP_DEBUGFS_H__ */ From 492f8d2030bec7ab6d0adf7f41808d73871f86e7 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Wed, 29 Jan 2025 09:41:37 -0800 Subject: [PATCH 0419/1443] drm/xe/pxp: Enable PXP for MTL and LNL Now that are the pieces are there, we can turn the feature on. Signed-off-by: Daniele Ceraolo Spurio Reviewed-by: John Harrison Link: https://patchwork.freedesktop.org/patch/msgid/20250129174140.948829-14-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/xe_pci.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 0a6e58d55682..feb46a707281 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -333,11 +333,13 @@ static const struct xe_device_desc mtl_desc = { .require_force_probe = true, PLATFORM(METEORLAKE), .has_display = true, + .has_pxp = true, }; static const struct xe_device_desc lnl_desc = { PLATFORM(LUNARLAKE), .has_display = true, + .has_pxp = true, }; static const struct xe_device_desc bmg_desc = { From 1ddee69108d305bbc059cbf31c0b47626796be77 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 17 Jan 2025 16:21:45 +0200 Subject: [PATCH 0420/1443] pinctrl: cy8c95x0: Respect IRQ trigger settings from firmware Some of the platforms may connect the INT pin via inversion logic effectively make the triggering to be active-low. Remove explicit trigger flag to respect the settings from firmware. Without this change even idling chip produces spurious interrupts and kernel disables the line in the result: irq 33: nobody cared (try booting with the "irqpoll" option) CPU: 0 UID: 0 PID: 125 Comm: irq/33-i2c-INT3 Not tainted 6.12.0-00236-g8b874ed11dae #64 Hardware name: Intel Corp. QUARK/Galileo, BIOS 0x01000900 01/01/2014 ... handlers: [<86e86bea>] irq_default_primary_handler threaded [] cy8c95x0_irq_handler [pinctrl_cy8c95x0] Disabling IRQ #33 Fixes: e6cbbe42944d ("pinctrl: Add Cypress cy8c95x0 support") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/20250117142304.596106-2-andriy.shevchenko@linux.intel.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinctrl-cy8c95x0.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/pinctrl-cy8c95x0.c b/drivers/pinctrl/pinctrl-cy8c95x0.c index 75100a9fb8e4..d73004b4a45e 100644 --- a/drivers/pinctrl/pinctrl-cy8c95x0.c +++ b/drivers/pinctrl/pinctrl-cy8c95x0.c @@ -1355,7 +1355,7 @@ static int cy8c95x0_irq_setup(struct cy8c95x0_pinctrl *chip, int irq) ret = devm_request_threaded_irq(chip->dev, irq, NULL, cy8c95x0_irq_handler, - IRQF_ONESHOT | IRQF_SHARED | IRQF_TRIGGER_HIGH, + IRQF_ONESHOT | IRQF_SHARED, dev_name(chip->dev), chip); if (ret) { dev_err(chip->dev, "failed to request irq %d\n", irq); From 902e09c8acde117b00369521f54df817a983d4ab Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 3 Feb 2025 16:16:09 -0500 Subject: [PATCH 0421/1443] fix braino in "9p: fix ->rename_sem exclusion" ->d_op can bloody well be NULL Fucked-up-by: Al Viro Fixes: 30d61efe118c "9p: fix ->rename_sem exclusion" Signed-off-by: Al Viro --- fs/dcache.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 903142b324e9..8a605681b26f 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2967,11 +2967,11 @@ static int __d_unalias(struct dentry *dentry, struct dentry *alias) goto out_err; m2 = &alias->d_parent->d_inode->i_rwsem; out_unalias: - if (alias->d_op->d_unalias_trylock && + if (alias->d_op && alias->d_op->d_unalias_trylock && !alias->d_op->d_unalias_trylock(alias)) goto out_err; __d_move(alias, dentry, false); - if (alias->d_op->d_unalias_unlock) + if (alias->d_op && alias->d_op->d_unalias_unlock) alias->d_op->d_unalias_unlock(alias); ret = 0; out_err: From 3a4e7193ec37ee2476ce726589de4495a066b565 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 1 Feb 2025 16:50:24 -0800 Subject: [PATCH 0422/1443] MAINTAINERS: list openvswitch docs under its entry Submissions to the docs seem to not get properly CCed. Acked-by: Ilya Maximets Link: https://patch.msgid.link/20250202005024.964262-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index d1086e53a317..c7b8c6535a1e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -17706,6 +17706,7 @@ L: netdev@vger.kernel.org L: dev@openvswitch.org S: Maintained W: http://openvswitch.org +F: Documentation/networking/openvswitch.rst F: include/uapi/linux/openvswitch.h F: net/openvswitch/ F: tools/testing/selftests/net/openvswitch/ From 4d896b35394144c246daaeb5280a015a630958e7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 1 Feb 2025 17:47:26 -0800 Subject: [PATCH 0423/1443] MAINTAINERS: add Kuniyuki Iwashima to TCP reviewers List Kuniyuki as an official TCP reviewer. Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250202014728.1005003-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index c7b8c6535a1e..48677d61c97b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16614,6 +16614,7 @@ F: tools/testing/selftests/net/mptcp/ NETWORKING [TCP] M: Eric Dumazet M: Neal Cardwell +R: Kuniyuki Iwashima L: netdev@vger.kernel.org S: Maintained F: Documentation/networking/net_cachelines/tcp_sock.rst From ae0585b04ab741b536b0db20c12baf24bf7118d2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 1 Feb 2025 17:47:27 -0800 Subject: [PATCH 0424/1443] MAINTAINERS: add a general entry for BSD sockets Create a MAINTAINERS entry for BSD sockets. List the top 3 reviewers as maintainers. The entry is meant to cover core socket code (of which there isn't much) but also reviews of any new socket families. Reviewed-by: Simon Horman Acked-by: Willem de Bruijn Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250202014728.1005003-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 48677d61c97b..438d85bb97a2 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16642,6 +16642,22 @@ F: include/net/tls.h F: include/uapi/linux/tls.h F: net/tls/* +NETWORKING [SOCKETS] +M: Eric Dumazet +M: Kuniyuki Iwashima +M: Paolo Abeni +M: Willem de Bruijn +S: Maintained +F: include/linux/sock_diag.h +F: include/linux/socket.h +F: include/linux/sockptr.h +F: include/net/sock.h +F: include/net/sock_reuseport.h +F: include/uapi/linux/socket.h +F: net/core/*sock* +F: net/core/scm.c +F: net/socket.c + NETXEN (1/10) GbE SUPPORT M: Manish Chopra M: Rahul Verma From 8a2e22f665a0b5c212057031e94b75cfdc11a4a6 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 1 Feb 2025 17:47:28 -0800 Subject: [PATCH 0425/1443] MAINTAINERS: add entry for UNIX sockets Add a MAINTAINERS entry for UNIX socket, Kuniyuki has been the de-facto maintainer of this code for a while. Reviewed-by: Simon Horman Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250202014728.1005003-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 438d85bb97a2..74b09dad4662 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16658,6 +16658,15 @@ F: net/core/*sock* F: net/core/scm.c F: net/socket.c +NETWORKING [UNIX SOCKETS] +M: Kuniyuki Iwashima +S: Maintained +F: include/net/af_unix.h +F: include/net/netns/unix.h +F: include/uapi/linux/unix_diag.h +F: net/unix/ +F: tools/testing/selftests/net/af_unix/ + NETXEN (1/10) GbE SUPPORT M: Manish Chopra M: Rahul Verma From 45ab5166a82d038c898985b0ad43ead69c1f9573 Mon Sep 17 00:00:00 2001 From: Charles Han Date: Thu, 14 Nov 2024 17:19:47 +0800 Subject: [PATCH 0426/1443] HID: winwing: Add NULL check in winwing_init_led() devm_kasprintf() can return a NULL pointer on failure,but this returned value in winwing_init_led() is not checked. Add NULL check in winwing_init_led(), to handle kernel NULL pointer dereference error. Fixes: 266c990debad ("HID: Add WinWing Orion2 throttle support") Signed-off-by: Charles Han Signed-off-by: Jiri Kosina --- drivers/hid/hid-winwing.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hid/hid-winwing.c b/drivers/hid/hid-winwing.c index 831b760c66ea..d4afbbd27807 100644 --- a/drivers/hid/hid-winwing.c +++ b/drivers/hid/hid-winwing.c @@ -106,6 +106,8 @@ static int winwing_init_led(struct hid_device *hdev, "%s::%s", dev_name(&input->dev), info->led_name); + if (!led->cdev.name) + return -ENOMEM; ret = devm_led_classdev_register(&hdev->dev, &led->cdev); if (ret) From 9b8e2220d3a052a690b1d1b23019673e612494c5 Mon Sep 17 00:00:00 2001 From: Charles Han Date: Fri, 15 Nov 2024 14:26:21 +0800 Subject: [PATCH 0427/1443] HID: multitouch: Add NULL check in mt_input_configured devm_kasprintf() can return a NULL pointer on failure,but this returned value in mt_input_configured() is not checked. Add NULL check in mt_input_configured(), to handle kernel NULL pointer dereference error. Fixes: 479439463529 ("HID: multitouch: Correct devm device reference for hidinput input_dev name") Signed-off-by: Charles Han Signed-off-by: Jiri Kosina --- drivers/hid/hid-multitouch.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 82900857bfd8..e50887a6d22c 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -1679,9 +1679,12 @@ static int mt_input_configured(struct hid_device *hdev, struct hid_input *hi) break; } - if (suffix) + if (suffix) { hi->input->name = devm_kasprintf(&hdev->dev, GFP_KERNEL, "%s %s", hdev->name, suffix); + if (!hi->input->name) + return -ENOMEM; + } return 0; } From e4afdef60562014bc4a5c74384cfe9de84590ca5 Mon Sep 17 00:00:00 2001 From: Sai Teja Pottumuttu Date: Thu, 30 Jan 2025 14:28:03 +0530 Subject: [PATCH 0428/1443] drm/xe: Refactor dma_mask_size dma_mask_size is more related to the platform than the GT IP. Thus move it to platform descriptors. v2: - Rebase Signed-off-by: Sai Teja Pottumuttu Reviewed-by: Matt Roper Link: https://patchwork.freedesktop.org/patch/msgid/20250130085804.4136497-2-sai.teja.pottumuttu@intel.com Signed-off-by: Matt Roper --- drivers/gpu/drm/xe/xe_pci.c | 22 ++++++++++++++++------ drivers/gpu/drm/xe/xe_pci_types.h | 1 - 2 files changed, 16 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index feb46a707281..38e6ba091ea5 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -55,6 +55,8 @@ struct xe_device_desc { enum xe_platform platform; + u8 dma_mask_size; + u8 require_force_probe:1; u8 is_dgfx:1; @@ -85,7 +87,6 @@ static const struct xe_graphics_desc graphics_xelp = { .hw_engine_mask = BIT(XE_HW_ENGINE_RCS0) | BIT(XE_HW_ENGINE_BCS0), - .dma_mask_size = 39, .va_bits = 48, .vm_max_level = 3, }; @@ -97,14 +98,12 @@ static const struct xe_graphics_desc graphics_xelpp = { .hw_engine_mask = BIT(XE_HW_ENGINE_RCS0) | BIT(XE_HW_ENGINE_BCS0), - .dma_mask_size = 39, .va_bits = 48, .vm_max_level = 3, }; #define XE_HP_FEATURES \ .has_range_tlb_invalidation = true, \ - .dma_mask_size = 46, \ .va_bits = 48, \ .vm_max_level = 3 @@ -139,7 +138,6 @@ static const struct xe_graphics_desc graphics_xehpc = { BIT(XE_HW_ENGINE_CCS2) | BIT(XE_HW_ENGINE_CCS3), XE_HP_FEATURES, - .dma_mask_size = 52, .max_remote_tiles = 1, .va_bits = 57, .vm_max_level = 4, @@ -160,7 +158,6 @@ static const struct xe_graphics_desc graphics_xelpg = { }; #define XE2_GFX_FEATURES \ - .dma_mask_size = 46, \ .has_asid = 1, \ .has_atomic_enable_pte_bit = 1, \ .has_flat_ccs = 1, \ @@ -220,6 +217,7 @@ static const struct xe_device_desc tgl_desc = { .graphics = &graphics_xelp, .media = &media_xem, PLATFORM(TIGERLAKE), + .dma_mask_size = 39, .has_display = true, .has_llc = true, .require_force_probe = true, @@ -229,6 +227,7 @@ static const struct xe_device_desc rkl_desc = { .graphics = &graphics_xelp, .media = &media_xem, PLATFORM(ROCKETLAKE), + .dma_mask_size = 39, .has_display = true, .has_llc = true, .require_force_probe = true, @@ -240,6 +239,7 @@ static const struct xe_device_desc adl_s_desc = { .graphics = &graphics_xelp, .media = &media_xem, PLATFORM(ALDERLAKE_S), + .dma_mask_size = 39, .has_display = true, .has_llc = true, .require_force_probe = true, @@ -255,6 +255,7 @@ static const struct xe_device_desc adl_p_desc = { .graphics = &graphics_xelp, .media = &media_xem, PLATFORM(ALDERLAKE_P), + .dma_mask_size = 39, .has_display = true, .has_llc = true, .require_force_probe = true, @@ -268,6 +269,7 @@ static const struct xe_device_desc adl_n_desc = { .graphics = &graphics_xelp, .media = &media_xem, PLATFORM(ALDERLAKE_N), + .dma_mask_size = 39, .has_display = true, .has_llc = true, .require_force_probe = true, @@ -281,6 +283,7 @@ static const struct xe_device_desc dg1_desc = { .media = &media_xem, DGFX_FEATURES, PLATFORM(DG1), + .dma_mask_size = 39, .has_display = true, .has_heci_gscfi = 1, .require_force_probe = true, @@ -304,6 +307,7 @@ static const u16 dg2_g12_ids[] = { INTEL_DG2_G12_IDS(NOP), 0 }; static const struct xe_device_desc ats_m_desc = { .graphics = &graphics_xehpg, .media = &media_xehpm, + .dma_mask_size = 46, .require_force_probe = true, DG2_FEATURES, @@ -313,6 +317,7 @@ static const struct xe_device_desc ats_m_desc = { static const struct xe_device_desc dg2_desc = { .graphics = &graphics_xehpg, .media = &media_xehpm, + .dma_mask_size = 46, .require_force_probe = true, DG2_FEATURES, @@ -323,6 +328,7 @@ static const __maybe_unused struct xe_device_desc pvc_desc = { .graphics = &graphics_xehpc, DGFX_FEATURES, PLATFORM(PVC), + .dma_mask_size = 52, .has_display = false, .has_heci_gscfi = 1, .require_force_probe = true, @@ -332,12 +338,14 @@ static const struct xe_device_desc mtl_desc = { /* .graphics and .media determined via GMD_ID */ .require_force_probe = true, PLATFORM(METEORLAKE), + .dma_mask_size = 46, .has_display = true, .has_pxp = true, }; static const struct xe_device_desc lnl_desc = { PLATFORM(LUNARLAKE), + .dma_mask_size = 46, .has_display = true, .has_pxp = true, }; @@ -345,12 +353,14 @@ static const struct xe_device_desc lnl_desc = { static const struct xe_device_desc bmg_desc = { DGFX_FEATURES, PLATFORM(BATTLEMAGE), + .dma_mask_size = 46, .has_display = true, .has_heci_cscfi = 1, }; static const struct xe_device_desc ptl_desc = { PLATFORM(PANTHERLAKE), + .dma_mask_size = 46, .has_display = true, .require_force_probe = true, }; @@ -617,6 +627,7 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.subplatform = subplatform_desc ? subplatform_desc->subplatform : XE_SUBPLATFORM_NONE; + xe->info.dma_mask_size = desc->dma_mask_size; xe->info.is_dgfx = desc->is_dgfx; xe->info.has_heci_gscfi = desc->has_heci_gscfi; xe->info.has_heci_cscfi = desc->has_heci_cscfi; @@ -682,7 +693,6 @@ static int xe_info_init(struct xe_device *xe, xe->info.graphics_name = graphics_desc->name; xe->info.media_name = media_desc ? media_desc->name : "none"; - xe->info.dma_mask_size = graphics_desc->dma_mask_size; xe->info.vram_flags = graphics_desc->vram_flags; xe->info.va_bits = graphics_desc->va_bits; xe->info.vm_max_level = graphics_desc->vm_max_level; diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index 873efec5cdee..7437415a54d5 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -13,7 +13,6 @@ struct xe_graphics_desc { u8 ver; u8 rel; - u8 dma_mask_size; /* available DMA address bits */ u8 va_bits; u8 vm_max_level; u8 vram_flags; From 206fa53fe1f1f73eae61c4330aa31dd6393d959e Mon Sep 17 00:00:00 2001 From: Sai Teja Pottumuttu Date: Thu, 30 Jan 2025 14:28:04 +0530 Subject: [PATCH 0429/1443] drm/xe: Refactor max_remote_tiles max_remote_tiles is more related to the platform than the GT IP. Thus move it to platform descriptor from graphics descriptor. Note that the FIXME is no more required, thus it can be dropped. v2: Rebase v3: Change the position of comment (MattR) Signed-off-by: Sai Teja Pottumuttu Reviewed-by: Matt Roper Link: https://patchwork.freedesktop.org/patch/msgid/20250130085804.4136497-3-sai.teja.pottumuttu@intel.com Signed-off-by: Matt Roper --- drivers/gpu/drm/xe/xe_pci.c | 21 +++++++++------------ drivers/gpu/drm/xe/xe_pci_types.h | 2 -- 2 files changed, 9 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 38e6ba091ea5..cfa1ad7a1f2d 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -56,6 +56,7 @@ struct xe_device_desc { enum xe_platform platform; u8 dma_mask_size; + u8 max_remote_tiles:2; u8 require_force_probe:1; u8 is_dgfx:1; @@ -138,7 +139,6 @@ static const struct xe_graphics_desc graphics_xehpc = { BIT(XE_HW_ENGINE_CCS2) | BIT(XE_HW_ENGINE_CCS3), XE_HP_FEATURES, - .max_remote_tiles = 1, .va_bits = 57, .vm_max_level = 4, .vram_flags = XE_VRAM_FLAGS_NEED64K, @@ -331,6 +331,7 @@ static const __maybe_unused struct xe_device_desc pvc_desc = { .dma_mask_size = 52, .has_display = false, .has_heci_gscfi = 1, + .max_remote_tiles = 1, .require_force_probe = true, }; @@ -641,6 +642,7 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.probe_display = IS_ENABLED(CONFIG_DRM_XE_DISPLAY) && xe_modparam.probe_display && desc->has_display; + xe->info.tile_count = 1 + desc->max_remote_tiles; err = xe_tile_init_early(xe_device_get_root_tile(xe), xe, 0); if (err) @@ -707,17 +709,6 @@ static int xe_info_init(struct xe_device *xe, xe->info.has_range_tlb_invalidation = graphics_desc->has_range_tlb_invalidation; xe->info.has_usm = graphics_desc->has_usm; - /* - * All platforms have at least one primary GT. Any platform with media - * version 13 or higher has an additional dedicated media GT. And - * depending on the graphics IP there may be additional "remote tiles." - * All of these together determine the overall GT count. - * - * FIXME: 'tile_count' here is misnamed since the rest of the driver - * treats it as the number of GTs rather than just the number of tiles. - */ - xe->info.tile_count = 1 + graphics_desc->max_remote_tiles; - for_each_remote_tile(tile, xe, id) { int err; @@ -726,6 +717,12 @@ static int xe_info_init(struct xe_device *xe, return err; } + /* + * All platforms have at least one primary GT. Any platform with media + * version 13 or higher has an additional dedicated media GT. And + * depending on the graphics IP there may be additional "remote tiles." + * All of these together determine the overall GT count. + */ for_each_tile(tile, xe, id) { gt = tile->primary_gt; gt->info.id = xe->info.gt_count++; diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index 7437415a54d5..b96423844952 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -19,8 +19,6 @@ struct xe_graphics_desc { u64 hw_engine_mask; /* hardware engines provided by graphics IP */ - u8 max_remote_tiles:2; - u8 has_asid:1; u8 has_atomic_enable_pte_bit:1; u8 has_flat_ccs:1; From 3d4114a1d34413dfffa0094c2eb7b95e61087abd Mon Sep 17 00:00:00 2001 From: Avri Altman Date: Tue, 28 Jan 2025 09:12:06 +0200 Subject: [PATCH 0430/1443] scsi: ufs: core: Ensure clk_gating.lock is used only after initialization Address a lockdep warning triggered by the use of the clk_gating.lock before it is properly initialized. The warning is as follows: [ 4.388838] INFO: trying to register non-static key. [ 4.395673] The code is fine but needs lockdep annotation, or maybe [ 4.402118] you didn't initialize this object before use? [ 4.407673] turning off the locking correctness validator. [ 4.413334] CPU: 5 UID: 0 PID: 58 Comm: kworker/u32:1 Not tainted 6.12-rc1 #185 [ 4.413343] Hardware name: Qualcomm Technologies, Inc. Robotics RB5 (DT) [ 4.413362] Call trace: [ 4.413364] show_stack+0x18/0x24 (C) [ 4.413374] dump_stack_lvl+0x90/0xd0 [ 4.413384] dump_stack+0x18/0x24 [ 4.413392] register_lock_class+0x498/0x4a8 [ 4.413400] __lock_acquire+0xb4/0x1b90 [ 4.413406] lock_acquire+0x114/0x310 [ 4.413413] _raw_spin_lock_irqsave+0x60/0x88 [ 4.413423] ufshcd_setup_clocks+0x2c0/0x490 [ 4.413433] ufshcd_init+0x198/0x10ec [ 4.413437] ufshcd_pltfrm_init+0x600/0x7c0 [ 4.413444] ufs_qcom_probe+0x20/0x58 [ 4.413449] platform_probe+0x68/0xd8 [ 4.413459] really_probe+0xbc/0x268 [ 4.413466] __driver_probe_device+0x78/0x12c [ 4.413473] driver_probe_device+0x40/0x11c [ 4.413481] __device_attach_driver+0xb8/0xf8 [ 4.413489] bus_for_each_drv+0x84/0xe4 [ 4.413495] __device_attach+0xfc/0x18c [ 4.413502] device_initial_probe+0x14/0x20 [ 4.413510] bus_probe_device+0xb0/0xb4 [ 4.413517] deferred_probe_work_func+0x8c/0xc8 [ 4.413524] process_scheduled_works+0x250/0x658 [ 4.413534] worker_thread+0x15c/0x2c8 [ 4.413542] kthread+0x134/0x200 [ 4.413550] ret_from_fork+0x10/0x20 To fix this issue, ensure that the spinlock is only used after it has been properly initialized before using it in ufshcd_setup_clocks(). Do that unconditionally as initializing a spinlock is a fast operation. Fixes: 209f4e43b806 ("scsi: ufs: core: Introduce a new clock_gating lock") Reported-by: Dmitry Baryshkov Tested-by: Geert Uytterhoeven Signed-off-by: Avri Altman Link: https://lore.kernel.org/r/20250128071207.75494-2-avri.altman@wdc.com Reviewed-by: Bean Huo Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index 356e1be33f53..db20b1d505b7 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -2120,8 +2120,6 @@ static void ufshcd_init_clk_gating(struct ufs_hba *hba) INIT_DELAYED_WORK(&hba->clk_gating.gate_work, ufshcd_gate_work); INIT_WORK(&hba->clk_gating.ungate_work, ufshcd_ungate_work); - spin_lock_init(&hba->clk_gating.lock); - hba->clk_gating.clk_gating_workq = alloc_ordered_workqueue( "ufs_clk_gating_%d", WQ_MEM_RECLAIM | WQ_HIGHPRI, hba->host->host_no); @@ -10411,6 +10409,12 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq) hba->irq = irq; hba->vps = &ufs_hba_vps; + /* + * Initialize clk_gating.lock early since it is being used in + * ufshcd_setup_clocks() + */ + spin_lock_init(&hba->clk_gating.lock); + err = ufshcd_hba_init(hba); if (err) goto out_error; From 839a74b5649c9f41d939a05059b5ca6b17156d03 Mon Sep 17 00:00:00 2001 From: Avri Altman Date: Tue, 28 Jan 2025 09:12:07 +0200 Subject: [PATCH 0431/1443] scsi: ufs: Fix toggling of clk_gating.state when clock gating is not allowed This commit addresses an issue where clk_gating.state is being toggled in ufshcd_setup_clocks() even if clock gating is not allowed. The fix is to add a check for hba->clk_gating.is_initialized before toggling clk_gating.state in ufshcd_setup_clocks(). Since clk_gating.lock is now initialized unconditionally, it can no longer lead to the spinlock being used before it is properly initialized, but instead it is mostly for documentation purposes. Fixes: 1ab27c9cf8b6 ("ufs: Add support for clock gating") Reported-by: Geert Uytterhoeven Tested-by: Geert Uytterhoeven Signed-off-by: Avri Altman Link: https://lore.kernel.org/r/20250128071207.75494-3-avri.altman@wdc.com Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index db20b1d505b7..d3741b1f4382 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -9140,7 +9140,7 @@ static int ufshcd_setup_clocks(struct ufs_hba *hba, bool on) if (!IS_ERR_OR_NULL(clki->clk) && clki->enabled) clk_disable_unprepare(clki->clk); } - } else if (!ret && on) { + } else if (!ret && on && hba->clk_gating.is_initialized) { scoped_guard(spinlock_irqsave, &hba->clk_gating.lock) hba->clk_gating.state = CLKS_ON; trace_ufshcd_clk_gating(dev_name(hba->dev), From 18c966b62819b9d3b99eac8fb8cdc8950826e0c2 Mon Sep 17 00:00:00 2001 From: Zhang Lixu Date: Thu, 23 Jan 2025 09:30:44 +0800 Subject: [PATCH 0432/1443] HID: intel-ish-hid: ipc: Add Panther Lake PCI device IDs Add device IDs of Panther Lake-H and Panther Lake-P into ishtp support list. Signed-off-by: Zhang Lixu Acked-by: Srinivas Pandruvada Signed-off-by: Jiri Kosina --- drivers/hid/intel-ish-hid/ipc/hw-ish.h | 2 ++ drivers/hid/intel-ish-hid/ipc/pci-ish.c | 7 +++++++ 2 files changed, 9 insertions(+) diff --git a/drivers/hid/intel-ish-hid/ipc/hw-ish.h b/drivers/hid/intel-ish-hid/ipc/hw-ish.h index cdd80c653918..07e90d51f073 100644 --- a/drivers/hid/intel-ish-hid/ipc/hw-ish.h +++ b/drivers/hid/intel-ish-hid/ipc/hw-ish.h @@ -36,6 +36,8 @@ #define PCI_DEVICE_ID_INTEL_ISH_ARL_H 0x7745 #define PCI_DEVICE_ID_INTEL_ISH_ARL_S 0x7F78 #define PCI_DEVICE_ID_INTEL_ISH_LNL_M 0xA845 +#define PCI_DEVICE_ID_INTEL_ISH_PTL_H 0xE345 +#define PCI_DEVICE_ID_INTEL_ISH_PTL_P 0xE445 #define REVISION_ID_CHT_A0 0x6 #define REVISION_ID_CHT_Ax_SI 0x0 diff --git a/drivers/hid/intel-ish-hid/ipc/pci-ish.c b/drivers/hid/intel-ish-hid/ipc/pci-ish.c index 9e2401291a2f..ff0fc8010072 100644 --- a/drivers/hid/intel-ish-hid/ipc/pci-ish.c +++ b/drivers/hid/intel-ish-hid/ipc/pci-ish.c @@ -26,9 +26,11 @@ enum ishtp_driver_data_index { ISHTP_DRIVER_DATA_NONE, ISHTP_DRIVER_DATA_LNL_M, + ISHTP_DRIVER_DATA_PTL, }; #define ISH_FW_GEN_LNL_M "lnlm" +#define ISH_FW_GEN_PTL "ptl" #define ISH_FIRMWARE_PATH(gen) "intel/ish/ish_" gen ".bin" #define ISH_FIRMWARE_PATH_ALL "intel/ish/ish_*.bin" @@ -37,6 +39,9 @@ static struct ishtp_driver_data ishtp_driver_data[] = { [ISHTP_DRIVER_DATA_LNL_M] = { .fw_generation = ISH_FW_GEN_LNL_M, }, + [ISHTP_DRIVER_DATA_PTL] = { + .fw_generation = ISH_FW_GEN_PTL, + }, }; static const struct pci_device_id ish_pci_tbl[] = { @@ -63,6 +68,8 @@ static const struct pci_device_id ish_pci_tbl[] = { {PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ISH_ARL_H)}, {PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ISH_ARL_S)}, {PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ISH_LNL_M), .driver_data = ISHTP_DRIVER_DATA_LNL_M}, + {PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ISH_PTL_H), .driver_data = ISHTP_DRIVER_DATA_PTL}, + {PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ISH_PTL_P), .driver_data = ISHTP_DRIVER_DATA_PTL}, {} }; MODULE_DEVICE_TABLE(pci, ish_pci_tbl); From 9271af9d846c7e49c8709b58d5853cb73c00b193 Mon Sep 17 00:00:00 2001 From: Daniel Brackenbury Date: Tue, 28 Jan 2025 20:08:49 -0500 Subject: [PATCH 0433/1443] HID: topre: Fix n-key rollover on Realforce R3S TKL boards Newer model R3* Topre Realforce keyboards share an issue with their older R2 cousins where a report descriptor fixup is needed in order for n-key rollover to work correctly, otherwise only 6-key rollover is available. This patch adds some new hardware IDs for the R3S 87-key keyboard and makes amendments to the existing hid-topre driver in order to change the correct byte in the new model. Signed-off-by: Daniel Brackenbury Signed-off-by: Jiri Kosina --- drivers/hid/Kconfig | 3 ++- drivers/hid/hid-ids.h | 1 + drivers/hid/hid-topre.c | 7 +++++++ 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig index ed657ef7281c..dfc245867a46 100644 --- a/drivers/hid/Kconfig +++ b/drivers/hid/Kconfig @@ -1169,7 +1169,8 @@ config HID_TOPRE tristate "Topre REALFORCE keyboards" depends on HID help - Say Y for N-key rollover support on Topre REALFORCE R2 108/87 key keyboards. + Say Y for N-key rollover support on Topre REALFORCE R2 108/87 key and + Topre REALFORCE R3S 87 key keyboards. config HID_THINGM tristate "ThingM blink(1) USB RGB LED" diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 7debfe0c5cb9..ed1d7f9e8caf 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -1302,6 +1302,7 @@ #define USB_VENDOR_ID_TOPRE 0x0853 #define USB_DEVICE_ID_TOPRE_REALFORCE_R2_108 0x0148 #define USB_DEVICE_ID_TOPRE_REALFORCE_R2_87 0x0146 +#define USB_DEVICE_ID_TOPRE_REALFORCE_R3S_87 0x0313 #define USB_VENDOR_ID_TOPSEED 0x0766 #define USB_DEVICE_ID_TOPSEED_CYBERLINK 0x0204 diff --git a/drivers/hid/hid-topre.c b/drivers/hid/hid-topre.c index 848361f6225d..ccedf8721722 100644 --- a/drivers/hid/hid-topre.c +++ b/drivers/hid/hid-topre.c @@ -29,6 +29,11 @@ static const __u8 *topre_report_fixup(struct hid_device *hdev, __u8 *rdesc, hid_info(hdev, "fixing up Topre REALFORCE keyboard report descriptor\n"); rdesc[72] = 0x02; + } else if (*rsize >= 106 && rdesc[28] == 0x29 && rdesc[29] == 0xe7 && + rdesc[30] == 0x81 && rdesc[31] == 0x00) { + hid_info(hdev, + "fixing up Topre REALFORCE keyboard report descriptor\n"); + rdesc[31] = 0x02; } return rdesc; } @@ -38,6 +43,8 @@ static const struct hid_device_id topre_id_table[] = { USB_DEVICE_ID_TOPRE_REALFORCE_R2_108) }, { HID_USB_DEVICE(USB_VENDOR_ID_TOPRE, USB_DEVICE_ID_TOPRE_REALFORCE_R2_87) }, + { HID_USB_DEVICE(USB_VENDOR_ID_TOPRE, + USB_DEVICE_ID_TOPRE_REALFORCE_R3S_87) }, { } }; MODULE_DEVICE_TABLE(hid, topre_id_table); From 5363ee9d110e139584c2d92a0b640bc210588506 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Tue, 28 Jan 2025 16:35:39 -0500 Subject: [PATCH 0434/1443] scsi: core: Use GFP_NOIO to avoid circular locking dependency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Filesystems can write to disk from page reclaim with __GFP_FS set. Marc found a case where scsi_realloc_sdev_budget_map() ends up in page reclaim with GFP_KERNEL, where it could try to take filesystem locks again, leading to a deadlock. WARNING: possible circular locking dependency detected 6.13.0 #1 Not tainted ------------------------------------------------------ kswapd0/70 is trying to acquire lock: ffff8881025d5d78 (&q->q_usage_counter(io)){++++}-{0:0}, at: blk_mq_submit_bio+0x461/0x6e0 but task is already holding lock: ffffffff81ef5f40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x9f/0x760 The full lockdep splat can be found in Marc's report: https://lkml.org/lkml/2025/1/24/1101 Avoid the potential deadlock by doing the allocation with GFP_NOIO, which prevents both filesystem and block layer recursion. Reported-by: Marc Aurèle La France Signed-off-by: Rik van Riel Link: https://lore.kernel.org/r/20250129104525.0ae8421e@fangorn Reviewed-by: Christoph Hellwig Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_scan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 087fcbfc9aaa..96d7e1a9a7c7 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -246,7 +246,7 @@ static int scsi_realloc_sdev_budget_map(struct scsi_device *sdev, } ret = sbitmap_init_node(&sdev->budget_map, scsi_device_max_queue_depth(sdev), - new_shift, GFP_KERNEL, + new_shift, GFP_NOIO, sdev->request_queue->node, false, true); if (!ret) sbitmap_resize(&sdev->budget_map, depth); From 9ff7c383b8ac0c482a1da7989f703406d78445c6 Mon Sep 17 00:00:00 2001 From: Igor Pylypiv Date: Fri, 31 Jan 2025 10:44:07 -0800 Subject: [PATCH 0435/1443] scsi: core: Do not retry I/Os during depopulation Fail I/Os instead of retry to prevent user space processes from being blocked on the I/O completion for several minutes. Retrying I/Os during "depopulation in progress" or "depopulation restore in progress" results in a continuous retry loop until the depopulation completes or until the I/O retry loop is aborted due to a timeout by the scsi_cmd_runtime_exceeced(). Depopulation is slow and can take 24+ hours to complete on 20+ TB HDDs. Most I/Os in the depopulation retry loop end up taking several minutes before returning the failure to user space. Cc: stable@vger.kernel.org # 4.18.x: 2bbeb8d scsi: core: Handle depopulation and restoration in progress Cc: stable@vger.kernel.org # 4.18.x Fixes: e37c7d9a0341 ("scsi: core: sanitize++ in progress") Signed-off-by: Igor Pylypiv Link: https://lore.kernel.org/r/20250131184408.859579-1-ipylypiv@google.com Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_lib.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index d776f13cd160..be0890e4e706 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -872,13 +872,18 @@ static void scsi_io_completion_action(struct scsi_cmnd *cmd, int result) case 0x1a: /* start stop unit in progress */ case 0x1b: /* sanitize in progress */ case 0x1d: /* configuration in progress */ - case 0x24: /* depopulation in progress */ - case 0x25: /* depopulation restore in progress */ action = ACTION_DELAYED_RETRY; break; case 0x0a: /* ALUA state transition */ action = ACTION_DELAYED_REPREP; break; + /* + * Depopulation might take many hours, + * thus it is not worthwhile to retry. + */ + case 0x24: /* depopulation in progress */ + case 0x25: /* depopulation restore in progress */ + fallthrough; default: action = ACTION_FAIL; break; From f8fb2403ddebb5eea0033d90d9daae4c88749ada Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Andr=C3=A9=20Draszik?= Date: Fri, 24 Jan 2025 15:09:00 +0000 Subject: [PATCH 0436/1443] scsi: ufs: core: Fix use-after free in init error and remove paths MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit devm_blk_crypto_profile_init() registers a cleanup handler to run when the associated (platform-) device is being released. For UFS, the crypto private data and pointers are stored as part of the ufs_hba's data structure 'struct ufs_hba::crypto_profile'. This structure is allocated as part of the underlying ufshcd and therefore Scsi_host allocation. During driver release or during error handling in ufshcd_pltfrm_init(), this structure is released as part of ufshcd_dealloc_host() before the (platform-) device associated with the crypto call above is released. Once this device is released, the crypto cleanup code will run, using the just-released 'struct ufs_hba::crypto_profile'. This causes a use-after-free situation: Call trace: kfree+0x60/0x2d8 (P) kvfree+0x44/0x60 blk_crypto_profile_destroy_callback+0x28/0x70 devm_action_release+0x1c/0x30 release_nodes+0x6c/0x108 devres_release_all+0x98/0x100 device_unbind_cleanup+0x20/0x70 really_probe+0x218/0x2d0 In other words, the initialisation code flow is: platform-device probe ufshcd_pltfrm_init() ufshcd_alloc_host() scsi_host_alloc() allocation of struct ufs_hba creation of scsi-host devices devm_blk_crypto_profile_init() devm registration of cleanup handler using platform-device and during error handling of ufshcd_pltfrm_init() or during driver removal: ufshcd_dealloc_host() scsi_host_put() put_device(scsi-host) release of struct ufs_hba put_device(platform-device) crypto cleanup handler To fix this use-after free, change ufshcd_alloc_host() to register a devres action to automatically cleanup the underlying SCSI device on ufshcd destruction, without requiring explicit calls to ufshcd_dealloc_host(). This way: * the crypto profile and all other ufs_hba-owned resources are destroyed before SCSI (as they've been registered after) * a memleak is plugged in tc-dwc-g210-pci.c remove() as a side-effect * EXPORT_SYMBOL_GPL(ufshcd_dealloc_host) can be removed fully as it's not needed anymore * no future drivers using ufshcd_alloc_host() could ever forget adding the cleanup Fixes: cb77cb5abe1f ("blk-crypto: rename blk_keyslot_manager to blk_crypto_profile") Fixes: d76d9d7d1009 ("scsi: ufs: use devm_blk_ksm_init()") Cc: stable@vger.kernel.org Signed-off-by: André Draszik Link: https://lore.kernel.org/r/20250124-ufshcd-fix-v4-1-c5d0144aae59@linaro.org Reviewed-by: Bean Huo Reviewed-by: Manivannan Sadhasivam Acked-by: Eric Biggers Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 31 +++++++++++++++++++++---------- drivers/ufs/host/ufshcd-pci.c | 2 -- drivers/ufs/host/ufshcd-pltfrm.c | 28 +++++++++------------------- include/ufs/ufshcd.h | 1 - 4 files changed, 30 insertions(+), 32 deletions(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index d3741b1f4382..d2de80b2bba4 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -10226,16 +10226,6 @@ int ufshcd_system_thaw(struct device *dev) EXPORT_SYMBOL_GPL(ufshcd_system_thaw); #endif /* CONFIG_PM_SLEEP */ -/** - * ufshcd_dealloc_host - deallocate Host Bus Adapter (HBA) - * @hba: pointer to Host Bus Adapter (HBA) - */ -void ufshcd_dealloc_host(struct ufs_hba *hba) -{ - scsi_host_put(hba->host); -} -EXPORT_SYMBOL_GPL(ufshcd_dealloc_host); - /** * ufshcd_set_dma_mask - Set dma mask based on the controller * addressing capability @@ -10254,12 +10244,26 @@ static int ufshcd_set_dma_mask(struct ufs_hba *hba) return dma_set_mask_and_coherent(hba->dev, DMA_BIT_MASK(32)); } +/** + * ufshcd_devres_release - devres cleanup handler, invoked during release of + * hba->dev + * @host: pointer to SCSI host + */ +static void ufshcd_devres_release(void *host) +{ + scsi_host_put(host); +} + /** * ufshcd_alloc_host - allocate Host Bus Adapter (HBA) * @dev: pointer to device handle * @hba_handle: driver private handle * * Return: 0 on success, non-zero value on failure. + * + * NOTE: There is no corresponding ufshcd_dealloc_host() because this function + * keeps track of its allocations using devres and deallocates everything on + * device removal automatically. */ int ufshcd_alloc_host(struct device *dev, struct ufs_hba **hba_handle) { @@ -10281,6 +10285,13 @@ int ufshcd_alloc_host(struct device *dev, struct ufs_hba **hba_handle) err = -ENOMEM; goto out_error; } + + err = devm_add_action_or_reset(dev, ufshcd_devres_release, + host); + if (err) + return dev_err_probe(dev, err, + "failed to add ufshcd dealloc action\n"); + host->nr_maps = HCTX_TYPE_POLL + 1; hba = shost_priv(host); hba->host = host; diff --git a/drivers/ufs/host/ufshcd-pci.c b/drivers/ufs/host/ufshcd-pci.c index ea39c5d5b8cf..9cfcaad23cf9 100644 --- a/drivers/ufs/host/ufshcd-pci.c +++ b/drivers/ufs/host/ufshcd-pci.c @@ -562,7 +562,6 @@ static void ufshcd_pci_remove(struct pci_dev *pdev) pm_runtime_forbid(&pdev->dev); pm_runtime_get_noresume(&pdev->dev); ufshcd_remove(hba); - ufshcd_dealloc_host(hba); } /** @@ -605,7 +604,6 @@ ufshcd_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) err = ufshcd_init(hba, mmio_base, pdev->irq); if (err) { dev_err(&pdev->dev, "Initialization failed\n"); - ufshcd_dealloc_host(hba); return err; } diff --git a/drivers/ufs/host/ufshcd-pltfrm.c b/drivers/ufs/host/ufshcd-pltfrm.c index 505572d4fa87..ffe5d1d2b215 100644 --- a/drivers/ufs/host/ufshcd-pltfrm.c +++ b/drivers/ufs/host/ufshcd-pltfrm.c @@ -465,21 +465,17 @@ int ufshcd_pltfrm_init(struct platform_device *pdev, struct device *dev = &pdev->dev; mmio_base = devm_platform_ioremap_resource(pdev, 0); - if (IS_ERR(mmio_base)) { - err = PTR_ERR(mmio_base); - goto out; - } + if (IS_ERR(mmio_base)) + return PTR_ERR(mmio_base); irq = platform_get_irq(pdev, 0); - if (irq < 0) { - err = irq; - goto out; - } + if (irq < 0) + return irq; err = ufshcd_alloc_host(dev, &hba); if (err) { dev_err(dev, "Allocation failed\n"); - goto out; + return err; } hba->vops = vops; @@ -488,13 +484,13 @@ int ufshcd_pltfrm_init(struct platform_device *pdev, if (err) { dev_err(dev, "%s: clock parse failed %d\n", __func__, err); - goto dealloc_host; + return err; } err = ufshcd_parse_regulator_info(hba); if (err) { dev_err(dev, "%s: regulator init failed %d\n", __func__, err); - goto dealloc_host; + return err; } ufshcd_init_lanes_per_dir(hba); @@ -502,25 +498,20 @@ int ufshcd_pltfrm_init(struct platform_device *pdev, err = ufshcd_parse_operating_points(hba); if (err) { dev_err(dev, "%s: OPP parse failed %d\n", __func__, err); - goto dealloc_host; + return err; } err = ufshcd_init(hba, mmio_base, irq); if (err) { dev_err_probe(dev, err, "Initialization failed with error %d\n", err); - goto dealloc_host; + return err; } pm_runtime_set_active(dev); pm_runtime_enable(dev); return 0; - -dealloc_host: - ufshcd_dealloc_host(hba); -out: - return err; } EXPORT_SYMBOL_GPL(ufshcd_pltfrm_init); @@ -534,7 +525,6 @@ void ufshcd_pltfrm_remove(struct platform_device *pdev) pm_runtime_get_sync(&pdev->dev); ufshcd_remove(hba); - ufshcd_dealloc_host(hba); pm_runtime_disable(&pdev->dev); pm_runtime_put_noidle(&pdev->dev); } diff --git a/include/ufs/ufshcd.h b/include/ufs/ufshcd.h index 650ff238cd74..8bf31e6ca4e5 100644 --- a/include/ufs/ufshcd.h +++ b/include/ufs/ufshcd.h @@ -1309,7 +1309,6 @@ static inline void ufshcd_rmwl(struct ufs_hba *hba, u32 mask, u32 val, u32 reg) void ufshcd_enable_irq(struct ufs_hba *hba); void ufshcd_disable_irq(struct ufs_hba *hba); int ufshcd_alloc_host(struct device *, struct ufs_hba **); -void ufshcd_dealloc_host(struct ufs_hba *); int ufshcd_hba_enable(struct ufs_hba *hba); int ufshcd_init(struct ufs_hba *, void __iomem *, unsigned int); int ufshcd_link_recovery(struct ufs_hba *hba); From 87c4b5e8a6b65189abd9ea5010ab308941f964a4 Mon Sep 17 00:00:00 2001 From: Long Li Date: Wed, 22 Jan 2025 19:07:22 -0800 Subject: [PATCH 0437/1443] scsi: storvsc: Set correct data length for sending SCSI command without payload In StorVSC, payload->range.len is used to indicate if this SCSI command carries payload. This data is allocated as part of the private driver data by the upper layer and may get passed to lower driver uninitialized. For example, the SCSI error handling mid layer may send TEST_UNIT_READY or REQUEST_SENSE while reusing the buffer from a failed command. The private data section may have stale data from the previous command. If the SCSI command doesn't carry payload, the driver may use this value as is for communicating with host, resulting in possible corruption. Fix this by always initializing this value. Fixes: be0cf6ca301c ("scsi: storvsc: Set the tablesize based on the information given by the host") Cc: stable@kernel.org Tested-by: Roman Kisel Reviewed-by: Roman Kisel Reviewed-by: Michael Kelley Signed-off-by: Long Li Link: https://lore.kernel.org/r/1737601642-7759-1-git-send-email-longli@linuxonhyperv.com Signed-off-by: Martin K. Petersen --- drivers/scsi/storvsc_drv.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c index 5a101ac06c47..a8614e54544e 100644 --- a/drivers/scsi/storvsc_drv.c +++ b/drivers/scsi/storvsc_drv.c @@ -1800,6 +1800,7 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd) length = scsi_bufflen(scmnd); payload = (struct vmbus_packet_mpb_array *)&cmd_request->mpb; + payload->range.len = 0; payload_sz = 0; if (scsi_sg_count(scmnd)) { From 1a78a56ea65252bb089e0daace989167227f2d31 Mon Sep 17 00:00:00 2001 From: Seunghui Lee Date: Sat, 18 Jan 2025 11:38:08 +0900 Subject: [PATCH 0438/1443] scsi: ufs: core: Fix error return with query response There is currently no mechanism to return error from query responses. Return the error and print the corresponding error message with it. Signed-off-by: Seunghui Lee Link: https://lore.kernel.org/r/20250118023808.24726-1-sh043.lee@samsung.com Reviewed-by: Bean Huo Reviewed-by: Bart Van Assche Signed-off-by: Martin K. Petersen --- drivers/ufs/core/ufshcd.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c index d2de80b2bba4..1893a7ad9531 100644 --- a/drivers/ufs/core/ufshcd.c +++ b/drivers/ufs/core/ufshcd.c @@ -3104,8 +3104,13 @@ ufshcd_dev_cmd_completion(struct ufs_hba *hba, struct ufshcd_lrb *lrbp) case UPIU_TRANSACTION_QUERY_RSP: { u8 response = lrbp->ucd_rsp_ptr->header.response; - if (response == 0) + if (response == 0) { err = ufshcd_copy_query_response(hba, lrbp); + } else { + err = -EINVAL; + dev_err(hba->dev, "%s: unexpected response in Query RSP: %x\n", + __func__, response); + } break; } case UPIU_TRANSACTION_REJECT_UPIU: From 5233e3235dec3065ccc632729675575dbe3c6b8a Mon Sep 17 00:00:00 2001 From: Magnus Lindholm Date: Sat, 25 Jan 2025 10:49:22 +0100 Subject: [PATCH 0439/1443] scsi: qla1280: Fix kernel oops when debug level > 2 A null dereference or oops exception will eventually occur when qla1280.c driver is compiled with DEBUG_QLA1280 enabled and ql_debug_level > 2. I think its clear from the code that the intention here is sg_dma_len(s) not length of sg_next(s) when printing the debug info. Signed-off-by: Magnus Lindholm Link: https://lore.kernel.org/r/20250125095033.26188-1-linmag7@gmail.com Signed-off-by: Martin K. Petersen --- drivers/scsi/qla1280.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/qla1280.c b/drivers/scsi/qla1280.c index 1fd2da0264e3..47d74f881948 100644 --- a/drivers/scsi/qla1280.c +++ b/drivers/scsi/qla1280.c @@ -2867,7 +2867,7 @@ qla1280_64bit_start_scsi(struct scsi_qla_host *ha, struct srb * sp) dprintk(3, "S/G Segment phys_addr=%x %x, len=0x%x\n", cpu_to_le32(upper_32_bits(dma_handle)), cpu_to_le32(lower_32_bits(dma_handle)), - cpu_to_le32(sg_dma_len(sg_next(s)))); + cpu_to_le32(sg_dma_len(s))); remseg--; } dprintk(5, "qla1280_64bit_start_scsi: Scatter/gather " From 1b0332a42656b798bea867631d739de023633ec6 Mon Sep 17 00:00:00 2001 From: Yu-Chun Lin Date: Thu, 30 Jan 2025 22:48:49 +0800 Subject: [PATCH 0440/1443] kthread: Fix return value on kzalloc() failure in kthread_affine_preferred() kthread_affine_preferred() incorrectly returns 0 instead of -ENOMEM when kzalloc() fails. Return 'ret' to ensure the correct error code is propagated. Fixes: 4d13f4304fa4 ("kthread: Implement preferred affinity") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501301528.t0cZVbnq-lkp@intel.com/ Signed-off-by: Yu-Chun Lin Signed-off-by: Frederic Weisbecker --- kernel/kthread.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/kthread.c b/kernel/kthread.c index 4005b13ebd7f..5dc5b0d7238e 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -859,7 +859,7 @@ int kthread_affine_preferred(struct task_struct *p, const struct cpumask *mask) struct kthread *kthread = to_kthread(p); cpumask_var_t affinity; unsigned long flags; - int ret; + int ret = 0; if (!wait_task_inactive(p, TASK_UNINTERRUPTIBLE) || kthread->started) { WARN_ON(1); @@ -892,7 +892,7 @@ int kthread_affine_preferred(struct task_struct *p, const struct cpumask *mask) out: free_cpumask_var(affinity); - return 0; + return ret; } /* From 244f8aa46fa9e2f4ea5fe0e04988b395d5e30fc7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 31 Jan 2025 17:30:37 -0800 Subject: [PATCH 0441/1443] ethtool: rss: fix hiding unsupported fields in dumps Commit ec6e57beaf8b ("ethtool: rss: don't report key if device doesn't support it") intended to stop reporting key fields for additional rss contexts if device has a global hashing key. Later we added dump support and the filtering wasn't properly added there. So we end up reporting the key fields in dumps but not in dos: # ./pyynl/cli.py --spec netlink/specs/ethtool.yaml --do rss-get \ --json '{"header": {"dev-index":2}, "context": 1 }' { "header": { ... }, "context": 1, "indir": [0, 1, 2, 3, ...]] } # ./pyynl/cli.py --spec netlink/specs/ethtool.yaml --dump rss-get [ ... snip context 0 ... { "header": { ... }, "context": 1, "indir": [0, 1, 2, 3, ...], -> "input_xfrm": 255, -> "hfunc": 1, -> "hkey": "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000" } ] Hide these fields correctly. The drivers/net/hw/rss_ctx.py selftest catches this when run on a device with single key, already: # Check| At /root/./ksft-net-drv/drivers/net/hw/rss_ctx.py, line 381, in test_rss_context_dump: # Check| ksft_ne(set(data.get('hkey', [1])), {0}, "key is all zero") # Check failed {0} == {0} key is all zero not ok 8 rss_ctx.test_rss_context_dump Fixes: f6122900f4e2 ("ethtool: rss: support dumping RSS contexts") Reviewed-by: Gal Pressman Reviewed-by: Joe Damato Link: https://patch.msgid.link/20250201013040.725123-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 7cb106b590ab..58df9ad02ce8 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -107,6 +107,8 @@ rss_prepare_ctx(const struct rss_req_info *request, struct net_device *dev, u32 total_size, indir_bytes; u8 *rss_config; + data->no_key_fields = !dev->ethtool_ops->rxfh_per_ctx_key; + ctx = xa_load(&dev->ethtool->rss_ctx, request->rss_context); if (!ctx) return -ENOENT; @@ -153,7 +155,6 @@ rss_prepare_data(const struct ethnl_req_info *req_base, if (!ops->cap_rss_ctx_supported && !ops->create_rxfh_context) return -EOPNOTSUPP; - data->no_key_fields = !ops->rxfh_per_ctx_key; return rss_prepare_ctx(request, dev, data, info); } From 2b91cc1214b165c25ac9b0885db89a0d3224028a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 31 Jan 2025 17:30:38 -0800 Subject: [PATCH 0442/1443] ethtool: ntuple: fix rss + ring_cookie check The info.flow_type is for RXFH commands, ntuple flow_type is inside the flow spec. The check currently does nothing, as info.flow_type is 0 (or even uninitialized by user space) for ETHTOOL_SRXCLSRLINS. Fixes: 9e43ad7a1ede ("net: ethtool: only allow set_rxnfc with rss + ring_cookie if driver opts in") Reviewed-by: Gal Pressman Reviewed-by: Joe Damato Link: https://patch.msgid.link/20250201013040.725123-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index 34bee42e1247..7609ce2b2c5e 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -993,7 +993,7 @@ static noinline_for_stack int ethtool_set_rxnfc(struct net_device *dev, return rc; /* Nonzero ring with RSS only makes sense if NIC adds them together */ - if (cmd == ETHTOOL_SRXCLSRLINS && info.flow_type & FLOW_RSS && + if (cmd == ETHTOOL_SRXCLSRLINS && info.fs.flow_type & FLOW_RSS && !ops->cap_rss_rxnfc_adds && ethtool_get_flow_spec_ring(info.fs.ring_cookie)) return -EINVAL; From de379dfd9ada2995699052f4a1ecebe5d8f8d70f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 31 Jan 2025 17:30:39 -0800 Subject: [PATCH 0443/1443] selftests: drv-net: rss_ctx: add missing cleanup in queue reconfigure Commit under Fixes adds ntuple rules but never deletes them. Fixes: 29a4bc1fe961 ("selftest: extend test_rss_context_queue_reconfigure for action addition") Reviewed-by: Joe Damato Link: https://patch.msgid.link/20250201013040.725123-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/rss_ctx.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py index ca8a7edff3dd..27e24e20749f 100755 --- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py +++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py @@ -252,6 +252,7 @@ def test_rss_queue_reconfigure(cfg, main_ctx=True): try: # this targets queue 4, which doesn't exist ntuple2 = ethtool_create(cfg, "-N", flow) + defer(ethtool, f"-N {cfg.ifname} delete {ntuple2}") except CmdExitFailure: pass else: @@ -260,6 +261,7 @@ def test_rss_queue_reconfigure(cfg, main_ctx=True): ethtool(f"-X {cfg.ifname} {ctx_ref} weight 1 0 1 0") # ntuple rule therefore targets queues 1 and 3 ntuple2 = ethtool_create(cfg, "-N", flow) + defer(ethtool, f"-N {cfg.ifname} delete {ntuple2}") # should replace existing filter ksft_eq(ntuple, ntuple2) _send_traffic_check(cfg, port, ctx_ref, { 'target': (1, 3), From c3da585509aeb8476886adf75a266c81a9b0df6c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 31 Jan 2025 17:30:40 -0800 Subject: [PATCH 0444/1443] selftests: drv-net: rss_ctx: don't fail reconfigure test if queue offset not supported Vast majority of drivers does not support queue offset. Simply return if the rss context + queue ntuple fails. Reviewed-by: Joe Damato Link: https://patch.msgid.link/20250201013040.725123-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/drivers/net/hw/rss_ctx.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py index 27e24e20749f..319aaa004c40 100755 --- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py +++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py @@ -260,7 +260,12 @@ def test_rss_queue_reconfigure(cfg, main_ctx=True): # change the table to target queues 0 and 2 ethtool(f"-X {cfg.ifname} {ctx_ref} weight 1 0 1 0") # ntuple rule therefore targets queues 1 and 3 - ntuple2 = ethtool_create(cfg, "-N", flow) + try: + ntuple2 = ethtool_create(cfg, "-N", flow) + except CmdExitFailure: + ksft_pr("Driver does not support rss + queue offset") + return + defer(ethtool, f"-N {cfg.ifname} delete {ntuple2}") # should replace existing filter ksft_eq(ntuple, ntuple2) From d3ed6dee73c560fad0a8e152c8e233b3fb3a2e44 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Sat, 1 Feb 2025 19:02:51 +0100 Subject: [PATCH 0445/1443] net: harmonize tstats and dstats After the blamed commits below, some UDP tunnel use dstats for accounting. On the xmit path, all the UDP-base tunnels ends up using iptunnel_xmit_stats() for stats accounting, and the latter assumes the relevant (tunnel) network device uses tstats. The end result is some 'funny' stat report for the mentioned UDP tunnel, e.g. when no packet is actually dropped and a bunch of packets are transmitted: gnv2: mtu 1450 qdisc noqueue \ state UNKNOWN mode DEFAULT group default qlen 1000 link/ether ee:7d:09:87:90:ea brd ff:ff:ff:ff:ff:ff RX: bytes packets errors dropped missed mcast 14916 23 0 15 0 0 TX: bytes packets errors dropped carrier collsns 0 1566 0 0 0 0 Address the issue ensuring the same binary layout for the overlapping fields of dstats and tstats. While this solution is a bit hackish, is smaller and with no performance pitfall compared to other alternatives i.e. supporting both dstat and tstat in iptunnel_xmit_stats() or reverting the blamed commit. With time we should possibly move all the IP-based tunnel (and virtual devices) to dstats. Fixes: c77200c07491 ("bareudp: Handle stats using NETDEV_PCPU_STAT_DSTATS.") Fixes: 6fa6de302246 ("geneve: Handle stats using NETDEV_PCPU_STAT_DSTATS.") Fixes: be226352e8dc ("vxlan: Handle stats using NETDEV_PCPU_STAT_DSTATS.") Signed-off-by: Paolo Abeni Reviewed-by: Guillaume Nault Link: https://patch.msgid.link/2e1c444cf0f63ae472baff29862c4c869be17031.1738432804.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 2 +- net/core/dev.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 2a59034a5fa2..03bb584c62cf 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2904,9 +2904,9 @@ struct pcpu_sw_netstats { struct pcpu_dstats { u64_stats_t rx_packets; u64_stats_t rx_bytes; - u64_stats_t rx_drops; u64_stats_t tx_packets; u64_stats_t tx_bytes; + u64_stats_t rx_drops; u64_stats_t tx_drops; struct u64_stats_sync syncp; } __aligned(8 * sizeof(u64)); diff --git a/net/core/dev.c b/net/core/dev.c index c0021cbd28fc..b91658e8aedb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -11286,6 +11286,20 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, const struct net_device_ops *ops = dev->netdev_ops; const struct net_device_core_stats __percpu *p; + /* + * IPv{4,6} and udp tunnels share common stat helpers and use + * different stat type (NETDEV_PCPU_STAT_TSTATS vs + * NETDEV_PCPU_STAT_DSTATS). Ensure the accounting is consistent. + */ + BUILD_BUG_ON(offsetof(struct pcpu_sw_netstats, rx_bytes) != + offsetof(struct pcpu_dstats, rx_bytes)); + BUILD_BUG_ON(offsetof(struct pcpu_sw_netstats, rx_packets) != + offsetof(struct pcpu_dstats, rx_packets)); + BUILD_BUG_ON(offsetof(struct pcpu_sw_netstats, tx_bytes) != + offsetof(struct pcpu_dstats, tx_bytes)); + BUILD_BUG_ON(offsetof(struct pcpu_sw_netstats, tx_packets) != + offsetof(struct pcpu_dstats, tx_packets)); + if (ops->ndo_get_stats64) { memset(storage, 0, sizeof(*storage)); ops->ndo_get_stats64(dev, storage); From a787ab73e2e43c0a3df10bc8d9b9b7a679129d49 Mon Sep 17 00:00:00 2001 From: Jithu Joseph Date: Fri, 31 Jan 2025 12:53:15 -0800 Subject: [PATCH 0446/1443] platform/x86/intel/ifs: Update documentation with image download path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The documentation previously listed the path to download In Field Scan (IFS) test images as "TBD". Update the documentation to include the correct image download location. Also move the download link to the appropriate section within the documentation. Reported-by: Anisse Astier Signed-off-by: Jithu Joseph Link: https://lore.kernel.org/r/20250131205315.1585663-1-jithu.joseph@intel.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/intel/ifs/ifs.h | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/platform/x86/intel/ifs/ifs.h b/drivers/platform/x86/intel/ifs/ifs.h index 5c3c0dfa1bf8..f369fb0d3d82 100644 --- a/drivers/platform/x86/intel/ifs/ifs.h +++ b/drivers/platform/x86/intel/ifs/ifs.h @@ -23,12 +23,14 @@ * IFS Image * --------- * - * Intel provides a firmware file containing the scan tests via - * github [#f1]_. Similar to microcode there is a separate file for each + * Intel provides firmware files containing the scan tests via the webpage [#f1]_. + * Look under "In-Field Scan Test Images Download" section towards the + * end of the page. Similar to microcode, there are separate files for each * family-model-stepping. IFS Images are not applicable for some test types. * Wherever applicable the sysfs directory would provide a "current_batch" file * (see below) for loading the image. * + * .. [#f1] https://intel.com/InFieldScan * * IFS Image Loading * ----------------- @@ -125,9 +127,6 @@ * 2) Hardware allows for some number of cores to be tested in parallel. * The driver does not make use of this, it only tests one core at a time. * - * .. [#f1] https://github.com/intel/TBD - * - * * Structural Based Functional Test at Field (SBAF): * ------------------------------------------------- * From 1739cafdb8decad538410b05a4640055408826de Mon Sep 17 00:00:00 2001 From: Jinghao Jia Date: Mon, 3 Feb 2025 02:55:04 -0600 Subject: [PATCH 0447/1443] samples/hid: remove unnecessary -I flags from libbpf EXTRA_CFLAGS Commit 5a6ea7022ff4 ("samples/bpf: Remove unnecessary -I flags from libbpf EXTRA_CFLAGS") fixed the build error caused by redundant include path for samples/bpf, but not samples/hid. Apply the same fix on samples/hid as well. Fixes: 13b25489b6f8 ("kbuild: change working directory to external module directory with M=") Tested-by: Ruowen Qin Signed-off-by: Jinghao Jia Link: https://patch.msgid.link/20250203085506.220297-2-jinghao7@illinois.edu Signed-off-by: Benjamin Tissoires --- samples/hid/Makefile | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/samples/hid/Makefile b/samples/hid/Makefile index 8ea59e9631a3..69159c81d045 100644 --- a/samples/hid/Makefile +++ b/samples/hid/Makefile @@ -40,16 +40,17 @@ BPF_EXTRA_CFLAGS += -I$(srctree)/arch/mips/include/asm/mach-generic endif endif -TPROGS_CFLAGS += -Wall -O2 -TPROGS_CFLAGS += -Wmissing-prototypes -TPROGS_CFLAGS += -Wstrict-prototypes +COMMON_CFLAGS += -Wall -O2 +COMMON_CFLAGS += -Wmissing-prototypes +COMMON_CFLAGS += -Wstrict-prototypes +TPROGS_CFLAGS += $(COMMON_CFLAGS) TPROGS_CFLAGS += -I$(objtree)/usr/include TPROGS_CFLAGS += -I$(LIBBPF_INCLUDE) TPROGS_CFLAGS += -I$(srctree)/tools/include ifdef SYSROOT -TPROGS_CFLAGS += --sysroot=$(SYSROOT) +COMMON_CFLAGS += --sysroot=$(SYSROOT) TPROGS_LDFLAGS := -L$(SYSROOT)/usr/lib endif @@ -112,7 +113,7 @@ clean: $(LIBBPF): $(wildcard $(LIBBPF_SRC)/*.[ch] $(LIBBPF_SRC)/Makefile) | $(LIBBPF_OUTPUT) # Fix up variables inherited from Kbuild that tools/ build system won't like - $(MAKE) -C $(LIBBPF_SRC) RM='rm -rf' EXTRA_CFLAGS="$(TPROGS_CFLAGS)" \ + $(MAKE) -C $(LIBBPF_SRC) RM='rm -rf' EXTRA_CFLAGS="$(COMMON_CFLAGS)" \ LDFLAGS=$(TPROGS_LDFLAGS) srctree=$(HID_SAMPLES_PATH)/../../ \ O= OUTPUT=$(LIBBPF_OUTPUT)/ DESTDIR=$(LIBBPF_DESTDIR) prefix= \ $@ install_headers From 8b125949df58a00e8797c6e6d3f3d3dc08f4d939 Mon Sep 17 00:00:00 2001 From: Jinghao Jia Date: Mon, 3 Feb 2025 02:55:06 -0600 Subject: [PATCH 0448/1443] samples/hid: fix broken vmlinux path for VMLINUX_BTF Commit 13b25489b6f8 ("kbuild: change working directory to external module directory with M=") changed kbuild working directory of hid-bpf sample programs to samples/hid, which broke the vmlinux path for VMLINUX_BTF, as the Makefiles assume the current work directory to be the kernel output directory and use a relative path (i.e., ./vmlinux): Makefile:173: *** Cannot find a vmlinux for VMLINUX_BTF at any of " /path/to/linux/samples/hid/vmlinux", build the kernel or set VMLINUX_BTF or VMLINUX_H variable. Stop. Correctly refer to the kernel output directory using $(objtree). Fixes: 13b25489b6f8 ("kbuild: change working directory to external module directory with M=") Tested-by: Ruowen Qin Suggested-by: Daniel Borkmann Suggested-by: Andrii Nakryiko Signed-off-by: Jinghao Jia Link: https://patch.msgid.link/20250203085506.220297-4-jinghao7@illinois.edu Signed-off-by: Benjamin Tissoires --- samples/hid/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/samples/hid/Makefile b/samples/hid/Makefile index 69159c81d045..db5a077c77fc 100644 --- a/samples/hid/Makefile +++ b/samples/hid/Makefile @@ -164,7 +164,7 @@ $(obj)/hid_surface_dial.o: $(obj)/hid_surface_dial.skel.h VMLINUX_BTF_PATHS ?= $(abspath $(if $(O),$(O)/vmlinux)) \ $(abspath $(if $(KBUILD_OUTPUT),$(KBUILD_OUTPUT)/vmlinux)) \ - $(abspath ./vmlinux) + $(abspath $(objtree)/vmlinux) VMLINUX_BTF ?= $(abspath $(firstword $(wildcard $(VMLINUX_BTF_PATHS)))) $(obj)/vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL) From 32392e04cb50d87bb7a6a7d9213f44a1a0961820 Mon Sep 17 00:00:00 2001 From: Oliver Upton Date: Mon, 3 Feb 2025 15:15:43 -0800 Subject: [PATCH 0449/1443] KVM: arm64: Fail protected mode init if no vgic hardware is present Protected mode assumes that at minimum vgic-v3 is present, however KVM fails to actually enforce this at the time of initialization. As such, when running protected mode in a half-baked state on GICv2 hardware we see the hyp go belly up at vcpu_load() when it tries to restore the vgic-v3 cpuif: $ ./arch_timer_edge_cases [ 130.599140] kvm [4518]: nVHE hyp panic at: [] __kvm_nvhe___vgic_v3_restore_vmcr_aprs+0x8/0x84! [ 130.603685] kvm [4518]: Cannot dump pKVM nVHE stacktrace: !CONFIG_PROTECTED_NVHE_STACKTRACE [ 130.611962] kvm [4518]: Hyp Offset: 0xfffeca95ed000000 [ 130.617053] Kernel panic - not syncing: HYP panic: [ 130.617053] PS:800003c9 PC:0000b56a94102b58 ESR:0000000002000000 [ 130.617053] FAR:ffff00007b98d4d0 HPFAR:00000000007b98d0 PAR:0000000000000000 [ 130.617053] VCPU:0000000000000000 [ 130.638013] CPU: 0 UID: 0 PID: 4518 Comm: arch_timer_edge Tainted: G C 6.13.0-rc3-00009-gf7d03fcbf1f4 #1 [ 130.648790] Tainted: [C]=CRAP [ 130.651721] Hardware name: Libre Computer AML-S905X-CC (DT) [ 130.657242] Call trace: [ 130.659656] show_stack+0x18/0x24 (C) [ 130.663279] dump_stack_lvl+0x38/0x90 [ 130.666900] dump_stack+0x18/0x24 [ 130.670178] panic+0x388/0x3e8 [ 130.673196] nvhe_hyp_panic_handler+0x104/0x208 [ 130.677681] kvm_arch_vcpu_load+0x290/0x548 [ 130.681821] vcpu_load+0x50/0x80 [ 130.685013] kvm_arch_vcpu_ioctl_run+0x30/0x868 [ 130.689498] kvm_vcpu_ioctl+0x2e0/0x974 [ 130.693293] __arm64_sys_ioctl+0xb4/0xec [ 130.697174] invoke_syscall+0x48/0x110 [ 130.700883] el0_svc_common.constprop.0+0x40/0xe0 [ 130.705540] do_el0_svc+0x1c/0x28 [ 130.708818] el0_svc+0x30/0xd0 [ 130.711837] el0t_64_sync_handler+0x10c/0x138 [ 130.716149] el0t_64_sync+0x198/0x19c [ 130.719774] SMP: stopping secondary CPUs [ 130.723660] Kernel Offset: disabled [ 130.727103] CPU features: 0x000,00000800,02800000,0200421b [ 130.732537] Memory Limit: none [ 130.735561] ---[ end Kernel panic - not syncing: HYP panic: [ 130.735561] PS:800003c9 PC:0000b56a94102b58 ESR:0000000002000000 [ 130.735561] FAR:ffff00007b98d4d0 HPFAR:00000000007b98d0 PAR:0000000000000000 [ 130.735561] VCPU:0000000000000000 ]--- Fix it by failing KVM initialization if the system doesn't implement vgic-v3, as protected mode will never do anything useful on such hardware. Reported-by: Mark Brown Closes: https://lore.kernel.org/kvmarm/5ca7588c-7bf2-4352-8661-e4a56a9cd9aa@sirena.org.uk/ Signed-off-by: Oliver Upton Link: https://lore.kernel.org/r/20250203231543.233511-1-oliver.upton@linux.dev Signed-off-by: Marc Zyngier --- arch/arm64/kvm/arm.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 0725a0b50a3e..62c650c2f7b6 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -2290,6 +2290,19 @@ static int __init init_subsystems(void) break; case -ENODEV: case -ENXIO: + /* + * No VGIC? No pKVM for you. + * + * Protected mode assumes that VGICv3 is present, so no point + * in trying to hobble along if vgic initialization fails. + */ + if (is_protected_kvm_enabled()) + goto out; + + /* + * Otherwise, userspace could choose to implement a GIC for its + * guest on non-cooperative hardware. + */ vgic_present = false; err = 0; break; From 3648027de1fa91a0c80cffd3ecff263d06e62605 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Fri, 10 Jan 2025 18:51:46 +0100 Subject: [PATCH 0450/1443] arm64: Fix 5-level paging support in kexec/hibernate trampoline Add the missing code to allocate P4D level page tables when cloning the the kernel page tables. This fixes a crash that may be observed when attempting to resume from hibernation on an LPA2 capable system with 4k pages, which therefore uses 5 levels of paging. Presumably, kexec is equally affected. Signed-off-by: Ard Biesheuvel Link: https://lore.kernel.org/r/20250110175145.785702-2-ardb+git@google.com Signed-off-by: Will Deacon --- arch/arm64/mm/trans_pgd.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/arm64/mm/trans_pgd.c b/arch/arm64/mm/trans_pgd.c index 19c67ed1a21f..18543b603c77 100644 --- a/arch/arm64/mm/trans_pgd.c +++ b/arch/arm64/mm/trans_pgd.c @@ -162,6 +162,13 @@ static int copy_p4d(struct trans_pgd_info *info, pgd_t *dst_pgdp, unsigned long next; unsigned long addr = start; + if (pgd_none(READ_ONCE(*dst_pgdp))) { + dst_p4dp = trans_alloc(info); + if (!dst_p4dp) + return -ENOMEM; + pgd_populate(NULL, dst_pgdp, dst_p4dp); + } + dst_p4dp = p4d_offset(dst_pgdp, start); src_p4dp = p4d_offset(src_pgdp, start); do { From f458b2165d7ac0f2401fff48f19c8f864e7e1e38 Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Fri, 17 Jan 2025 07:55:22 -0500 Subject: [PATCH 0451/1443] arm64: Kconfig: Remove selecting replaced HAVE_FUNCTION_GRAPH_RETVAL Commit a3ed4157b7d8 ("fgraph: Replace fgraph_ret_regs with ftrace_regs") replaces the config HAVE_FUNCTION_GRAPH_RETVAL with the config HAVE_FUNCTION_GRAPH_FREGS, and it replaces all the select commands in the various architecture Kconfig files. In the arm64 architecture, the commit adds the 'select HAVE_FUNCTION_GRAPH_FREGS', but misses to remove the 'select HAVE_FUNCTION_GRAPH_RETVAL', i.e., the select on the replaced config. Remove selecting the replaced config. No functional change, just cleanup. Fixes: a3ed4157b7d8 ("fgraph: Replace fgraph_ret_regs with ftrace_regs") Signed-off-by: Lukas Bulwahn Link: https://lore.kernel.org/r/20250117125522.99071-1-lukas.bulwahn@redhat.com Signed-off-by: Will Deacon --- arch/arm64/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index fcdd0ed3eca8..940343beb3d4 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -225,7 +225,6 @@ config ARM64 select HAVE_FUNCTION_ERROR_INJECTION select HAVE_FUNCTION_GRAPH_FREGS select HAVE_FUNCTION_GRAPH_TRACER - select HAVE_FUNCTION_GRAPH_RETVAL select HAVE_GCC_PLUGINS select HAVE_HARDLOCKUP_DETECTOR_PERF if PERF_EVENTS && \ HW_PERF_EVENTS && HAVE_PERF_EVENTS_NMI From f64f9dddd1f58c41c140034f7d2b0beeef1bc548 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 24 Jan 2025 17:33:22 +0000 Subject: [PATCH 0452/1443] arm64/gcs: Fix documentation for HWCAP In one of the renumberings of the GCS hwcap a stray reference to HWCAP2 was left, fix it. Reported-by: David Spickett Fixes: 7058bf87cd59 ("arm64/gcs: Document the ABI for Guarded Control Stacks") Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20250124-arm64-gcs-hwcap-doc-v1-1-fa9368b01ca6@kernel.org Signed-off-by: Will Deacon --- Documentation/arch/arm64/gcs.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/arch/arm64/gcs.rst b/Documentation/arch/arm64/gcs.rst index 1f65a3193e77..226c0b008456 100644 --- a/Documentation/arch/arm64/gcs.rst +++ b/Documentation/arch/arm64/gcs.rst @@ -37,7 +37,7 @@ intended to be exhaustive. shadow stacks rather than GCS. * Support for GCS is reported to userspace via HWCAP_GCS in the aux vector - AT_HWCAP2 entry. + AT_HWCAP entry. * GCS is enabled per thread. While there is support for disabling GCS at runtime this should be done with great care. From e5e8367d25abcf1ed5f9f84df7c856bd4e1e12c9 Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Wed, 22 Jan 2025 11:03:56 +0530 Subject: [PATCH 0453/1443] drm/dp: Add the DPCD register required for Extended wake timeout Add DPCD registers required to configure Extended Wake Timeout for LTTPR. Signed-off-by: Suraj Kandpal Reviewed-by: Arun R Murthy Link: https://patchwork.freedesktop.org/patch/msgid/20250122053358.1545039-2-suraj.kandpal@intel.com --- include/drm/display/drm_dp.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h index a6f8b098c56f..480370bba1de 100644 --- a/include/drm/display/drm_dp.h +++ b/include/drm/display/drm_dp.h @@ -696,6 +696,9 @@ #define DP_UPSTREAM_DEVICE_DP_PWR_NEED 0x118 /* 1.2 */ # define DP_PWR_NOT_NEEDED (1 << 0) +#define DP_EXTENDED_DPRX_SLEEP_WAKE_TIMEOUT_GRANT 0x119 /* 1.4a */ +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_GRANTED (1 << 0) + #define DP_FEC_CONFIGURATION 0x120 /* 1.4 */ # define DP_FEC_READY (1 << 0) # define DP_FEC_ERR_COUNT_SEL_MASK (7 << 1) @@ -1168,6 +1171,15 @@ # define DP_VSC_EXT_CEA_SDP_SUPPORTED (1 << 6) /* DP 1.4 */ # define DP_VSC_EXT_CEA_SDP_CHAINING_SUPPORTED (1 << 7) /* DP 1.4 */ +#define DP_EXTENDED_DPRX_SLEEP_WAKE_TIMEOUT_REQUEST 0x2211 /* 1.4a */ +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_MASK 0xff +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_1_MS 0x00 +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_20_MS 0x01 +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_40_MS 0x02 +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_60_MS 0x03 +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_80_MS 0x04 +# define DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_100_MS 0x05 + #define DP_DPRX_FEATURE_ENUMERATION_LIST_CONT_1 0x2214 /* 2.0 E11 */ # define DP_ADAPTIVE_SYNC_SDP_SUPPORTED (1 << 0) # define DP_ADAPTIVE_SYNC_SDP_OPERATION_MODE GENMASK(1, 0) @@ -1473,6 +1485,8 @@ #define DP_MAX_LANE_COUNT_PHY_REPEATER 0xf0004 /* 1.4a */ #define DP_Repeater_FEC_CAPABILITY 0xf0004 /* 1.4 */ #define DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT 0xf0005 /* 1.4a */ +# define DP_EXTENDED_WAKE_TIMEOUT_REQUEST_MASK 0x7f +# define DP_EXTENDED_WAKE_TIMEOUT_GRANT (1 << 7) #define DP_MAIN_LINK_CHANNEL_CODING_PHY_REPEATER 0xf0006 /* 2.0 */ # define DP_PHY_REPEATER_128B132B_SUPPORTED (1 << 0) /* See DP_128B132B_SUPPORTED_LINK_RATES for values */ From eaf53ac4901fbb06a94cc0b7842567b3d13c0492 Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Wed, 22 Jan 2025 11:03:57 +0530 Subject: [PATCH 0454/1443] drm/display/dp: Define function to setup Extended wake time Extended wake timeout request helps to give additional time by reading the DPCD register through which sink requests the minimal amount of time required to wake the sink up. Source device shall keep retying the AUX tansaction till the extended timeout that is being granted for LTTPRs from the sink device. --v2 -Add documentation [Dmitry] Spec: DP v2.1 Section 3.6.12.3 Signed-off-by: Suraj Kandpal Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20250122053358.1545039-3-suraj.kandpal@intel.com --- drivers/gpu/drm/display/drm_dp_helper.c | 58 +++++++++++++++++++++++++ include/drm/display/drm_dp_helper.h | 1 + 2 files changed, 59 insertions(+) diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c index da3c8521a7fa..c488d160a3c1 100644 --- a/drivers/gpu/drm/display/drm_dp_helper.c +++ b/drivers/gpu/drm/display/drm_dp_helper.c @@ -459,6 +459,64 @@ void drm_dp_lttpr_link_train_channel_eq_delay(const struct drm_dp_aux *aux, } EXPORT_SYMBOL(drm_dp_lttpr_link_train_channel_eq_delay); +/** + * drm_dp_lttpr_wake_timeout_setup() - Grant extended time for sink to wake up + * @aux: The DP AUX channel to use + * @transparent_mode: This is true if lttpr is in transparent mode + * + * This function checks if the sink needs any extended wake time, if it does + * it grants this request. Post this setup the source device can keep trying + * the Aux transaction till the granted wake timeout. + * If this function is not called all Aux transactions are expected to take + * a default of 1ms before they throw an error. + */ +void drm_dp_lttpr_wake_timeout_setup(struct drm_dp_aux *aux, bool transparent_mode) +{ + u8 val = 1; + int ret; + + if (transparent_mode) { + static const u8 timeout_mapping[] = { + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_1_MS] = 1, + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_20_MS] = 20, + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_40_MS] = 40, + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_60_MS] = 60, + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_80_MS] = 80, + [DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_100_MS] = 100, + }; + + ret = drm_dp_dpcd_readb(aux, DP_EXTENDED_DPRX_SLEEP_WAKE_TIMEOUT_REQUEST, &val); + if (ret != 1) { + drm_dbg_kms(aux->drm_dev, + "Failed to read Extended sleep wake timeout request\n"); + return; + } + + val = (val < sizeof(timeout_mapping) && timeout_mapping[val]) ? + timeout_mapping[val] : 1; + + if (val > 1) + drm_dp_dpcd_writeb(aux, + DP_EXTENDED_DPRX_SLEEP_WAKE_TIMEOUT_GRANT, + DP_DPRX_SLEEP_WAKE_TIMEOUT_PERIOD_GRANTED); + } else { + ret = drm_dp_dpcd_readb(aux, DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT, &val); + if (ret != 1) { + drm_dbg_kms(aux->drm_dev, + "Failed to read Extended sleep wake timeout request\n"); + return; + } + + val = (val & DP_EXTENDED_WAKE_TIMEOUT_REQUEST_MASK) ? + (val & DP_EXTENDED_WAKE_TIMEOUT_REQUEST_MASK) * 10 : 1; + + if (val > 1) + drm_dp_dpcd_writeb(aux, DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT, + DP_EXTENDED_WAKE_TIMEOUT_GRANT); + } +} +EXPORT_SYMBOL(drm_dp_lttpr_wake_timeout_setup); + u8 drm_dp_link_rate_to_bw_code(int link_rate) { switch (link_rate) { diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h index 8f4054a56039..89a34dff85a4 100644 --- a/include/drm/display/drm_dp_helper.h +++ b/include/drm/display/drm_dp_helper.h @@ -633,6 +633,7 @@ int drm_dp_lttpr_max_link_rate(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]); int drm_dp_lttpr_max_lane_count(const u8 caps[DP_LTTPR_COMMON_CAP_SIZE]); bool drm_dp_lttpr_voltage_swing_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]); bool drm_dp_lttpr_pre_emphasis_level_3_supported(const u8 caps[DP_LTTPR_PHY_CAP_SIZE]); +void drm_dp_lttpr_wake_timeout_setup(struct drm_dp_aux *aux, bool transparent_mode); void drm_dp_remote_aux_init(struct drm_dp_aux *aux); void drm_dp_aux_init(struct drm_dp_aux *aux); From 242d9bf59a0a8293acf3e67264875abd0b573614 Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Wed, 22 Jan 2025 11:03:58 +0530 Subject: [PATCH 0455/1443] drm/i915/lttpr: Enable Extended Wake Timeout Usually retimers take around 30 to 40ms to exit all devices from sleep state. Extended wake timeout mechanism helps to give that additional time. --v2 -Grant the requested time only if greater than 1ms [Arun/Jani] -Reframe commit message [Arun] --v3 -Move the function to drm_core [Dmitry/Jani] Spec: DP v2.1 Section 3.6.12.3 Signed-off-by: Suraj Kandpal Reviewed-by: Arun R Murthy Acked-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/20250122053358.1545039-4-suraj.kandpal@intel.com --- drivers/gpu/drm/i915/display/intel_ddi.c | 4 ++++ drivers/gpu/drm/i915/display/intel_dp_link_training.c | 2 +- drivers/gpu/drm/i915/display/intel_dp_link_training.h | 1 + 3 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 49b5cc01ce40..a70e7f263a89 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -2546,6 +2546,7 @@ static void mtl_ddi_pre_enable_dp(struct intel_atomic_state *state, { struct intel_dp *intel_dp = enc_to_intel_dp(encoder); bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST); + bool transparent_mode; intel_dp_set_link_params(intel_dp, crtc_state->port_clock, @@ -2600,6 +2601,9 @@ static void mtl_ddi_pre_enable_dp(struct intel_atomic_state *state, if (!is_mst) intel_dp_set_power(intel_dp, DP_SET_POWER_D0); + transparent_mode = intel_dp_lttpr_transparent_mode_enabled(intel_dp); + drm_dp_lttpr_wake_timeout_setup(&intel_dp->aux, transparent_mode); + intel_dp_configure_protocol_converter(intel_dp, crtc_state); if (!is_mst) intel_dp_sink_enable_decompression(state, diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c index 397cc4ebae52..c911d4a19e62 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c @@ -127,7 +127,7 @@ intel_dp_set_lttpr_transparent_mode(struct intel_dp *intel_dp, bool enable) return true; } -static bool intel_dp_lttpr_transparent_mode_enabled(struct intel_dp *intel_dp) +bool intel_dp_lttpr_transparent_mode_enabled(struct intel_dp *intel_dp) { return intel_dp->lttpr_common_caps[DP_PHY_REPEATER_MODE - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV] == diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.h b/drivers/gpu/drm/i915/display/intel_dp_link_training.h index 2066b9146762..46614124569f 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.h +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.h @@ -15,6 +15,7 @@ struct intel_dp; int intel_dp_read_dprx_caps(struct intel_dp *intel_dp, u8 dpcd[DP_RECEIVER_CAP_SIZE]); int intel_dp_init_lttpr_and_dprx_caps(struct intel_dp *intel_dp); +bool intel_dp_lttpr_transparent_mode_enabled(struct intel_dp *intel_dp); void intel_dp_link_training_set_mode(struct intel_dp *intel_dp, int link_rate, bool is_vrr); From 21fed7c223e20e694b91dbf25936d922a50c8b19 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Mon, 3 Feb 2025 20:11:04 +0000 Subject: [PATCH 0456/1443] arm64/hwcap: Remove stray references to SF8MMx Due to SME currently being disabled when removing the SF8MMx support it wasn't noticed that there were some stray references in the hwcap table, delete them. Fixes: 819935464cb2 ("arm64/hwcap: Describe 2024 dpISA extensions to userspace") Signed-off-by: Mark Brown Link: https://lore.kernel.org/r/20250203-arm64-remove-sf8mmx-v1-1-6f1da3dbff82@kernel.org Signed-off-by: Will Deacon --- arch/arm64/kernel/cpufeature.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c index 4eb7c6698ae4..f0910f20fbf8 100644 --- a/arch/arm64/kernel/cpufeature.c +++ b/arch/arm64/kernel/cpufeature.c @@ -3180,8 +3180,6 @@ static const struct arm64_cpu_capabilities arm64_elf_hwcaps[] = { HWCAP_CAP(ID_AA64SMFR0_EL1, SF8FMA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8FMA), HWCAP_CAP(ID_AA64SMFR0_EL1, SF8DP4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP4), HWCAP_CAP(ID_AA64SMFR0_EL1, SF8DP2, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP2), - HWCAP_CAP(ID_AA64SMFR0_EL1, SF8MM8, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8MM8), - HWCAP_CAP(ID_AA64SMFR0_EL1, SF8MM4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8MM4), HWCAP_CAP(ID_AA64SMFR0_EL1, SBitPerm, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SBITPERM), HWCAP_CAP(ID_AA64SMFR0_EL1, AES, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_AES), HWCAP_CAP(ID_AA64SMFR0_EL1, SFEXPA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SFEXPA), From ba69e0750b0362870294adab09339a0c39c3beaf Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Sat, 1 Feb 2025 18:21:35 +0100 Subject: [PATCH 0457/1443] efi: Avoid cold plugged memory for placing the kernel UEFI 2.11 introduced EFI_MEMORY_HOT_PLUGGABLE to annotate system memory regions that are 'cold plugged' at boot, i.e., hot pluggable memory that is available from early boot, and described as system RAM by the firmware. Existing loaders and EFI applications running in the boot context will happily use this memory for allocating data structures that cannot be freed or moved at runtime, and this prevents the memory from being unplugged. Going forward, the new EFI_MEMORY_HOT_PLUGGABLE attribute should be tested, and memory annotated as such should be avoided for such allocations. In the EFI stub, there are a couple of occurrences where, instead of the high-level AllocatePages() UEFI boot service, a low-level code sequence is used that traverses the EFI memory map and carves out the requested number of pages from a free region. This is needed, e.g., for allocating as low as possible, or for allocating pages at random. While AllocatePages() should presumably avoid special purpose memory and cold plugged regions, this manual approach needs to incorporate this logic itself, in order to prevent the kernel itself from ending up in a hot unpluggable region, preventing it from being unplugged. So add the EFI_MEMORY_HOTPLUGGABLE macro definition, and check for it where appropriate. Cc: stable@vger.kernel.org Signed-off-by: Ard Biesheuvel --- drivers/firmware/efi/efi.c | 6 ++++-- drivers/firmware/efi/libstub/randomalloc.c | 3 +++ drivers/firmware/efi/libstub/relocate.c | 3 +++ include/linux/efi.h | 1 + 4 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c index 8296bf985d1d..7309394b8fc9 100644 --- a/drivers/firmware/efi/efi.c +++ b/drivers/firmware/efi/efi.c @@ -934,13 +934,15 @@ char * __init efi_md_typeattr_format(char *buf, size_t size, EFI_MEMORY_WB | EFI_MEMORY_UCE | EFI_MEMORY_RO | EFI_MEMORY_WP | EFI_MEMORY_RP | EFI_MEMORY_XP | EFI_MEMORY_NV | EFI_MEMORY_SP | EFI_MEMORY_CPU_CRYPTO | - EFI_MEMORY_RUNTIME | EFI_MEMORY_MORE_RELIABLE)) + EFI_MEMORY_MORE_RELIABLE | EFI_MEMORY_HOT_PLUGGABLE | + EFI_MEMORY_RUNTIME)) snprintf(pos, size, "|attr=0x%016llx]", (unsigned long long)attr); else snprintf(pos, size, - "|%3s|%2s|%2s|%2s|%2s|%2s|%2s|%2s|%2s|%3s|%2s|%2s|%2s|%2s]", + "|%3s|%2s|%2s|%2s|%2s|%2s|%2s|%2s|%2s|%2s|%3s|%2s|%2s|%2s|%2s]", attr & EFI_MEMORY_RUNTIME ? "RUN" : "", + attr & EFI_MEMORY_HOT_PLUGGABLE ? "HP" : "", attr & EFI_MEMORY_MORE_RELIABLE ? "MR" : "", attr & EFI_MEMORY_CPU_CRYPTO ? "CC" : "", attr & EFI_MEMORY_SP ? "SP" : "", diff --git a/drivers/firmware/efi/libstub/randomalloc.c b/drivers/firmware/efi/libstub/randomalloc.c index e5872e38d9a4..5a732018be36 100644 --- a/drivers/firmware/efi/libstub/randomalloc.c +++ b/drivers/firmware/efi/libstub/randomalloc.c @@ -25,6 +25,9 @@ static unsigned long get_entry_num_slots(efi_memory_desc_t *md, if (md->type != EFI_CONVENTIONAL_MEMORY) return 0; + if (md->attribute & EFI_MEMORY_HOT_PLUGGABLE) + return 0; + if (efi_soft_reserve_enabled() && (md->attribute & EFI_MEMORY_SP)) return 0; diff --git a/drivers/firmware/efi/libstub/relocate.c b/drivers/firmware/efi/libstub/relocate.c index 99b45d1cd624..d4264bfb6dc1 100644 --- a/drivers/firmware/efi/libstub/relocate.c +++ b/drivers/firmware/efi/libstub/relocate.c @@ -53,6 +53,9 @@ efi_status_t efi_low_alloc_above(unsigned long size, unsigned long align, if (desc->type != EFI_CONVENTIONAL_MEMORY) continue; + if (desc->attribute & EFI_MEMORY_HOT_PLUGGABLE) + continue; + if (efi_soft_reserve_enabled() && (desc->attribute & EFI_MEMORY_SP)) continue; diff --git a/include/linux/efi.h b/include/linux/efi.h index 053c57e61869..db293d7de686 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -128,6 +128,7 @@ typedef struct { #define EFI_MEMORY_RO ((u64)0x0000000000020000ULL) /* read-only */ #define EFI_MEMORY_SP ((u64)0x0000000000040000ULL) /* soft reserved */ #define EFI_MEMORY_CPU_CRYPTO ((u64)0x0000000000080000ULL) /* supports encryption */ +#define EFI_MEMORY_HOT_PLUGGABLE BIT_ULL(20) /* supports unplugging at runtime */ #define EFI_MEMORY_RUNTIME ((u64)0x8000000000000000ULL) /* range requires runtime mapping */ #define EFI_MEMORY_DESCRIPTOR_VERSION 1 From bbc4578537e350d5bf8a7a2c7d054d6b163b3c41 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Sat, 1 Feb 2025 18:21:36 +0100 Subject: [PATCH 0458/1443] efi: Use BIT_ULL() constants for memory attributes For legibility, use the existing BIT_ULL() to generate the u64 type EFI memory attribute macros. Signed-off-by: Ard Biesheuvel --- include/linux/efi.h | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/include/linux/efi.h b/include/linux/efi.h index db293d7de686..7d63d1d75f22 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -114,22 +114,22 @@ typedef struct { #define EFI_MAX_MEMORY_TYPE 16 /* Attribute values: */ -#define EFI_MEMORY_UC ((u64)0x0000000000000001ULL) /* uncached */ -#define EFI_MEMORY_WC ((u64)0x0000000000000002ULL) /* write-coalescing */ -#define EFI_MEMORY_WT ((u64)0x0000000000000004ULL) /* write-through */ -#define EFI_MEMORY_WB ((u64)0x0000000000000008ULL) /* write-back */ -#define EFI_MEMORY_UCE ((u64)0x0000000000000010ULL) /* uncached, exported */ -#define EFI_MEMORY_WP ((u64)0x0000000000001000ULL) /* write-protect */ -#define EFI_MEMORY_RP ((u64)0x0000000000002000ULL) /* read-protect */ -#define EFI_MEMORY_XP ((u64)0x0000000000004000ULL) /* execute-protect */ -#define EFI_MEMORY_NV ((u64)0x0000000000008000ULL) /* non-volatile */ -#define EFI_MEMORY_MORE_RELIABLE \ - ((u64)0x0000000000010000ULL) /* higher reliability */ -#define EFI_MEMORY_RO ((u64)0x0000000000020000ULL) /* read-only */ -#define EFI_MEMORY_SP ((u64)0x0000000000040000ULL) /* soft reserved */ -#define EFI_MEMORY_CPU_CRYPTO ((u64)0x0000000000080000ULL) /* supports encryption */ +#define EFI_MEMORY_UC BIT_ULL(0) /* uncached */ +#define EFI_MEMORY_WC BIT_ULL(1) /* write-coalescing */ +#define EFI_MEMORY_WT BIT_ULL(2) /* write-through */ +#define EFI_MEMORY_WB BIT_ULL(3) /* write-back */ +#define EFI_MEMORY_UCE BIT_ULL(4) /* uncached, exported */ +#define EFI_MEMORY_WP BIT_ULL(12) /* write-protect */ +#define EFI_MEMORY_RP BIT_ULL(13) /* read-protect */ +#define EFI_MEMORY_XP BIT_ULL(14) /* execute-protect */ +#define EFI_MEMORY_NV BIT_ULL(15) /* non-volatile */ +#define EFI_MEMORY_MORE_RELIABLE BIT_ULL(16) /* higher reliability */ +#define EFI_MEMORY_RO BIT_ULL(17) /* read-only */ +#define EFI_MEMORY_SP BIT_ULL(18) /* soft reserved */ +#define EFI_MEMORY_CPU_CRYPTO BIT_ULL(19) /* supports encryption */ #define EFI_MEMORY_HOT_PLUGGABLE BIT_ULL(20) /* supports unplugging at runtime */ -#define EFI_MEMORY_RUNTIME ((u64)0x8000000000000000ULL) /* range requires runtime mapping */ +#define EFI_MEMORY_RUNTIME BIT_ULL(63) /* range requires runtime mapping */ + #define EFI_MEMORY_DESCRIPTOR_VERSION 1 #define EFI_PAGE_SHIFT 12 From f269e5eac4dbf55d89409257ec794c9c3c51e515 Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Tue, 4 Feb 2025 09:51:58 +0100 Subject: [PATCH 0459/1443] drm/drm_mode_object: fix typo in kerneldoc Remove unintended extra word. Reviewed-by: Dmitry Baryshkov Acked-by: Louis Chauvet Signed-off-by: Luca Ceresoli Link: https://patchwork.freedesktop.org/patch/msgid/20250204-drm-small-improvements-v4-1-d6bbc92f12f1@bootlin.com Signed-off-by: Louis Chauvet --- include/drm/drm_mode_object.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/drm/drm_mode_object.h b/include/drm/drm_mode_object.h index 08d7a7f0188f..c68edbd126d0 100644 --- a/include/drm/drm_mode_object.h +++ b/include/drm/drm_mode_object.h @@ -35,7 +35,7 @@ struct drm_file; * @id: userspace visible identifier * @type: type of the object, one of DRM_MODE_OBJECT\_\* * @properties: properties attached to this object, including values - * @refcount: reference count for objects which with dynamic lifetime + * @refcount: reference count for objects with dynamic lifetime * @free_cb: free function callback, only set for objects with dynamic lifetime * * Base structure for modeset objects visible to userspace. Objects can be From 51d262a96bc6c87f5dababf5136c978afa59f28e Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Tue, 4 Feb 2025 09:51:59 +0100 Subject: [PATCH 0460/1443] drm/atomic-helper: improve CRTC enabled/connectors mismatch logging message This message reports a mismatch between new_crtc_state->enable and has_connectors, which should be either both true or both false. However it does not mention which one is true and which is false, which can be useful for debugging. Add the value of both avriables to the log message. Reviewed-by: Dmitry Baryshkov Acked-by: Louis Chauvet Signed-off-by: Luca Ceresoli Link: https://patchwork.freedesktop.org/patch/msgid/20250204-drm-small-improvements-v4-2-d6bbc92f12f1@bootlin.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/drm_atomic_helper.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 3034ba09c0ee..26bf2465cee7 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -690,8 +690,9 @@ drm_atomic_helper_check_modeset(struct drm_device *dev, } if (new_crtc_state->enable != has_connectors) { - drm_dbg_atomic(dev, "[CRTC:%d:%s] enabled/connectors mismatch\n", - crtc->base.id, crtc->name); + drm_dbg_atomic(dev, "[CRTC:%d:%s] enabled/connectors mismatch (%d/%d)\n", + crtc->base.id, crtc->name, + new_crtc_state->enable, has_connectors); return -EINVAL; } From e8ed246ded863eb862806c5591afdcf70012ab5e Mon Sep 17 00:00:00 2001 From: Andre Werner Date: Tue, 21 Jan 2025 08:18:19 +0100 Subject: [PATCH 0461/1443] serial: sc16is7xx: Fix IRQ number check behavior The logical meaning of the previous version is wrong due to a typo. If the IRQ equals 0, no interrupt pin is available and polling mode shall be used. Additionally, this fix adds a check for IRQ < 0 to increase robustness, because documentation still says that negative IRQ values cannot be absolutely ruled-out. Fixes: 104c1b9dde9d ("serial: sc16is7xx: Add polling mode if no IRQ pin is available") Signed-off-by: Andre Werner Reviewed-by: Jiri Slaby Reviewed-by: Andy Shevchenko Reviewed-by: Maarten Brock Reviewed-by: Hugo Villeneuve Link: https://lore.kernel.org/r/20250121071819.1346672-1-andre.werner@systec-electronic.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/sc16is7xx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/sc16is7xx.c b/drivers/tty/serial/sc16is7xx.c index 7b51cdc274fd..560f45ed19ae 100644 --- a/drivers/tty/serial/sc16is7xx.c +++ b/drivers/tty/serial/sc16is7xx.c @@ -1561,7 +1561,7 @@ int sc16is7xx_probe(struct device *dev, const struct sc16is7xx_devtype *devtype, /* Always ask for fixed clock rate from a property. */ device_property_read_u32(dev, "clock-frequency", &uartclk); - s->polling = !!irq; + s->polling = (irq <= 0); if (s->polling) dev_dbg(dev, "No interrupt pin definition, falling back to polling mode\n"); From 8cca475b8085e551b3eebf4f2164e56c800402e0 Mon Sep 17 00:00:00 2001 From: Paz Zcharya Date: Wed, 29 Jan 2025 14:20:58 +0000 Subject: [PATCH 0462/1443] drm/vkms: Add support for ABGR8888 pixel format Add support for pixel format ABGR8888, which is the default format on Android devices. This will allow us to use VKMS as the default display driver in Android Emulator (Cuttlefish) and increase VKMS adoption. Signed-off-by: Paz Zcharya Reviewed-by: Louis Chauvet Link: https://patchwork.freedesktop.org/patch/msgid/20250129142238.562999-1-pazz@google.com Signed-off-by: Louis Chauvet --- drivers/gpu/drm/vkms/vkms_formats.c | 32 +++++++++++++++++++++++++++ drivers/gpu/drm/vkms/vkms_plane.c | 1 + drivers/gpu/drm/vkms/vkms_writeback.c | 1 + 3 files changed, 34 insertions(+) diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c index 39b1d7c97d45..30a64ecca87c 100644 --- a/drivers/gpu/drm/vkms/vkms_formats.c +++ b/drivers/gpu/drm/vkms/vkms_formats.c @@ -253,6 +253,26 @@ static void XRGB8888_read_line(const struct vkms_plane_state *plane, int x_start } } +static void ABGR8888_read_line(const struct vkms_plane_state *plane, int x_start, int y_start, + enum pixel_read_direction direction, int count, + struct pixel_argb_u16 out_pixel[]) +{ + struct pixel_argb_u16 *end = out_pixel + count; + u8 *src_pixels; + + packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels); + + int step = get_block_step_bytes(plane->frame_info->fb, direction, 0); + + while (out_pixel < end) { + u8 *px = (u8 *)src_pixels; + /* Switch blue and red pixels. */ + *out_pixel = argb_u16_from_u8888(px[3], px[0], px[1], px[2]); + out_pixel += 1; + src_pixels += step; + } +} + static void ARGB16161616_read_line(const struct vkms_plane_state *plane, int x_start, int y_start, enum pixel_read_direction direction, int count, struct pixel_argb_u16 out_pixel[]) @@ -344,6 +364,14 @@ static void argb_u16_to_XRGB8888(u8 *out_pixel, const struct pixel_argb_u16 *in_ out_pixel[0] = DIV_ROUND_CLOSEST(in_pixel->b, 257); } +static void argb_u16_to_ABGR8888(u8 *out_pixel, const struct pixel_argb_u16 *in_pixel) +{ + out_pixel[3] = DIV_ROUND_CLOSEST(in_pixel->a, 257); + out_pixel[2] = DIV_ROUND_CLOSEST(in_pixel->b, 257); + out_pixel[1] = DIV_ROUND_CLOSEST(in_pixel->g, 257); + out_pixel[0] = DIV_ROUND_CLOSEST(in_pixel->r, 257); +} + static void argb_u16_to_ARGB16161616(u8 *out_pixel, const struct pixel_argb_u16 *in_pixel) { __le16 *pixel = (__le16 *)out_pixel; @@ -420,6 +448,8 @@ pixel_read_line_t get_pixel_read_line_function(u32 format) return &ARGB8888_read_line; case DRM_FORMAT_XRGB8888: return &XRGB8888_read_line; + case DRM_FORMAT_ABGR8888: + return &ABGR8888_read_line; case DRM_FORMAT_ARGB16161616: return &ARGB16161616_read_line; case DRM_FORMAT_XRGB16161616: @@ -453,6 +483,8 @@ pixel_write_t get_pixel_write_function(u32 format) return &argb_u16_to_ARGB8888; case DRM_FORMAT_XRGB8888: return &argb_u16_to_XRGB8888; + case DRM_FORMAT_ABGR8888: + return &argb_u16_to_ABGR8888; case DRM_FORMAT_ARGB16161616: return &argb_u16_to_ARGB16161616; case DRM_FORMAT_XRGB16161616: diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c index e2fce471870f..e34f8c7f83c3 100644 --- a/drivers/gpu/drm/vkms/vkms_plane.c +++ b/drivers/gpu/drm/vkms/vkms_plane.c @@ -15,6 +15,7 @@ static const u32 vkms_formats[] = { DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB8888, + DRM_FORMAT_ABGR8888, DRM_FORMAT_XRGB16161616, DRM_FORMAT_ARGB16161616, DRM_FORMAT_RGB565 diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c index e9b5c74d7c58..fe163271d5b5 100644 --- a/drivers/gpu/drm/vkms/vkms_writeback.c +++ b/drivers/gpu/drm/vkms/vkms_writeback.c @@ -17,6 +17,7 @@ static const u32 vkms_wb_formats[] = { DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB8888, + DRM_FORMAT_ABGR8888, DRM_FORMAT_XRGB16161616, DRM_FORMAT_ARGB16161616, DRM_FORMAT_RGB565 From 166ac2bba167d575e7146beaa66093bc7c072f43 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:46 +0200 Subject: [PATCH 0463/1443] serial: port: Assign ->iotype correctly when ->iobase is set Currently the ->iotype is always assigned to the UPIO_MEM when the respective property is not found. However, this will not support the cases when user wants to have UPIO_PORT to be set or preserved. Support this scenario by checking ->iobase value and default the ->iotype respectively. Fixes: 1117a6fdc7c1 ("serial: 8250_of: Switch to use uart_read_port_properties()") Fixes: e894b6005dce ("serial: port: Introduce a common helper to read properties") Cc: stable Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-2-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/serial_port.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/serial_port.c b/drivers/tty/serial/serial_port.c index d35f1d24156c..f28d0633fe6b 100644 --- a/drivers/tty/serial/serial_port.c +++ b/drivers/tty/serial/serial_port.c @@ -173,6 +173,7 @@ EXPORT_SYMBOL(uart_remove_one_port); * The caller is responsible to initialize the following fields of the @port * ->dev (must be valid) * ->flags + * ->iobase * ->mapbase * ->mapsize * ->regshift (if @use_defaults is false) @@ -214,7 +215,7 @@ static int __uart_read_properties(struct uart_port *port, bool use_defaults) /* Read the registers I/O access type (default: MMIO 8-bit) */ ret = device_property_read_u32(dev, "reg-io-width", &value); if (ret) { - port->iotype = UPIO_MEM; + port->iotype = port->iobase ? UPIO_PORT : UPIO_MEM; } else { switch (value) { case 1: From e8486bd50ecf63c9a1e25271f258a8d959f2672f Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:47 +0200 Subject: [PATCH 0464/1443] serial: port: Always update ->iotype in __uart_read_properties() The documentation of the __uart_read_properties() states that ->iotype member is always altered after the function call, but the code doesn't do that in the case when use_defaults == false and the value of reg-io-width is unsupported. Make sure the code follows the documentation. Note, the current users of the uart_read_and_validate_port_properties() will fail and the change doesn't affect their behaviour, neither users of uart_read_port_properties() will be affected since the alteration happens there even in the current code flow. Fixes: e894b6005dce ("serial: port: Introduce a common helper to read properties") Cc: stable Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-3-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/serial_port.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/serial_port.c b/drivers/tty/serial/serial_port.c index f28d0633fe6b..85285c56fabf 100644 --- a/drivers/tty/serial/serial_port.c +++ b/drivers/tty/serial/serial_port.c @@ -228,11 +228,11 @@ static int __uart_read_properties(struct uart_port *port, bool use_defaults) port->iotype = device_is_big_endian(dev) ? UPIO_MEM32BE : UPIO_MEM32; break; default: + port->iotype = UPIO_UNKNOWN; if (!use_defaults) { dev_err(dev, "Unsupported reg-io-width (%u)\n", value); return -EINVAL; } - port->iotype = UPIO_UNKNOWN; break; } } From 12397549b5014071e1d2b315509f68eb93ef9144 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:48 +0200 Subject: [PATCH 0465/1443] serial: port: Make ->iotype validation global in __uart_read_properties() In order to make code robust against potential changes in the future move ->iotype validation outside of switch in __uart_read_properties(). If any code will be added in between that might leave the ->iotype value unknown the validation catches this up. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-4-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/serial_port.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/tty/serial/serial_port.c b/drivers/tty/serial/serial_port.c index 85285c56fabf..2fc48cd63f6c 100644 --- a/drivers/tty/serial/serial_port.c +++ b/drivers/tty/serial/serial_port.c @@ -229,14 +229,15 @@ static int __uart_read_properties(struct uart_port *port, bool use_defaults) break; default: port->iotype = UPIO_UNKNOWN; - if (!use_defaults) { - dev_err(dev, "Unsupported reg-io-width (%u)\n", value); - return -EINVAL; - } break; } } + if (!use_defaults && port->iotype == UPIO_UNKNOWN) { + dev_err(dev, "Unsupported reg-io-width (%u)\n", value); + return -EINVAL; + } + /* Read the address mapping base offset (default: no offset) */ ret = device_property_read_u32(dev, "reg-offset", &value); if (ret) From fe310f75327edbc042c7cc0df32c6b9ec29eb93a Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:49 +0200 Subject: [PATCH 0466/1443] serial: 8250_of: Remove unneeded ->iotype assignment If ->iobase is set the default will be UPIO_PORT for ->iotype after the uart_read_and_validate_port_properties() call. Hence no need to assign that explicitly. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-5-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_of.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/tty/serial/8250/8250_of.c b/drivers/tty/serial/8250/8250_of.c index 64aed7efc569..11c860ea80f6 100644 --- a/drivers/tty/serial/8250/8250_of.c +++ b/drivers/tty/serial/8250/8250_of.c @@ -110,7 +110,6 @@ static int of_platform_serial_setup(struct platform_device *ofdev, spin_lock_init(&port->lock); if (resource_type(&resource) == IORESOURCE_IO) { - port->iotype = UPIO_PORT; port->iobase = resource.start; } else { port->mapbase = resource.start; From 34bbb5d5137f32be3186a995a5ad4c60aaad11a7 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:50 +0200 Subject: [PATCH 0467/1443] serial: 8250_platform: Remove unneeded ->iotype assignment If ->iobase is set the default will be UPIO_PORT for ->iotype after the uart_read_and_validate_port_properties() call. Hence no need to assign that explicitly. Otherwise it will be UPIO_MEM. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-6-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_platform.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/tty/serial/8250/8250_platform.c b/drivers/tty/serial/8250/8250_platform.c index 8bdc1879d952..c0343bfb8064 100644 --- a/drivers/tty/serial/8250/8250_platform.c +++ b/drivers/tty/serial/8250/8250_platform.c @@ -112,7 +112,6 @@ static int serial8250_probe_acpi(struct platform_device *pdev) struct device *dev = &pdev->dev; struct uart_8250_port uart = { }; struct resource *regs; - unsigned char iotype; int ret, line; regs = platform_get_mem_or_io(pdev, 0); @@ -122,13 +121,11 @@ static int serial8250_probe_acpi(struct platform_device *pdev) switch (resource_type(regs)) { case IORESOURCE_IO: uart.port.iobase = regs->start; - iotype = UPIO_PORT; break; case IORESOURCE_MEM: uart.port.mapbase = regs->start; uart.port.mapsize = resource_size(regs); uart.port.flags = UPF_IOREMAP; - iotype = UPIO_MEM; break; default: return -EINVAL; @@ -147,12 +144,6 @@ static int serial8250_probe_acpi(struct platform_device *pdev) if (ret) return ret; - /* - * The previous call may not set iotype correctly when reg-io-width - * property is absent and it doesn't support IO port resource. - */ - uart.port.iotype = iotype; - line = serial8250_register_8250_port(&uart); if (line < 0) return line; From 0f3fd9cf6491f5beecbb65abb41556c56135340c Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 24 Jan 2025 18:10:51 +0200 Subject: [PATCH 0468/1443] serial: 8250_pnp: Remove unneeded ->iotype assignment If ->iobase is set the default will be UPIO_PORT for ->iotype after the uart_read_and_validate_port_properties() call. Hence no need to assign that explicitly. Otherwise it will be UPIO_MEM. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250124161530.398361-7-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_pnp.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/tty/serial/8250/8250_pnp.c b/drivers/tty/serial/8250/8250_pnp.c index 7c06ae79d8e2..7a837fdf9df1 100644 --- a/drivers/tty/serial/8250/8250_pnp.c +++ b/drivers/tty/serial/8250/8250_pnp.c @@ -436,7 +436,6 @@ serial_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) { struct uart_8250_port uart, *port; int ret, flags = dev_id->driver_data; - unsigned char iotype; long line; if (flags & UNKNOWN_DEV) { @@ -448,14 +447,11 @@ serial_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) memset(&uart, 0, sizeof(uart)); if ((flags & CIR_PORT) && pnp_port_valid(dev, 2)) { uart.port.iobase = pnp_port_start(dev, 2); - iotype = UPIO_PORT; } else if (pnp_port_valid(dev, 0)) { uart.port.iobase = pnp_port_start(dev, 0); - iotype = UPIO_PORT; } else if (pnp_mem_valid(dev, 0)) { uart.port.mapbase = pnp_mem_start(dev, 0); uart.port.mapsize = pnp_mem_len(dev, 0); - iotype = UPIO_MEM; uart.port.flags = UPF_IOREMAP; } else return -ENODEV; @@ -471,12 +467,6 @@ serial_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *dev_id) if (ret) return ret; - /* - * The previous call may not set iotype correctly when reg-io-width - * property is absent and it doesn't support IO port resource. - */ - uart.port.iotype = iotype; - if (flags & CIR_PORT) { uart.port.flags |= UPF_FIXED_PORT | UPF_FIXED_TYPE; uart.port.type = PORT_8250_CIR; From 78d5d1e20d1de9422f013338a0f2311448588ba7 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 31 Jan 2025 16:37:13 +0100 Subject: [PATCH 0469/1443] drm/xe/relay: Don't use GFP_KERNEL for new transactions VFs use a relay transaction during the resume/reset flow and use of the GFP_KERNEL flag may conflict with the reclaim: -> #0 (fs_reclaim){+.+.}-{0:0}: [ ] __lock_acquire+0x1874/0x2bc0 [ ] lock_acquire+0xd2/0x310 [ ] fs_reclaim_acquire+0xc5/0x100 [ ] mempool_alloc_noprof+0x5c/0x1b0 [ ] __relay_get_transaction+0xdc/0xa10 [xe] [ ] relay_send_to+0x251/0xe50 [xe] [ ] xe_guc_relay_send_to_pf+0x79/0x3a0 [xe] [ ] xe_gt_sriov_vf_connect+0x90/0x4d0 [xe] [ ] xe_uc_init_hw+0x157/0x3b0 [xe] [ ] do_gt_restart+0x1ae/0x650 [xe] [ ] xe_gt_resume+0xb6/0x120 [xe] [ ] xe_pm_runtime_resume+0x15b/0x370 [xe] [ ] xe_pci_runtime_resume+0x73/0x90 [xe] [ ] pci_pm_runtime_resume+0xa0/0x100 [ ] __rpm_callback+0x4d/0x170 [ ] rpm_callback+0x64/0x70 [ ] rpm_resume+0x594/0x790 [ ] __pm_runtime_resume+0x4e/0x90 [ ] xe_pm_runtime_get_ioctl+0x9c/0x160 [xe] Since we have a preallocated pool of relay transactions, which should cover all our normal relay use cases, we may use the GFP_NOWAIT flag when allocating new outgoing transactions. Signed-off-by: Michal Wajdeczko Tested-by: Marcin Bernatowicz Reviewed-by: Marcin Bernatowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250131153713.808-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_guc_relay.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_guc_relay.c b/drivers/gpu/drm/xe/xe_guc_relay.c index 8f62de026724..e5dc94f3e618 100644 --- a/drivers/gpu/drm/xe/xe_guc_relay.c +++ b/drivers/gpu/drm/xe/xe_guc_relay.c @@ -225,7 +225,7 @@ __relay_get_transaction(struct xe_guc_relay *relay, bool incoming, u32 remote, u * with CTB lock held which is marked as used in the reclaim path. * Btw, that's one of the reason why we use mempool here! */ - txn = mempool_alloc(&relay->pool, incoming ? GFP_ATOMIC : GFP_KERNEL); + txn = mempool_alloc(&relay->pool, incoming ? GFP_ATOMIC : GFP_NOWAIT); if (!txn) return ERR_PTR(-ENOMEM); From 4241a702e0d0c2ca9364cfac08dbf134264962de Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 3 Feb 2025 11:03:04 +0000 Subject: [PATCH 0470/1443] rxrpc: Fix the rxrpc_connection attend queue handling The rxrpc_connection attend queue is never used because conn::attend_link is never initialised and so is always NULL'd out and thus always appears to be busy. This requires the following fix: (1) Fix this the attend queue problem by initialising conn::attend_link. And, consequently, two further fixes for things masked by the above bug: (2) Fix rxrpc_input_conn_event() to handle being invoked with a NULL sk_buff pointer - something that can now happen with the above change. (3) Fix the RXRPC_SKB_MARK_SERVICE_CONN_SECURED message to carry a pointer to the connection and a ref on it. Signed-off-by: David Howells cc: Marc Dionne cc: Jakub Kicinski cc: "David S. Miller" cc: Eric Dumazet cc: Paolo Abeni cc: Simon Horman cc: linux-afs@lists.infradead.org cc: netdev@vger.kernel.org Fixes: f2cce89a074e ("rxrpc: Implement a mechanism to send an event notification to a connection") Link: https://patch.msgid.link/20250203110307.7265-3-dhowells@redhat.com Signed-off-by: Paolo Abeni --- include/trace/events/rxrpc.h | 1 + net/rxrpc/conn_event.c | 17 ++++++++++------- net/rxrpc/conn_object.c | 1 + 3 files changed, 12 insertions(+), 7 deletions(-) diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 2f119d18a061..cad50d91077e 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -219,6 +219,7 @@ EM(rxrpc_conn_get_conn_input, "GET inp-conn") \ EM(rxrpc_conn_get_idle, "GET idle ") \ EM(rxrpc_conn_get_poke_abort, "GET pk-abort") \ + EM(rxrpc_conn_get_poke_secured, "GET secured ") \ EM(rxrpc_conn_get_poke_timer, "GET poke ") \ EM(rxrpc_conn_get_service_conn, "GET svc-conn") \ EM(rxrpc_conn_new_client, "NEW client ") \ diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 713e04394ceb..74bb49b936cd 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -272,6 +272,7 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, * we've already received the packet, put it on the * front of the queue. */ + sp->conn = rxrpc_get_connection(conn, rxrpc_conn_get_poke_secured); skb->mark = RXRPC_SKB_MARK_SERVICE_CONN_SECURED; rxrpc_get_skb(skb, rxrpc_skb_get_conn_secured); skb_queue_head(&conn->local->rx_queue, skb); @@ -437,14 +438,16 @@ void rxrpc_input_conn_event(struct rxrpc_connection *conn, struct sk_buff *skb) if (test_and_clear_bit(RXRPC_CONN_EV_ABORT_CALLS, &conn->events)) rxrpc_abort_calls(conn); - switch (skb->mark) { - case RXRPC_SKB_MARK_SERVICE_CONN_SECURED: - if (conn->state != RXRPC_CONN_SERVICE) - break; + if (skb) { + switch (skb->mark) { + case RXRPC_SKB_MARK_SERVICE_CONN_SECURED: + if (conn->state != RXRPC_CONN_SERVICE) + break; - for (loop = 0; loop < RXRPC_MAXCALLS; loop++) - rxrpc_call_is_secure(conn->channels[loop].call); - break; + for (loop = 0; loop < RXRPC_MAXCALLS; loop++) + rxrpc_call_is_secure(conn->channels[loop].call); + break; + } } /* Process delayed ACKs whose time has come. */ diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 7eba4d7d9a38..2f1fd1e2e7e4 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -67,6 +67,7 @@ struct rxrpc_connection *rxrpc_alloc_connection(struct rxrpc_net *rxnet, INIT_WORK(&conn->destructor, rxrpc_clean_up_connection); INIT_LIST_HEAD(&conn->proc_link); INIT_LIST_HEAD(&conn->link); + INIT_LIST_HEAD(&conn->attend_link); mutex_init(&conn->security_lock); mutex_init(&conn->tx_data_alloc_lock); skb_queue_head_init(&conn->rx_queue); From 459777724d306315070d24608fcd89aea85516d6 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Fri, 31 Jan 2025 19:25:02 +0100 Subject: [PATCH 0471/1443] drm/xe/vf: Don't try to trigger a full GT reset if VF VFs don't have access to the GDRST(0x941c) register that driver uses to reset a GT. Attempt to trigger a reset using debugfs: $ cat /sys/kernel/debug/dri/0000:00:02.1/gt0/force_reset or due to a hang condition detected by the driver leads to: [ ] xe 0000:00:02.1: [drm] GT0: trying reset from force_reset [xe] [ ] xe 0000:00:02.1: [drm] GT0: reset queued [ ] xe 0000:00:02.1: [drm] GT0: reset started [ ] ------------[ cut here ]------------ [ ] xe 0000:00:02.1: [drm] GT0: VF is trying to write 0x1 to an inaccessible register 0x941c+0x0 [ ] WARNING: CPU: 3 PID: 3069 at drivers/gpu/drm/xe/xe_gt_sriov_vf.c:996 xe_gt_sriov_vf_write32+0xc6/0x580 [xe] [ ] RIP: 0010:xe_gt_sriov_vf_write32+0xc6/0x580 [xe] [ ] Call Trace: [ ] [ ] ? show_regs+0x6c/0x80 [ ] ? __warn+0x93/0x1c0 [ ] ? xe_gt_sriov_vf_write32+0xc6/0x580 [xe] [ ] ? report_bug+0x182/0x1b0 [ ] ? handle_bug+0x6e/0xb0 [ ] ? exc_invalid_op+0x18/0x80 [ ] ? asm_exc_invalid_op+0x1b/0x20 [ ] ? xe_gt_sriov_vf_write32+0xc6/0x580 [xe] [ ] ? xe_gt_sriov_vf_write32+0xc6/0x580 [xe] [ ] ? xe_gt_tlb_invalidation_reset+0xef/0x110 [xe] [ ] ? __mutex_unlock_slowpath+0x41/0x2e0 [ ] xe_mmio_write32+0x64/0x150 [xe] [ ] do_gt_reset+0x2f/0xa0 [xe] [ ] gt_reset_worker+0x14e/0x1e0 [xe] [ ] process_one_work+0x21c/0x740 [ ] worker_thread+0x1db/0x3c0 Fix that by sending H2G VF_RESET(0x5507) action instead. Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4078 Signed-off-by: Michal Wajdeczko Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20250131182502.852-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_gt.c | 4 ++++ drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 16 ++++++++++++++++ drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 1 + 3 files changed, 21 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 01a4a852b8f4..9fb8f1e678dc 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -32,6 +32,7 @@ #include "xe_gt_pagefault.h" #include "xe_gt_printk.h" #include "xe_gt_sriov_pf.h" +#include "xe_gt_sriov_vf.h" #include "xe_gt_sysfs.h" #include "xe_gt_tlb_invalidation.h" #include "xe_gt_topology.h" @@ -679,6 +680,9 @@ static int do_gt_reset(struct xe_gt *gt) { int err; + if (IS_SRIOV_VF(gt_to_xe(gt))) + return xe_gt_sriov_vf_reset(gt); + xe_gsc_wa_14015076503(gt, true); xe_mmio_write32(>->mmio, GDRST, GRDOM_FULL); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c index 6671030439fd..4831549da319 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c @@ -58,6 +58,22 @@ static int vf_reset_guc_state(struct xe_gt *gt) return err; } +/** + * xe_gt_sriov_vf_reset - Reset GuC VF internal state. + * @gt: the &xe_gt + * + * It requires functional `GuC MMIO based communication`_. + * + * Return: 0 on success or a negative error code on failure. + */ +int xe_gt_sriov_vf_reset(struct xe_gt *gt) +{ + if (!xe_device_uc_enabled(gt_to_xe(gt))) + return -ENODEV; + + return vf_reset_guc_state(gt); +} + static int guc_action_match_version(struct xe_guc *guc, u32 wanted_branch, u32 wanted_major, u32 wanted_minor, u32 *branch, u32 *major, u32 *minor, u32 *patch) diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h index 912d20814261..ba6c5d74e326 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h @@ -12,6 +12,7 @@ struct drm_printer; struct xe_gt; struct xe_reg; +int xe_gt_sriov_vf_reset(struct xe_gt *gt); int xe_gt_sriov_vf_bootstrap(struct xe_gt *gt); int xe_gt_sriov_vf_query_config(struct xe_gt *gt); int xe_gt_sriov_vf_connect(struct xe_gt *gt); From 5417a2e9b130a78bf48cb4cf92630efcee5ccf38 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 4 Feb 2025 14:55:54 +0000 Subject: [PATCH 0472/1443] KVM: arm64: Fix nested S2 MMU structures reallocation For each vcpu that userspace creates, we allocate a number of s2_mmu structures that will eventually contain our shadow S2 page tables. Since this is a dynamically allocated array, we reallocate the array and initialise the newly allocated elements. Once everything is correctly initialised, we adjust pointer and size in the kvm structure, and move on. But should that initialisation fail *and* the reallocation triggered a copy to another location, we end-up returning early, with the kvm structure still containing the (now stale) old pointer. Weeee! Cure it by assigning the pointer early, and use this to perform the initialisation. If everything succeeds, we adjust the size. Otherwise, we just leave the size as it was, no harm done, and the new memory is as good as the ol' one (we hope...). Fixes: 4f128f8e1aaac ("KVM: arm64: nv: Support multiple nested Stage-2 mmu structures") Reported-by: Alexander Potapenko Tested-by: Alexander Potapenko Link: https://lore.kernel.org/r/20250204145554.774427-1-maz@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/kvm/nested.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 33d2ace68665..0c9387d2f507 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -67,26 +67,27 @@ int kvm_vcpu_init_nested(struct kvm_vcpu *vcpu) if (!tmp) return -ENOMEM; + swap(kvm->arch.nested_mmus, tmp); + /* * If we went through a realocation, adjust the MMU back-pointers in * the previously initialised kvm_pgtable structures. */ if (kvm->arch.nested_mmus != tmp) for (int i = 0; i < kvm->arch.nested_mmus_size; i++) - tmp[i].pgt->mmu = &tmp[i]; + kvm->arch.nested_mmus[i].pgt->mmu = &kvm->arch.nested_mmus[i]; for (int i = kvm->arch.nested_mmus_size; !ret && i < num_mmus; i++) - ret = init_nested_s2_mmu(kvm, &tmp[i]); + ret = init_nested_s2_mmu(kvm, &kvm->arch.nested_mmus[i]); if (ret) { for (int i = kvm->arch.nested_mmus_size; i < num_mmus; i++) - kvm_free_stage2_pgd(&tmp[i]); + kvm_free_stage2_pgd(&kvm->arch.nested_mmus[i]); return ret; } kvm->arch.nested_mmus_size = num_mmus; - kvm->arch.nested_mmus = tmp; return 0; } From b450dcce93bc2cf6d2bfaf5a0de88a94ebad8f89 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 4 Feb 2025 11:00:48 +0000 Subject: [PATCH 0473/1443] KVM: arm64: timer: Always evaluate the need for a soft timer When updating the interrupt state for an emulated timer, we return early and skip the setup of a soft timer that runs in parallel with the guest. While this is OK if we have set the interrupt pending, it is pretty wrong if the guest moved CVAL into the future. In that case, no timer is armed and the guest can wait for a very long time (it will take a full put/load cycle for the situation to resolve). This is specially visible with EDK2 running at EL2, but still using the EL1 virtual timer, which in that case is fully emulated. Any key-press takes ages to be captured, as there is no UART interrupt and EDK2 relies on polling from a timer... The fix is simply to drop the early return. If the timer interrupt is pending, we will still return early, and otherwise arm the soft timer. Fixes: 4d74ecfa6458b ("KVM: arm64: Don't arm a hrtimer for an already pending timer") Cc: stable@vger.kernel.org Tested-by: Dmytro Terletskyi Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20250204110050.150560-2-maz@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/kvm/arch_timer.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c index d3d243366536..035e43f5d4f9 100644 --- a/arch/arm64/kvm/arch_timer.c +++ b/arch/arm64/kvm/arch_timer.c @@ -471,10 +471,8 @@ static void timer_emulate(struct arch_timer_context *ctx) trace_kvm_timer_emulate(ctx, should_fire); - if (should_fire != ctx->irq.level) { + if (should_fire != ctx->irq.level) kvm_timer_update_irq(ctx->vcpu, should_fire, ctx); - return; - } kvm_timer_update_status(ctx, should_fire); From 1b8705ad5365b5333240b46d5cd24e88ef2ddb14 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 4 Feb 2025 11:00:49 +0000 Subject: [PATCH 0474/1443] KVM: arm64: timer: Correctly handle EL1 timer emulation when !FEAT_ECV Both Wei-Lin Chang and Volodymyr Babchuk report that the way we handle the emulation of EL1 timers with NV is completely wrong, specially in the case of HCR_EL2.E2H==0. There are three problems in about as many lines of code: - With E2H==0, the EL1 timers are overwritten with the EL1 state, while they should actually contain the EL2 state (as per the timer map) - With E2H==1, we run the full EL1 timer emulation even when ECV is present, hiding a bug in timer_emulate() (see previous patch) - The comments are actively misleading, and say all the wrong things. This is only attributable to the code having been initially written for FEAT_NV, hacked up to handle FEAT_NV2 *in parallel*, and vaguely hacked again to be FEAT_NV2 only. Oh, and yours truly being a gold plated idiot. The fix is obvious: just delete most of the E2H==0 code, have a unified handling of the timers (because they really are E2H agnostic), and make sure we don't execute any of that when FEAT_ECV is present. Fixes: 4bad3068cfa9f ("KVM: arm64: nv: Sync nested timer state with FEAT_NV2") Reported-by: Wei-Lin Chang Reported-by: Volodymyr Babchuk Link: https://lore.kernel.org/r/fqiqfjzwpgbzdtouu2pwqlu7llhnf5lmy4hzv5vo6ph4v3vyls@jdcfy3fjjc5k Link: https://lore.kernel.org/r/87frl51tse.fsf@epam.com Tested-by: Dmytro Terletskyi Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20250204110050.150560-3-maz@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/kvm/arch_timer.c | 30 ++++++++++-------------------- 1 file changed, 10 insertions(+), 20 deletions(-) diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c index 035e43f5d4f9..e59836e0260c 100644 --- a/arch/arm64/kvm/arch_timer.c +++ b/arch/arm64/kvm/arch_timer.c @@ -974,31 +974,21 @@ void kvm_timer_sync_nested(struct kvm_vcpu *vcpu) * which allows trapping of the timer registers even with NV2. * Still, this is still worse than FEAT_NV on its own. Meh. */ - if (!vcpu_el2_e2h_is_set(vcpu)) { - if (cpus_have_final_cap(ARM64_HAS_ECV)) - return; - - /* - * A non-VHE guest hypervisor doesn't have any direct access - * to its timers: the EL2 registers trap (and the HW is - * fully emulated), while the EL0 registers access memory - * despite the access being notionally direct. Boo. - * - * We update the hardware timer registers with the - * latest value written by the guest to the VNCR page - * and let the hardware take care of the rest. - */ - write_sysreg_el0(__vcpu_sys_reg(vcpu, CNTV_CTL_EL0), SYS_CNTV_CTL); - write_sysreg_el0(__vcpu_sys_reg(vcpu, CNTV_CVAL_EL0), SYS_CNTV_CVAL); - write_sysreg_el0(__vcpu_sys_reg(vcpu, CNTP_CTL_EL0), SYS_CNTP_CTL); - write_sysreg_el0(__vcpu_sys_reg(vcpu, CNTP_CVAL_EL0), SYS_CNTP_CVAL); - } else { + if (!cpus_have_final_cap(ARM64_HAS_ECV)) { /* * For a VHE guest hypervisor, the EL2 state is directly - * stored in the host EL1 timers, while the emulated EL0 + * stored in the host EL1 timers, while the emulated EL1 * state is stored in the VNCR page. The latter could have * been updated behind our back, and we must reset the * emulation of the timers. + * + * A non-VHE guest hypervisor doesn't have any direct access + * to its timers: the EL2 registers trap despite being + * notionally direct (we use the EL1 HW, as for VHE), while + * the EL1 registers access memory. + * + * In both cases, process the emulated timers on each guest + * exit. Boo. */ struct timer_map map; get_timer_map(vcpu, &map); From 0e459810285503fb354537e84049e212c5917c33 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 4 Feb 2025 11:00:50 +0000 Subject: [PATCH 0475/1443] KVM: arm64: timer: Don't adjust the EL2 virtual timer offset The way we deal with the EL2 virtual timer is a bit odd. We try to cope with E2H being flipped, and adjust which offset applies to that timer depending on the current E2H value. But that's a complexity we shouldn't have to worry about. What we have to deal with is either E2H being RES1, in which case there is no offset, or E2H being RES0, and the virtual timer simply does not exist. Drop the adjusting of the timer offset, which makes things a bit simpler. At the same time, make sure that accessing the HV timer when E2H is RES0 results in an UNDEF in the guest. Suggested-by: Oliver Upton Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20250204110050.150560-4-maz@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/kvm/arch_timer.c | 15 --------------- arch/arm64/kvm/sys_regs.c | 16 +++++++++++++--- 2 files changed, 13 insertions(+), 18 deletions(-) diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c index e59836e0260c..231c0cd9c7b4 100644 --- a/arch/arm64/kvm/arch_timer.c +++ b/arch/arm64/kvm/arch_timer.c @@ -759,21 +759,6 @@ static void kvm_timer_vcpu_load_nested_switch(struct kvm_vcpu *vcpu, timer_irq(map->direct_ptimer), &arch_timer_irq_ops); WARN_ON_ONCE(ret); - - /* - * The virtual offset behaviour is "interesting", as it - * always applies when HCR_EL2.E2H==0, but only when - * accessed from EL1 when HCR_EL2.E2H==1. So make sure we - * track E2H when putting the HV timer in "direct" mode. - */ - if (map->direct_vtimer == vcpu_hvtimer(vcpu)) { - struct arch_timer_offset *offs = &map->direct_vtimer->offset; - - if (vcpu_el2_e2h_is_set(vcpu)) - offs->vcpu_offset = NULL; - else - offs->vcpu_offset = &__vcpu_sys_reg(vcpu, CNTVOFF_EL2); - } } } diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 526d66f24e34..7968bee0d27e 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -1452,6 +1452,16 @@ static bool access_arch_timer(struct kvm_vcpu *vcpu, return true; } +static bool access_hv_timer(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + if (!vcpu_el2_e2h_is_set(vcpu)) + return undef_access(vcpu, p, r); + + return access_arch_timer(vcpu, p, r); +} + static s64 kvm_arm64_ftr_safe_value(u32 id, const struct arm64_ftr_bits *ftrp, s64 new, s64 cur) { @@ -3099,9 +3109,9 @@ static const struct sys_reg_desc sys_reg_descs[] = { EL2_REG(CNTHP_CTL_EL2, access_arch_timer, reset_val, 0), EL2_REG(CNTHP_CVAL_EL2, access_arch_timer, reset_val, 0), - { SYS_DESC(SYS_CNTHV_TVAL_EL2), access_arch_timer }, - EL2_REG(CNTHV_CTL_EL2, access_arch_timer, reset_val, 0), - EL2_REG(CNTHV_CVAL_EL2, access_arch_timer, reset_val, 0), + { SYS_DESC(SYS_CNTHV_TVAL_EL2), access_hv_timer }, + EL2_REG(CNTHV_CTL_EL2, access_hv_timer, reset_val, 0), + EL2_REG(CNTHV_CVAL_EL2, access_hv_timer, reset_val, 0), { SYS_DESC(SYS_CNTKCTL_EL12), access_cntkctl_el12 }, From ee3a66f431d689b796b9cb48aefd3d223540381c Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 4 Feb 2025 11:13:24 -0500 Subject: [PATCH 0476/1443] kvm: x86: SRSO_USER_KERNEL_NO is not synthesized SYNTHESIZED_F() generally is used together with setup_force_cpu_cap(), i.e. when it makes sense to present the feature even if cpuid does not have it *and* the VM is not able to see the difference. For example, it can be used when mitigations on the host automatically protect the guest as well. The "SYNTHESIZED_F(SRSO_USER_KERNEL_NO)" line came in as a conflict resolution between the CPUID overhaul from the KVM tree and support for the feature in the x86 tree. Using it right now does not hurt, or make a difference for that matter, because there is no setup_force_cpu_cap(X86_FEATURE_SRSO_USER_KERNEL_NO). However, it is a little less future proof in case such a setup_force_cpu_cap() appears later, for a case where the kernel somehow is not vulnerable but the guest would have to apply the mitigation. Signed-off-by: Paolo Bonzini --- arch/x86/kvm/cpuid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 2cbb3874ad39..8eb3a88707f2 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -1180,7 +1180,7 @@ void kvm_set_cpu_caps(void) SYNTHESIZED_F(SBPB), SYNTHESIZED_F(IBPB_BRTYPE), SYNTHESIZED_F(SRSO_NO), - SYNTHESIZED_F(SRSO_USER_KERNEL_NO), + F(SRSO_USER_KERNEL_NO), ); kvm_cpu_cap_init(CPUID_8000_0022_EAX, From 203a53029a9c6935ad38e0343d048e51488797cf Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Tue, 4 Feb 2025 10:56:47 +0000 Subject: [PATCH 0477/1443] KVM: selftests: Fix spelling mistake "initally" -> "initially" There is a spelling mistake in a literal string and in the function test_get_inital_dirty. Fix them. Signed-off-by: Colin Ian King Message-ID: <20250204105647.367743-1-colin.i.king@gmail.com> Signed-off-by: Paolo Bonzini --- tools/testing/selftests/kvm/s390/cmma_test.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/kvm/s390/cmma_test.c b/tools/testing/selftests/kvm/s390/cmma_test.c index e32dd59703a0..85cc8c18d6e7 100644 --- a/tools/testing/selftests/kvm/s390/cmma_test.c +++ b/tools/testing/selftests/kvm/s390/cmma_test.c @@ -444,7 +444,7 @@ static void assert_no_pages_cmma_dirty(struct kvm_vm *vm) ); } -static void test_get_inital_dirty(void) +static void test_get_initial_dirty(void) { struct kvm_vm *vm = create_vm_two_memslots(); struct kvm_vcpu *vcpu; @@ -651,7 +651,7 @@ struct testdef { } testlist[] = { { "migration mode and dirty tracking", test_migration_mode }, { "GET_CMMA_BITS: basic calls", test_get_cmma_basic }, - { "GET_CMMA_BITS: all pages are dirty initally", test_get_inital_dirty }, + { "GET_CMMA_BITS: all pages are dirty initially", test_get_initial_dirty }, { "GET_CMMA_BITS: holes are skipped", test_get_skip_holes }, }; From fc016ef7da64fd473d73ee6c261ba1b0b47afe2b Mon Sep 17 00:00:00 2001 From: Richard Fitzgerald Date: Tue, 4 Feb 2025 13:39:41 +0800 Subject: [PATCH 0478/1443] ASoC: Intel: sof_sdw: Add lookup of quirk using PCI subsystem ID Add lookup of PCI subsystem vendor:device ID to find a quirk. The subsystem ID (SSID) is part of the PCI specification to uniquely identify a particular system-specific implementation of a hardware device. Unlike DMI information, it identifies the sound hardware itself, rather than a specific model of PC. SSID can be more reliable and stable than DMI strings, and is preferred by some vendors as the way to identify the actual sound hardware. Signed-off-by: Richard Fitzgerald Reviewed-by: Liam Girdwood Signed-off-by: Bard Liao Link: https://patch.msgid.link/20250204053943.93596-2-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/boards/sof_sdw.c | 30 ++++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/sound/soc/intel/boards/sof_sdw.c b/sound/soc/intel/boards/sof_sdw.c index b0d35fda7b17..381fae5943fe 100644 --- a/sound/soc/intel/boards/sof_sdw.c +++ b/sound/soc/intel/boards/sof_sdw.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include "sof_sdw_common.h" #include "../../codecs/rt711.h" @@ -751,6 +752,22 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = { {} }; +static const struct snd_pci_quirk sof_sdw_ssid_quirk_table[] = { + {} +}; + +static void sof_sdw_check_ssid_quirk(const struct snd_soc_acpi_mach *mach) +{ + const struct snd_pci_quirk *quirk_entry; + + quirk_entry = snd_pci_quirk_lookup_id(mach->mach_params.subsystem_vendor, + mach->mach_params.subsystem_device, + sof_sdw_ssid_quirk_table); + + if (quirk_entry) + sof_sdw_quirk = quirk_entry->value; +} + static struct snd_soc_dai_link_component platform_component[] = { { /* name might be overridden during probe */ @@ -1278,6 +1295,13 @@ static int mc_probe(struct platform_device *pdev) snd_soc_card_set_drvdata(card, ctx); + if (mach->mach_params.subsystem_id_set) { + snd_soc_card_set_pci_ssid(card, + mach->mach_params.subsystem_vendor, + mach->mach_params.subsystem_device); + sof_sdw_check_ssid_quirk(mach); + } + dmi_check_system(sof_sdw_quirk_table); if (quirk_override != -1) { @@ -1293,12 +1317,6 @@ static int mc_probe(struct platform_device *pdev) for (i = 0; i < ctx->codec_info_list_count; i++) codec_info_list[i].amp_num = 0; - if (mach->mach_params.subsystem_id_set) { - snd_soc_card_set_pci_ssid(card, - mach->mach_params.subsystem_vendor, - mach->mach_params.subsystem_device); - } - ret = sof_card_dai_links_create(card); if (ret < 0) return ret; From 0843449708085c4fb45a3c325c2fbced556f6abf Mon Sep 17 00:00:00 2001 From: Richard Fitzgerald Date: Tue, 4 Feb 2025 13:39:42 +0800 Subject: [PATCH 0479/1443] ASoC: Intel: sof_sdw: Add quirk for Asus Zenbook S14 Asus laptops with sound PCI subsystem ID 1043:1e13 have the DMICs connected to the host instead of the CS42L43 so need the SOC_SDW_CODEC_MIC quirk. Signed-off-by: Richard Fitzgerald Reviewed-by: Liam Girdwood Signed-off-by: Bard Liao Link: https://patch.msgid.link/20250204053943.93596-3-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/boards/sof_sdw.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/intel/boards/sof_sdw.c b/sound/soc/intel/boards/sof_sdw.c index 381fae5943fe..683e15b459a1 100644 --- a/sound/soc/intel/boards/sof_sdw.c +++ b/sound/soc/intel/boards/sof_sdw.c @@ -753,6 +753,7 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = { }; static const struct snd_pci_quirk sof_sdw_ssid_quirk_table[] = { + SND_PCI_QUIRK(0x1043, 0x1e13, "ASUS Zenbook S14", SOC_SDW_CODEC_MIC), {} }; From d8989106287d3735c7e7fc6acb3811d62ebb666c Mon Sep 17 00:00:00 2001 From: Uday M Bhat Date: Tue, 4 Feb 2025 13:39:43 +0800 Subject: [PATCH 0480/1443] ASoC: Intel: sof_sdw: Add support for Fatcat board with BT offload enabled in PTL platform This change adds an entry for fatcat boards in soundwire quirk table and also, enables BT offload for PTL RVP. Signed-off-by: Uday M Bhat Signed-off-by: Jairaj Arava Reviewed-by: Ranjani Sridharan Signed-off-by: Bard Liao Link: https://patch.msgid.link/20250204053943.93596-4-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/boards/sof_sdw.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/sound/soc/intel/boards/sof_sdw.c b/sound/soc/intel/boards/sof_sdw.c index 683e15b459a1..203b07d4d833 100644 --- a/sound/soc/intel/boards/sof_sdw.c +++ b/sound/soc/intel/boards/sof_sdw.c @@ -749,6 +749,16 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = { }, .driver_data = (void *)(SOC_SDW_PCH_DMIC), }, + { + .callback = sof_sdw_quirk_cb, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Google"), + DMI_MATCH(DMI_PRODUCT_NAME, "Fatcat"), + }, + .driver_data = (void *)(SOC_SDW_PCH_DMIC | + SOF_BT_OFFLOAD_SSP(2) | + SOF_SSP_BT_OFFLOAD_PRESENT), + }, {} }; From 3588b76db7ba798f54dee39a55708b16e1c61de4 Mon Sep 17 00:00:00 2001 From: Bard Liao Date: Tue, 4 Feb 2025 11:31:33 +0800 Subject: [PATCH 0481/1443] ASoC: Intel: soc-acpi-intel-tgl-match: declare adr as ull MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The adr is u64. Signed-off-by: Bard Liao Reviewed-by: Péter Ujfalusi Reviewed-by: Ranjani Sridharan Link: https://patch.msgid.link/20250204033134.92332-2-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- .../soc/intel/common/soc-acpi-intel-tgl-match.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/sound/soc/intel/common/soc-acpi-intel-tgl-match.c b/sound/soc/intel/common/soc-acpi-intel-tgl-match.c index 6f8c06413665..b77aafb0bfb6 100644 --- a/sound/soc/intel/common/soc-acpi-intel-tgl-match.c +++ b/sound/soc/intel/common/soc-acpi-intel-tgl-match.c @@ -658,25 +658,25 @@ static const struct snd_soc_acpi_endpoint cs35l56_7_fb_endpoints[] = { static const struct snd_soc_acpi_adr_device cs35l56_sdw_eight_1_4_fb_adr[] = { { - .adr = 0x00003301fa355601, + .adr = 0x00003301fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_l_fb_endpoints), .endpoints = cs35l56_l_fb_endpoints, .name_prefix = "AMP1" }, { - .adr = 0x00003201fa355601, + .adr = 0x00003201fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_2_fb_endpoints), .endpoints = cs35l56_2_fb_endpoints, .name_prefix = "AMP2" }, { - .adr = 0x00003101fa355601, + .adr = 0x00003101fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_4_fb_endpoints), .endpoints = cs35l56_4_fb_endpoints, .name_prefix = "AMP3" }, { - .adr = 0x00003001fa355601, + .adr = 0x00003001fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_6_fb_endpoints), .endpoints = cs35l56_6_fb_endpoints, .name_prefix = "AMP4" @@ -685,25 +685,25 @@ static const struct snd_soc_acpi_adr_device cs35l56_sdw_eight_1_4_fb_adr[] = { static const struct snd_soc_acpi_adr_device cs35l56_sdw_eight_5_8_fb_adr[] = { { - .adr = 0x00013701fa355601, + .adr = 0x00013701fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_r_fb_endpoints), .endpoints = cs35l56_r_fb_endpoints, .name_prefix = "AMP8" }, { - .adr = 0x00013601fa355601, + .adr = 0x00013601fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_3_fb_endpoints), .endpoints = cs35l56_3_fb_endpoints, .name_prefix = "AMP7" }, { - .adr = 0x00013501fa355601, + .adr = 0x00013501fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_5_fb_endpoints), .endpoints = cs35l56_5_fb_endpoints, .name_prefix = "AMP6" }, { - .adr = 0x00013401fa355601, + .adr = 0x00013401fa355601ull, .num_endpoints = ARRAY_SIZE(cs35l56_7_fb_endpoints), .endpoints = cs35l56_7_fb_endpoints, .name_prefix = "AMP5" From 20efccc53abf99fa52ea30a43dec758f6b6b9940 Mon Sep 17 00:00:00 2001 From: Bard Liao Date: Tue, 4 Feb 2025 11:31:34 +0800 Subject: [PATCH 0482/1443] ASoC: Intel: soc-acpi-intel-mtl-match: declare adr as ull MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The adr is u64. Signed-off-by: Bard Liao Reviewed-by: Péter Ujfalusi Reviewed-by: Ranjani Sridharan Link: https://patch.msgid.link/20250204033134.92332-3-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/common/soc-acpi-intel-mtl-match.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/intel/common/soc-acpi-intel-mtl-match.c b/sound/soc/intel/common/soc-acpi-intel-mtl-match.c index 770e2194a283..9e611e3667ad 100644 --- a/sound/soc/intel/common/soc-acpi-intel-mtl-match.c +++ b/sound/soc/intel/common/soc-acpi-intel-mtl-match.c @@ -330,7 +330,7 @@ static const struct snd_soc_acpi_adr_device rt1316_3_single_adr[] = { static const struct snd_soc_acpi_adr_device rt1318_1_single_adr[] = { { - .adr = 0x000130025D131801, + .adr = 0x000130025D131801ull, .num_endpoints = 1, .endpoints = &single_endpoint, .name_prefix = "rt1318-1" From 6b24e67b4056ba83b1e95e005b7e50fdb1cc6cf4 Mon Sep 17 00:00:00 2001 From: John Keeping Date: Tue, 4 Feb 2025 16:13:10 +0000 Subject: [PATCH 0483/1443] ASoC: rockchip: i2s-tdm: fix shift config for SND_SOC_DAIFMT_DSP_[AB] Commit 2f45a4e289779 ("ASoC: rockchip: i2s_tdm: Fixup config for SND_SOC_DAIFMT_DSP_A/B") applied a partial change to fix the configuration for DSP A and DSP B formats. The shift control also needs updating to set the correct offset for frame data compared to LRCK. Set the correct values. Fixes: 081068fd64140 ("ASoC: rockchip: add support for i2s-tdm controller") Signed-off-by: John Keeping Link: https://patch.msgid.link/20250204161311.2117240-1-jkeeping@inmusicbrands.com Signed-off-by: Mark Brown --- sound/soc/rockchip/rockchip_i2s_tdm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/rockchip/rockchip_i2s_tdm.c b/sound/soc/rockchip/rockchip_i2s_tdm.c index 7f5fcaecee4b..78ab88843f86 100644 --- a/sound/soc/rockchip/rockchip_i2s_tdm.c +++ b/sound/soc/rockchip/rockchip_i2s_tdm.c @@ -451,11 +451,11 @@ static int rockchip_i2s_tdm_set_fmt(struct snd_soc_dai *cpu_dai, break; case SND_SOC_DAIFMT_DSP_A: val = I2S_TXCR_TFS_TDM_PCM; - tdm_val = TDM_SHIFT_CTRL(0); + tdm_val = TDM_SHIFT_CTRL(2); break; case SND_SOC_DAIFMT_DSP_B: val = I2S_TXCR_TFS_TDM_PCM; - tdm_val = TDM_SHIFT_CTRL(2); + tdm_val = TDM_SHIFT_CTRL(4); break; default: ret = -EINVAL; From 6f61269495260531e15d84d090ee63618110c470 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Fri, 24 Jan 2025 10:26:22 -0500 Subject: [PATCH 0484/1443] KVM: remove kvm_arch_post_init_vm The only statement in a kvm_arch_post_init_vm implementation can be moved into the x86 kvm_arch_init_vm. Do so and remove all traces from architecture-independent code. Signed-off-by: Paolo Bonzini --- arch/x86/kvm/x86.c | 7 +------ include/linux/kvm_host.h | 1 - virt/kvm/kvm_main.c | 15 --------------- 3 files changed, 1 insertion(+), 22 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6d4a6734b2d6..8e77e61d4fbd 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -12741,6 +12741,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) "does not run without ignore_msrs=1, please report it to kvm@vger.kernel.org.\n"); } + once_init(&kvm->arch.nx_once); return 0; out_uninit_mmu: @@ -12750,12 +12751,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) return ret; } -int kvm_arch_post_init_vm(struct kvm *kvm) -{ - once_init(&kvm->arch.nx_once); - return 0; -} - static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu) { vcpu_load(vcpu); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 3cb9a32a6330..f34f4cfaa513 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -1615,7 +1615,6 @@ int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu); bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu); bool kvm_arch_dy_has_pending_interrupt(struct kvm_vcpu *vcpu); bool kvm_arch_vcpu_preempted_in_kernel(struct kvm_vcpu *vcpu); -int kvm_arch_post_init_vm(struct kvm *kvm); void kvm_arch_pre_destroy_vm(struct kvm *kvm); void kvm_arch_create_vm_debugfs(struct kvm *kvm); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 3f04cd5e3a8c..ba0327e2d0d3 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1070,15 +1070,6 @@ static int kvm_create_vm_debugfs(struct kvm *kvm, const char *fdname) return ret; } -/* - * Called after the VM is otherwise initialized, but just before adding it to - * the vm_list. - */ -int __weak kvm_arch_post_init_vm(struct kvm *kvm) -{ - return 0; -} - /* * Called just after removing the VM from the vm_list, but before doing any * other destruction. @@ -1199,10 +1190,6 @@ static struct kvm *kvm_create_vm(unsigned long type, const char *fdname) if (r) goto out_err_no_debugfs; - r = kvm_arch_post_init_vm(kvm); - if (r) - goto out_err; - mutex_lock(&kvm_lock); list_add(&kvm->vm_list, &vm_list); mutex_unlock(&kvm_lock); @@ -1212,8 +1199,6 @@ static struct kvm *kvm_create_vm(unsigned long type, const char *fdname) return kvm; -out_err: - kvm_destroy_vm_debugfs(kvm); out_err_no_debugfs: kvm_coalesced_mmio_free(kvm); out_no_coalesced_mmio: From 43fb96ae78551d7bfa4ecca956b258f085d67c40 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 24 Jan 2025 15:46:23 -0800 Subject: [PATCH 0485/1443] KVM: x86/mmu: Ensure NX huge page recovery thread is alive before waking When waking a VM's NX huge page recovery thread, ensure the thread is actually alive before trying to wake it. Now that the thread is spawned on-demand during KVM_RUN, a VM without a recovery thread is reachable via the related module params. BUG: kernel NULL pointer dereference, address: 0000000000000040 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015 RIP: 0010:vhost_task_wake+0x5/0x10 Call Trace: set_nx_huge_pages+0xcc/0x1e0 [kvm] param_attr_store+0x8a/0xd0 module_attr_store+0x1a/0x30 kernfs_fop_write_iter+0x12f/0x1e0 vfs_write+0x233/0x3e0 ksys_write+0x60/0xd0 do_syscall_64+0x5b/0x160 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7f3b52710104 Modules linked in: kvm_intel kvm CR2: 0000000000000040 Fixes: 931656b9e2ff ("kvm: defer huge page recovery vhost task to later") Cc: stable@vger.kernel.org Cc: Keith Busch Signed-off-by: Sean Christopherson Message-ID: <20250124234623.3609069-1-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/mmu/mmu.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index a45ae60e84ab..74c20dbb92da 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -7120,6 +7120,19 @@ static void mmu_destroy_caches(void) kmem_cache_destroy(mmu_page_header_cache); } +static void kvm_wake_nx_recovery_thread(struct kvm *kvm) +{ + /* + * The NX recovery thread is spawned on-demand at the first KVM_RUN and + * may not be valid even though the VM is globally visible. Do nothing, + * as such a VM can't have any possible NX huge pages. + */ + struct vhost_task *nx_thread = READ_ONCE(kvm->arch.nx_huge_page_recovery_thread); + + if (nx_thread) + vhost_task_wake(nx_thread); +} + static int get_nx_huge_pages(char *buffer, const struct kernel_param *kp) { if (nx_hugepage_mitigation_hard_disabled) @@ -7180,7 +7193,7 @@ static int set_nx_huge_pages(const char *val, const struct kernel_param *kp) kvm_mmu_zap_all_fast(kvm); mutex_unlock(&kvm->slots_lock); - vhost_task_wake(kvm->arch.nx_huge_page_recovery_thread); + kvm_wake_nx_recovery_thread(kvm); } mutex_unlock(&kvm_lock); } @@ -7315,7 +7328,7 @@ static int set_nx_huge_pages_recovery_param(const char *val, const struct kernel mutex_lock(&kvm_lock); list_for_each_entry(kvm, &vm_list, vm_list) - vhost_task_wake(kvm->arch.nx_huge_page_recovery_thread); + kvm_wake_nx_recovery_thread(kvm); mutex_unlock(&kvm_lock); } @@ -7451,14 +7464,20 @@ static void kvm_mmu_start_lpage_recovery(struct once *once) { struct kvm_arch *ka = container_of(once, struct kvm_arch, nx_once); struct kvm *kvm = container_of(ka, struct kvm, arch); + struct vhost_task *nx_thread; kvm->arch.nx_huge_page_last = get_jiffies_64(); - kvm->arch.nx_huge_page_recovery_thread = vhost_task_create( - kvm_nx_huge_page_recovery_worker, kvm_nx_huge_page_recovery_worker_kill, - kvm, "kvm-nx-lpage-recovery"); + nx_thread = vhost_task_create(kvm_nx_huge_page_recovery_worker, + kvm_nx_huge_page_recovery_worker_kill, + kvm, "kvm-nx-lpage-recovery"); - if (kvm->arch.nx_huge_page_recovery_thread) - vhost_task_start(kvm->arch.nx_huge_page_recovery_thread); + if (!nx_thread) + return; + + vhost_task_start(nx_thread); + + /* Make the task visible only once it is fully started. */ + WRITE_ONCE(kvm->arch.nx_huge_page_recovery_thread, nx_thread); } int kvm_mmu_post_init_vm(struct kvm *kvm) From 2a03d2da55b4cd5c86b360db0e917ee93b3f0cb9 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Thu, 9 Jan 2025 06:35:49 +0200 Subject: [PATCH 0486/1443] dt-bindings: nvmem: qcom,qfprom: Add SAR2130P compatible Document compatible for the QFPROM on SAR2130P platform. Signed-off-by: Dmitry Baryshkov Acked-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20250109-sar2130p-nvmem-v4-5-633739fe5f11@linaro.org Signed-off-by: Rob Herring (Arm) --- Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml b/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml index d37f544ab8aa..39c209249c9c 100644 --- a/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml +++ b/Documentation/devicetree/bindings/nvmem/qcom,qfprom.yaml @@ -36,6 +36,7 @@ properties: - qcom,qcs404-qfprom - qcom,qcs615-qfprom - qcom,qcs8300-qfprom + - qcom,sar2130p-qfprom - qcom,sc7180-qfprom - qcom,sc7280-qfprom - qcom,sc8280xp-qfprom From 230b19bc2bcc5897d0e20b4ce7e9790a469a2db0 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 31 Jan 2025 14:49:54 +0200 Subject: [PATCH 0487/1443] drm/i915/dp: Iterate DSC BPP from high to low on all platforms Commit 1c56e9a39833 ("drm/i915/dp: Get optimal link config to have best compressed bpp") tries to find the best compressed bpp for the link. However, it iterates from max to min bpp on display 13+, and from min to max on other platforms. This presumably leads to minimum compressed bpp always being chosen on display 11-12. Iterate from high to low on all platforms to actually use the best possible compressed bpp. Fixes: 1c56e9a39833 ("drm/i915/dp: Get optimal link config to have best compressed bpp") Cc: Ankit Nautiyal Cc: Imre Deak Cc: # v6.7+ Reviewed-by: Imre Deak Reviewed-by: Ankit Nautiyal Signed-off-by: Jani Nikula Link: https://patchwork.freedesktop.org/patch/msgid/3bba67923cbcd13a59d26ef5fa4bb042b13c8a9b.1738327620.git.jani.nikula@intel.com (cherry picked from commit 56b0337d429356c3b9ecc36a03023c8cc856b196) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_dp.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index be07034bfcc6..aa77ddcee42c 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -2072,11 +2072,10 @@ icl_dsc_compute_link_config(struct intel_dp *intel_dp, /* Compressed BPP should be less than the Input DSC bpp */ dsc_max_bpp = min(dsc_max_bpp, pipe_bpp - 1); - for (i = 0; i < ARRAY_SIZE(valid_dsc_bpp); i++) { - if (valid_dsc_bpp[i] < dsc_min_bpp) + for (i = ARRAY_SIZE(valid_dsc_bpp) - 1; i >= 0; i--) { + if (valid_dsc_bpp[i] < dsc_min_bpp || + valid_dsc_bpp[i] > dsc_max_bpp) continue; - if (valid_dsc_bpp[i] > dsc_max_bpp) - break; ret = dsc_compute_link_config(intel_dp, pipe_config, From ecee4d0695067ae04959b121028b42a588e75370 Mon Sep 17 00:00:00 2001 From: Mario Limonciello Date: Tue, 4 Feb 2025 11:40:25 -0600 Subject: [PATCH 0488/1443] accel/amdxdna: Add MODULE_FIRMWARE() declarations Initramfs building tools such as dracut will look for a MODULE_FIRMWARE() declaration to determine which firmware to include in the initramfs when a driver is included in the initramfs. As amdxdna doesn't declare any firmware this causes the driver to fail to load with -ENOENT when in the initramfs. Add the missing declaration for possible firmware. Reported-by: Renjith Pananchikkal Suggested-by: Alexander Deucher Fixes: 8c9ff1b181ba ("accel/amdxdna: Add a new driver for AMD AI Engine") Reviewed-by: Lizhi Hou Link: https://lore.kernel.org/r/20250204174031.3425762-1-superm1@kernel.org Signed-off-by: Mario Limonciello Link: https://patchwork.freedesktop.org/patch/msgid/20250204174031.3425762-1-superm1@kernel.org --- drivers/accel/amdxdna/amdxdna_pci_drv.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.c b/drivers/accel/amdxdna/amdxdna_pci_drv.c index 97d4a032171f..f5b8497cf5ad 100644 --- a/drivers/accel/amdxdna/amdxdna_pci_drv.c +++ b/drivers/accel/amdxdna/amdxdna_pci_drv.c @@ -21,6 +21,11 @@ #define AMDXDNA_AUTOSUSPEND_DELAY 5000 /* milliseconds */ +MODULE_FIRMWARE("amdnpu/1502_00/npu.sbin"); +MODULE_FIRMWARE("amdnpu/17f0_10/npu.sbin"); +MODULE_FIRMWARE("amdnpu/17f0_11/npu.sbin"); +MODULE_FIRMWARE("amdnpu/17f0_20/npu.sbin"); + /* * Bind the driver base on (vendor_id, device_id) pair and later use the * (device_id, rev_id) pair as a key to select the devices. The devices with From aff2355d260e47e780cd96af127beaab18a664b1 Mon Sep 17 00:00:00 2001 From: Mark Lord Date: Tue, 4 Feb 2025 19:45:06 +0200 Subject: [PATCH 0489/1443] spi: pxa2xx: Fix regression when toggling chip select on LPSS devices The commit 78b435c9044a ("spi: pxa2xx: Introduce __lpss_ssp_update_priv() helper") broke speaker output on my ASUS UX5304MA laptop. The problem is in inverted value that got written in the private register. Simple bug, simple fix. Fixes: 78b435c9044a ("spi: pxa2xx: Introduce __lpss_ssp_update_priv() helper") Signed-off-by: Mark Lord Tested-by: Mark Lord Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20250204174506.149978-1-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown --- drivers/spi/spi-pxa2xx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index 5f9cac41baff..06711a62fa3d 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -399,7 +399,7 @@ static void lpss_ssp_cs_control(struct spi_device *spi, bool enable) lpss_ssp_select_cs(spi, config); mask = LPSS_CS_CONTROL_CS_HIGH; - __lpss_ssp_update_priv(drv_data, config->reg_cs_ctrl, mask, enable ? mask : 0); + __lpss_ssp_update_priv(drv_data, config->reg_cs_ctrl, mask, enable ? 0 : mask); if (config->cs_clk_stays_gated) { /* * Changing CS alone when dynamic clock gating is on won't From fd079124112c6e11c1bca2e7c71470a2d60bc363 Mon Sep 17 00:00:00 2001 From: Bharadwaj Raju Date: Wed, 5 Feb 2025 00:59:53 +0530 Subject: [PATCH 0490/1443] selftests/cgroup: use bash in test_cpuset_v1_hp.sh The script uses non-POSIX features like `[[` for conditionals and hence does not work when run with a POSIX /bin/sh. Change the shebang to /bin/bash instead, like the other tests in cgroup. Signed-off-by: Bharadwaj Raju Signed-off-by: Tejun Heo --- tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh b/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh index 3f45512fb512..7406c24be1ac 100755 --- a/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh +++ b/tools/testing/selftests/cgroup/test_cpuset_v1_hp.sh @@ -1,4 +1,4 @@ -#!/bin/sh +#!/bin/bash # SPDX-License-Identifier: GPL-2.0 # # Test the special cpuset v1 hotplug case where a cpuset become empty of From d6179f6c6204f9932aed3a7a2100b4a295dfed9d Mon Sep 17 00:00:00 2001 From: Mark Tomlinson Date: Thu, 6 Jun 2024 15:31:02 +1200 Subject: [PATCH 0491/1443] gpio: pca953x: Improve interrupt support The GPIO drivers with latch interrupt support (typically types starting with PCAL) have interrupt status registers to determine which particular inputs have caused an interrupt. Unfortunately there is no atomic operation to read these registers and clear the interrupt. Clearing the interrupt is done by reading the input registers. The code was reading the interrupt status registers, and then reading the input registers. If an input changed between these two events it was lost. The solution in this patch is to revert to the non-latch version of code, i.e. remembering the previous input status, and looking for the changes. This system results in no more I2C transfers, so is no slower. The latch property of the device still means interrupts will still be noticed if the input changes back to its initial state. Fixes: 44896beae605 ("gpio: pca953x: add PCAL9535 interrupt support for Galileo Gen2") Signed-off-by: Mark Tomlinson Reviewed-by: Andy Shevchenko Link: https://lore.kernel.org/r/20240606033102.2271916-1-mark.tomlinson@alliedtelesis.co.nz Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-pca953x.c | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c index be4c9981ebc4..d63c1030e6ac 100644 --- a/drivers/gpio/gpio-pca953x.c +++ b/drivers/gpio/gpio-pca953x.c @@ -841,25 +841,6 @@ static bool pca953x_irq_pending(struct pca953x_chip *chip, unsigned long *pendin DECLARE_BITMAP(trigger, MAX_LINE); int ret; - if (chip->driver_data & PCA_PCAL) { - /* Read the current interrupt status from the device */ - ret = pca953x_read_regs(chip, PCAL953X_INT_STAT, trigger); - if (ret) - return false; - - /* Check latched inputs and clear interrupt status */ - ret = pca953x_read_regs(chip, chip->regs->input, cur_stat); - if (ret) - return false; - - /* Apply filter for rising/falling edge selection */ - bitmap_replace(new_stat, chip->irq_trig_fall, chip->irq_trig_raise, cur_stat, gc->ngpio); - - bitmap_and(pending, new_stat, trigger, gc->ngpio); - - return !bitmap_empty(pending, gc->ngpio); - } - ret = pca953x_read_regs(chip, chip->regs->input, cur_stat); if (ret) return false; From 028676bb189ed6d1b550a0fc570a9d695b6acfd3 Mon Sep 17 00:00:00 2001 From: Jacob Moroni Date: Mon, 3 Feb 2025 09:36:05 -0500 Subject: [PATCH 0492/1443] net: atlantic: fix warning during hot unplug Firmware deinitialization performs MMIO accesses which are not necessary if the device has already been removed. In some cases, these accesses happen via readx_poll_timeout_atomic which ends up timing out, resulting in a warning at hw_atl2_utils_fw.c:112: [ 104.595913] Call Trace: [ 104.595915] [ 104.595918] ? show_regs+0x6c/0x80 [ 104.595923] ? __warn+0x8d/0x150 [ 104.595925] ? aq_a2_fw_deinit+0xcf/0xe0 [atlantic] [ 104.595934] ? report_bug+0x182/0x1b0 [ 104.595938] ? handle_bug+0x6e/0xb0 [ 104.595940] ? exc_invalid_op+0x18/0x80 [ 104.595942] ? asm_exc_invalid_op+0x1b/0x20 [ 104.595944] ? aq_a2_fw_deinit+0xcf/0xe0 [atlantic] [ 104.595952] ? aq_a2_fw_deinit+0xcf/0xe0 [atlantic] [ 104.595959] aq_nic_deinit.part.0+0xbd/0xf0 [atlantic] [ 104.595964] aq_nic_deinit+0x17/0x30 [atlantic] [ 104.595970] aq_ndev_close+0x2b/0x40 [atlantic] [ 104.595975] __dev_close_many+0xad/0x160 [ 104.595978] dev_close_many+0x99/0x170 [ 104.595979] unregister_netdevice_many_notify+0x18b/0xb20 [ 104.595981] ? __call_rcu_common+0xcd/0x700 [ 104.595984] unregister_netdevice_queue+0xc6/0x110 [ 104.595986] unregister_netdev+0x1c/0x30 [ 104.595988] aq_pci_remove+0xb1/0xc0 [atlantic] Fix this by skipping firmware deinitialization altogether if the PCI device is no longer present. Tested with an AQC113 attached via Thunderbolt by performing repeated unplug cycles while traffic was running via iperf. Fixes: 97bde5c4f909 ("net: ethernet: aquantia: Support for NIC-specific code") Signed-off-by: Jacob Moroni Reviewed-by: Igor Russkikh Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250203143604.24930-3-mail@jakemoroni.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index fe0e3e2a8117..71e50fc65c14 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -1441,7 +1441,9 @@ void aq_nic_deinit(struct aq_nic_s *self, bool link_down) aq_ptp_ring_free(self); aq_ptp_free(self); - if (likely(self->aq_fw_ops->deinit) && link_down) { + /* May be invoked during hot unplug. */ + if (pci_device_is_present(self->pdev) && + likely(self->aq_fw_ops->deinit) && link_down) { mutex_lock(&self->fwreq_mutex); self->aq_fw_ops->deinit(self->aq_hw); mutex_unlock(&self->fwreq_mutex); From a1300691aed9ee852b0a9192e29e2bdc2411a7e6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 3 Feb 2025 17:08:38 +0000 Subject: [PATCH 0493/1443] net: rose: lock the socket in rose_bind() syzbot reported a soft lockup in rose_loopback_timer(), with a repro calling bind() from multiple threads. rose_bind() must lock the socket to avoid this issue. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot+7ff41b5215f0c534534e@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/67a0f78d.050a0220.d7c5a.00a0.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Link: https://patch.msgid.link/20250203170838.3521361-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/rose/af_rose.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 72c65d938a15..a4a668b88a8f 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -701,11 +701,9 @@ static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) struct net_device *dev; ax25_address *source; ax25_uid_assoc *user; + int err = -EINVAL; int n; - if (!sock_flag(sk, SOCK_ZAPPED)) - return -EINVAL; - if (addr_len != sizeof(struct sockaddr_rose) && addr_len != sizeof(struct full_sockaddr_rose)) return -EINVAL; @@ -718,8 +716,15 @@ static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if ((unsigned int) addr->srose_ndigis > ROSE_MAX_DIGIS) return -EINVAL; - if ((dev = rose_dev_get(&addr->srose_addr)) == NULL) - return -EADDRNOTAVAIL; + lock_sock(sk); + + if (!sock_flag(sk, SOCK_ZAPPED)) + goto out_release; + + err = -EADDRNOTAVAIL; + dev = rose_dev_get(&addr->srose_addr); + if (!dev) + goto out_release; source = &addr->srose_call; @@ -730,7 +735,8 @@ static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) } else { if (ax25_uid_policy && !capable(CAP_NET_BIND_SERVICE)) { dev_put(dev); - return -EACCES; + err = -EACCES; + goto out_release; } rose->source_call = *source; } @@ -753,8 +759,10 @@ static int rose_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) rose_insert_socket(sk); sock_reset_flag(sk, SOCK_ZAPPED); - - return 0; + err = 0; +out_release: + release_sock(sk); + return err; } static int rose_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags) From 820ccf8cb2b145ab9fc12651f7f80339614fa46c Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Fri, 31 Jan 2025 15:31:19 -0700 Subject: [PATCH 0494/1443] drm/amd/display: Respect user's CONFIG_FRAME_WARN more for dml files Currently, there are several files in drm/amd/display that aim to have a higher -Wframe-larger-than value to avoid instances of that warning with a lower value from the user's configuration. However, with the way that it is currently implemented, it does not respect the user's request via CONFIG_FRAME_WARN for a higher stack frame limit, which can cause pain when new instances of the warning appear and break the build due to CONFIG_WERROR. Adjust the logic to switch from a hard coded -Wframe-larger-than value to only using the value as a minimum clamp and deferring to the requested value from CONFIG_FRAME_WARN if it is higher. Suggested-by: Harry Wentland Reported-by: Greg Kroah-Hartman Closes: https://lore.kernel.org/2025013003-audience-opposing-7f95@gregkh/ Signed-off-by: Nathan Chancellor Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/dml/Makefile | 14 ++++++++----- drivers/gpu/drm/amd/display/dc/dml2/Makefile | 22 ++++++++++++-------- 2 files changed, 22 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 46f9c05de16e..e1d500633dfa 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -29,11 +29,15 @@ dml_ccflags := $(CC_FLAGS_FPU) dml_rcflags := $(CC_FLAGS_NO_FPU) ifneq ($(CONFIG_FRAME_WARN),0) -ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y) -frame_warn_flag := -Wframe-larger-than=3072 -else -frame_warn_flag := -Wframe-larger-than=2048 -endif + ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y) + frame_warn_limit := 3072 + else + frame_warn_limit := 2048 + endif + + ifeq ($(call test-lt, $(CONFIG_FRAME_WARN), $(frame_warn_limit)),y) + frame_warn_flag := -Wframe-larger-than=$(frame_warn_limit) + endif endif CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/Makefile b/drivers/gpu/drm/amd/display/dc/dml2/Makefile index 91c4f3b4bd5f..21fd466dba26 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml2/Makefile @@ -28,15 +28,19 @@ dml2_ccflags := $(CC_FLAGS_FPU) dml2_rcflags := $(CC_FLAGS_NO_FPU) ifneq ($(CONFIG_FRAME_WARN),0) -ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y) -ifeq ($(CONFIG_CC_IS_CLANG)$(CONFIG_COMPILE_TEST),yy) -frame_warn_flag := -Wframe-larger-than=4096 -else -frame_warn_flag := -Wframe-larger-than=3072 -endif -else -frame_warn_flag := -Wframe-larger-than=2048 -endif + ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y) + ifeq ($(CONFIG_CC_IS_CLANG)$(CONFIG_COMPILE_TEST),yy) + frame_warn_limit := 4096 + else + frame_warn_limit := 3072 + endif + else + frame_warn_limit := 2048 + endif + + ifeq ($(call test-lt, $(CONFIG_FRAME_WARN), $(frame_warn_limit)),y) + frame_warn_flag := -Wframe-larger-than=$(frame_warn_limit) + endif endif subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/dml2 From f245b400a223a71d6d5f4c72a2cb9b573a7fc2b6 Mon Sep 17 00:00:00 2001 From: Tom Chung Date: Tue, 4 Feb 2025 15:07:44 +0800 Subject: [PATCH 0495/1443] Revert "drm/amd/display: Use HW lock mgr for PSR1" This reverts commit a2b5a9956269 ("drm/amd/display: Use HW lock mgr for PSR1") Because it may cause system hang while connect with two edp panel. Acked-by: Wayne Lin Signed-off-by: Tom Chung Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c index 5bb8b78bf250..bf636b28e3e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_hw_lock_mgr.c @@ -63,8 +63,7 @@ void dmub_hw_lock_mgr_inbox0_cmd(struct dc_dmub_srv *dmub_srv, bool should_use_dmub_lock(struct dc_link *link) { - if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1 || - link->psr_settings.psr_version == DC_PSR_VERSION_1) + if (link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) return true; if (link->replay_settings.replay_feature_enabled) From 9c9dc9ba4a00510c624588d9860968b26803a2b8 Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Mon, 3 Feb 2025 15:48:57 -0800 Subject: [PATCH 0496/1443] drm/xe/pxp: Fail the load if PXP fails to initialize The PXP implementation mimics the i915 approach of allowing the load to continue even if PXP init has failed. On Xe however we're taking an harder stance on boot error and only allowing the load to complete if everything is working, so update the code to fail if anything goes wrong during PXP init. While at it, update the return code in case of PXP not supported to be 0 instead of EOPNOTSUPP, to follow the standard of functions called by xe_device_probe where every non-zero value means failure. Suggested-by: Lucas De Marchi Signed-off-by: Daniele Ceraolo Spurio Cc: Lucas De Marchi Cc: John Harrison Reviewed-by: Lucas De Marchi Link: https://patchwork.freedesktop.org/patch/msgid/20250203234857.1419637-1-daniele.ceraolospurio@intel.com --- drivers/gpu/drm/xe/xe_device.c | 4 ++-- drivers/gpu/drm/xe/xe_pxp.c | 20 ++++++++++++-------- 2 files changed, 14 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index f30f8f668dee..c641c802d4fb 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -864,8 +864,8 @@ int xe_device_probe(struct xe_device *xe) /* A PXP init failure is not fatal */ err = xe_pxp_init(xe); - if (err && err != -EOPNOTSUPP) - drm_err(&xe->drm, "PXP initialization failed: %pe\n", ERR_PTR(err)); + if (err) + goto err_fini_display; err = drm_dev_register(&xe->drm, 0); if (err) diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c index 5e7a1688a771..3cd3f83e86b0 100644 --- a/drivers/gpu/drm/xe/xe_pxp.c +++ b/drivers/gpu/drm/xe/xe_pxp.c @@ -372,8 +372,8 @@ static void pxp_fini(void *arg) * are performed asynchronously as part of the GSC init. PXP can only be used * after both this function and the async worker have completed. * - * Returns -EOPNOTSUPP if PXP is not supported, 0 if PXP initialization is - * successful, other errno value if there is an error during the init. + * Returns 0 if PXP is not supported or if PXP initialization is successful, + * other errno value if there is an error during the init. */ int xe_pxp_init(struct xe_device *xe) { @@ -382,26 +382,28 @@ int xe_pxp_init(struct xe_device *xe) int err; if (!xe_pxp_is_supported(xe)) - return -EOPNOTSUPP; + return 0; /* we only support PXP on single tile devices with a media GT */ if (xe->info.tile_count > 1 || !gt) - return -EOPNOTSUPP; + return 0; /* The GSCCS is required for submissions to the GSC FW */ if (!(gt->info.engine_mask & BIT(XE_HW_ENGINE_GSCCS0))) - return -EOPNOTSUPP; + return 0; /* PXP requires both GSC and HuC firmwares to be available */ if (!xe_uc_fw_is_loadable(>->uc.gsc.fw) || !xe_uc_fw_is_loadable(>->uc.huc.fw)) { drm_info(&xe->drm, "skipping PXP init due to missing FW dependencies"); - return -EOPNOTSUPP; + return 0; } pxp = drmm_kzalloc(&xe->drm, sizeof(struct xe_pxp), GFP_KERNEL); - if (!pxp) - return -ENOMEM; + if (!pxp) { + err = -ENOMEM; + goto out; + } INIT_LIST_HEAD(&pxp->queues.list); spin_lock_init(&pxp->queues.lock); @@ -448,6 +450,8 @@ int xe_pxp_init(struct xe_device *xe) destroy_workqueue(pxp->irq.wq); out_free: drmm_kfree(&xe->drm, pxp); +out: + drm_err(&xe->drm, "PXP initialization failed: %pe\n", ERR_PTR(err)); return err; } From dabbd325b25edb5cdd99c94391817202dd54b651 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Tue, 4 Feb 2025 23:50:08 +0000 Subject: [PATCH 0497/1443] ASoC: simple-card-utils.c: add missing dlc->of_node commit 90de551c1bf ("ASoC: simple-card-utils.c: enable multi Component support") added muiti Component support, but was missing to add dlc->of_node. Because of it, Sound device list will indicates strange name if it was DPCM connection and driver supports dai->driver->dai_args, like below > aplay -l card X: sndulcbmix [xxxx], device 0: fe.(null).rsnd-dai.0 (*) [] ... ^^^^^^ It will be fixed by this patch > aplay -l card X: sndulcbmix [xxxx], device 0: fe.sound@ec500000.rsnd-dai.0 (*) [] ... ^^^^^^^^^^^^^^ Signed-off-by: Kuninori Morimoto Reviewed-by: Daniel Baluta Link: https://patch.msgid.link/87ikpp2rtb.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown --- sound/soc/generic/simple-card-utils.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/generic/simple-card-utils.c b/sound/soc/generic/simple-card-utils.c index dd414634b4ac..c2445c5ccd84 100644 --- a/sound/soc/generic/simple-card-utils.c +++ b/sound/soc/generic/simple-card-utils.c @@ -1092,6 +1092,7 @@ int graph_util_parse_dai(struct device *dev, struct device_node *ep, args.np = ep; dai = snd_soc_get_dai_via_args(&args); if (dai) { + dlc->of_node = node; dlc->dai_name = snd_soc_dai_name_get(dai); dlc->dai_args = snd_soc_copy_dai_args(dev, &args); if (!dlc->dai_args) From 32ffed055dcee17f6705f545b069e44a66067808 Mon Sep 17 00:00:00 2001 From: Jiasheng Jiang Date: Wed, 5 Feb 2025 00:43:43 +0000 Subject: [PATCH 0498/1443] regmap-irq: Add missing kfree() Add kfree() for "d->main_status_buf" to the error-handling path to prevent a memory leak. Fixes: a2d21848d921 ("regmap: regmap-irq: Add main status register support") Cc: stable@vger.kernel.org # v5.1+ Signed-off-by: Jiasheng Jiang Link: https://patch.msgid.link/20250205004343.14413-1-jiashengjiangcool@gmail.com Signed-off-by: Mark Brown --- drivers/base/regmap/regmap-irq.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/base/regmap/regmap-irq.c b/drivers/base/regmap/regmap-irq.c index 0bcd81389a29..978613407ea3 100644 --- a/drivers/base/regmap/regmap-irq.c +++ b/drivers/base/regmap/regmap-irq.c @@ -906,6 +906,7 @@ int regmap_add_irq_chip_fwnode(struct fwnode_handle *fwnode, kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); + kfree(d->main_status_buf); kfree(d->status_buf); kfree(d->status_reg_buf); if (d->config_buf) { @@ -981,6 +982,7 @@ void regmap_del_irq_chip(int irq, struct regmap_irq_chip_data *d) kfree(d->wake_buf); kfree(d->mask_buf_def); kfree(d->mask_buf); + kfree(d->main_status_buf); kfree(d->status_reg_buf); kfree(d->status_buf); if (d->config_buf) { From 76b0a22d4cf7dc9091129560fdc04e73eb9db4cb Mon Sep 17 00:00:00 2001 From: Edson Juliano Drosdeck Date: Sat, 1 Feb 2025 11:39:30 -0300 Subject: [PATCH 0499/1443] ALSA: hda/realtek: Limit mic boost on Positivo ARN50 The internal mic boost on the Positivo ARN50 is too high. Fix this by applying the ALC269_FIXUP_LIMIT_INT_MIC_BOOST fixup to the machine to limit the gain. Signed-off-by: Edson Juliano Drosdeck Link: https://patch.msgid.link/20250201143930.25089-1-edson.drosdeck@gmail.com Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 8192be394d0d..ae0beb52e7b0 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -11045,6 +11045,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1d72, 0x1945, "Redmi G", ALC256_FIXUP_ASUS_HEADSET_MIC), SND_PCI_QUIRK(0x1d72, 0x1947, "RedmiBook Air", ALC255_FIXUP_XIAOMI_HEADSET_MIC), SND_PCI_QUIRK(0x1f66, 0x0105, "Ayaneo Portable Game Player", ALC287_FIXUP_CS35L41_I2C_2), + SND_PCI_QUIRK(0x2014, 0x800a, "Positivo ARN50", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x2782, 0x0214, "VAIO VJFE-CL", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x2782, 0x0228, "Infinix ZERO BOOK 13", ALC269VB_FIXUP_INFINIX_ZERO_BOOK_13), SND_PCI_QUIRK(0x2782, 0x0232, "CHUWI CoreBook XPro", ALC269VB_FIXUP_CHUWI_COREBOOK_XPRO), From b0eddc21900fb44f8c5db95710479865e3700fbd Mon Sep 17 00:00:00 2001 From: Varadarajan Narayanan Date: Wed, 5 Feb 2025 13:16:56 +0530 Subject: [PATCH 0500/1443] regulator: qcom_smd: Add l2, l5 sub-node to mp5496 regulator Adding l2, l5 sub-node entry to mp5496 regulator node. Cc: stable@vger.kernel.org Acked-by: Rob Herring Signed-off-by: Varadarajan Narayanan Link: https://patch.msgid.link/20250205074657.4142365-2-quic_varada@quicinc.com Signed-off-by: Mark Brown --- .../devicetree/bindings/regulator/qcom,smd-rpm-regulator.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/regulator/qcom,smd-rpm-regulator.yaml b/Documentation/devicetree/bindings/regulator/qcom,smd-rpm-regulator.yaml index f2fd2df68a9e..b7241ce975b9 100644 --- a/Documentation/devicetree/bindings/regulator/qcom,smd-rpm-regulator.yaml +++ b/Documentation/devicetree/bindings/regulator/qcom,smd-rpm-regulator.yaml @@ -22,7 +22,7 @@ description: Each sub-node is identified using the node's name, with valid values listed for each of the pmics below. - For mp5496, s1, s2 + For mp5496, s1, s2, l2, l5 For pm2250, s1, s2, s3, s4, l1, l2, l3, l4, l5, l6, l7, l8, l9, l10, l11, l12, l13, l14, l15, l16, l17, l18, l19, l20, l21, l22 From 796106e29e5df6cd4b4e2b51262a8a19e9fa0625 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 5 Feb 2025 00:20:36 +0000 Subject: [PATCH 0501/1443] ASoC: rsnd: indicate unsupported clock rate It will indicate "unsupported clock rate" when setup clock failed. But it is unclear what kind of rate was failed. Indicate it. Signed-off-by: Kuninori Morimoto Reviewed-by: Yoshihiro Shimoda Link: https://patch.msgid.link/874j192qej.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown --- sound/soc/renesas/rcar/ssi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/soc/renesas/rcar/ssi.c b/sound/soc/renesas/rcar/ssi.c index b3d4e8ae07ef..0c6424a1fcac 100644 --- a/sound/soc/renesas/rcar/ssi.c +++ b/sound/soc/renesas/rcar/ssi.c @@ -336,7 +336,8 @@ static int rsnd_ssi_master_clk_start(struct rsnd_mod *mod, return 0; rate_err: - dev_err(dev, "unsupported clock rate\n"); + dev_err(dev, "unsupported clock rate (%d)\n", rate); + return ret; } From c3fc002b206c6c83d1e3702b979733002ba6fb2c Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 5 Feb 2025 00:20:42 +0000 Subject: [PATCH 0502/1443] ASoC: rsnd: don't indicate warning on rsnd_kctrl_accept_runtime() rsnd_kctrl_accept_runtime() (1) is used for runtime convert rate (= Synchronous SRC Mode). Now, rsnd driver has 2 kctrls for it (A): "SRC Out Rate Switch" (B): "SRC Out Rate" // it calls (1) (A): can be called anytime (B): can be called only runtime, and will indicate warning if it was used at non-runtime. To use runtime convert rate (= Synchronous SRC Mode), user might uses command in below order. (X): > amixer set "SRC Out Rate" on > aplay xxx.wav & (Y): > amixer set "SRC Out Rate" 48010 // convert rate to 48010Hz (Y): calls B (X): calls both A and B. In this case, when user calls (X), it calls both (A) and (B), but it is not yet start running. So, (B) will indicate warning. This warning was added by commit b5c088689847 ("ASoC: rsnd: add warning message to rsnd_kctrl_accept_runtime()"), but the message sounds like the operation was not correct. Let's update warning message. The message is very SRC specific, implement it in src.c Signed-off-by: Kuninori Morimoto Reviewed-by: Yoshihiro Shimoda Link: https://patch.msgid.link/8734gt2qed.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown --- sound/soc/renesas/rcar/core.c | 14 -------------- sound/soc/renesas/rcar/rsnd.h | 1 - sound/soc/renesas/rcar/src.c | 18 +++++++++++++++++- 3 files changed, 17 insertions(+), 16 deletions(-) diff --git a/sound/soc/renesas/rcar/core.c b/sound/soc/renesas/rcar/core.c index d3709fd0409e..f3f0c3f0bb9f 100644 --- a/sound/soc/renesas/rcar/core.c +++ b/sound/soc/renesas/rcar/core.c @@ -1770,20 +1770,6 @@ int rsnd_kctrl_accept_anytime(struct rsnd_dai_stream *io) return 1; } -int rsnd_kctrl_accept_runtime(struct rsnd_dai_stream *io) -{ - struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); - struct rsnd_priv *priv = rsnd_io_to_priv(io); - struct device *dev = rsnd_priv_to_dev(priv); - - if (!runtime) { - dev_warn(dev, "Can't update kctrl when idle\n"); - return 0; - } - - return 1; -} - struct rsnd_kctrl_cfg *rsnd_kctrl_init_m(struct rsnd_kctrl_cfg_m *cfg) { cfg->cfg.val = cfg->val; diff --git a/sound/soc/renesas/rcar/rsnd.h b/sound/soc/renesas/rcar/rsnd.h index a5f54b65313c..04c70690f7a2 100644 --- a/sound/soc/renesas/rcar/rsnd.h +++ b/sound/soc/renesas/rcar/rsnd.h @@ -742,7 +742,6 @@ struct rsnd_kctrl_cfg_s { #define rsnd_kctrl_vals(x) ((x).val) /* = (x).cfg.val[0] */ int rsnd_kctrl_accept_anytime(struct rsnd_dai_stream *io); -int rsnd_kctrl_accept_runtime(struct rsnd_dai_stream *io); struct rsnd_kctrl_cfg *rsnd_kctrl_init_m(struct rsnd_kctrl_cfg_m *cfg); struct rsnd_kctrl_cfg *rsnd_kctrl_init_s(struct rsnd_kctrl_cfg_s *cfg); int rsnd_kctrl_new(struct rsnd_mod *mod, diff --git a/sound/soc/renesas/rcar/src.c b/sound/soc/renesas/rcar/src.c index e7f86db0d94c..309918029772 100644 --- a/sound/soc/renesas/rcar/src.c +++ b/sound/soc/renesas/rcar/src.c @@ -531,6 +531,22 @@ static irqreturn_t rsnd_src_interrupt(int irq, void *data) return IRQ_HANDLED; } +static int rsnd_src_kctrl_accept_runtime(struct rsnd_dai_stream *io) +{ + struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); + + if (!runtime) { + struct rsnd_priv *priv = rsnd_io_to_priv(io); + struct device *dev = rsnd_priv_to_dev(priv); + + dev_warn(dev, "\"SRC Out Rate\" can use during running\n"); + + return 0; + } + + return 1; +} + static int rsnd_src_probe_(struct rsnd_mod *mod, struct rsnd_dai_stream *io, struct rsnd_priv *priv) @@ -594,7 +610,7 @@ static int rsnd_src_pcm_new(struct rsnd_mod *mod, rsnd_io_is_play(io) ? "SRC Out Rate" : "SRC In Rate", - rsnd_kctrl_accept_runtime, + rsnd_src_kctrl_accept_runtime, rsnd_src_set_convert_rate, &src->sync, 192000); From 89f9cf185885d4358aa92b48e51d0f09b71775aa Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 5 Feb 2025 00:20:48 +0000 Subject: [PATCH 0503/1443] ASoC: rsnd: adjust convert rate limitation Current rsnd driver supports Synchronous SRC Mode, but HW allow to update rate only within 1% from current rate. Adjust to it. Becially, this feature is used to fine-tune subtle difference that occur during sampling rate conversion in SRC. So, it should be called within 1% margin of rate difference. If there was difference over 1%, it will apply with 1% increments by using loop without indicating error message. Cc: Yoshihiro Shimoda Signed-off-by: Kuninori Morimoto Reviewed-by: Yoshihiro Shimoda Tested-by: Yoshihiro Shimoda Link: https://patch.msgid.link/871pwd2qe8.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown --- sound/soc/renesas/rcar/src.c | 98 ++++++++++++++++++++++++++++-------- 1 file changed, 76 insertions(+), 22 deletions(-) diff --git a/sound/soc/renesas/rcar/src.c b/sound/soc/renesas/rcar/src.c index 309918029772..7d73b183bda6 100644 --- a/sound/soc/renesas/rcar/src.c +++ b/sound/soc/renesas/rcar/src.c @@ -35,6 +35,7 @@ struct rsnd_src { struct rsnd_mod *dma; struct rsnd_kctrl_cfg_s sen; /* sync convert enable */ struct rsnd_kctrl_cfg_s sync; /* sync convert */ + u32 current_sync_rate; int irq; }; @@ -100,7 +101,7 @@ static u32 rsnd_src_convert_rate(struct rsnd_dai_stream *io, if (!rsnd_src_sync_is_enabled(mod)) return rsnd_io_converted_rate(io); - convert_rate = src->sync.val; + convert_rate = src->current_sync_rate; if (!convert_rate) convert_rate = rsnd_io_converted_rate(io); @@ -201,13 +202,73 @@ static const u32 chan222222[] = { static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, struct rsnd_mod *mod) { + struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); + struct rsnd_priv *priv = rsnd_mod_to_priv(mod); + struct rsnd_src *src = rsnd_mod_to_src(mod); + u32 fin, fout, new_rate; + int inc, cnt, rate; + u64 base, val; + + if (!runtime) + return; + + if (!rsnd_src_sync_is_enabled(mod)) + return; + + fin = rsnd_src_get_in_rate(priv, io); + fout = rsnd_src_get_out_rate(priv, io); + + new_rate = src->sync.val; + + if (!new_rate) + new_rate = fout; + + /* Do nothing if no diff */ + if (new_rate == src->current_sync_rate) + return; + + /* + * SRCm_IFSVR::INTIFS can change within 1% + * see + * SRCm_IFSVR::INTIFS Note + */ + inc = fout / 100; + cnt = abs(new_rate - fout) / inc; + if (fout > new_rate) + inc *= -1; + + /* + * After start running SRC, we can update only SRC_IFSVR + * for Synchronous Mode + */ + base = (u64)0x0400000 * fin; + rate = fout; + for (int i = 0; i < cnt; i++) { + val = base; + rate += inc; + do_div(val, rate); + + rsnd_mod_write(mod, SRC_IFSVR, val); + } + val = base; + do_div(val, new_rate); + + rsnd_mod_write(mod, SRC_IFSVR, val); + + /* update current_sync_rate */ + src->current_sync_rate = new_rate; +} + +static void rsnd_src_init_convert_rate(struct rsnd_dai_stream *io, + struct rsnd_mod *mod) +{ + struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); struct rsnd_priv *priv = rsnd_mod_to_priv(mod); struct device *dev = rsnd_priv_to_dev(priv); - struct snd_pcm_runtime *runtime = rsnd_io_to_runtime(io); int is_play = rsnd_io_is_play(io); int use_src = 0; u32 fin, fout; - u32 ifscr, fsrate, adinr; + u32 ifscr, adinr; u32 cr, route; u32 i_busif, o_busif, tmp; const u32 *bsdsr_table; @@ -245,26 +306,15 @@ static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, adinr = rsnd_get_adinr_bit(mod, io) | chan; /* - * SRC_IFSCR / SRC_IFSVR - */ - ifscr = 0; - fsrate = 0; - if (use_src) { - u64 n; - - ifscr = 1; - n = (u64)0x0400000 * fin; - do_div(n, fout); - fsrate = n; - } - - /* + * SRC_IFSCR * SRC_SRCCR / SRC_ROUTE_MODE0 */ + ifscr = 0; cr = 0x00011110; route = 0x0; if (use_src) { route = 0x1; + ifscr = 0x1; if (rsnd_src_sync_is_enabled(mod)) { cr |= 0x1; @@ -335,7 +385,6 @@ static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, rsnd_mod_write(mod, SRC_SRCIR, 1); /* initialize */ rsnd_mod_write(mod, SRC_ADINR, adinr); rsnd_mod_write(mod, SRC_IFSCR, ifscr); - rsnd_mod_write(mod, SRC_IFSVR, fsrate); rsnd_mod_write(mod, SRC_SRCCR, cr); rsnd_mod_write(mod, SRC_BSDSR, bsdsr_table[idx]); rsnd_mod_write(mod, SRC_BSISR, bsisr_table[idx]); @@ -348,6 +397,9 @@ static void rsnd_src_set_convert_rate(struct rsnd_dai_stream *io, rsnd_adg_set_src_timesel_gen2(mod, io, fin, fout); + /* update SRC_IFSVR */ + rsnd_src_set_convert_rate(io, mod); + return; convert_rate_err: @@ -467,7 +519,8 @@ static int rsnd_src_init(struct rsnd_mod *mod, int ret; /* reset sync convert_rate */ - src->sync.val = 0; + src->sync.val = + src->current_sync_rate = 0; ret = rsnd_mod_power_on(mod); if (ret < 0) @@ -475,7 +528,7 @@ static int rsnd_src_init(struct rsnd_mod *mod, rsnd_src_activation(mod); - rsnd_src_set_convert_rate(io, mod); + rsnd_src_init_convert_rate(io, mod); rsnd_src_status_clear(mod); @@ -493,7 +546,8 @@ static int rsnd_src_quit(struct rsnd_mod *mod, rsnd_mod_power_off(mod); /* reset sync convert_rate */ - src->sync.val = 0; + src->sync.val = + src->current_sync_rate = 0; return 0; } @@ -601,7 +655,7 @@ static int rsnd_src_pcm_new(struct rsnd_mod *mod, "SRC Out Rate Switch" : "SRC In Rate Switch", rsnd_kctrl_accept_anytime, - rsnd_src_set_convert_rate, + rsnd_src_init_convert_rate, &src->sen, 1); if (ret < 0) return ret; From 550b82651bbd4339ac415933bc21e979d439cab4 Mon Sep 17 00:00:00 2001 From: Vignesh Raman Date: Wed, 5 Feb 2025 13:46:46 +0530 Subject: [PATCH 0504/1443] drm/ci: uprev mesa Uprev mesa to adapt to the latest changes in mesa ci which includes new container jobs and stages. Also update lava-submit script to adapt to the recent changes in mesa to use LAVA rootfs overlays. Signed-off-by: Vignesh Raman Signed-off-by: Helen Koike Link: https://patchwork.freedesktop.org/patch/msgid/20250205081652.1928927-2-vignesh.raman@collabora.com --- drivers/gpu/drm/ci/build.sh | 2 +- drivers/gpu/drm/ci/build.yml | 103 +++++++++++++++++++++++++++++- drivers/gpu/drm/ci/container.yml | 22 +++---- drivers/gpu/drm/ci/gitlab-ci.yml | 81 +++++++++++++++-------- drivers/gpu/drm/ci/igt_runner.sh | 13 ++-- drivers/gpu/drm/ci/image-tags.yml | 11 +++- drivers/gpu/drm/ci/lava-submit.sh | 100 +++++++++++++++++++++-------- drivers/gpu/drm/ci/test.yml | 17 +++-- 8 files changed, 264 insertions(+), 85 deletions(-) diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh index 139b81db6312..19fe01257ab9 100644 --- a/drivers/gpu/drm/ci/build.sh +++ b/drivers/gpu/drm/ci/build.sh @@ -132,7 +132,7 @@ fi # Pass needed files to the test stage mkdir -p install cp -rfv .gitlab-ci/* install/. -cp -rfv ci/* install/. +cp -rfv bin/ci/* install/. cp -rfv install/common install/ci-common cp -rfv drivers/gpu/drm/ci/* install/. diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml index 9c198239033d..a2baa00a20db 100644 --- a/drivers/gpu/drm/ci/build.yml +++ b/drivers/gpu/drm/ci/build.yml @@ -2,7 +2,7 @@ extends: - .build-rules - .container+build-rules - stage: build + stage: build-only artifacts: paths: - artifacts @@ -110,3 +110,104 @@ build-nodebugfs:arm64: build:x86_64: extends: .build:x86_64 + +# Disable build jobs that we won't use +alpine-build-testing: + rules: + - when: never + +debian-android: + rules: + - when: never + +debian-arm32: + rules: + - when: never + +debian-arm32-asan: + rules: + - when: never + +debian-arm64: + rules: + - when: never + +debian-arm64-asan: + rules: + - when: never + +debian-arm64-build-test: + rules: + - when: never + +debian-arm64-release: + rules: + - when: never + +debian-build-testing: + rules: + - when: never + +debian-clang: + rules: + - when: never + +debian-clang-release: + rules: + - when: never + +debian-no-libdrm: + rules: + - when: never + +debian-ppc64el: + rules: + - when: never + +debian-release: + rules: + - when: never + +debian-s390x: + rules: + - when: never + +debian-testing: + rules: + - when: never + +debian-testing-asan: + rules: + - when: never + +debian-testing-msan: + rules: + - when: never + +debian-vulkan: + rules: + - when: never + +debian-x86_32: + rules: + - when: never + +fedora-release: + rules: + - when: never + +rustfmt: + rules: + - when: never + +shader-db: + rules: + - when: never + +windows-msvc: + rules: + - when: never + +yaml-toml-shell-py-test: + rules: + - when: never diff --git a/drivers/gpu/drm/ci/container.yml b/drivers/gpu/drm/ci/container.yml index 2a94f54ce4cf..07dc13ff865d 100644 --- a/drivers/gpu/drm/ci/container.yml +++ b/drivers/gpu/drm/ci/container.yml @@ -24,7 +24,7 @@ alpine/x86_64_build: rules: - when: never -debian/x86_64_test-vk: +debian/arm64_test-gl: rules: - when: never @@ -32,7 +32,15 @@ debian/arm64_test-vk: rules: - when: never -debian/arm64_test-gl: +debian/ppc64el_build: + rules: + - when: never + +debian/s390x_build: + rules: + - when: never + +debian/x86_64_test-vk: rules: - when: never @@ -56,14 +64,6 @@ windows_test_msvc: rules: - when: never -.debian/x86_64_build-mingw: - rules: - - when: never - -rustfmt: - rules: - - when: never - windows_msvc: rules: - - when: never \ No newline at end of file + - when: never diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml index b7214718723c..48ab2de9c90b 100644 --- a/drivers/gpu/drm/ci/gitlab-ci.yml +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -1,6 +1,6 @@ variables: DRM_CI_PROJECT_PATH: &drm-ci-project-path mesa/mesa - DRM_CI_COMMIT_SHA: &drm-ci-commit-sha c6a9a9c3bce90923f7700219354e0b6e5a3c9ba6 + DRM_CI_COMMIT_SHA: &drm-ci-commit-sha 7d3062470f3ccc6cb40540e772e902c7e2248024 UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git TARGET_BRANCH: drm-next @@ -20,6 +20,9 @@ variables: rm download-git-cache.sh set +o xtrace S3_JWT_FILE: /s3_jwt + S3_JWT_FILE_SCRIPT: |- + echo -n '${S3_JWT}' > '${S3_JWT_FILE}' && + unset CI_JOB_JWT S3_JWT # Unsetting vulnerable env variables S3_HOST: s3.freedesktop.org # This bucket is used to fetch the kernel image S3_KERNEL_BUCKET: mesa-rootfs @@ -31,11 +34,7 @@ variables: PIPELINE_ARTIFACTS_BASE: ${S3_HOST}/${S3_ARTIFACTS_BUCKET}/${CI_PROJECT_PATH}/${CI_PIPELINE_ID} # per-job artifact storage on MinIO JOB_ARTIFACTS_BASE: ${PIPELINE_ARTIFACTS_BASE}/${CI_JOB_ID} - # default kernel for rootfs before injecting the current kernel tree - KERNEL_REPO: "gfx-ci/linux" - KERNEL_TAG: "v6.6.21-mesa-f8ea" KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG} - PKG_REPO_REV: "3cc12a2a" LAVA_TAGS: subset-1-gfx LAVA_JOB_PRIORITY: 30 ARTIFACTS_BASE_URL: https://${CI_PROJECT_ROOT_NAMESPACE}.${CI_PAGES_DOMAIN}/-/${CI_PROJECT_NAME}/-/jobs/${CI_JOB_ID}/artifacts @@ -50,16 +49,13 @@ default: - export SCRIPTS_DIR=$(mktemp -d) - curl -L -s --retry 4 -f --retry-all-errors --retry-delay 60 -O --output-dir "${SCRIPTS_DIR}" "${DRM_CI_PROJECT_URL}/-/raw/${DRM_CI_COMMIT_SHA}/.gitlab-ci/setup-test-env.sh" - source ${SCRIPTS_DIR}/setup-test-env.sh - - echo -e "\e[0Ksection_start:$(date +%s):unset_env_vars_section[collapsed=true]\r\e[0KUnsetting vulnerable environment variables" - - echo -n "${S3_JWT}" > "${S3_JWT_FILE}" - - unset CI_JOB_JWT S3_JWT - - echo -e "\e[0Ksection_end:$(date +%s):unset_env_vars_section\r\e[0K" + - eval "$S3_JWT_FILE_SCRIPT" - echo -e "\e[0Ksection_start:$(date +%s):drm_ci_download_section[collapsed=true]\r\e[0KDownloading mesa from $DRM_CI_PROJECT_URL/-/archive/$DRM_CI_COMMIT_SHA/mesa-$DRM_CI_COMMIT_SHA.tar.gz" - cd $CI_PROJECT_DIR - curl --output - $DRM_CI_PROJECT_URL/-/archive/$DRM_CI_COMMIT_SHA/mesa-$DRM_CI_COMMIT_SHA.tar.gz | tar -xz - mv mesa-$DRM_CI_COMMIT_SHA/.gitlab-ci* . - - mv mesa-$DRM_CI_COMMIT_SHA/bin/ci . + - mv mesa-$DRM_CI_COMMIT_SHA/bin . - rm -rf mesa-$DRM_CI_COMMIT_SHA/ - echo -e "\e[0Ksection_end:$(date +%s):drm_ci_download_section\r\e[0K" @@ -85,6 +81,7 @@ include: - project: *drm-ci-project-path ref: *drm-ci-commit-sha file: + - '/.gitlab-ci/build/gitlab-ci.yml' - '/.gitlab-ci/container/gitlab-ci.yml' - '/.gitlab-ci/farm-rules.yml' - '/.gitlab-ci/lava/lava-gitlab-ci.yml' @@ -115,9 +112,10 @@ include: stages: - sanity - container - - code-validation - git-archive - - build + - build-for-tests + - build-only + - code-validation - amdgpu - i915 - mediatek @@ -197,7 +195,7 @@ stages: .ci-deqp-artifacts: artifacts: - name: "mesa_${CI_JOB_NAME}" + name: "${CI_PROJECT_NAME}_${CI_JOB_NAME}" when: always untracked: false paths: @@ -264,30 +262,63 @@ sanity: rules: - if: *is-pre-merge when: on_success - # Other cases default to never + - when: never variables: GIT_STRATEGY: none script: # ci-fairy check-commits --junit-xml=check-commits.xml - ci-fairy check-merge-request --require-allow-collaboration --junit-xml=check-merge-request.xml + - | + set -eu + image_tags=( + ALPINE_X86_64_LAVA_SSH_TAG + CONTAINER_TAG + DEBIAN_BASE_TAG + DEBIAN_BUILD_TAG + DEBIAN_PYUTILS_TAG + DEBIAN_TEST_GL_TAG + KERNEL_ROOTFS_TAG + KERNEL_TAG + PKG_REPO_REV + ) + for var in "${image_tags[@]}" + do + if [ "$(echo -n "${!var}" | wc -c)" -gt 20 ] + then + echo "$var is too long; please make sure it is at most 20 chars." + exit 1 + fi + done artifacts: when: on_failure reports: junit: check-*.xml + tags: + - placeholder-job -# Rules for tests that should not block merging, but should be available to -# optionally run with the "play" button in the UI in pre-merge non-marge -# pipelines. This should appear in "extends:" after any includes of -# test-source-dep.yml rules, so that these rules replace those. -.test-manual-mr: + +mr-label-maker-test: + extends: + - .fdo.ci-fairy + stage: sanity rules: - - !reference [.no_scheduled_pipelines-rules, rules] - - if: *is-forked-branch-or-pre-merge-not-for-marge - when: manual + - !reference [.mr-label-maker-rules, rules] variables: - JOB_TIMEOUT: 80 - + GIT_STRATEGY: fetch + timeout: 10m + script: + - set -eu + - python3 -m venv .venv + - source .venv/bin/activate + - pip install git+https://gitlab.freedesktop.org/freedesktop/mr-label-maker + - mr-label-maker --dry-run --mr $CI_MERGE_REQUEST_IID # Jobs that need to pass before spending hardware resources on further testing .required-for-hardware-jobs: - needs: [] + needs: + - job: clang-format + optional: true + - job: rustfmt + optional: true + - job: toml-lint + optional: true diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh index f38836ec837c..68b042e43b7f 100755 --- a/drivers/gpu/drm/ci/igt_runner.sh +++ b/drivers/gpu/drm/ci/igt_runner.sh @@ -47,7 +47,7 @@ else ARCH="x86_64" fi -curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s ${FDO_HTTP_CACHE_URI:-}$PIPELINE_ARTIFACTS_BASE/$ARCH/igt.tar.gz | tar --zstd -v -x -C / +curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s $PIPELINE_ARTIFACTS_BASE/$ARCH/igt.tar.gz | tar --zstd -v -x -C / TESTLIST="/igt/libexec/igt-gpu-tools/ci-testlist.txt" @@ -69,7 +69,7 @@ igt-runner \ run \ --igt-folder /igt/libexec/igt-gpu-tools \ --caselist $TESTLIST \ - --output /results \ + --output $RESULTS_DIR \ -vvvv \ $IGT_SKIPS \ $IGT_FLAKES \ @@ -80,13 +80,10 @@ set -e deqp-runner junit \ --testsuite IGT \ - --results /results/failures.csv \ - --output /results/junit.xml \ + --results $RESULTS_DIR/failures.csv \ + --output $RESULTS_DIR/junit.xml \ --limit 50 \ - --template "See https://$CI_PROJECT_ROOT_NAMESPACE.pages.freedesktop.org/-/$CI_PROJECT_NAME/-/jobs/$CI_JOB_ID/artifacts/results/{{testcase}}.xml" - -# Store the results also in the simpler format used by the runner in ChromeOS CI -#sed -r 's/(dmesg-warn|pass)/success/g' /results/results.txt > /results/results_simple.txt + --template "See $ARTIFACTS_BASE_URL/results/{{testcase}}.xml" cd $oldpath exit $ret diff --git a/drivers/gpu/drm/ci/image-tags.yml b/drivers/gpu/drm/ci/image-tags.yml index 8d8b9e71852e..20049f3626b2 100644 --- a/drivers/gpu/drm/ci/image-tags.yml +++ b/drivers/gpu/drm/ci/image-tags.yml @@ -1,5 +1,5 @@ variables: - CONTAINER_TAG: "2024-09-09-uprevs" + CONTAINER_TAG: "20250204-mesa-uprev" DEBIAN_X86_64_BUILD_BASE_IMAGE: "debian/x86_64_build-base" DEBIAN_BASE_TAG: "${CONTAINER_TAG}" @@ -7,9 +7,16 @@ variables: DEBIAN_BUILD_TAG: "${CONTAINER_TAG}" KERNEL_ROOTFS_TAG: "${CONTAINER_TAG}" + # default kernel for rootfs before injecting the current kernel tree + KERNEL_TAG: "v6.13-rc4-mesa-5e77" + KERNEL_REPO: "gfx-ci/linux" + PKG_REPO_REV: "bca9635d" DEBIAN_X86_64_TEST_BASE_IMAGE: "debian/x86_64_test-base" DEBIAN_X86_64_TEST_IMAGE_GL_PATH: "debian/x86_64_test-gl" DEBIAN_TEST_GL_TAG: "${CONTAINER_TAG}" - ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}" \ No newline at end of file + DEBIAN_PYUTILS_IMAGE: "debian/x86_64_pyutils" + DEBIAN_PYUTILS_TAG: "${CONTAINER_TAG}" + + ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}" diff --git a/drivers/gpu/drm/ci/lava-submit.sh b/drivers/gpu/drm/ci/lava-submit.sh index 6add15083c78..6e5ac51e8c0a 100755 --- a/drivers/gpu/drm/ci/lava-submit.sh +++ b/drivers/gpu/drm/ci/lava-submit.sh @@ -1,58 +1,102 @@ -#!/bin/bash +#!/usr/bin/env bash # SPDX-License-Identifier: MIT +# shellcheck disable=SC2086 # we want word splitting +# shellcheck disable=SC1091 # paths only become valid at runtime -set -e -set -x +# If we run in the fork (not from mesa or Marge-bot), reuse mainline kernel and rootfs, if exist. +_check_artifact_path() { + _url="https://${1}/${2}" + if curl -s -o /dev/null -I -L -f --retry 4 --retry-delay 15 "${_url}"; then + echo -n "${_url}" + fi +} -# Try to use the kernel and rootfs built in mainline first, so we're more -# likely to hit cache -if curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s "https://${BASE_SYSTEM_MAINLINE_HOST_PATH}/done"; then - BASE_SYSTEM_HOST_PATH="${BASE_SYSTEM_MAINLINE_HOST_PATH}" -else - BASE_SYSTEM_HOST_PATH="${BASE_SYSTEM_FORK_HOST_PATH}" -fi +get_path_to_artifact() { + _mainline_artifact="$(_check_artifact_path ${BASE_SYSTEM_MAINLINE_HOST_PATH} ${1})" + if [ -n "${_mainline_artifact}" ]; then + echo -n "${_mainline_artifact}" + return + fi + _fork_artifact="$(_check_artifact_path ${BASE_SYSTEM_FORK_HOST_PATH} ${1})" + if [ -n "${_fork_artifact}" ]; then + echo -n "${_fork_artifact}" + return + fi + set +x + error "Sorry, I couldn't find a viable built path for ${1} in either mainline or a fork." >&2 + echo "" >&2 + echo "If you're working on CI, this probably means that you're missing a dependency:" >&2 + echo "this job ran ahead of the job which was supposed to upload that artifact." >&2 + echo "" >&2 + echo "If you aren't working on CI, please ping @mesa/ci-helpers to see if we can help." >&2 + echo "" >&2 + echo "This job is going to fail, because I can't find the resources I need. Sorry." >&2 + set -x + exit 1 +} + +. "${SCRIPTS_DIR}/setup-test-env.sh" + +section_start prepare_rootfs "Preparing root filesystem" + +set -ex + +section_switch rootfs "Assembling root filesystem" +ROOTFS_URL="$(get_path_to_artifact lava-rootfs.tar.zst)" +[ $? != 1 ] || exit 1 rm -rf results mkdir -p results/job-rootfs-overlay/ -cp artifacts/ci-common/capture-devcoredump.sh results/job-rootfs-overlay/ +artifacts/ci-common/generate-env.sh > results/job-rootfs-overlay/set-job-env-vars.sh cp artifacts/ci-common/init-*.sh results/job-rootfs-overlay/ -cp artifacts/ci-common/intel-gpu-freq.sh results/job-rootfs-overlay/ cp "$SCRIPTS_DIR"/setup-test-env.sh results/job-rootfs-overlay/ -# Prepare env vars for upload. -section_start variables "Variables passed through:" -KERNEL_IMAGE_BASE="https://${BASE_SYSTEM_HOST_PATH}" \ - artifacts/ci-common/generate-env.sh | tee results/job-rootfs-overlay/set-job-env-vars.sh -section_end variables - tar zcf job-rootfs-overlay.tar.gz -C results/job-rootfs-overlay/ . ci-fairy s3cp --token-file "${S3_JWT_FILE}" job-rootfs-overlay.tar.gz "https://${JOB_ROOTFS_OVERLAY_PATH}" +# Prepare env vars for upload. +section_switch variables "Environment variables passed through to device:" +cat results/job-rootfs-overlay/set-job-env-vars.sh + +section_switch lava_submit "Submitting job for scheduling" + touch results/lava.log tail -f results/lava.log & - PYTHONPATH=artifacts/ artifacts/lava/lava_job_submitter.py \ - submit \ + --farm "${FARM}" \ + --device-type "${DEVICE_TYPE}" \ + --boot-method "${BOOT_METHOD}" \ + --job-timeout-min $((CI_JOB_TIMEOUT/60 - 5)) \ --dump-yaml \ --pipeline-info "$CI_JOB_NAME: $CI_PIPELINE_URL on $CI_COMMIT_REF_NAME ${CI_NODE_INDEX}/${CI_NODE_TOTAL}" \ - --rootfs-url-prefix "https://${BASE_SYSTEM_HOST_PATH}" \ + --rootfs-url "${ROOTFS_URL}" \ --kernel-url-prefix "https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}" \ - --build-url "${FDO_HTTP_CACHE_URI:-}https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/kernel-files.tar.zst" \ - --job-rootfs-overlay-url "${FDO_HTTP_CACHE_URI:-}https://${JOB_ROOTFS_OVERLAY_PATH}" \ - --job-timeout-min ${JOB_TIMEOUT:-80} \ + --kernel-external "${EXTERNAL_KERNEL_TAG}" \ --first-stage-init artifacts/ci-common/init-stage1.sh \ - --ci-project-dir "${CI_PROJECT_DIR}" \ - --device-type "${DEVICE_TYPE}" \ - --farm "${FARM}" \ --dtb-filename "${DTB}" \ --jwt-file "${S3_JWT_FILE}" \ --kernel-image-name "${KERNEL_IMAGE_NAME}" \ --kernel-image-type "${KERNEL_IMAGE_TYPE}" \ - --boot-method "${BOOT_METHOD}" \ --visibility-group "${VISIBILITY_GROUP}" \ --lava-tags "${LAVA_TAGS}" \ --mesa-job-name "$CI_JOB_NAME" \ --structured-log-file "results/lava_job_detail.json" \ --ssh-client-image "${LAVA_SSH_CLIENT_IMAGE}" \ + --project-name "${CI_PROJECT_NAME}" \ + --starting-section "${CURRENT_SECTION}" \ + --job-submitted-at "${CI_JOB_STARTED_AT}" \ + - append-overlay \ + --name=kernel-build \ + --url="${FDO_HTTP_CACHE_URI:-}https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/kernel-files.tar.zst" \ + --compression=zstd \ + --path="${CI_PROJECT_DIR}" \ + --format=tar \ + - append-overlay \ + --name=job-overlay \ + --url="https://${JOB_ROOTFS_OVERLAY_PATH}" \ + --compression=gz \ + --path="/" \ + --format=tar \ + - submit \ >> results/lava.log diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml index f0ef60c8f56d..c5c786bf419b 100644 --- a/drivers/gpu/drm/ci/test.yml +++ b/drivers/gpu/drm/ci/test.yml @@ -1,21 +1,21 @@ .test-rules: rules: - - if: '$FD_FARM == "offline" && $RUNNER_TAG =~ /^google-freedreno-/' - when: never - - if: '$COLLABORA_FARM == "offline" && $RUNNER_TAG =~ /^mesa-ci-x86-64-lava-/' - when: never - !reference [.no_scheduled_pipelines-rules, rules] + - !reference [.collabora-farm-rules, rules] + - !reference [.google-freedreno-farm-rules, rules] - when: on_success .lava-test: extends: - .test-rules + - .build-rules + - .container+build-rules timeout: "1h30m" script: # Note: Build dir (and thus install) may be dirty due to GIT_STRATEGY - rm -rf install - tar -xf artifacts/install.tar - - mv install/* artifacts/. + - mv -n install/* artifacts/. # Override it with our lava-submit.sh script - ./artifacts/lava-submit.sh @@ -32,6 +32,7 @@ - alpine/x86_64_lava_ssh_client - kernel+rootfs_arm32 - debian/x86_64_build + - python-artifacts - testing:arm32 - igt:arm32 @@ -48,6 +49,7 @@ - alpine/x86_64_lava_ssh_client - kernel+rootfs_arm64 - debian/x86_64_build + - python-artifacts - testing:arm64 - igt:arm64 @@ -64,6 +66,7 @@ - alpine/x86_64_lava_ssh_client - kernel+rootfs_x86_64 - debian/x86_64_build + - python-artifacts - testing:x86_64 - igt:x86_64 @@ -453,8 +456,6 @@ virtio_gpu:none: script: - ln -sf $CI_PROJECT_DIR/install /install - mv install/bzImage /lava-files/bzImage - - mkdir -p $CI_PROJECT_DIR/results - - ln -sf $CI_PROJECT_DIR/results /results - install/crosvm-runner.sh install/igt_runner.sh needs: - debian/x86_64_test-gl @@ -476,8 +477,6 @@ vkms:none: - ln -sf $CI_PROJECT_DIR/install /install - mv install/bzImage /lava-files/bzImage - mkdir -p /lib/modules - - mkdir -p $CI_PROJECT_DIR/results - - ln -sf $CI_PROJECT_DIR/results /results - ./install/crosvm-runner.sh ./install/igt_runner.sh needs: - debian/x86_64_test-gl From df54f04f2020dd750d7c2d6c336ef91375d0db02 Mon Sep 17 00:00:00 2001 From: Vignesh Raman Date: Wed, 5 Feb 2025 13:46:47 +0530 Subject: [PATCH 0505/1443] drm/ci: update gitlab rules Update gitlab rules to include scheduled pipelines. Signed-off-by: Vignesh Raman Signed-off-by: Helen Koike Link: https://patchwork.freedesktop.org/patch/msgid/20250205081652.1928927-3-vignesh.raman@collabora.com --- drivers/gpu/drm/ci/build.yml | 1 - drivers/gpu/drm/ci/gitlab-ci.yml | 116 +++++++++++++------------------ drivers/gpu/drm/ci/test.yml | 26 +++---- 3 files changed, 63 insertions(+), 80 deletions(-) diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml index a2baa00a20db..274f118533a7 100644 --- a/drivers/gpu/drm/ci/build.yml +++ b/drivers/gpu/drm/ci/build.yml @@ -1,6 +1,5 @@ .build: extends: - - .build-rules - .container+build-rules stage: build-only artifacts: diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml index 48ab2de9c90b..f04aabe8327c 100644 --- a/drivers/gpu/drm/ci/gitlab-ci.yml +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -41,6 +41,7 @@ variables: # Python scripts for structured logger PYTHONPATH: "$PYTHONPATH:$CI_PROJECT_DIR/install" + default: id_tokens: S3_JWT: @@ -67,6 +68,7 @@ default: export S3_JWT="$(<${S3_JWT_FILE})" && rm "${S3_JWT_FILE}" + include: - project: 'freedesktop/ci-templates' ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811 @@ -126,33 +128,27 @@ stages: - rockchip - software-driver + # YAML anchors for rule conditions # -------------------------------- .rules-anchors: rules: - # Pipeline for forked project branch - - if: &is-forked-branch '$CI_COMMIT_BRANCH && $CI_PROJECT_NAMESPACE != "mesa"' - when: manual - # Forked project branch / pre-merge pipeline not for Marge bot - - if: &is-forked-branch-or-pre-merge-not-for-marge '$CI_PROJECT_NAMESPACE != "mesa" || ($GITLAB_USER_LOGIN != "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event")' - when: manual - # Pipeline runs for the main branch of the upstream Mesa project - - if: &is-mesa-main '$CI_PROJECT_NAMESPACE == "mesa" && $CI_COMMIT_REF_NAME == $CI_DEFAULT_BRANCH && $CI_COMMIT_BRANCH' - when: always - # Post-merge pipeline - - if: &is-post-merge '$CI_PROJECT_NAMESPACE == "mesa" && $CI_COMMIT_BRANCH' - when: on_success - # Post-merge pipeline, not for Marge Bot - - if: &is-post-merge-not-for-marge '$CI_PROJECT_NAMESPACE == "mesa" && $GITLAB_USER_LOGIN != "marge-bot" && $CI_COMMIT_BRANCH' - when: on_success + # do not duplicate pipelines on merge pipelines + - if: $CI_COMMIT_BRANCH && $CI_OPEN_MERGE_REQUESTS && $CI_PIPELINE_SOURCE == "push" + when: never + # merge pipeline + - if: &is-merge-attempt $GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event" + # post-merge pipeline + - if: &is-post-merge $GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "push" # Pre-merge pipeline - - if: &is-pre-merge '$CI_PIPELINE_SOURCE == "merge_request_event"' - when: on_success - # Pre-merge pipeline for Marge Bot - - if: &is-pre-merge-for-marge '$GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event"' - when: on_success + - if: &is-pre-merge $CI_PIPELINE_SOURCE == "merge_request_event" # Push to a branch on a fork - - &is-fork-push '$CI_PROJECT_NAMESPACE != "mesa" && $CI_PIPELINE_SOURCE == "push"' + - if: &is-fork-push $CI_PROJECT_NAMESPACE != "mesa" && $CI_PIPELINE_SOURCE == "push" + # nightly pipeline + - if: &is-scheduled-pipeline $CI_PIPELINE_SOURCE == "schedule" + # pipeline for direct pushes that bypassed the CI + - if: &is-direct-push $CI_PROJECT_NAMESPACE == "mesa" && $CI_PIPELINE_SOURCE == "push" && $GITLAB_USER_LOGIN != "marge-bot" + # Rules applied to every job in the pipeline .common-rules: @@ -160,39 +156,48 @@ stages: - if: *is-fork-push when: manual + .never-post-merge-rules: rules: - if: *is-post-merge when: never -# Rule to filter for only scheduled pipelines. -.scheduled_pipeline-rules: - rules: - - if: &is-scheduled-pipeline '$CI_PIPELINE_SOURCE == "schedule"' - when: on_success -# Generic rule to not run the job during scheduled pipelines. Jobs that aren't -# something like a nightly run should include this rule. -.no_scheduled_pipelines-rules: - rules: - - if: *is-scheduled-pipeline - when: never - -# When to automatically run the CI for build jobs -.build-rules: - rules: - - !reference [.no_scheduled_pipelines-rules, rules] - - !reference [.never-post-merge-rules, rules] - # Run automatically once all dependency jobs have passed - - when: on_success - -# When to automatically run the CI for container jobs .container+build-rules: rules: - - !reference [.no_scheduled_pipelines-rules, rules] + - !reference [.common-rules, rules] + # Run when re-enabling a disabled farm, but not when disabling it + - !reference [.disable-farm-mr-rules, rules] + # Never run immediately after merging, as we just ran everything - !reference [.never-post-merge-rules, rules] + # Build everything in merge pipelines, if any files affecting the pipeline + # were changed + - if: *is-merge-attempt + changes: &all_paths + - drivers/gpu/drm/ci/**/* + when: on_success + # Same as above, but for pre-merge pipelines + - if: *is-pre-merge + changes: + *all_paths + when: manual + # Skip everything for pre-merge and merge pipelines which don't change + # anything in the build + - if: *is-merge-attempt + when: never + - if: *is-pre-merge + when: never + # Build everything after someone bypassed the CI + - if: *is-direct-push + when: on_success + # Build everything in scheduled pipelines + - if: *is-scheduled-pipeline + when: on_success + # Allow building everything in fork pipelines, but build nothing unless + # manually triggered - when: manual + .ci-deqp-artifacts: artifacts: name: "${CI_PROJECT_NAME}_${CI_JOB_NAME}" @@ -206,31 +211,7 @@ stages: - _build/meson-logs/strace -.container-rules: - rules: - - !reference [.no_scheduled_pipelines-rules, rules] - - !reference [.never-post-merge-rules, rules] - # Run pipeline by default in the main project if any CI pipeline - # configuration files were changed, to ensure docker images are up to date - - if: *is-post-merge - changes: - - drivers/gpu/drm/ci/**/* - when: on_success - # Run pipeline by default if it was triggered by Marge Bot, is for a - # merge request, and any files affecting the pipeline were changed - - if: *is-pre-merge-for-marge - when: on_success - # Run pipeline by default in the main project if it was not triggered by - # Marge Bot, and any files affecting the pipeline were changed - - if: *is-post-merge-not-for-marge - when: on_success - # Allow triggering jobs manually in other cases - - when: manual - - - # Git archive - make git archive: extends: - .fdo.ci-fairy @@ -313,6 +294,7 @@ mr-label-maker-test: - pip install git+https://gitlab.freedesktop.org/freedesktop/mr-label-maker - mr-label-maker --dry-run --mr $CI_MERGE_REQUEST_IID + # Jobs that need to pass before spending hardware resources on further testing .required-for-hardware-jobs: needs: diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml index c5c786bf419b..6a1e059858e5 100644 --- a/drivers/gpu/drm/ci/test.yml +++ b/drivers/gpu/drm/ci/test.yml @@ -1,16 +1,11 @@ -.test-rules: - rules: - - !reference [.no_scheduled_pipelines-rules, rules] - - !reference [.collabora-farm-rules, rules] - - !reference [.google-freedreno-farm-rules, rules] - - when: on_success - .lava-test: extends: - - .test-rules - - .build-rules - .container+build-rules timeout: "1h30m" + rules: + - !reference [.scheduled_pipeline-rules, rules] + - !reference [.collabora-farm-rules, rules] + - when: on_success script: # Note: Build dir (and thus install) may be dirty due to GIT_STRATEGY - rm -rf install @@ -74,8 +69,11 @@ extends: - .baremetal-test-arm64 - .use-debian/baremetal_arm64_test - - .test-rules timeout: "1h30m" + rules: + - !reference [.scheduled_pipeline-rules, rules] + - !reference [.google-freedreno-farm-rules, rules] + - when: on_success variables: FDO_CI_CONCURRENT: 10 HWCI_TEST_SCRIPT: "/install/igt_runner.sh" @@ -444,13 +442,15 @@ panfrost:g12b: virtio_gpu:none: stage: software-driver timeout: "1h30m" + rules: + - !reference [.scheduled_pipeline-rules, rules] + - when: on_success variables: CROSVM_GALLIUM_DRIVER: llvmpipe DRIVER_NAME: virtio_gpu GPU_VERSION: none extends: - .test-gl - - .test-rules tags: - kvm script: @@ -465,12 +465,14 @@ virtio_gpu:none: vkms:none: stage: software-driver timeout: "1h30m" + rules: + - !reference [.scheduled_pipeline-rules, rules] + - when: on_success variables: DRIVER_NAME: vkms GPU_VERSION: none extends: - .test-gl - - .test-rules tags: - kvm script: From c5d82ed17eeaad79460e402da7ae58a9e37d8f94 Mon Sep 17 00:00:00 2001 From: Vignesh Raman Date: Wed, 5 Feb 2025 13:46:48 +0530 Subject: [PATCH 0506/1443] drm/ci: update expectation files Update expectation files for the mesa uprev. Signed-off-by: Vignesh Raman Signed-off-by: Helen Koike Link: https://patchwork.freedesktop.org/patch/msgid/20250205081652.1928927-4-vignesh.raman@collabora.com --- drivers/gpu/drm/ci/xfails/vkms-none-fails.txt | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt b/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt index 85647eca4e55..6ebcc7d89fbd 100644 --- a/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt +++ b/drivers/gpu/drm/ci/xfails/vkms-none-fails.txt @@ -13,6 +13,8 @@ kms_cursor_legacy@flip-vs-cursor-crc-legacy,Fail kms_cursor_legacy@flip-vs-cursor-legacy,Fail kms_flip@flip-vs-modeset-vs-hang,Fail kms_flip@flip-vs-panning-vs-hang,Fail +kms_flip@flip-vs-suspend,Fail +kms_flip@flip-vs-suspend-interruptible,Fail kms_lease@lease-uevent,Fail kms_writeback@writeback-check-output,Fail kms_writeback@writeback-check-output-XRGB2101010,Fail From c4d3dfd8ccaef2cbd374860e307f1e056854a472 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Wed, 5 Feb 2025 14:21:36 +0100 Subject: [PATCH 0507/1443] Revert "i2c: Replace list-based mechanism for handling userspace-created clients" This reverts commit 3cfe39b3a845593a485ab1c716615979004ef9f6. Mux handling is not sufficiently implemented. It needs more time. Signed-off-by: Wolfram Sang --- drivers/i2c/i2c-core-base.c | 63 +++++++++++++++++++++++-------------- include/linux/i2c.h | 7 ++++- 2 files changed, 46 insertions(+), 24 deletions(-) diff --git a/drivers/i2c/i2c-core-base.c b/drivers/i2c/i2c-core-base.c index 5546184df05f..ddac2f155718 100644 --- a/drivers/i2c/i2c-core-base.c +++ b/drivers/i2c/i2c-core-base.c @@ -1300,12 +1300,14 @@ new_device_store(struct device *dev, struct device_attribute *attr, info.flags |= I2C_CLIENT_SLAVE; } - info.flags |= I2C_CLIENT_USER; - client = i2c_new_client_device(adap, &info); if (IS_ERR(client)) return PTR_ERR(client); + /* Keep track of the added device */ + mutex_lock(&adap->userspace_clients_lock); + list_add_tail(&client->detected, &adap->userspace_clients); + mutex_unlock(&adap->userspace_clients_lock); dev_info(dev, "%s: Instantiated device %s at 0x%02hx\n", "new_device", info.type, info.addr); @@ -1313,15 +1315,6 @@ new_device_store(struct device *dev, struct device_attribute *attr, } static DEVICE_ATTR_WO(new_device); -static int __i2c_find_user_addr(struct device *dev, const void *addrp) -{ - struct i2c_client *client = i2c_verify_client(dev); - unsigned short addr = *(unsigned short *)addrp; - - return client && client->flags & I2C_CLIENT_USER && - i2c_encode_flags_to_addr(client) == addr; -} - /* * And of course let the users delete the devices they instantiated, if * they got it wrong. This interface can only be used to delete devices @@ -1336,7 +1329,7 @@ delete_device_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct i2c_adapter *adap = to_i2c_adapter(dev); - struct device *child_dev; + struct i2c_client *client, *next; unsigned short addr; char end; int res; @@ -1352,19 +1345,28 @@ delete_device_store(struct device *dev, struct device_attribute *attr, return -EINVAL; } - mutex_lock(&core_lock); /* Make sure the device was added through sysfs */ - child_dev = device_find_child(&adap->dev, &addr, __i2c_find_user_addr); - if (child_dev) { - i2c_unregister_device(i2c_verify_client(child_dev)); - put_device(child_dev); - } else { - dev_err(dev, "Can't find userspace-created device at %#x\n", addr); - count = -ENOENT; - } - mutex_unlock(&core_lock); + res = -ENOENT; + mutex_lock_nested(&adap->userspace_clients_lock, + i2c_adapter_depth(adap)); + list_for_each_entry_safe(client, next, &adap->userspace_clients, + detected) { + if (i2c_encode_flags_to_addr(client) == addr) { + dev_info(dev, "%s: Deleting device %s at 0x%02hx\n", + "delete_device", client->name, client->addr); - return count; + list_del(&client->detected); + i2c_unregister_device(client); + res = count; + break; + } + } + mutex_unlock(&adap->userspace_clients_lock); + + if (res < 0) + dev_err(dev, "%s: Can't find device in list\n", + "delete_device"); + return res; } static DEVICE_ATTR_IGNORE_LOCKDEP(delete_device, S_IWUSR, NULL, delete_device_store); @@ -1535,6 +1537,8 @@ static int i2c_register_adapter(struct i2c_adapter *adap) adap->locked_flags = 0; rt_mutex_init(&adap->bus_lock); rt_mutex_init(&adap->mux_lock); + mutex_init(&adap->userspace_clients_lock); + INIT_LIST_HEAD(&adap->userspace_clients); /* Set default timeout to 1 second if not already set */ if (adap->timeout == 0) @@ -1726,6 +1730,7 @@ static int __unregister_dummy(struct device *dev, void *dummy) void i2c_del_adapter(struct i2c_adapter *adap) { struct i2c_adapter *found; + struct i2c_client *client, *next; /* First make sure that this adapter was ever added */ mutex_lock(&core_lock); @@ -1738,6 +1743,18 @@ void i2c_del_adapter(struct i2c_adapter *adap) i2c_acpi_remove_space_handler(adap); + /* Remove devices instantiated from sysfs */ + mutex_lock_nested(&adap->userspace_clients_lock, + i2c_adapter_depth(adap)); + list_for_each_entry_safe(client, next, &adap->userspace_clients, + detected) { + dev_dbg(&adap->dev, "Removing %s at 0x%x\n", client->name, + client->addr); + list_del(&client->detected); + i2c_unregister_device(client); + } + mutex_unlock(&adap->userspace_clients_lock); + /* Detach any active clients. This can't fail, thus we do not * check the returned value. This is a two-pass process, because * we can't remove the dummy devices during the first pass: they diff --git a/include/linux/i2c.h b/include/linux/i2c.h index c31fd1dba3bd..4955d9e76c5f 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -313,6 +313,8 @@ struct i2c_driver { * @dev: Driver model device node for the slave. * @init_irq: IRQ that was set at initialization * @irq: indicates the IRQ generated by this device (if any) + * @detected: member of an i2c_driver.clients list or i2c-core's + * userspace_devices list * @slave_cb: Callback when I2C slave mode of an adapter is used. The adapter * calls it to pass on slave events to the slave driver. * @devres_group_id: id of the devres group that will be created for resources @@ -333,7 +335,6 @@ struct i2c_client { #define I2C_CLIENT_HOST_NOTIFY 0x40 /* We want to use I2C host notify */ #define I2C_CLIENT_WAKE 0x80 /* for board_info; true iff can wake */ #define I2C_CLIENT_AUTO 0x100 /* client was auto-detected */ -#define I2C_CLIENT_USER 0x200 /* client was userspace-created */ #define I2C_CLIENT_SCCB 0x9000 /* Use Omnivision SCCB protocol */ /* Must match I2C_M_STOP|IGNORE_NAK */ @@ -345,6 +346,7 @@ struct i2c_client { struct device dev; /* the device structure */ int init_irq; /* irq set at initialization */ int irq; /* irq issued by device */ + struct list_head detected; #if IS_ENABLED(CONFIG_I2C_SLAVE) i2c_slave_cb_t slave_cb; /* callback for slave mode */ #endif @@ -751,6 +753,9 @@ struct i2c_adapter { char name[48]; struct completion dev_released; + struct mutex userspace_clients_lock; + struct list_head userspace_clients; + struct i2c_bus_recovery_info *bus_recovery_info; const struct i2c_adapter_quirks *quirks; From 3bfa08fe9ec8dd79e183c88e1275be74191e7bc8 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Wed, 5 Feb 2025 14:22:12 +0100 Subject: [PATCH 0508/1443] Revert "i2c: Replace list-based mechanism for handling auto-detected clients" This reverts commit 56a50667cbcfaf95eea9128d5676af94e54b51a8. Mux handling is not sufficiently implemented. It needs more time. Signed-off-by: Wolfram Sang --- drivers/i2c/i2c-core-base.c | 52 +++++++++++++++++++++++++------------ include/linux/i2c.h | 3 ++- 2 files changed, 38 insertions(+), 17 deletions(-) diff --git a/drivers/i2c/i2c-core-base.c b/drivers/i2c/i2c-core-base.c index ddac2f155718..35a221e2c11c 100644 --- a/drivers/i2c/i2c-core-base.c +++ b/drivers/i2c/i2c-core-base.c @@ -1704,6 +1704,23 @@ int i2c_add_numbered_adapter(struct i2c_adapter *adap) } EXPORT_SYMBOL_GPL(i2c_add_numbered_adapter); +static void i2c_do_del_adapter(struct i2c_driver *driver, + struct i2c_adapter *adapter) +{ + struct i2c_client *client, *_n; + + /* Remove the devices we created ourselves as the result of hardware + * probing (using a driver's detect method) */ + list_for_each_entry_safe(client, _n, &driver->clients, detected) { + if (client->adapter == adapter) { + dev_dbg(&adapter->dev, "Removing %s at 0x%x\n", + client->name, client->addr); + list_del(&client->detected); + i2c_unregister_device(client); + } + } +} + static int __unregister_client(struct device *dev, void *dummy) { struct i2c_client *client = i2c_verify_client(dev); @@ -1719,6 +1736,12 @@ static int __unregister_dummy(struct device *dev, void *dummy) return 0; } +static int __process_removed_adapter(struct device_driver *d, void *data) +{ + i2c_do_del_adapter(to_i2c_driver(d), data); + return 0; +} + /** * i2c_del_adapter - unregister I2C adapter * @adap: the adapter being unregistered @@ -1742,6 +1765,11 @@ void i2c_del_adapter(struct i2c_adapter *adap) } i2c_acpi_remove_space_handler(adap); + /* Tell drivers about this removal */ + mutex_lock(&core_lock); + bus_for_each_drv(&i2c_bus_type, NULL, adap, + __process_removed_adapter); + mutex_unlock(&core_lock); /* Remove devices instantiated from sysfs */ mutex_lock_nested(&adap->userspace_clients_lock, @@ -1760,10 +1788,8 @@ void i2c_del_adapter(struct i2c_adapter *adap) * we can't remove the dummy devices during the first pass: they * could have been instantiated by real devices wishing to clean * them up properly, so we give them a chance to do that first. */ - mutex_lock(&core_lock); device_for_each_child(&adap->dev, NULL, __unregister_client); device_for_each_child(&adap->dev, NULL, __unregister_dummy); - mutex_unlock(&core_lock); /* device name is gone after device_unregister */ dev_dbg(&adap->dev, "adapter [%s] unregistered\n", adap->name); @@ -1983,6 +2009,7 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver) /* add the driver to the list of i2c drivers in the driver core */ driver->driver.owner = owner; driver->driver.bus = &i2c_bus_type; + INIT_LIST_HEAD(&driver->clients); /* When registration returns, the driver core * will have called probe() for all matching-but-unbound devices. @@ -2000,13 +2027,10 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver) } EXPORT_SYMBOL(i2c_register_driver); -static int __i2c_unregister_detected_client(struct device *dev, void *argp) +static int __process_removed_driver(struct device *dev, void *data) { - struct i2c_client *client = i2c_verify_client(dev); - - if (client && client->flags & I2C_CLIENT_AUTO) - i2c_unregister_device(client); - + if (dev->type == &i2c_adapter_type) + i2c_do_del_adapter(data, to_i2c_adapter(dev)); return 0; } @@ -2017,12 +2041,7 @@ static int __i2c_unregister_detected_client(struct device *dev, void *argp) */ void i2c_del_driver(struct i2c_driver *driver) { - mutex_lock(&core_lock); - /* Satisfy __must_check, function can't fail */ - if (driver_for_each_device(&driver->driver, NULL, NULL, - __i2c_unregister_detected_client)) { - } - mutex_unlock(&core_lock); + i2c_for_each_dev(driver, __process_removed_driver); driver_unregister(&driver->driver); pr_debug("driver [%s] unregistered\n", driver->driver.name); @@ -2449,7 +2468,6 @@ static int i2c_detect_address(struct i2c_client *temp_client, /* Finally call the custom detection function */ memset(&info, 0, sizeof(struct i2c_board_info)); info.addr = addr; - info.flags = I2C_CLIENT_AUTO; err = driver->detect(temp_client, &info); if (err) { /* -ENODEV is returned if the detection fails. We catch it @@ -2476,7 +2494,9 @@ static int i2c_detect_address(struct i2c_client *temp_client, dev_dbg(&adapter->dev, "Creating %s at 0x%02x\n", info.type, info.addr); client = i2c_new_client_device(adapter, &info); - if (IS_ERR(client)) + if (!IS_ERR(client)) + list_add_tail(&client->detected, &driver->clients); + else dev_err(&adapter->dev, "Failed creating %s at 0x%02x\n", info.type, info.addr); } diff --git a/include/linux/i2c.h b/include/linux/i2c.h index 4955d9e76c5f..2b2af24d2a43 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -244,6 +244,7 @@ enum i2c_driver_flags { * @id_table: List of I2C devices supported by this driver * @detect: Callback for device detection * @address_list: The I2C addresses to probe (for detect) + * @clients: List of detected clients we created (for i2c-core use only) * @flags: A bitmask of flags defined in &enum i2c_driver_flags * * The driver.owner field should be set to the module owner of this driver. @@ -298,6 +299,7 @@ struct i2c_driver { /* Device detection callback for automatic device creation */ int (*detect)(struct i2c_client *client, struct i2c_board_info *info); const unsigned short *address_list; + struct list_head clients; u32 flags; }; @@ -334,7 +336,6 @@ struct i2c_client { #define I2C_CLIENT_SLAVE 0x20 /* we are the slave */ #define I2C_CLIENT_HOST_NOTIFY 0x40 /* We want to use I2C host notify */ #define I2C_CLIENT_WAKE 0x80 /* for board_info; true iff can wake */ -#define I2C_CLIENT_AUTO 0x100 /* client was auto-detected */ #define I2C_CLIENT_SCCB 0x9000 /* Use Omnivision SCCB protocol */ /* Must match I2C_M_STOP|IGNORE_NAK */ From 015b7dae084fa95465ff89f6cbf15fe49906a370 Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Mon, 3 Feb 2025 12:01:23 +0100 Subject: [PATCH 0509/1443] gpio: sim: lock hog configfs items if present Depending on the user config, the leaf entry may be the hog directory, not line. Check it and lock the correct item. Fixes: 8bd76b3d3f3a ("gpio: sim: lock up configfs that an instantiated device depends on") Tested-by: Koichiro Den Link: https://lore.kernel.org/r/20250203110123.87701-1-brgl@bgdev.pl Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-sim.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/gpio/gpio-sim.c b/drivers/gpio/gpio-sim.c index a086087ada17..b6c230fab840 100644 --- a/drivers/gpio/gpio-sim.c +++ b/drivers/gpio/gpio-sim.c @@ -1028,20 +1028,23 @@ gpio_sim_device_lockup_configfs(struct gpio_sim_device *dev, bool lock) struct configfs_subsystem *subsys = dev->group.cg_subsys; struct gpio_sim_bank *bank; struct gpio_sim_line *line; + struct config_item *item; /* - * The device only needs to depend on leaf line entries. This is + * The device only needs to depend on leaf entries. This is * sufficient to lock up all the configfs entries that the * instantiated, alive device depends on. */ list_for_each_entry(bank, &dev->bank_list, siblings) { list_for_each_entry(line, &bank->line_list, siblings) { + item = line->hog ? &line->hog->item + : &line->group.cg_item; + if (lock) - WARN_ON(configfs_depend_item_unlocked( - subsys, &line->group.cg_item)); + WARN_ON(configfs_depend_item_unlocked(subsys, + item)); else - configfs_undepend_item_unlocked( - &line->group.cg_item); + configfs_undepend_item_unlocked(item); } } } From 5393f40a640b8c4f716bf87e7b0d4328bf1f22b2 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 5 Feb 2025 14:05:03 +0100 Subject: [PATCH 0510/1443] gpio: GPIO_GRGPIO should depend on OF While the Aeroflex Gaisler GRGPIO driver has no build-time dependency on gpiolib-of, it supports only DT-based configuration, and is used only on DT systems. Hence add a dependency on OF, to prevent asking the user about this driver when configuring a kernel without DT support. Fixes: bc40668def384256 ("gpio: grgpio: drop Kconfig dependency on OF_GPIO") Signed-off-by: Geert Uytterhoeven Reviewed-by: Andreas Larsson Link: https://lore.kernel.org/r/db6da3d11bf850d89f199e5c740d8f133e38078d.1738760539.git.geert+renesas@glider.be Signed-off-by: Bartosz Golaszewski --- drivers/gpio/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig index add5ad29a673..98b4d1633b25 100644 --- a/drivers/gpio/Kconfig +++ b/drivers/gpio/Kconfig @@ -338,6 +338,7 @@ config GPIO_GRANITERAPIDS config GPIO_GRGPIO tristate "Aeroflex Gaisler GRGPIO support" + depends on OF || COMPILE_TEST select GPIO_GENERIC select IRQ_DOMAIN help From 59ff2040f0a58923c787fdba5999100667338230 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 4 Feb 2025 13:45:15 +0200 Subject: [PATCH 0511/1443] MAINTAINERS: Use my kernel.org address for ACPI GPIO work Switch to use my kernel.org address for ACPI GPIO work. Signed-off-by: Mika Westerberg Link: https://lore.kernel.org/r/20250204114515.3971923-1-mika.westerberg@linux.intel.com Signed-off-by: Bartosz Golaszewski --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 896a307fa065..d1389ca6699d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9878,7 +9878,7 @@ S: Maintained F: drivers/staging/gpib/ GPIO ACPI SUPPORT -M: Mika Westerberg +M: Mika Westerberg M: Andy Shevchenko L: linux-gpio@vger.kernel.org L: linux-acpi@vger.kernel.org From dac328dea701ed1fde2b67176fe3c3a818536441 Mon Sep 17 00:00:00 2001 From: Raag Jadav Date: Fri, 31 Jan 2025 11:15:02 +0530 Subject: [PATCH 0512/1443] drm/xe/hwmon: expose package and vram temperature Add hwmon support for temp2_input and temp3_input attributes, which will expose package and vram temperature in millidegree Celsius. With this in place we can monitor temperature using lm-sensors tool. v2: Reuse existing channels (Badal, Karthik) Signed-off-by: Raag Jadav Reviewed-by: Andi Shyti Reviewed-by: Badal Nilawar Link: https://patchwork.freedesktop.org/patch/msgid/20250131054502.1528555-1-raag.jadav@intel.com Signed-off-by: Rodrigo Vivi --- .../ABI/testing/sysfs-driver-intel-xe-hwmon | 16 +++++ drivers/gpu/drm/xe/regs/xe_mchbar_regs.h | 3 + drivers/gpu/drm/xe/regs/xe_pcode_regs.h | 2 + drivers/gpu/drm/xe/xe_hwmon.c | 60 +++++++++++++++++++ 4 files changed, 81 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon index d792a56f59ac..9bce281314df 100644 --- a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon +++ b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon @@ -108,3 +108,19 @@ Contact: intel-xe@lists.freedesktop.org Description: RO. Package current voltage in millivolt. Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/temp2_input +Date: March 2025 +KernelVersion: 6.14 +Contact: intel-xe@lists.freedesktop.org +Description: RO. Package temperature in millidegree Celsius. + + Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/temp3_input +Date: March 2025 +KernelVersion: 6.14 +Contact: intel-xe@lists.freedesktop.org +Description: RO. VRAM temperature in millidegree Celsius. + + Only supported for particular Intel Xe graphics platforms. diff --git a/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h b/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h index 519dd1067a19..f5e5234857c1 100644 --- a/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h @@ -34,6 +34,9 @@ #define PCU_CR_PACKAGE_ENERGY_STATUS XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x593c) +#define PCU_CR_PACKAGE_TEMPERATURE XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x5978) +#define TEMP_MASK REG_GENMASK(7, 0) + #define PCU_CR_PACKAGE_RAPL_LIMIT XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x59a0) #define PKG_PWR_LIM_1 REG_GENMASK(14, 0) #define PKG_PWR_LIM_1_EN REG_BIT(15) diff --git a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h index 0b0b49d850ae..8846eb9ce2a4 100644 --- a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h @@ -21,6 +21,8 @@ #define BMG_PACKAGE_POWER_SKU XE_REG(0x138098) #define BMG_PACKAGE_POWER_SKU_UNIT XE_REG(0x1380dc) #define BMG_PACKAGE_ENERGY_STATUS XE_REG(0x138120) +#define BMG_VRAM_TEMPERATURE XE_REG(0x1382c0) +#define BMG_PACKAGE_TEMPERATURE XE_REG(0x138434) #define BMG_PACKAGE_RAPL_LIMIT XE_REG(0x138440) #define BMG_PLATFORM_ENERGY_STATUS XE_REG(0x138458) #define BMG_PLATFORM_POWER_LIMIT XE_REG(0x138460) diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c index fde56dad3ab7..7f327e334212 100644 --- a/drivers/gpu/drm/xe/xe_hwmon.c +++ b/drivers/gpu/drm/xe/xe_hwmon.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include "regs/xe_gt_regs.h" @@ -20,6 +21,7 @@ #include "xe_pm.h" enum xe_hwmon_reg { + REG_TEMP, REG_PKG_RAPL_LIMIT, REG_PKG_POWER_SKU, REG_PKG_POWER_SKU_UNIT, @@ -36,6 +38,7 @@ enum xe_hwmon_reg_operation { enum xe_hwmon_channel { CHANNEL_CARD, CHANNEL_PKG, + CHANNEL_VRAM, CHANNEL_MAX, }; @@ -84,6 +87,19 @@ static struct xe_reg xe_hwmon_get_reg(struct xe_hwmon *hwmon, enum xe_hwmon_reg struct xe_device *xe = hwmon->xe; switch (hwmon_reg) { + case REG_TEMP: + if (xe->info.platform == XE_BATTLEMAGE) { + if (channel == CHANNEL_PKG) + return BMG_PACKAGE_TEMPERATURE; + else if (channel == CHANNEL_VRAM) + return BMG_VRAM_TEMPERATURE; + } else if (xe->info.platform == XE_DG2) { + if (channel == CHANNEL_PKG) + return PCU_CR_PACKAGE_TEMPERATURE; + else if (channel == CHANNEL_VRAM) + return BMG_VRAM_TEMPERATURE; + } + break; case REG_PKG_RAPL_LIMIT: if (xe->info.platform == XE_BATTLEMAGE) { if (channel == CHANNEL_PKG) @@ -431,6 +447,8 @@ static const struct attribute_group *hwmon_groups[] = { }; static const struct hwmon_channel_info * const hwmon_info[] = { + HWMON_CHANNEL_INFO(temp, HWMON_T_LABEL, HWMON_T_INPUT | HWMON_T_LABEL, + HWMON_T_INPUT | HWMON_T_LABEL), HWMON_CHANNEL_INFO(power, HWMON_P_MAX | HWMON_P_RATED_MAX | HWMON_P_LABEL, HWMON_P_MAX | HWMON_P_RATED_MAX | HWMON_P_CRIT | HWMON_P_LABEL), HWMON_CHANNEL_INFO(curr, HWMON_C_LABEL, HWMON_C_CRIT | HWMON_C_LABEL), @@ -506,6 +524,36 @@ static void xe_hwmon_get_voltage(struct xe_hwmon *hwmon, int channel, long *valu *value = DIV_ROUND_CLOSEST(REG_FIELD_GET(VOLTAGE_MASK, reg_val) * 2500, SF_VOLTAGE); } +static umode_t +xe_hwmon_temp_is_visible(struct xe_hwmon *hwmon, u32 attr, int channel) +{ + switch (attr) { + case hwmon_temp_input: + case hwmon_temp_label: + return xe_reg_is_valid(xe_hwmon_get_reg(hwmon, REG_TEMP, channel)) ? 0444 : 0; + default: + return 0; + } +} + +static int +xe_hwmon_temp_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *val) +{ + struct xe_mmio *mmio = xe_root_tile_mmio(hwmon->xe); + u64 reg_val; + + switch (attr) { + case hwmon_temp_input: + reg_val = xe_mmio_read32(mmio, xe_hwmon_get_reg(hwmon, REG_TEMP, channel)); + + /* HW register value is in degrees Celsius, convert to millidegrees. */ + *val = REG_FIELD_GET(TEMP_MASK, reg_val) * MILLIDEGREE_PER_DEGREE; + return 0; + default: + return -EOPNOTSUPP; + } +} + static umode_t xe_hwmon_power_is_visible(struct xe_hwmon *hwmon, u32 attr, int channel) { @@ -667,6 +715,9 @@ xe_hwmon_is_visible(const void *drvdata, enum hwmon_sensor_types type, xe_pm_runtime_get(hwmon->xe); switch (type) { + case hwmon_temp: + ret = xe_hwmon_temp_is_visible(hwmon, attr, channel); + break; case hwmon_power: ret = xe_hwmon_power_is_visible(hwmon, attr, channel); break; @@ -699,6 +750,9 @@ xe_hwmon_read(struct device *dev, enum hwmon_sensor_types type, u32 attr, xe_pm_runtime_get(hwmon->xe); switch (type) { + case hwmon_temp: + ret = xe_hwmon_temp_read(hwmon, attr, channel, val); + break; case hwmon_power: ret = xe_hwmon_power_read(hwmon, attr, channel, val); break; @@ -752,6 +806,12 @@ static int xe_hwmon_read_label(struct device *dev, u32 attr, int channel, const char **str) { switch (type) { + case hwmon_temp: + if (channel == CHANNEL_PKG) + *str = "pkg"; + else if (channel == CHANNEL_VRAM) + *str = "vram"; + return 0; case hwmon_power: case hwmon_energy: case hwmon_curr: From d8d99c3b5c485f339864aeaa29f76269cc0ea975 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Wed, 5 Feb 2025 15:52:31 +0200 Subject: [PATCH 0513/1443] ASoC: SOF: stream-ipc: Check for cstream nullity in sof_ipc_msg_data() The nullity of sps->cstream should be checked similarly as it is done in sof_set_stream_data_offset() function. Assuming that it is not NULL if sps->stream is NULL is incorrect and can lead to NULL pointer dereference. Fixes: 090349a9feba ("ASoC: SOF: Add support for compress API for stream data/offset") Cc: stable@vger.kernel.org Reported-by: Curtis Malainey Closes: https://github.com/thesofproject/linux/pull/5214 Signed-off-by: Peter Ujfalusi Reviewed-by: Daniel Baluta Reviewed-by: Ranjani Sridharan Reviewed-by: Bard Liao Reviewed-by: Curtis Malainey Link: https://patch.msgid.link/20250205135232.19762-2-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/sof/stream-ipc.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/sound/soc/sof/stream-ipc.c b/sound/soc/sof/stream-ipc.c index 794c7bbccbaf..8262443ac89a 100644 --- a/sound/soc/sof/stream-ipc.c +++ b/sound/soc/sof/stream-ipc.c @@ -43,7 +43,7 @@ int sof_ipc_msg_data(struct snd_sof_dev *sdev, return -ESTRPIPE; posn_offset = stream->posn_offset; - } else { + } else if (sps->cstream) { struct sof_compr_stream *sstream = sps->cstream->runtime->private_data; @@ -51,6 +51,10 @@ int sof_ipc_msg_data(struct snd_sof_dev *sdev, return -ESTRPIPE; posn_offset = sstream->posn_offset; + + } else { + dev_err(sdev->dev, "%s: No stream opened\n", __func__); + return -EINVAL; } snd_sof_dsp_mailbox_read(sdev, posn_offset, p, sz); From 46c7b901e2a03536df5a3cb40b3b26e2be505df6 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Wed, 5 Feb 2025 15:52:32 +0200 Subject: [PATCH 0514/1443] ASoC: SOF: pcm: Clear the susbstream pointer to NULL on close The spcm->stream[substream->stream].substream is set during open and was left untouched. After the first PCM stream it will never be NULL and we have code which checks for substream NULLity as indication if the stream is active or not. For the compressed cstream pointer the same has been done, this change will correct the handling of PCM streams. Fixes: 090349a9feba ("ASoC: SOF: Add support for compress API for stream data/offset") Cc: stable@vger.kernel.org Reported-by: Curtis Malainey Closes: https://github.com/thesofproject/linux/pull/5214 Signed-off-by: Peter Ujfalusi Reviewed-by: Daniel Baluta Reviewed-by: Ranjani Sridharan Reviewed-by: Bard Liao Reviewed-by: Curtis Malainey Link: https://patch.msgid.link/20250205135232.19762-3-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/sof/pcm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/soc/sof/pcm.c b/sound/soc/sof/pcm.c index 35a7462d8b69..c5c6353f18ce 100644 --- a/sound/soc/sof/pcm.c +++ b/sound/soc/sof/pcm.c @@ -511,6 +511,8 @@ static int sof_pcm_close(struct snd_soc_component *component, */ } + spcm->stream[substream->stream].substream = NULL; + return 0; } From 738fc998b639407346a9e026514f0562301462cd Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Fri, 31 Jan 2025 15:55:28 -0700 Subject: [PATCH 0515/1443] scripts/Makefile.extrawarn: Do not show clang's non-kprintf warnings at W=1 Clang's -Wformat-overflow and -Wformat-truncation have chosen to check '%p' unlike GCC but it does not know about the kernel's pointer extensions in lib/vsprintf.c, so the developers split that part of the warning out for the kernel to disable because there will always be false positives. Commit 908dd508276d ("kbuild: enable -Wformat-truncation on clang") did disabled these warnings but only in a block that would be called when W=1 was not passed, so they would appear with W=1. Move the disabling of the non-kprintf warnings to a block that always runs so that they are never seen, regardless of warning level. Fixes: 908dd508276d ("kbuild: enable -Wformat-truncation on clang") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501291646.VtwF98qd-lkp@intel.com/ Signed-off-by: Nathan Chancellor Signed-off-by: Masahiro Yamada --- scripts/Makefile.extrawarn | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn index eb719f6d8d53..e976790f84dc 100644 --- a/scripts/Makefile.extrawarn +++ b/scripts/Makefile.extrawarn @@ -31,6 +31,11 @@ KBUILD_CFLAGS-$(CONFIG_CC_NO_ARRAY_BOUNDS) += -Wno-array-bounds ifdef CONFIG_CC_IS_CLANG # The kernel builds with '-std=gnu11' so use of GNU extensions is acceptable. KBUILD_CFLAGS += -Wno-gnu + +# Clang checks for overflow/truncation with '%p', while GCC does not: +# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111219 +KBUILD_CFLAGS += $(call cc-disable-warning, format-overflow-non-kprintf) +KBUILD_CFLAGS += $(call cc-disable-warning, format-truncation-non-kprintf) else # gcc inanely warns about local variables called 'main' @@ -105,11 +110,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, packed-not-aligned) KBUILD_CFLAGS += $(call cc-disable-warning, format-overflow) ifdef CONFIG_CC_IS_GCC KBUILD_CFLAGS += $(call cc-disable-warning, format-truncation) -else -# Clang checks for overflow/truncation with '%p', while GCC does not: -# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111219 -KBUILD_CFLAGS += $(call cc-disable-warning, format-overflow-non-kprintf) -KBUILD_CFLAGS += $(call cc-disable-warning, format-truncation-non-kprintf) endif KBUILD_CFLAGS += $(call cc-disable-warning, stringop-truncation) From 4c56eb33e603c3b9eb4bd24efbfdd0283c1c37e4 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Sun, 2 Feb 2025 03:51:41 +0900 Subject: [PATCH 0516/1443] kbuild: keep symbols for symbol_get() even with CONFIG_TRIM_UNUSED_KSYMS Linus observed that the symbol_request(utf8_data_table) call fails when CONFIG_UNICODE=y and CONFIG_TRIM_UNUSED_KSYMS=y. symbol_get() relies on the symbol data being present in the ksymtab for symbol lookups. However, EXPORT_SYMBOL_GPL(utf8_data_table) is dropped due to CONFIG_TRIM_UNUSED_KSYMS, as no module references it in this case. Probably, this has been broken since commit dbacb0ef670d ("kconfig option for TRIM_UNUSED_KSYMS"). This commit addresses the issue by leveraging modpost. Symbol names passed to symbol_get() are recorded in the special .no_trim_symbol section, which is then parsed by modpost to forcibly keep such symbols. The .no_trim_symbol section is discarded by the linker scripts, so there is no impact on the size of the final vmlinux or modules. This commit cannot resolve the issue for direct calls to __symbol_get() because the symbol name is not known at compile-time. Although symbol_get() may eventually be deprecated, this workaround should be good enough meanwhile. Reported-by: Linus Torvalds Suggested-by: Linus Torvalds Signed-off-by: Masahiro Yamada --- include/asm-generic/vmlinux.lds.h | 1 + include/linux/module.h | 5 ++++- scripts/mod/modpost.c | 35 +++++++++++++++++++++++++++++++ scripts/mod/modpost.h | 6 ++++++ scripts/module.lds.S | 1 + 5 files changed, 47 insertions(+), 1 deletion(-) diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 54504013c749..02a4adb4a999 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -1038,6 +1038,7 @@ defined(CONFIG_AUTOFDO_CLANG) || defined(CONFIG_PROPELLER_CLANG) *(.discard) \ *(.discard.*) \ *(.export_symbol) \ + *(.no_trim_symbol) \ *(.modinfo) \ /* ld.bfd warns about .gnu.version* even when not emitted */ \ *(.gnu.version*) \ diff --git a/include/linux/module.h b/include/linux/module.h index 23792d5d7b74..30e5b19bafa9 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -306,7 +306,10 @@ extern int modules_disabled; /* for sysctl */ /* Get/put a kernel symbol (calls must be symmetric) */ void *__symbol_get(const char *symbol); void *__symbol_get_gpl(const char *symbol); -#define symbol_get(x) ((typeof(&x))(__symbol_get(__stringify(x)))) +#define symbol_get(x) ({ \ + static const char __notrim[] \ + __used __section(".no_trim_symbol") = __stringify(x); \ + (typeof(&x))(__symbol_get(__stringify(x))); }) /* modules using other modules: kdb wants to see this. */ struct module_use { diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index e18ae7dc8140..36b28987a2f0 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -507,6 +507,9 @@ static int parse_elf(struct elf_info *info, const char *filename) info->modinfo_len = sechdrs[i].sh_size; } else if (!strcmp(secname, ".export_symbol")) { info->export_symbol_secndx = i; + } else if (!strcmp(secname, ".no_trim_symbol")) { + info->no_trim_symbol = (void *)hdr + sechdrs[i].sh_offset; + info->no_trim_symbol_len = sechdrs[i].sh_size; } if (sechdrs[i].sh_type == SHT_SYMTAB) { @@ -1566,6 +1569,14 @@ static void read_symbols(const char *modname) /* strip trailing .o */ mod = new_module(modname, strlen(modname) - strlen(".o")); + /* save .no_trim_symbol section for later use */ + if (info.no_trim_symbol_len) { + mod->no_trim_symbol = xmalloc(info.no_trim_symbol_len); + memcpy(mod->no_trim_symbol, info.no_trim_symbol, + info.no_trim_symbol_len); + mod->no_trim_symbol_len = info.no_trim_symbol_len; + } + if (!mod->is_vmlinux) { license = get_modinfo(&info, "license"); if (!license) @@ -1728,6 +1739,28 @@ static void handle_white_list_exports(const char *white_list) free(buf); } +/* + * Keep symbols recorded in the .no_trim_symbol section. This is necessary to + * prevent CONFIG_TRIM_UNUSED_KSYMS from dropping EXPORT_SYMBOL because + * symbol_get() relies on the symbol being present in the ksymtab for lookups. + */ +static void keep_no_trim_symbols(struct module *mod) +{ + unsigned long size = mod->no_trim_symbol_len; + + for (char *s = mod->no_trim_symbol; s; s = next_string(s , &size)) { + struct symbol *sym; + + /* + * If find_symbol() returns NULL, this symbol is not provided + * by any module, and symbol_get() will fail. + */ + sym = find_symbol(s); + if (sym) + sym->used = true; + } +} + static void check_modname_len(struct module *mod) { const char *mod_name; @@ -2254,6 +2287,8 @@ int main(int argc, char **argv) read_symbols_from_files(files_source); list_for_each_entry(mod, &modules, list) { + keep_no_trim_symbols(mod); + if (mod->dump_file || mod->is_vmlinux) continue; diff --git a/scripts/mod/modpost.h b/scripts/mod/modpost.h index ffd0a52a606e..59366f456b76 100644 --- a/scripts/mod/modpost.h +++ b/scripts/mod/modpost.h @@ -111,6 +111,8 @@ struct module_alias { * * @dump_file: path to the .symvers file if loaded from a file * @aliases: list head for module_aliases + * @no_trim_symbol: .no_trim_symbol section data + * @no_trim_symbol_len: length of the .no_trim_symbol section */ struct module { struct list_head list; @@ -128,6 +130,8 @@ struct module { // Actual imported namespaces struct list_head imported_namespaces; struct list_head aliases; + char *no_trim_symbol; + unsigned int no_trim_symbol_len; char name[]; }; @@ -141,6 +145,8 @@ struct elf_info { char *strtab; char *modinfo; unsigned int modinfo_len; + char *no_trim_symbol; + unsigned int no_trim_symbol_len; /* support for 32bit section numbers */ diff --git a/scripts/module.lds.S b/scripts/module.lds.S index c2f80f9141d4..450f1088d5fd 100644 --- a/scripts/module.lds.S +++ b/scripts/module.lds.S @@ -16,6 +16,7 @@ SECTIONS { *(.discard) *(.discard.*) *(.export_symbol) + *(.no_trim_symbol) } __ksymtab 0 : ALIGN(8) { *(SORT(___ksymtab+*)) } From 98a5cfd2320966f40fe049a9855f8787f0126825 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Wed, 5 Feb 2025 09:43:31 +0100 Subject: [PATCH 0517/1443] x86/xen: fix xen_hypercall_hvm() to not clobber %rbx xen_hypercall_hvm(), which is used when running as a Xen PVH guest at most only once during early boot, is clobbering %rbx. Depending on whether the caller relies on %rbx to be preserved across the call or not, this clobbering might result in an early crash of the system. This can be avoided by using an already saved register instead of %rbx. Fixes: b4845bb63838 ("x86/xen: add central hypercall functions") Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich Reviewed-by: Andrew Cooper Signed-off-by: Juergen Gross --- arch/x86/xen/xen-head.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index 9252652afe59..72f28d66e0e5 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -117,8 +117,8 @@ SYM_FUNC_START(xen_hypercall_hvm) pop %ebx pop %eax #else - lea xen_hypercall_amd(%rip), %rbx - cmp %rax, %rbx + lea xen_hypercall_amd(%rip), %rcx + cmp %rax, %rcx #ifdef CONFIG_FRAME_POINTER pop %rax /* Dummy pop. */ #endif From 0bd797b801bd8ee06c822844e20d73aaea0878dd Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Wed, 5 Feb 2025 10:07:56 +0100 Subject: [PATCH 0518/1443] x86/xen: add FRAME_END to xen_hypercall_hvm() xen_hypercall_hvm() is missing a FRAME_END at the end, add it. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202502030848.HTNTTuo9-lkp@intel.com/ Fixes: b4845bb63838 ("x86/xen: add central hypercall functions") Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich Reviewed-by: Andrew Cooper Signed-off-by: Juergen Gross --- arch/x86/xen/xen-head.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index 72f28d66e0e5..4e481b0eefc9 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -132,6 +132,7 @@ SYM_FUNC_START(xen_hypercall_hvm) pop %rcx pop %rax #endif + FRAME_END /* Use correct hypercall function. */ jz xen_hypercall_amd jmp xen_hypercall_intel From aaf5eefd374b6e006e1c224a2b37bac9d3737aa2 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Wed, 5 Feb 2025 11:24:47 +0100 Subject: [PATCH 0519/1443] x86/xen: remove unneeded dummy push from xen_hypercall_hvm() Stack alignment of the kernel in 64-bit mode is 8, not 16, so the dummy push in xen_hypercall_hvm() for aligning the stack to 16 bytes can be removed. Signed-off-by: Juergen Gross Reviewed-by: Jan Beulich Reviewed-by: Andrew Cooper Signed-off-by: Juergen Gross --- arch/x86/xen/xen-head.S | 6 ------ 1 file changed, 6 deletions(-) diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index 4e481b0eefc9..894edf8d6d62 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -100,9 +100,6 @@ SYM_FUNC_START(xen_hypercall_hvm) push %r10 push %r9 push %r8 -#ifdef CONFIG_FRAME_POINTER - pushq $0 /* Dummy push for stack alignment. */ -#endif #endif /* Set the vendor specific function. */ call __xen_hypercall_setfunc @@ -119,9 +116,6 @@ SYM_FUNC_START(xen_hypercall_hvm) #else lea xen_hypercall_amd(%rip), %rcx cmp %rax, %rcx -#ifdef CONFIG_FRAME_POINTER - pop %rax /* Dummy pop. */ -#endif pop %r8 pop %r9 pop %r10 From d364eee14c682b141f4667efc3c65191339d88bd Mon Sep 17 00:00:00 2001 From: Dhananjay Ugwekar Date: Wed, 5 Feb 2025 11:25:12 +0000 Subject: [PATCH 0520/1443] cpufreq/amd-pstate: Remove the goto label in amd_pstate_update_limits Scope based guard/cleanup macros should not be used together with goto labels. Hence, remove the goto label. Fixes: 6c093d5a5b73 ("cpufreq/amd-pstate: convert mutex use to guard()") Signed-off-by: Dhananjay Ugwekar Reviewed-by: Mario Limonciello Link: https://lore.kernel.org/r/20250205112523.201101-2-dhananjay.ugwekar@amd.com Signed-off-by: Mario Limonciello --- drivers/cpufreq/amd-pstate.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 7120f035c0be..b163c1699821 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -838,8 +838,10 @@ static void amd_pstate_update_limits(unsigned int cpu) guard(mutex)(&amd_pstate_driver_lock); ret = amd_get_highest_perf(cpu, &cur_high); - if (ret) - goto free_cpufreq_put; + if (ret) { + cpufreq_cpu_put(policy); + return; + } prev_high = READ_ONCE(cpudata->prefcore_ranking); highest_perf_changed = (prev_high != cur_high); @@ -849,8 +851,6 @@ static void amd_pstate_update_limits(unsigned int cpu) if (cur_high < CPPC_MAX_PERF) sched_set_itmt_core_prio((int)cur_high, cpu); } - -free_cpufreq_put: cpufreq_cpu_put(policy); if (!highest_perf_changed) From 55db9b73c3a77544efc671d5e796d9674772c330 Mon Sep 17 00:00:00 2001 From: Dhananjay Ugwekar Date: Wed, 5 Feb 2025 11:25:13 +0000 Subject: [PATCH 0521/1443] cpufreq/amd-pstate: Fix max_perf updation with schedutil In adjust_perf() callback, we are setting the max_perf to highest_perf, as opposed to the correct limit value i.e. max_limit_perf. Fix that. Fixes: 3f7b835fa4d0 ("cpufreq/amd-pstate: Move limit updating code") Signed-off-by: Dhananjay Ugwekar Reviewed-by: Mario Limonciello Link: https://lore.kernel.org/r/20250205112523.201101-3-dhananjay.ugwekar@amd.com Signed-off-by: Mario Limonciello --- drivers/cpufreq/amd-pstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index b163c1699821..9dc3933bc326 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -699,7 +699,7 @@ static void amd_pstate_adjust_perf(unsigned int cpu, if (min_perf < lowest_nonlinear_perf) min_perf = lowest_nonlinear_perf; - max_perf = cap_perf; + max_perf = cpudata->max_limit_perf; if (max_perf < min_perf) max_perf = min_perf; From 038e33fcd40e59b60cdca561c2a39998e6759e08 Mon Sep 17 00:00:00 2001 From: Lukasz Majewski Date: Thu, 9 Jan 2025 16:41:49 +0100 Subject: [PATCH 0522/1443] dt-bindings: display: Add powertip,{st7272|hx8238a} as DT Schema description This patch provides the DT Schema description of: - powertip,st7272 320 x 240 LCD display - powertip,hx8238a 320 x 240 LCD display Used with the different HW revisions of btt3 devices. Signed-off-by: Lukasz Majewski Reviewed-by: Rob Herring Link: https://lore.kernel.org/r/20250109154149.1212631-1-lukma@denx.de Signed-off-by: Rob Herring (Arm) --- .../display/panel/powertip,hx8238a.yaml | 29 +++++++++++++++++++ .../display/panel/powertip,st7272.yaml | 29 +++++++++++++++++++ 2 files changed, 58 insertions(+) create mode 100644 Documentation/devicetree/bindings/display/panel/powertip,hx8238a.yaml create mode 100644 Documentation/devicetree/bindings/display/panel/powertip,st7272.yaml diff --git a/Documentation/devicetree/bindings/display/panel/powertip,hx8238a.yaml b/Documentation/devicetree/bindings/display/panel/powertip,hx8238a.yaml new file mode 100644 index 000000000000..b7d74faeb5d5 --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/powertip,hx8238a.yaml @@ -0,0 +1,29 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/powertip,hx8238a.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Powertip Electronic Technology Co. 320 x 240 LCD panel + +maintainers: + - Lukasz Majewski + +allOf: + - $ref: panel-dpi.yaml# + +properties: + compatible: + items: + - const: powertip,hx8238a + - {} # panel-dpi, but not listed here to avoid false select + + height-mm: true + panel-timing: true + port: true + power-supply: true + width-mm: true + +additionalProperties: false + +... diff --git a/Documentation/devicetree/bindings/display/panel/powertip,st7272.yaml b/Documentation/devicetree/bindings/display/panel/powertip,st7272.yaml new file mode 100644 index 000000000000..f3622800f13f --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/powertip,st7272.yaml @@ -0,0 +1,29 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/powertip,st7272.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Powertip Electronic Technology Co. 320 x 240 LCD panel + +maintainers: + - Lukasz Majewski + +allOf: + - $ref: panel-dpi.yaml# + +properties: + compatible: + items: + - const: powertip,st7272 + - {} # panel-dpi, but not listed here to avoid false select + + height-mm: true + panel-timing: true + port: true + power-supply: true + width-mm: true + +additionalProperties: false + +... From 069504f1fcfa1532e4e221290df428b15bd9d284 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Tue, 4 Feb 2025 17:49:25 +0200 Subject: [PATCH 0523/1443] drm/i915/dp: Fix potential infinite loop in 128b/132b SST Passing 0 as the step only works when there are other reasons to break out of the BPP loop in intel_dp_mtp_tu_compute_config(). Otherwise, an infinite loop might occur. Fix it by explicitly checking for 0 step. Fixes: ef0a0757bbea ("drm/i915/dp: compute config for 128b/132b SST w/o DSC") Reported-by: Imre Deak Closes: https://lore.kernel.org/r/Z6I0knh2Kt5T0JrT@ideak-desk.fi.intel.com Reviewed-by: Imre Deak Link: https://patchwork.freedesktop.org/patch/msgid/20250204154925.3001781-1-jani.nikula@intel.com Signed-off-by: Jani Nikula (cherry picked from commit a40e718d34d3d02c781c295466b013415f68c4f1) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_dp_mst.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 0c44fc7dd86c..a65cf97ad12d 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -341,6 +341,10 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, break; } + + /* Allow using zero step to indicate one try */ + if (!step) + break; } if (slots < 0) { From 90508a1bb8f00618fa12cb2ad2276bc783656fc5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 3 Feb 2025 16:24:15 +0530 Subject: [PATCH 0524/1443] cpufreq: airoha: modify CONFIG_OF dependency Compile-testing without CONFIG_OF leads to a harmless build warning: drivers/cpufreq/airoha-cpufreq.c:109:34: error: 'airoha_cpufreq_match_list' defined but not used [-Werror=unused-const-variable=] 109 | static const struct of_device_id airoha_cpufreq_match_list[] __initconst = { | ^~~~~~~~~~~~~~~~~~~~~~~~~ It would be possible to mark the variable as __maybe_unused to shut up that warning, but a Kconfig dependency seems more appropriate as this still allows build testing in allmodconfig and randconfig builds on all architectures. An earlier commit, b865a8404642 ("cpufreq: airoha: Depends on OF"), tried to fix it incorrectly. ARCH_AIROHA already requires CONFIG_OF, so this change does nothing, and the dependency is still missing for the COMPILE_TEST case. Fix it properly. Fixes: 84cf9e541ccc ("cpufreq: airoha: Add EN7581 CPUFreq SMCCC driver") Fixes: b865a8404642 ("cpufreq: airoha: Depends on OF") Signed-off-by: Arnd Bergmann [ Viresh: updated commit log and fixed rebase conflict ] Signed-off-by: Viresh Kumar Link: https://patch.msgid.link/9d51d2710061dfa7f2568287c6ed125b858b7318.1738580005.git.viresh.kumar@linaro.org Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/Kconfig.arm | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm index 0ee5c691fb36..9e46960f6a86 100644 --- a/drivers/cpufreq/Kconfig.arm +++ b/drivers/cpufreq/Kconfig.arm @@ -17,7 +17,8 @@ config ARM_ALLWINNER_SUN50I_CPUFREQ_NVMEM config ARM_AIROHA_SOC_CPUFREQ tristate "Airoha EN7581 SoC CPUFreq support" - depends on (ARCH_AIROHA && OF) || COMPILE_TEST + depends on ARCH_AIROHA || COMPILE_TEST + depends on OF select PM_OPP default ARCH_AIROHA help From 0813fd2e14ca6ecd4e6ba005a9766f08e26020d7 Mon Sep 17 00:00:00 2001 From: Aboorva Devarajan Date: Wed, 5 Feb 2025 23:43:47 +0530 Subject: [PATCH 0525/1443] cpufreq: prevent NULL dereference in cpufreq_online() Ensure cpufreq_driver->set_boost is non-NULL before using it in cpufreq_online() to prevent a potential NULL pointer dereference. Reported-by: Gautam Menghani Closes: https://lore.kernel.org/all/c9e56c5f54cc33338762c94e9bed7b5a0d5de812.camel@linux.ibm.com/ Fixes: dd016f379ebc ("cpufreq: Introduce a more generic way to set default per-policy boost flag") Suggested-by: Viresh Kumar Signed-off-by: Aboorva Devarajan Link: https://patch.msgid.link/20250205181347.2079272-1-aboorvad@linux.ibm.com [ rjw: Minor edits in the subject and changelog ] Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/cpufreq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index e0048856ecee..30ffbddc7ece 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1571,7 +1571,8 @@ static int cpufreq_online(unsigned int cpu) policy->cdev = of_cpufreq_cooling_register(policy); /* Let the per-policy boost flag mirror the cpufreq_driver boost during init */ - if (policy->boost_enabled != cpufreq_boost_enabled()) { + if (cpufreq_driver->set_boost && + policy->boost_enabled != cpufreq_boost_enabled()) { policy->boost_enabled = cpufreq_boost_enabled(); ret = cpufreq_driver->set_boost(policy, policy->boost_enabled); if (ret) { From ab930483eca9f3e816c35824b5868599af0c61d7 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 3 Feb 2025 21:46:29 +0200 Subject: [PATCH 0526/1443] ACPI: property: Fix return value for nval == 0 in acpi_data_prop_read() While analysing code for software and OF node for the corner case when caller asks to read zero items in the supposed to be an array of values I found that ACPI behaves differently to what OF does, i.e. 1. It returns -EINVAL when caller asks to read zero items from integer array, while OF returns 0, if no other errors happened. 2. It returns -EINVAL when caller asks to read zero items from string array, while OF returns -ENODATA, if no other errors happened. Amend ACPI implementation to follow what OF does. Fixes: b31384fa5de3 ("Driver core: Unified device properties interface for platform firmware") Signed-off-by: Andy Shevchenko Link: https://patch.msgid.link/20250203194629.3731895-1-andriy.shevchenko@linux.intel.com [ rjw: Added empty line after a conditional ] Signed-off-by: Rafael J. Wysocki --- drivers/acpi/property.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/acpi/property.c b/drivers/acpi/property.c index 98d93ed58315..436019d96027 100644 --- a/drivers/acpi/property.c +++ b/drivers/acpi/property.c @@ -1187,8 +1187,6 @@ static int acpi_data_prop_read(const struct acpi_device_data *data, } break; } - if (nval == 0) - return -EINVAL; if (obj->type == ACPI_TYPE_BUFFER) { if (proptype != DEV_PROP_U8) @@ -1212,9 +1210,11 @@ static int acpi_data_prop_read(const struct acpi_device_data *data, ret = acpi_copy_property_array_uint(items, (u64 *)val, nval); break; case DEV_PROP_STRING: - ret = acpi_copy_property_array_string( - items, (char **)val, - min_t(u32, nval, obj->package.count)); + nval = min_t(u32, nval, obj->package.count); + if (nval == 0) + return -ENODATA; + + ret = acpi_copy_property_array_string(items, (char **)val, nval); break; default: ret = -EINVAL; From 607ab6f85f4194b644ea95ac5fe660ef575db3b4 Mon Sep 17 00:00:00 2001 From: Gannon Kolding Date: Mon, 27 Jan 2025 02:39:02 -0700 Subject: [PATCH 0527/1443] ACPI: resource: IRQ override for Eluktronics MECH-17 The Eluktronics MECH-17 (GM7RG7N) needs IRQ overriding for the keyboard to work. Adding a DMI_MATCH entry for this laptop model makes the internal keyboard function normally. Signed-off-by: Gannon Kolding Link: https://patch.msgid.link/20250127093902.328361-1-gannon.kolding@gmail.com Signed-off-by: Rafael J. Wysocki --- drivers/acpi/resource.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 90aaec923889..b4cd14e7fa76 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -563,6 +563,12 @@ static const struct dmi_system_id irq1_edge_low_force_override[] = { DMI_MATCH(DMI_BOARD_NAME, "RP-15"), }, }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Eluktronics Inc."), + DMI_MATCH(DMI_BOARD_NAME, "MECH-17"), + }, + }, { /* TongFang GM6XGxX/TUXEDO Stellaris 16 Gen5 AMD */ .matches = { From 7f5704b6a143b8eca640cba820968e798d065e91 Mon Sep 17 00:00:00 2001 From: Aubrey Li Date: Sun, 26 Jan 2025 10:22:50 +0800 Subject: [PATCH 0528/1443] ACPI: PRM: Remove unnecessary strict handler address checks Commit 088984c8d54c ("ACPI: PRM: Find EFI_MEMORY_RUNTIME block for PRM handler and context") added unnecessary strict handler address checks, causing the PRM module to fail in translating memory error addresses. Both static data buffer address and ACPI parameter buffer address may be NULL if they are not needed, as described in section 4.1.2 PRM Handler Information Structure of Platform Runtime Mechanism specification [1]. Here are two examples from real hardware: ----PRMT.dsl---- - staic data address is not used [10Ch 0268 2] Revision : 0000 [10Eh 0270 2] Length : 002C [110h 0272 16] Handler GUID : F6A58D47-E04F-4F5A-86B8-2A50D4AA109B [120h 0288 8] Handler address : 0000000065CE51F4 [128h 0296 8] Satic Data Address : 0000000000000000 [130h 0304 8] ACPI Parameter Address : 000000006522A718 - ACPI parameter address is not used [1B0h 0432 2] Revision : 0000 [1B2h 0434 2] Length : 002C [1B4h 0436 16] Handler GUID : 657E8AE6-A8FC-4877-BB28-42E7DE1899A5 [1C4h 0452 8] Handler address : 0000000065C567C8 [1CCh 0460 8] Satic Data Address : 000000006113FB98 [1D4h 0468 8] ACPI Parameter Address : 0000000000000000 Fixes: 088984c8d54c ("ACPI: PRM: Find EFI_MEMORY_RUNTIME block for PRM handler and context") Reported-and-tested-by: Shi Liu Cc: All applicable Signed-off-by: Aubrey Li Link: https://uefi.org/sites/default/files/resources/Platform%20Runtime%20Mechanism%20-%20with%20legal%20notice.pdf # [1] Reviewed-by: Koba Ko Acked-by: Ard Biesheuvel Link: https://patch.msgid.link/20250126022250.3014210-1-aubrey.li@linux.intel.com [ rjw: Minor changelog edits ] Signed-off-by: Rafael J. Wysocki --- drivers/acpi/prmt.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/acpi/prmt.c b/drivers/acpi/prmt.c index 747f83f7114d..e549914a636c 100644 --- a/drivers/acpi/prmt.c +++ b/drivers/acpi/prmt.c @@ -287,9 +287,7 @@ static acpi_status acpi_platformrt_space_handler(u32 function, if (!handler || !module) goto invalid_guid; - if (!handler->handler_addr || - !handler->static_data_buffer_addr || - !handler->acpi_param_buffer_addr) { + if (!handler->handler_addr) { buffer->prm_status = PRM_HANDLER_ERROR; return AE_OK; } From 0e1daf2b32dc85559cb1bb67e91c5a42db332cba Mon Sep 17 00:00:00 2001 From: Damon Ding Date: Wed, 5 Feb 2025 17:47:34 +0800 Subject: [PATCH 0529/1443] drm/edp-panel: Add LG Display panel model LP079QX1-SP0V The raw edid for LP079QX1-SP0V panel model is: 00 ff ff ff ff ff ff 00 16 83 00 00 00 00 00 00 04 17 01 00 a5 10 0c 78 06 ef 05 a3 54 4c 99 26 0f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 01 ea 4e 00 4c 60 00 14 80 0c 10 84 00 78 a0 00 00 00 18 00 00 00 10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fe 00 4c 50 30 37 39 51 58 31 2d 53 50 30 56 00 00 00 fc 00 43 6f 6c 6f 72 20 4c 43 44 0a 20 20 20 00 3f Signed-off-by: Damon Ding Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250205094734.562992-1-damon.ding@rock-chips.com --- drivers/gpu/drm/panel/panel-edp.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 25bb7e454fc9..13cca7ebf0f7 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1814,6 +1814,14 @@ static const struct panel_delay delay_200_500_e250 = { .enable = 250, }; +static const struct panel_delay delay_50_500_e200_d200_po2e335 = { + .hpd_absent = 50, + .unprepare = 500, + .enable = 200, + .disable = 200, + .powered_on_to_enable = 335, +}; + #define EDP_PANEL_ENTRY(vend_chr_0, vend_chr_1, vend_chr_2, product_id, _delay, _name) \ { \ .ident = { \ @@ -1965,6 +1973,8 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50, "MNB601LS1-4"), EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"), + EDP_PANEL_ENTRY('E', 'T', 'C', 0x0000, &delay_50_500_e200_d200_po2e335, "LP079QX1-SP0V"), + EDP_PANEL_ENTRY('H', 'K', 'C', 0x2d51, &delay_200_500_e200, "Unknown"), EDP_PANEL_ENTRY('H', 'K', 'C', 0x2d5b, &delay_200_500_e200, "MB116AN01"), EDP_PANEL_ENTRY('H', 'K', 'C', 0x2d5c, &delay_200_500_e200, "MB116AN01-2"), From 7e3bf00047cdfe2b09e9a0d77b99bedc15406bfa Mon Sep 17 00:00:00 2001 From: Tejas Vipin Date: Sun, 2 Feb 2025 00:24:39 +0530 Subject: [PATCH 0530/1443] drm/panel: sharp-ls060t1sx01: transition to mipi_dsi wrapped functions Changes the sharp-ls060t1sx01 panel to use multi style functions for improved error handling. Signed-off-by: Tejas Vipin Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250201185439.157613-1-tejasvipin76@gmail.com --- .../gpu/drm/panel/panel-sharp-ls060t1sx01.c | 59 +++++-------------- 1 file changed, 16 insertions(+), 43 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-sharp-ls060t1sx01.c b/drivers/gpu/drm/panel/panel-sharp-ls060t1sx01.c index 74c760ee0c2d..0b4e0983639b 100644 --- a/drivers/gpu/drm/panel/panel-sharp-ls060t1sx01.c +++ b/drivers/gpu/drm/panel/panel-sharp-ls060t1sx01.c @@ -44,60 +44,39 @@ static void sharp_ls060_reset(struct sharp_ls060 *ctx) static int sharp_ls060_on(struct sharp_ls060 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dsi->mode_flags |= MIPI_DSI_MODE_LPM; - mipi_dsi_dcs_write_seq(dsi, 0xbb, 0x13); - mipi_dsi_dcs_write_seq(dsi, MIPI_DCS_WRITE_MEMORY_START); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xbb, 0x13); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_MEMORY_START); - ret = mipi_dsi_dcs_exit_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to exit sleep mode: %d\n", ret); - return ret; - } - msleep(120); + mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 120); - ret = mipi_dsi_dcs_set_display_on(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display on: %d\n", ret); - return ret; - } - msleep(50); + mipi_dsi_dcs_set_display_on_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 50); - return 0; + return dsi_ctx.accum_err; } -static int sharp_ls060_off(struct sharp_ls060 *ctx) +static void sharp_ls060_off(struct sharp_ls060 *ctx) { struct mipi_dsi_device *dsi = ctx->dsi; - struct device *dev = &dsi->dev; - int ret; + struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi }; dsi->mode_flags &= ~MIPI_DSI_MODE_LPM; - ret = mipi_dsi_dcs_set_display_off(dsi); - if (ret < 0) { - dev_err(dev, "Failed to set display off: %d\n", ret); - return ret; - } - usleep_range(2000, 3000); + mipi_dsi_dcs_set_display_off_multi(&dsi_ctx); + mipi_dsi_usleep_range(&dsi_ctx, 2000, 3000); - ret = mipi_dsi_dcs_enter_sleep_mode(dsi); - if (ret < 0) { - dev_err(dev, "Failed to enter sleep mode: %d\n", ret); - return ret; - } - msleep(121); - - return 0; + mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 121); } static int sharp_ls060_prepare(struct drm_panel *panel) { struct sharp_ls060 *ctx = to_sharp_ls060(panel); - struct device *dev = &ctx->dsi->dev; int ret; ret = regulator_enable(ctx->vddi_supply); @@ -125,10 +104,8 @@ static int sharp_ls060_prepare(struct drm_panel *panel) sharp_ls060_reset(ctx); ret = sharp_ls060_on(ctx); - if (ret < 0) { - dev_err(dev, "Failed to initialize panel: %d\n", ret); + if (ret < 0) goto err_on; - } return 0; @@ -154,12 +131,8 @@ static int sharp_ls060_prepare(struct drm_panel *panel) static int sharp_ls060_unprepare(struct drm_panel *panel) { struct sharp_ls060 *ctx = to_sharp_ls060(panel); - struct device *dev = &ctx->dsi->dev; - int ret; - ret = sharp_ls060_off(ctx); - if (ret < 0) - dev_err(dev, "Failed to un-initialize panel: %d\n", ret); + sharp_ls060_off(ctx); regulator_disable(ctx->vddh_supply); From b1749432a52d3605151634b000fec0361ad45067 Mon Sep 17 00:00:00 2001 From: Tamir Duberstein Date: Sat, 1 Feb 2025 12:40:38 -0500 Subject: [PATCH 0531/1443] rust: kbuild: use host dylib naming in rusttestlib-kernel There seems to have been merge skew between commit b2c261fa8629 ("rust: kbuild: expand rusttest target for macros") and commit 0730422bced5 ("rust: use host dylib naming convention to support macOS") ; the latter replaced `libmacros.so` with `$(libmacros_name)` and the former added an instance of `libmacros.so`. The former was not yet applied when the latter was sent, resulting in a stray `libmacros.so`. Replace the stray with `$(libmacros_name)` to allow `rusttest` to build on macOS. Fixes: 0730422bced5 ("rust: use host dylib naming convention to support macOS") Signed-off-by: Tamir Duberstein Link: https://lore.kernel.org/r/20250201-fix-mac-build-again-v1-1-ca665f5d7de7@gmail.com [ Slightly reworded title. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rust/Makefile b/rust/Makefile index 8fcfd60447bc..ff4343ca3f7c 100644 --- a/rust/Makefile +++ b/rust/Makefile @@ -144,7 +144,7 @@ rusttestlib-kernel: private rustc_target_flags = --extern ffi \ --extern bindings --extern uapi rusttestlib-kernel: $(src)/kernel/lib.rs \ rusttestlib-bindings rusttestlib-uapi rusttestlib-build_error \ - $(obj)/libmacros.so $(obj)/bindings.o FORCE + $(obj)/$(libmacros_name) $(obj)/bindings.o FORCE +$(call if_changed,rustc_test_library) rusttestlib-bindings: private rustc_target_flags = --extern ffi From c21bdb3d8a850afdfa4afe77eea39ae9533629b0 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Tue, 21 Jan 2025 21:09:34 +0100 Subject: [PATCH 0532/1443] rust: init: use explicit ABI to clean warning in future compilers Starting with Rust 1.86.0 (currently in nightly, to be released on 2025-04-03), the `missing_abi` lint is warn-by-default [1]: error: extern declarations without an explicit ABI are deprecated --> rust/doctests_kernel_generated.rs:3158:1 | 3158 | extern { | ^^^^^^ help: explicitly specify the C ABI: `extern "C"` | = note: `-D missing-abi` implied by `-D warnings` = help: to override `-D warnings` add `#[allow(missing_abi)]` Thus clean it up. Cc: # Needed in 6.12.y and 6.13.y only (Rust is pinned in older LTSs). Fixes: 7f8977a7fe6d ("rust: init: add `{pin_}chain` functions to `{Pin}Init`") Link: https://github.com/rust-lang/rust/pull/132397 [1] Reviewed-by: Gary Guo Reviewed-by: Alice Ryhl Reviewed-by: Fiona Behrens Link: https://lore.kernel.org/r/20250121200934.222075-1-ojeda@kernel.org [ Added 6.13.y to Cc: stable tag. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/kernel/init.rs | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rust/kernel/init.rs b/rust/kernel/init.rs index 3f9236c1c9d5..7fd1ea8265a5 100644 --- a/rust/kernel/init.rs +++ b/rust/kernel/init.rs @@ -870,7 +870,7 @@ pub unsafe trait PinInit: Sized { /// use kernel::{types::Opaque, init::pin_init_from_closure}; /// #[repr(C)] /// struct RawFoo([u8; 16]); - /// extern { + /// extern "C" { /// fn init_foo(_: *mut RawFoo); /// } /// From 5368a67307b3b2c347dc8965ac55b888be665934 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 4 Feb 2025 23:19:53 +0100 Subject: [PATCH 0533/1443] selftests: mptcp: connect: -f: no reconnect The '-f' parameter is there to force the kernel to emit MPTCP FASTCLOSE by closing the connection with unread bytes in the receive queue. The xdisconnect() helper was used to stop the connection, but it does more than that: it will shut it down, then wait before reconnecting to the same address. This causes the mptcp_join's "fastclose test" to fail all the time. This failure is due to a recent change, with commit 218cc166321f ("selftests: mptcp: avoid spurious errors on disconnect"), but that went unnoticed because the test is currently ignored. The recent modification only shown an existing issue: xdisconnect() doesn't need to be used here, only the shutdown() part is needed. Fixes: 6bf41020b72b ("selftests: mptcp: update and extend fastclose test-cases") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20250204-net-mptcp-sft-conn-f-v1-1-6b470c72fffa@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_connect.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c index 414addef9a45..d240d02fa443 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_connect.c +++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c @@ -1302,7 +1302,7 @@ int main_loop(void) return ret; if (cfg_truncate > 0) { - xdisconnect(fd); + shutdown(fd, SHUT_WR); } else if (--cfg_repeat > 0) { xdisconnect(fd); From 647cef20e649c576dff271e018d5d15d998b629d Mon Sep 17 00:00:00 2001 From: Quang Le Date: Mon, 3 Feb 2025 16:58:38 -0800 Subject: [PATCH 0534/1443] pfifo_tail_enqueue: Drop new packet when sch->limit == 0 Expected behaviour: In case we reach scheduler's limit, pfifo_tail_enqueue() will drop a packet in scheduler's queue and decrease scheduler's qlen by one. Then, pfifo_tail_enqueue() enqueue new packet and increase scheduler's qlen by one. Finally, pfifo_tail_enqueue() return `NET_XMIT_CN` status code. Weird behaviour: In case we set `sch->limit == 0` and trigger pfifo_tail_enqueue() on a scheduler that has no packet, the 'drop a packet' step will do nothing. This means the scheduler's qlen still has value equal 0. Then, we continue to enqueue new packet and increase scheduler's qlen by one. In summary, we can leverage pfifo_tail_enqueue() to increase qlen by one and return `NET_XMIT_CN` status code. The problem is: Let's say we have two qdiscs: Qdisc_A and Qdisc_B. - Qdisc_A's type must have '->graft()' function to create parent/child relationship. Let's say Qdisc_A's type is `hfsc`. Enqueue packet to this qdisc will trigger `hfsc_enqueue`. - Qdisc_B's type is pfifo_head_drop. Enqueue packet to this qdisc will trigger `pfifo_tail_enqueue`. - Qdisc_B is configured to have `sch->limit == 0`. - Qdisc_A is configured to route the enqueued's packet to Qdisc_B. Enqueue packet through Qdisc_A will lead to: - hfsc_enqueue(Qdisc_A) -> pfifo_tail_enqueue(Qdisc_B) - Qdisc_B->q.qlen += 1 - pfifo_tail_enqueue() return `NET_XMIT_CN` - hfsc_enqueue() check for `NET_XMIT_SUCCESS` and see `NET_XMIT_CN` => hfsc_enqueue() don't increase qlen of Qdisc_A. The whole process lead to a situation where Qdisc_A->q.qlen == 0 and Qdisc_B->q.qlen == 1. Replace 'hfsc' with other type (for example: 'drr') still lead to the same problem. This violate the design where parent's qlen should equal to the sum of its childrens'qlen. Bug impact: This issue can be used for user->kernel privilege escalation when it is reachable. Fixes: 57dbb2d83d10 ("sched: add head drop fifo queue") Reported-by: Quang Le Signed-off-by: Quang Le Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250204005841.223511-2-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_fifo.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index b50b2c2cc09b..e6bfd39ff339 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -40,6 +40,9 @@ static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch, { unsigned int prev_backlog; + if (unlikely(READ_ONCE(sch->limit) == 0)) + return qdisc_drop(skb, sch, to_free); + if (likely(sch->q.qlen < READ_ONCE(sch->limit))) return qdisc_enqueue_tail(skb, sch); From 3fe5648d1df1798ce14b5464b2ea49f10cd9db31 Mon Sep 17 00:00:00 2001 From: Quang Le Date: Mon, 3 Feb 2025 16:58:39 -0800 Subject: [PATCH 0535/1443] selftests/tc-testing: Add a test case for pfifo_head_drop qdisc when limit==0 When limit == 0, pfifo_tail_enqueue() must drop new packet and increase dropped packets count of the qdisc. All test results: 1..16 ok 1 a519 - Add bfifo qdisc with system default parameters on egress ok 2 585c - Add pfifo qdisc with system default parameters on egress ok 3 a86e - Add bfifo qdisc with system default parameters on egress with handle of maximum value ok 4 9ac8 - Add bfifo qdisc on egress with queue size of 3000 bytes ok 5 f4e6 - Add pfifo qdisc on egress with queue size of 3000 packets ok 6 b1b1 - Add bfifo qdisc with system default parameters on egress with invalid handle exceeding maximum value ok 7 8d5e - Add bfifo qdisc on egress with unsupported argument ok 8 7787 - Add pfifo qdisc on egress with unsupported argument ok 9 c4b6 - Replace bfifo qdisc on egress with new queue size ok 10 3df6 - Replace pfifo qdisc on egress with new queue size ok 11 7a67 - Add bfifo qdisc on egress with queue size in invalid format ok 12 1298 - Add duplicate bfifo qdisc on egress ok 13 45a0 - Delete nonexistent bfifo qdisc ok 14 972b - Add prio qdisc on egress with invalid format for handles ok 15 4d39 - Delete bfifo qdisc twice ok 16 d774 - Check pfifo_head_drop qdisc enqueue behaviour when limit == 0 Signed-off-by: Quang Le Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250204005841.223511-3-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- .../tc-testing/tc-tests/qdiscs/fifo.json | 23 +++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fifo.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fifo.json index ae3d286a32b2..6f20d033670d 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fifo.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/fifo.json @@ -313,6 +313,29 @@ "matchPattern": "qdisc bfifo 1: root", "matchCount": "0", "teardown": [ + ] + }, + { + "id": "d774", + "name": "Check pfifo_head_drop qdisc enqueue behaviour when limit == 0", + "category": [ + "qdisc", + "pfifo_head_drop" + ], + "plugins": { + "requires": "nsPlugin" + }, + "setup": [ + "$IP addr add 10.10.10.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY root handle 1: pfifo_head_drop limit 0", + "$IP link set dev $DUMMY up || true" + ], + "cmdUnderTest": "ping -c2 -W0.01 -I $DUMMY 10.10.10.1", + "expExitCode": "1", + "verifyCmd": "$TC -s qdisc show dev $DUMMY", + "matchPattern": "dropped 2", + "matchCount": "1", + "teardown": [ ] } ] From 638ba5089324796c2ee49af10427459c2de35f71 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 3 Feb 2025 16:58:40 -0800 Subject: [PATCH 0536/1443] netem: Update sch->q.qlen before qdisc_tree_reduce_backlog() qdisc_tree_reduce_backlog() notifies parent qdisc only if child qdisc becomes empty, therefore we need to reduce the backlog of the child qdisc before calling it. Otherwise it would miss the opportunity to call cops->qlen_notify(), in the case of DRR, it resulted in UAF since DRR uses ->qlen_notify() to maintain its active list. Fixes: f8d4bc455047 ("net/sched: netem: account for backlog updates from child qdisc") Cc: Martin Ottens Reported-by: Mingi Cho Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250204005841.223511-4-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 71ec9986ed37..fdd79d3ccd8c 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -749,9 +749,9 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch) if (err != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(err)) qdisc_qstats_drop(sch); - qdisc_tree_reduce_backlog(sch, 1, pkt_len); sch->qstats.backlog -= pkt_len; sch->q.qlen--; + qdisc_tree_reduce_backlog(sch, 1, pkt_len); } goto tfifo_dequeue; } From 91aadc16ee73cf958be6b0896da3caea49b7f414 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 3 Feb 2025 16:58:41 -0800 Subject: [PATCH 0537/1443] selftests/tc-testing: Add a test case for qdisc_tree_reduce_backlog() Integrate the test case provided by Mingi Cho into TDC. All test results: 1..4 ok 1 ca5e - Check class delete notification for ffff: ok 2 e4b7 - Check class delete notification for root ffff: ok 3 33a9 - Check ingress is not searchable on backlog update ok 4 a4b9 - Test class qlen notification Cc: Mingi Cho Signed-off-by: Cong Wang Link: https://patch.msgid.link/20250204005841.223511-5-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski --- .../tc-testing/tc-tests/infra/qdiscs.json | 34 ++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json index d3dd65b05b5f..9044ac054167 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json +++ b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json @@ -94,5 +94,37 @@ "$TC qdisc del dev $DUMMY ingress", "$IP addr del 10.10.10.10/24 dev $DUMMY" ] - } + }, + { + "id": "a4b9", + "name": "Test class qlen notification", + "category": [ + "qdisc" + ], + "plugins": { + "requires": "nsPlugin" + }, + "setup": [ + "$IP link set dev $DUMMY up || true", + "$IP addr add 10.10.10.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY root handle 1: drr", + "$TC filter add dev $DUMMY parent 1: basic classid 1:1", + "$TC class add dev $DUMMY parent 1: classid 1:1 drr", + "$TC qdisc add dev $DUMMY parent 1:1 handle 2: netem", + "$TC qdisc add dev $DUMMY parent 2: handle 3: drr", + "$TC filter add dev $DUMMY parent 3: basic action drop", + "$TC class add dev $DUMMY parent 3: classid 3:1 drr", + "$TC class del dev $DUMMY classid 1:1", + "$TC class add dev $DUMMY parent 1: classid 1:1 drr" + ], + "cmdUnderTest": "ping -c1 -W0.01 -I $DUMMY 10.10.10.1", + "expExitCode": "1", + "verifyCmd": "$TC qdisc ls dev $DUMMY", + "matchPattern": "drr 1: root", + "matchCount": "1", + "teardown": [ + "$TC qdisc del dev $DUMMY root handle 1: drr", + "$IP addr del 10.10.10.10/24 dev $DUMMY" + ] + } ] From a70c7b3cbc0688016810bb2e0b9b8a0d6a530045 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Tue, 4 Feb 2025 11:10:06 -0500 Subject: [PATCH 0538/1443] tun: revert fix group permission check This reverts commit 3ca459eaba1bf96a8c7878de84fa8872259a01e3. The blamed commit caused a regression when neither tun->owner nor tun->group is set. This is intended to be allowed, but now requires CAP_NET_ADMIN. Discussion in the referenced thread pointed out that the original issue that prompted this patch can be resolved in userspace. The relaxed access control may also make a device accessible when it previously wasn't, while existing users may depend on it to not be. This is a clean pure git revert, except for fixing the indentation on the gid_valid line that checkpatch correctly flagged. Fixes: 3ca459eaba1b ("tun: fix group permission check") Link: https://lore.kernel.org/netdev/CAFqZXNtkCBT4f+PwyVRmQGoT3p1eVa01fCG_aNtpt6dakXncUg@mail.gmail.com/ Signed-off-by: Willem de Bruijn Cc: Ondrej Mosnacek Cc: Stas Sergeev Link: https://patch.msgid.link/20250204161015.739430-1-willemdebruijn.kernel@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 28624cca91f8..acf96f262488 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -574,18 +574,14 @@ static u16 tun_select_queue(struct net_device *dev, struct sk_buff *skb, return ret; } -static inline bool tun_capable(struct tun_struct *tun) +static inline bool tun_not_capable(struct tun_struct *tun) { const struct cred *cred = current_cred(); struct net *net = dev_net(tun->dev); - if (ns_capable(net->user_ns, CAP_NET_ADMIN)) - return 1; - if (uid_valid(tun->owner) && uid_eq(cred->euid, tun->owner)) - return 1; - if (gid_valid(tun->group) && in_egroup_p(tun->group)) - return 1; - return 0; + return ((uid_valid(tun->owner) && !uid_eq(cred->euid, tun->owner)) || + (gid_valid(tun->group) && !in_egroup_p(tun->group))) && + !ns_capable(net->user_ns, CAP_NET_ADMIN); } static void tun_set_real_num_queues(struct tun_struct *tun) @@ -2782,7 +2778,7 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) !!(tun->flags & IFF_MULTI_QUEUE)) return -EINVAL; - if (!tun_capable(tun)) + if (tun_not_capable(tun)) return -EPERM; err = security_tun_dev_open(tun->security); if (err < 0) From 811b8f534fd85e17077bd2ac0413bcd16cc8fb9b Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 4 Feb 2025 14:38:39 +0200 Subject: [PATCH 0539/1443] net: sched: Fix truncation of offloaded action statistics In case of tc offload, when user space queries the kernel for tc action statistics, tc will query the offloaded statistics from device drivers. Among other statistics, drivers are expected to pass the number of packets that hit the action since the last query as a 64-bit number. Unfortunately, tc treats the number of packets as a 32-bit number, leading to truncation and incorrect statistics when the number of packets since the last query exceeds 0xffffffff: $ tc -s filter show dev swp2 ingress filter protocol all pref 1 flower chain 0 filter protocol all pref 1 flower chain 0 handle 0x1 skip_sw in_hw in_hw_count 1 action order 1: mirred (Egress Redirect to device swp1) stolen index 1 ref 1 bind 1 installed 58 sec used 0 sec Action statistics: Sent 1133877034176 bytes 536959475 pkt (dropped 0, overlimits 0 requeues 0) [...] According to the above, 2111-byte packets were redirected which is impossible as only 64-byte packets were transmitted and the MTU was 1500. Fix by treating packets as a 64-bit number: $ tc -s filter show dev swp2 ingress filter protocol all pref 1 flower chain 0 filter protocol all pref 1 flower chain 0 handle 0x1 skip_sw in_hw in_hw_count 1 action order 1: mirred (Egress Redirect to device swp1) stolen index 1 ref 1 bind 1 installed 61 sec used 0 sec Action statistics: Sent 1370624380864 bytes 21416005951 pkt (dropped 0, overlimits 0 requeues 0) [...] Which shows that only 64-byte packets were redirected (1370624380864 / 21416005951 = 64). Fixes: 380407023526 ("net/sched: Enable netdev drivers to update statistics of offloaded actions") Reported-by: Joe Botha Signed-off-by: Ido Schimmel Reviewed-by: Petr Machata Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250204123839.1151804-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- include/net/sch_generic.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index d635c5b47eba..d48c657191cd 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -851,7 +851,7 @@ static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, } static inline void _bstats_update(struct gnet_stats_basic_sync *bstats, - __u64 bytes, __u32 packets) + __u64 bytes, __u64 packets) { u64_stats_update_begin(&bstats->syncp); u64_stats_add(&bstats->bytes, bytes); From 41b996ce83bf944de5569d6263c8dbd5513e7ed0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 4 Feb 2025 23:05:53 +0000 Subject: [PATCH 0540/1443] rxrpc: Fix call state set to not include the SERVER_SECURING state The RXRPC_CALL_SERVER_SECURING state doesn't really belong with the other states in the call's state set as the other states govern the call's Rx/Tx phase transition and govern when packets can and can't be received or transmitted. The "Securing" state doesn't actually govern the reception of packets and would need to be split depending on whether or not we've received the last packet yet (to mirror RECV_REQUEST/ACK_REQUEST). The "Securing" state is more about whether or not we can start forwarding packets to the application as recvmsg will need to decode them and the decoding can't take place until the challenge/response exchange has completed. Fix this by removing the RXRPC_CALL_SERVER_SECURING state from the state set and, instead, using a flag, RXRPC_CALL_CONN_CHALLENGING, to track whether or not we can queue the call for reception by recvmsg() or notify the kernel app that data is ready. In the event that we've already received all the packets, the connection event handler will poke the app layer in the appropriate manner. Also there's a race whereby the app layer sees the last packet before rxrpc has managed to end the rx phase and change the state to one amenable to allowing a reply. Fix this by queuing the packet after calling rxrpc_end_rx_phase(). Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: David Howells cc: Marc Dionne cc: Simon Horman cc: linux-afs@lists.infradead.org Link: https://patch.msgid.link/20250204230558.712536-2-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/ar-internal.h | 2 +- net/rxrpc/call_object.c | 6 ++---- net/rxrpc/conn_event.c | 4 +--- net/rxrpc/input.c | 2 +- net/rxrpc/sendmsg.c | 2 +- 5 files changed, 6 insertions(+), 10 deletions(-) diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 718193df9d2e..f251845fe532 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -582,6 +582,7 @@ enum rxrpc_call_flag { RXRPC_CALL_EXCLUSIVE, /* The call uses a once-only connection */ RXRPC_CALL_RX_IS_IDLE, /* recvmsg() is idle - send an ACK */ RXRPC_CALL_RECVMSG_READ_ALL, /* recvmsg() read all of the received data */ + RXRPC_CALL_CONN_CHALLENGING, /* The connection is being challenged */ }; /* @@ -602,7 +603,6 @@ enum rxrpc_call_state { RXRPC_CALL_CLIENT_AWAIT_REPLY, /* - client awaiting reply */ RXRPC_CALL_CLIENT_RECV_REPLY, /* - client receiving reply phase */ RXRPC_CALL_SERVER_PREALLOC, /* - service preallocation */ - RXRPC_CALL_SERVER_SECURING, /* - server securing request connection */ RXRPC_CALL_SERVER_RECV_REQUEST, /* - server receiving request */ RXRPC_CALL_SERVER_ACK_REQUEST, /* - server pending ACK of request */ RXRPC_CALL_SERVER_SEND_REPLY, /* - server sending reply */ diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 5a543c3f6fb0..c4c8b46a68c6 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -22,7 +22,6 @@ const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = { [RXRPC_CALL_CLIENT_AWAIT_REPLY] = "ClAwtRpl", [RXRPC_CALL_CLIENT_RECV_REPLY] = "ClRcvRpl", [RXRPC_CALL_SERVER_PREALLOC] = "SvPrealc", - [RXRPC_CALL_SERVER_SECURING] = "SvSecure", [RXRPC_CALL_SERVER_RECV_REQUEST] = "SvRcvReq", [RXRPC_CALL_SERVER_ACK_REQUEST] = "SvAckReq", [RXRPC_CALL_SERVER_SEND_REPLY] = "SvSndRpl", @@ -453,17 +452,16 @@ void rxrpc_incoming_call(struct rxrpc_sock *rx, call->cong_tstamp = skb->tstamp; __set_bit(RXRPC_CALL_EXPOSED, &call->flags); - rxrpc_set_call_state(call, RXRPC_CALL_SERVER_SECURING); + rxrpc_set_call_state(call, RXRPC_CALL_SERVER_RECV_REQUEST); spin_lock(&conn->state_lock); switch (conn->state) { case RXRPC_CONN_SERVICE_UNSECURED: case RXRPC_CONN_SERVICE_CHALLENGING: - rxrpc_set_call_state(call, RXRPC_CALL_SERVER_SECURING); + __set_bit(RXRPC_CALL_CONN_CHALLENGING, &call->flags); break; case RXRPC_CONN_SERVICE: - rxrpc_set_call_state(call, RXRPC_CALL_SERVER_RECV_REQUEST); break; case RXRPC_CONN_ABORTED: diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 74bb49b936cd..4d9c5e21ba78 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -228,10 +228,8 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn) */ static void rxrpc_call_is_secure(struct rxrpc_call *call) { - if (call && __rxrpc_call_state(call) == RXRPC_CALL_SERVER_SECURING) { - rxrpc_set_call_state(call, RXRPC_CALL_SERVER_RECV_REQUEST); + if (call && __test_and_clear_bit(RXRPC_CALL_CONN_CHALLENGING, &call->flags)) rxrpc_notify_socket(call); - } } /* diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 4974b5accafa..4a152f3c831f 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -657,7 +657,7 @@ static bool rxrpc_input_split_jumbo(struct rxrpc_call *call, struct sk_buff *skb rxrpc_propose_delay_ACK(call, sp->hdr.serial, rxrpc_propose_ack_input_data); } - if (notify) { + if (notify && !test_bit(RXRPC_CALL_CONN_CHALLENGING, &call->flags)) { trace_rxrpc_notify_socket(call->debug_id, sp->hdr.serial); rxrpc_notify_socket(call); } diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 0e8da909d4f2..584397aba4a0 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -707,7 +707,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) } else { switch (rxrpc_call_state(call)) { case RXRPC_CALL_CLIENT_AWAIT_CONN: - case RXRPC_CALL_SERVER_SECURING: + case RXRPC_CALL_SERVER_RECV_REQUEST: if (p.command == RXRPC_CMD_SEND_ABORT) break; fallthrough; From 2d7b30aef34dae942e9ab7812b288ce14658ae66 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 4 Feb 2025 23:05:54 +0000 Subject: [PATCH 0541/1443] rxrpc: Fix race in call state changing vs recvmsg() There's a race in between the rxrpc I/O thread recording the end of the receive phase of a call and recvmsg() examining the state of the call to determine whether it has completed. The problem is that call->_state records the I/O thread's view of the call, not the application's view (which may lag), so that alone is not sufficient. To this end, the application also checks whether there is anything left in call->recvmsg_queue for it to pick up. The call must be in state RXRPC_CALL_COMPLETE and the recvmsg_queue empty for the call to be considered fully complete. In rxrpc_input_queue_data(), the latest skbuff is added to the queue and then, if it was marked as LAST_PACKET, the state is advanced... But this is two separate operations with no locking around them. As a consequence, the lack of locking means that sendmsg() can jump into the gap on a service call and attempt to send the reply - but then get rejected because the I/O thread hasn't advanced the state yet. Simply flipping the order in which things are done isn't an option as that impacts the client side, causing the checks in rxrpc_kernel_check_life() as to whether the call is still alive to race instead. Fix this by moving the update of call->_state inside the skb queue spinlocked section where the packet is queued on the I/O thread side. rxrpc's recvmsg() will then automatically sync against this because it has to take the call->recvmsg_queue spinlock in order to dequeue the last packet. rxrpc's sendmsg() doesn't need amending as the app shouldn't be calling it to send a reply until recvmsg() indicates it has returned all of the request. Fixes: 93368b6bd58a ("rxrpc: Move call state changes from recvmsg to I/O thread") Signed-off-by: David Howells cc: Marc Dionne cc: Simon Horman cc: linux-afs@lists.infradead.org Link: https://patch.msgid.link/20250204230558.712536-3-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/input.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 4a152f3c831f..9047ba13bd31 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -448,11 +448,19 @@ static void rxrpc_input_queue_data(struct rxrpc_call *call, struct sk_buff *skb, struct rxrpc_skb_priv *sp = rxrpc_skb(skb); bool last = sp->hdr.flags & RXRPC_LAST_PACKET; - skb_queue_tail(&call->recvmsg_queue, skb); + spin_lock_irq(&call->recvmsg_queue.lock); + + __skb_queue_tail(&call->recvmsg_queue, skb); rxrpc_input_update_ack_window(call, window, wtop); trace_rxrpc_receive(call, last ? why + 1 : why, sp->hdr.serial, sp->hdr.seq); if (last) + /* Change the state inside the lock so that recvmsg syncs + * correctly with it and using sendmsg() to send a reply + * doesn't race. + */ rxrpc_end_rx_phase(call, sp->hdr.serial); + + spin_unlock_irq(&call->recvmsg_queue.lock); } /* From 77c2e45dbf9d2ced21d2cf6cc3b2a048d57ab7ad Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Wed, 5 Feb 2025 13:03:33 -0300 Subject: [PATCH 0542/1443] smb: client: don't trust DFSREF_STORAGE_SERVER bit Some servers don't respect the DFSREF_STORAGE_SERVER bit, so unconditionally tree connect to DFS link target and then decide whether or not continue chasing DFS referrals for DFS interlinks. Otherwise the client would fail to mount such shares. Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/dfs.c | 30 ++++++++++++++++-------------- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c index dad521336b5e..f65a8a90ba27 100644 --- a/fs/smb/client/dfs.c +++ b/fs/smb/client/dfs.c @@ -150,25 +150,27 @@ static int __dfs_referral_walk(struct dfs_ref_walk *rw) if (rc) continue; - if (tgt.flags & DFSREF_STORAGE_SERVER) { - rc = cifs_mount_get_tcon(mnt_ctx); - if (!rc) - rc = cifs_is_path_remote(mnt_ctx); + rc = cifs_mount_get_tcon(mnt_ctx); + if (rc) { + if (tgt.server_type == DFS_TYPE_LINK && + DFS_INTERLINK(tgt.flags)) + rc = -EREMOTE; + } else { + rc = cifs_is_path_remote(mnt_ctx); if (!rc) { ref_walk_set_tgt_hint(rw); break; } - if (rc != -EREMOTE) - continue; } - - rc = ref_walk_advance(rw); - if (!rc) { - rc = setup_dfs_ref(&tgt, rw); - if (rc) - break; - ref_walk_mark_end(rw); - goto again; + if (rc == -EREMOTE) { + rc = ref_walk_advance(rw); + if (!rc) { + rc = setup_dfs_ref(&tgt, rw); + if (rc) + break; + ref_walk_mark_end(rw); + goto again; + } } } } while (rc && ref_walk_descend(rw)); From 773dc23ff81838b6f74d7fabba5a441cc6a93982 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Wed, 5 Feb 2025 13:22:11 -0300 Subject: [PATCH 0543/1443] smb: client: fix noisy when tree connecting to DFS interlink targets When the client attempts to tree connect to a domain-based DFS namespace from a DFS interlink target, the server will return STATUS_BAD_NETWORK_NAME and the following will appear on dmesg: CIFS: VFS: BAD_NETWORK_NAME: \\dom\dfs Since a DFS share might contain several DFS interlinks and they expire after 10 minutes, the above message might end up being flooded on dmesg when mounting or accessing them. Print this only once per share. Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/smb2pdu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 40ad9e79437a..78395195e016 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -2169,7 +2169,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, tcon_error_exit: if (rsp && rsp->hdr.Status == STATUS_BAD_NETWORK_NAME) - cifs_tcon_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree); + cifs_dbg(VFS | ONCE, "BAD_NETWORK_NAME: %s\n", tree); goto tcon_exit; } From be1963dd4ce4e467f062b023d1e696f40c926a04 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Wed, 5 Feb 2025 13:41:32 -0300 Subject: [PATCH 0544/1443] smb: client: get rid of kstrdup() in get_ses_refpath() After commit 36008fe6e3dc ("smb: client: don't try following DFS links in cifs_tree_connect()"), TCP_Server_Info::leaf_fullpath will no longer be changed, so there is no need to kstrdup() it. Signed-off-by: Paulo Alcantara (Red Hat) Signed-off-by: Steve French --- fs/smb/client/dfs.h | 7 +++++++ fs/smb/client/dfs_cache.c | 27 +++++---------------------- 2 files changed, 12 insertions(+), 22 deletions(-) diff --git a/fs/smb/client/dfs.h b/fs/smb/client/dfs.h index ed4cd7cf1ec6..e60f0a24a8a1 100644 --- a/fs/smb/client/dfs.h +++ b/fs/smb/client/dfs.h @@ -188,4 +188,11 @@ static inline void dfs_put_root_smb_sessions(struct list_head *head) } } +static inline const char *dfs_ses_refpath(struct cifs_ses *ses) +{ + const char *path = ses->server->leaf_fullpath; + + return path ? path + 1 : ERR_PTR(-ENOENT); +} + #endif /* _CIFS_DFS_H */ diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c index 5022bb1f122a..4dada26d56b5 100644 --- a/fs/smb/client/dfs_cache.c +++ b/fs/smb/client/dfs_cache.c @@ -1136,33 +1136,19 @@ static bool is_ses_good(struct cifs_ses *ses) return ret; } -static char *get_ses_refpath(struct cifs_ses *ses) -{ - struct TCP_Server_Info *server = ses->server; - char *path = ERR_PTR(-ENOENT); - - if (server->leaf_fullpath) { - path = kstrdup(server->leaf_fullpath + 1, GFP_KERNEL); - if (!path) - path = ERR_PTR(-ENOMEM); - } - return path; -} - /* Refresh dfs referral of @ses */ static void refresh_ses_referral(struct cifs_ses *ses) { struct cache_entry *ce; unsigned int xid; - char *path; + const char *path; int rc = 0; xid = get_xid(); - path = get_ses_refpath(ses); + path = dfs_ses_refpath(ses); if (IS_ERR(path)) { rc = PTR_ERR(path); - path = NULL; goto out; } @@ -1181,7 +1167,6 @@ static void refresh_ses_referral(struct cifs_ses *ses) out: free_xid(xid); - kfree(path); } static int __refresh_tcon_referral(struct cifs_tcon *tcon, @@ -1231,19 +1216,18 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh) struct dfs_info3_param *refs = NULL; struct cache_entry *ce; struct cifs_ses *ses; - unsigned int xid; bool needs_refresh; - char *path; + const char *path; + unsigned int xid; int numrefs = 0; int rc = 0; xid = get_xid(); ses = tcon->ses; - path = get_ses_refpath(ses); + path = dfs_ses_refpath(ses); if (IS_ERR(path)) { rc = PTR_ERR(path); - path = NULL; goto out; } @@ -1271,7 +1255,6 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh) out: free_xid(xid); - kfree(path); free_dfs_info_array(refs, numrefs); } From c7691aec5e991cec9c5c5fdab08c24856a1fc56f Mon Sep 17 00:00:00 2001 From: Vaishnav Achath Date: Wed, 5 Feb 2025 11:52:29 +0530 Subject: [PATCH 0545/1443] arm64: defconfig: Enable TISCI Interrupt Router and Aggregator Enable TISCI Interrupt Router and Interrupt Aggregator drivers. These IPs are found in all TI K3 SoCs like J721E, AM62X and is required for core functionality like DMA, GPIO Interrupts which is necessary during boot, thus make them built-in. bloat-o-meter summary on vmlinux: add/remove: 460/1 grow/shrink: 4/0 up/down: 162483/-8 (162475) ... Total: Before=31615984, After=31778459, chg +0.51% These configs were previously selected for ARCH_K3 in respective Kconfigs till commit b8b26ae398c4 ("irqchip/ti-sci-inta : Add module build support") and commit 2d95ffaecbc2 ("irqchip/ti-sci-intr: Add module build support") dropped them and few driver configs (TI_K3_UDMA, TI_K3_RINGACC) dependent on these also got disabled due to this. While re-enabling the TI_SCI_INT_*_IRQCHIP configs, these configs with missing dependencies (which are already part of arm64 defconfig) also get re-enabled which explains the slightly larger size increase from the bloat-o-meter summary. Fixes: 2d95ffaecbc2 ("irqchip/ti-sci-intr: Add module build support") Fixes: b8b26ae398c4 ("irqchip/ti-sci-inta : Add module build support") Signed-off-by: Vaishnav Achath Tested-by: Dhruva Gole Reviewed-by: Dhruva Gole Link: https://lore.kernel.org/r/20250205062229.3869081-1-vaishnav.a@ti.com Signed-off-by: Nishanth Menon --- arch/arm64/configs/defconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index cb7da4415599..1f25423de383 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -1551,6 +1551,8 @@ CONFIG_PWM_VISCONTI=m CONFIG_SL28CPLD_INTC=y CONFIG_QCOM_PDC=y CONFIG_QCOM_MPM=y +CONFIG_TI_SCI_INTR_IRQCHIP=y +CONFIG_TI_SCI_INTA_IRQCHIP=y CONFIG_RESET_GPIO=m CONFIG_RESET_IMX7=y CONFIG_RESET_QCOM_AOSS=y From 0af4c120f5e7a1ea70aff7da2dfb65b6148a3e84 Mon Sep 17 00:00:00 2001 From: Claudiu Beznea Date: Wed, 5 Feb 2025 12:10:58 +0200 Subject: [PATCH 0546/1443] pinctrl: pinconf-generic: Print unsigned value if a format is registered Commit 3ba11e684d16 ("pinctrl: pinconf-generic: print hex value") unconditionally switched to printing hex values in pinconf_generic_dump_one(). However, if a dump format is registered for the dumped pin, the hex value is printed as well. This hex value does not necessarily correspond 1:1 with the hardware register value (as noted by commit 3ba11e684d16 ("pinctrl: pinconf-generic: print hex value")). As a result, user-facing output may include information like: output drive strength (0x100 uA). To address this, check if a dump format is registered for the dumped property, and print the unsigned value instead when applicable. Fixes: 3ba11e684d16 ("pinctrl: pinconf-generic: print hex value") Signed-off-by: Claudiu Beznea Link: https://lore.kernel.org/20250205101058.2034860-1-claudiu.beznea.uj@bp.renesas.com Signed-off-by: Linus Walleij --- drivers/pinctrl/pinconf-generic.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/pinctrl/pinconf-generic.c b/drivers/pinctrl/pinconf-generic.c index 0b13d7f17b32..42547f64453e 100644 --- a/drivers/pinctrl/pinconf-generic.c +++ b/drivers/pinctrl/pinconf-generic.c @@ -89,12 +89,12 @@ static void pinconf_generic_dump_one(struct pinctrl_dev *pctldev, seq_puts(s, items[i].display); /* Print unit if available */ if (items[i].has_arg) { - seq_printf(s, " (0x%x", - pinconf_to_config_argument(config)); + u32 val = pinconf_to_config_argument(config); + if (items[i].format) - seq_printf(s, " %s)", items[i].format); + seq_printf(s, " (%u %s)", val, items[i].format); else - seq_puts(s, ")"); + seq_printf(s, " (0x%x)", val); } } } From 1e3835a8aea5118d58ff9daa656395e69c8806b2 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 4 Feb 2025 13:57:29 -0800 Subject: [PATCH 0547/1443] MAINTAINERS: add entry for ethtool Michal did an amazing job converting ethtool to Netlink, but never added an entry to MAINTAINERS for himself. Create a formal entry so that we can delegate (portions) of this code to folks. Over the last 3 years majority of the reviews have been done by Andrew and I. I suppose Michal didn't want to be on the receiving end of the flood of patches. Reviewed-by: Simon Horman Signed-off-by: Jakub Kicinski Link: https://patch.msgid.link/20250204215729.168992-1-kuba@kernel.org Signed-off-by: Paolo Abeni --- MAINTAINERS | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 74b09dad4662..20c8daf3ce62 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16455,6 +16455,16 @@ F: include/net/dsa.h F: net/dsa/ F: tools/testing/selftests/drivers/net/dsa/ +NETWORKING [ETHTOOL] +M: Andrew Lunn +M: Jakub Kicinski +F: Documentation/netlink/specs/ethtool.yaml +F: Documentation/networking/ethtool-netlink.rst +F: include/linux/ethtool* +F: include/uapi/linux/ethtool* +F: net/ethtool/ +F: tools/testing/selftests/drivers/net/*/ethtool* + NETWORKING [GENERAL] M: "David S. Miller" M: Eric Dumazet From 82b02a7c459922bbf80e45d5f7e2c4cfef617943 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 4 Feb 2025 13:57:50 -0800 Subject: [PATCH 0548/1443] MAINTAINERS: add a sample ethtool section entry I feel like we don't do a good enough keeping authors of driver APIs around. The ethtool code base was very nicely compartmentalized by Michal. Establish a precedent of creating MAINTAINERS entries for "sections" of the ethtool API. Use Andrew and cable test as a sample entry. The entry should ideally cover 3 elements: a core file, test(s), and keywords. The last one is important because we intend the entries to cover core code *and* reviews of drivers implementing given API! Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Link: https://patch.msgid.link/20250204215750.169249-1-kuba@kernel.org Signed-off-by: Paolo Abeni --- MAINTAINERS | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 20c8daf3ce62..bd705e9123a3 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16465,6 +16465,12 @@ F: include/uapi/linux/ethtool* F: net/ethtool/ F: tools/testing/selftests/drivers/net/*/ethtool* +NETWORKING [ETHTOOL CABLE TEST] +M: Andrew Lunn +F: net/ethtool/cabletest.c +F: tools/testing/selftests/drivers/net/*/ethtool* +K: cable_test + NETWORKING [GENERAL] M: "David S. Miller" M: Eric Dumazet From ba958ac74800573f7f54dbe2a7a7b9a9a523ed52 Mon Sep 17 00:00:00 2001 From: Oleh Zadorozhnyi Date: Tue, 4 Feb 2025 07:17:30 +0200 Subject: [PATCH 0549/1443] kbuild: fix misspelling in scripts/Makefile.lib Signed-off-by: Oleh Zadorozhnyi Signed-off-by: Masahiro Yamada --- scripts/Makefile.lib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index ad55ef201aac..cad20f0e66ee 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -305,7 +305,7 @@ endef # These are shared by some Makefile.* files. ifdef CONFIG_LTO_CLANG -# Run $(LD) here to covert LLVM IR to ELF in the following cases: +# Run $(LD) here to convert LLVM IR to ELF in the following cases: # - when this object needs objtool processing, as objtool cannot process LLVM IR # - when this is a single-object module, as modpost cannot process LLVM IR cmd_ld_single = $(if $(objtool-enabled)$(is-single-obj-m), ; $(LD) $(ld_flags) -r -o $(tmp-target) $@; mv $(tmp-target) $@) From 5da7e15fb5a12e78de974d8908f348e279922ce9 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 31 Jan 2025 19:01:42 -0800 Subject: [PATCH 0550/1443] net: Add rx_skb of kfree_skb to raw_tp_null_args[]. Yan Zhai reported a BPF prog could trigger a null-ptr-deref [0] in trace_kfree_skb if the prog does not check if rx_sk is NULL. Commit c53795d48ee8 ("net: add rx_sk to trace_kfree_skb") added rx_sk to trace_kfree_skb, but rx_sk is optional and could be NULL. Let's add kfree_skb to raw_tp_null_args[] to let the BPF verifier validate such a prog and prevent the issue. Now we fail to load such a prog: libbpf: prog 'drop': -- BEGIN PROG LOAD LOG -- 0: R1=ctx() R10=fp0 ; int BPF_PROG(drop, struct sk_buff *skb, void *location, @ kfree_skb_sk_null.bpf.c:21 0: (79) r3 = *(u64 *)(r1 +24) func 'kfree_skb' arg3 has btf_id 5253 type STRUCT 'sock' 1: R1=ctx() R3_w=trusted_ptr_or_null_sock(id=1) ; bpf_printk("sk: %d, %d\n", sk, sk->__sk_common.skc_family); @ kfree_skb_sk_null.bpf.c:24 1: (69) r4 = *(u16 *)(r3 +16) R3 invalid mem access 'trusted_ptr_or_null_' processed 2 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0 -- END PROG LOAD LOG -- Note this fix requires commit 838a10bd2ebf ("bpf: Augment raw_tp arguments with PTR_MAYBE_NULL"). [0]: BUG: kernel NULL pointer dereference, address: 0000000000000010 PF: supervisor read access in kernel mode PF: error_code(0x0000) - not-present page PGD 0 P4D 0 PREEMPT SMP RIP: 0010:bpf_prog_5e21a6db8fcff1aa_drop+0x10/0x2d Call Trace: ? __die+0x1f/0x60 ? page_fault_oops+0x148/0x420 ? search_bpf_extables+0x5b/0x70 ? fixup_exception+0x27/0x2c0 ? exc_page_fault+0x75/0x170 ? asm_exc_page_fault+0x22/0x30 ? bpf_prog_5e21a6db8fcff1aa_drop+0x10/0x2d bpf_trace_run4+0x68/0xd0 ? unix_stream_connect+0x1f4/0x6f0 sk_skb_reason_drop+0x90/0x120 unix_stream_connect+0x1f4/0x6f0 __sys_connect+0x7f/0xb0 __x64_sys_connect+0x14/0x20 do_syscall_64+0x47/0xc30 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Fixes: c53795d48ee8 ("net: add rx_sk to trace_kfree_skb") Reported-by: Yan Zhai Closes: https://lore.kernel.org/netdev/Z50zebTRzI962e6X@debian.debian/ Signed-off-by: Kuniyuki Iwashima Tested-by: Yan Zhai Acked-by: Jiri Olsa Link: https://lore.kernel.org/r/20250201030142.62703-1-kuniyu@amazon.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/btf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 9de6acddd479..c3223e0db2f5 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -6507,6 +6507,8 @@ static const struct bpf_raw_tp_null_args raw_tp_null_args[] = { /* rxrpc */ { "rxrpc_recvdata", 0x1 }, { "rxrpc_resend", 0x10 }, + /* skb */ + {"kfree_skb", 0x1000}, /* sunrpc */ { "xs_stream_read_data", 0x1 }, /* ... from xprt_cong_event event class */ From 81dde32e7266e7132076b886337bd29b4648e542 Mon Sep 17 00:00:00 2001 From: Damon Ding Date: Thu, 6 Feb 2025 11:03:29 +0800 Subject: [PATCH 0551/1443] dt-bindings: display: rockchip: Fix label name of hdptxphy for RK3588 HDMI TX Controller The hdptxphy is a combo transmit-PHY for HDMI2.1 TMDS Link, FRL Link, DP and eDP Link. Therefore, it is better to name it hdptxphy0 other than hdptxphy_hdmi0, which will be referenced by both hdmi0 and edp0 nodes. Acked-by: Rob Herring (Arm) Signed-off-by: Damon Ding Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20250206030330.680424-2-damon.ding@rock-chips.com --- .../bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml index b838b5ec2dca..96b4b088eebe 100644 --- a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml @@ -157,7 +157,7 @@ examples: , ; interrupt-names = "avp", "cec", "earc", "main", "hpd"; - phys = <&hdptxphy_hdmi0>; + phys = <&hdptxphy0>; power-domains = <&power RK3588_PD_VO1>; resets = <&cru SRST_HDMITX0_REF>, <&cru SRST_HDMIHDP0>; reset-names = "ref", "hdp"; From 2a64c96356c87aa8af826605943e5524bf45e24d Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 5 Feb 2025 12:57:47 +0000 Subject: [PATCH 0552/1443] Revert "net: stmmac: Specify hardware capability value when FIFO size isn't specified" This reverts commit 8865d22656b4, which caused breakage for platforms which are not using xgmac2 or gmac4. Only these two cores have the capability of providing the FIFO sizes from hardware capability fields (which are provided in priv->dma_cap.[tr]x_fifo_size.) All other cores can not, which results in these two fields containing zero. We also have platforms that do not provide a value in priv->plat->[tr]x_fifo_size, resulting in these also being zero. This causes the new tests introduced by the reverted commit to fail, and produce e.g.: stmmaceth f0804000.eth: Can't specify Rx FIFO size An example of such a platform which fails is QEMU's npcm750-evb. This uses dwmac1000 which, as noted above, does not have the capability to provide the FIFO sizes from hardware. Therefore, revert the commit to maintain compatibility with the way the driver used to work. Reported-by: Guenter Roeck Link: https://lore.kernel.org/r/4e98f967-f636-46fb-9eca-d383b9495b86@roeck-us.net Signed-off-by: Russell King (Oracle) Tested-by: Steven Price Fixes: 8865d22656b4 ("net: stmmac: Specify hardware capability value when FIFO size isn't specified") Link: https://patch.msgid.link/E1tfeyR-003YGJ-Gb@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni --- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 35 +++++++++---------- 1 file changed, 17 insertions(+), 18 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index d04543e5697b..b34ebb916b89 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2424,6 +2424,11 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv) u32 chan = 0; u8 qmode = 0; + if (rxfifosz == 0) + rxfifosz = priv->dma_cap.rx_fifo_size; + if (txfifosz == 0) + txfifosz = priv->dma_cap.tx_fifo_size; + /* Split up the shared Tx/Rx FIFO memory on DW QoS Eth and DW XGMAC */ if (priv->plat->has_gmac4 || priv->plat->has_xgmac) { rxfifosz /= rx_channels_count; @@ -2892,6 +2897,11 @@ static void stmmac_set_dma_operation_mode(struct stmmac_priv *priv, u32 txmode, int rxfifosz = priv->plat->rx_fifo_size; int txfifosz = priv->plat->tx_fifo_size; + if (rxfifosz == 0) + rxfifosz = priv->dma_cap.rx_fifo_size; + if (txfifosz == 0) + txfifosz = priv->dma_cap.tx_fifo_size; + /* Adjust for real per queue fifo size */ rxfifosz /= rx_channels_count; txfifosz /= tx_channels_count; @@ -5868,6 +5878,9 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu) const int mtu = new_mtu; int ret; + if (txfifosz == 0) + txfifosz = priv->dma_cap.tx_fifo_size; + txfifosz /= priv->plat->tx_queues_to_use; if (stmmac_xdp_is_enabled(priv) && new_mtu > ETH_DATA_LEN) { @@ -7219,29 +7232,15 @@ static int stmmac_hw_init(struct stmmac_priv *priv) priv->plat->tx_queues_to_use = priv->dma_cap.number_tx_queues; } - if (!priv->plat->rx_fifo_size) { - if (priv->dma_cap.rx_fifo_size) { - priv->plat->rx_fifo_size = priv->dma_cap.rx_fifo_size; - } else { - dev_err(priv->device, "Can't specify Rx FIFO size\n"); - return -ENODEV; - } - } else if (priv->dma_cap.rx_fifo_size && - priv->plat->rx_fifo_size > priv->dma_cap.rx_fifo_size) { + if (priv->dma_cap.rx_fifo_size && + priv->plat->rx_fifo_size > priv->dma_cap.rx_fifo_size) { dev_warn(priv->device, "Rx FIFO size (%u) exceeds dma capability\n", priv->plat->rx_fifo_size); priv->plat->rx_fifo_size = priv->dma_cap.rx_fifo_size; } - if (!priv->plat->tx_fifo_size) { - if (priv->dma_cap.tx_fifo_size) { - priv->plat->tx_fifo_size = priv->dma_cap.tx_fifo_size; - } else { - dev_err(priv->device, "Can't specify Tx FIFO size\n"); - return -ENODEV; - } - } else if (priv->dma_cap.tx_fifo_size && - priv->plat->tx_fifo_size > priv->dma_cap.tx_fifo_size) { + if (priv->dma_cap.tx_fifo_size && + priv->plat->tx_fifo_size > priv->dma_cap.tx_fifo_size) { dev_warn(priv->device, "Tx FIFO size (%u) exceeds dma capability\n", priv->plat->tx_fifo_size); From 79982cbac896768c3860c241df2028c3e75f5a6b Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Tue, 4 Feb 2025 14:40:04 +0200 Subject: [PATCH 0553/1443] dt-bindings: display: vop2: Add optional PLL clock properties On RK3588, HDMI PHY PLL can be used as an alternative and more accurate pixel clock source for VOP2 video ports 0, 1 and 2. Document the optional PLL clock properties corresponding to the two HDMI PHYs available on the SoC. Acked-by: Rob Herring (Arm) Signed-off-by: Cristian Ciocaltea Tested-by: FUKAUMI Naoki Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20250204-vop2-hdmi0-disp-modes-v3-1-d71c6a196e58@collabora.com --- .../devicetree/bindings/display/rockchip/rockchip-vop2.yaml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml index 2531726af306..46d956e63338 100644 --- a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml @@ -53,6 +53,8 @@ properties: - description: Pixel clock for video port 2. - description: Pixel clock for video port 3. - description: Peripheral(vop grf/dsi) clock. + - description: Alternative pixel clock provided by HDMI0 PHY PLL. + - description: Alternative pixel clock provided by HDMI1 PHY PLL. clock-names: minItems: 5 @@ -64,6 +66,8 @@ properties: - const: dclk_vp2 - const: dclk_vp3 - const: pclk_vop + - const: pll_hdmiphy0 + - const: pll_hdmiphy1 rockchip,grf: $ref: /schemas/types.yaml#/definitions/phandle From 9f40d7a94427a503e303b2a2d8db227d615e32c1 Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Tue, 4 Feb 2025 14:40:05 +0200 Subject: [PATCH 0554/1443] drm/rockchip: vop2: Drop unnecessary if_pixclk_rate computation The if_pixclk_rate variable is not being used outside of the if-block in rk3588_calc_cru_cfg(), hence move the superfluous assignment from the first branch to the inner comment-block. Signed-off-by: Cristian Ciocaltea Tested-by: FUKAUMI Naoki Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20250204-vop2-hdmi0-disp-modes-v3-2-d71c6a196e58@collabora.com --- drivers/gpu/drm/rockchip/rockchip_drm_vop2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index 17a98845fd31..2455d4a55abd 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c @@ -1905,8 +1905,8 @@ static unsigned long rk3588_calc_cru_cfg(struct vop2_video_port *vp, int id, K = 2; } - if_pixclk_rate = (dclk_core_rate << 1) / K; /* + * if_pixclk_rate = (dclk_core_rate << 1) / K; * if_dclk_rate = dclk_core_rate / K; * *if_pixclk_div = dclk_rate / if_pixclk_rate; * *if_dclk_div = dclk_rate / if_dclk_rate; From 2c1268e7aad0819f38e56134bbc2095fd95fde1b Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Tue, 4 Feb 2025 14:40:06 +0200 Subject: [PATCH 0555/1443] drm/rockchip: vop2: Improve display modes handling on RK3588 HDMI0 The RK3588 specific implementation is currently quite limited in terms of handling the full range of display modes supported by the connected screens, e.g. 2560x1440@75Hz, 2048x1152@60Hz, 1024x768@60Hz are just a few of them. Additionally, it doesn't cope well with non-integer refresh rates like 59.94, 29.97, 23.98, etc. Make use of HDMI0 PHY PLL as a more accurate DCLK source to handle all display modes up to 4K@60Hz. Tested-by: FUKAUMI Naoki Signed-off-by: Cristian Ciocaltea Signed-off-by: Heiko Stuebner Link: https://patchwork.freedesktop.org/patch/msgid/20250204-vop2-hdmi0-disp-modes-v3-3-d71c6a196e58@collabora.com --- drivers/gpu/drm/rockchip/rockchip_drm_vop2.c | 34 ++++++++++++++++++++ 1 file changed, 34 insertions(+) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index 2455d4a55abd..afc946ead870 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c @@ -159,6 +159,7 @@ struct vop2_video_port { struct drm_crtc crtc; struct vop2 *vop2; struct clk *dclk; + struct clk *dclk_src; unsigned int id; const struct vop2_video_port_data *data; @@ -214,6 +215,7 @@ struct vop2 { struct clk *hclk; struct clk *aclk; struct clk *pclk; + struct clk *pll_hdmiphy0; /* optional internal rgb encoder */ struct rockchip_rgb *rgb; @@ -222,6 +224,8 @@ struct vop2 { struct vop2_win win[]; }; +#define VOP2_MAX_DCLK_RATE 600000000 + #define vop2_output_if_is_hdmi(x) ((x) == ROCKCHIP_VOP2_EP_HDMI0 || \ (x) == ROCKCHIP_VOP2_EP_HDMI1) @@ -1155,6 +1159,9 @@ static void vop2_crtc_atomic_disable(struct drm_crtc *crtc, vop2_crtc_disable_irq(vp, VP_INT_DSP_HOLD_VALID); + if (vp->dclk_src) + clk_set_parent(vp->dclk, vp->dclk_src); + clk_disable_unprepare(vp->dclk); vop2->enable_count--; @@ -2259,6 +2266,27 @@ static void vop2_crtc_atomic_enable(struct drm_crtc *crtc, vop2_vp_write(vp, RK3568_VP_MIPI_CTRL, 0); + /* + * Switch to HDMI PHY PLL as DCLK source for display modes up + * to 4K@60Hz, if available, otherwise keep using the system CRU. + */ + if (vop2->pll_hdmiphy0 && clock <= VOP2_MAX_DCLK_RATE) { + drm_for_each_encoder_mask(encoder, crtc->dev, crtc_state->encoder_mask) { + struct rockchip_encoder *rkencoder = to_rockchip_encoder(encoder); + + if (rkencoder->crtc_endpoint_id == ROCKCHIP_VOP2_EP_HDMI0) { + if (!vp->dclk_src) + vp->dclk_src = clk_get_parent(vp->dclk); + + ret = clk_set_parent(vp->dclk, vop2->pll_hdmiphy0); + if (ret < 0) + drm_warn(vop2->drm, + "Could not switch to HDMI0 PHY PLL: %d\n", ret); + break; + } + } + } + clk_set_rate(vp->dclk, clock); vop2_post_config(crtc); @@ -3699,6 +3727,12 @@ static int vop2_bind(struct device *dev, struct device *master, void *data) return PTR_ERR(vop2->pclk); } + vop2->pll_hdmiphy0 = devm_clk_get_optional(vop2->dev, "pll_hdmiphy0"); + if (IS_ERR(vop2->pll_hdmiphy0)) { + drm_err(vop2->drm, "failed to get pll_hdmiphy0\n"); + return PTR_ERR(vop2->pll_hdmiphy0); + } + vop2->irq = platform_get_irq(pdev, 0); if (vop2->irq < 0) { drm_err(vop2->drm, "cannot find irq for vop2\n"); From 679074942c2502a95842a80471d8fb718165ac77 Mon Sep 17 00:00:00 2001 From: Vitaly Rodionov Date: Wed, 5 Feb 2025 16:08:46 +0000 Subject: [PATCH 0556/1443] ASoC: arizona/madera: use fsleep() in up/down DAPM event delays. Using `fsleep` instead of `msleep` resolves some customer complaints regarding the precision of up/down DAPM event timing. `fsleep()` automatically selects the appropriate sleep function, making the delay time more predictable. Signed-off-by: Vitaly Rodionov Link: https://patch.msgid.link/20250205160849.500306-1-vitalyr@opensource.cirrus.com Reviewed-by: Charles Keepax Signed-off-by: Mark Brown --- sound/soc/codecs/arizona.c | 14 +++++++------- sound/soc/codecs/madera.c | 10 +++++----- sound/soc/codecs/wm5110.c | 8 ++++---- 3 files changed, 16 insertions(+), 16 deletions(-) diff --git a/sound/soc/codecs/arizona.c b/sound/soc/codecs/arizona.c index 402b9a2ff024..68cdb1027d0c 100644 --- a/sound/soc/codecs/arizona.c +++ b/sound/soc/codecs/arizona.c @@ -967,7 +967,7 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, case ARIZONA_OUT3L_ENA_SHIFT: case ARIZONA_OUT3R_ENA_SHIFT: priv->out_up_pending++; - priv->out_up_delay += 17; + priv->out_up_delay += 17000; break; case ARIZONA_OUT4L_ENA_SHIFT: case ARIZONA_OUT4R_ENA_SHIFT: @@ -977,7 +977,7 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, case WM8997: break; default: - priv->out_up_delay += 10; + priv->out_up_delay += 10000; break; } break; @@ -999,7 +999,7 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, if (!priv->out_up_pending && priv->out_up_delay) { dev_dbg(component->dev, "Power up delay: %d\n", priv->out_up_delay); - msleep(priv->out_up_delay); + fsleep(priv->out_up_delay); priv->out_up_delay = 0; } break; @@ -1017,7 +1017,7 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, case ARIZONA_OUT3L_ENA_SHIFT: case ARIZONA_OUT3R_ENA_SHIFT: priv->out_down_pending++; - priv->out_down_delay++; + priv->out_down_delay += 1000; break; case ARIZONA_OUT4L_ENA_SHIFT: case ARIZONA_OUT4R_ENA_SHIFT: @@ -1028,10 +1028,10 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, break; case WM8998: case WM1814: - priv->out_down_delay += 5; + priv->out_down_delay += 5000; break; default: - priv->out_down_delay++; + priv->out_down_delay += 1000; break; } break; @@ -1053,7 +1053,7 @@ int arizona_out_ev(struct snd_soc_dapm_widget *w, if (!priv->out_down_pending && priv->out_down_delay) { dev_dbg(component->dev, "Power down delay: %d\n", priv->out_down_delay); - msleep(priv->out_down_delay); + fsleep(priv->out_down_delay); priv->out_down_delay = 0; } break; diff --git a/sound/soc/codecs/madera.c b/sound/soc/codecs/madera.c index a840a2eb92b9..af109761f359 100644 --- a/sound/soc/codecs/madera.c +++ b/sound/soc/codecs/madera.c @@ -2323,10 +2323,10 @@ int madera_out_ev(struct snd_soc_dapm_widget *w, case CS42L92: case CS47L92: case CS47L93: - out_up_delay = 6; + out_up_delay = 6000; break; default: - out_up_delay = 17; + out_up_delay = 17000; break; } @@ -2357,7 +2357,7 @@ int madera_out_ev(struct snd_soc_dapm_widget *w, case MADERA_OUT3R_ENA_SHIFT: priv->out_up_pending--; if (!priv->out_up_pending) { - msleep(priv->out_up_delay); + fsleep(priv->out_up_delay); priv->out_up_delay = 0; } break; @@ -2376,7 +2376,7 @@ int madera_out_ev(struct snd_soc_dapm_widget *w, case MADERA_OUT3L_ENA_SHIFT: case MADERA_OUT3R_ENA_SHIFT: priv->out_down_pending++; - priv->out_down_delay++; + priv->out_down_delay += 1000; break; default: break; @@ -2393,7 +2393,7 @@ int madera_out_ev(struct snd_soc_dapm_widget *w, case MADERA_OUT3R_ENA_SHIFT: priv->out_down_pending--; if (!priv->out_down_pending) { - msleep(priv->out_down_delay); + fsleep(priv->out_down_delay); priv->out_down_delay = 0; } break; diff --git a/sound/soc/codecs/wm5110.c b/sound/soc/codecs/wm5110.c index 502196253d42..64eee0d2347d 100644 --- a/sound/soc/codecs/wm5110.c +++ b/sound/soc/codecs/wm5110.c @@ -302,7 +302,7 @@ static int wm5110_hp_pre_enable(struct snd_soc_dapm_widget *w) } else { wseq = wm5110_no_dre_left_enable; nregs = ARRAY_SIZE(wm5110_no_dre_left_enable); - priv->out_up_delay += 10; + priv->out_up_delay += 10000; } break; case ARIZONA_OUT1R_ENA_SHIFT: @@ -312,7 +312,7 @@ static int wm5110_hp_pre_enable(struct snd_soc_dapm_widget *w) } else { wseq = wm5110_no_dre_right_enable; nregs = ARRAY_SIZE(wm5110_no_dre_right_enable); - priv->out_up_delay += 10; + priv->out_up_delay += 10000; } break; default: @@ -338,7 +338,7 @@ static int wm5110_hp_pre_disable(struct snd_soc_dapm_widget *w) snd_soc_component_update_bits(component, ARIZONA_SPARE_TRIGGERS, ARIZONA_WS_TRG1, 0); - priv->out_down_delay += 27; + priv->out_down_delay += 27000; } break; case ARIZONA_OUT1R_ENA_SHIFT: @@ -350,7 +350,7 @@ static int wm5110_hp_pre_disable(struct snd_soc_dapm_widget *w) snd_soc_component_update_bits(component, ARIZONA_SPARE_TRIGGERS, ARIZONA_WS_TRG2, 0); - priv->out_down_delay += 27; + priv->out_down_delay += 27000; } break; default: From 1d44a30ae3f9195cb4eb7d81bb9ced2776232094 Mon Sep 17 00:00:00 2001 From: Stefan Binding Date: Wed, 5 Feb 2025 16:48:04 +0000 Subject: [PATCH 0557/1443] ASoC: cs35l41: Fallback to using HID for system_name if no SUB is available MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For systems which load firmware on the cs35l41 which use ACPI, the _SUB value is used to differentiate firmware and tuning files for the individual systems. In the case where a system does not have a _SUB defined in ACPI node for cs35l41, there needs to be a fallback to allow the files for that system to be differentiated. Since all ACPI nodes for cs35l41 should have a HID defined, the HID should be a safe option. Signed-off-by: Stefan Binding Reviewed-by: André Almeida Tested-by: André Almeida Link: https://patch.msgid.link/20250205164806.414020-1-sbinding@opensource.cirrus.com Signed-off-by: Mark Brown --- sound/soc/codecs/cs35l41.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/sound/soc/codecs/cs35l41.c b/sound/soc/codecs/cs35l41.c index 07a5cab35fe1..30b89018b113 100644 --- a/sound/soc/codecs/cs35l41.c +++ b/sound/soc/codecs/cs35l41.c @@ -1150,19 +1150,28 @@ static int cs35l41_dsp_init(struct cs35l41_private *cs35l41) static int cs35l41_acpi_get_name(struct cs35l41_private *cs35l41) { - acpi_handle handle = ACPI_HANDLE(cs35l41->dev); + struct acpi_device *adev = ACPI_COMPANION(cs35l41->dev); + acpi_handle handle = acpi_device_handle(adev); + const char *hid; const char *sub; - /* If there is no ACPI_HANDLE, there is no ACPI for this system, return 0 */ - if (!handle) + /* If there is no acpi_device, there is no ACPI for this system, return 0 */ + if (!adev) return 0; sub = acpi_get_subsystem_id(handle); if (IS_ERR(sub)) { - /* If bad ACPI, return 0 and fallback to legacy firmware path, otherwise fail */ - if (PTR_ERR(sub) == -ENODATA) - return 0; - else + /* If no _SUB, fallback to _HID, otherwise fail */ + if (PTR_ERR(sub) == -ENODATA) { + hid = acpi_device_hid(adev); + /* If dummy hid, return 0 and fallback to legacy firmware path */ + if (!strcmp(hid, "device")) + return 0; + sub = kstrdup(hid, GFP_KERNEL); + if (!sub) + sub = ERR_PTR(-ENOMEM); + + } else return PTR_ERR(sub); } From 6fd60136d256b3b948333ebdb3835f41a95ab7ef Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Thu, 6 Feb 2025 10:46:42 +0200 Subject: [PATCH 0558/1443] ASoC: SOF: ipc4-topology: Harden loops for looking up ALH copiers Other, non DAI copier widgets could have the same stream name (sname) as the ALH copier and in that case the copier->data is NULL, no alh_data is attached, which could lead to NULL pointer dereference. We could check for this NULL pointer in sof_ipc4_prepare_copier_module() and avoid the crash, but a similar loop in sof_ipc4_widget_setup_comp_dai() will miscalculate the ALH device count, causing broken audio. The correct fix is to harden the matching logic by making sure that the 1. widget is a DAI widget - so dai = w->private is valid 2. the dai (and thus the copier) is ALH copier Fixes: a150345aa758 ("ASoC: SOF: ipc4-topology: add SoundWire/ALH aggregation support") Reported-by: Seppo Ingalsuo Link: https://github.com/thesofproject/sof/pull/9652 Signed-off-by: Peter Ujfalusi Reviewed-by: Liam Girdwood Reviewed-by: Ranjani Sridharan Reviewed-by: Bard Liao Link: https://patch.msgid.link/20250206084642.14988-1-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/sof/ipc4-topology.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/sound/soc/sof/ipc4-topology.c b/sound/soc/sof/ipc4-topology.c index c04c62478827..6d5cda813e48 100644 --- a/sound/soc/sof/ipc4-topology.c +++ b/sound/soc/sof/ipc4-topology.c @@ -765,10 +765,16 @@ static int sof_ipc4_widget_setup_comp_dai(struct snd_sof_widget *swidget) } list_for_each_entry(w, &sdev->widget_list, list) { - if (w->widget->sname && + struct snd_sof_dai *alh_dai; + + if (!WIDGET_IS_DAI(w->id) || !w->widget->sname || strcmp(w->widget->sname, swidget->widget->sname)) continue; + alh_dai = w->private; + if (alh_dai->type != SOF_DAI_INTEL_ALH) + continue; + blob->alh_cfg.device_count++; } @@ -2061,11 +2067,13 @@ sof_ipc4_prepare_copier_module(struct snd_sof_widget *swidget, list_for_each_entry(w, &sdev->widget_list, list) { u32 node_type; - if (w->widget->sname && + if (!WIDGET_IS_DAI(w->id) || !w->widget->sname || strcmp(w->widget->sname, swidget->widget->sname)) continue; dai = w->private; + if (dai->type != SOF_DAI_INTEL_ALH) + continue; alh_copier = (struct sof_ipc4_copier *)dai->private; alh_data = &alh_copier->data; node_type = SOF_IPC4_GET_NODE_TYPE(alh_data->gtw_cfg.node_id); From 33b7dc7843dbdc9b90c91d11ba30b107f9138ffd Mon Sep 17 00:00:00 2001 From: Terry Cheong Date: Thu, 6 Feb 2025 11:47:23 +0200 Subject: [PATCH 0559/1443] ASoC: SOF: Intel: hda: add softdep pre to snd-hda-codec-hdmi module MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In enviornment without KMOD requesting module may fail to load snd-hda-codec-hdmi, resulting in HDMI audio not usable. Add softdep to loading HDMI codec module first to ensure we can load it correctly. Signed-off-by: Terry Cheong Reviewed-by: Bard Liao Reviewed-by: Johny Lin Reviewed-by: Péter Ujfalusi Signed-off-by: Peter Ujfalusi Link: https://patch.msgid.link/20250206094723.18013-1-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/sof/intel/hda-codec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/sof/intel/hda-codec.c b/sound/soc/sof/intel/hda-codec.c index 568f3dfe822f..2f9925830d1d 100644 --- a/sound/soc/sof/intel/hda-codec.c +++ b/sound/soc/sof/intel/hda-codec.c @@ -454,6 +454,7 @@ int hda_codec_i915_exit(struct snd_sof_dev *sdev) } EXPORT_SYMBOL_NS_GPL(hda_codec_i915_exit, "SND_SOC_SOF_HDA_AUDIO_CODEC_I915"); +MODULE_SOFTDEP("pre: snd-hda-codec-hdmi"); #endif MODULE_LICENSE("Dual BSD/GPL"); From 3588b1c0fde2f58d166e3f94a5a58d64b893526c Mon Sep 17 00:00:00 2001 From: Kunihiko Hayashi Date: Thu, 6 Feb 2025 17:57:47 +0900 Subject: [PATCH 0560/1443] spi: sn-f-ospi: Fix division by zero When there is no dummy cycle in the spi-nor commands, both dummy bus cycle bytes and width are zero. Because of the cpu's warning when divided by zero, the warning should be avoided. Return just zero to avoid such calculations. Fixes: 1b74dd64c861 ("spi: Add Socionext F_OSPI SPI flash controller driver") Co-developed-by: Kohei Ito Signed-off-by: Kohei Ito Signed-off-by: Kunihiko Hayashi Link: https://patch.msgid.link/20250206085747.3834148-1-hayashi.kunihiko@socionext.com Signed-off-by: Mark Brown --- drivers/spi/spi-sn-f-ospi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/spi/spi-sn-f-ospi.c b/drivers/spi/spi-sn-f-ospi.c index 6ad4b729897e..c4969f66a0ba 100644 --- a/drivers/spi/spi-sn-f-ospi.c +++ b/drivers/spi/spi-sn-f-ospi.c @@ -116,6 +116,9 @@ struct f_ospi { static u32 f_ospi_get_dummy_cycle(const struct spi_mem_op *op) { + if (!op->dummy.nbytes) + return 0; + return (op->dummy.nbytes * 8) / op->dummy.buswidth; } From 517e8a7835e8cfb398a0aeb0133de50e31cae32b Mon Sep 17 00:00:00 2001 From: Alan Maguire Date: Wed, 5 Feb 2025 17:00:59 +0000 Subject: [PATCH 0561/1443] bpf: Fix softlockup in arena_map_free on 64k page kernel On an aarch64 kernel with CONFIG_PAGE_SIZE_64KB=y, arena_htab tests cause a segmentation fault and soft lockup. The same failure is not observed with 4k pages on aarch64. It turns out arena_map_free() is calling apply_to_existing_page_range() with the address returned by bpf_arena_get_kern_vm_start(). If this address is not page-aligned the code ends up calling apply_to_pte_range() with that unaligned address causing soft lockup. Fix it by round up GUARD_SZ to PAGE_SIZE << 1 so that the division by 2 in bpf_arena_get_kern_vm_start() returns a page-aligned value. Fixes: 317460317a02 ("bpf: Introduce bpf_arena.") Reported-by: Colm Harrington Suggested-by: Alexei Starovoitov Signed-off-by: Alan Maguire Link: https://lore.kernel.org/r/20250205170059.427458-1-alan.maguire@oracle.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/arena.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/bpf/arena.c b/kernel/bpf/arena.c index 870aeb51d70a..095a9554e1de 100644 --- a/kernel/bpf/arena.c +++ b/kernel/bpf/arena.c @@ -39,7 +39,7 @@ */ /* number of bytes addressable by LDX/STX insn with 16-bit 'off' field */ -#define GUARD_SZ (1ull << sizeof_field(struct bpf_insn, off) * 8) +#define GUARD_SZ round_up(1ull << sizeof_field(struct bpf_insn, off) * 8, PAGE_SIZE << 1) #define KERN_VM_SZ (SZ_4G + GUARD_SZ) struct bpf_arena { From fb97bc2e47f694f79d6358d981ae0428db8e8088 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 29 Jan 2025 15:21:53 +0100 Subject: [PATCH 0562/1443] drm/tests: hdmi: Fix WW_MUTEX_SLOWPATH failures The light_up_connector helper function in the HDMI infrastructure unit tests uses drm_atomic_set_crtc_for_connector(), but fails when it returns an error. This function can return EDEADLK though if the sequence needs to be restarted, and WW_MUTEX_SLOWPATH is meant to test that we handle it properly. Let's handle EDEADLK and restart the sequence in our tests as well. Fixes: eb66d34d793e ("drm/tests: Add output bpc tests") Reported-by: Dave Airlie Closes: https://lore.kernel.org/r/CAPM=9tzJ4-ERDxvuwrCyUPY0=+P44orhp1kLWVGL7MCfpQjMEQ@mail.gmail.com/ Link: https://lore.kernel.org/r/20241031091558.2435850-1-mripard@kernel.org Reviewed-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250129-test-kunit-v2-1-fe59c43805d5@kernel.org Signed-off-by: Maxime Ripard --- drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c index b976a5e9aef5..8e6eb94075a5 100644 --- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c +++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c @@ -70,10 +70,17 @@ static int light_up_connector(struct kunit *test, state = drm_kunit_helper_atomic_state_alloc(test, drm, ctx); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, state); +retry: conn_state = drm_atomic_get_connector_state(state, connector); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, conn_state); ret = drm_atomic_set_crtc_for_connector(conn_state, crtc); + if (ret == -EDEADLK) { + drm_atomic_state_clear(state); + ret = drm_modeset_backoff(ctx); + if (!ret) + goto retry; + } KUNIT_EXPECT_EQ(test, ret, 0); crtc_state = drm_atomic_get_crtc_state(state, crtc); From bb4f929a8875b4801db95b8cf3b2c527c1e475e0 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 29 Jan 2025 15:21:54 +0100 Subject: [PATCH 0563/1443] drm/tests: hdmi: Remove redundant assignments Some tests have the drm pointer assigned multiple times to the same value. Drop the redundant assignments. Reviewed-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250129-test-kunit-v2-2-fe59c43805d5@kernel.org Signed-off-by: Maxime Ripard --- drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c index 8e6eb94075a5..a36422aa9e27 100644 --- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c +++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c @@ -481,7 +481,6 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode_vic_1(struct kunit *test) mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1); KUNIT_ASSERT_NOT_NULL(test, mode); - drm = &priv->drm; crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, mode, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -595,7 +594,6 @@ static void drm_test_check_broadcast_rgb_full_cea_mode_vic_1(struct kunit *test) mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1); KUNIT_ASSERT_NOT_NULL(test, mode); - drm = &priv->drm; crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, mode, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -711,7 +709,6 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode_vic_1(struct kunit *te mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1); KUNIT_ASSERT_NOT_NULL(test, mode); - drm = &priv->drm; crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, mode, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1313,7 +1310,6 @@ static void drm_test_check_output_bpc_format_vic_1(struct kunit *test) rate = mode->clock * 1500; KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, mode, ctx); KUNIT_EXPECT_EQ(test, ret, 0); From 6b6bfd63e1626ceedc738b2a06505aa5b46c1481 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 29 Jan 2025 15:21:55 +0100 Subject: [PATCH 0564/1443] drm/tests: hdmi: Reorder DRM entities variables assignment The tests all deviate slightly in how they assign their local pointers to DRM entities. This makes refactoring pretty difficult, so let's just move the assignment as soon as the entities are allocated. Reviewed-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250129-test-kunit-v2-3-fe59c43805d5@kernel.org Signed-off-by: Maxime Ripard --- .../drm/tests/drm_hdmi_state_helper_test.c | 81 ++++++++++--------- 1 file changed, 42 insertions(+), 39 deletions(-) diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c index a36422aa9e27..925724b57878 100644 --- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c +++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c @@ -289,15 +289,16 @@ static void drm_test_check_broadcast_rgb_crtc_mode_changed(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; + conn = &priv->connector; + ctx = drm_kunit_helper_acquire_ctx_alloc(test); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - conn = &priv->connector; preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -352,15 +353,16 @@ static void drm_test_check_broadcast_rgb_crtc_mode_not_changed(struct kunit *tes 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; + conn = &priv->connector; + ctx = drm_kunit_helper_acquire_ctx_alloc(test); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - conn = &priv->connector; preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -415,6 +417,8 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); @@ -425,8 +429,6 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -526,6 +528,8 @@ static void drm_test_check_broadcast_rgb_full_cea_mode(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); @@ -536,8 +540,6 @@ static void drm_test_check_broadcast_rgb_full_cea_mode(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -641,6 +643,8 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); @@ -651,8 +655,6 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -758,6 +760,8 @@ static void drm_test_check_output_bpc_crtc_mode_changed(struct kunit *test) 10); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz, @@ -770,8 +774,6 @@ static void drm_test_check_output_bpc_crtc_mode_changed(struct kunit *test) preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -832,6 +834,8 @@ static void drm_test_check_output_bpc_crtc_mode_not_changed(struct kunit *test) 10); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz, @@ -844,8 +848,6 @@ static void drm_test_check_output_bpc_crtc_mode_not_changed(struct kunit *test) preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -903,6 +905,8 @@ static void drm_test_check_output_bpc_dvi(struct kunit *test) 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_dvi_1080p, @@ -918,8 +922,6 @@ static void drm_test_check_output_bpc_dvi(struct kunit *test) preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -950,6 +952,8 @@ static void drm_test_check_tmds_char_rate_rgb_8bpc(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_max_200mhz, @@ -963,8 +967,6 @@ static void drm_test_check_tmds_char_rate_rgb_8bpc(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -997,6 +999,8 @@ static void drm_test_check_tmds_char_rate_rgb_10bpc(struct kunit *test) 10); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_340mhz, @@ -1010,8 +1014,6 @@ static void drm_test_check_tmds_char_rate_rgb_10bpc(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1044,6 +1046,8 @@ static void drm_test_check_tmds_char_rate_rgb_12bpc(struct kunit *test) 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_340mhz, @@ -1057,8 +1061,6 @@ static void drm_test_check_tmds_char_rate_rgb_12bpc(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1095,15 +1097,16 @@ static void drm_test_check_hdmi_funcs_reject_rate(struct kunit *test) 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; + conn = &priv->connector; + ctx = drm_kunit_helper_acquire_ctx_alloc(test); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - conn = &priv->connector; preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1151,6 +1154,8 @@ static void drm_test_check_max_tmds_rate_bpc_fallback(struct kunit *test) 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz, @@ -1174,8 +1179,6 @@ static void drm_test_check_max_tmds_rate_bpc_fallback(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 10, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1220,6 +1223,8 @@ static void drm_test_check_max_tmds_rate_format_fallback(struct kunit *test) 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz, @@ -1246,8 +1251,6 @@ static void drm_test_check_max_tmds_rate_format_fallback(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1343,6 +1346,8 @@ static void drm_test_check_output_bpc_format_driver_rgb_only(struct kunit *test) 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz, @@ -1374,8 +1379,6 @@ static void drm_test_check_output_bpc_format_driver_rgb_only(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1410,6 +1413,8 @@ static void drm_test_check_output_bpc_format_display_rgb_only(struct kunit *test 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_max_200mhz, @@ -1441,8 +1446,6 @@ static void drm_test_check_output_bpc_format_display_rgb_only(struct kunit *test rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1476,6 +1479,8 @@ static void drm_test_check_output_bpc_format_driver_8bpc_only(struct kunit *test 8); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_yuv_dc_max_340mhz, @@ -1499,8 +1504,6 @@ static void drm_test_check_output_bpc_format_driver_8bpc_only(struct kunit *test rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1536,6 +1539,8 @@ static void drm_test_check_output_bpc_format_display_8bpc_only(struct kunit *tes 12); KUNIT_ASSERT_NOT_NULL(test, priv); + drm = &priv->drm; + crtc = priv->crtc; conn = &priv->connector; ret = set_connector_edid(test, conn, test_edid_hdmi_1080p_rgb_max_340mhz, @@ -1559,8 +1564,6 @@ static void drm_test_check_output_bpc_format_display_8bpc_only(struct kunit *tes rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); - drm = &priv->drm; - crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); From 5d14c08a47460e8eedf0185a28b116420ea7f29d Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 29 Jan 2025 15:21:56 +0100 Subject: [PATCH 0565/1443] drm/tests: hdmi: Fix recursive locking The find_preferred_mode() functions takes the mode_config mutex, but due to the order most tests have, is called with the crtc_ww_class_mutex taken. This raises a warning for a circular dependency when running the tests with lockdep. Reorder the tests to call find_preferred_mode before the acquire context has been created to avoid the issue. Reviewed-by: Simona Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20250129-test-kunit-v2-4-fe59c43805d5@kernel.org Signed-off-by: Maxime Ripard --- .../drm/tests/drm_hdmi_state_helper_test.c | 114 +++++++++--------- 1 file changed, 57 insertions(+), 57 deletions(-) diff --git a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c index 925724b57878..23ecc00accb2 100644 --- a/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c +++ b/drivers/gpu/drm/tests/drm_hdmi_state_helper_test.c @@ -293,12 +293,12 @@ static void drm_test_check_broadcast_rgb_crtc_mode_changed(struct kunit *test) crtc = priv->crtc; conn = &priv->connector; - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -357,12 +357,12 @@ static void drm_test_check_broadcast_rgb_crtc_mode_not_changed(struct kunit *tes crtc = priv->crtc; conn = &priv->connector; - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -422,13 +422,13 @@ static void drm_test_check_broadcast_rgb_auto_cea_mode(struct kunit *test) conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -533,13 +533,13 @@ static void drm_test_check_broadcast_rgb_full_cea_mode(struct kunit *test) conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -648,13 +648,13 @@ static void drm_test_check_broadcast_rgb_limited_cea_mode(struct kunit *test) conn = &priv->connector; KUNIT_ASSERT_TRUE(test, conn->display_info.is_hdmi); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_NE(test, drm_match_cea_mode(preferred), 1); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -768,12 +768,12 @@ static void drm_test_check_output_bpc_crtc_mode_changed(struct kunit *test) ARRAY_SIZE(test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz)); KUNIT_ASSERT_GT(test, ret, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -842,12 +842,12 @@ static void drm_test_check_output_bpc_crtc_mode_not_changed(struct kunit *test) ARRAY_SIZE(test_edid_hdmi_1080p_rgb_yuv_dc_max_200mhz)); KUNIT_ASSERT_GT(test, ret, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -916,12 +916,12 @@ static void drm_test_check_output_bpc_dvi(struct kunit *test) info = &conn->display_info; KUNIT_ASSERT_FALSE(test, info->is_hdmi); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -960,13 +960,13 @@ static void drm_test_check_tmds_char_rate_rgb_8bpc(struct kunit *test) ARRAY_SIZE(test_edid_hdmi_1080p_rgb_max_200mhz)); KUNIT_ASSERT_GT(test, ret, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1007,13 +1007,13 @@ static void drm_test_check_tmds_char_rate_rgb_10bpc(struct kunit *test) ARRAY_SIZE(test_edid_hdmi_1080p_rgb_yuv_dc_max_340mhz)); KUNIT_ASSERT_GT(test, ret, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1054,13 +1054,13 @@ static void drm_test_check_tmds_char_rate_rgb_12bpc(struct kunit *test) ARRAY_SIZE(test_edid_hdmi_1080p_rgb_yuv_dc_max_340mhz)); KUNIT_ASSERT_GT(test, ret, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1101,12 +1101,12 @@ static void drm_test_check_hdmi_funcs_reject_rate(struct kunit *test) crtc = priv->crtc; conn = &priv->connector; - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_ASSERT_EQ(test, ret, 0); @@ -1166,9 +1166,6 @@ static void drm_test_check_max_tmds_rate_bpc_fallback(struct kunit *test) KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); @@ -1179,6 +1176,9 @@ static void drm_test_check_max_tmds_rate_bpc_fallback(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 10, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1235,9 +1235,6 @@ static void drm_test_check_max_tmds_rate_format_fallback(struct kunit *test) KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); KUNIT_ASSERT_FALSE(test, preferred->flags & DRM_MODE_FLAG_DBLCLK); @@ -1251,6 +1248,9 @@ static void drm_test_check_max_tmds_rate_format_fallback(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1297,9 +1297,6 @@ static void drm_test_check_output_bpc_format_vic_1(struct kunit *test) KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - mode = drm_kunit_display_mode_from_cea_vic(test, drm, 1); KUNIT_ASSERT_NOT_NULL(test, mode); @@ -1313,6 +1310,9 @@ static void drm_test_check_output_bpc_format_vic_1(struct kunit *test) rate = mode->clock * 1500; KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + crtc = priv->crtc; ret = light_up_connector(test, drm, crtc, conn, mode, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1358,9 +1358,6 @@ static void drm_test_check_output_bpc_format_driver_rgb_only(struct kunit *test) KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); @@ -1379,6 +1376,9 @@ static void drm_test_check_output_bpc_format_driver_rgb_only(struct kunit *test) rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1425,9 +1425,6 @@ static void drm_test_check_output_bpc_format_display_rgb_only(struct kunit *test KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); @@ -1446,6 +1443,9 @@ static void drm_test_check_output_bpc_format_display_rgb_only(struct kunit *test rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_YUV422); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1491,9 +1491,6 @@ static void drm_test_check_output_bpc_format_driver_8bpc_only(struct kunit *test KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); @@ -1504,6 +1501,9 @@ static void drm_test_check_output_bpc_format_driver_8bpc_only(struct kunit *test rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); @@ -1551,9 +1551,6 @@ static void drm_test_check_output_bpc_format_display_8bpc_only(struct kunit *tes KUNIT_ASSERT_TRUE(test, info->is_hdmi); KUNIT_ASSERT_GT(test, info->max_tmds_clock, 0); - ctx = drm_kunit_helper_acquire_ctx_alloc(test); - KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); - preferred = find_preferred_mode(conn); KUNIT_ASSERT_NOT_NULL(test, preferred); @@ -1564,6 +1561,9 @@ static void drm_test_check_output_bpc_format_display_8bpc_only(struct kunit *tes rate = drm_hdmi_compute_mode_clock(preferred, 12, HDMI_COLORSPACE_RGB); KUNIT_ASSERT_LT(test, rate, info->max_tmds_clock * 1000); + ctx = drm_kunit_helper_acquire_ctx_alloc(test); + KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx); + ret = light_up_connector(test, drm, crtc, conn, preferred, ctx); KUNIT_EXPECT_EQ(test, ret, 0); From a5072fc77fb9e38fa9fd883642c83c3720049159 Mon Sep 17 00:00:00 2001 From: Hermes Wu Date: Tue, 21 Jan 2025 15:01:51 +0800 Subject: [PATCH 0566/1443] drm/bridge: it6505: fix HDCP V match check is not performed correctly Fix a typo where V compare incorrectly compares av[] with av[] itself, which can result in HDCP failure. The loop of V compare is expected to iterate for 5 times which compare V array form av[0][] to av[4][]. It should check loop counter reach the last statement "i == 5" before return true Fixes: 0989c02c7a5c ("drm/bridge: it6505: fix HDCP CTS compare V matching") Signed-off-by: Hermes Wu Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250121-fix-hdcp-v-comp-v4-1-185f45c728dc@ite.com.tw --- drivers/gpu/drm/bridge/ite-it6505.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c index 88ef76a37fe6..76dabca04d0d 100644 --- a/drivers/gpu/drm/bridge/ite-it6505.c +++ b/drivers/gpu/drm/bridge/ite-it6505.c @@ -2250,12 +2250,13 @@ static bool it6505_hdcp_part2_ksvlist_check(struct it6505 *it6505) continue; } - for (i = 0; i < 5; i++) { + for (i = 0; i < 5; i++) if (bv[i][3] != av[i][0] || bv[i][2] != av[i][1] || - av[i][1] != av[i][2] || bv[i][0] != av[i][3]) + bv[i][1] != av[i][2] || bv[i][0] != av[i][3]) break; - DRM_DEV_DEBUG_DRIVER(dev, "V' all match!! %d, %d", retry, i); + if (i == 5) { + DRM_DEV_DEBUG_DRIVER(dev, "V' all match!! %d", retry); return true; } } From 0b06000704f8ae72056ad777a67742b7799d6660 Mon Sep 17 00:00:00 2001 From: Shenghao Ding Date: Thu, 6 Feb 2025 20:38:08 +0800 Subject: [PATCH 0567/1443] ASoC: tas2781: drop a redundant code Report from internal ticket, priv->cali_data.data devm_kzalloc twice, drop the first one, it is the unnecessary one. Signed-off-by: Shenghao Ding Link: https://patch.msgid.link/20250206123808.1590-1-shenghao-ding@ti.com Signed-off-by: Mark Brown --- sound/soc/codecs/tas2781-i2c.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/sound/soc/codecs/tas2781-i2c.c b/sound/soc/codecs/tas2781-i2c.c index a730ab6ad4e3..90c5b2e74d12 100644 --- a/sound/soc/codecs/tas2781-i2c.c +++ b/sound/soc/codecs/tas2781-i2c.c @@ -2,7 +2,7 @@ // // ALSA SoC Texas Instruments TAS2563/TAS2781 Audio Smart Amplifier // -// Copyright (C) 2022 - 2024 Texas Instruments Incorporated +// Copyright (C) 2022 - 2025 Texas Instruments Incorporated // https://www.ti.com // // The TAS2563/TAS2781 driver implements a flexible and configurable @@ -1260,8 +1260,6 @@ static int tasdevice_create_cali_ctrls(struct tasdevice_priv *priv) (cali_data->cali_dat_sz_per_dev + 1) + 1 + 15 + 1; priv->cali_data.total_sz = priv->ndev * (cali_data->cali_dat_sz_per_dev + 1); - priv->cali_data.data = devm_kzalloc(priv->dev, - ext_cali_data->max, GFP_KERNEL); cali_ctrls[i].name = cali_name; cali_ctrls[i].iface = SNDRV_CTL_ELEM_IFACE_MIXER; cali_ctrls[i].info = snd_soc_bytes_info_ext; From b029628be267cba3c7684ec684749fe3e4372398 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sun, 12 Jan 2025 23:39:01 -0600 Subject: [PATCH 0568/1443] alpha/elf: Fix misc/setarch test of util-linux by removing 32bit support Richard Henderson writes[1]: > There was a Spec benchmark (I forget which) which was memory bound and ran > twice as fast with 32-bit pointers. > > I copied the idea from DEC to the ELF abi, but never did all the other work > to allow the toolchain to take advantage. > > Amusingly, a later Spec changed the benchmark data sets to not fit into a > 32-bit address space, specifically because of this. > > I expect one could delete the ELF bit and personality and no one would > notice. Not even the 10 remaining Alpha users. In [2] it was pointed out that parts of setarch weren't working properly on alpha because it has it's own SET_PERSONALITY implementation. In the discussion that followed Richard Henderson pointed out that the 32bit pointer support for alpha was never completed. Fix this by removing alpha's 32bit pointer support. As a bit of paranoia refuse to execute any alpha binaries that have the EF_ALPHA_32BIT flag set. Just in case someone somewhere has binaries that try to use alpha's 32bit pointer support. Link: https://lkml.kernel.org/r/CAFXwXrkgu=4Qn-v1PjnOR4SG0oUb9LSa0g6QXpBq4ttm52pJOQ@mail.gmail.com [1] Link: https://lkml.kernel.org/r/20250103140148.370368-1-glaubitz@physik.fu-berlin.de [2] Signed-off-by: Eric W. Biederman Reviewed-by: Richard Henderson Reviewed-by: Arnd Bergmann Reviewed-by: John Paul Adrian Glaubitz Tested-by: John Paul Adrian Glaubitz Link: https://lore.kernel.org/r/87y0zfs26i.fsf_-_@email.froward.int.ebiederm.org Signed-off-by: Kees Cook --- arch/alpha/include/asm/elf.h | 6 +----- arch/alpha/include/asm/pgtable.h | 2 +- arch/alpha/include/asm/processor.h | 8 ++------ arch/alpha/kernel/osf_sys.c | 11 ++--------- 4 files changed, 6 insertions(+), 21 deletions(-) diff --git a/arch/alpha/include/asm/elf.h b/arch/alpha/include/asm/elf.h index 4d7c46f50382..50c82187e60e 100644 --- a/arch/alpha/include/asm/elf.h +++ b/arch/alpha/include/asm/elf.h @@ -74,7 +74,7 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG]; /* * This is used to ensure we don't load something for the wrong architecture. */ -#define elf_check_arch(x) ((x)->e_machine == EM_ALPHA) +#define elf_check_arch(x) (((x)->e_machine == EM_ALPHA) && !((x)->e_flags & EF_ALPHA_32BIT)) /* * These are used to set parameters in the core dumps. @@ -137,10 +137,6 @@ extern int dump_elf_task(elf_greg_t *dest, struct task_struct *task); : amask (AMASK_CIX) ? "ev6" : "ev67"); \ }) -#define SET_PERSONALITY(EX) \ - set_personality(((EX).e_flags & EF_ALPHA_32BIT) \ - ? PER_LINUX_32BIT : PER_LINUX) - extern int alpha_l1i_cacheshape; extern int alpha_l1d_cacheshape; extern int alpha_l2_cacheshape; diff --git a/arch/alpha/include/asm/pgtable.h b/arch/alpha/include/asm/pgtable.h index 635f0a5f5bbd..02e8817a8921 100644 --- a/arch/alpha/include/asm/pgtable.h +++ b/arch/alpha/include/asm/pgtable.h @@ -360,7 +360,7 @@ static inline pte_t pte_swp_clear_exclusive(pte_t pte) extern void paging_init(void); -/* We have our own get_unmapped_area to cope with ADDR_LIMIT_32BIT. */ +/* We have our own get_unmapped_area */ #define HAVE_ARCH_UNMAPPED_AREA #endif /* _ALPHA_PGTABLE_H */ diff --git a/arch/alpha/include/asm/processor.h b/arch/alpha/include/asm/processor.h index 55bb1c09fd39..5dce5518a211 100644 --- a/arch/alpha/include/asm/processor.h +++ b/arch/alpha/include/asm/processor.h @@ -8,23 +8,19 @@ #ifndef __ASM_ALPHA_PROCESSOR_H #define __ASM_ALPHA_PROCESSOR_H -#include /* for ADDR_LIMIT_32BIT */ - /* * We have a 42-bit user address space: 4TB user VM... */ #define TASK_SIZE (0x40000000000UL) -#define STACK_TOP \ - (current->personality & ADDR_LIMIT_32BIT ? 0x80000000 : 0x00120000000UL) +#define STACK_TOP (0x00120000000UL) #define STACK_TOP_MAX 0x00120000000UL /* This decides where the kernel will search for a free chunk of vm * space during mmap's. */ -#define TASK_UNMAPPED_BASE \ - ((current->personality & ADDR_LIMIT_32BIT) ? 0x40000000 : TASK_SIZE / 2) +#define TASK_UNMAPPED_BASE (TASK_SIZE / 2) /* This is dead. Everything has been moved to thread_info. */ struct thread_struct { }; diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index c0424de9e7cd..077a1407be6d 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -1211,8 +1211,7 @@ SYSCALL_DEFINE1(old_adjtimex, struct timex32 __user *, txc_p) return ret; } -/* Get an address range which is currently unmapped. Similar to the - generic version except that we know how to honor ADDR_LIMIT_32BIT. */ +/* Get an address range which is currently unmapped. */ static unsigned long arch_get_unmapped_area_1(unsigned long addr, unsigned long len, @@ -1231,13 +1230,7 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags, vm_flags_t vm_flags) { - unsigned long limit; - - /* "32 bit" actually means 31 bit, since pointers sign extend. */ - if (current->personality & ADDR_LIMIT_32BIT) - limit = 0x80000000; - else - limit = TASK_SIZE; + unsigned long limit = TASK_SIZE; if (len > limit) return -ENOMEM; From 7507eb3e7bfac7c3baef8dd377fdf5871eefd42b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Fri, 31 Jan 2025 17:29:13 +0200 Subject: [PATCH 0569/1443] PCI/ASPM: Fix L1SS saving MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 1db806ec06b7 ("PCI/ASPM: Save parent L1SS config in pci_save_aspm_l1ss_state()") aimed to perform L1SS config save for both the Upstream Port and its upstream bridge when handling an Upstream Port, which matches what the L1SS restore side does. However, parent->state_saved can be set true at an earlier time when the upstream bridge saved other parts of its state. Then later when attempting to save the L1SS config while handling the Upstream Port, parent->state_saved is true in pci_save_aspm_l1ss_state() resulting in early return and skipping saving bridge's L1SS config because it is assumed to be already saved. Later on restore, junk is written into L1SS config which causes issues with some devices. Remove parent->state_saved check and unconditionally save L1SS config also for the upstream bridge from an Upstream Port which ought to be harmless from correctness point of view. With the Upstream Port check now present, saving the L1SS config more than once for the bridge is no longer a problem (unlike when the parent->state_saved check got introduced into the fix during its development). Link: https://lore.kernel.org/r/20250131152913.2507-1-ilpo.jarvinen@linux.intel.com Fixes: 1db806ec06b7 ("PCI/ASPM: Save parent L1SS config in pci_save_aspm_l1ss_state()") Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219731 Reported-by: Niklāvs Koļesņikovs Reported by: Rafael J. Wysocki Closes: https://lore.kernel.org/r/CAJZ5v0iKmynOQ5vKSQbg1J_FmavwZE-nRONovOZ0mpMVauheWg@mail.gmail.com Reported-by: Paul Menzel Closes: https://lore.kernel.org/r/d7246feb-4f3f-4d0c-bb64-89566b170671@molgen.mpg.de Signed-off-by: Ilpo Järvinen Signed-off-by: Bjorn Helgaas Tested-by: Niklāvs Koļesņikovs Tested-by: Paul Menzel # Dell XPS 13 9360 --- drivers/pci/pcie/aspm.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index e0bc90597dca..da3e7edcf49d 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -108,9 +108,6 @@ void pci_save_aspm_l1ss_state(struct pci_dev *pdev) pci_read_config_dword(pdev, pdev->l1ss + PCI_L1SS_CTL2, cap++); pci_read_config_dword(pdev, pdev->l1ss + PCI_L1SS_CTL1, cap++); - if (parent->state_saved) - return; - /* * Save parent's L1 substate configuration so we have it for * pci_restore_aspm_l1ss_state(pdev) to restore. From 57e4a9bd61c308f607bc3e55e8fa02257b06b552 Mon Sep 17 00:00:00 2001 From: Meetakshi Setiya Date: Thu, 6 Feb 2025 01:50:41 -0500 Subject: [PATCH 0570/1443] smb: client: change lease epoch type from unsigned int to __u16 MS-SMB2 section 2.2.13.2.10 specifies that 'epoch' should be a 16-bit unsigned integer used to track lease state changes. Change the data type of all instances of 'epoch' from unsigned int to __u16. This simplifies the epoch change comparisons and makes the code more compliant with the protocol spec. Cc: stable@vger.kernel.org Signed-off-by: Meetakshi Setiya Reviewed-by: Shyam Prasad N Signed-off-by: Steve French --- fs/smb/client/cifsglob.h | 14 +++++++------- fs/smb/client/smb1ops.c | 2 +- fs/smb/client/smb2ops.c | 18 +++++++++--------- fs/smb/client/smb2pdu.c | 2 +- fs/smb/client/smb2proto.h | 2 +- 5 files changed, 19 insertions(+), 19 deletions(-) diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index a68434ad744a..ac1f890a0d54 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -357,7 +357,7 @@ struct smb_version_operations { int (*handle_cancelled_mid)(struct mid_q_entry *, struct TCP_Server_Info *); void (*downgrade_oplock)(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache); + __u16 epoch, bool *purge_cache); /* process transaction2 response */ bool (*check_trans2)(struct mid_q_entry *, struct TCP_Server_Info *, char *, int); @@ -552,12 +552,12 @@ struct smb_version_operations { /* if we can do cache read operations */ bool (*is_read_op)(__u32); /* set oplock level for the inode */ - void (*set_oplock_level)(struct cifsInodeInfo *, __u32, unsigned int, - bool *); + void (*set_oplock_level)(struct cifsInodeInfo *cinode, __u32 oplock, __u16 epoch, + bool *purge_cache); /* create lease context buffer for CREATE request */ char * (*create_lease_buf)(u8 *lease_key, u8 oplock); /* parse lease context buffer and return oplock/epoch info */ - __u8 (*parse_lease_buf)(void *buf, unsigned int *epoch, char *lkey); + __u8 (*parse_lease_buf)(void *buf, __u16 *epoch, char *lkey); ssize_t (*copychunk_range)(const unsigned int, struct cifsFileInfo *src_file, struct cifsFileInfo *target_file, @@ -1447,7 +1447,7 @@ struct cifs_fid { __u8 create_guid[16]; __u32 access; struct cifs_pending_open *pending_open; - unsigned int epoch; + __u16 epoch; #ifdef CONFIG_CIFS_DEBUG2 __u64 mid; #endif /* CIFS_DEBUG2 */ @@ -1480,7 +1480,7 @@ struct cifsFileInfo { bool oplock_break_cancelled:1; bool status_file_deleted:1; /* file has been deleted */ bool offload:1; /* offload final part of _put to a wq */ - unsigned int oplock_epoch; /* epoch from the lease break */ + __u16 oplock_epoch; /* epoch from the lease break */ __u32 oplock_level; /* oplock/lease level from the lease break */ int count; spinlock_t file_info_lock; /* protects four flag/count fields above */ @@ -1577,7 +1577,7 @@ struct cifsInodeInfo { spinlock_t open_file_lock; /* protects openFileList */ __u32 cifsAttrs; /* e.g. DOS archive bit, sparse, compressed, system */ unsigned int oplock; /* oplock/lease level we have */ - unsigned int epoch; /* used to track lease state changes */ + __u16 epoch; /* used to track lease state changes */ #define CIFS_INODE_PENDING_OPLOCK_BREAK (0) /* oplock break in progress */ #define CIFS_INODE_PENDING_WRITERS (1) /* Writes in progress */ #define CIFS_INODE_FLAG_UNUSED (2) /* Unused flag */ diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c index 9756b876a75e..d6e2fb669c40 100644 --- a/fs/smb/client/smb1ops.c +++ b/fs/smb/client/smb1ops.c @@ -377,7 +377,7 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr) static void cifs_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { cifs_set_oplock_level(cinode, oplock); } diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 77309217dab4..ec36bed54b0b 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -3904,22 +3904,22 @@ static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode, static void smb2_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { server->ops->set_oplock_level(cinode, oplock, 0, NULL); } static void smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache); + __u16 epoch, bool *purge_cache); static void smb3_downgrade_oplock(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { unsigned int old_state = cinode->oplock; - unsigned int old_epoch = cinode->epoch; + __u16 old_epoch = cinode->epoch; unsigned int new_state; if (epoch > old_epoch) { @@ -3939,7 +3939,7 @@ smb3_downgrade_oplock(struct TCP_Server_Info *server, static void smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { oplock &= 0xFF; cinode->lease_granted = false; @@ -3963,7 +3963,7 @@ smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, static void smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { char message[5] = {0}; unsigned int new_oplock = 0; @@ -4000,7 +4000,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, static void smb3_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, - unsigned int epoch, bool *purge_cache) + __u16 epoch, bool *purge_cache) { unsigned int old_oplock = cinode->oplock; @@ -4114,7 +4114,7 @@ smb3_create_lease_buf(u8 *lease_key, u8 oplock) } static __u8 -smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) +smb2_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key) { struct create_lease *lc = (struct create_lease *)buf; @@ -4125,7 +4125,7 @@ smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) } static __u8 -smb3_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key) +smb3_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key) { struct create_lease_v2 *lc = (struct create_lease_v2 *)buf; diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 78395195e016..ed7812247ebc 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -2329,7 +2329,7 @@ parse_posix_ctxt(struct create_context *cc, struct smb2_file_all_info *info, int smb2_parse_contexts(struct TCP_Server_Info *server, struct kvec *rsp_iov, - unsigned int *epoch, + __u16 *epoch, char *lease_key, __u8 *oplock, struct smb2_file_all_info *buf, struct create_posix_rsp *posix) diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 2336dfb23f36..4662c7e2d259 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -283,7 +283,7 @@ extern enum securityEnum smb2_select_sectype(struct TCP_Server_Info *, enum securityEnum); int smb2_parse_contexts(struct TCP_Server_Info *server, struct kvec *rsp_iov, - unsigned int *epoch, + __u16 *epoch, char *lease_key, __u8 *oplock, struct smb2_file_all_info *buf, struct create_posix_rsp *posix); From a9c621a217128eb3fb7522cf763992d9437fd5ba Mon Sep 17 00:00:00 2001 From: "Justin M. Forbes" Date: Wed, 29 Jan 2025 14:50:02 -0700 Subject: [PATCH 0571/1443] rust: kbuild: add -fzero-init-padding-bits to bindgen_skip_cflags This seems to break the build when building with gcc15: Unable to generate bindings: ClangDiagnostic("error: unknown argument: '-fzero-init-padding-bits=all'\n") Thus skip that flag. Signed-off-by: Justin M. Forbes Fixes: dce4aab8441d ("kbuild: Use -fzero-init-padding-bits=all") Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20250129215003.1736127-1-jforbes@fedoraproject.org [ Slightly reworded commit. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/rust/Makefile b/rust/Makefile index ff4343ca3f7c..ff8a5e810d65 100644 --- a/rust/Makefile +++ b/rust/Makefile @@ -240,6 +240,7 @@ bindgen_skip_c_flags := -mno-fp-ret-in-387 -mpreferred-stack-boundary=% \ -fzero-call-used-regs=% -fno-stack-clash-protection \ -fno-inline-functions-called-once -fsanitize=bounds-strict \ -fstrict-flex-arrays=% -fmin-function-alignment=% \ + -fzero-init-padding-bits=% \ --param=% --param asan-% # Derived from `scripts/Makefile.clang`. From 6f64b83d9fe9729000a0616830cb1606945465d8 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Wed, 5 Feb 2025 12:52:13 +0000 Subject: [PATCH 0572/1443] PCI/TPH: Restore TPH Requester Enable correctly When we reenable TPH after changing a Steering Tag value, we need the actual TPH Requester Enable value, not the ST Mode (which only happens to work out by chance for non-extended TPH in interrupt vector mode). Link: https://lore.kernel.org/r/13118098116d7bce07aa20b8c52e28c7d1847246.1738759933.git.robin.murphy@arm.com Fixes: d2e8a34876ce ("PCI/TPH: Add Steering Tag support") Signed-off-by: Robin Murphy Signed-off-by: Bjorn Helgaas Reviewed-by: Wei Huang --- drivers/pci/tph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/tph.c b/drivers/pci/tph.c index 1e604fbbda65..07de59ca2ebf 100644 --- a/drivers/pci/tph.c +++ b/drivers/pci/tph.c @@ -360,7 +360,7 @@ int pcie_tph_set_st_entry(struct pci_dev *pdev, unsigned int index, u16 tag) return err; } - set_ctrl_reg_req_en(pdev, pdev->tph_mode); + set_ctrl_reg_req_en(pdev, pdev->tph_req_type); pci_dbg(pdev, "set steering tag: %s table, index=%d, tag=%#04x\n", (loc == PCI_TPH_LOC_MSIX) ? "MSI-X" : "ST", index, tag); From 0e446e3145011b8fe39759b59bd69d39fb47cfeb Mon Sep 17 00:00:00 2001 From: Matthew Maurer Date: Wed, 22 Jan 2025 00:14:43 +0000 Subject: [PATCH 0573/1443] rust: kbuild: do not export generated KASAN ODR symbols ASAN generates special synthetic symbols to help check for ODR violations. These synthetic symbols lack debug information, so gendwarfksyms emits warnings when processing them. No code should ever have a dependency on these symbols, so we should not be exporting them, just like the __cfi symbols. Signed-off-by: Matthew Maurer Reviewed-by: Alice Ryhl Link: https://lore.kernel.org/r/20250122-gendwarfksyms-kasan-rust-v1-1-5ee5658f4fb6@google.com [ Fixed typo in commit message. Slightly reworded title. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rust/Makefile b/rust/Makefile index ff8a5e810d65..ea3849eb78f6 100644 --- a/rust/Makefile +++ b/rust/Makefile @@ -332,7 +332,7 @@ $(obj)/bindings/bindings_helpers_generated.rs: private bindgen_target_extra = ; $(obj)/bindings/bindings_helpers_generated.rs: $(src)/helpers/helpers.c FORCE $(call if_changed_dep,bindgen) -rust_exports = $(NM) -p --defined-only $(1) | awk '$$2~/(T|R|D|B)/ && $$3!~/__cfi/ { printf $(2),$$3 }' +rust_exports = $(NM) -p --defined-only $(1) | awk '$$2~/(T|R|D|B)/ && $$3!~/__cfi/ && $$3!~/__odr_asan/ { printf $(2),$$3 }' quiet_cmd_exports = EXPORTS $@ cmd_exports = \ From 8a734b9359cfa1bdb805f5ca23e20bd99dd18a0a Mon Sep 17 00:00:00 2001 From: Rodrigo Vivi Date: Thu, 9 Jan 2025 14:52:19 -0500 Subject: [PATCH 0574/1443] drm/xe: Fix PVC RPe and RPa information A simple lazy buggy copy and paste of the PVC comment has brought the attention to the incorrect masks of the PVC register for RPa and RPe. So, let's fix them all. Cc: Lucas De Marchi Cc: Vinay Belgaumkar Reviewed-by: Vinay Belgaumkar Link: https://patchwork.freedesktop.org/patch/msgid/20250109195219.658557-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_pc.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 43f9617baba2..02409eedb914 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -363,16 +363,17 @@ static void tgl_update_rpa_value(struct xe_guc_pc *pc) u32 reg; /* - * For PVC we still need to use fused RP1 as the approximation for RPe - * For other platforms than PVC we get the resolved RPe directly from + * For PVC we still need to use fused RP0 as the approximation for RPa + * For other platforms than PVC we get the resolved RPa directly from * PCODE at a different register */ - if (xe->info.platform == XE_PVC) + if (xe->info.platform == XE_PVC) { reg = xe_mmio_read32(>->mmio, PVC_RP_STATE_CAP); - else + pc->rpa_freq = REG_FIELD_GET(RP0_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + } else { reg = xe_mmio_read32(>->mmio, FREQ_INFO_REC); - - pc->rpa_freq = REG_FIELD_GET(RPA_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + pc->rpa_freq = REG_FIELD_GET(RPA_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + } } static void tgl_update_rpe_value(struct xe_guc_pc *pc) @@ -386,12 +387,13 @@ static void tgl_update_rpe_value(struct xe_guc_pc *pc) * For other platforms than PVC we get the resolved RPe directly from * PCODE at a different register */ - if (xe->info.platform == XE_PVC) + if (xe->info.platform == XE_PVC) { reg = xe_mmio_read32(>->mmio, PVC_RP_STATE_CAP); - else + pc->rpe_freq = REG_FIELD_GET(RP1_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + } else { reg = xe_mmio_read32(>->mmio, FREQ_INFO_REC); - - pc->rpe_freq = REG_FIELD_GET(RPE_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + pc->rpe_freq = REG_FIELD_GET(RPE_MASK, reg) * GT_FREQUENCY_MULTIPLIER; + } } static void pc_update_rp_values(struct xe_guc_pc *pc) From 3ace20038e19f23fe73259513f1f08d4bf1a3c83 Mon Sep 17 00:00:00 2001 From: Dhananjay Ugwekar Date: Wed, 5 Feb 2025 11:25:20 +0000 Subject: [PATCH 0575/1443] cpufreq/amd-pstate: Fix cpufreq_policy ref counting amd_pstate_update_limits() takes a cpufreq_policy reference but doesn't decrement the refcount in one of the exit paths, fix that. Fixes: 45722e777fd9 ("cpufreq: amd-pstate: Optimize amd_pstate_update_limits()") Signed-off-by: Dhananjay Ugwekar Reviewed-by: Mario Limonciello Link: https://lore.kernel.org/r/20250205112523.201101-10-dhananjay.ugwekar@amd.com Signed-off-by: Mario Limonciello --- drivers/cpufreq/amd-pstate.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 9dc3933bc326..313550fa62d4 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -821,20 +821,21 @@ static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata) static void amd_pstate_update_limits(unsigned int cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); + struct cpufreq_policy *policy = NULL; struct amd_cpudata *cpudata; u32 prev_high = 0, cur_high = 0; int ret; bool highest_perf_changed = false; + if (!amd_pstate_prefcore) + return; + + policy = cpufreq_cpu_get(cpu); if (!policy) return; cpudata = policy->driver_data; - if (!amd_pstate_prefcore) - return; - guard(mutex)(&amd_pstate_driver_lock); ret = amd_get_highest_perf(cpu, &cur_high); From bb5408801a5f2ecd76b61dcd539a5c466ebaac4c Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 4 Feb 2025 09:45:13 -0800 Subject: [PATCH 0576/1443] stackinit: Keep selftest union size small on m68k The stack frame on m68k is very sensitive to the size of what needs to be stored. Like done for long string testing, reduce the size of the large trailing struct in the union initialization testing. Reported-by: Geert Uytterhoeven Closes: https://lore.kernel.org/all/CAMuHMdXW8VbtOAixO7w+aDOG70aZtZ50j1Ybcr8B3eYnRUcrcA@mail.gmail.com Fixes: e71a29db79da ("stackinit: Add union initialization to selftests") Link: https://lore.kernel.org/r/20250204174509.work.711-kees@kernel.org Signed-off-by: Kees Cook Tested-by: Geert Uytterhoeven --- lib/stackinit_kunit.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/stackinit_kunit.c b/lib/stackinit_kunit.c index fbe910c9c825..967b345a98fd 100644 --- a/lib/stackinit_kunit.c +++ b/lib/stackinit_kunit.c @@ -75,8 +75,10 @@ static bool stackinit_range_contains(char *haystack_start, size_t haystack_size, */ #ifdef CONFIG_M68K #define FILL_SIZE_STRING 8 +#define FILL_SIZE_ARRAY 2 #else #define FILL_SIZE_STRING 16 +#define FILL_SIZE_ARRAY 8 #endif #define INIT_CLONE_SCALAR /**/ @@ -345,7 +347,7 @@ union test_small_start { short three; unsigned long four; struct big_struct { - unsigned long array[8]; + unsigned long array[FILL_SIZE_ARRAY]; } big; }; From 78bba6097b9318f4aa645afeade14024af86af4e Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 3 Feb 2025 15:34:07 +0100 Subject: [PATCH 0577/1443] stackinit: Fix comment for test_small_end In union test_small_end, the small members are three and four. Fixes: e71a29db79da1946 ("stackinit: Add union initialization to selftests") Closes: https://lore.kernel.org/CAMuHMdWvcKOc6v5o3-9-SqP_4oh5-GZQjZZb=-krhY=mVRED_Q@mail.gmail.com Signed-off-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/3f8faa2d7d0d6b36571093ab0fb1fd5157abd7bb.1738593178.git.geert+renesas@glider.be Signed-off-by: Kees Cook --- lib/stackinit_kunit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/stackinit_kunit.c b/lib/stackinit_kunit.c index 967b345a98fd..135322592faf 100644 --- a/lib/stackinit_kunit.c +++ b/lib/stackinit_kunit.c @@ -351,7 +351,7 @@ union test_small_start { } big; }; -/* Mismatched sizes, with one and two being small */ +/* Mismatched sizes, with three and four being small */ union test_small_end { short one; unsigned long two; From 01aebfaeff324d21f559a696016e13640f59b297 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Wed, 5 Feb 2025 13:01:50 +0100 Subject: [PATCH 0578/1443] drm/xe: Don't treat SR-IOV platforms as reclaim unsafe MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit a4d1c5d0b99b ("drm/xe/pf: Move VFs reprovisioning to worker") and commit 78d5d1e20d1d ("drm/xe/relay: Don't use GFP_KERNEL for new transactions") we should have no more lockdep dependencies on the reclaim path when running in the SRIOV mode so we believe that we can now mark SRIOV driver as reclaim safe. Signed-off-by: Michal Wajdeczko Cc: Thomas Hellström Cc: Jonathan Cavitt Cc: Matthew Brost Tested-by: Marcin Bernatowicz Reviewed-by: Marcin Bernatowicz Reviewed-by: Thomas Hellström Link: https://patchwork.freedesktop.org/patch/msgid/20250205120150.896-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_pm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c index 35dd1757b40b..12200be7b43d 100644 --- a/drivers/gpu/drm/xe/xe_pm.c +++ b/drivers/gpu/drm/xe/xe_pm.c @@ -91,7 +91,7 @@ static struct lockdep_map xe_pm_runtime_nod3cold_map = { */ bool xe_rpm_reclaim_safe(const struct xe_device *xe) { - return !xe->d3cold.capable && !xe->info.has_sriov; + return !xe->d3cold.capable; } static void xe_rpm_lockmap_acquire(const struct xe_device *xe) From 0db6b7d49b50c037b5eac19a8d8d1da986db80c6 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Mon, 3 Feb 2025 20:04:44 +0200 Subject: [PATCH 0579/1443] MAINTAINERS: wifi: ath: remove Kalle I'm stepping down as ath10k, ath11k and ath12k maintainer so remove me from MAINTAINERS file and Device Tree bindings. Jeff continues as the maintainer. As my quicinc.com email will not work anymore so add an entry to .mailmap file to direct the mail to my kernel.org address. Signed-off-by: Kalle Valo Acked-by: Jeff Johnson Signed-off-by: Kalle Valo Link: https://patch.msgid.link/20250203180445.1429640-1-kvalo@kernel.org --- .mailmap | 1 + .../devicetree/bindings/net/wireless/qcom,ath10k.yaml | 1 - .../devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml | 1 - .../devicetree/bindings/net/wireless/qcom,ath11k.yaml | 1 - .../devicetree/bindings/net/wireless/qcom,ath12k-wsi.yaml | 1 - .../devicetree/bindings/net/wireless/qcom,ath12k.yaml | 1 - MAINTAINERS | 4 ---- 7 files changed, 1 insertion(+), 9 deletions(-) diff --git a/.mailmap b/.mailmap index 42e42cabb36d..68d36bf46dcb 100644 --- a/.mailmap +++ b/.mailmap @@ -370,6 +370,7 @@ Juha Yrjola Julien Thierry Iskren Chernev Kalle Valo +Kalle Valo Kalyan Thota Karthikeyan Periyasamy Kathiravan T diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml index 070c4c9b8643..aace072e2d52 100644 --- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml @@ -7,7 +7,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Qualcomm Technologies ath10k wireless devices maintainers: - - Kalle Valo - Jeff Johnson description: diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml index a71fdf05bc1e..a4425cf196ab 100644 --- a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml @@ -8,7 +8,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Qualcomm Technologies ath11k wireless devices (PCIe) maintainers: - - Kalle Valo - Jeff Johnson description: | diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml index ff5763dc66a8..a69ffb7b3cb8 100644 --- a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml @@ -8,7 +8,6 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Qualcomm Technologies ath11k wireless devices maintainers: - - Kalle Valo - Jeff Johnson description: | diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-wsi.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-wsi.yaml index cbfb559f6b69..318f305405e3 100644 --- a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-wsi.yaml +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-wsi.yaml @@ -9,7 +9,6 @@ title: Qualcomm Technologies ath12k wireless devices (PCIe) with WSI interface maintainers: - Jeff Johnson - - Kalle Valo description: | Qualcomm Technologies IEEE 802.11be PCIe devices with WSI interface. diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml index 1b5884015b15..9e557cb838c7 100644 --- a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml @@ -9,7 +9,6 @@ title: Qualcomm Technologies ath12k wireless devices (PCIe) maintainers: - Jeff Johnson - - Kalle Valo description: Qualcomm Technologies IEEE 802.11be PCIe devices. diff --git a/MAINTAINERS b/MAINTAINERS index 5bcc78c0be70..2048c75c3c42 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3631,7 +3631,6 @@ F: Documentation/devicetree/bindings/phy/phy-ath79-usb.txt F: drivers/phy/qualcomm/phy-ath79-usb.c ATHEROS ATH GENERIC UTILITIES -M: Kalle Valo M: Jeff Johnson L: linux-wireless@vger.kernel.org S: Supported @@ -19192,7 +19191,6 @@ Q: http://patchwork.linuxtv.org/project/linux-media/list/ F: drivers/media/tuners/qt1010* QUALCOMM ATH12K WIRELESS DRIVER -M: Kalle Valo M: Jeff Johnson L: ath12k@lists.infradead.org S: Supported @@ -19202,7 +19200,6 @@ F: drivers/net/wireless/ath/ath12k/ N: ath12k QUALCOMM ATHEROS ATH10K WIRELESS DRIVER -M: Kalle Valo M: Jeff Johnson L: ath10k@lists.infradead.org S: Supported @@ -19212,7 +19209,6 @@ F: drivers/net/wireless/ath/ath10k/ N: ath10k QUALCOMM ATHEROS ATH11K WIRELESS DRIVER -M: Kalle Valo M: Jeff Johnson L: ath11k@lists.infradead.org S: Supported From b76adb9758f8eaaf22b824d0bcdd694551ce0557 Mon Sep 17 00:00:00 2001 From: Kalle Valo Date: Mon, 3 Feb 2025 20:04:45 +0200 Subject: [PATCH 0580/1443] MAINTAINERS: wifi: remove Kalle I'm stepping down as wireless driver maintainer. Johannes kindly voluntereed to be the "custodian"[1] for the drivers until a better solution is found. Link: https://lore.kernel.org/all/21896d2788b8bc6c7fcb534cd43e75671a57f494.camel@sipsolutions.net/ [1] Signed-off-by: Kalle Valo Link: https://patch.msgid.link/20250203180445.1429640-2-kvalo@kernel.org --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 2048c75c3c42..6401e59ec2cc 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16341,7 +16341,7 @@ X: drivers/net/can/ X: drivers/net/wireless/ NETWORKING DRIVERS (WIRELESS) -M: Kalle Valo +M: Johannes Berg L: linux-wireless@vger.kernel.org S: Maintained W: https://wireless.wiki.kernel.org/ From cf6cb56ef24410fb5308f9655087f1eddf4452e6 Mon Sep 17 00:00:00 2001 From: Eyal Birger Date: Sun, 2 Feb 2025 08:29:20 -0800 Subject: [PATCH 0581/1443] seccomp: passthrough uretprobe systemcall without filtering When attaching uretprobes to processes running inside docker, the attached process is segfaulted when encountering the retprobe. The reason is that now that uretprobe is a system call the default seccomp filters in docker block it as they only allow a specific set of known syscalls. This is true for other userspace applications which use seccomp to control their syscall surface. Since uretprobe is a "kernel implementation detail" system call which is not used by userspace application code directly, it is impractical and there's very little point in forcing all userspace applications to explicitly allow it in order to avoid crashing tracked processes. Pass this systemcall through seccomp without depending on configuration. Note: uretprobe is currently only x86_64 and isn't expected to ever be supported in i386. Fixes: ff474a78cef5 ("uprobe: Add uretprobe syscall to speed up return probe") Reported-by: Rafael Buchbinder Closes: https://lore.kernel.org/lkml/CAHsH6Gs3Eh8DFU0wq58c_LF8A4_+o6z456J7BidmcVY2AqOnHQ@mail.gmail.com/ Link: https://lore.kernel.org/lkml/20250121182939.33d05470@gandalf.local.home/T/#me2676c378eff2d6a33f3054fed4a5f3afa64e65b Link: https://lore.kernel.org/lkml/20250128145806.1849977-1-eyal.birger@gmail.com/ Cc: stable@vger.kernel.org Signed-off-by: Eyal Birger Link: https://lore.kernel.org/r/20250202162921.335813-2-eyal.birger@gmail.com [kees: minimized changes for easier backporting, tweaked commit log] Signed-off-by: Kees Cook --- kernel/seccomp.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/kernel/seccomp.c b/kernel/seccomp.c index f59381c4a2ff..7bbb408431eb 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -749,6 +749,15 @@ static bool seccomp_is_const_allow(struct sock_fprog_kern *fprog, if (WARN_ON_ONCE(!fprog)) return false; + /* Our single exception to filtering. */ +#ifdef __NR_uretprobe +#ifdef SECCOMP_ARCH_COMPAT + if (sd->arch == SECCOMP_ARCH_NATIVE) +#endif + if (sd->nr == __NR_uretprobe) + return true; +#endif + for (pc = 0; pc < fprog->len; pc++) { struct sock_filter *insn = &fprog->filter[pc]; u16 code = insn->code; @@ -1023,6 +1032,9 @@ static inline void seccomp_log(unsigned long syscall, long signr, u32 action, */ static const int mode1_syscalls[] = { __NR_seccomp_read, __NR_seccomp_write, __NR_seccomp_exit, __NR_seccomp_sigreturn, +#ifdef __NR_uretprobe + __NR_uretprobe, +#endif -1, /* negative terminated */ }; From c2debdb8544f415eaf9292a866d4073912eeb561 Mon Sep 17 00:00:00 2001 From: Eyal Birger Date: Sun, 2 Feb 2025 08:29:21 -0800 Subject: [PATCH 0582/1443] selftests/seccomp: validate uretprobe syscall passes through seccomp The uretprobe syscall is implemented as a performance enhancement on x86_64 by having the kernel inject a call to it on function exit; User programs cannot call this system call explicitly. As such, this syscall is considered a kernel implementation detail and should not be filtered by seccomp. Enhance the seccomp bpf test suite to check that uretprobes can be attached to processes without the killing the process regardless of seccomp policy. Signed-off-by: Eyal Birger Link: https://lore.kernel.org/r/20250202162921.335813-3-eyal.birger@gmail.com [kees: Skip archs without __NR_uretprobe] Signed-off-by: Kees Cook --- tools/testing/selftests/seccomp/seccomp_bpf.c | 199 ++++++++++++++++++ 1 file changed, 199 insertions(+) diff --git a/tools/testing/selftests/seccomp/seccomp_bpf.c b/tools/testing/selftests/seccomp/seccomp_bpf.c index 8c3a73461475..14ba51b52095 100644 --- a/tools/testing/selftests/seccomp/seccomp_bpf.c +++ b/tools/testing/selftests/seccomp/seccomp_bpf.c @@ -47,6 +47,7 @@ #include #include #include +#include #include #include @@ -68,6 +69,10 @@ # define PR_SET_PTRACER 0x59616d61 #endif +#ifndef noinline +#define noinline __attribute__((noinline)) +#endif + #ifndef PR_SET_NO_NEW_PRIVS #define PR_SET_NO_NEW_PRIVS 38 #define PR_GET_NO_NEW_PRIVS 39 @@ -4888,6 +4893,200 @@ TEST(tsync_vs_dead_thread_leader) EXPECT_EQ(0, status); } +noinline int probed(void) +{ + return 1; +} + +static int parse_uint_from_file(const char *file, const char *fmt) +{ + int err = -1, ret; + FILE *f; + + f = fopen(file, "re"); + if (f) { + err = fscanf(f, fmt, &ret); + fclose(f); + } + return err == 1 ? ret : err; +} + +static int determine_uprobe_perf_type(void) +{ + const char *file = "/sys/bus/event_source/devices/uprobe/type"; + + return parse_uint_from_file(file, "%d\n"); +} + +static int determine_uprobe_retprobe_bit(void) +{ + const char *file = "/sys/bus/event_source/devices/uprobe/format/retprobe"; + + return parse_uint_from_file(file, "config:%d\n"); +} + +static ssize_t get_uprobe_offset(const void *addr) +{ + size_t start, base, end; + bool found = false; + char buf[256]; + FILE *f; + + f = fopen("/proc/self/maps", "r"); + if (!f) + return -1; + + while (fscanf(f, "%zx-%zx %s %zx %*[^\n]\n", &start, &end, buf, &base) == 4) { + if (buf[2] == 'x' && (uintptr_t)addr >= start && (uintptr_t)addr < end) { + found = true; + break; + } + } + fclose(f); + return found ? (uintptr_t)addr - start + base : -1; +} + +FIXTURE(URETPROBE) { + int fd; +}; + +FIXTURE_VARIANT(URETPROBE) { + /* + * All of the URETPROBE behaviors can be tested with either + * uretprobe attached or not + */ + bool attach; +}; + +FIXTURE_VARIANT_ADD(URETPROBE, attached) { + .attach = true, +}; + +FIXTURE_VARIANT_ADD(URETPROBE, not_attached) { + .attach = false, +}; + +FIXTURE_SETUP(URETPROBE) +{ + const size_t attr_sz = sizeof(struct perf_event_attr); + struct perf_event_attr attr; + ssize_t offset; + int type, bit; + +#ifndef __NR_uretprobe + SKIP(return, "__NR_uretprobe syscall not defined"); +#endif + + if (!variant->attach) + return; + + memset(&attr, 0, attr_sz); + + type = determine_uprobe_perf_type(); + ASSERT_GE(type, 0); + bit = determine_uprobe_retprobe_bit(); + ASSERT_GE(bit, 0); + offset = get_uprobe_offset(probed); + ASSERT_GE(offset, 0); + + attr.config |= 1 << bit; + attr.size = attr_sz; + attr.type = type; + attr.config1 = ptr_to_u64("/proc/self/exe"); + attr.config2 = offset; + + self->fd = syscall(__NR_perf_event_open, &attr, + getpid() /* pid */, -1 /* cpu */, -1 /* group_fd */, + PERF_FLAG_FD_CLOEXEC); +} + +FIXTURE_TEARDOWN(URETPROBE) +{ + /* we could call close(self->fd), but we'd need extra filter for + * that and since we are calling _exit right away.. + */ +} + +static int run_probed_with_filter(struct sock_fprog *prog) +{ + if (prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0) || + seccomp(SECCOMP_SET_MODE_FILTER, 0, prog)) { + return -1; + } + + probed(); + return 0; +} + +TEST_F(URETPROBE, uretprobe_default_allow) +{ + struct sock_filter filter[] = { + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW), + }; + struct sock_fprog prog = { + .len = (unsigned short)ARRAY_SIZE(filter), + .filter = filter, + }; + + ASSERT_EQ(0, run_probed_with_filter(&prog)); +} + +TEST_F(URETPROBE, uretprobe_default_block) +{ + struct sock_filter filter[] = { + BPF_STMT(BPF_LD|BPF_W|BPF_ABS, + offsetof(struct seccomp_data, nr)), + BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_exit_group, 1, 0), + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL), + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW), + }; + struct sock_fprog prog = { + .len = (unsigned short)ARRAY_SIZE(filter), + .filter = filter, + }; + + ASSERT_EQ(0, run_probed_with_filter(&prog)); +} + +TEST_F(URETPROBE, uretprobe_block_uretprobe_syscall) +{ + struct sock_filter filter[] = { + BPF_STMT(BPF_LD|BPF_W|BPF_ABS, + offsetof(struct seccomp_data, nr)), +#ifdef __NR_uretprobe + BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_uretprobe, 0, 1), +#endif + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL), + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW), + }; + struct sock_fprog prog = { + .len = (unsigned short)ARRAY_SIZE(filter), + .filter = filter, + }; + + ASSERT_EQ(0, run_probed_with_filter(&prog)); +} + +TEST_F(URETPROBE, uretprobe_default_block_with_uretprobe_syscall) +{ + struct sock_filter filter[] = { + BPF_STMT(BPF_LD|BPF_W|BPF_ABS, + offsetof(struct seccomp_data, nr)), +#ifdef __NR_uretprobe + BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_uretprobe, 2, 0), +#endif + BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_exit_group, 1, 0), + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL), + BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW), + }; + struct sock_fprog prog = { + .len = (unsigned short)ARRAY_SIZE(filter), + .filter = filter, + }; + + ASSERT_EQ(0, run_probed_with_filter(&prog)); +} + /* * TODO: * - expand NNP testing From 6273a058383e05465083b535ed9469f2c8a48321 Mon Sep 17 00:00:00 2001 From: Alice Ryhl Date: Mon, 3 Feb 2025 08:40:57 +0000 Subject: [PATCH 0583/1443] x86: rust: set rustc-abi=x86-softfloat on rustc>=1.86.0 When using Rust on the x86 architecture, we are currently using the unstable target.json feature to specify the compilation target. Rustc is going to change how softfloat is specified in the target.json file on x86, thus update generate_rust_target.rs to specify softfloat using the new option. Note that if you enable this parameter with a compiler that does not recognize it, then that triggers a warning but it does not break the build. [ For future reference, this solves the following error: RUSTC L rust/core.o error: Error loading target specification: target feature `soft-float` is incompatible with the ABI but gets enabled in target spec. Run `rustc --print target-list` for a list of built-in targets - Miguel ] Cc: # Needed in 6.12.y and 6.13.y only (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust/pull/136146 Signed-off-by: Alice Ryhl Acked-by: Dave Hansen # for x86 Link: https://lore.kernel.org/r/20250203-rustc-1-86-x86-softfloat-v1-1-220a72a5003e@google.com [ Added 6.13.y too to Cc: stable tag and added reasoning to avoid over-backporting. - Miguel ] Signed-off-by: Miguel Ojeda --- scripts/generate_rust_target.rs | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/scripts/generate_rust_target.rs b/scripts/generate_rust_target.rs index 0d00ac3723b5..4fd6b6ab3e32 100644 --- a/scripts/generate_rust_target.rs +++ b/scripts/generate_rust_target.rs @@ -165,6 +165,18 @@ fn has(&self, option: &str) -> bool { let option = "CONFIG_".to_owned() + option; self.0.contains_key(&option) } + + /// Is the rustc version at least `major.minor.patch`? + fn rustc_version_atleast(&self, major: u32, minor: u32, patch: u32) -> bool { + let check_version = 100000 * major + 100 * minor + patch; + let actual_version = self + .0 + .get("CONFIG_RUSTC_VERSION") + .unwrap() + .parse::() + .unwrap(); + check_version <= actual_version + } } fn main() { @@ -182,6 +194,9 @@ fn main() { } } else if cfg.has("X86_64") { ts.push("arch", "x86_64"); + if cfg.rustc_version_atleast(1, 86, 0) { + ts.push("rustc-abi", "x86-softfloat"); + } ts.push( "data-layout", "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128", @@ -215,6 +230,9 @@ fn main() { panic!("32-bit x86 only works under UML"); } ts.push("arch", "x86"); + if cfg.rustc_version_atleast(1, 86, 0) { + ts.push("rustc-abi", "x86-softfloat"); + } ts.push( "data-layout", "e-m:e-p:32:32-p270:32:32-p271:32:32-p272:64:64-i128:128-f64:32:64-f80:32-n8:16:32-S128", From 482ad2a4ace2740ca0ff1cbc8f3c7f862f3ab507 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:09 +0000 Subject: [PATCH 0584/1443] net: add dev_net_rcu() helper dev->nd_net can change, readers should either use rcu_read_lock() or RTNL. We currently use a generic helper, dev_net() with no debugging support. We probably have many hidden bugs. Add dev_net_rcu() helper for callers using rcu_read_lock() protection. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 6 ++++++ include/net/net_namespace.h | 2 +- 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 03bb584c62cf..c0a86afb85da 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2663,6 +2663,12 @@ struct net *dev_net(const struct net_device *dev) return read_pnet(&dev->nd_net); } +static inline +struct net *dev_net_rcu(const struct net_device *dev) +{ + return read_pnet_rcu(&dev->nd_net); +} + static inline void dev_net_set(struct net_device *dev, struct net *net) { diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 0f5eb9db0c62..7ba1402ca779 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -398,7 +398,7 @@ static inline struct net *read_pnet(const possible_net_t *pnet) #endif } -static inline struct net *read_pnet_rcu(possible_net_t *pnet) +static inline struct net *read_pnet_rcu(const possible_net_t *pnet) { #ifdef CONFIG_NET_NS return rcu_dereference(pnet->net); From 469308552ca4560176cfc100e7ca84add1bebd7c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:10 +0000 Subject: [PATCH 0585/1443] ipv4: add RCU protection to ip4_dst_hoplimit() ip4_dst_hoplimit() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: fa50d974d104 ("ipv4: Namespaceify ip_default_ttl sysctl knob") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/route.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/include/net/route.h b/include/net/route.h index f86775be3e29..c605fd5ec0c0 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -382,10 +382,15 @@ static inline int inet_iif(const struct sk_buff *skb) static inline int ip4_dst_hoplimit(const struct dst_entry *dst) { int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT); - struct net *net = dev_net(dst->dev); - if (hoplimit == 0) + if (hoplimit == 0) { + const struct net *net; + + rcu_read_lock(); + net = dev_net_rcu(dst->dev); hoplimit = READ_ONCE(net->ipv4.sysctl_ip_default_ttl); + rcu_read_unlock(); + } return hoplimit; } From 071d8012869b6af352acca346ade13e7be90a49f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:11 +0000 Subject: [PATCH 0586/1443] ipv4: use RCU protection in ip_dst_mtu_maybe_forward() ip_dst_mtu_maybe_forward() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: f87c10a8aa1e8 ("ipv4: introduce ip_dst_mtu_maybe_forward and protect forwarding path against pmtu spoofing") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/ip.h | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/include/net/ip.h b/include/net/ip.h index 9f5e33e371fc..ba7b43447775 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -471,9 +471,12 @@ static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, bool forwarding) { const struct rtable *rt = dst_rtable(dst); - struct net *net = dev_net(dst->dev); - unsigned int mtu; + unsigned int mtu, res; + struct net *net; + rcu_read_lock(); + + net = dev_net_rcu(dst->dev); if (READ_ONCE(net->ipv4.sysctl_ip_fwd_use_pmtu) || ip_mtu_locked(dst) || !forwarding) { @@ -497,7 +500,11 @@ static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, out: mtu = min_t(unsigned int, mtu, IP_MAX_MTU); - return mtu - lwtunnel_headroom(dst->lwtstate, mtu); + res = mtu - lwtunnel_headroom(dst->lwtstate, mtu); + + rcu_read_unlock(); + + return res; } static inline unsigned int ip_skb_dst_mtu(struct sock *sk, From 71b8471c93fa0bcab911fcb65da1eb6c4f5f735f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:12 +0000 Subject: [PATCH 0587/1443] ipv4: use RCU protection in ipv4_default_advmss() ipv4_default_advmss() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: 2e9589ff809e ("ipv4: Namespaceify min_adv_mss sysctl knob") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/route.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 577b88a43293..74c074f45758 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1307,10 +1307,15 @@ static void set_class_tag(struct rtable *rt, u32 tag) static unsigned int ipv4_default_advmss(const struct dst_entry *dst) { - struct net *net = dev_net(dst->dev); unsigned int header_size = sizeof(struct tcphdr) + sizeof(struct iphdr); - unsigned int advmss = max_t(unsigned int, ipv4_mtu(dst) - header_size, - net->ipv4.ip_rt_min_advmss); + unsigned int advmss; + struct net *net; + + rcu_read_lock(); + net = dev_net_rcu(dst->dev); + advmss = max_t(unsigned int, ipv4_mtu(dst) - header_size, + net->ipv4.ip_rt_min_advmss); + rcu_read_unlock(); return min(advmss, IPV4_MAX_PMTU - header_size); } From dd205fcc33d92d54eee4d7f21bb073af9bd5ce2b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:13 +0000 Subject: [PATCH 0588/1443] ipv4: use RCU protection in rt_is_expired() rt_is_expired() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: e84f84f27647 ("netns: place rt_genid into struct net") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-6-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/route.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 74c074f45758..e959327c0ba8 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -390,7 +390,13 @@ static inline int ip_rt_proc_init(void) static inline bool rt_is_expired(const struct rtable *rth) { - return rth->rt_genid != rt_genid_ipv4(dev_net(rth->dst.dev)); + bool res; + + rcu_read_lock(); + res = rth->rt_genid != rt_genid_ipv4(dev_net_rcu(rth->dst.dev)); + rcu_read_unlock(); + + return res; } void rt_cache_flush(struct net *net) From 719817cd293e4fa389e1f69c396f3f816ed5aa41 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:14 +0000 Subject: [PATCH 0589/1443] ipv4: use RCU protection in inet_select_addr() inet_select_addr() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: c4544c724322 ("[NETNS]: Process inet_select_addr inside a namespace.") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250205155120.1676781-7-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/devinet.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index c8b3cf5fba4c..55b8151759bc 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1371,10 +1371,11 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope) __be32 addr = 0; unsigned char localnet_scope = RT_SCOPE_HOST; struct in_device *in_dev; - struct net *net = dev_net(dev); + struct net *net; int master_idx; rcu_read_lock(); + net = dev_net_rcu(dev); in_dev = __in_dev_get_rcu(dev); if (!in_dev) goto no_in_dev; From 139512191bd06f1b496117c76372b2ce372c9a41 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:15 +0000 Subject: [PATCH 0590/1443] ipv4: use RCU protection in __ip_rt_update_pmtu() __ip_rt_update_pmtu() must use RCU protection to make sure the net structure it reads does not disappear. Fixes: 2fbc6e89b2f1 ("ipv4: Update exception handling for multipath routes via same device") Fixes: 1de6b15a434c ("Namespaceify min_pmtu sysctl") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250205155120.1676781-8-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/route.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index e959327c0ba8..753704f75b2c 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1008,9 +1008,9 @@ out: kfree_skb_reason(skb, reason); static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) { struct dst_entry *dst = &rt->dst; - struct net *net = dev_net(dst->dev); struct fib_result res; bool lock = false; + struct net *net; u32 old_mtu; if (ip_mtu_locked(dst)) @@ -1020,6 +1020,8 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (old_mtu < mtu) return; + rcu_read_lock(); + net = dev_net_rcu(dst->dev); if (mtu < net->ipv4.ip_rt_min_pmtu) { lock = true; mtu = min(old_mtu, net->ipv4.ip_rt_min_pmtu); @@ -1027,9 +1029,8 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (rt->rt_pmtu == mtu && !lock && time_before(jiffies, dst->expires - net->ipv4.ip_rt_mtu_expires / 2)) - return; + goto out; - rcu_read_lock(); if (fib_lookup(net, fl4, &res, 0) == 0) { struct fib_nh_common *nhc; @@ -1043,14 +1044,14 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) update_or_create_fnhe(nhc, fl4->daddr, 0, mtu, lock, jiffies + net->ipv4.ip_rt_mtu_expires); } - rcu_read_unlock(); - return; + goto out; } #endif /* CONFIG_IP_ROUTE_MULTIPATH */ nhc = FIB_RES_NHC(res); update_or_create_fnhe(nhc, fl4->daddr, 0, mtu, lock, jiffies + net->ipv4.ip_rt_mtu_expires); } +out: rcu_read_unlock(); } From 4b8474a0951e605d2a27a2c483da4eb4b8c63760 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:16 +0000 Subject: [PATCH 0591/1443] ipv4: icmp: convert to dev_net_rcu() __icmp_send() must ensure rcu_read_lock() is held, as spotted by Jakub. Other ICMP uses of dev_net() seem safe, change them to dev_net_rcu() to get LOCKDEP support. Fixes: dde1bc0e6f86 ("[NETNS]: Add namespace for ICMP replying code.") Closes: https://lore.kernel.org/netdev/20250203153633.46ce0337@kernel.org/ Reported-by: Jakub Kicinski Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250205155120.1676781-9-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/icmp.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 094084b61bff..5482edb5aade 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -399,10 +399,10 @@ static void icmp_push_reply(struct sock *sk, static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) { - struct ipcm_cookie ipc; struct rtable *rt = skb_rtable(skb); - struct net *net = dev_net(rt->dst.dev); + struct net *net = dev_net_rcu(rt->dst.dev); bool apply_ratelimit = false; + struct ipcm_cookie ipc; struct flowi4 fl4; struct sock *sk; struct inet_sock *inet; @@ -608,12 +608,14 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, struct sock *sk; if (!rt) - goto out; + return; + + rcu_read_lock(); if (rt->dst.dev) - net = dev_net(rt->dst.dev); + net = dev_net_rcu(rt->dst.dev); else if (skb_in->dev) - net = dev_net(skb_in->dev); + net = dev_net_rcu(skb_in->dev); else goto out; @@ -785,7 +787,8 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, icmp_xmit_unlock(sk); out_bh_enable: local_bh_enable(); -out:; +out: + rcu_read_unlock(); } EXPORT_SYMBOL(__icmp_send); @@ -834,7 +837,7 @@ static void icmp_socket_deliver(struct sk_buff *skb, u32 info) * avoid additional coding at protocol handlers. */ if (!pskb_may_pull(skb, iph->ihl * 4 + 8)) { - __ICMP_INC_STATS(dev_net(skb->dev), ICMP_MIB_INERRORS); + __ICMP_INC_STATS(dev_net_rcu(skb->dev), ICMP_MIB_INERRORS); return; } @@ -868,7 +871,7 @@ static enum skb_drop_reason icmp_unreach(struct sk_buff *skb) struct net *net; u32 info = 0; - net = dev_net(skb_dst(skb)->dev); + net = dev_net_rcu(skb_dst(skb)->dev); /* * Incomplete header ? @@ -979,7 +982,7 @@ static enum skb_drop_reason icmp_unreach(struct sk_buff *skb) static enum skb_drop_reason icmp_redirect(struct sk_buff *skb) { if (skb->len < sizeof(struct iphdr)) { - __ICMP_INC_STATS(dev_net(skb->dev), ICMP_MIB_INERRORS); + __ICMP_INC_STATS(dev_net_rcu(skb->dev), ICMP_MIB_INERRORS); return SKB_DROP_REASON_PKT_TOO_SMALL; } @@ -1011,7 +1014,7 @@ static enum skb_drop_reason icmp_echo(struct sk_buff *skb) struct icmp_bxm icmp_param; struct net *net; - net = dev_net(skb_dst(skb)->dev); + net = dev_net_rcu(skb_dst(skb)->dev); /* should there be an ICMP stat for ignored echos? */ if (READ_ONCE(net->ipv4.sysctl_icmp_echo_ignore_all)) return SKB_NOT_DROPPED_YET; @@ -1040,9 +1043,9 @@ static enum skb_drop_reason icmp_echo(struct sk_buff *skb) bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr) { + struct net *net = dev_net_rcu(skb->dev); struct icmp_ext_hdr *ext_hdr, _ext_hdr; struct icmp_ext_echo_iio *iio, _iio; - struct net *net = dev_net(skb->dev); struct inet6_dev *in6_dev; struct in_device *in_dev; struct net_device *dev; @@ -1181,7 +1184,7 @@ static enum skb_drop_reason icmp_timestamp(struct sk_buff *skb) return SKB_NOT_DROPPED_YET; out_err: - __ICMP_INC_STATS(dev_net(skb_dst(skb)->dev), ICMP_MIB_INERRORS); + __ICMP_INC_STATS(dev_net_rcu(skb_dst(skb)->dev), ICMP_MIB_INERRORS); return SKB_DROP_REASON_PKT_TOO_SMALL; } @@ -1198,7 +1201,7 @@ int icmp_rcv(struct sk_buff *skb) { enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED; struct rtable *rt = skb_rtable(skb); - struct net *net = dev_net(rt->dst.dev); + struct net *net = dev_net_rcu(rt->dst.dev); struct icmphdr *icmph; if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { @@ -1371,9 +1374,9 @@ int icmp_err(struct sk_buff *skb, u32 info) struct iphdr *iph = (struct iphdr *)skb->data; int offset = iph->ihl<<2; struct icmphdr *icmph = (struct icmphdr *)(skb->data + offset); + struct net *net = dev_net_rcu(skb->dev); int type = icmp_hdr(skb)->type; int code = icmp_hdr(skb)->code; - struct net *net = dev_net(skb->dev); /* * Use ping_err to handle all icmp errors except those From afec62cd0a4191cde6dd3a75382be4d51a38ce9b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:17 +0000 Subject: [PATCH 0592/1443] flow_dissector: use RCU protection to fetch dev_net() __skb_flow_dissect() can be called from arbitrary contexts. It must extend its RCU protection section to include the call to dev_net(), which can become dev_net_rcu(). This makes sure the net structure can not disappear under us. Fixes: 9b52e3f267a6 ("flow_dissector: handle no-skb use case") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-10-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/flow_dissector.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 0e638a37aa09..5db41bf2ed93 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -1108,10 +1108,12 @@ bool __skb_flow_dissect(const struct net *net, FLOW_DISSECTOR_KEY_BASIC, target_container); + rcu_read_lock(); + if (skb) { if (!net) { if (skb->dev) - net = dev_net(skb->dev); + net = dev_net_rcu(skb->dev); else if (skb->sk) net = sock_net(skb->sk); } @@ -1122,7 +1124,6 @@ bool __skb_flow_dissect(const struct net *net, enum netns_bpf_attach_type type = NETNS_BPF_FLOW_DISSECTOR; struct bpf_prog_array *run_array; - rcu_read_lock(); run_array = rcu_dereference(init_net.bpf.run_array[type]); if (!run_array) run_array = rcu_dereference(net->bpf.run_array[type]); @@ -1150,17 +1151,17 @@ bool __skb_flow_dissect(const struct net *net, prog = READ_ONCE(run_array->items[0].prog); result = bpf_flow_dissect(prog, &ctx, n_proto, nhoff, hlen, flags); - if (result == BPF_FLOW_DISSECTOR_CONTINUE) - goto dissect_continue; - __skb_flow_bpf_to_target(&flow_keys, flow_dissector, - target_container); - rcu_read_unlock(); - return result == BPF_OK; + if (result != BPF_FLOW_DISSECTOR_CONTINUE) { + __skb_flow_bpf_to_target(&flow_keys, flow_dissector, + target_container); + rcu_read_unlock(); + return result == BPF_OK; + } } -dissect_continue: - rcu_read_unlock(); } + rcu_read_unlock(); + if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { struct ethhdr *eth = eth_hdr(skb); From 3c8ffcd248da34fc41e52a46e51505900115fc2a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:18 +0000 Subject: [PATCH 0593/1443] ipv6: use RCU protection in ip6_default_advmss() ip6_default_advmss() needs rcu protection to make sure the net structure it reads does not disappear. Fixes: 5578689a4e3c ("[NETNS][IPV6] route6 - make route6 per namespace") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205155120.1676781-11-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 78362822b907..ef2d23a1e3d5 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -3196,13 +3196,18 @@ static unsigned int ip6_default_advmss(const struct dst_entry *dst) { struct net_device *dev = dst->dev; unsigned int mtu = dst_mtu(dst); - struct net *net = dev_net(dev); + struct net *net; mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr); + rcu_read_lock(); + + net = dev_net_rcu(dev); if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss) mtu = net->ipv6.sysctl.ip6_rt_min_advmss; + rcu_read_unlock(); + /* * Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and * corresponding MSS is IPV6_MAXPLEN - tcp_header_size. From 34aef2b0ce3aa4eb4ef2e1f5cad3738d527032f5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:19 +0000 Subject: [PATCH 0594/1443] ipv6: icmp: convert to dev_net_rcu() icmp6_send() must acquire rcu_read_lock() sooner to ensure the dev_net() call done from a safe context. Other ICMPv6 uses of dev_net() seem safe, change them to dev_net_rcu() to get LOCKDEP support to catch bugs. Fixes: 9a43b709a230 ("[NETNS][IPV6] icmp6 - make icmpv6_socket per namespace") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250205155120.1676781-12-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/icmp.c | 42 +++++++++++++++++++++++------------------- 1 file changed, 23 insertions(+), 19 deletions(-) diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index a6984a29fdb9..4d14ab7f7e99 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -76,7 +76,7 @@ static int icmpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, { /* icmpv6_notify checks 8 bytes can be pulled, icmp6hdr is 8 bytes */ struct icmp6hdr *icmp6 = (struct icmp6hdr *) (skb->data + offset); - struct net *net = dev_net(skb->dev); + struct net *net = dev_net_rcu(skb->dev); if (type == ICMPV6_PKT_TOOBIG) ip6_update_pmtu(skb, net, info, skb->dev->ifindex, 0, sock_net_uid(net, NULL)); @@ -473,7 +473,10 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, if (!skb->dev) return; - net = dev_net(skb->dev); + + rcu_read_lock(); + + net = dev_net_rcu(skb->dev); mark = IP6_REPLY_MARK(net, skb->mark); /* * Make sure we respect the rules @@ -496,7 +499,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, !(type == ICMPV6_PARAMPROB && code == ICMPV6_UNK_OPTION && (opt_unrec(skb, info)))) - return; + goto out; saddr = NULL; } @@ -526,7 +529,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, if ((addr_type == IPV6_ADDR_ANY) || (addr_type & IPV6_ADDR_MULTICAST)) { net_dbg_ratelimited("icmp6_send: addr_any/mcast source [%pI6c > %pI6c]\n", &hdr->saddr, &hdr->daddr); - return; + goto out; } /* @@ -535,7 +538,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, if (is_ineligible(skb)) { net_dbg_ratelimited("icmp6_send: no reply to icmp error [%pI6c > %pI6c]\n", &hdr->saddr, &hdr->daddr); - return; + goto out; } /* Needed by both icmpv6_global_allow and icmpv6_xmit_lock */ @@ -582,7 +585,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, np = inet6_sk(sk); if (!icmpv6_xrlim_allow(sk, type, &fl6, apply_ratelimit)) - goto out; + goto out_unlock; tmp_hdr.icmp6_type = type; tmp_hdr.icmp6_code = code; @@ -600,7 +603,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, dst = icmpv6_route_lookup(net, skb, sk, &fl6); if (IS_ERR(dst)) - goto out; + goto out_unlock; ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); @@ -616,7 +619,6 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, goto out_dst_release; } - rcu_read_lock(); idev = __in6_dev_get(skb->dev); if (ip6_append_data(sk, icmpv6_getfrag, &msg, @@ -630,13 +632,15 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, icmpv6_push_pending_frames(sk, &fl6, &tmp_hdr, len + sizeof(struct icmp6hdr)); } - rcu_read_unlock(); + out_dst_release: dst_release(dst); -out: +out_unlock: icmpv6_xmit_unlock(sk); out_bh_enable: local_bh_enable(); +out: + rcu_read_unlock(); } EXPORT_SYMBOL(icmp6_send); @@ -679,8 +683,8 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, skb_pull(skb2, nhs); skb_reset_network_header(skb2); - rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, NULL, 0, - skb, 0); + rt = rt6_lookup(dev_net_rcu(skb->dev), &ipv6_hdr(skb2)->saddr, + NULL, 0, skb, 0); if (rt && rt->dst.dev) skb2->dev = rt->dst.dev; @@ -717,7 +721,7 @@ EXPORT_SYMBOL(ip6_err_gen_icmpv6_unreach); static enum skb_drop_reason icmpv6_echo_reply(struct sk_buff *skb) { - struct net *net = dev_net(skb->dev); + struct net *net = dev_net_rcu(skb->dev); struct sock *sk; struct inet6_dev *idev; struct ipv6_pinfo *np; @@ -832,7 +836,7 @@ enum skb_drop_reason icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info) { struct inet6_skb_parm *opt = IP6CB(skb); - struct net *net = dev_net(skb->dev); + struct net *net = dev_net_rcu(skb->dev); const struct inet6_protocol *ipprot; enum skb_drop_reason reason; int inner_offset; @@ -889,7 +893,7 @@ enum skb_drop_reason icmpv6_notify(struct sk_buff *skb, u8 type, static int icmpv6_rcv(struct sk_buff *skb) { enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED; - struct net *net = dev_net(skb->dev); + struct net *net = dev_net_rcu(skb->dev); struct net_device *dev = icmp6_dev(skb); struct inet6_dev *idev = __in6_dev_get(dev); const struct in6_addr *saddr, *daddr; @@ -921,7 +925,7 @@ static int icmpv6_rcv(struct sk_buff *skb) skb_set_network_header(skb, nh); } - __ICMP6_INC_STATS(dev_net(dev), idev, ICMP6_MIB_INMSGS); + __ICMP6_INC_STATS(dev_net_rcu(dev), idev, ICMP6_MIB_INMSGS); saddr = &ipv6_hdr(skb)->saddr; daddr = &ipv6_hdr(skb)->daddr; @@ -939,7 +943,7 @@ static int icmpv6_rcv(struct sk_buff *skb) type = hdr->icmp6_type; - ICMP6MSGIN_INC_STATS(dev_net(dev), idev, type); + ICMP6MSGIN_INC_STATS(dev_net_rcu(dev), idev, type); switch (type) { case ICMPV6_ECHO_REQUEST: @@ -1034,9 +1038,9 @@ static int icmpv6_rcv(struct sk_buff *skb) csum_error: reason = SKB_DROP_REASON_ICMP_CSUM; - __ICMP6_INC_STATS(dev_net(dev), idev, ICMP6_MIB_CSUMERRORS); + __ICMP6_INC_STATS(dev_net_rcu(dev), idev, ICMP6_MIB_CSUMERRORS); discard_it: - __ICMP6_INC_STATS(dev_net(dev), idev, ICMP6_MIB_INERRORS); + __ICMP6_INC_STATS(dev_net_rcu(dev), idev, ICMP6_MIB_INERRORS); drop_no_count: kfree_skb_reason(skb, reason); return 0; From b768294d449da6d7dc0667c1ec92dc4af6ef766b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Feb 2025 15:51:20 +0000 Subject: [PATCH 0595/1443] ipv6: Use RCU in ip6_input() Instead of grabbing rcu_read_lock() from ip6_input_finish(), do it earlier in is caller, so that ip6_input() access to dev_net() can be validated by LOCKDEP. Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20250205155120.1676781-13-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_input.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 70c0e16c0ae6..39da6a7ce5f1 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -477,9 +477,7 @@ void ip6_protocol_deliver_rcu(struct net *net, struct sk_buff *skb, int nexthdr, static int ip6_input_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { skb_clear_delivery_time(skb); - rcu_read_lock(); ip6_protocol_deliver_rcu(net, skb, 0, false); - rcu_read_unlock(); return 0; } @@ -487,9 +485,15 @@ static int ip6_input_finish(struct net *net, struct sock *sk, struct sk_buff *sk int ip6_input(struct sk_buff *skb) { - return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN, - dev_net(skb->dev), NULL, skb, skb->dev, NULL, - ip6_input_finish); + int res; + + rcu_read_lock(); + res = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_IN, + dev_net_rcu(skb->dev), NULL, skb, skb->dev, NULL, + ip6_input_finish); + rcu_read_unlock(); + + return res; } EXPORT_SYMBOL_GPL(ip6_input); From 6a774228e890ee04a0ee13f4e6e731ec8554b9c2 Mon Sep 17 00:00:00 2001 From: Kory Maincent Date: Wed, 5 Feb 2025 12:03:01 +0100 Subject: [PATCH 0596/1443] net: ethtool: tsconfig: Fix netlink type of hwtstamp flags Fix the netlink type for hardware timestamp flags, which are represented as a bitset of flags. Although only one flag is supported currently, the correct netlink bitset type should be used instead of u32 to keep consistency with other fields. Address this by adding a new named string set description for the hwtstamp flag structure. The code has been introduced in the current release so the uAPI change is still okay. Signed-off-by: Kory Maincent Fixes: 6e9e2eed4f39 ("net: ethtool: Add support for tsconfig command to get/set hwtstamp config") Link: https://patch.msgid.link/20250205110304.375086-1-kory.maincent@bootlin.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/ethtool.yaml | 3 ++- include/uapi/linux/ethtool.h | 2 ++ net/ethtool/common.c | 5 ++++ net/ethtool/common.h | 2 ++ net/ethtool/strset.c | 5 ++++ net/ethtool/tsconfig.c | 33 +++++++++++++++++------- 6 files changed, 39 insertions(+), 11 deletions(-) diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index 259cb211a338..655d8d10fe24 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -1524,7 +1524,8 @@ attribute-sets: nested-attributes: bitset - name: hwtstamp-flags - type: u32 + type: nest + nested-attributes: bitset operations: enum-model: directional diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index d1089b88efc7..9b18c4cfe56f 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -682,6 +682,7 @@ enum ethtool_link_ext_substate_module { * @ETH_SS_STATS_ETH_CTRL: names of IEEE 802.3 MAC Control statistics * @ETH_SS_STATS_RMON: names of RMON statistics * @ETH_SS_STATS_PHY: names of PHY(dev) statistics + * @ETH_SS_TS_FLAGS: hardware timestamping flags * * @ETH_SS_COUNT: number of defined string sets */ @@ -708,6 +709,7 @@ enum ethtool_stringset { ETH_SS_STATS_ETH_CTRL, ETH_SS_STATS_RMON, ETH_SS_STATS_PHY, + ETH_SS_TS_FLAGS, /* add new constants above here */ ETH_SS_COUNT diff --git a/net/ethtool/common.c b/net/ethtool/common.c index 2bd77c94f9f1..d88e9080643b 100644 --- a/net/ethtool/common.c +++ b/net/ethtool/common.c @@ -462,6 +462,11 @@ const char ts_rx_filter_names[][ETH_GSTRING_LEN] = { }; static_assert(ARRAY_SIZE(ts_rx_filter_names) == __HWTSTAMP_FILTER_CNT); +const char ts_flags_names[][ETH_GSTRING_LEN] = { + [const_ilog2(HWTSTAMP_FLAG_BONDED_PHC_INDEX)] = "bonded-phc-index", +}; +static_assert(ARRAY_SIZE(ts_flags_names) == __HWTSTAMP_FLAG_CNT); + const char udp_tunnel_type_names[][ETH_GSTRING_LEN] = { [ETHTOOL_UDP_TUNNEL_TYPE_VXLAN] = "vxlan", [ETHTOOL_UDP_TUNNEL_TYPE_GENEVE] = "geneve", diff --git a/net/ethtool/common.h b/net/ethtool/common.h index 850eadde4bfc..58e9e7db06f9 100644 --- a/net/ethtool/common.h +++ b/net/ethtool/common.h @@ -13,6 +13,7 @@ ETHTOOL_LINK_MODE_ ## speed ## base ## type ## _ ## duplex ## _BIT #define __SOF_TIMESTAMPING_CNT (const_ilog2(SOF_TIMESTAMPING_LAST) + 1) +#define __HWTSTAMP_FLAG_CNT (const_ilog2(HWTSTAMP_FLAG_LAST) + 1) struct link_mode_info { int speed; @@ -38,6 +39,7 @@ extern const char wol_mode_names[][ETH_GSTRING_LEN]; extern const char sof_timestamping_names[][ETH_GSTRING_LEN]; extern const char ts_tx_type_names[][ETH_GSTRING_LEN]; extern const char ts_rx_filter_names[][ETH_GSTRING_LEN]; +extern const char ts_flags_names[][ETH_GSTRING_LEN]; extern const char udp_tunnel_type_names[][ETH_GSTRING_LEN]; int __ethtool_get_link(struct net_device *dev); diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c index 818cf01f0911..6b76c05caba4 100644 --- a/net/ethtool/strset.c +++ b/net/ethtool/strset.c @@ -75,6 +75,11 @@ static const struct strset_info info_template[] = { .count = __HWTSTAMP_FILTER_CNT, .strings = ts_rx_filter_names, }, + [ETH_SS_TS_FLAGS] = { + .per_dev = false, + .count = __HWTSTAMP_FLAG_CNT, + .strings = ts_flags_names, + }, [ETH_SS_UDP_TUNNEL_TYPES] = { .per_dev = false, .count = __ETHTOOL_UDP_TUNNEL_TYPE_CNT, diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c index 9188e088fb2f..2be356bdfe87 100644 --- a/net/ethtool/tsconfig.c +++ b/net/ethtool/tsconfig.c @@ -54,7 +54,7 @@ static int tsconfig_prepare_data(const struct ethnl_req_info *req_base, data->hwtst_config.tx_type = BIT(cfg.tx_type); data->hwtst_config.rx_filter = BIT(cfg.rx_filter); - data->hwtst_config.flags = BIT(cfg.flags); + data->hwtst_config.flags = cfg.flags; data->hwprov_desc.index = -1; hwprov = rtnl_dereference(dev->hwprov); @@ -91,10 +91,16 @@ static int tsconfig_reply_size(const struct ethnl_req_info *req_base, BUILD_BUG_ON(__HWTSTAMP_TX_CNT > 32); BUILD_BUG_ON(__HWTSTAMP_FILTER_CNT > 32); + BUILD_BUG_ON(__HWTSTAMP_FLAG_CNT > 32); - if (data->hwtst_config.flags) - /* _TSCONFIG_HWTSTAMP_FLAGS */ - len += nla_total_size(sizeof(u32)); + if (data->hwtst_config.flags) { + ret = ethnl_bitset32_size(&data->hwtst_config.flags, + NULL, __HWTSTAMP_FLAG_CNT, + ts_flags_names, compact); + if (ret < 0) + return ret; + len += ret; /* _TSCONFIG_HWTSTAMP_FLAGS */ + } if (data->hwtst_config.tx_type) { ret = ethnl_bitset32_size(&data->hwtst_config.tx_type, @@ -130,8 +136,10 @@ static int tsconfig_fill_reply(struct sk_buff *skb, int ret; if (data->hwtst_config.flags) { - ret = nla_put_u32(skb, ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS, - data->hwtst_config.flags); + ret = ethnl_put_bitset32(skb, ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS, + &data->hwtst_config.flags, NULL, + __HWTSTAMP_FLAG_CNT, + ts_flags_names, compact); if (ret < 0) return ret; } @@ -180,7 +188,7 @@ const struct nla_policy ethnl_tsconfig_set_policy[ETHTOOL_A_TSCONFIG_MAX + 1] = [ETHTOOL_A_TSCONFIG_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy), [ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER] = NLA_POLICY_NESTED(ethnl_ts_hwtst_prov_policy), - [ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS] = { .type = NLA_U32 }, + [ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS] = { .type = NLA_NESTED }, [ETHTOOL_A_TSCONFIG_RX_FILTERS] = { .type = NLA_NESTED }, [ETHTOOL_A_TSCONFIG_TX_TYPES] = { .type = NLA_NESTED }, }; @@ -296,6 +304,7 @@ static int ethnl_set_tsconfig(struct ethnl_req_info *req_base, BUILD_BUG_ON(__HWTSTAMP_TX_CNT >= 32); BUILD_BUG_ON(__HWTSTAMP_FILTER_CNT >= 32); + BUILD_BUG_ON(__HWTSTAMP_FLAG_CNT > 32); if (!netif_device_present(dev)) return -ENODEV; @@ -377,9 +386,13 @@ static int ethnl_set_tsconfig(struct ethnl_req_info *req_base, } if (tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]) { - ethnl_update_u32(&hwtst_config.flags, - tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS], - &config_mod); + ret = ethnl_update_bitset32(&hwtst_config.flags, + __HWTSTAMP_FLAG_CNT, + tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS], + ts_flags_names, info->extack, + &config_mod); + if (ret < 0) + goto err_free_hwprov; } ret = net_hwtstamp_validate(&hwtst_config); From bca0902e61731a75fc4860c8720168d9f1bae3b6 Mon Sep 17 00:00:00 2001 From: Murad Masimov Date: Mon, 3 Feb 2025 12:12:03 +0300 Subject: [PATCH 0597/1443] ax25: Fix refcount leak caused by setting SO_BINDTODEVICE sockopt If an AX25 device is bound to a socket by setting the SO_BINDTODEVICE socket option, a refcount leak will occur in ax25_release(). Commit 9fd75b66b8f6 ("ax25: Fix refcount leaks caused by ax25_cb_del()") added decrement of device refcounts in ax25_release(). In order for that to work correctly the refcounts must already be incremented when the device is bound to the socket. An AX25 device can be bound to a socket by either calling ax25_bind() or setting SO_BINDTODEVICE socket option. In both cases the refcounts should be incremented, but in fact it is done only in ax25_bind(). This bug leads to the following issue reported by Syzkaller: ================================================================ refcount_t: decrement hit 0; leaking memory. WARNING: CPU: 1 PID: 5932 at lib/refcount.c:31 refcount_warn_saturate+0x1ed/0x210 lib/refcount.c:31 Modules linked in: CPU: 1 UID: 0 PID: 5932 Comm: syz-executor424 Not tainted 6.13.0-rc4-syzkaller-00110-g4099a71718b0 #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 RIP: 0010:refcount_warn_saturate+0x1ed/0x210 lib/refcount.c:31 Call Trace: __refcount_dec include/linux/refcount.h:336 [inline] refcount_dec include/linux/refcount.h:351 [inline] ref_tracker_free+0x710/0x820 lib/ref_tracker.c:236 netdev_tracker_free include/linux/netdevice.h:4156 [inline] netdev_put include/linux/netdevice.h:4173 [inline] netdev_put include/linux/netdevice.h:4169 [inline] ax25_release+0x33f/0xa10 net/ax25/af_ax25.c:1069 __sock_release+0xb0/0x270 net/socket.c:640 sock_close+0x1c/0x30 net/socket.c:1408 ... do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f ... ================================================================ Fix the implementation of ax25_setsockopt() by adding increment of refcounts for the new device bound, and decrement of refcounts for the old unbound device. Fixes: 9fd75b66b8f6 ("ax25: Fix refcount leaks caused by ax25_cb_del()") Reported-by: syzbot+33841dc6aa3e1d86b78a@syzkaller.appspotmail.com Signed-off-by: Murad Masimov Link: https://patch.msgid.link/20250203091203.1744-1-m.masimov@mt-integration.ru Signed-off-by: Jakub Kicinski --- net/ax25/af_ax25.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index aa6c714892ec..9f3b8b682adb 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -685,6 +685,15 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname, break; } + if (ax25->ax25_dev) { + if (dev == ax25->ax25_dev->dev) { + rcu_read_unlock(); + break; + } + netdev_put(ax25->ax25_dev->dev, &ax25->dev_tracker); + ax25_dev_put(ax25->ax25_dev); + } + ax25->ax25_dev = ax25_dev_ax25dev(dev); if (!ax25->ax25_dev) { rcu_read_unlock(); @@ -692,6 +701,8 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname, break; } ax25_fillin_cb(ax25, ax25->ax25_dev); + netdev_hold(dev, &ax25->dev_tracker, GFP_ATOMIC); + ax25_dev_hold(ax25->ax25_dev); rcu_read_unlock(); break; From 1438f5d07b9a7afb15e1d0e26df04a6fd4e56a3c Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Wed, 5 Feb 2025 23:10:37 +0100 Subject: [PATCH 0598/1443] rtnetlink: fix netns leak with rtnl_setlink() A call to rtnl_nets_destroy() is needed to release references taken on netns put in rtnl_nets. CC: stable@vger.kernel.org Fixes: 636af13f213b ("rtnetlink: Register rtnl_dellink() and rtnl_setlink() with RTNL_FLAG_DOIT_PERNET_WIP.") Signed-off-by: Nicolas Dichtel Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250205221037.2474426-1-nicolas.dichtel@6wind.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 1f4d4b5570ab..d1e559fce918 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3432,6 +3432,7 @@ static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, err = -ENODEV; rtnl_nets_unlock(&rtnl_nets); + rtnl_nets_destroy(&rtnl_nets); errout: return err; } From cb7380de9e4cbc9a24216b722ec50e092ae83036 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 5 Feb 2025 12:32:49 -0800 Subject: [PATCH 0599/1443] compiler.h: Move C string helpers into C-only kernel section The C kernel helpers for evaluating C Strings were positioned where they were visible to assembly inclusion, which was not intended. Move them into the kernel and C-only area of the header so future changes won't confuse the assembler. Fixes: d7a516c6eeae ("compiler.h: Fix undefined BUILD_BUG_ON_ZERO()") Fixes: 559048d156ff ("string: Check for "nonstring" attribute on strscpy() arguments") Reviewed-by: Miguel Ojeda Signed-off-by: Kees Cook --- include/linux/compiler.h | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 240c632c5b95..7af999a131cb 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -214,6 +214,19 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, __v; \ }) +#ifdef __CHECKER__ +#define __BUILD_BUG_ON_ZERO_MSG(e, msg) (0) +#else /* __CHECKER__ */ +#define __BUILD_BUG_ON_ZERO_MSG(e, msg) ((int)sizeof(struct {_Static_assert(!(e), msg);})) +#endif /* __CHECKER__ */ + +/* &a[0] degrades to a pointer: a different type from an array */ +#define __must_be_array(a) __BUILD_BUG_ON_ZERO_MSG(__same_type((a), &(a)[0]), "must be array") + +/* Require C Strings (i.e. NUL-terminated) lack the "nonstring" attribute. */ +#define __must_be_cstr(p) \ + __BUILD_BUG_ON_ZERO_MSG(__annotated(p, nonstring), "must be cstr (NUL-terminated)") + #endif /* __KERNEL__ */ /** @@ -254,19 +267,6 @@ static inline void *offset_to_ptr(const int *off) #define __ADDRESSABLE_ASM_STR(sym) __stringify(__ADDRESSABLE_ASM(sym)) -#ifdef __CHECKER__ -#define __BUILD_BUG_ON_ZERO_MSG(e, msg) (0) -#else /* __CHECKER__ */ -#define __BUILD_BUG_ON_ZERO_MSG(e, msg) ((int)sizeof(struct {_Static_assert(!(e), msg);})) -#endif /* __CHECKER__ */ - -/* &a[0] degrades to a pointer: a different type from an array */ -#define __must_be_array(a) __BUILD_BUG_ON_ZERO_MSG(__same_type((a), &(a)[0]), "must be array") - -/* Require C Strings (i.e. NUL-terminated) lack the "nonstring" attribute. */ -#define __must_be_cstr(p) \ - __BUILD_BUG_ON_ZERO_MSG(__annotated(p, nonstring), "must be cstr (NUL-terminated)") - /* * This returns a constant expression while determining if an argument is * a constant expression, most importantly without evaluating the argument. From 20e5cc26e56db09cc612721f90b4994cce5e5b7b Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 5 Feb 2025 12:48:07 -0800 Subject: [PATCH 0600/1443] compiler.h: Introduce __must_be_byte_array() In preparation for adding stricter type checking to the str/mem*() helpers, provide a way to check that a variable is a byte array via __must_be_byte_array(). Suggested-by: Kent Overstreet Signed-off-by: Kees Cook --- include/linux/compiler.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 7af999a131cb..1c0688319435 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -221,7 +221,13 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, #endif /* __CHECKER__ */ /* &a[0] degrades to a pointer: a different type from an array */ -#define __must_be_array(a) __BUILD_BUG_ON_ZERO_MSG(__same_type((a), &(a)[0]), "must be array") +#define __is_array(a) (!__same_type((a), &(a)[0])) +#define __must_be_array(a) __BUILD_BUG_ON_ZERO_MSG(!__is_array(a), \ + "must be array") + +#define __is_byte_array(a) (__is_array(a) && sizeof((a)[0]) == 1) +#define __must_be_byte_array(a) __BUILD_BUG_ON_ZERO_MSG(!__is_byte_array(a), \ + "must be byte array") /* Require C Strings (i.e. NUL-terminated) lack the "nonstring" attribute. */ #define __must_be_cstr(p) \ From 6270f4deba3fbd77d1717fb8634f1fc612ff69e2 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 5 Feb 2025 13:45:26 -0800 Subject: [PATCH 0601/1443] string.h: Use ARRAY_SIZE() for memtostr*()/strtomem*() The destination argument of memtostr*() and strtomem*() must be a fixed-size char array at compile time, so there is no need to use __builtin_object_size() (which is useful for when an argument is either a pointer or unknown). Instead use ARRAY_SIZE(), which has the benefit of working around a bug in Clang (fixed[1] in 15+) that got __builtin_object_size() wrong sometimes. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501310832.kiAeOt2z-lkp@intel.com/ Suggested-by: Kent Overstreet Link: https://github.com/llvm/llvm-project/commit/d8e0a6d5e9dd2311641f9a8a5d2bf90829951ddc [1] Tested-by: Suren Baghdasaryan Signed-off-by: Kees Cook --- include/linux/string.h | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/include/linux/string.h b/include/linux/string.h index 493ac4862c77..fc5ae145bd78 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -411,7 +411,8 @@ void memcpy_and_pad(void *dest, size_t dest_len, const void *src, size_t count, * must be discoverable by the compiler. */ #define strtomem_pad(dest, src, pad) do { \ - const size_t _dest_len = __builtin_object_size(dest, 1); \ + const size_t _dest_len = __must_be_byte_array(dest) + \ + ARRAY_SIZE(dest); \ const size_t _src_len = __builtin_object_size(src, 1); \ \ BUILD_BUG_ON(!__builtin_constant_p(_dest_len) || \ @@ -434,7 +435,8 @@ void memcpy_and_pad(void *dest, size_t dest_len, const void *src, size_t count, * must be discoverable by the compiler. */ #define strtomem(dest, src) do { \ - const size_t _dest_len = __builtin_object_size(dest, 1); \ + const size_t _dest_len = __must_be_byte_array(dest) + \ + ARRAY_SIZE(dest); \ const size_t _src_len = __builtin_object_size(src, 1); \ \ BUILD_BUG_ON(!__builtin_constant_p(_dest_len) || \ @@ -453,7 +455,8 @@ void memcpy_and_pad(void *dest, size_t dest_len, const void *src, size_t count, * Note that sizes of @dest and @src must be known at compile-time. */ #define memtostr(dest, src) do { \ - const size_t _dest_len = __builtin_object_size(dest, 1); \ + const size_t _dest_len = __must_be_byte_array(dest) + \ + ARRAY_SIZE(dest); \ const size_t _src_len = __builtin_object_size(src, 1); \ const size_t _src_chars = strnlen(src, _src_len); \ const size_t _copy_len = min(_dest_len - 1, _src_chars); \ @@ -478,7 +481,8 @@ void memcpy_and_pad(void *dest, size_t dest_len, const void *src, size_t count, * Note that sizes of @dest and @src must be known at compile-time. */ #define memtostr_pad(dest, src) do { \ - const size_t _dest_len = __builtin_object_size(dest, 1); \ + const size_t _dest_len = __must_be_byte_array(dest) + \ + ARRAY_SIZE(dest); \ const size_t _src_len = __builtin_object_size(src, 1); \ const size_t _src_chars = strnlen(src, _src_len); \ const size_t _copy_len = min(_dest_len - 1, _src_chars); \ From fdfd0ad82890f678398ee670c4e59747738540e7 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 1 Feb 2025 12:56:51 -0500 Subject: [PATCH 0602/1443] bcachefs docs: SubmittingPatches.rst Add an (initial?) patch submission checklist, focusing mainly on testing. Yes, all patches must be tested, and that starts (but does not end) with the patch author. Signed-off-by: Kent Overstreet --- .../bcachefs/SubmittingPatches.rst | 98 +++++++++++++++++++ Documentation/filesystems/bcachefs/index.rst | 1 + MAINTAINERS | 1 + 3 files changed, 100 insertions(+) create mode 100644 Documentation/filesystems/bcachefs/SubmittingPatches.rst diff --git a/Documentation/filesystems/bcachefs/SubmittingPatches.rst b/Documentation/filesystems/bcachefs/SubmittingPatches.rst new file mode 100644 index 000000000000..026b12ae0d6a --- /dev/null +++ b/Documentation/filesystems/bcachefs/SubmittingPatches.rst @@ -0,0 +1,98 @@ +Submitting patches to bcachefs: +=============================== + +Patches must be tested before being submitted, either with the xfstests suite +[0], or the full bcachefs test suite in ktest [1], depending on what's being +touched. Note that ktest wraps xfstests and will be an easier method to running +it for most users; it includes single-command wrappers for all the mainstream +in-kernel local filesystems. + +Patches will undergo more testing after being merged (including +lockdep/kasan/preempt/etc. variants), these are not generally required to be +run by the submitter - but do put some thought into what you're changing and +which tests might be relevant, e.g. are you dealing with tricky memory layout +work? kasan, are you doing locking work? then lockdep; and ktest includes +single-command variants for the debug build types you'll most likely need. + +The exception to this rule is incomplete WIP/RFC patches: if you're working on +something nontrivial, it's encouraged to send out a WIP patch to let people +know what you're doing and make sure you're on the right track. Just make sure +it includes a brief note as to what's done and what's incomplete, to avoid +confusion. + +Rigorous checkpatch.pl adherence is not required (many of its warnings are +considered out of date), but try not to deviate too much without reason. + +Focus on writing code that reads well and is organized well; code should be +aesthetically pleasing. + +CI: +=== + +Instead of running your tests locally, when running the full test suite it's +prefereable to let a server farm do it in parallel, and then have the results +in a nice test dashboard (which can tell you which failures are new, and +presents results in a git log view, avoiding the need for most bisecting). + +That exists [2], and community members may request an account. If you work for +a big tech company, you'll need to help out with server costs to get access - +but the CI is not restricted to running bcachefs tests: it runs any ktest test +(which generally makes it easy to wrap other tests that can run in qemu). + +Other things to think about: +============================ + +- How will we debug this code? Is there sufficient introspection to diagnose + when something starts acting wonky on a user machine? + + We don't necessarily need every single field of every data structure visible + with introspection, but having the important fields of all the core data + types wired up makes debugging drastically easier - a bit of thoughtful + foresight greatly reduces the need to have people build custom kernels with + debug patches. + + More broadly, think about all the debug tooling that might be needed. + +- Does it make the codebase more or less of a mess? Can we also try to do some + organizing, too? + +- Do new tests need to be written? New assertions? How do we know and verify + that the code is correct, and what happens if something goes wrong? + + We don't yet have automated code coverage analysis or easy fault injection - + but for now, pretend we did and ask what they might tell us. + + Assertions are hugely important, given that we don't yet have a systems + language that can do ergonomic embedded correctness proofs. Hitting an assert + in testing is much better than wandering off into undefined behaviour la-la + land - use them. Use them judiciously, and not as a replacement for proper + error handling, but use them. + +- Does it need to be performance tested? Should we add new peformance counters? + + bcachefs has a set of persistent runtime counters which can be viewed with + the 'bcachefs fs top' command; this should give users a basic idea of what + their filesystem is currently doing. If you're doing a new feature or looking + at old code, think if anything should be added. + +- If it's a new on disk format feature - have upgrades and downgrades been + tested? (Automated tests exists but aren't in the CI, due to the hassle of + disk image management; coordinate to have them run.) + +Mailing list, IRC: +================== + +Patches should hit the list [3], but much discussion and code review happens on +IRC as well [4]; many people appreciate the more conversational approach and +quicker feedback. + +Additionally, we have a lively user community doing excellent QA work, which +exists primarily on IRC. Please make use of that resource; user feedback is +important for any nontrivial feature, and documenting it in commit messages +would be a good idea. + +[0]: git://git.kernel.org/pub/scm/fs/xfs/xfstests-dev.git +[1]: https://evilpiepirate.org/git/ktest.git/ +[2]: https://evilpiepirate.org/~testdashboard/ci/ +[3]: linux-bcachefs@vger.kernel.org +[4]: irc.oftc.net#bcache, #bcachefs-dev diff --git a/Documentation/filesystems/bcachefs/index.rst b/Documentation/filesystems/bcachefs/index.rst index 95fc4b90739e..7db4d7ceab58 100644 --- a/Documentation/filesystems/bcachefs/index.rst +++ b/Documentation/filesystems/bcachefs/index.rst @@ -9,4 +9,5 @@ bcachefs Documentation :numbered: CodingStyle + SubmittingPatches errorcodes diff --git a/MAINTAINERS b/MAINTAINERS index 896a307fa065..c40d3d0c68c7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3955,6 +3955,7 @@ M: Kent Overstreet L: linux-bcachefs@vger.kernel.org S: Supported C: irc://irc.oftc.net/bcache +P: Documentation/filesystems/bcachefs/SubmittingPatches.rst T: git https://evilpiepirate.org/git/bcachefs.git F: fs/bcachefs/ F: Documentation/filesystems/bcachefs/ From 6b37037d6d1b42083642340efcf80f7a30203039 Mon Sep 17 00:00:00 2001 From: Jeongjun Park Date: Sat, 1 Feb 2025 01:20:31 +0900 Subject: [PATCH 0603/1443] bcachefs: fix incorrect pointer check in __bch2_subvolume_delete() For some unknown reason, checks on struct bkey_s_c_snapshot and struct bkey_s_c_snapshot_tree pointers are missing. Therefore, I think it would be appropriate to fix the incorrect pointer checking through this patch. Fixes: 4bd06f07bcb5 ("bcachefs: Fixes for snapshot_tree.master_subvol") Signed-off-by: Jeongjun Park Signed-off-by: Kent Overstreet --- fs/bcachefs/subvolume.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c index e3d0475232e5..b7b96283c316 100644 --- a/fs/bcachefs/subvolume.c +++ b/fs/bcachefs/subvolume.c @@ -428,7 +428,7 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) bch2_bkey_get_iter_typed(trans, &snapshot_iter, BTREE_ID_snapshots, POS(0, snapid), 0, snapshot); - ret = bkey_err(subvol); + ret = bkey_err(snapshot); bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, "missing snapshot %u", snapid); if (ret) @@ -440,6 +440,11 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid) bch2_bkey_get_iter_typed(trans, &snapshot_tree_iter, BTREE_ID_snapshot_trees, POS(0, treeid), 0, snapshot_tree); + ret = bkey_err(snapshot_tree); + bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c, + "missing snapshot tree %u", treeid); + if (ret) + goto err; if (le32_to_cpu(snapshot_tree.v->master_subvol) == subvolid) { struct bkey_i_snapshot_tree *snapshot_tree_mut = From 2ef995df0ce592f665d312008dbe1ad1c4bcf87f Mon Sep 17 00:00:00 2001 From: Jeongjun Park Date: Sun, 2 Feb 2025 15:13:51 +0900 Subject: [PATCH 0604/1443] bcachefs: fix deadlock in journal_entry_open() In the previous commit b3d82c2f2761, code was added to prevent journal sequence overflow. Among them, the code added to journal_entry_open() uses the bch2_fs_fatal_err_on() function to handle errors. However, __journal_res_get() , which calls journal_entry_open() , calls journal_entry_open() while holding journal->lock , but bch2_fs_fatal_err_on() internally tries to acquire journal->lock , which results in a deadlock. So we need to add a locked helper to handle fatal errors even when the journal->lock is held. Fixes: b3d82c2f2761 ("bcachefs: Guard against journal seq overflow") Signed-off-by: Jeongjun Park Signed-off-by: Kent Overstreet --- fs/bcachefs/journal.c | 17 +++++++++++++++-- fs/bcachefs/journal.h | 1 + fs/bcachefs/super.c | 11 +++++++++++ fs/bcachefs/super.h | 1 + 4 files changed, 28 insertions(+), 2 deletions(-) diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index cb2c3722f674..0a943a27ef44 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -319,6 +319,16 @@ void bch2_journal_halt(struct journal *j) spin_unlock(&j->lock); } +void bch2_journal_halt_locked(struct journal *j) +{ + lockdep_assert_held(&j->lock); + + __journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL, true); + if (!j->err_seq) + j->err_seq = journal_cur_seq(j); + journal_wake(j); +} + static bool journal_entry_want_write(struct journal *j) { bool ret = !journal_entry_is_open(j) || @@ -381,9 +391,12 @@ static int journal_entry_open(struct journal *j) if (nr_unwritten_journal_entries(j) == ARRAY_SIZE(j->buf)) return JOURNAL_ERR_max_in_flight; - if (bch2_fs_fatal_err_on(journal_cur_seq(j) >= JOURNAL_SEQ_MAX, - c, "cannot start: journal seq overflow")) + if (journal_cur_seq(j) >= JOURNAL_SEQ_MAX) { + bch_err(c, "cannot start: journal seq overflow"); + if (bch2_fs_emergency_read_only_locked(c)) + bch_err(c, "fatal error - emergency read only"); return JOURNAL_ERR_insufficient_devices; /* -EROFS */ + } BUG_ON(!j->cur_entry_sectors); diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h index dccddd5420ad..107f7f901cd9 100644 --- a/fs/bcachefs/journal.h +++ b/fs/bcachefs/journal.h @@ -409,6 +409,7 @@ bool bch2_journal_noflush_seq(struct journal *, u64, u64); int bch2_journal_meta(struct journal *); void bch2_journal_halt(struct journal *); +void bch2_journal_halt_locked(struct journal *); static inline int bch2_journal_error(struct journal *j) { diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c index d97ea7bd1171..6d97d412fed9 100644 --- a/fs/bcachefs/super.c +++ b/fs/bcachefs/super.c @@ -411,6 +411,17 @@ bool bch2_fs_emergency_read_only(struct bch_fs *c) return ret; } +bool bch2_fs_emergency_read_only_locked(struct bch_fs *c) +{ + bool ret = !test_and_set_bit(BCH_FS_emergency_ro, &c->flags); + + bch2_journal_halt_locked(&c->journal); + bch2_fs_read_only_async(c); + + wake_up(&bch2_read_only_wait); + return ret; +} + static int bch2_fs_read_write_late(struct bch_fs *c) { int ret; diff --git a/fs/bcachefs/super.h b/fs/bcachefs/super.h index fa6d52216510..04f8287eff5c 100644 --- a/fs/bcachefs/super.h +++ b/fs/bcachefs/super.h @@ -29,6 +29,7 @@ int bch2_dev_resize(struct bch_fs *, struct bch_dev *, u64); struct bch_dev *bch2_dev_lookup(struct bch_fs *, const char *); bool bch2_fs_emergency_read_only(struct bch_fs *); +bool bch2_fs_emergency_read_only_locked(struct bch_fs *); void bch2_fs_read_only(struct bch_fs *); int bch2_fs_read_write(struct bch_fs *); From 9e9033522ad1e4bb697c9493aa449630fa2c98d2 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 27 Jan 2025 01:21:44 -0500 Subject: [PATCH 0605/1443] bcachefs: Fix discard path journal flushing The discard path is supposed to issue journal flushes when there's too many buckets empty buckets that need a journal commit before they can be written to again, but at some point this code seems to have been lost. Bring it back with a new optimization to make sure we don't issue too many journal flushes: the journal now tracks the sequence number of the most recent flush in progress, which the discard path uses when deciding which buckets need a journal flush. Signed-off-by: Kent Overstreet --- fs/bcachefs/alloc_background.c | 47 ++++++++++++----------- fs/bcachefs/alloc_foreground.c | 10 +++-- fs/bcachefs/alloc_types.h | 1 + fs/bcachefs/buckets_waiting_for_journal.c | 12 +++--- fs/bcachefs/buckets_waiting_for_journal.h | 4 +- fs/bcachefs/journal.c | 1 + fs/bcachefs/journal_types.h | 1 + fs/bcachefs/trace.h | 14 ++++++- 8 files changed, 55 insertions(+), 35 deletions(-) diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c index fc2ef33b67b3..3ea809990ef1 100644 --- a/fs/bcachefs/alloc_background.c +++ b/fs/bcachefs/alloc_background.c @@ -1803,7 +1803,6 @@ struct discard_buckets_state { u64 open; u64 need_journal_commit; u64 discarded; - u64 need_journal_commit_this_dev; }; static int bch2_discard_one_bucket(struct btree_trans *trans, @@ -1827,11 +1826,11 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, goto out; } - if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, - pos.inode, pos.offset)) { - s->need_journal_commit++; - s->need_journal_commit_this_dev++; + u64 seq_ready = bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal, + pos.inode, pos.offset); + if (seq_ready > c->journal.flushed_seq_ondisk) { + if (seq_ready > c->journal.flushing_seq) + s->need_journal_commit++; goto out; } @@ -1865,23 +1864,24 @@ static int bch2_discard_one_bucket(struct btree_trans *trans, discard_locked = true; } - if (!bkey_eq(*discard_pos_done, iter.pos) && - ca->mi.discard && !c->opts.nochanges) { - /* - * This works without any other locks because this is the only - * thread that removes items from the need_discard tree - */ - bch2_trans_unlock_long(trans); - blkdev_issue_discard(ca->disk_sb.bdev, - k.k->p.offset * ca->mi.bucket_size, - ca->mi.bucket_size, - GFP_KERNEL); - *discard_pos_done = iter.pos; + if (!bkey_eq(*discard_pos_done, iter.pos)) { s->discarded++; + *discard_pos_done = iter.pos; - ret = bch2_trans_relock_notrace(trans); - if (ret) - goto out; + if (ca->mi.discard && !c->opts.nochanges) { + /* + * This works without any other locks because this is the only + * thread that removes items from the need_discard tree + */ + bch2_trans_unlock_long(trans); + blkdev_issue_discard(ca->disk_sb.bdev, + k.k->p.offset * ca->mi.bucket_size, + ca->mi.bucket_size, + GFP_KERNEL); + ret = bch2_trans_relock_notrace(trans); + if (ret) + goto out; + } } SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false); @@ -1929,6 +1929,9 @@ static void bch2_do_discards_work(struct work_struct *work) POS(ca->dev_idx, U64_MAX), 0, k, bch2_discard_one_bucket(trans, ca, &iter, &discard_pos_done, &s, false))); + if (s.need_journal_commit > dev_buckets_available(ca, BCH_WATERMARK_normal)) + bch2_journal_flush_async(&c->journal, NULL); + trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); @@ -2024,7 +2027,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work) break; } - trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); + trace_discard_buckets_fast(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret)); bch2_trans_put(trans); percpu_ref_put(&ca->io_ref); diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c index 6df41c331a52..5a781fb4c794 100644 --- a/fs/bcachefs/alloc_foreground.c +++ b/fs/bcachefs/alloc_foreground.c @@ -205,8 +205,12 @@ static inline bool may_alloc_bucket(struct bch_fs *c, return false; } - if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal, - c->journal.flushed_seq_ondisk, bucket.inode, bucket.offset)) { + u64 journal_seq_ready = + bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal, + bucket.inode, bucket.offset); + if (journal_seq_ready > c->journal.flushed_seq_ondisk) { + if (journal_seq_ready > c->journal.flushing_seq) + s->need_journal_commit++; s->skipped_need_journal_commit++; return false; } @@ -570,7 +574,7 @@ static struct open_bucket *bch2_bucket_alloc_trans(struct btree_trans *trans, ? bch2_bucket_alloc_freelist(trans, ca, watermark, &s, cl) : bch2_bucket_alloc_early(trans, ca, watermark, &s, cl); - if (s.skipped_need_journal_commit * 2 > avail) + if (s.need_journal_commit * 2 > avail) bch2_journal_flush_async(&c->journal, NULL); if (!ob && s.btree_bitmap != BTREE_BITMAP_ANY) { diff --git a/fs/bcachefs/alloc_types.h b/fs/bcachefs/alloc_types.h index 9bbb28e90b93..4aa8ee026cb8 100644 --- a/fs/bcachefs/alloc_types.h +++ b/fs/bcachefs/alloc_types.h @@ -18,6 +18,7 @@ struct bucket_alloc_state { u64 buckets_seen; u64 skipped_open; u64 skipped_need_journal_commit; + u64 need_journal_commit; u64 skipped_nocow; u64 skipped_nouse; u64 skipped_mi_btree_bitmap; diff --git a/fs/bcachefs/buckets_waiting_for_journal.c b/fs/bcachefs/buckets_waiting_for_journal.c index f9fb150eda70..c8a488e6b7b8 100644 --- a/fs/bcachefs/buckets_waiting_for_journal.c +++ b/fs/bcachefs/buckets_waiting_for_journal.c @@ -22,23 +22,21 @@ static void bucket_table_init(struct buckets_waiting_for_journal_table *t, size_ memset(t->d, 0, sizeof(t->d[0]) << t->bits); } -bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *b, - u64 flushed_seq, - unsigned dev, u64 bucket) +u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *b, + unsigned dev, u64 bucket) { struct buckets_waiting_for_journal_table *t; u64 dev_bucket = (u64) dev << 56 | bucket; - bool ret = false; - unsigned i; + u64 ret = 0; mutex_lock(&b->lock); t = b->t; - for (i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) { + for (unsigned i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) { struct bucket_hashed *h = bucket_hash(t, i, dev_bucket); if (h->dev_bucket == dev_bucket) { - ret = h->journal_seq > flushed_seq; + ret = h->journal_seq; break; } } diff --git a/fs/bcachefs/buckets_waiting_for_journal.h b/fs/bcachefs/buckets_waiting_for_journal.h index d2ae19cbe18c..365619ca44c8 100644 --- a/fs/bcachefs/buckets_waiting_for_journal.h +++ b/fs/bcachefs/buckets_waiting_for_journal.h @@ -4,8 +4,8 @@ #include "buckets_waiting_for_journal_types.h" -bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *, - u64, unsigned, u64); +u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *, + unsigned, u64); int bch2_set_bucket_needs_journal_commit(struct buckets_waiting_for_journal *, u64, unsigned, u64, u64); diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c index 0a943a27ef44..24c294d4634e 100644 --- a/fs/bcachefs/journal.c +++ b/fs/bcachefs/journal.c @@ -796,6 +796,7 @@ int bch2_journal_flush_seq_async(struct journal *j, u64 seq, } buf->must_flush = true; + j->flushing_seq = max(j->flushing_seq, seq); if (parent && !closure_wait(&buf->wait, parent)) BUG(); diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h index 3ba433a48eb8..a198a81d7478 100644 --- a/fs/bcachefs/journal_types.h +++ b/fs/bcachefs/journal_types.h @@ -237,6 +237,7 @@ struct journal { /* seq, last_seq from the most recent journal entry successfully written */ u64 seq_ondisk; u64 flushed_seq_ondisk; + u64 flushing_seq; u64 last_seq_ondisk; u64 err_seq; u64 last_empty_seq; diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 56a5a7fbc0fd..c1b51009edf6 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -727,7 +727,7 @@ DEFINE_EVENT(fs_str, bucket_alloc_fail, TP_ARGS(c, str) ); -TRACE_EVENT(discard_buckets, +DECLARE_EVENT_CLASS(discard_buckets_class, TP_PROTO(struct bch_fs *c, u64 seen, u64 open, u64 need_journal_commit, u64 discarded, const char *err), TP_ARGS(c, seen, open, need_journal_commit, discarded, err), @@ -759,6 +759,18 @@ TRACE_EVENT(discard_buckets, __entry->err) ); +DEFINE_EVENT(discard_buckets_class, discard_buckets, + TP_PROTO(struct bch_fs *c, u64 seen, u64 open, + u64 need_journal_commit, u64 discarded, const char *err), + TP_ARGS(c, seen, open, need_journal_commit, discarded, err) +); + +DEFINE_EVENT(discard_buckets_class, discard_buckets_fast, + TP_PROTO(struct bch_fs *c, u64 seen, u64 open, + u64 need_journal_commit, u64 discarded, const char *err), + TP_ARGS(c, seen, open, need_journal_commit, discarded, err) +); + TRACE_EVENT(bucket_invalidate, TP_PROTO(struct bch_fs *c, unsigned dev, u64 bucket, u32 sectors), TP_ARGS(c, dev, bucket, sectors), From 3539880ef1a5f8c970d0f69a6fdcfeffc000e63d Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 3 Feb 2025 11:35:11 -0500 Subject: [PATCH 0606/1443] bcachefs: Fix rcu imbalance in bch2_fs_btree_key_cache_exit() Spotted by sparse. Signed-off-by: Kent Overstreet --- fs/bcachefs/btree_key_cache.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c index c378b97ebeca..1821f40c161a 100644 --- a/fs/bcachefs/btree_key_cache.c +++ b/fs/bcachefs/btree_key_cache.c @@ -748,7 +748,6 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc) rcu_read_unlock(); mutex_lock(&bc->table.mutex); mutex_unlock(&bc->table.mutex); - rcu_read_lock(); continue; } for (i = 0; i < tbl->size; i++) From 4be214c26936813b636eed2fac906f585ddbf0f9 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Sat, 25 Jan 2025 21:29:45 -0500 Subject: [PATCH 0607/1443] bcachefs: bch2_bkey_sectors_need_rebalance() now only depends on bch_extent_rebalance Previously, bch2_bkey_sectors_need_rebalance() called bch2_target_accepts_data(), checking whether the target is writable. However, this means that adding or removing devices from a target would change the value of bch2_bkey_sectors_need_rebalance() for an existing extent; this needs to be invariant so that the extent trigger can correctly maintain rebalance_work accounting. Instead, check target_accepts_data() in io_opts_to_rebalance_opts(), before creating the bch_extent_rebalance entry. This fixes (one?) cause of rebalance_work accounting being off. Signed-off-by: Kent Overstreet --- fs/bcachefs/inode.h | 4 +++- fs/bcachefs/opts.h | 14 -------------- fs/bcachefs/rebalance.c | 8 +++----- fs/bcachefs/rebalance.h | 20 ++++++++++++++++++++ 4 files changed, 26 insertions(+), 20 deletions(-) diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h index d2e134528f0e..428b9be6af34 100644 --- a/fs/bcachefs/inode.h +++ b/fs/bcachefs/inode.h @@ -285,12 +285,14 @@ void bch2_inode_opts_get(struct bch_io_opts *, struct bch_fs *, struct bch_inode_unpacked *); int bch2_inum_opts_get(struct btree_trans*, subvol_inum, struct bch_io_opts *); +#include "rebalance.h" + static inline struct bch_extent_rebalance bch2_inode_rebalance_opts_get(struct bch_fs *c, struct bch_inode_unpacked *inode) { struct bch_io_opts io_opts; bch2_inode_opts_get(&io_opts, c, inode); - return io_opts_to_rebalance_opts(&io_opts); + return io_opts_to_rebalance_opts(c, &io_opts); } int bch2_inode_rm_snapshot(struct btree_trans *, u64, u32); diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h index a182b5d454ba..9d397fc2a1f0 100644 --- a/fs/bcachefs/opts.h +++ b/fs/bcachefs/opts.h @@ -659,18 +659,4 @@ static inline void bch2_io_opts_fixups(struct bch_io_opts *opts) struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts); bool bch2_opt_is_inode_opt(enum bch_opt_id); -/* rebalance opts: */ - -static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_io_opts *opts) -{ - return (struct bch_extent_rebalance) { - .type = BIT(BCH_EXTENT_ENTRY_rebalance), -#define x(_name) \ - ._name = opts->_name, \ - ._name##_from_inode = opts->_name##_from_inode, - BCH_REBALANCE_OPTS() -#undef x - }; -}; - #endif /* _BCACHEFS_OPTS_H */ diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c index 4adc74cd3f70..d0a1f5cd5c2b 100644 --- a/fs/bcachefs/rebalance.c +++ b/fs/bcachefs/rebalance.c @@ -121,12 +121,10 @@ u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k) } } incompressible: - if (opts->background_target && - bch2_target_accepts_data(c, BCH_DATA_user, opts->background_target)) { + if (opts->background_target) bkey_for_each_ptr_decode(k.k, ptrs, p, entry) if (!p.ptr.cached && !bch2_dev_in_target(c, p.ptr.dev, opts->background_target)) sectors += p.crc.compressed_size; - } return sectors; } @@ -140,7 +138,7 @@ static bool bch2_bkey_rebalance_needs_update(struct bch_fs *c, struct bch_io_opt const struct bch_extent_rebalance *old = bch2_bkey_rebalance_opts(k); if (k.k->type == KEY_TYPE_reflink_v || bch2_bkey_ptrs_need_rebalance(c, opts, k)) { - struct bch_extent_rebalance new = io_opts_to_rebalance_opts(opts); + struct bch_extent_rebalance new = io_opts_to_rebalance_opts(c, opts); return old == NULL || memcmp(old, &new, sizeof(new)); } else { return old != NULL; @@ -163,7 +161,7 @@ int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bch_io_opts *opts, k.k->u64s += sizeof(*old) / sizeof(u64); } - *old = io_opts_to_rebalance_opts(opts); + *old = io_opts_to_rebalance_opts(c, opts); } else { if (old) extent_entry_drop(k, (union bch_extent_entry *) old); diff --git a/fs/bcachefs/rebalance.h b/fs/bcachefs/rebalance.h index 0a0821ab895d..62a3859d3823 100644 --- a/fs/bcachefs/rebalance.h +++ b/fs/bcachefs/rebalance.h @@ -4,8 +4,28 @@ #include "compress.h" #include "disk_groups.h" +#include "opts.h" #include "rebalance_types.h" +static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_fs *c, + struct bch_io_opts *opts) +{ + struct bch_extent_rebalance r = { + .type = BIT(BCH_EXTENT_ENTRY_rebalance), +#define x(_name) \ + ._name = opts->_name, \ + ._name##_from_inode = opts->_name##_from_inode, + BCH_REBALANCE_OPTS() +#undef x + }; + + if (r.background_target && + !bch2_target_accepts_data(c, BCH_DATA_user, r.background_target)) + r.background_target = 0; + + return r; +}; + u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *, struct bkey_s_c); int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bch_io_opts *, struct bkey_i *); int bch2_get_update_rebalance_opts(struct btree_trans *, From 29a61a1f40637ae010b828745fb41f60301c3a3d Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 5 Feb 2025 15:22:56 +0100 Subject: [PATCH 0608/1443] genirq: Remove leading space from irq_chip::irq_print_chip() callbacks The space separator was factored out from the multiple chip name prints, but several irq_chip::irq_print_chip() callbacks still print a leading space. Remove the superfluous double spaces. Fixes: 9d9f204bdf7243bf ("genirq/proc: Add missing space separator back") Signed-off-by: Geert Uytterhoeven Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/893f7e9646d8933cd6786d5a1ef3eb076d263768.1738764803.git.geert+renesas@glider.be --- arch/powerpc/sysdev/fsl_msi.c | 2 +- drivers/bus/moxtet.c | 2 +- drivers/irqchip/irq-partition-percpu.c | 2 +- drivers/soc/qcom/smp2p.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c index 1aa0cb097c9c..7b9a5ea9cad9 100644 --- a/arch/powerpc/sysdev/fsl_msi.c +++ b/arch/powerpc/sysdev/fsl_msi.c @@ -75,7 +75,7 @@ static void fsl_msi_print_chip(struct irq_data *irqd, struct seq_file *p) srs = (hwirq >> msi_data->srs_shift) & MSI_SRS_MASK; cascade_virq = msi_data->cascade_array[srs]->virq; - seq_printf(p, " fsl-msi-%d", cascade_virq); + seq_printf(p, "fsl-msi-%d", cascade_virq); } diff --git a/drivers/bus/moxtet.c b/drivers/bus/moxtet.c index 6276551d7968..1e57ebfb7622 100644 --- a/drivers/bus/moxtet.c +++ b/drivers/bus/moxtet.c @@ -657,7 +657,7 @@ static void moxtet_irq_print_chip(struct irq_data *d, struct seq_file *p) id = moxtet->modules[pos->idx]; - seq_printf(p, " moxtet-%s.%i#%i", mox_module_name(id), pos->idx, + seq_printf(p, "moxtet-%s.%i#%i", mox_module_name(id), pos->idx, pos->bit); } diff --git a/drivers/irqchip/irq-partition-percpu.c b/drivers/irqchip/irq-partition-percpu.c index 8e76d2913e6b..4441ffe149ea 100644 --- a/drivers/irqchip/irq-partition-percpu.c +++ b/drivers/irqchip/irq-partition-percpu.c @@ -98,7 +98,7 @@ static void partition_irq_print_chip(struct irq_data *d, struct seq_file *p) struct irq_chip *chip = irq_desc_get_chip(part->chained_desc); struct irq_data *data = irq_desc_get_irq_data(part->chained_desc); - seq_printf(p, " %5s-%lu", chip->name, data->hwirq); + seq_printf(p, "%5s-%lu", chip->name, data->hwirq); } static struct irq_chip partition_irq_chip = { diff --git a/drivers/soc/qcom/smp2p.c b/drivers/soc/qcom/smp2p.c index 4783ab1adb8d..a3e88ced328a 100644 --- a/drivers/soc/qcom/smp2p.c +++ b/drivers/soc/qcom/smp2p.c @@ -365,7 +365,7 @@ static void smp2p_irq_print_chip(struct irq_data *irqd, struct seq_file *p) { struct smp2p_entry *entry = irq_data_get_irq_chip_data(irqd); - seq_printf(p, " %8s", dev_name(entry->smp2p->dev)); + seq_printf(p, "%8s", dev_name(entry->smp2p->dev)); } static struct irq_chip smp2p_irq_chip = { From 868c9037df626b3c245ee26a290a03ae1f9f58d3 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 5 Feb 2025 17:02:20 +0100 Subject: [PATCH 0609/1443] timers/migration: Fix off-by-one root mis-connection Before attaching a new root to the old root, the children counter of the new root is checked to verify that only the upcoming CPU's top group have been connected to it. However since the recently added commit b729cc1ec21a ("timers/migration: Fix another race between hotplug and idle entry/exit") this check is not valid anymore because the old root is pre-accounted as a child to the new root. Therefore after connecting the upcoming CPU's top group to the new root, the children count to be expected must be 2 and not 1 anymore. This omission results in the old root to not be connected to the new root. Then eventually the system may run with more than one top level, which defeats the purpose of a single idle migrator. Also the old root is pre-accounted but not connected upon the new root creation. But it can be connected to the new root later on. Therefore the old root may be accounted twice to the new root. The propagation of such overcommit can end up creating a double final top-level root with a groupmask incorrectly initialized. Although harmless given that the final top level roots will never have a parent to walk up to, this oddity opportunistically reported the core issue: WARNING: CPU: 8 PID: 0 at kernel/time/timer_migration.c:543 tmigr_requires_handle_remote CPU: 8 UID: 0 PID: 0 Comm: swapper/8 RIP: 0010:tmigr_requires_handle_remote Call Trace: ? tmigr_requires_handle_remote ? hrtimer_run_queues update_process_times tick_periodic tick_handle_periodic __sysvec_apic_timer_interrupt sysvec_apic_timer_interrupt Fix the problem by taking the old root into account in the children count of the new root so the connection is not omitted. Also warn when more than one top level group exists to better detect similar issues in the future. Fixes: b729cc1ec21a ("timers/migration: Fix another race between hotplug and idle entry/exit") Reported-by: Matt Fleming Signed-off-by: Frederic Weisbecker Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20250205160220.39467-1-frederic@kernel.org --- kernel/time/timer_migration.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c index 9cb9b6584ea1..2f6330831f08 100644 --- a/kernel/time/timer_migration.c +++ b/kernel/time/timer_migration.c @@ -1675,6 +1675,9 @@ static int tmigr_setup_groups(unsigned int cpu, unsigned int node) } while (i < tmigr_hierarchy_levels); + /* Assert single root */ + WARN_ON_ONCE(!err && !group->parent && !list_is_singular(&tmigr_level_list[top])); + while (i > 0) { group = stack[--i]; @@ -1716,7 +1719,12 @@ static int tmigr_setup_groups(unsigned int cpu, unsigned int node) WARN_ON_ONCE(top == 0); lvllist = &tmigr_level_list[top]; - if (group->num_children == 1 && list_is_singular(lvllist)) { + + /* + * Newly created root level should have accounted the upcoming + * CPU's child group and pre-accounted the old root. + */ + if (group->num_children == 2 && list_is_singular(lvllist)) { /* * The target CPU must never do the prepare work, except * on early boot when the boot CPU is the target. Otherwise From 0fac3ed473dd2955053be6671cdd747807f5e488 Mon Sep 17 00:00:00 2001 From: Su Hui Date: Sun, 19 Jan 2025 10:59:47 +0800 Subject: [PATCH 0610/1443] fs/stat.c: avoid harmless garbage value problem in vfs_statx_path() Clang static checker(scan-build) warning: fs/stat.c:287:21: warning: The left expression of the compound assignment is an uninitialized value. The computed value will also be garbage. 287 | stat->result_mask |= STATX_MNT_ID_UNIQUE; | ~~~~~~~~~~~~~~~~~ ^ fs/stat.c:290:21: warning: The left expression of the compound assignment is an uninitialized value. The computed value will also be garbage. 290 | stat->result_mask |= STATX_MNT_ID; When vfs_getattr() failed because of security_inode_getattr(), 'stat' is uninitialized. In this case, there is a harmless garbage problem in vfs_statx_path(). It's better to return error directly when vfs_getattr() failed, avoiding garbage value and more clearly. Signed-off-by: Su Hui Link: https://lore.kernel.org/r/20250119025946.1168957-1-suhui@nfschina.com Signed-off-by: Christian Brauner --- fs/stat.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/stat.c b/fs/stat.c index 2c0e111a098a..f13308bfdc98 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -281,6 +281,8 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat, u32 request_mask) { int error = vfs_getattr(path, stat, request_mask, flags); + if (error) + return error; if (request_mask & STATX_MNT_ID_UNIQUE) { stat->mnt_id = real_mount(path->mnt)->mnt_id_unique; @@ -302,7 +304,7 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat, if (S_ISBLK(stat->mode)) bdev_statx(path, stat, request_mask); - return error; + return 0; } static int vfs_statx_fd(int fd, int flags, struct kstat *stat, From 4e7487245abcbc5a1a1aea54e4d3b33c53804bda Mon Sep 17 00:00:00 2001 From: Brahmajit Das Date: Tue, 21 Jan 2025 21:56:48 +0530 Subject: [PATCH 0611/1443] vboxsf: fix building with GCC 15 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Building with GCC 15 results in build error fs/vboxsf/super.c:24:54: error: initializer-string for array of ‘unsigned char’ is too long [-Werror=unterminated-string-initialization] 24 | static const unsigned char VBSF_MOUNT_SIGNATURE[4] = "\000\377\376\375"; | ^~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors Due to GCC having enabled -Werror=unterminated-string-initialization[0] by default. Separately initializing each array element of VBSF_MOUNT_SIGNATURE to ensure NUL termination, thus satisfying GCC 15 and fixing the build error. [0]: https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Wno-unterminated-string-initialization Signed-off-by: Brahmajit Das Link: https://lore.kernel.org/r/20250121162648.1408743-1-brahmajit.xyz@gmail.com Reviewed-by: Hans de Goede Signed-off-by: Christian Brauner --- fs/vboxsf/super.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c index e95b8a48d8a0..1d94bb784108 100644 --- a/fs/vboxsf/super.c +++ b/fs/vboxsf/super.c @@ -21,7 +21,8 @@ #define VBOXSF_SUPER_MAGIC 0x786f4256 /* 'VBox' little endian */ -static const unsigned char VBSF_MOUNT_SIGNATURE[4] = "\000\377\376\375"; +static const unsigned char VBSF_MOUNT_SIGNATURE[4] = { '\000', '\377', '\376', + '\375' }; static int follow_symlinks; module_param(follow_symlinks, int, 0444); From e52e97f09fb66fd868260d05bd6b74a9a3db39ee Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 30 Jan 2025 13:15:00 +0100 Subject: [PATCH 0612/1443] statmount: let unset strings be empty Just like it's normal for unset values to be zero, unset strings should be empty instead of containing random values. It seems to be a typical mistake that the mask returned by statmount is not checked, which can result in various bugs. With this fix, these bugs are prevented, since it is highly likely that userspace would just want to turn the missing mask case into an empty string anyway (most of the recently found cases are of this type). Link: https://lore.kernel.org/all/CAJfpegsVCPfCn2DpM8iiYSS5DpMsLB8QBUCHecoj6s0Vxf4jzg@mail.gmail.com/ Fixes: 68385d77c05b ("statmount: simplify string option retrieval") Fixes: 46eae99ef733 ("add statmount(2) syscall") Cc: stable@vger.kernel.org # v6.8 Signed-off-by: Miklos Szeredi Link: https://lore.kernel.org/r/20250130121500.113446-1-mszeredi@redhat.com Reviewed-by: Jeff Layton Signed-off-by: Christian Brauner --- fs/namespace.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index a3ed3f2980cb..9c4d307a82cd 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -5191,39 +5191,45 @@ static int statmount_string(struct kstatmount *s, u64 flag) size_t kbufsize; struct seq_file *seq = &s->seq; struct statmount *sm = &s->sm; - u32 start = seq->count; + u32 start, *offp; + + /* Reserve an empty string at the beginning for any unset offsets */ + if (!seq->count) + seq_putc(seq, 0); + + start = seq->count; switch (flag) { case STATMOUNT_FS_TYPE: - sm->fs_type = start; + offp = &sm->fs_type; ret = statmount_fs_type(s, seq); break; case STATMOUNT_MNT_ROOT: - sm->mnt_root = start; + offp = &sm->mnt_root; ret = statmount_mnt_root(s, seq); break; case STATMOUNT_MNT_POINT: - sm->mnt_point = start; + offp = &sm->mnt_point; ret = statmount_mnt_point(s, seq); break; case STATMOUNT_MNT_OPTS: - sm->mnt_opts = start; + offp = &sm->mnt_opts; ret = statmount_mnt_opts(s, seq); break; case STATMOUNT_OPT_ARRAY: - sm->opt_array = start; + offp = &sm->opt_array; ret = statmount_opt_array(s, seq); break; case STATMOUNT_OPT_SEC_ARRAY: - sm->opt_sec_array = start; + offp = &sm->opt_sec_array; ret = statmount_opt_sec_array(s, seq); break; case STATMOUNT_FS_SUBTYPE: - sm->fs_subtype = start; + offp = &sm->fs_subtype; statmount_fs_subtype(s, seq); break; case STATMOUNT_SB_SOURCE: - sm->sb_source = start; + offp = &sm->sb_source; ret = statmount_sb_source(s, seq); break; default: @@ -5251,6 +5257,7 @@ static int statmount_string(struct kstatmount *s, u64 flag) seq->buf[seq->count++] = '\0'; sm->mask |= flag; + *offp = start; return 0; } From d9b3a3c70df2c2b87c83ca3f6e8ab49bd092fdbd Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 30 Jan 2025 14:56:21 +0100 Subject: [PATCH 0613/1443] gfs2: use lockref_init for gl_lockref Move the initialization of gl_lockref from gfs2_init_glock_once() to gfs2_glock_get(). This allows to use lockref_init() there. Reviewed-by: Christoph Hellwig Signed-off-by: Andreas Gruenbacher Link: https://lore.kernel.org/r/20250130135624.1899988-2-agruenba@redhat.com Signed-off-by: Christian Brauner --- fs/gfs2/glock.c | 2 +- fs/gfs2/main.c | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 8c4c1f871a88..b29eb71e3e29 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1201,8 +1201,8 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, if (glops->go_instantiate) gl->gl_flags |= BIT(GLF_INSTANTIATE_NEEDED); gl->gl_name = name; + lockref_init(&gl->gl_lockref, 1); lockdep_set_subclass(&gl->gl_lockref.lock, glops->go_subclass); - gl->gl_lockref.count = 1; gl->gl_state = LM_ST_UNLOCKED; gl->gl_target = LM_ST_UNLOCKED; gl->gl_demote_state = LM_ST_EXCLUSIVE; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 04cadc02e5a6..0727f60ad028 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -51,7 +51,6 @@ static void gfs2_init_glock_once(void *foo) { struct gfs2_glock *gl = foo; - spin_lock_init(&gl->gl_lockref.lock); INIT_LIST_HEAD(&gl->gl_holders); INIT_LIST_HEAD(&gl->gl_lru); INIT_LIST_HEAD(&gl->gl_ail_list); From 34ad6fa2add2b38f2a89d28518de0142bff8fb43 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 30 Jan 2025 14:56:22 +0100 Subject: [PATCH 0614/1443] gfs2: switch to lockref_init(..., 1) In qd_alloc(), initialize the lockref count to 1 to cover the common case. Compensate for that in gfs2_quota_init() by adjusting the count back down to 0; this only occurs when mounting the filesystem rw. Reviewed-by: Christoph Hellwig Signed-off-by: Andreas Gruenbacher Link: https://lore.kernel.org/r/20250130135624.1899988-3-agruenba@redhat.com Signed-off-by: Christian Brauner --- fs/gfs2/quota.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 58bc5013ca49..6ae529a5388b 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -236,7 +236,7 @@ static struct gfs2_quota_data *qd_alloc(unsigned hash, struct gfs2_sbd *sdp, str return NULL; qd->qd_sbd = sdp; - lockref_init(&qd->qd_lockref, 0); + lockref_init(&qd->qd_lockref, 1); qd->qd_id = qid; qd->qd_slot = -1; INIT_LIST_HEAD(&qd->qd_lru); @@ -297,7 +297,6 @@ static int qd_get(struct gfs2_sbd *sdp, struct kqid qid, spin_lock_bucket(hash); *qdp = qd = gfs2_qd_search_bucket(hash, sdp, qid); if (qd == NULL) { - new_qd->qd_lockref.count++; *qdp = new_qd; list_add(&new_qd->qd_list, &sdp->sd_quota_list); hlist_bl_add_head_rcu(&new_qd->qd_hlist, &qd_hash_table[hash]); @@ -1450,6 +1449,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp) if (qd == NULL) goto fail_brelse; + qd->qd_lockref.count = 0; set_bit(QDF_CHANGE, &qd->qd_flags); qd->qd_change = qc_change; qd->qd_slot = slot; From bb504b4d64266fa0d7460c218c85afed371db03a Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Thu, 30 Jan 2025 14:56:23 +0100 Subject: [PATCH 0615/1443] lockref: remove count argument of lockref_init All users of lockref_init() now initialize the count to 1, so hardcode that and remove the count argument. Reviewed-by: Christoph Hellwig Signed-off-by: Andreas Gruenbacher Link: https://lore.kernel.org/r/20250130135624.1899988-4-agruenba@redhat.com Signed-off-by: Christian Brauner --- fs/dcache.c | 2 +- fs/erofs/zdata.c | 2 +- fs/gfs2/glock.c | 2 +- fs/gfs2/quota.c | 2 +- include/linux/lockref.h | 7 ++++--- 5 files changed, 8 insertions(+), 7 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 9cc0d47da321..7dee242b4195 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1700,7 +1700,7 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name) smp_store_release(&dentry->d_name.name, dname); /* ^^^ */ dentry->d_flags = 0; - lockref_init(&dentry->d_lockref, 1); + lockref_init(&dentry->d_lockref); seqcount_spinlock_init(&dentry->d_seq, &dentry->d_lock); dentry->d_inode = NULL; dentry->d_parent = dentry; diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 29f8963bb523..d771e06db738 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -726,7 +726,7 @@ static int z_erofs_register_pcluster(struct z_erofs_frontend *fe) if (IS_ERR(pcl)) return PTR_ERR(pcl); - lockref_init(&pcl->lockref, 1); /* one ref for this request */ + lockref_init(&pcl->lockref); /* one ref for this request */ pcl->algorithmformat = map->m_algorithmformat; pcl->length = 0; pcl->partial = true; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index b29eb71e3e29..65c07aa95718 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1201,7 +1201,7 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number, if (glops->go_instantiate) gl->gl_flags |= BIT(GLF_INSTANTIATE_NEEDED); gl->gl_name = name; - lockref_init(&gl->gl_lockref, 1); + lockref_init(&gl->gl_lockref); lockdep_set_subclass(&gl->gl_lockref.lock, glops->go_subclass); gl->gl_state = LM_ST_UNLOCKED; gl->gl_target = LM_ST_UNLOCKED; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 6ae529a5388b..2298e06797ac 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -236,7 +236,7 @@ static struct gfs2_quota_data *qd_alloc(unsigned hash, struct gfs2_sbd *sdp, str return NULL; qd->qd_sbd = sdp; - lockref_init(&qd->qd_lockref, 1); + lockref_init(&qd->qd_lockref); qd->qd_id = qid; qd->qd_slot = -1; INIT_LIST_HEAD(&qd->qd_lru); diff --git a/include/linux/lockref.h b/include/linux/lockref.h index c39f119659ba..676721ee878d 100644 --- a/include/linux/lockref.h +++ b/include/linux/lockref.h @@ -37,12 +37,13 @@ struct lockref { /** * lockref_init - Initialize a lockref * @lockref: pointer to lockref structure - * @count: initial count + * + * Initializes @lockref->count to 1. */ -static inline void lockref_init(struct lockref *lockref, unsigned int count) +static inline void lockref_init(struct lockref *lockref) { spin_lock_init(&lockref->lock); - lockref->count = count; + lockref->count = 1; } void lockref_get(struct lockref *lockref); From 95101401bb50ae2cf9deee1bbf4d2b28d0dfdc26 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Mon, 3 Feb 2025 23:32:03 +0100 Subject: [PATCH 0616/1443] fsnotify: use accessor to set FMODE_NONOTIFY_* The FMODE_NONOTIFY_* bits are a 2-bits mode. Open coding manipulation of those bits is risky. Use an accessor file_set_fsnotify_mode() to set the mode. Rename file_set_fsnotify_mode() => file_set_fsnotify_mode_from_watchers() to make way for the simple accessor name. Signed-off-by: Amir Goldstein Link: https://lore.kernel.org/r/20250203223205.861346-2-amir73il@gmail.com Signed-off-by: Christian Brauner --- drivers/tty/pty.c | 2 +- fs/notify/fsnotify.c | 18 ++++++++++++------ fs/open.c | 7 ++++--- include/linux/fs.h | 7 ++++++- include/linux/fsnotify.h | 4 ++-- 5 files changed, 25 insertions(+), 13 deletions(-) diff --git a/drivers/tty/pty.c b/drivers/tty/pty.c index df08f13052ff..8bb1a01fef2a 100644 --- a/drivers/tty/pty.c +++ b/drivers/tty/pty.c @@ -798,7 +798,7 @@ static int ptmx_open(struct inode *inode, struct file *filp) nonseekable_open(inode, filp); /* We refuse fsnotify events on ptmx, since it's a shared resource */ - filp->f_mode |= FMODE_NONOTIFY; + file_set_fsnotify_mode(filp, FMODE_NONOTIFY); retval = tty_alloc_file(filp); if (retval) diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 8ee495a58d0a..fae1b6d397ea 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -648,7 +648,7 @@ EXPORT_SYMBOL_GPL(fsnotify); * Later, fsnotify permission hooks do not check if there are permission event * watches, but that there were permission event watches at open time. */ -void file_set_fsnotify_mode(struct file *file) +void file_set_fsnotify_mode_from_watchers(struct file *file) { struct dentry *dentry = file->f_path.dentry, *parent; struct super_block *sb = dentry->d_sb; @@ -665,7 +665,7 @@ void file_set_fsnotify_mode(struct file *file) */ if (likely(!fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_CONTENT))) { - file->f_mode |= FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY_PERM); return; } @@ -676,7 +676,7 @@ void file_set_fsnotify_mode(struct file *file) if ((!d_is_dir(dentry) && !d_is_reg(dentry)) || likely(!fsnotify_sb_has_priority_watchers(sb, FSNOTIFY_PRIO_PRE_CONTENT))) { - file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM); return; } @@ -686,19 +686,25 @@ void file_set_fsnotify_mode(struct file *file) */ mnt_mask = READ_ONCE(real_mount(file->f_path.mnt)->mnt_fsnotify_mask); if (unlikely(fsnotify_object_watched(d_inode(dentry), mnt_mask, - FSNOTIFY_PRE_CONTENT_EVENTS))) + FSNOTIFY_PRE_CONTENT_EVENTS))) { + /* Enable pre-content events */ + file_set_fsnotify_mode(file, 0); return; + } /* Is parent watching for pre-content events on this file? */ if (dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED) { parent = dget_parent(dentry); p_mask = fsnotify_inode_watches_children(d_inode(parent)); dput(parent); - if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS) + if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS) { + /* Enable pre-content events */ + file_set_fsnotify_mode(file, 0); return; + } } /* Nobody watching for pre-content events from this file */ - file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM; + file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM); } #endif diff --git a/fs/open.c b/fs/open.c index 932e5a6de63b..3fcbfff8aede 100644 --- a/fs/open.c +++ b/fs/open.c @@ -905,7 +905,8 @@ static int do_dentry_open(struct file *f, f->f_sb_err = file_sample_sb_err(f); if (unlikely(f->f_flags & O_PATH)) { - f->f_mode = FMODE_PATH | FMODE_OPENED | FMODE_NONOTIFY; + f->f_mode = FMODE_PATH | FMODE_OPENED; + file_set_fsnotify_mode(f, FMODE_NONOTIFY); f->f_op = &empty_fops; return 0; } @@ -938,7 +939,7 @@ static int do_dentry_open(struct file *f, * If FMODE_NONOTIFY was already set for an fanotify fd, this doesn't * change anything. */ - file_set_fsnotify_mode(f); + file_set_fsnotify_mode_from_watchers(f); error = fsnotify_open_perm(f); if (error) goto cleanup_all; @@ -1122,7 +1123,7 @@ struct file *dentry_open_nonotify(const struct path *path, int flags, if (!IS_ERR(f)) { int error; - f->f_mode |= FMODE_NONOTIFY; + file_set_fsnotify_mode(f, FMODE_NONOTIFY); error = vfs_open(path, f); if (error) { fput(f); diff --git a/include/linux/fs.h b/include/linux/fs.h index be3ad155ec9f..7620547432a8 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -222,7 +222,6 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, #define FMODE_FSNOTIFY_HSM(mode) 0 #endif - /* * Attribute flags. These should be or-ed together to figure out what * has been changed! @@ -3140,6 +3139,12 @@ static inline void exe_file_allow_write_access(struct file *exe_file) allow_write_access(exe_file); } +static inline void file_set_fsnotify_mode(struct file *file, fmode_t mode) +{ + file->f_mode &= ~FMODE_FSNOTIFY_MASK; + file->f_mode |= mode; +} + static inline bool inode_is_open_for_write(const struct inode *inode) { return atomic_read(&inode->i_writecount) > 0; diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 1a9ef8f6784d..6a33288bd6a1 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -129,7 +129,7 @@ static inline int fsnotify_file(struct file *file, __u32 mask) #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS -void file_set_fsnotify_mode(struct file *file); +void file_set_fsnotify_mode_from_watchers(struct file *file); /* * fsnotify_file_area_perm - permission hook before access to file range @@ -213,7 +213,7 @@ static inline int fsnotify_open_perm(struct file *file) } #else -static inline void file_set_fsnotify_mode(struct file *file) +static inline void file_set_fsnotify_mode_from_watchers(struct file *file) { } From 5eb987105357cb7cfa7cf3b1e2f66d5c0977e412 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 29 Jan 2025 16:12:53 +0100 Subject: [PATCH 0617/1443] fs: fix adding security options to statmount.mnt_opt Prepending security options was made conditional on sb->s_op->show_options, but security options are independent of sb options. Fixes: 056d33137bf9 ("fs: prepend statmount.mnt_opts string with security_sb_mnt_opts()") Fixes: f9af549d1fd3 ("fs: export mount options via statmount()") Cc: stable@vger.kernel.org # v6.11 Signed-off-by: Miklos Szeredi Link: https://lore.kernel.org/r/20250129151253.33241-1-mszeredi@redhat.com Signed-off-by: Christian Brauner --- fs/namespace.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-) diff --git a/fs/namespace.c b/fs/namespace.c index 9c4d307a82cd..8f1000f9f3df 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -5087,31 +5087,30 @@ static int statmount_mnt_opts(struct kstatmount *s, struct seq_file *seq) { struct vfsmount *mnt = s->mnt; struct super_block *sb = mnt->mnt_sb; + size_t start = seq->count; int err; + err = security_sb_show_options(seq, sb); + if (err) + return err; + if (sb->s_op->show_options) { - size_t start = seq->count; - - err = security_sb_show_options(seq, sb); - if (err) - return err; - err = sb->s_op->show_options(seq, mnt->mnt_root); if (err) return err; - - if (unlikely(seq_has_overflowed(seq))) - return -EAGAIN; - - if (seq->count == start) - return 0; - - /* skip leading comma */ - memmove(seq->buf + start, seq->buf + start + 1, - seq->count - start - 1); - seq->count--; } + if (unlikely(seq_has_overflowed(seq))) + return -EAGAIN; + + if (seq->count == start) + return 0; + + /* skip leading comma */ + memmove(seq->buf + start, seq->buf + start + 1, + seq->count - start - 1); + seq->count--; + return 0; } From 2a42754b3104d78a2bc7a2ad8844427411c76ca6 Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Mon, 3 Feb 2025 23:32:04 +0100 Subject: [PATCH 0618/1443] fsnotify: disable notification by default for all pseudo files Most pseudo files are not applicable for fsnotify events at all, let alone to the new pre-content events. Disable notifications to all files allocated with alloc_file_pseudo() and enable legacy inotify events for the specific cases of pipe and socket, which have known users of inotify events. Pre-content events are also kept disabled for sockets and pipes. Fixes: 20bf82a898b6 ("mm: don't allow huge faults for files with pre content watches") Reported-by: Alex Williamson Closes: https://lore.kernel.org/linux-fsdevel/20250131121703.1e4d00a7.alex.williamson@redhat.com/ Suggested-by: Linus Torvalds Link: https://lore.kernel.org/linux-fsdevel/CAHk-=wi2pThSVY=zhO=ZKxViBj5QCRX-=AS2+rVknQgJnHXDFg@mail.gmail.com/ Tested-by: Alex Williamson Signed-off-by: Amir Goldstein Link: https://lore.kernel.org/r/20250203223205.861346-3-amir73il@gmail.com Signed-off-by: Christian Brauner --- fs/file_table.c | 11 +++++++++++ fs/open.c | 4 ++-- fs/pipe.c | 6 ++++++ net/socket.c | 5 +++++ 4 files changed, 24 insertions(+), 2 deletions(-) diff --git a/fs/file_table.c b/fs/file_table.c index f0291a66f9db..35b93da6c5cb 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -375,7 +375,13 @@ struct file *alloc_file_pseudo(struct inode *inode, struct vfsmount *mnt, if (IS_ERR(file)) { ihold(inode); path_put(&path); + return file; } + /* + * Disable all fsnotify events for pseudo files by default. + * They may be enabled by caller with file_set_fsnotify_mode(). + */ + file_set_fsnotify_mode(file, FMODE_NONOTIFY); return file; } EXPORT_SYMBOL(alloc_file_pseudo); @@ -400,6 +406,11 @@ struct file *alloc_file_pseudo_noaccount(struct inode *inode, return file; } file_init_path(file, &path, fops); + /* + * Disable all fsnotify events for pseudo files by default. + * They may be enabled by caller with file_set_fsnotify_mode(). + */ + file_set_fsnotify_mode(file, FMODE_NONOTIFY); return file; } EXPORT_SYMBOL_GPL(alloc_file_pseudo_noaccount); diff --git a/fs/open.c b/fs/open.c index 3fcbfff8aede..1be20de9f283 100644 --- a/fs/open.c +++ b/fs/open.c @@ -936,8 +936,8 @@ static int do_dentry_open(struct file *f, /* * Set FMODE_NONOTIFY_* bits according to existing permission watches. - * If FMODE_NONOTIFY was already set for an fanotify fd, this doesn't - * change anything. + * If FMODE_NONOTIFY mode was already set for an fanotify fd or for a + * pseudo file, this call will not change the mode. */ file_set_fsnotify_mode_from_watchers(f); error = fsnotify_open_perm(f); diff --git a/fs/pipe.c b/fs/pipe.c index 94b59045ab44..ce1af7592780 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -960,6 +960,12 @@ int create_pipe_files(struct file **res, int flags) res[1] = f; stream_open(inode, res[0]); stream_open(inode, res[1]); + /* + * Disable permission and pre-content events, but enable legacy + * inotify events for legacy users. + */ + file_set_fsnotify_mode(res[0], FMODE_NONOTIFY_PERM); + file_set_fsnotify_mode(res[1], FMODE_NONOTIFY_PERM); return 0; } diff --git a/net/socket.c b/net/socket.c index 262a28b59c7f..28bae5a94234 100644 --- a/net/socket.c +++ b/net/socket.c @@ -479,6 +479,11 @@ struct file *sock_alloc_file(struct socket *sock, int flags, const char *dname) sock->file = file; file->private_data = sock; stream_open(SOCK_INODE(sock), file); + /* + * Disable permission and pre-content events, but enable legacy + * inotify events for legacy users. + */ + file_set_fsnotify_mode(file, FMODE_NONOTIFY_PERM); return file; } EXPORT_SYMBOL(sock_alloc_file); From 2cc02059fbc79306b53a44b1f1a4444aa3c76598 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 29 Jan 2025 17:06:41 +0100 Subject: [PATCH 0619/1443] selftests: always check mask returned by statmount(2) STATMOUNT_MNT_OPTS can actually be missing if there are no options. This is a change of behavior since 75ead69a7173 ("fs: don't let statmount return empty strings"). The other checks shouldn't actually trigger, but add them for correctness and for easier debugging if the test fails. Signed-off-by: Miklos Szeredi Link: https://lore.kernel.org/r/20250129160641.35485-1-mszeredi@redhat.com Signed-off-by: Christian Brauner --- .../filesystems/statmount/statmount_test.c | 22 ++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/filesystems/statmount/statmount_test.c b/tools/testing/selftests/filesystems/statmount/statmount_test.c index 8eb6aa606a0d..46d289611ce8 100644 --- a/tools/testing/selftests/filesystems/statmount/statmount_test.c +++ b/tools/testing/selftests/filesystems/statmount/statmount_test.c @@ -383,6 +383,10 @@ static void test_statmount_mnt_point(void) return; } + if (!(sm->mask & STATMOUNT_MNT_POINT)) { + ksft_test_result_fail("missing STATMOUNT_MNT_POINT in mask\n"); + return; + } if (strcmp(sm->str + sm->mnt_point, "/") != 0) { ksft_test_result_fail("unexpected mount point: '%s' != '/'\n", sm->str + sm->mnt_point); @@ -408,6 +412,10 @@ static void test_statmount_mnt_root(void) strerror(errno)); return; } + if (!(sm->mask & STATMOUNT_MNT_ROOT)) { + ksft_test_result_fail("missing STATMOUNT_MNT_ROOT in mask\n"); + return; + } mnt_root = sm->str + sm->mnt_root; last_root = strrchr(mnt_root, '/'); if (last_root) @@ -437,6 +445,10 @@ static void test_statmount_fs_type(void) strerror(errno)); return; } + if (!(sm->mask & STATMOUNT_FS_TYPE)) { + ksft_test_result_fail("missing STATMOUNT_FS_TYPE in mask\n"); + return; + } fs_type = sm->str + sm->fs_type; for (s = known_fs; s != NULL; s++) { if (strcmp(fs_type, *s) == 0) @@ -464,6 +476,11 @@ static void test_statmount_mnt_opts(void) return; } + if (!(sm->mask & STATMOUNT_MNT_BASIC)) { + ksft_test_result_fail("missing STATMOUNT_MNT_BASIC in mask\n"); + return; + } + while (getline(&line, &len, f_mountinfo) != -1) { int i; char *p, *p2; @@ -514,7 +531,10 @@ static void test_statmount_mnt_opts(void) if (p2) *p2 = '\0'; - statmount_opts = sm->str + sm->mnt_opts; + if (sm->mask & STATMOUNT_MNT_OPTS) + statmount_opts = sm->str + sm->mnt_opts; + else + statmount_opts = ""; if (strcmp(statmount_opts, p) != 0) ksft_test_result_fail( "unexpected mount options: '%s' != '%s'\n", From 711f9b8fbe4f4936302804e246e206f0829f628f Mon Sep 17 00:00:00 2001 From: Amir Goldstein Date: Mon, 3 Feb 2025 23:32:05 +0100 Subject: [PATCH 0620/1443] fsnotify: disable pre-content and permission events by default After introducing pre-content events, we had a regression related to disabling huge faults on files that should never have pre-content events enabled. This happened because the default f_mode of allocated files (0) does not disable pre-content events. Pre-content events are disabled in file_set_fsnotify_mode_by_watchers() but internal files may not get to call this helper. Initialize f_mode to disable permission and pre-content events for all files and if needed they will be enabled for the callers of file_set_fsnotify_mode_by_watchers(). Fixes: 20bf82a898b6 ("mm: don't allow huge faults for files with pre content watches") Reported-by: Alex Williamson Closes: https://lore.kernel.org/linux-fsdevel/20250131121703.1e4d00a7.alex.williamson@redhat.com/ Tested-by: Alex Williamson Signed-off-by: Amir Goldstein Link: https://lore.kernel.org/r/20250203223205.861346-4-amir73il@gmail.com Signed-off-by: Christian Brauner --- fs/file_table.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/fs/file_table.c b/fs/file_table.c index 35b93da6c5cb..5c00dc38558d 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -194,6 +194,11 @@ static int init_file(struct file *f, int flags, const struct cred *cred) * refcount bumps we should reinitialize the reused file first. */ file_ref_init(&f->f_ref, 1); + /* + * Disable permission and pre-content events for all files by default. + * They may be enabled later by file_set_fsnotify_mode_from_watchers(). + */ + file_set_fsnotify_mode(f, FMODE_NONOTIFY_PERM); return 0; } From 091ee63e36e8289f9067f659a48d497911e49d6f Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Tue, 4 Feb 2025 14:51:20 +0100 Subject: [PATCH 0621/1443] pidfs: improve ioctl handling Pidfs supports extensible and non-extensible ioctls. The extensible ioctls need to check for the ioctl number itself not just the ioctl command otherwise both backward- and forward compatibility are broken. The pidfs ioctl handler also needs to look at the type of the ioctl command to guard against cases where "[...] a daemon receives some random file descriptor from a (potentially less privileged) client and expects the FD to be of some specific type, it might call ioctl() on this FD with some type-specific command and expect the call to fail if the FD is of the wrong type; but due to the missing type check, the kernel instead performs some action that userspace didn't expect." (cf. [1]] Link: https://lore.kernel.org/r/20250204-work-pidfs-ioctl-v1-1-04987d239575@kernel.org Link: https://lore.kernel.org/r/CAG48ez2K9A5GwtgqO31u9ZL292we8ZwAA=TJwwEv7wRuJ3j4Lw@mail.gmail.com [1] Fixes: 8ce352818820 ("pidfs: check for valid ioctl commands") Acked-by: Luca Boccassi Reported-by: Jann Horn Cc: stable@vger.kernel.org # v6.13; please backport with 8ce352818820 ("pidfs: check for valid ioctl commands") Signed-off-by: Christian Brauner --- fs/pidfs.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/pidfs.c b/fs/pidfs.c index 049352f973de..63f9699ebac3 100644 --- a/fs/pidfs.c +++ b/fs/pidfs.c @@ -287,7 +287,6 @@ static bool pidfs_ioctl_valid(unsigned int cmd) switch (cmd) { case FS_IOC_GETVERSION: case PIDFD_GET_CGROUP_NAMESPACE: - case PIDFD_GET_INFO: case PIDFD_GET_IPC_NAMESPACE: case PIDFD_GET_MNT_NAMESPACE: case PIDFD_GET_NET_NAMESPACE: @@ -300,6 +299,17 @@ static bool pidfs_ioctl_valid(unsigned int cmd) return true; } + /* Extensible ioctls require some more careful checks. */ + switch (_IOC_NR(cmd)) { + case _IOC_NR(PIDFD_GET_INFO): + /* + * Try to prevent performing a pidfd ioctl when someone + * erronously mistook the file descriptor for a pidfd. + * This is not perfect but will catch most cases. + */ + return (_IOC_TYPE(cmd) == _IOC_TYPE(PIDFD_GET_INFO)); + } + return false; } From 37d11cfc63604b3886308e2111d845d148ced8bc Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Tue, 4 Feb 2025 22:32:07 +0100 Subject: [PATCH 0622/1443] vfs: sanity check the length passed to inode_set_cached_link() This costs a strlen() call when instatianating a symlink. Preferably it would be hidden behind VFS_WARN_ON (or compatible), but there is no such facility at the moment. With the facility in place the call can be patched out in production kernels. In the meantime, since the cost is being paid unconditionally, use the result to a fixup the bad caller. This is not expected to persist in the long run (tm). Sample splat: bad length passed for symlink [/tmp/syz-imagegen43743633/file0/file0] (got 131109, expected 37) [rest of WARN blurp goes here] Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/r/20250204213207.337980-1-mjguzik@gmail.com Signed-off-by: Christian Brauner --- include/linux/fs.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/include/linux/fs.h b/include/linux/fs.h index 7620547432a8..2c3b2f8a621f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -790,6 +790,19 @@ struct inode { static inline void inode_set_cached_link(struct inode *inode, char *link, int linklen) { + int testlen; + + /* + * TODO: patch it into a debug-only check if relevant macros show up. + * In the meantime, since we are suffering strlen even on production kernels + * to find the right length, do a fixup if the wrong value got passed. + */ + testlen = strlen(link); + if (testlen != linklen) { + WARN_ONCE(1, "bad length passed for symlink [%s] (got %d, expected %d)", + link, linklen, testlen); + linklen = testlen; + } inode->i_link = link; inode->i_linklen = linklen; inode->i_opflags |= IOP_CACHED_LINK; From ca0f4fe7cf7183bfbdc67ca2de56ae1fc3a8db2b Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Thu, 6 Feb 2025 10:21:38 -0700 Subject: [PATCH 0623/1443] arm64: Handle .ARM.attributes section in linker scripts A recent LLVM commit [1] started generating an .ARM.attributes section similar to the one that exists for 32-bit, which results in orphan section warnings (or errors if CONFIG_WERROR is enabled) from the linker because it is not handled in the arm64 linker scripts. ld.lld: error: arch/arm64/kernel/vdso/vgettimeofday.o:(.ARM.attributes) is being placed in '.ARM.attributes' ld.lld: error: arch/arm64/kernel/vdso/vgetrandom.o:(.ARM.attributes) is being placed in '.ARM.attributes' ld.lld: error: vmlinux.a(lib/vsprintf.o):(.ARM.attributes) is being placed in '.ARM.attributes' ld.lld: error: vmlinux.a(lib/win_minmax.o):(.ARM.attributes) is being placed in '.ARM.attributes' ld.lld: error: vmlinux.a(lib/xarray.o):(.ARM.attributes) is being placed in '.ARM.attributes' Discard the new sections in the necessary linker scripts to resolve the warnings, as the kernel and vDSO do not need to retain it, similar to the .note.gnu.property section. Cc: stable@vger.kernel.org Fixes: b3e5d80d0c48 ("arm64/build: Warn on orphan section placement") Link: https://github.com/llvm/llvm-project/commit/ee99c4d4845db66c4daa2373352133f4b237c942 [1] Signed-off-by: Nathan Chancellor Link: https://lore.kernel.org/r/20250206-arm64-handle-arm-attributes-in-linker-script-v3-1-d53d169913eb@kernel.org Signed-off-by: Will Deacon --- arch/arm64/kernel/vdso/vdso.lds.S | 1 + arch/arm64/kernel/vmlinux.lds.S | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/arm64/kernel/vdso/vdso.lds.S b/arch/arm64/kernel/vdso/vdso.lds.S index 4ec32e86a8da..47ad6944f9f0 100644 --- a/arch/arm64/kernel/vdso/vdso.lds.S +++ b/arch/arm64/kernel/vdso/vdso.lds.S @@ -41,6 +41,7 @@ SECTIONS */ /DISCARD/ : { *(.note.GNU-stack .note.gnu.property) + *(.ARM.attributes) } .note : { *(.note.*) } :text :note diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S index f84c71f04d9e..e73326bd3ff7 100644 --- a/arch/arm64/kernel/vmlinux.lds.S +++ b/arch/arm64/kernel/vmlinux.lds.S @@ -162,6 +162,7 @@ SECTIONS /DISCARD/ : { *(.interp .dynamic) *(.dynsym .dynstr .hash .gnu.hash) + *(.ARM.attributes) } . = KIMAGE_VADDR; From 875d742cf5327c93cba1f11e12b08d3cce7a88d2 Mon Sep 17 00:00:00 2001 From: Radu Rendec Date: Thu, 6 Feb 2025 12:44:20 -0500 Subject: [PATCH 0624/1443] arm64: cacheinfo: Avoid out-of-bounds write to cacheinfo array The loop that detects/populates cache information already has a bounds check on the array size but does not account for cache levels with separate data/instructions cache. Fix this by incrementing the index for any populated leaf (instead of any populated level). Fixes: 5d425c186537 ("arm64: kernel: add support for cpu cache information") Signed-off-by: Radu Rendec Link: https://lore.kernel.org/r/20250206174420.2178724-1-rrendec@redhat.com Signed-off-by: Will Deacon --- arch/arm64/kernel/cacheinfo.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/arch/arm64/kernel/cacheinfo.c b/arch/arm64/kernel/cacheinfo.c index d9c9218fa1fd..309942b06c5b 100644 --- a/arch/arm64/kernel/cacheinfo.c +++ b/arch/arm64/kernel/cacheinfo.c @@ -101,16 +101,18 @@ int populate_cache_leaves(unsigned int cpu) unsigned int level, idx; enum cache_type type; struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); - struct cacheinfo *this_leaf = this_cpu_ci->info_list; + struct cacheinfo *infos = this_cpu_ci->info_list; for (idx = 0, level = 1; level <= this_cpu_ci->num_levels && - idx < this_cpu_ci->num_leaves; idx++, level++) { + idx < this_cpu_ci->num_leaves; level++) { type = get_cache_type(level); if (type == CACHE_TYPE_SEPARATE) { - ci_leaf_init(this_leaf++, CACHE_TYPE_DATA, level); - ci_leaf_init(this_leaf++, CACHE_TYPE_INST, level); + if (idx + 1 >= this_cpu_ci->num_leaves) + break; + ci_leaf_init(&infos[idx++], CACHE_TYPE_DATA, level); + ci_leaf_init(&infos[idx++], CACHE_TYPE_INST, level); } else { - ci_leaf_init(this_leaf++, type, level); + ci_leaf_init(&infos[idx++], type, level); } } return 0; From 5b380838930f1fb74d38f34ca4bff43416db01c3 Mon Sep 17 00:00:00 2001 From: Francois Dugast Date: Thu, 6 Feb 2025 14:45:50 +0100 Subject: [PATCH 0625/1443] drm/xe: Add stats for vma page faults Add new entries in stats for vma page faults. If CONFIG_DEBUG_FS is enabled, the count and number of bytes can be viewed per GT in the stat debugfs file. This helps when testing, to confirm page faults have been triggered as expected. It also helps when looking at the performance impact of page faults. Data is simply collected when entering the page fault handler so there is no indication whether it completed successfully, with or without retries, etc. Example output: cat /sys/kernel/debug/dri/0/gt0/stats tlb_inval_count: 129 vma_pagefault_count: 12 vma_pagefault_bytes: 98304 v2: Rebase Reviewed-by: Jonathan Cavitt Link: https://patchwork.freedesktop.org/patch/msgid/20250206134551.1321265-1-francois.dugast@intel.com Signed-off-by: Francois Dugast --- drivers/gpu/drm/xe/xe_gt_pagefault.c | 10 +++++++--- drivers/gpu/drm/xe/xe_gt_stats.c | 2 ++ drivers/gpu/drm/xe/xe_gt_stats_types.h | 2 ++ 3 files changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c index cb92fb5cbc75..46701ca11ce0 100644 --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c @@ -14,6 +14,7 @@ #include "abi/guc_actions_abi.h" #include "xe_bo.h" #include "xe_gt.h" +#include "xe_gt_stats.h" #include "xe_gt_tlb_invalidation.h" #include "xe_guc.h" #include "xe_guc_ct.h" @@ -124,16 +125,20 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma, return 0; } -static int handle_vma_pagefault(struct xe_tile *tile, struct pagefault *pf, +static int handle_vma_pagefault(struct xe_gt *gt, struct pagefault *pf, struct xe_vma *vma) { struct xe_vm *vm = xe_vma_vm(vma); + struct xe_tile *tile = gt_to_tile(gt); struct drm_exec exec; struct dma_fence *fence; ktime_t end = 0; int err; bool atomic; + xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT, 1); + xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_BYTES, xe_vma_size(vma)); + trace_xe_vma_pagefault(vma); atomic = access_is_atomic(pf->access_type); @@ -202,7 +207,6 @@ static struct xe_vm *asid_to_vm(struct xe_device *xe, u32 asid) static int handle_pagefault(struct xe_gt *gt, struct pagefault *pf) { struct xe_device *xe = gt_to_xe(gt); - struct xe_tile *tile = gt_to_tile(gt); struct xe_vm *vm; struct xe_vma *vma = NULL; int err; @@ -231,7 +235,7 @@ static int handle_pagefault(struct xe_gt *gt, struct pagefault *pf) goto unlock_vm; } - err = handle_vma_pagefault(tile, pf, vma); + err = handle_vma_pagefault(gt, pf, vma); unlock_vm: if (!err) diff --git a/drivers/gpu/drm/xe/xe_gt_stats.c b/drivers/gpu/drm/xe/xe_gt_stats.c index 7a6c1d808e41..2e9879ea4674 100644 --- a/drivers/gpu/drm/xe/xe_gt_stats.c +++ b/drivers/gpu/drm/xe/xe_gt_stats.c @@ -28,6 +28,8 @@ void xe_gt_stats_incr(struct xe_gt *gt, const enum xe_gt_stats_id id, int incr) static const char *const stat_description[__XE_GT_STATS_NUM_IDS] = { "tlb_inval_count", + "vma_pagefault_count", + "vma_pagefault_bytes", }; /** diff --git a/drivers/gpu/drm/xe/xe_gt_stats_types.h b/drivers/gpu/drm/xe/xe_gt_stats_types.h index 2fc055e39f27..b072bd80c4b9 100644 --- a/drivers/gpu/drm/xe/xe_gt_stats_types.h +++ b/drivers/gpu/drm/xe/xe_gt_stats_types.h @@ -8,6 +8,8 @@ enum xe_gt_stats_id { XE_GT_STATS_ID_TLB_INVAL, + XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT, + XE_GT_STATS_ID_VMA_PAGEFAULT_BYTES, /* must be the last entry */ __XE_GT_STATS_NUM_IDS, }; From 2eca617f12586abff62038db1c14cb3aa60a15aa Mon Sep 17 00:00:00 2001 From: Tvrtko Ursulin Date: Wed, 5 Feb 2025 11:04:10 +0000 Subject: [PATCH 0626/1443] drm/scheduler: Remove some unused prototypes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As far as I can tell some removed prototypes were introduced by probably bad conflict resolution in fc58764bbf60 ("Merge tag 'amd-drm-next-6.2-2022-11-18' of https://gitlab.freedesktop.org/agd5f/linux into drm-next"). Remove them. Signed-off-by: Tvrtko Ursulin Cc: Christian König Cc: Danilo Krummrich Cc: Matthew Brost Cc: Philipp Stanner Reviewed-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20250205110410.7941-1-tvrtko.ursulin@igalia.com --- include/drm/gpu_scheduler.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h index a0ff08123f07..cf88f2bd020f 100644 --- a/include/drm/gpu_scheduler.h +++ b/include/drm/gpu_scheduler.h @@ -581,10 +581,6 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad); void drm_sched_start(struct drm_gpu_scheduler *sched, int errno); void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched); void drm_sched_increase_karma(struct drm_sched_job *bad); -void drm_sched_reset_karma(struct drm_sched_job *bad); -void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type); -bool drm_sched_dependency_optimized(struct dma_fence* fence, - struct drm_sched_entity *entity); void drm_sched_fault(struct drm_gpu_scheduler *sched); void drm_sched_rq_add_entity(struct drm_sched_rq *rq, From 2813e00dcd748cef47d2bffaa04071de93fddf00 Mon Sep 17 00:00:00 2001 From: Ievgen Vovk Date: Sun, 12 Jan 2025 13:13:14 +0900 Subject: [PATCH 0627/1443] HID: hid-apple: Apple Magic Keyboard a3203 USB-C support Add Apple Magic Keyboard 2024 model (with USB-C port) device ID (0320) to those recognized by the hid-apple driver. Keyboard is otherwise compatible with the existing implementation for its earlier 2021 model. Signed-off-by: Ievgen Vovk Signed-off-by: Jiri Kosina --- drivers/hid/hid-apple.c | 5 +++++ drivers/hid/hid-ids.h | 1 + 2 files changed, 6 insertions(+) diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c index 7e1ae2a2bcc2..3c3f67d0bfcf 100644 --- a/drivers/hid/hid-apple.c +++ b/drivers/hid/hid-apple.c @@ -474,6 +474,7 @@ static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input, hid->product == USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_NUMPAD_2015) table = magic_keyboard_2015_fn_keys; else if (hid->product == USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2021 || + hid->product == USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2024 || hid->product == USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_FINGERPRINT_2021 || hid->product == USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_NUMPAD_2021) table = apple2021_fn_keys; @@ -1150,6 +1151,10 @@ static const struct hid_device_id apple_devices[] = { .driver_data = APPLE_HAS_FN | APPLE_ISO_TILDE_QUIRK | APPLE_RDESC_BATTERY }, { HID_BLUETOOTH_DEVICE(BT_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2021), .driver_data = APPLE_HAS_FN | APPLE_ISO_TILDE_QUIRK }, + { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2024), + .driver_data = APPLE_HAS_FN | APPLE_ISO_TILDE_QUIRK | APPLE_RDESC_BATTERY }, + { HID_BLUETOOTH_DEVICE(BT_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2024), + .driver_data = APPLE_HAS_FN | APPLE_ISO_TILDE_QUIRK }, { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_FINGERPRINT_2021), .driver_data = APPLE_HAS_FN | APPLE_ISO_TILDE_QUIRK | APPLE_RDESC_BATTERY }, { HID_BLUETOOTH_DEVICE(BT_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_FINGERPRINT_2021), diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index ed1d7f9e8caf..7e400624908e 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -184,6 +184,7 @@ #define USB_DEVICE_ID_APPLE_IRCONTROL4 0x8242 #define USB_DEVICE_ID_APPLE_IRCONTROL5 0x8243 #define USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2021 0x029c +#define USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_2024 0x0320 #define USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_FINGERPRINT_2021 0x029a #define USB_DEVICE_ID_APPLE_MAGIC_KEYBOARD_NUMPAD_2021 0x029f #define USB_DEVICE_ID_APPLE_TOUCHBAR_BACKLIGHT 0x8102 From 819083cb6eedcc8495cbf84845877bcc741b93b3 Mon Sep 17 00:00:00 2001 From: Alex Henrie Date: Thu, 16 Jan 2025 23:12:17 -0700 Subject: [PATCH 0628/1443] HID: apple: fix up the F6 key on the Omoton KB066 keyboard The Omoton KB066 is an Apple A1255 keyboard clone (HID product code 05ac:022c). On both keyboards, the F6 key becomes Num Lock when the Fn key is held. But unlike its Apple exemplar, when the Omoton's F6 key is pressed without Fn, it sends the usage code 0xC0301 from the reserved section of the consumer page instead of the standard F6 usage code 0x7003F from the keyboard page. The nonstandard code is translated to KEY_UNKNOWN and becomes useless on Linux. The Omoton KB066 is a pretty popular keyboard, judging from its 29,058 reviews on Amazon at time of writing, so let's account for its quirk to make it more usable. By the way, it would be nice if we could automatically set fnmode to 0 for Omoton keyboards because they handle the Fn key internally and the kernel's Fn key handling creates undesirable side effects such as making F1 and F2 always Brightness Up and Brightness Down in fnmode=1 (the default) or always F1 and F2 in fnmode=2. Unfortunately I don't think there's a way to identify Bluetooth keyboards more specifically than the HID product code which is obviously inaccurate. Users of Omoton keyboards will just have to set fnmode to 0 manually to get full Fn key functionality. Signed-off-by: Alex Henrie Signed-off-by: Jiri Kosina --- drivers/hid/hid-apple.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c index 3c3f67d0bfcf..49812a76b7ed 100644 --- a/drivers/hid/hid-apple.c +++ b/drivers/hid/hid-apple.c @@ -546,6 +546,9 @@ static int hidinput_apple_event(struct hid_device *hid, struct input_dev *input, } } + if (usage->hid == 0xc0301) /* Omoton KB066 quirk */ + code = KEY_F6; + if (usage->code != code) { input_event_with_scancode(input, usage->type, code, usage->hid, value); From 0b43d98ff29be3144e86294486b1373b5df74c0e Mon Sep 17 00:00:00 2001 From: Tulio Fernandes Date: Wed, 5 Feb 2025 18:50:34 -0300 Subject: [PATCH 0629/1443] HID: hid-thrustmaster: fix stack-out-of-bounds read in usb_check_int_endpoints() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Syzbot[1] has detected a stack-out-of-bounds read of the ep_addr array from hid-thrustmaster driver. This array is passed to usb_check_int_endpoints function from usb.c core driver, which executes a for loop that iterates over the elements of the passed array. Not finding a null element at the end of the array, it tries to read the next, non-existent element, crashing the kernel. To fix this, a 0 element was added at the end of the array to break the for loop. [1] https://syzkaller.appspot.com/bug?extid=9c9179ac46169c56c1ad Reported-by: syzbot+9c9179ac46169c56c1ad@syzkaller.appspotmail.com Fixes: 50420d7c79c3 ("HID: hid-thrustmaster: Fix warning in thrustmaster_probe by adding endpoint check") Signed-off-by: Túlio Fernandes Signed-off-by: Jiri Kosina --- drivers/hid/hid-thrustmaster.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/hid-thrustmaster.c b/drivers/hid/hid-thrustmaster.c index 6c3e758bbb09..3b81468a1df2 100644 --- a/drivers/hid/hid-thrustmaster.c +++ b/drivers/hid/hid-thrustmaster.c @@ -171,7 +171,7 @@ static void thrustmaster_interrupts(struct hid_device *hdev) b_ep = ep->desc.bEndpointAddress; /* Are the expected endpoints present? */ - u8 ep_addr[1] = {b_ep}; + u8 ep_addr[2] = {b_ep, 0}; if (!usb_check_int_endpoints(usbif, ep_addr)) { hid_err(hdev, "Unexpected non-int endpoint\n"); From 79504249d7e27cad4a3eeb9afc6386e418728ce0 Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Tue, 4 Feb 2025 19:55:27 -0800 Subject: [PATCH 0630/1443] HID: hid-steam: Move hidraw input (un)registering to work Due to an interplay between locking in the input and hid transport subsystems, attempting to register or deregister the relevant input devices during the hidraw open/close events can lead to a lock ordering issue. Though this shouldn't cause a deadlock, this commit moves the input device manipulation to deferred work to sidestep the issue. Fixes: 385a4886778f6 ("HID: steam: remove input device when a hid client is running.") Signed-off-by: Vicki Pfau Signed-off-by: Jiri Kosina --- drivers/hid/hid-steam.c | 38 +++++++++++++++++++++++++++++++------- 1 file changed, 31 insertions(+), 7 deletions(-) diff --git a/drivers/hid/hid-steam.c b/drivers/hid/hid-steam.c index b008fd0834b9..5a17714fedea 100644 --- a/drivers/hid/hid-steam.c +++ b/drivers/hid/hid-steam.c @@ -313,6 +313,7 @@ struct steam_device { u16 rumble_left; u16 rumble_right; unsigned int sensor_timestamp_us; + struct work_struct unregister_work; }; static int steam_recv_report(struct steam_device *steam, @@ -1072,6 +1073,31 @@ static void steam_mode_switch_cb(struct work_struct *work) } } +static void steam_work_unregister_cb(struct work_struct *work) +{ + struct steam_device *steam = container_of(work, struct steam_device, + unregister_work); + unsigned long flags; + bool connected; + bool opened; + + spin_lock_irqsave(&steam->lock, flags); + opened = steam->client_opened; + connected = steam->connected; + spin_unlock_irqrestore(&steam->lock, flags); + + if (connected) { + if (opened) { + steam_sensors_unregister(steam); + steam_input_unregister(steam); + } else { + steam_set_lizard_mode(steam, lizard_mode); + steam_input_register(steam); + steam_sensors_register(steam); + } + } +} + static bool steam_is_valve_interface(struct hid_device *hdev) { struct hid_report_enum *rep_enum; @@ -1117,8 +1143,7 @@ static int steam_client_ll_open(struct hid_device *hdev) steam->client_opened++; spin_unlock_irqrestore(&steam->lock, flags); - steam_sensors_unregister(steam); - steam_input_unregister(steam); + schedule_work(&steam->unregister_work); return 0; } @@ -1135,11 +1160,7 @@ static void steam_client_ll_close(struct hid_device *hdev) connected = steam->connected && !steam->client_opened; spin_unlock_irqrestore(&steam->lock, flags); - if (connected) { - steam_set_lizard_mode(steam, lizard_mode); - steam_input_register(steam); - steam_sensors_register(steam); - } + schedule_work(&steam->unregister_work); } static int steam_client_ll_raw_request(struct hid_device *hdev, @@ -1231,6 +1252,7 @@ static int steam_probe(struct hid_device *hdev, INIT_LIST_HEAD(&steam->list); INIT_WORK(&steam->rumble_work, steam_haptic_rumble_cb); steam->sensor_timestamp_us = 0; + INIT_WORK(&steam->unregister_work, steam_work_unregister_cb); /* * With the real steam controller interface, do not connect hidraw. @@ -1291,6 +1313,7 @@ static int steam_probe(struct hid_device *hdev, cancel_work_sync(&steam->work_connect); cancel_delayed_work_sync(&steam->mode_switch); cancel_work_sync(&steam->rumble_work); + cancel_work_sync(&steam->unregister_work); return ret; } @@ -1307,6 +1330,7 @@ static void steam_remove(struct hid_device *hdev) cancel_delayed_work_sync(&steam->mode_switch); cancel_work_sync(&steam->work_connect); cancel_work_sync(&steam->rumble_work); + cancel_work_sync(&steam->unregister_work); hid_destroy_device(steam->client_hdev); steam->client_hdev = NULL; steam->client_opened = 0; From b051ffa2aeb2a60e092387b6fb2af1ad42f51a3c Mon Sep 17 00:00:00 2001 From: Vicki Pfau Date: Tue, 4 Feb 2025 19:55:29 -0800 Subject: [PATCH 0631/1443] HID: hid-steam: Don't use cancel_delayed_work_sync in IRQ context Lockdep reported that, as steam_do_deck_input_event is called from steam_raw_event inside of an IRQ context, it can lead to issues if that IRQ occurs while the work to be cancelled is running. By using cancel_delayed_work, this issue can be avoided. The exact ordering of the work and the event processing is not super important, so this is safe. Fixes: cd438e57dd05 ("HID: hid-steam: Add gamepad-only mode switched to by holding options") Signed-off-by: Vicki Pfau Signed-off-by: Jiri Kosina --- drivers/hid/hid-steam.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/hid-steam.c b/drivers/hid/hid-steam.c index 5a17714fedea..c9e65e9088b3 100644 --- a/drivers/hid/hid-steam.c +++ b/drivers/hid/hid-steam.c @@ -1617,7 +1617,7 @@ static void steam_do_deck_input_event(struct steam_device *steam, if (!(b9 & BIT(6)) && steam->did_mode_switch) { steam->did_mode_switch = false; - cancel_delayed_work_sync(&steam->mode_switch); + cancel_delayed_work(&steam->mode_switch); } else if (!steam->client_opened && (b9 & BIT(6)) && !steam->did_mode_switch) { steam->did_mode_switch = true; schedule_delayed_work(&steam->mode_switch, 45 * HZ / 100); From 3ddf003c7e41c038c877ee4661d4900d5eb05268 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Adri=C3=A1n=20Larumbe?= Date: Thu, 30 Jan 2025 17:28:09 +0000 Subject: [PATCH 0632/1443] Documentation/gpu: Clarify format of driver-specific fidnfo keys MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This change reflects de facto usage by amdgpu, as exemplified by commit d6530c33a978 ("drm/amdgpu: expose more memory stats in fdinfo"). Cc: Tvrtko Ursulin Acked-by: Tvrtko Ursulin Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-2-adrian.larumbe@collabora.com --- Documentation/gpu/drm-usage-stats.rst | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst index b7fc106dad99..63d6b2abe5ad 100644 --- a/Documentation/gpu/drm-usage-stats.rst +++ b/Documentation/gpu/drm-usage-stats.rst @@ -21,7 +21,10 @@ File format specification - File shall contain one key value pair per one line of text. - Colon character (`:`) must be used to delimit keys and values. -- All keys shall be prefixed with `drm-`. +- All standardised keys shall be prefixed with `drm-`. +- Driver-specific keys shall be prefixed with `driver_name-`, where + driver_name should ideally be the same as the `name` field in + `struct drm_driver`, although this is not mandatory. - Whitespace between the delimiter and first non-whitespace character shall be ignored when parsing. - Keys are not allowed to contain whitespace characters. From af6c2b7c46e16701fba44a21326cb634786e3e71 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Adri=C3=A1n=20Larumbe?= Date: Thu, 30 Jan 2025 17:28:10 +0000 Subject: [PATCH 0633/1443] drm/file: Add fdinfo helper for printing regions with prefix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is motivated by the desire of some drivers (eg. Panthor) to print the size of internal memory regions with a prefix that reflects the driver name, as suggested in the previous documentation commit. That means adding a new argument to print_size and making it available for DRM users. Cc: Tvrtko Ursulin Reviewed-by: Tvrtko Ursulin Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-3-adrian.larumbe@collabora.com --- drivers/gpu/drm/drm_file.c | 26 ++++++++++++++++++-------- include/drm/drm_file.h | 5 +++++ 2 files changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index 2289e71e2fa2..c299cd94d3f7 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -830,8 +830,11 @@ void drm_send_event(struct drm_device *dev, struct drm_pending_event *e) } EXPORT_SYMBOL(drm_send_event); -static void print_size(struct drm_printer *p, const char *stat, - const char *region, u64 sz) +void drm_fdinfo_print_size(struct drm_printer *p, + const char *prefix, + const char *stat, + const char *region, + u64 sz) { const char *units[] = {"", " KiB", " MiB"}; unsigned u; @@ -842,8 +845,10 @@ static void print_size(struct drm_printer *p, const char *stat, sz = div_u64(sz, SZ_1K); } - drm_printf(p, "drm-%s-%s:\t%llu%s\n", stat, region, sz, units[u]); + drm_printf(p, "%s-%s-%s:\t%llu%s\n", + prefix, stat, region, sz, units[u]); } +EXPORT_SYMBOL(drm_fdinfo_print_size); int drm_memory_stats_is_zero(const struct drm_memory_stats *stats) { @@ -868,17 +873,22 @@ void drm_print_memory_stats(struct drm_printer *p, enum drm_gem_object_status supported_status, const char *region) { - print_size(p, "total", region, stats->private + stats->shared); - print_size(p, "shared", region, stats->shared); + const char *prefix = "drm"; + + drm_fdinfo_print_size(p, prefix, "total", region, + stats->private + stats->shared); + drm_fdinfo_print_size(p, prefix, "shared", region, stats->shared); if (supported_status & DRM_GEM_OBJECT_ACTIVE) - print_size(p, "active", region, stats->active); + drm_fdinfo_print_size(p, prefix, "active", region, stats->active); if (supported_status & DRM_GEM_OBJECT_RESIDENT) - print_size(p, "resident", region, stats->resident); + drm_fdinfo_print_size(p, prefix, "resident", region, + stats->resident); if (supported_status & DRM_GEM_OBJECT_PURGEABLE) - print_size(p, "purgeable", region, stats->purgeable); + drm_fdinfo_print_size(p, prefix, "purgeable", region, + stats->purgeable); } EXPORT_SYMBOL(drm_print_memory_stats); diff --git a/include/drm/drm_file.h b/include/drm/drm_file.h index ef817926cddd..94d365b22505 100644 --- a/include/drm/drm_file.h +++ b/include/drm/drm_file.h @@ -495,6 +495,11 @@ struct drm_memory_stats { enum drm_gem_object_status; int drm_memory_stats_is_zero(const struct drm_memory_stats *stats); +void drm_fdinfo_print_size(struct drm_printer *p, + const char *prefix, + const char *stat, + const char *region, + u64 sz); void drm_print_memory_stats(struct drm_printer *p, const struct drm_memory_stats *stats, enum drm_gem_object_status supported_status, From 434e5ca5b5d7ea415670d4fcb399d90d355a1e38 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Adri=C3=A1n=20Larumbe?= Date: Thu, 30 Jan 2025 17:28:11 +0000 Subject: [PATCH 0634/1443] drm/panthor: Expose size of driver internal BO's over fdinfo MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This will display the sizes of kenrel BO's bound to an open file, which are otherwise not exposed to UM through a handle. The sizes recorded are as follows: - Per group: suspend buffer, protm-suspend buffer, syncobjcs - Per queue: ringbuffer, profiling slots, firmware interface - For all heaps in all heap pools across all VM's bound to an open file, record size of all heap chuks, and for each pool the gpu_context BO too. This does not record the size of FW regions, as these aren't bound to a specific open file and remain active through the whole life of the driver. Reviewed-by: Liviu Dudau Reviewed-by: Mihail Atanassov Reviewed-by: Steven Price Reviewed-by: Boris Brezillon Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-4-adrian.larumbe@collabora.com --- drivers/gpu/drm/panthor/panthor_drv.c | 14 +++++++ drivers/gpu/drm/panthor/panthor_heap.c | 26 ++++++++++++ drivers/gpu/drm/panthor/panthor_heap.h | 2 + drivers/gpu/drm/panthor/panthor_mmu.c | 33 +++++++++++++++ drivers/gpu/drm/panthor/panthor_mmu.h | 3 ++ drivers/gpu/drm/panthor/panthor_sched.c | 56 ++++++++++++++++++++++++- drivers/gpu/drm/panthor/panthor_sched.h | 3 ++ 7 files changed, 136 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/panthor/panthor_drv.c b/drivers/gpu/drm/panthor/panthor_drv.c index d5dcd3d1b33a..310bb44abe1a 100644 --- a/drivers/gpu/drm/panthor/panthor_drv.c +++ b/drivers/gpu/drm/panthor/panthor_drv.c @@ -1457,12 +1457,26 @@ static void panthor_gpu_show_fdinfo(struct panthor_device *ptdev, drm_printf(p, "drm-curfreq-panthor:\t%lu Hz\n", ptdev->current_frequency); } +static void panthor_show_internal_memory_stats(struct drm_printer *p, struct drm_file *file) +{ + char *drv_name = file->minor->dev->driver->name; + struct panthor_file *pfile = file->driver_priv; + struct drm_memory_stats stats = {0}; + + panthor_fdinfo_gather_group_mem_info(pfile, &stats); + panthor_vm_heaps_sizes(pfile, &stats); + + drm_fdinfo_print_size(p, drv_name, "resident", "memory", stats.resident); + drm_fdinfo_print_size(p, drv_name, "active", "memory", stats.active); +} + static void panthor_show_fdinfo(struct drm_printer *p, struct drm_file *file) { struct drm_device *dev = file->minor->dev; struct panthor_device *ptdev = container_of(dev, struct panthor_device, base); panthor_gpu_show_fdinfo(ptdev, file->driver_priv, p); + panthor_show_internal_memory_stats(p, file); drm_show_memory_stats(p, file); } diff --git a/drivers/gpu/drm/panthor/panthor_heap.c b/drivers/gpu/drm/panthor/panthor_heap.c index 3796a9eb22af..db0285ce5812 100644 --- a/drivers/gpu/drm/panthor/panthor_heap.c +++ b/drivers/gpu/drm/panthor/panthor_heap.c @@ -603,3 +603,29 @@ void panthor_heap_pool_destroy(struct panthor_heap_pool *pool) panthor_heap_pool_put(pool); } + +/** + * panthor_heap_pool_size() - Calculate size of all chunks across all heaps in a pool + * @pool: Pool whose total chunk size to calculate. + * + * This function adds the size of all heap chunks across all heaps in the + * argument pool. It also adds the size of the gpu contexts kernel bo. + * It is meant to be used by fdinfo for displaying the size of internal + * driver BO's that aren't exposed to userspace through a GEM handle. + * + */ +size_t panthor_heap_pool_size(struct panthor_heap_pool *pool) +{ + struct panthor_heap *heap; + unsigned long i; + size_t size = 0; + + down_read(&pool->lock); + xa_for_each(&pool->xa, i, heap) + size += heap->chunk_size * heap->chunk_count; + up_read(&pool->lock); + + size += pool->gpu_contexts->obj->size; + + return size; +} diff --git a/drivers/gpu/drm/panthor/panthor_heap.h b/drivers/gpu/drm/panthor/panthor_heap.h index 25a5f2bba445..e3358d4e8edb 100644 --- a/drivers/gpu/drm/panthor/panthor_heap.h +++ b/drivers/gpu/drm/panthor/panthor_heap.h @@ -27,6 +27,8 @@ struct panthor_heap_pool * panthor_heap_pool_get(struct panthor_heap_pool *pool); void panthor_heap_pool_put(struct panthor_heap_pool *pool); +size_t panthor_heap_pool_size(struct panthor_heap_pool *pool); + int panthor_heap_grow(struct panthor_heap_pool *pool, u64 heap_gpu_va, u32 renderpasses_in_flight, diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c index 5ce80e2532d5..0a4e352b5505 100644 --- a/drivers/gpu/drm/panthor/panthor_mmu.c +++ b/drivers/gpu/drm/panthor/panthor_mmu.c @@ -1944,6 +1944,39 @@ struct panthor_heap_pool *panthor_vm_get_heap_pool(struct panthor_vm *vm, bool c return pool; } +/** + * panthor_vm_heaps_sizes() - Calculate size of all heap chunks across all + * heaps over all the heap pools in a VM + * @pfile: File. + * @stats: Memory stats to be updated. + * + * Calculate all heap chunk sizes in all heap pools bound to a VM. If the VM + * is active, record the size as active as well. + */ +void panthor_vm_heaps_sizes(struct panthor_file *pfile, struct drm_memory_stats *stats) +{ + struct panthor_vm *vm; + unsigned long i; + + if (!pfile->vms) + return; + + xa_lock(&pfile->vms->xa); + xa_for_each(&pfile->vms->xa, i, vm) { + size_t size = 0; + + mutex_lock(&vm->heaps.lock); + if (vm->heaps.pool) + size = panthor_heap_pool_size(vm->heaps.pool); + mutex_unlock(&vm->heaps.lock); + + stats->resident += size; + if (vm->as.id >= 0) + stats->active += size; + } + xa_unlock(&pfile->vms->xa); +} + static u64 mair_to_memattr(u64 mair, bool coherent) { u64 memattr = 0; diff --git a/drivers/gpu/drm/panthor/panthor_mmu.h b/drivers/gpu/drm/panthor/panthor_mmu.h index 8d21e83d8aba..fc274637114e 100644 --- a/drivers/gpu/drm/panthor/panthor_mmu.h +++ b/drivers/gpu/drm/panthor/panthor_mmu.h @@ -9,6 +9,7 @@ struct drm_exec; struct drm_sched_job; +struct drm_memory_stats; struct panthor_gem_object; struct panthor_heap_pool; struct panthor_vm; @@ -37,6 +38,8 @@ int panthor_vm_flush_all(struct panthor_vm *vm); struct panthor_heap_pool * panthor_vm_get_heap_pool(struct panthor_vm *vm, bool create); +void panthor_vm_heaps_sizes(struct panthor_file *pfile, struct drm_memory_stats *stats); + struct panthor_vm *panthor_vm_get(struct panthor_vm *vm); void panthor_vm_put(struct panthor_vm *vm); struct panthor_vm *panthor_vm_create(struct panthor_device *ptdev, bool for_mcu, diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c index 5844a7f639e0..da87e6e81938 100644 --- a/drivers/gpu/drm/panthor/panthor_sched.c +++ b/drivers/gpu/drm/panthor/panthor_sched.c @@ -625,7 +625,7 @@ struct panthor_group { */ struct panthor_kernel_bo *syncobjs; - /** @fdinfo: Per-file total cycle and timestamp values reference. */ + /** @fdinfo: Per-file info exposed through /proc//fdinfo */ struct { /** @data: Total sampled values for jobs in queues from this group. */ struct panthor_gpu_usage data; @@ -635,6 +635,9 @@ struct panthor_group { * and job post-completion processing function */ struct mutex lock; + + /** @fdinfo.kbo_sizes: Aggregate size of private kernel BO's held by the group. */ + size_t kbo_sizes; } fdinfo; /** @state: Group state. */ @@ -3378,6 +3381,29 @@ group_create_queue(struct panthor_group *group, return ERR_PTR(ret); } +static void add_group_kbo_sizes(struct panthor_device *ptdev, + struct panthor_group *group) +{ + struct panthor_queue *queue; + int i; + + if (drm_WARN_ON(&ptdev->base, IS_ERR_OR_NULL(group))) + return; + if (drm_WARN_ON(&ptdev->base, ptdev != group->ptdev)) + return; + + group->fdinfo.kbo_sizes += group->suspend_buf->obj->size; + group->fdinfo.kbo_sizes += group->protm_suspend_buf->obj->size; + group->fdinfo.kbo_sizes += group->syncobjs->obj->size; + + for (i = 0; i < group->queue_count; i++) { + queue = group->queues[i]; + group->fdinfo.kbo_sizes += queue->ringbuf->obj->size; + group->fdinfo.kbo_sizes += queue->iface.mem->obj->size; + group->fdinfo.kbo_sizes += queue->profiling.slots->obj->size; + } +} + #define MAX_GROUPS_PER_POOL 128 int panthor_group_create(struct panthor_file *pfile, @@ -3502,6 +3528,7 @@ int panthor_group_create(struct panthor_file *pfile, } mutex_unlock(&sched->reset.lock); + add_group_kbo_sizes(group->ptdev, group); mutex_init(&group->fdinfo.lock); return gid; @@ -3621,6 +3648,33 @@ void panthor_group_pool_destroy(struct panthor_file *pfile) pfile->groups = NULL; } +/** + * panthor_fdinfo_gather_group_mem_info() - Retrieve aggregate size of all private kernel BO's + * belonging to all the groups owned by an open Panthor file + * @pfile: File. + * @stats: Memory statistics to be updated. + * + */ +void +panthor_fdinfo_gather_group_mem_info(struct panthor_file *pfile, + struct drm_memory_stats *stats) +{ + struct panthor_group_pool *gpool = pfile->groups; + struct panthor_group *group; + unsigned long i; + + if (IS_ERR_OR_NULL(gpool)) + return; + + xa_lock(&gpool->xa); + xa_for_each(&gpool->xa, i, group) { + stats->resident += group->fdinfo.kbo_sizes; + if (group->csg_id >= 0) + stats->active += group->fdinfo.kbo_sizes; + } + xa_unlock(&gpool->xa); +} + static void job_release(struct kref *ref) { struct panthor_job *job = container_of(ref, struct panthor_job, refcount); diff --git a/drivers/gpu/drm/panthor/panthor_sched.h b/drivers/gpu/drm/panthor/panthor_sched.h index 5ae6b4bde7c5..e650a445cf50 100644 --- a/drivers/gpu/drm/panthor/panthor_sched.h +++ b/drivers/gpu/drm/panthor/panthor_sched.h @@ -9,6 +9,7 @@ struct dma_fence; struct drm_file; struct drm_gem_object; struct drm_sched_job; +struct drm_memory_stats; struct drm_panthor_group_create; struct drm_panthor_queue_create; struct drm_panthor_group_get_state; @@ -36,6 +37,8 @@ void panthor_job_update_resvs(struct drm_exec *exec, struct drm_sched_job *job); int panthor_group_pool_create(struct panthor_file *pfile); void panthor_group_pool_destroy(struct panthor_file *pfile); +void panthor_fdinfo_gather_group_mem_info(struct panthor_file *pfile, + struct drm_memory_stats *stats); int panthor_sched_init(struct panthor_device *ptdev); void panthor_sched_unplug(struct panthor_device *ptdev); From f780ef56f72912e39bdf0596e13c01d8e450dcc9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Adri=C3=A1n=20Larumbe?= Date: Thu, 30 Jan 2025 17:28:12 +0000 Subject: [PATCH 0635/1443] Documentation/gpu: Add fdinfo meanings of panthor-*-memory tags MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A previous commit enabled display of driver-internal kernel BO sizes through the device file's fdinfo interface. Expand the description of the relevant driver-specific key:value pairs with the definitions of the new panthor-*-memory ones. Reviewed-by: Mihail Atanassov Reviewed-by: Liviu Dudau Reviewed-by: Steven Price Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-5-adrian.larumbe@collabora.com --- Documentation/gpu/panthor.rst | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/Documentation/gpu/panthor.rst b/Documentation/gpu/panthor.rst index 3f8979fa2b86..7a841741278f 100644 --- a/Documentation/gpu/panthor.rst +++ b/Documentation/gpu/panthor.rst @@ -26,6 +26,8 @@ the currently possible format options: drm-cycles-panthor: 94439687187 drm-maxfreq-panthor: 1000000000 Hz drm-curfreq-panthor: 1000000000 Hz + panthor-resident-memory: 10396 KiB + panthor-active-memory: 10396 KiB drm-total-memory: 16480 KiB drm-shared-memory: 0 drm-active-memory: 16200 KiB @@ -44,3 +46,11 @@ driver by writing into the appropriate sysfs node:: Where `N` is a bit mask where cycle and timestamp sampling are respectively enabled by the first and second bits. + +Possible `panthor-*-memory` keys are: `active` and `resident`. +These values convey the sizes of the internal driver-owned shmem BO's that +aren't exposed to user-space through a DRM handle, like queue ring buffers, +sync object arrays and heap chunks. Because they are all allocated and pinned +at creation time, only `panthor-resident-memory` is necessary to tell us their +size. `panthor-active-memory` shows the size of kernel BO's associated with +VM's and groups currently being scheduled for execution by the GPU. From 0590c94c3596d6c1a3d549ae611366f2ad4e1d8d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Adri=C3=A1n=20Larumbe?= Date: Thu, 30 Jan 2025 17:28:13 +0000 Subject: [PATCH 0636/1443] drm/panthor: Fix race condition when gathering fdinfo group samples MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit e16635d88fa0 ("drm/panthor: add DRM fdinfo support") failed to protect access to groups with an xarray lock, which could lead to use-after-free errors. Fixes: e16635d88fa0 ("drm/panthor: add DRM fdinfo support") Reviewed-by: Boris Brezillon Reviewed-by: Steven Price Signed-off-by: Adrián Larumbe Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250130172851.941597-6-adrian.larumbe@collabora.com Link: https://patchwork.freedesktop.org/patch/msgid/20250107173310.88329-1-florent.tomasin@arm.com --- drivers/gpu/drm/panthor/panthor_sched.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c index da87e6e81938..0b93bf83a9b2 100644 --- a/drivers/gpu/drm/panthor/panthor_sched.c +++ b/drivers/gpu/drm/panthor/panthor_sched.c @@ -2878,6 +2878,7 @@ void panthor_fdinfo_gather_group_samples(struct panthor_file *pfile) if (IS_ERR_OR_NULL(gpool)) return; + xa_lock(&gpool->xa); xa_for_each(&gpool->xa, i, group) { mutex_lock(&group->fdinfo.lock); pfile->stats.cycles += group->fdinfo.data.cycles; @@ -2886,6 +2887,7 @@ void panthor_fdinfo_gather_group_samples(struct panthor_file *pfile) group->fdinfo.data.time = 0; mutex_unlock(&group->fdinfo.lock); } + xa_unlock(&gpool->xa); } static void group_sync_upd_work(struct work_struct *work) From 02458fbfaa0170aabf8506f7d4ed054f02414251 Mon Sep 17 00:00:00 2001 From: Rupinderjit Singh Date: Thu, 6 Feb 2025 15:58:03 +0000 Subject: [PATCH 0637/1443] gpu: host1x: Fix a use of uninitialized mutex commit c8347f915e67 ("gpu: host1x: Fix boot regression for Tegra") caused a use of uninitialized mutex leading to below warning when CONFIG_DEBUG_MUTEXES and CONFIG_DEBUG_LOCK_ALLOC are enabled. [ 41.662843] ------------[ cut here ]------------ [ 41.663012] DEBUG_LOCKS_WARN_ON(lock->magic != lock) [ 41.663035] WARNING: CPU: 4 PID: 794 at kernel/locking/mutex.c:587 __mutex_lock+0x670/0x878 [ 41.663458] Modules linked in: rtw88_8822c(+) bluetooth(+) rtw88_pci rtw88_core mac80211 aquantia libarc4 crc_itu_t cfg80211 tegra194_cpufreq dwmac_tegra(+) arm_dsu_pmu stmmac_platform stmmac pcs_xpcs rfkill at24 host1x(+) tegra_bpmp_thermal ramoops reed_solomon fuse loop nfnetlink xfs mmc_block rpmb_core ucsi_ccg ina3221 crct10dif_ce xhci_tegra ghash_ce lm90 sha2_ce sha256_arm64 sha1_ce sdhci_tegra pwm_fan sdhci_pltfm sdhci gpio_keys rtc_tegra cqhci mmc_core phy_tegra_xusb i2c_tegra tegra186_gpc_dma i2c_tegra_bpmp spi_tegra114 dm_mirror dm_region_hash dm_log dm_mod [ 41.665078] CPU: 4 UID: 0 PID: 794 Comm: (udev-worker) Not tainted 6.11.0-29.31_1538613708.el10.aarch64+debug #1 [ 41.665838] Hardware name: NVIDIA NVIDIA Jetson AGX Orin Developer Kit/Jetson, BIOS 36.3.0-gcid-35594366 02/26/2024 [ 41.672555] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 41.679636] pc : __mutex_lock+0x670/0x878 [ 41.683834] lr : __mutex_lock+0x670/0x878 [ 41.688035] sp : ffff800084b77090 [ 41.691446] x29: ffff800084b77160 x28: ffffdd4bebf7b000 x27: ffffdd4be96b1000 [ 41.698799] x26: 1fffe0002308361c x25: 1ffff0001096ee18 x24: 0000000000000000 [ 41.706149] x23: 0000000000000000 x22: 0000000000000002 x21: ffffdd4be6e3c7a0 [ 41.713500] x20: ffff800084b770f0 x19: ffff00011841b1e8 x18: 0000000000000000 [ 41.720675] x17: 0000000000000000 x16: 0000000000000000 x15: 0720072007200720 [ 41.728023] x14: 0000000000000000 x13: 0000000000000001 x12: ffff6001a96eaab3 [ 41.735375] x11: 1fffe001a96eaab2 x10: ffff6001a96eaab2 x9 : ffffdd4be4838bbc [ 41.742723] x8 : 00009ffe5691554e x7 : ffff000d4b755593 x6 : 0000000000000001 [ 41.749985] x5 : ffff000d4b755590 x4 : 1fffe0001d88f001 x3 : dfff800000000000 [ 41.756988] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000ec478000 [ 41.764251] Call trace: [ 41.766695] __mutex_lock+0x670/0x878 [ 41.770373] mutex_lock_nested+0x2c/0x40 [ 41.774134] host1x_intr_start+0x54/0xf8 [host1x] [ 41.778863] host1x_runtime_resume+0x150/0x228 [host1x] [ 41.783935] pm_generic_runtime_resume+0x84/0xc8 [ 41.788485] __rpm_callback+0xa0/0x478 [ 41.792422] rpm_callback+0x15c/0x1a8 [ 41.795922] rpm_resume+0x698/0xc08 [ 41.799597] __pm_runtime_resume+0xa8/0x140 [ 41.803621] host1x_probe+0x810/0xbc0 [host1x] [ 41.807909] platform_probe+0xcc/0x1a8 [ 41.811845] really_probe+0x188/0x800 [ 41.815347] __driver_probe_device+0x164/0x360 [ 41.819810] driver_probe_device+0x64/0x1a8 [ 41.823834] __driver_attach+0x180/0x490 [ 41.827773] bus_for_each_dev+0x104/0x1a0 [ 41.831797] driver_attach+0x44/0x68 [ 41.835296] bus_add_driver+0x23c/0x4e8 [ 41.839235] driver_register+0x15c/0x3a8 [ 41.843170] __platform_register_drivers+0xa4/0x208 [ 41.848159] tegra_host1x_init+0x4c/0xff8 [host1x] [ 41.853147] do_one_initcall+0xd4/0x380 [ 41.856997] do_init_module+0x1dc/0x698 [ 41.860758] load_module+0xc70/0x1300 [ 41.864435] __do_sys_init_module+0x1a8/0x1d0 [ 41.868721] __arm64_sys_init_module+0x74/0xb0 [ 41.873183] invoke_syscall.constprop.0+0xdc/0x1e8 [ 41.877997] do_el0_svc+0x154/0x1d0 [ 41.881671] el0_svc+0x54/0x140 [ 41.884820] el0t_64_sync_handler+0x120/0x130 [ 41.889285] el0t_64_sync+0x1a4/0x1a8 [ 41.892960] irq event stamp: 69737 [ 41.896370] hardirqs last enabled at (69737): [] _raw_spin_unlock_irqrestore+0x44/0xe8 [ 41.905739] hardirqs last disabled at (69736): [] clk_enable_lock+0x98/0x198 [ 41.914314] softirqs last enabled at (68082): [] handle_softirqs+0x4c8/0x890 [ 41.922977] softirqs last disabled at (67945): [] __do_softirq+0x1c/0x28 [ 41.931289] ---[ end trace 0000000000000000 ]--- Inside the probe function when pm_runtime_enable() is called, the PM core invokes a resume callback if the device Host1x is in a suspended state. As it can be seen in the logs above, this leads to host1x_intr_start() function call which is trying to acquire a mutex lock. But, the function host_intr_init() only gets called after the pm_runtime_enable() where mutex is initialised leading to the use of mutex prior to its initialisation. Fix this by moving the mutex initialisation prior to the runtime PM enablement function pm_runtime_enable() in probe. Fixes: c8347f915e67 ("gpu: host1x: Fix boot regression for Tegra") Signed-off-by: Rupinderjit Singh Reviewed-by: Jon Hunter Tested-by: Jon Hunter Signed-off-by: Thierry Reding Link: https://patchwork.ozlabs.org/project/linux-tegra/patch/20250206155803.201942-1-rusingh@redhat.com/ --- drivers/gpu/host1x/dev.c | 2 ++ drivers/gpu/host1x/intr.c | 2 -- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 7b1d091f3c09..46cae925b095 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -619,6 +619,8 @@ static int host1x_probe(struct platform_device *pdev) goto free_contexts; } + mutex_init(&host->intr_mutex); + pm_runtime_enable(&pdev->dev); err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); diff --git a/drivers/gpu/host1x/intr.c b/drivers/gpu/host1x/intr.c index b3285dd10180..f77a678949e9 100644 --- a/drivers/gpu/host1x/intr.c +++ b/drivers/gpu/host1x/intr.c @@ -104,8 +104,6 @@ int host1x_intr_init(struct host1x *host) unsigned int id; int i, err; - mutex_init(&host->intr_mutex); - for (id = 0; id < host1x_syncpt_nb_pts(host); ++id) { struct host1x_syncpt *syncpt = &host->syncpt[id]; From 3b32b7f638fe61e9d29290960172f4e360e38233 Mon Sep 17 00:00:00 2001 From: Su Hui Date: Sun, 19 Jan 2025 10:58:29 +0800 Subject: [PATCH 0638/1443] drm/panthor: avoid garbage value in panthor_ioctl_dev_query() 'priorities_info' is uninitialized, and the uninitialized value is copied to user object when calling PANTHOR_UOBJ_SET(). Using memset to initialize 'priorities_info' to avoid this garbage value problem. Fixes: f70000ef2352 ("drm/panthor: Add DEV_QUERY_GROUP_PRIORITIES_INFO dev query") Signed-off-by: Su Hui Reviewed-by: Dan Carpenter Reviewed-by: Boris Brezillon Reviewed-by: Steven Price Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250119025828.1168419-1-suhui@nfschina.com --- drivers/gpu/drm/panthor/panthor_drv.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/panthor/panthor_drv.c b/drivers/gpu/drm/panthor/panthor_drv.c index d5dcd3d1b33a..08136e790ca0 100644 --- a/drivers/gpu/drm/panthor/panthor_drv.c +++ b/drivers/gpu/drm/panthor/panthor_drv.c @@ -802,6 +802,7 @@ static void panthor_query_group_priorities_info(struct drm_file *file, { int prio; + memset(arg, 0, sizeof(*arg)); for (prio = PANTHOR_GROUP_PRIORITY_REALTIME; prio >= 0; prio--) { if (!group_priority_permit(file, prio)) arg->allowed_mask |= BIT(prio); From dbed4a797e00d98a4eaa971f1585091fad076a74 Mon Sep 17 00:00:00 2001 From: Jocelyn Falempe Date: Tue, 28 Jan 2025 17:52:10 +0100 Subject: [PATCH 0639/1443] drm/panic: Better binary encoding in QR code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current encoding, is done by converting 13bits of input into 4 decimal digits, that are then encoded efficiently using the numeric encoding of the QR code specification. The Fido v2.2 specification [1] uses a similar approach for its QR-initiated authentication. The only difference is that it converts 7 bytes (56bits) of input into 17 decimal digits. The benefit is that the algorithm doesn't require to split input bytes into 13bits chunk, and the ratio is a bit better. This improvement was proposed by Jó Ágila Bitsch in [2]. drm_panic is still young, and the QR code feature is not widely used, so it's still time to switch to a common algorithm, shared with a widely used standard. I also changed the name of the url parameter, from zl= to z=, so the website can keep backward compatibility if needed. [1] https://fidoalliance.org/specs/fido-v2.2-rd-20230321/fido-client-to-authenticator-protocol-v2.2-rd-20230321.html#hybrid-qr-initiated [2] https://github.com/kdj0c/panic_report/issues/2 Signed-off-by: Jocelyn Falempe Acked-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20250128165254.893204-1-jfalempe@redhat.com --- drivers/gpu/drm/drm_panic.c | 2 +- drivers/gpu/drm/drm_panic_qr.rs | 108 +++++++++++--------------------- 2 files changed, 36 insertions(+), 74 deletions(-) diff --git a/drivers/gpu/drm/drm_panic.c b/drivers/gpu/drm/drm_panic.c index f128d345b16d..ab42a2b1567d 100644 --- a/drivers/gpu/drm/drm_panic.c +++ b/drivers/gpu/drm/drm_panic.c @@ -499,7 +499,7 @@ static int drm_panic_get_qr_code_url(u8 **qr_image) char *kmsg; int max_qr_data_size, url_len; - url_len = snprintf(url, sizeof(url), CONFIG_DRM_PANIC_SCREEN_QR_CODE_URL "?a=%s&v=%s&zl=", + url_len = snprintf(url, sizeof(url), CONFIG_DRM_PANIC_SCREEN_QR_CODE_URL "?a=%s&v=%s&z=", utsname()->machine, utsname()->release); max_qr_data_size = drm_panic_qr_max_data_size(panic_qr_version, url_len); diff --git a/drivers/gpu/drm/drm_panic_qr.rs b/drivers/gpu/drm/drm_panic_qr.rs index bcf248f69252..62cb8a162483 100644 --- a/drivers/gpu/drm/drm_panic_qr.rs +++ b/drivers/gpu/drm/drm_panic_qr.rs @@ -13,12 +13,13 @@ //! The binary data must be a valid URL parameter, so the easiest way is //! to use base64 encoding. But this wastes 25% of data space, so the //! whole stack trace won't fit in the QR code. So instead it encodes -//! every 13bits of input into 4 decimal digits, and then uses the +//! every 7 bytes of input into 17 decimal digits, and then uses the //! efficient numeric encoding, that encode 3 decimal digits into -//! 10bits. This makes 39bits of compressed data into 12 decimal digits, -//! into 40bits in the QR code, so wasting only 2.5%. And the numbers are +//! 10bits. This makes 168bits of compressed data into 51 decimal digits, +//! into 170bits in the QR code, so wasting only 1.17%. And the numbers are //! valid URL parameter, so the website can do the reverse, to get the -//! binary data. +//! binary data. This is the same algorithm used by Fido v2.2 QR-initiated +//! authentication specification. //! //! Inspired by these 3 projects, all under MIT license: //! @@ -26,7 +27,6 @@ //! * //! * -use core::cmp; use kernel::str::CStr; #[derive(Debug, Clone, Copy, PartialEq, Eq, Ord, PartialOrd)] @@ -296,35 +296,11 @@ fn version_info(&self) -> u32 { /// Padding bytes. const PADDING: [u8; 2] = [236, 17]; -/// Get the next 13 bits of data, starting at specified offset (in bits). -fn get_next_13b(data: &[u8], offset: usize) -> Option<(u16, usize)> { - if offset < data.len() * 8 { - let size = cmp::min(13, data.len() * 8 - offset); - let byte_off = offset / 8; - let bit_off = offset % 8; - // `b` is 20 at max (`bit_off` <= 7 and `size` <= 13). - let b = (bit_off + size) as u16; - - let first_byte = (data[byte_off] << bit_off >> bit_off) as u16; - - let number = match b { - 0..=8 => first_byte >> (8 - b), - 9..=16 => (first_byte << (b - 8)) + (data[byte_off + 1] >> (16 - b)) as u16, - _ => { - (first_byte << (b - 8)) - + ((data[byte_off + 1] as u16) << (b - 16)) - + (data[byte_off + 2] >> (24 - b)) as u16 - } - }; - Some((number, size)) - } else { - None - } -} - /// Number of bits to encode characters in numeric mode. const NUM_CHARS_BITS: [usize; 4] = [0, 4, 7, 10]; -const POW10: [u16; 4] = [1, 10, 100, 1000]; +/// Number of decimal digits required to encode n bytes of binary data. +/// eg: you need 15 decimal digits to fit 6 bytes of binary data. +const BYTES_TO_DIGITS: [usize; 8] = [0, 3, 5, 8, 10, 13, 15, 17]; enum Segment<'a> { Numeric(&'a [u8]), @@ -360,13 +336,9 @@ fn character_count(&self) -> usize { match self { Segment::Binary(data) => data.len(), Segment::Numeric(data) => { - let data_bits = data.len() * 8; - let last_chars = match data_bits % 13 { - 1 => 1, - k => (k + 1) / 3, - }; - // 4 decimal numbers per 13bits + remainder. - 4 * (data_bits / 13) + last_chars + let last_chars = BYTES_TO_DIGITS[data.len() % 7]; + // 17 decimal numbers per 7bytes + remainder. + 17 * (data.len() / 7) + last_chars } } } @@ -403,7 +375,7 @@ fn iter(&self) -> SegmentIterator<'_> { struct SegmentIterator<'a> { segment: &'a Segment<'a>, offset: usize, - carry: u16, + carry: u64, carry_len: usize, } @@ -422,40 +394,30 @@ fn next(&mut self) -> Option { } } Segment::Numeric(data) => { - if self.carry_len == 3 { - let out = (self.carry, NUM_CHARS_BITS[self.carry_len]); - self.carry_len = 0; - self.carry = 0; - Some(out) - } else if let Some((bits, size)) = get_next_13b(data, self.offset) { - self.offset += size; - let new_chars = match size { - 1 => 1, - k => (k + 1) / 3, - }; - if self.carry_len + new_chars > 3 { - self.carry_len = new_chars + self.carry_len - 3; - let out = ( - self.carry * POW10[new_chars - self.carry_len] - + bits / POW10[self.carry_len], - NUM_CHARS_BITS[3], - ); - self.carry = bits % POW10[self.carry_len]; - Some(out) - } else { - let out = ( - self.carry * POW10[new_chars] + bits, - NUM_CHARS_BITS[self.carry_len + new_chars], - ); - self.carry_len = 0; - Some(out) + if self.carry_len < 3 && self.offset < data.len() { + // If there are less than 3 decimal digits in the carry, + // take the next 7 bytes of input, and add them to the carry. + let mut buf = [0u8; 8]; + let len = 7.min(data.len() - self.offset); + buf[..len].copy_from_slice(&data[self.offset..self.offset + len]); + let chunk = u64::from_le_bytes(buf); + let pow = u64::pow(10, BYTES_TO_DIGITS[len] as u32); + self.carry = chunk + self.carry * pow; + self.offset += len; + self.carry_len += BYTES_TO_DIGITS[len]; + } + match self.carry_len { + 0 => None, + len => { + // take the next 3 decimal digits of the carry + // and return 10bits of numeric data. + let out_len = 3.min(len); + self.carry_len -= out_len; + let pow = u64::pow(10, self.carry_len as u32); + let out = (self.carry / pow) as u16; + self.carry = self.carry % pow; + Some((out, NUM_CHARS_BITS[out_len])) } - } else if self.carry_len > 0 { - let out = (self.carry, NUM_CHARS_BITS[self.carry_len]); - self.carry_len = 0; - Some(out) - } else { - None } } } From 511121a48bbd12df4ae50a099a8936e833df8c46 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Wed, 5 Feb 2025 19:42:01 +0100 Subject: [PATCH 0640/1443] MAINTAINERS: Move Pavel to kernel.org address I need to filter my emails better, switch to pavel@kernel.org address to help with that. Signed-off-by: Pavel Machek Signed-off-by: Linus Torvalds --- CREDITS | 6 ++---- MAINTAINERS | 10 +++++----- 2 files changed, 7 insertions(+), 9 deletions(-) diff --git a/CREDITS b/CREDITS index 1f9f0f078b4a..53d11a46fd69 100644 --- a/CREDITS +++ b/CREDITS @@ -2515,11 +2515,9 @@ D: SLS distribution D: Initial implementation of VC's, pty's and select() N: Pavel Machek -E: pavel@ucw.cz +E: pavel@kernel.org P: 4096R/92DFCE96 4FA7 9EEF FCD4 C44F C585 B8C7 C060 2241 92DF CE96 -D: Softcursor for vga, hypertech cdrom support, vcsa bugfix, nbd, -D: sun4/330 port, capabilities for elf, speedup for rm on ext2, USB, -D: work on suspend-to-ram/disk, killing duplicates from ioctl32, +D: NBD, Sun4/330 port, USB, work on suspend-to-ram/disk, D: Altera SoCFPGA and Nokia N900 support. S: Czech Republic diff --git a/MAINTAINERS b/MAINTAINERS index 873aa2cce4d7..157818de0b55 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9418,7 +9418,7 @@ F: fs/freevxfs/ FREEZER M: "Rafael J. Wysocki" -M: Pavel Machek +M: Pavel Machek L: linux-pm@vger.kernel.org S: Supported F: Documentation/power/freezing-of-tasks.rst @@ -10253,7 +10253,7 @@ F: drivers/video/fbdev/hgafb.c HIBERNATION (aka Software Suspend, aka swsusp) M: "Rafael J. Wysocki" -M: Pavel Machek +M: Pavel Machek L: linux-pm@vger.kernel.org S: Supported B: https://bugzilla.kernel.org @@ -13124,8 +13124,8 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git for-next/har F: scripts/leaking_addresses.pl LED SUBSYSTEM -M: Pavel Machek M: Lee Jones +M: Pavel Machek L: linux-leds@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/lee/leds.git @@ -16823,7 +16823,7 @@ F: include/linux/tick.h F: kernel/time/tick*.* NOKIA N900 CAMERA SUPPORT (ET8EK8 SENSOR, AD5820 FOCUS) -M: Pavel Machek +M: Pavel Machek M: Sakari Ailus L: linux-media@vger.kernel.org S: Maintained @@ -22849,7 +22849,7 @@ F: drivers/sh/ SUSPEND TO RAM M: "Rafael J. Wysocki" M: Len Brown -M: Pavel Machek +M: Pavel Machek L: linux-pm@vger.kernel.org S: Supported B: https://bugzilla.kernel.org From 1b3291f00013c86a9bb349d6158a9a7a4f0334fe Mon Sep 17 00:00:00 2001 From: Hector Martin Date: Fri, 7 Feb 2025 03:21:46 +0900 Subject: [PATCH 0641/1443] MAINTAINERS: Remove myself I no longer have any faith left in the kernel development process or community management approach. Apple/ARM platform development will continue downstream. If I feel like sending some patches upstream in the future myself for whatever subtree I may, or I may not. Anyone who feels like fighting the upstreaming fight themselves is welcome to do so. Signed-off-by: Hector Martin Signed-off-by: Linus Torvalds --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 157818de0b55..20c9e0871215 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2209,7 +2209,6 @@ F: sound/soc/codecs/cs42l84.* F: sound/soc/codecs/ssm3515.c ARM/APPLE MACHINE SUPPORT -M: Hector Martin M: Sven Peter R: Alyssa Rosenzweig L: asahi@lists.linux.dev From 6bb05b3631bd81227298b291b651dc7ec6ee0145 Mon Sep 17 00:00:00 2001 From: Michal Wajdeczko Date: Thu, 6 Feb 2025 22:45:45 +0100 Subject: [PATCH 0642/1443] drm/xe: Enable SR-IOV for PTL MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We should now have sufficient changes in the driver to run it on PTL platforms in the SR-IOV Physical Function (PF) mode, that would allow us to enable SR-IOV Virtual Functions (VFs), and successfully probe our driver in the VF mode on enabled VF devices. To unblock SR-IOV PF mode you need to load xe with modparam: xe.max_vfs=7 Then to enable VFs it is sufficient to use: $ echo 7 > /sys/bus/pci/devices/0000:00:02.0/sriov_numvfs Note that in default auto-provisioning all VFs are allocated with some amount of shared resources (like unlimited GPU execution and preemption times, fair GGTT space, fair GuC context IDs range, ...) However with CONFIG_DEBUG_FS enabled it is possible to tweak most of the SR-IOV configuration parameters using attributes like: /sys/kernel/debug/dri/0000:00:02.0/gt0/ ├── pf │   ├── contexts_spare │   ├── doorbells_spare │   ├── exec_quantum_ms │   ├── ggtt_spare │   ├── preempt_timeout_us │   ├── sched_priority │   └── ... ├── vf1 │   ├── contexts_quota │   ├── doorbells_quota │   ├── exec_quantum_ms │   ├── ggtt_quota │   ├── preempt_timeout_us │   ├── sched_priority │   └── ... ├── vf2 │   └── ... : Signed-off-by: Michal Wajdeczko Cc: Rodrigo Vivi Cc: Lucas De Marchi Cc: Thomas Hellstrom Acked-by: Rodrigo Vivi Reviewed-by: Jakub Kolakowski Tested-by: Marcin Bernatowicz Reviewed-by: Marcin Bernatowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250206214545.940-1-michal.wajdeczko@intel.com --- drivers/gpu/drm/xe/xe_pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index cfa1ad7a1f2d..663bfc29cddc 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -363,6 +363,7 @@ static const struct xe_device_desc ptl_desc = { PLATFORM(PANTHERLAKE), .dma_mask_size = 46, .has_display = true, + .has_sriov = true, .require_force_probe = true, }; From 4f5f701c55c1dfa287c6ad55a4bcb4bc6cad974a Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Tue, 4 Feb 2025 17:28:24 -0600 Subject: [PATCH 0643/1443] drm/panthor: Convert IOCTL defines to an enum Use an enum instead of #defines for panthor IOCTLs. This allows the header to be used with Rust code as bindgen can't handle complex defines. Cc: Beata Michalska Signed-off-by: Rob Herring (Arm) Reviewed-by: Alice Ryhl Acked-by: Boris Brezillon Reviewed-by: Steven Price Signed-off-by: Boris Brezillon Link: https://patchwork.freedesktop.org/patch/msgid/20250204232824.3819437-1-robh@kernel.org --- include/uapi/drm/panthor_drm.h | 86 +++++++++++++++++----------------- 1 file changed, 44 insertions(+), 42 deletions(-) diff --git a/include/uapi/drm/panthor_drm.h b/include/uapi/drm/panthor_drm.h index b99763cbae48..97e2c4510e69 100644 --- a/include/uapi/drm/panthor_drm.h +++ b/include/uapi/drm/panthor_drm.h @@ -129,48 +129,6 @@ enum drm_panthor_ioctl_id { DRM_PANTHOR_TILER_HEAP_DESTROY, }; -/** - * DRM_IOCTL_PANTHOR() - Build a Panthor IOCTL number - * @__access: Access type. Must be R, W or RW. - * @__id: One of the DRM_PANTHOR_xxx id. - * @__type: Suffix of the type being passed to the IOCTL. - * - * Don't use this macro directly, use the DRM_IOCTL_PANTHOR_xxx - * values instead. - * - * Return: An IOCTL number to be passed to ioctl() from userspace. - */ -#define DRM_IOCTL_PANTHOR(__access, __id, __type) \ - DRM_IO ## __access(DRM_COMMAND_BASE + DRM_PANTHOR_ ## __id, \ - struct drm_panthor_ ## __type) - -#define DRM_IOCTL_PANTHOR_DEV_QUERY \ - DRM_IOCTL_PANTHOR(WR, DEV_QUERY, dev_query) -#define DRM_IOCTL_PANTHOR_VM_CREATE \ - DRM_IOCTL_PANTHOR(WR, VM_CREATE, vm_create) -#define DRM_IOCTL_PANTHOR_VM_DESTROY \ - DRM_IOCTL_PANTHOR(WR, VM_DESTROY, vm_destroy) -#define DRM_IOCTL_PANTHOR_VM_BIND \ - DRM_IOCTL_PANTHOR(WR, VM_BIND, vm_bind) -#define DRM_IOCTL_PANTHOR_VM_GET_STATE \ - DRM_IOCTL_PANTHOR(WR, VM_GET_STATE, vm_get_state) -#define DRM_IOCTL_PANTHOR_BO_CREATE \ - DRM_IOCTL_PANTHOR(WR, BO_CREATE, bo_create) -#define DRM_IOCTL_PANTHOR_BO_MMAP_OFFSET \ - DRM_IOCTL_PANTHOR(WR, BO_MMAP_OFFSET, bo_mmap_offset) -#define DRM_IOCTL_PANTHOR_GROUP_CREATE \ - DRM_IOCTL_PANTHOR(WR, GROUP_CREATE, group_create) -#define DRM_IOCTL_PANTHOR_GROUP_DESTROY \ - DRM_IOCTL_PANTHOR(WR, GROUP_DESTROY, group_destroy) -#define DRM_IOCTL_PANTHOR_GROUP_SUBMIT \ - DRM_IOCTL_PANTHOR(WR, GROUP_SUBMIT, group_submit) -#define DRM_IOCTL_PANTHOR_GROUP_GET_STATE \ - DRM_IOCTL_PANTHOR(WR, GROUP_GET_STATE, group_get_state) -#define DRM_IOCTL_PANTHOR_TILER_HEAP_CREATE \ - DRM_IOCTL_PANTHOR(WR, TILER_HEAP_CREATE, tiler_heap_create) -#define DRM_IOCTL_PANTHOR_TILER_HEAP_DESTROY \ - DRM_IOCTL_PANTHOR(WR, TILER_HEAP_DESTROY, tiler_heap_destroy) - /** * DOC: IOCTL arguments */ @@ -1019,6 +977,50 @@ struct drm_panthor_tiler_heap_destroy { __u32 pad; }; +/** + * DRM_IOCTL_PANTHOR() - Build a Panthor IOCTL number + * @__access: Access type. Must be R, W or RW. + * @__id: One of the DRM_PANTHOR_xxx id. + * @__type: Suffix of the type being passed to the IOCTL. + * + * Don't use this macro directly, use the DRM_IOCTL_PANTHOR_xxx + * values instead. + * + * Return: An IOCTL number to be passed to ioctl() from userspace. + */ +#define DRM_IOCTL_PANTHOR(__access, __id, __type) \ + DRM_IO ## __access(DRM_COMMAND_BASE + DRM_PANTHOR_ ## __id, \ + struct drm_panthor_ ## __type) + +enum { + DRM_IOCTL_PANTHOR_DEV_QUERY = + DRM_IOCTL_PANTHOR(WR, DEV_QUERY, dev_query), + DRM_IOCTL_PANTHOR_VM_CREATE = + DRM_IOCTL_PANTHOR(WR, VM_CREATE, vm_create), + DRM_IOCTL_PANTHOR_VM_DESTROY = + DRM_IOCTL_PANTHOR(WR, VM_DESTROY, vm_destroy), + DRM_IOCTL_PANTHOR_VM_BIND = + DRM_IOCTL_PANTHOR(WR, VM_BIND, vm_bind), + DRM_IOCTL_PANTHOR_VM_GET_STATE = + DRM_IOCTL_PANTHOR(WR, VM_GET_STATE, vm_get_state), + DRM_IOCTL_PANTHOR_BO_CREATE = + DRM_IOCTL_PANTHOR(WR, BO_CREATE, bo_create), + DRM_IOCTL_PANTHOR_BO_MMAP_OFFSET = + DRM_IOCTL_PANTHOR(WR, BO_MMAP_OFFSET, bo_mmap_offset), + DRM_IOCTL_PANTHOR_GROUP_CREATE = + DRM_IOCTL_PANTHOR(WR, GROUP_CREATE, group_create), + DRM_IOCTL_PANTHOR_GROUP_DESTROY = + DRM_IOCTL_PANTHOR(WR, GROUP_DESTROY, group_destroy), + DRM_IOCTL_PANTHOR_GROUP_SUBMIT = + DRM_IOCTL_PANTHOR(WR, GROUP_SUBMIT, group_submit), + DRM_IOCTL_PANTHOR_GROUP_GET_STATE = + DRM_IOCTL_PANTHOR(WR, GROUP_GET_STATE, group_get_state), + DRM_IOCTL_PANTHOR_TILER_HEAP_CREATE = + DRM_IOCTL_PANTHOR(WR, TILER_HEAP_CREATE, tiler_heap_create), + DRM_IOCTL_PANTHOR_TILER_HEAP_DESTROY = + DRM_IOCTL_PANTHOR(WR, TILER_HEAP_DESTROY, tiler_heap_destroy), +}; + #if defined(__cplusplus) } #endif From 75e80af7d62e0dde006c23aec67b9ec8037324db Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Thu, 6 Feb 2025 21:12:58 +0800 Subject: [PATCH 0644/1443] dt-bindings: vendor-prefixes: add csot Add "csot" to the Devicetree Vendor Prefix Registry. Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Acked-by: Conor Dooley Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250206131300.1295111-2-yelangyan@huaqin.corp-partner.google.com --- Documentation/devicetree/bindings/vendor-prefixes.yaml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml index 5079ca6ce1d1..88adeb7e28db 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.yaml +++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml @@ -338,6 +338,8 @@ patternProperties: description: Crystalfontz America, Inc. "^csky,.*": description: Hangzhou C-SKY Microsystems Co., Ltd + "^csot,.*": + description: Guangzhou China Star Optoelectronics Technology Co., Ltd "^csq,.*": description: Shenzen Chuangsiqi Technology Co.,Ltd. "^ctera,.*": From 36947218ca90c1f560f43c88c438124f8df884d4 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Thu, 6 Feb 2025 21:12:59 +0800 Subject: [PATCH 0645/1443] dt-bindings: display: panel: Add compatible for CSOT PNA957QT1-1 Add a new compatible for the panel CSOT PNA957QT1-1. This panel uses HX83102 IC, so add the compatible to the hx83102 binding files. Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Acked-by: Conor Dooley Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250206131300.1295111-3-yelangyan@huaqin.corp-partner.google.com --- .../devicetree/bindings/display/panel/himax,hx83102.yaml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml b/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml index c649fb085833..d876269e1fac 100644 --- a/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml +++ b/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml @@ -18,6 +18,8 @@ properties: - enum: # Boe nv110wum-l60 11.0" WUXGA TFT LCD panel - boe,nv110wum-l60 + # CSOT pna957qt1-1 10.95" WUXGA TFT LCD panel + - csot,pna957qt1-1 # IVO t109nw41 11.0" WUXGA TFT LCD panel - ivo,t109nw41 # STARRY himax83102-j02 10.51" WUXGA TFT LCD panel From 05345cea4ff5a857612df3f10144dec685c07e6d Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Thu, 6 Feb 2025 21:13:00 +0800 Subject: [PATCH 0646/1443] drm/panel: panel-himax-hx83102: support for csot-pna957qt1-1 MIPI-DSI panel The csot-pna957qt1-1 is a 10.95" TFT panel. The MIPI controller on this panel is the same as the other panels here, so add this panel to this driver. Signed-off-by: Langyan Ye Reviewed-by: Neil Armstrong Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250206131300.1295111-4-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-himax-hx83102.c | 123 ++++++++++++++++++++ 1 file changed, 123 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-himax-hx83102.c b/drivers/gpu/drm/panel/panel-himax-hx83102.c index 3644a7544b93..c3f4f5ee175d 100644 --- a/drivers/gpu/drm/panel/panel-himax-hx83102.c +++ b/drivers/gpu/drm/panel/panel-himax-hx83102.c @@ -43,6 +43,7 @@ #define HX83102_SETGIP1 0xd5 #define HX83102_SETGIP2 0xd6 #define HX83102_SETGIP3 0xd8 +#define HX83102_UNKNOWN_D9 0xd9 #define HX83102_SETGMA 0xe0 #define HX83102_UNKNOWN_E1 0xe1 #define HX83102_SETTP1 0xe7 @@ -291,6 +292,103 @@ static int boe_nv110wum_init(struct hx83102 *ctx) return dsi_ctx.accum_err; }; +static int csot_pna957qt1_1_init(struct hx83102 *ctx) +{ + struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; + + msleep(60); + + hx83102_enable_extended_cmds(&dsi_ctx, true); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D9, 0xd2); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x2c, 0xb3, 0xb3, 0x31, 0xf1, 0x33, + 0xe0, 0x54, 0x36, 0x36, 0x3a, 0x3a, 0x32, 0x8b, 0x11, 0xe5, + 0x98); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xd9); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x8b, 0x33); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETDISP, 0x00, 0x47, 0xb0, 0x80, 0x00, 0x2c, + 0x80, 0x3c, 0x9f, 0x22, 0x20, 0x00, 0x00, 0x98, 0x51); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x41, 0x41, 0x41, 0x41, 0x64, 0x64, + 0x40, 0x84, 0x64, 0x84, 0x01, 0x9d, 0x01, 0x02, 0x01, 0x00, + 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETVDC, 0x1b, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_BE, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPTBA, 0xfc, 0xc4, 0x80, 0x9c, 0x36, 0x00, + 0x0d, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSTBA, 0x32, 0x32, 0x22, 0x11, 0x22, 0xa0, + 0x31, 0x08, 0xf5, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcc); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTCON, 0x80); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETRAMDMY, 0x97); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPWM, 0x00, 0x1e, 0x13, 0x88, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x08, 0x13, 0x07, 0x00, 0x0f, + 0x36); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPANEL, 0x02, 0x03, 0x44); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPCTRL, 0x07, 0x06, 0x00, 0x02, 0x04, 0x2c, + 0xff); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x06, 0x00, 0x00, 0x00, 0x40, 0x04, + 0x08, 0x04, 0x08, 0x37, 0x07, 0x44, 0x37, 0x2b, 0x2b, 0x03, + 0x03, 0x32, 0x10, 0x22, 0x00, 0x25, 0x32, 0x10, 0x29, 0x00, + 0x29, 0x32, 0x10, 0x08, 0x00, 0x08, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP1, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, + 0x18, 0x18, 0x18, 0x18, 0x07, 0x06, 0x07, 0x06, 0x05, 0x04, + 0x05, 0x04, 0x03, 0x02, 0x03, 0x02, 0x01, 0x00, 0x01, 0x00, + 0x18, 0x18, 0x25, 0x24, 0x25, 0x24, 0x1f, 0x1f, 0x1f, 0x1f, + 0x1e, 0x1e, 0x1e, 0x1e, 0x20, 0x20, 0x20, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGMA, 0x0a, 0x0e, 0x1a, 0x21, 0x28, 0x46, + 0x5c, 0x61, 0x63, 0x5e, 0x78, 0x7d, 0x80, 0x8e, 0x89, 0x90, + 0x98, 0xaa, 0xa8, 0x52, 0x59, 0x60, 0x6f, 0x06, 0x0a, 0x16, + 0x1d, 0x24, 0x46, 0x5c, 0x61, 0x6b, 0x66, 0x7c, 0x7d, 0x80, + 0x8e, 0x89, 0x90, 0x98, 0xaa, 0xa8, 0x52, 0x59, 0x60, 0x6f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0xe0, 0x10, 0x10, 0x0d, 0x1e, 0x9d, + 0x02, 0x52, 0x9d, 0x14, 0x14); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x01, 0x7f, 0x11, 0xfd); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x4f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x86); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D2, 0x64); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, 0x05, 0x15, 0x55, 0x45, + 0x55, 0x50, 0x05, 0x15, 0x55, 0x45, 0x55, 0x50); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0x02, 0x00, 0x24, 0x01, 0x7e, 0x0f, + 0x7c, 0x10, 0xa0, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x03, 0x07, 0x00, 0x10, 0x7b); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0f, 0x3f, 0xff, 0xcf, 0xff, 0xf0, + 0x0f, 0x3f, 0xff, 0xcf, 0xff, 0xf0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0xfe, 0x01, 0xfe, 0x01, 0xfe, 0x01, + 0x00, 0x00, 0x00, 0x23, 0x00, 0x23, 0x81, 0x02, 0x40, 0x00, + 0x20, 0x9d, 0x02, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x01, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETDISP, 0x66, 0x81); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x03, 0xff, 0xf8); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, 0x0f, 0x2a, 0xaa, 0x8a, + 0xaa, 0xf0, 0x0f, 0x2a, 0xaa, 0x8a, 0xaa, 0xf0, 0x0a, 0x2a, + 0xaa, 0x8a, 0xaa, 0xa0, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x00); + hx83102_enable_extended_cmds(&dsi_ctx, false); + + mipi_dsi_msleep(&dsi_ctx, 60); + + return dsi_ctx.accum_err; +}; + static int ivo_t109nw41_init(struct hx83102 *ctx) { struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; @@ -440,6 +538,28 @@ static const struct hx83102_panel_desc boe_nv110wum_desc = { .init = boe_nv110wum_init, }; +static const struct drm_display_mode csot_pna957qt1_1_default_mode = { + .clock = 177958, + .hdisplay = 1200, + .hsync_start = 1200 + 124, + .hsync_end = 1200 + 124 + 80, + .htotal = 1200 + 124 + 80 + 40, + .vdisplay = 1920, + .vsync_start = 1920 + 88, + .vsync_end = 1920 + 88 + 8, + .vtotal = 1920 + 88 + 8 + 38, + .type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED, +}; + +static const struct hx83102_panel_desc csot_pna957qt1_1_desc = { + .modes = &csot_pna957qt1_1_default_mode, + .size = { + .width_mm = 147, + .height_mm = 235, + }, + .init = csot_pna957qt1_1_init, +}; + static const struct drm_display_mode ivo_t109nw41_default_mode = { .clock = 167700, .hdisplay = 1200, @@ -683,6 +803,9 @@ static const struct of_device_id hx83102_of_match[] = { { .compatible = "boe,nv110wum-l60", .data = &boe_nv110wum_desc }, + { .compatible = "csot,pna957qt1-1", + .data = &csot_pna957qt1_1_desc + }, { .compatible = "ivo,t109nw41", .data = &ivo_t109nw41_desc }, From f354fc88a72ae83dacd68370f6fa040e5733bcfe Mon Sep 17 00:00:00 2001 From: WangYuli Date: Fri, 7 Feb 2025 15:08:55 +0800 Subject: [PATCH 0647/1443] kbuild: install-extmod-build: add missing quotation marks for CC variable While attempting to build a Debian packages with CC="ccache gcc", I saw the following error as builddeb builds linux-headers-$KERNELVERSION: make HOSTCC=ccache gcc VPATH= srcroot=. -f ./scripts/Makefile.build obj=debian/linux-headers-6.14.0-rc1/usr/src/linux-headers-6.14.0-rc1/scripts make[6]: *** No rule to make target 'gcc'. Stop. Upon investigation, it seems that one instance of $(CC) variable reference in ./scripts/package/install-extmod-build was missing quotation marks, causing the above error. Add the missing quotation marks around $(CC) to fix build. Fixes: 5f73e7d0386d ("kbuild: refactor cross-compiling linux-headers package") Co-developed-by: Mingcong Bai Signed-off-by: Mingcong Bai Tested-by: WangYuli Signed-off-by: WangYuli Signed-off-by: Masahiro Yamada --- scripts/package/install-extmod-build | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/package/install-extmod-build b/scripts/package/install-extmod-build index bb6e23c1174e..b724626ea0ca 100755 --- a/scripts/package/install-extmod-build +++ b/scripts/package/install-extmod-build @@ -63,7 +63,7 @@ if [ "${CC}" != "${HOSTCC}" ]; then # Clear VPATH and srcroot because the source files reside in the output # directory. # shellcheck disable=SC2016 # $(MAKE), $(CC), and $(build) will be expanded by Make - "${MAKE}" run-command KBUILD_RUN_COMMAND='+$(MAKE) HOSTCC=$(CC) VPATH= srcroot=. $(build)='"${destdir}"/scripts + "${MAKE}" run-command KBUILD_RUN_COMMAND='+$(MAKE) HOSTCC="$(CC)" VPATH= srcroot=. $(build)='"${destdir}"/scripts rm -f "${destdir}/scripts/Kbuild" fi From 595170d4b660640289003d1881a9a6ef8ded6865 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 6 Feb 2025 22:28:41 -0500 Subject: [PATCH 0648/1443] bcachefs: Fix marking reflink pointers to missing indirect extents reflink pointers to missing indirect extents aren't deleted, they just have an error bit set - in case the indirect extent somehow reappears. fsck/mark and sweep thus needs to ignore these errors. Also, they can be marked AUTOFIX now. Reported-by: Roland Vet Signed-off-by: Kent Overstreet --- fs/bcachefs/reflink.c | 2 ++ fs/bcachefs/sb-errors_format.h | 4 ++-- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c index 93ba4f4e47ca..376fd0a6e868 100644 --- a/fs/bcachefs/reflink.c +++ b/fs/bcachefs/reflink.c @@ -381,6 +381,8 @@ static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans, not_found: if (flags & BTREE_TRIGGER_check_repair) { ret = bch2_indirect_extent_missing_error(trans, p, *idx, next_idx, false); + if (ret == -BCH_ERR_missing_indirect_extent) + ret = 0; if (ret) goto err; } diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h index ea0a18364751..b86ec013d7d7 100644 --- a/fs/bcachefs/sb-errors_format.h +++ b/fs/bcachefs/sb-errors_format.h @@ -180,9 +180,9 @@ enum bch_fsck_flags { x(ptr_crc_nonce_mismatch, 162, 0) \ x(ptr_stripe_redundant, 163, 0) \ x(reservation_key_nr_replicas_invalid, 164, 0) \ - x(reflink_v_refcount_wrong, 165, 0) \ + x(reflink_v_refcount_wrong, 165, FSCK_AUTOFIX) \ x(reflink_v_pos_bad, 292, 0) \ - x(reflink_p_to_missing_reflink_v, 166, 0) \ + x(reflink_p_to_missing_reflink_v, 166, FSCK_AUTOFIX) \ x(reflink_refcount_underflow, 293, 0) \ x(stripe_pos_bad, 167, 0) \ x(stripe_val_size_bad, 168, 0) \ From 01af106a076352182b2916b143fc50272600bd81 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 21 Jan 2025 05:40:51 +0000 Subject: [PATCH 0649/1443] btrfs: fix two misuses of folio_shift() It is meaningless to shift a byte count by folio_shift(). The folio index is in units of PAGE_SIZE, not folio_size(). We can use folio_contains() to make this work for arbitrary-order folios, so remove the assertion that the folios are of order 0. Reviewed-by: Qu Wenruo Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent_io.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d9f856358704..6f64ee16744d 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -523,8 +523,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) u64 end; u32 len; - /* For now only order 0 folios are supported for data. */ - ASSERT(folio_order(folio) == 0); btrfs_debug(fs_info, "%s: bi_sector=%llu, err=%d, mirror=%u", __func__, bio->bi_iter.bi_sector, bio->bi_status, @@ -552,7 +550,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) if (likely(uptodate)) { loff_t i_size = i_size_read(inode); - pgoff_t end_index = i_size >> folio_shift(folio); /* * Zero out the remaining part if this range straddles @@ -561,9 +558,11 @@ static void end_bbio_data_read(struct btrfs_bio *bbio) * Here we should only zero the range inside the folio, * not touch anything else. * - * NOTE: i_size is exclusive while end is inclusive. + * NOTE: i_size is exclusive while end is inclusive and + * folio_contains() takes PAGE_SIZE units. */ - if (folio_index(folio) == end_index && i_size <= end) { + if (folio_contains(folio, i_size >> PAGE_SHIFT) && + i_size <= end) { u32 zero_start = max(offset_in_folio(folio, i_size), offset_in_folio(folio, start)); u32 zero_len = offset_in_folio(folio, end) + 1 - @@ -956,7 +955,7 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached, return ret; } - if (folio->index == last_byte >> folio_shift(folio)) { + if (folio_contains(folio, last_byte >> PAGE_SHIFT)) { size_t zero_offset = offset_in_folio(folio, last_byte); if (zero_offset) { From cb827db50a88aebec516151681adb6db10b688ee Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 6 Feb 2025 08:30:51 +0000 Subject: [PATCH 0650/1443] net: fib_rules: annotate data-races around rule->[io]ifindex rule->iifindex and rule->oifindex can be read without holding RTNL. Add READ_ONCE()/WRITE_ONCE() annotations where needed. Fixes: 32affa5578f0 ("fib: rules: no longer hold RTNL in fib_nl_dumprule()") Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20250206083051.2494877-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/fib_rules.c | 24 +++++++++++++----------- 1 file changed, 13 insertions(+), 11 deletions(-) diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index e684ba3ebb38..94a7872ab231 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -37,8 +37,8 @@ static const struct fib_kuid_range fib_kuid_range_unset = { bool fib_rule_matchall(const struct fib_rule *rule) { - if (rule->iifindex || rule->oifindex || rule->mark || rule->tun_id || - rule->flags) + if (READ_ONCE(rule->iifindex) || READ_ONCE(rule->oifindex) || + rule->mark || rule->tun_id || rule->flags) return false; if (rule->suppress_ifgroup != -1 || rule->suppress_prefixlen != -1) return false; @@ -261,12 +261,14 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, struct flowi *fl, int flags, struct fib_lookup_arg *arg) { - int ret = 0; + int iifindex, oifindex, ret = 0; - if (rule->iifindex && (rule->iifindex != fl->flowi_iif)) + iifindex = READ_ONCE(rule->iifindex); + if (iifindex && (iifindex != fl->flowi_iif)) goto out; - if (rule->oifindex && (rule->oifindex != fl->flowi_oif)) + oifindex = READ_ONCE(rule->oifindex); + if (oifindex && (oifindex != fl->flowi_oif)) goto out; if ((rule->mark ^ fl->flowi_mark) & rule->mark_mask) @@ -1041,14 +1043,14 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, if (rule->iifname[0]) { if (nla_put_string(skb, FRA_IIFNAME, rule->iifname)) goto nla_put_failure; - if (rule->iifindex == -1) + if (READ_ONCE(rule->iifindex) == -1) frh->flags |= FIB_RULE_IIF_DETACHED; } if (rule->oifname[0]) { if (nla_put_string(skb, FRA_OIFNAME, rule->oifname)) goto nla_put_failure; - if (rule->oifindex == -1) + if (READ_ONCE(rule->oifindex) == -1) frh->flags |= FIB_RULE_OIF_DETACHED; } @@ -1220,10 +1222,10 @@ static void attach_rules(struct list_head *rules, struct net_device *dev) list_for_each_entry(rule, rules, list) { if (rule->iifindex == -1 && strcmp(dev->name, rule->iifname) == 0) - rule->iifindex = dev->ifindex; + WRITE_ONCE(rule->iifindex, dev->ifindex); if (rule->oifindex == -1 && strcmp(dev->name, rule->oifname) == 0) - rule->oifindex = dev->ifindex; + WRITE_ONCE(rule->oifindex, dev->ifindex); } } @@ -1233,9 +1235,9 @@ static void detach_rules(struct list_head *rules, struct net_device *dev) list_for_each_entry(rule, rules, list) { if (rule->iifindex == dev->ifindex) - rule->iifindex = -1; + WRITE_ONCE(rule->iifindex, -1); if (rule->oifindex == dev->ifindex) - rule->oifindex = -1; + WRITE_ONCE(rule->oifindex, -1); } } From db5fd3cf8bf41b84b577b8ad5234ea95f327c9be Mon Sep 17 00:00:00 2001 From: Muhammad Adeel Date: Fri, 7 Feb 2025 14:24:32 +0000 Subject: [PATCH 0651/1443] cgroup: Remove steal time from usage_usec MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The CPU usage time is the time when user, system or both are using the CPU. Steal time is the time when CPU is waiting to be run by the Hypervisor. It should not be added to the CPU usage time, hence removing it from the usage_usec entry. Fixes: 936f2a70f2077 ("cgroup: add cpu.stat file to root cgroup") Acked-by: Axel Busch Acked-by: Michal Koutný Signed-off-by: Muhammad Adeel Signed-off-by: Tejun Heo --- kernel/cgroup/rstat.c | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c index 5877974ece92..aac91466279f 100644 --- a/kernel/cgroup/rstat.c +++ b/kernel/cgroup/rstat.c @@ -590,7 +590,6 @@ static void root_cgroup_cputime(struct cgroup_base_stat *bstat) cputime->sum_exec_runtime += user; cputime->sum_exec_runtime += sys; - cputime->sum_exec_runtime += cpustat[CPUTIME_STEAL]; #ifdef CONFIG_SCHED_CORE bstat->forceidle_sum += cpustat[CPUTIME_FORCEIDLE]; From 884c3a18dadfda326dffa364477cc027728219de Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Tue, 4 Feb 2025 10:25:16 -0700 Subject: [PATCH 0652/1443] bpf: verifier: Do not extract constant map keys for irrelevant maps Previously, we were trying to extract constant map keys for all bpf_map_lookup_elem(), regardless of map type. This is an issue if the map has a u64 key and the value is very high, as it can be interpreted as a negative signed value. This in turn is treated as an error value by check_func_arg() which causes a valid program to be incorrectly rejected. Fix by only extracting constant map keys for relevant maps. This fix works because nullness elision is only allowed for {PERCPU_}ARRAY maps, and keys for these are within u32 range. See next commit for an example via selftest. Acked-by: Eduard Zingerman Reported-by: Marc Hartmayer Reported-by: Ilya Leoshkevich Tested-by: Marc Hartmayer Signed-off-by: Daniel Xu Link: https://lore.kernel.org/r/aa868b642b026ff87ba6105ea151bc8693b35932.1738689872.git.dxu@dxuuu.xyz Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 9971c03adfd5..e9176a5ce215 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9206,6 +9206,8 @@ static s64 get_constant_map_key(struct bpf_verifier_env *env, return reg->var_off.value; } +static bool can_elide_value_nullness(enum bpf_map_type type); + static int check_func_arg(struct bpf_verifier_env *env, u32 arg, struct bpf_call_arg_meta *meta, const struct bpf_func_proto *fn, @@ -9354,9 +9356,11 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 arg, err = check_helper_mem_access(env, regno, key_size, BPF_READ, false, NULL); if (err) return err; - meta->const_map_key = get_constant_map_key(env, reg, key_size); - if (meta->const_map_key < 0 && meta->const_map_key != -EOPNOTSUPP) - return meta->const_map_key; + if (can_elide_value_nullness(meta->map_ptr->map_type)) { + meta->const_map_key = get_constant_map_key(env, reg, key_size); + if (meta->const_map_key < 0 && meta->const_map_key != -EOPNOTSUPP) + return meta->const_map_key; + } break; case ARG_PTR_TO_MAP_VALUE: if (type_may_be_null(arg_type) && register_is_null(reg)) From 973cb1382ead401c476c82f20525e593ae84788f Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Tue, 4 Feb 2025 10:25:17 -0700 Subject: [PATCH 0653/1443] bpf: selftests: Test constant key extraction on irrelevant maps Test that very high constant map keys are not interpreted as an error value by the verifier. This would previously fail. Acked-by: Eduard Zingerman Signed-off-by: Daniel Xu Link: https://lore.kernel.org/r/c0590b62eb9303f389b2f52c0c7e9cf22a358a30.1738689872.git.dxu@dxuuu.xyz Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/progs/verifier_array_access.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/verifier_array_access.c b/tools/testing/selftests/bpf/progs/verifier_array_access.c index 29eb9568633f..0a187ff725cc 100644 --- a/tools/testing/selftests/bpf/progs/verifier_array_access.c +++ b/tools/testing/selftests/bpf/progs/verifier_array_access.c @@ -713,4 +713,19 @@ unsigned int non_stack_key_lookup(void) return val->index; } +SEC("socket") +__description("doesn't reject UINT64_MAX as s64 for irrelevant maps") +__success __retval(42) +unsigned int doesnt_reject_irrelevant_maps(void) +{ + __u64 key = 0xFFFFFFFFFFFFFFFF; + struct test_val *val; + + val = bpf_map_lookup_elem(&map_hash_48b, &key); + if (val) + return val->index; + + return 42; +} + char _license[] SEC("license") = "GPL"; From 7968c6581507052c1c6484ee6c5cbe07381e2dbc Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Tue, 4 Feb 2025 10:25:18 -0700 Subject: [PATCH 0654/1443] bpf: verifier: Disambiguate get_constant_map_key() errors Refactor get_constant_map_key() to disambiguate the constant key value from potential error values. In the case that the key is negative, it could be confused for an error. It's not currently an issue, as the verifier seems to track s32 spills as u32. So even if the program wrongly uses a negative value for an arraymap key, the verifier just thinks it's an impossibly high value which gets correctly discarded. Refactor anyways to make things cleaner and prevent potential future issues. Acked-by: Eduard Zingerman Signed-off-by: Daniel Xu Link: https://lore.kernel.org/r/dfe144259ae7cfc98aa63e1b388a14869a10632a.1738689872.git.dxu@dxuuu.xyz Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e9176a5ce215..98354d781678 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9149,10 +9149,11 @@ static int check_reg_const_str(struct bpf_verifier_env *env, return 0; } -/* Returns constant key value if possible, else negative error */ -static s64 get_constant_map_key(struct bpf_verifier_env *env, +/* Returns constant key value in `value` if possible, else negative error */ +static int get_constant_map_key(struct bpf_verifier_env *env, struct bpf_reg_state *key, - u32 key_size) + u32 key_size, + s64 *value) { struct bpf_func_state *state = func(env, key); struct bpf_reg_state *reg; @@ -9179,8 +9180,10 @@ static s64 get_constant_map_key(struct bpf_verifier_env *env, /* First handle precisely tracked STACK_ZERO */ for (i = off; i >= 0 && stype[i] == STACK_ZERO; i--) zero_size++; - if (zero_size >= key_size) + if (zero_size >= key_size) { + *value = 0; return 0; + } /* Check that stack contains a scalar spill of expected size */ if (!is_spilled_scalar_reg(&state->stack[spi])) @@ -9203,7 +9206,8 @@ static s64 get_constant_map_key(struct bpf_verifier_env *env, if (err < 0) return err; - return reg->var_off.value; + *value = reg->var_off.value; + return 0; } static bool can_elide_value_nullness(enum bpf_map_type type); @@ -9357,9 +9361,14 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 arg, if (err) return err; if (can_elide_value_nullness(meta->map_ptr->map_type)) { - meta->const_map_key = get_constant_map_key(env, reg, key_size); - if (meta->const_map_key < 0 && meta->const_map_key != -EOPNOTSUPP) - return meta->const_map_key; + err = get_constant_map_key(env, reg, key_size, &meta->const_map_key); + if (err < 0) { + meta->const_map_key = -1; + if (err == -EOPNOTSUPP) + err = 0; + else + return err; + } } break; case ARG_PTR_TO_MAP_VALUE: From 011b0335903832facca86cd8ed05d7d8d94c9c76 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Thu, 6 Feb 2025 22:28:48 +0100 Subject: [PATCH 0655/1443] Revert "net: skb: introduce and use a single page frag cache" This reverts commit dbae2b062824 ("net: skb: introduce and use a single page frag cache"). The intended goal of such change was to counter a performance regression introduced by commit 3226b158e67c ("net: avoid 32 x truesize under-estimation for tiny skbs"). Unfortunately, the blamed commit introduces another regression for the virtio_net driver. Such a driver calls napi_alloc_skb() with a tiny size, so that the whole head frag could fit a 512-byte block. The single page frag cache uses a 1K fragment for such allocation, and the additional overhead, under small UDP packets flood, makes the page allocator a bottleneck. Thanks to commit bf9f1baa279f ("net: add dedicated kmem_cache for typical/small skb->head"), this revert does not re-introduce the original regression. Actually, in the relevant test on top of this revert, I measure a small but noticeable positive delta, just above noise level. The revert itself required some additional mangling due to the introduction of the SKB_HEAD_ALIGN() helper and local lock infra in the affected code. Suggested-by: Eric Dumazet Fixes: dbae2b062824 ("net: skb: introduce and use a single page frag cache") Signed-off-by: Paolo Abeni Link: https://patch.msgid.link/e649212fde9f0fdee23909ca0d14158d32bb7425.1738877290.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 1 - net/core/dev.c | 17 +++++++ net/core/skbuff.c | 103 ++------------------------------------ 3 files changed, 22 insertions(+), 99 deletions(-) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c0a86afb85da..365f0e2098d1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4115,7 +4115,6 @@ void netif_receive_skb_list(struct list_head *head); gro_result_t napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb); void napi_gro_flush(struct napi_struct *napi, bool flush_old); struct sk_buff *napi_get_frags(struct napi_struct *napi); -void napi_get_frags_check(struct napi_struct *napi); gro_result_t napi_gro_frags(struct napi_struct *napi); static inline void napi_free_frags(struct napi_struct *napi) diff --git a/net/core/dev.c b/net/core/dev.c index b91658e8aedb..55e356a68db6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6920,6 +6920,23 @@ netif_napi_dev_list_add(struct net_device *dev, struct napi_struct *napi) list_add_rcu(&napi->dev_list, higher); /* adds after higher */ } +/* Double check that napi_get_frags() allocates skbs with + * skb->head being backed by slab, not a page fragment. + * This is to make sure bug fixed in 3226b158e67c + * ("net: avoid 32 x truesize under-estimation for tiny skbs") + * does not accidentally come back. + */ +static void napi_get_frags_check(struct napi_struct *napi) +{ + struct sk_buff *skb; + + local_bh_disable(); + skb = napi_get_frags(napi); + WARN_ON_ONCE(skb && skb->head_frag); + napi_free_frags(napi); + local_bh_enable(); +} + void netif_napi_add_weight_locked(struct net_device *dev, struct napi_struct *napi, int (*poll)(struct napi_struct *, int), diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a441613a1e6c..6a99c453397f 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -220,67 +220,9 @@ static void skb_under_panic(struct sk_buff *skb, unsigned int sz, void *addr) #define NAPI_SKB_CACHE_BULK 16 #define NAPI_SKB_CACHE_HALF (NAPI_SKB_CACHE_SIZE / 2) -#if PAGE_SIZE == SZ_4K - -#define NAPI_HAS_SMALL_PAGE_FRAG 1 -#define NAPI_SMALL_PAGE_PFMEMALLOC(nc) ((nc).pfmemalloc) - -/* specialized page frag allocator using a single order 0 page - * and slicing it into 1K sized fragment. Constrained to systems - * with a very limited amount of 1K fragments fitting a single - * page - to avoid excessive truesize underestimation - */ - -struct page_frag_1k { - void *va; - u16 offset; - bool pfmemalloc; -}; - -static void *page_frag_alloc_1k(struct page_frag_1k *nc, gfp_t gfp) -{ - struct page *page; - int offset; - - offset = nc->offset - SZ_1K; - if (likely(offset >= 0)) - goto use_frag; - - page = alloc_pages_node(NUMA_NO_NODE, gfp, 0); - if (!page) - return NULL; - - nc->va = page_address(page); - nc->pfmemalloc = page_is_pfmemalloc(page); - offset = PAGE_SIZE - SZ_1K; - page_ref_add(page, offset / SZ_1K); - -use_frag: - nc->offset = offset; - return nc->va + offset; -} -#else - -/* the small page is actually unused in this build; add dummy helpers - * to please the compiler and avoid later preprocessor's conditionals - */ -#define NAPI_HAS_SMALL_PAGE_FRAG 0 -#define NAPI_SMALL_PAGE_PFMEMALLOC(nc) false - -struct page_frag_1k { -}; - -static void *page_frag_alloc_1k(struct page_frag_1k *nc, gfp_t gfp_mask) -{ - return NULL; -} - -#endif - struct napi_alloc_cache { local_lock_t bh_lock; struct page_frag_cache page; - struct page_frag_1k page_small; unsigned int skb_count; void *skb_cache[NAPI_SKB_CACHE_SIZE]; }; @@ -290,23 +232,6 @@ static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache) = { .bh_lock = INIT_LOCAL_LOCK(bh_lock), }; -/* Double check that napi_get_frags() allocates skbs with - * skb->head being backed by slab, not a page fragment. - * This is to make sure bug fixed in 3226b158e67c - * ("net: avoid 32 x truesize under-estimation for tiny skbs") - * does not accidentally come back. - */ -void napi_get_frags_check(struct napi_struct *napi) -{ - struct sk_buff *skb; - - local_bh_disable(); - skb = napi_get_frags(napi); - WARN_ON_ONCE(!NAPI_HAS_SMALL_PAGE_FRAG && skb && skb->head_frag); - napi_free_frags(napi); - local_bh_enable(); -} - void *__napi_alloc_frag_align(unsigned int fragsz, unsigned int align_mask) { struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache); @@ -813,10 +738,8 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len) /* If requested length is either too small or too big, * we use kmalloc() for skb->head allocation. - * When the small frag allocator is available, prefer it over kmalloc - * for small fragments */ - if ((!NAPI_HAS_SMALL_PAGE_FRAG && len <= SKB_WITH_OVERHEAD(1024)) || + if (len <= SKB_WITH_OVERHEAD(1024) || len > SKB_WITH_OVERHEAD(PAGE_SIZE) || (gfp_mask & (__GFP_DIRECT_RECLAIM | GFP_DMA))) { skb = __alloc_skb(len, gfp_mask, SKB_ALLOC_RX | SKB_ALLOC_NAPI, @@ -826,32 +749,16 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len) goto skb_success; } + len = SKB_HEAD_ALIGN(len); + if (sk_memalloc_socks()) gfp_mask |= __GFP_MEMALLOC; local_lock_nested_bh(&napi_alloc_cache.bh_lock); nc = this_cpu_ptr(&napi_alloc_cache); - if (NAPI_HAS_SMALL_PAGE_FRAG && len <= SKB_WITH_OVERHEAD(1024)) { - /* we are artificially inflating the allocation size, but - * that is not as bad as it may look like, as: - * - 'len' less than GRO_MAX_HEAD makes little sense - * - On most systems, larger 'len' values lead to fragment - * size above 512 bytes - * - kmalloc would use the kmalloc-1k slab for such values - * - Builds with smaller GRO_MAX_HEAD will very likely do - * little networking, as that implies no WiFi and no - * tunnels support, and 32 bits arches. - */ - len = SZ_1K; - data = page_frag_alloc_1k(&nc->page_small, gfp_mask); - pfmemalloc = NAPI_SMALL_PAGE_PFMEMALLOC(nc->page_small); - } else { - len = SKB_HEAD_ALIGN(len); - - data = page_frag_alloc(&nc->page, len, gfp_mask); - pfmemalloc = page_frag_cache_is_pfmemalloc(&nc->page); - } + data = page_frag_alloc(&nc->page, len, gfp_mask); + pfmemalloc = page_frag_cache_is_pfmemalloc(&nc->page); local_unlock_nested_bh(&napi_alloc_cache.bh_lock); if (unlikely(!data)) From 8784714d7f27045c7cb72456cf66705b73fbc804 Mon Sep 17 00:00:00 2001 From: Kumar Kartikeya Dwivedi Date: Thu, 6 Feb 2025 02:54:31 -0800 Subject: [PATCH 0656/1443] bpf: Handle allocation failure in acquire_lock_state The acquire_lock_state function needs to handle possible NULL values returned by acquire_reference_state, and return -ENOMEM. Fixes: 769b0f1c8214 ("bpf: Refactor {acquire,release}_reference_state") Signed-off-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20250206105435.2159977-24-memxor@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 98354d781678..60611df77957 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1501,6 +1501,8 @@ static int acquire_lock_state(struct bpf_verifier_env *env, int insn_idx, enum r struct bpf_reference_state *s; s = acquire_reference_state(env, insn_idx); + if (!s) + return -ENOMEM; s->type = type; s->id = id; s->ptr = ptr; From 8f6629c004b193d23612641c3607e785819e97ab Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Thu, 17 Oct 2024 10:09:22 -0700 Subject: [PATCH 0657/1443] kbuild: Move -Wenum-enum-conversion to W=2 -Wenum-enum-conversion was strengthened in clang-19 to warn for C, which caused the kernel to move it to W=1 in commit 75b5ab134bb5 ("kbuild: Move -Wenum-{compare-conditional,enum-conversion} into W=1") because there were numerous instances that would break builds with -Werror. Unfortunately, this is not a full solution, as more and more developers, subsystems, and distributors are building with W=1 as well, so they continue to see the numerous instances of this warning. Since the move to W=1, there have not been many new instances that have appeared through various build reports and the ones that have appeared seem to be following similar existing patterns, suggesting that most instances of this warning will not be real issues. The only alternatives for silencing this warning are adding casts (which is generally seen as an ugly practice) or refactoring the enums to macro defines or a unified enum (which may be undesirable because of type safety in other parts of the code). Move the warning to W=2, where warnings that occur frequently but may be relevant should reside. Cc: stable@vger.kernel.org Fixes: 75b5ab134bb5 ("kbuild: Move -Wenum-{compare-conditional,enum-conversion} into W=1") Link: https://lore.kernel.org/ZwRA9SOcOjjLJcpi@google.com/ Signed-off-by: Nathan Chancellor Acked-by: Arnd Bergmann Signed-off-by: Linus Torvalds --- scripts/Makefile.extrawarn | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn index eb719f6d8d53..a7003c1e66c7 100644 --- a/scripts/Makefile.extrawarn +++ b/scripts/Makefile.extrawarn @@ -133,7 +133,6 @@ KBUILD_CFLAGS += $(call cc-disable-warning, pointer-to-enum-cast) KBUILD_CFLAGS += -Wno-tautological-constant-out-of-range-compare KBUILD_CFLAGS += $(call cc-disable-warning, unaligned-access) KBUILD_CFLAGS += -Wno-enum-compare-conditional -KBUILD_CFLAGS += -Wno-enum-enum-conversion endif endif @@ -157,6 +156,10 @@ KBUILD_CFLAGS += -Wno-missing-field-initializers KBUILD_CFLAGS += -Wno-type-limits KBUILD_CFLAGS += -Wno-shift-negative-value +ifdef CONFIG_CC_IS_CLANG +KBUILD_CFLAGS += -Wno-enum-enum-conversion +endif + ifdef CONFIG_CC_IS_GCC KBUILD_CFLAGS += -Wno-maybe-uninitialized endif From d0b197b6505fe3788860fc2a81b3ce53cbecc69c Mon Sep 17 00:00:00 2001 From: Reyders Morales Date: Mon, 3 Feb 2025 23:47:20 +0100 Subject: [PATCH 0658/1443] Documentation/networking: fix basic node example document ISO 15765-2 In the current struct sockaddr_can tp is member of can_addr. tp is not member of struct sockaddr_can. Signed-off-by: Reyders Morales Reviewed-by: Simon Horman Acked-by: Oliver Hartkopp Link: https://patch.msgid.link/20250203224720.42530-1-reyders1@gmail.com Fixes: 67711e04254c ("Documentation: networking: document ISO 15765-2") Signed-off-by: Marc Kleine-Budde --- Documentation/networking/iso15765-2.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/networking/iso15765-2.rst b/Documentation/networking/iso15765-2.rst index 0e9d96074178..37ebb2c417cb 100644 --- a/Documentation/networking/iso15765-2.rst +++ b/Documentation/networking/iso15765-2.rst @@ -369,8 +369,8 @@ to their default. addr.can_family = AF_CAN; addr.can_ifindex = if_nametoindex("can0"); - addr.tp.tx_id = 0x18DA42F1 | CAN_EFF_FLAG; - addr.tp.rx_id = 0x18DAF142 | CAN_EFF_FLAG; + addr.can_addr.tp.tx_id = 0x18DA42F1 | CAN_EFF_FLAG; + addr.can_addr.tp.rx_id = 0x18DAF142 | CAN_EFF_FLAG; ret = bind(s, (struct sockaddr *)&addr, sizeof(addr)); if (ret < 0) From 44de577e61ed239db09f0da9d436866bef9b77dd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Alexander=20H=C3=B6lzl?= Date: Wed, 5 Feb 2025 18:46:51 +0100 Subject: [PATCH 0659/1443] can: j1939: j1939_sk_send_loop(): fix unable to send messages with data length zero MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The J1939 standard requires the transmission of messages of length 0. For example proprietary messages are specified with a data length of 0 to 1785. The transmission of such messages is not possible. Sending results in no error being returned but no corresponding can frame being generated. Enable the transmission of zero length J1939 messages. In order to facilitate this two changes are necessary: 1) If the transmission of a new message is requested from user space the message is segmented in j1939_sk_send_loop(). Let the segmentation take into account zero length messages, do not terminate immediately, queue the corresponding skb. 2) j1939_session_skb_get_by_offset() selects the next skb to transmit for a session. Take into account that there might be zero length skbs in the queue. Signed-off-by: Alexander Hölzl Acked-by: Oleksij Rempel Link: https://patch.msgid.link/20250205174651.103238-1-alexander.hoelzl@gmx.net Fixes: 9d71dd0c7009 ("can: add support of SAE J1939 protocol") Cc: stable@vger.kernel.org [mkl: commit message rephrased] Signed-off-by: Marc Kleine-Budde --- net/can/j1939/socket.c | 4 ++-- net/can/j1939/transport.c | 5 +++-- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c index 305dd72c844c..17226b2341d0 100644 --- a/net/can/j1939/socket.c +++ b/net/can/j1939/socket.c @@ -1132,7 +1132,7 @@ static int j1939_sk_send_loop(struct j1939_priv *priv, struct sock *sk, todo_size = size; - while (todo_size) { + do { struct j1939_sk_buff_cb *skcb; segment_size = min_t(size_t, J1939_MAX_TP_PACKET_SIZE, @@ -1177,7 +1177,7 @@ static int j1939_sk_send_loop(struct j1939_priv *priv, struct sock *sk, todo_size -= segment_size; session->total_queued_size += segment_size; - } + } while (todo_size); switch (ret) { case 0: /* OK */ diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c index 95f7a7e65a73..9b72d118d756 100644 --- a/net/can/j1939/transport.c +++ b/net/can/j1939/transport.c @@ -382,8 +382,9 @@ sk_buff *j1939_session_skb_get_by_offset(struct j1939_session *session, skb_queue_walk(&session->skb_queue, do_skb) { do_skcb = j1939_skb_to_cb(do_skb); - if (offset_start >= do_skcb->offset && - offset_start < (do_skcb->offset + do_skb->len)) { + if ((offset_start >= do_skcb->offset && + offset_start < (do_skcb->offset + do_skb->len)) || + (offset_start == 0 && do_skcb->offset == 0 && do_skb->len == 0)) { skb = do_skb; } } From 9bd24927e3eeb85642c7baa3b28be8bea6c2a078 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Tue, 14 Jan 2025 18:21:38 +0300 Subject: [PATCH 0660/1443] can: ctucanfd: handle skb allocation failure If skb allocation fails, the pointer to struct can_frame is NULL. This is actually handled everywhere inside ctucan_err_interrupt() except for the only place. Add the missed NULL check. Found by Linux Verification Center (linuxtesting.org) with SVACE static analysis tool. Fixes: 2dcb8e8782d8 ("can: ctucanfd: add support for CTU CAN FD open-source IP core - bus independent part.") Cc: stable@vger.kernel.org Signed-off-by: Fedor Pchelkin Acked-by: Pavel Pisa Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20250114152138.139580-1-pchelkin@ispras.ru Signed-off-by: Marc Kleine-Budde --- drivers/net/can/ctucanfd/ctucanfd_base.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/can/ctucanfd/ctucanfd_base.c b/drivers/net/can/ctucanfd/ctucanfd_base.c index 64c349fd4600..f65c1a1e05cc 100644 --- a/drivers/net/can/ctucanfd/ctucanfd_base.c +++ b/drivers/net/can/ctucanfd/ctucanfd_base.c @@ -867,10 +867,12 @@ static void ctucan_err_interrupt(struct net_device *ndev, u32 isr) } break; case CAN_STATE_ERROR_ACTIVE: - cf->can_id |= CAN_ERR_CNT; - cf->data[1] = CAN_ERR_CRTL_ACTIVE; - cf->data[6] = bec.txerr; - cf->data[7] = bec.rxerr; + if (skb) { + cf->can_id |= CAN_ERR_CNT; + cf->data[1] = CAN_ERR_CRTL_ACTIVE; + cf->data[6] = bec.txerr; + cf->data[7] = bec.rxerr; + } break; default: netdev_warn(ndev, "unhandled error state (%d:%s)!\n", From 257a2cd3eb578ee63d6bf90475dc4f4b16984139 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 12 Jan 2025 13:41:52 +0100 Subject: [PATCH 0661/1443] can: c_can: fix unbalanced runtime PM disable in error path Runtime PM is enabled as one of the last steps of probe(), so all earlier gotos to "exit_free_device" label were not correct and were leading to unbalanced runtime PM disable depth. Fixes: 6e2fe01dd6f9 ("can: c_can: move runtime PM enable/disable to c_can_platform") Cc: stable@vger.kernel.org Signed-off-by: Krzysztof Kozlowski Reviewed-by: Vincent Mailhol Link: https://patch.msgid.link/20250112-syscon-phandle-args-can-v1-1-314d9549906f@linaro.org Signed-off-by: Marc Kleine-Budde --- drivers/net/can/c_can/c_can_platform.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/c_can/c_can_platform.c b/drivers/net/can/c_can/c_can_platform.c index 6cba9717a6d8..399844809bbe 100644 --- a/drivers/net/can/c_can/c_can_platform.c +++ b/drivers/net/can/c_can/c_can_platform.c @@ -385,15 +385,16 @@ static int c_can_plat_probe(struct platform_device *pdev) if (ret) { dev_err(&pdev->dev, "registering %s failed (err=%d)\n", KBUILD_MODNAME, ret); - goto exit_free_device; + goto exit_pm_runtime; } dev_info(&pdev->dev, "%s device registered (regs=%p, irq=%d)\n", KBUILD_MODNAME, priv->base, dev->irq); return 0; -exit_free_device: +exit_pm_runtime: pm_runtime_disable(priv->device); +exit_free_device: free_c_can_dev(dev); exit: dev_err(&pdev->dev, "probe failed\n"); From a1ad2109ce41c9e3912dadd07ad8a9c640064ffb Mon Sep 17 00:00:00 2001 From: Vincent Mailhol Date: Wed, 5 Feb 2025 00:48:15 +0900 Subject: [PATCH 0662/1443] can: etas_es58x: fix potential NULL pointer dereference on udev->serial The driver assumed that es58x_dev->udev->serial could never be NULL. While this is true on commercially available devices, an attacker could spoof the device identity providing a NULL USB serial number. That would trigger a NULL pointer dereference. Add a check on es58x_dev->udev->serial before accessing it. Reported-by: yan kang Reported-by: yue sun Closes: https://lore.kernel.org/linux-can/SY8P300MB0421E0013C0EBD2AA46BA709A1F42@SY8P300MB0421.AUSP300.PROD.OUTLOOK.COM/ Fixes: 9f06631c3f1f ("can: etas_es58x: export product information through devlink_ops::info_get()") Signed-off-by: Vincent Mailhol Link: https://patch.msgid.link/20250204154859.9797-2-mailhol.vincent@wanadoo.fr Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- drivers/net/can/usb/etas_es58x/es58x_devlink.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/can/usb/etas_es58x/es58x_devlink.c b/drivers/net/can/usb/etas_es58x/es58x_devlink.c index eee20839d96f..0d155eb1b9e9 100644 --- a/drivers/net/can/usb/etas_es58x/es58x_devlink.c +++ b/drivers/net/can/usb/etas_es58x/es58x_devlink.c @@ -248,7 +248,11 @@ static int es58x_devlink_info_get(struct devlink *devlink, return ret; } - return devlink_info_serial_number_put(req, es58x_dev->udev->serial); + if (es58x_dev->udev->serial) + ret = devlink_info_serial_number_put(req, + es58x_dev->udev->serial); + + return ret; } const struct devlink_ops es58x_dl_ops = { From f7f0adfe64de08803990dc4cbecd2849c04e314a Mon Sep 17 00:00:00 2001 From: Robin van der Gracht Date: Mon, 27 Jan 2025 13:16:44 +0100 Subject: [PATCH 0663/1443] can: rockchip: rkcanfd_handle_rx_fifo_overflow_int(): bail out if skb cannot be allocated Fix NULL pointer check in rkcanfd_handle_rx_fifo_overflow_int() to bail out if skb cannot be allocated. Fixes: ff60bfbaf67f ("can: rockchip_canfd: add driver for Rockchip CAN-FD controller") Cc: stable@vger.kernel.org Signed-off-by: Robin van der Gracht Link: https://patch.msgid.link/20250208-fix-rockchip-canfd-v1-1-ec533c8a9895@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rockchip/rockchip_canfd-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/rockchip/rockchip_canfd-core.c b/drivers/net/can/rockchip/rockchip_canfd-core.c index df18c85fc078..d9a937ba126c 100644 --- a/drivers/net/can/rockchip/rockchip_canfd-core.c +++ b/drivers/net/can/rockchip/rockchip_canfd-core.c @@ -622,7 +622,7 @@ rkcanfd_handle_rx_fifo_overflow_int(struct rkcanfd_priv *priv) netdev_dbg(priv->ndev, "RX-FIFO overflow\n"); skb = rkcanfd_alloc_can_err_skb(priv, &cf, ×tamp); - if (skb) + if (!skb) return 0; rkcanfd_get_berr_counter_corrected(priv, &bec); From c8c9b1d2d5b4377c72a979f5a26e842a869aefc9 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Sat, 8 Feb 2025 00:15:11 -0500 Subject: [PATCH 0664/1443] fgraph: Fix set_graph_notrace with setting TRACE_GRAPH_NOTRACE_BIT The code was restructured where the function graph notrace code, that would not trace a function and all its children is done by setting a NOTRACE flag when the function that is not to be traced is hit. There's a TRACE_GRAPH_NOTRACE_BIT which defines the bit in the flags and a TRACE_GRAPH_NOTRACE which is the mask with that bit set. But the restructuring used TRACE_GRAPH_NOTRACE_BIT when it should have used TRACE_GRAPH_NOTRACE. For example: # cd /sys/kernel/tracing # echo set_track_prepare stack_trace_save > set_graph_notrace # echo function_graph > current_tracer # cat trace [..] 0) | __slab_free() { 0) | free_to_partial_list() { 0) | arch_stack_walk() { 0) | __unwind_start() { 0) 0.501 us | get_stack_info(); Where a non filter trace looks like: # echo > set_graph_notrace # cat trace 0) | free_to_partial_list() { 0) | set_track_prepare() { 0) | stack_trace_save() { 0) | arch_stack_walk() { 0) | __unwind_start() { Where the filter should look like: # cat trace 0) | free_to_partial_list() { 0) | _raw_spin_lock_irqsave() { 0) 0.350 us | preempt_count_add(); 0) 0.351 us | do_raw_spin_lock(); 0) 2.440 us | } Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Link: https://lore.kernel.org/20250208001511.535be150@batman.local.home Fixes: b84214890a9bc ("function_graph: Move graph notrace bit to shadow stack global var") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace_functions_graph.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 54d850997c0a..136c750b0b4d 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -198,7 +198,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace, * returning from the function. */ if (ftrace_graph_notrace_addr(trace->func)) { - *task_var |= TRACE_GRAPH_NOTRACE_BIT; + *task_var |= TRACE_GRAPH_NOTRACE; /* * Need to return 1 to have the return called * that will clear the NOTRACE bit. From 3724062ca2b1364f02cf44dbea1a552227844ad1 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Tue, 14 Jan 2025 13:57:58 -0800 Subject: [PATCH 0665/1443] objtool: Ignore dangling jump table entries Clang sometimes leaves dangling unused jump table entries which point to the end of the function. Ignore them. Closes: https://lore.kernel.org/20250113235835.vqgvb7cdspksy5dn@jpoimboe Reported-by: Klaus Kusche Signed-off-by: Josh Poimboeuf Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/ee25c0b7e80113e950bd1d4c208b671d35774ff4.1736891751.git.jpoimboe@kernel.org --- tools/objtool/check.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 753dbc4f8198..3520a45ebde8 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -1975,6 +1975,14 @@ static int add_jump_table(struct objtool_file *file, struct instruction *insn, reloc_addend(reloc) == pfunc->offset) break; + /* + * Clang sometimes leaves dangling unused jump table entries + * which point to the end of the function. Ignore them. + */ + if (reloc->sym->sec == pfunc->sec && + reloc_addend(reloc) == pfunc->offset + pfunc->len) + goto next; + dest_insn = find_insn(file, reloc->sym->sec, reloc_addend(reloc)); if (!dest_insn) break; @@ -1992,6 +2000,7 @@ static int add_jump_table(struct objtool_file *file, struct instruction *insn, alt->insn = dest_insn; alt->next = insn->alts; insn->alts = alt; +next: prev_offset = reloc_offset(reloc); } From 7e501637bd5b702a2fa627e903a0025654110e1e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 6 Feb 2025 11:12:08 +0100 Subject: [PATCH 0666/1443] objtool: Move dodgy linker warn to verbose The lld.ld borkage is fixed in the latest llvm release (?) but will not be backported, meaning we're stuck with broken linker for a fair while. Lets not spam all clang build logs and move warning to verbose. Signed-off-by: Peter Zijlstra (Intel) --- tools/objtool/check.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 3520a45ebde8..497cb8dfb3eb 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -2273,7 +2273,7 @@ static int read_annotate(struct objtool_file *file, if (sec->sh.sh_entsize != 8) { static bool warned = false; - if (!warned) { + if (!warned && opts.verbose) { WARN("%s: dodgy linker, sh_entsize != 8", sec->name); warned = true; } From bcc6244e13b4d4903511a1ea84368abf925031c0 Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Wed, 29 Jan 2025 20:53:03 +0100 Subject: [PATCH 0667/1443] sched: Clarify wake_up_q()'s write to task->wake_q.next Clarify that wake_up_q() does an atomic write to task->wake_q.next, after which a concurrent __wake_q_add() can immediately overwrite task->wake_q.next again. Signed-off-by: Jann Horn Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20250129-sched-wakeup-prettier-v1-1-2f51f5f663fa@google.com --- kernel/sched/core.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 3e5a6bf587f9..8931d9b1e895 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1055,9 +1055,10 @@ void wake_up_q(struct wake_q_head *head) struct task_struct *task; task = container_of(node, struct task_struct, wake_q); - /* Task can safely be re-inserted now: */ node = node->next; - task->wake_q.next = NULL; + /* pairs with cmpxchg_relaxed() in __wake_q_add() */ + WRITE_ONCE(task->wake_q.next, NULL); + /* Task can safely be re-inserted now. */ /* * wake_up_process() executes a full barrier, which pairs with From 469c76a83bb9f6b2c7b2989c46617c4fe01fee79 Mon Sep 17 00:00:00 2001 From: Dhananjay Ugwekar Date: Wed, 29 Jan 2025 08:05:14 +0000 Subject: [PATCH 0668/1443] perf/x86/rapl: Fix the error checking order After the commit b4943b8bfc41 ("perf/x86/rapl: Add core energy counter support for AMD CPUs"), the default "perf record"/"perf top" command is broken in systems where there isn't a PMU registered for type PERF_TYPE_RAW. This is due to the change in order of error checks in rapl_pmu_event_init() Due to which we return -EINVAL instead of -ENOENT, when we reach here from the fallback loop in perf_init_event(). Move the "PMU and event type match" back to the beginning of the function so that we return -ENOENT early on. Closes: https://lore.kernel.org/all/uv7mz6vew2bzgre5jdpmwldxljp5djzmuiksqdcdwipfm4zm7w@ribobcretidk/ Fixes: b4943b8bfc41 ("perf/x86/rapl: Add core energy counter support for AMD CPUs") Reported-by: Koichiro Den Signed-off-by: Dhananjay Ugwekar Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20250129080513.30353-1-dhananjay.ugwekar@amd.com --- arch/x86/events/rapl.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/arch/x86/events/rapl.c b/arch/x86/events/rapl.c index d3bb3865c1b1..4952faf03e82 100644 --- a/arch/x86/events/rapl.c +++ b/arch/x86/events/rapl.c @@ -370,6 +370,10 @@ static int rapl_pmu_event_init(struct perf_event *event) unsigned int rapl_pmu_idx; struct rapl_pmus *rapl_pmus; + /* only look at RAPL events */ + if (event->attr.type != event->pmu->type) + return -ENOENT; + /* unsupported modes and filters */ if (event->attr.sample_period) /* no sampling */ return -EINVAL; @@ -387,10 +391,6 @@ static int rapl_pmu_event_init(struct perf_event *event) rapl_pmus_scope = rapl_pmus->pmu.scope; if (rapl_pmus_scope == PERF_PMU_SCOPE_PKG || rapl_pmus_scope == PERF_PMU_SCOPE_DIE) { - /* only look at RAPL package events */ - if (event->attr.type != rapl_pmus_pkg->pmu.type) - return -ENOENT; - cfg = array_index_nospec((long)cfg, NR_RAPL_PKG_DOMAINS + 1); if (!cfg || cfg >= NR_RAPL_PKG_DOMAINS + 1) return -EINVAL; @@ -398,10 +398,6 @@ static int rapl_pmu_event_init(struct perf_event *event) bit = cfg - 1; event->hw.event_base = rapl_model->rapl_pkg_msrs[bit].msr; } else if (rapl_pmus_scope == PERF_PMU_SCOPE_CORE) { - /* only look at RAPL core events */ - if (event->attr.type != rapl_pmus_core->pmu.type) - return -ENOENT; - cfg = array_index_nospec((long)cfg, NR_RAPL_CORE_DOMAINS + 1); if (!cfg || cfg >= NR_RAPL_PKG_DOMAINS + 1) return -EINVAL; From 0a5561501397e2bbd0fb0e300eb489f72a90597a Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Wed, 29 Jan 2025 07:48:18 -0800 Subject: [PATCH 0669/1443] perf/x86/intel: Clean up PEBS-via-PT on hybrid The PEBS-via-PT feature is exposed for the e-core of some hybrid platforms, e.g., ADL and MTL. But it never works. $ dmesg | grep PEBS [ 1.793888] core: cpu_atom PMU driver: PEBS-via-PT $ perf record -c 1000 -e '{intel_pt/branch=0/, cpu_atom/cpu-cycles,aux-output/pp}' -C8 Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (cpu_atom/cpu-cycles,aux-output/pp). "dmesg | grep -i perf" may provide additional information. The "PEBS-via-PT" is printed if the corresponding bit of per-PMU capabilities is set. Since the feature is supported by the e-core HW, perf sets the bit for e-core. However, for Intel PT, if a feature is not supported on all CPUs, it is not supported at all. The PEBS-via-PT event cannot be created successfully. The PEBS-via-PT is no longer enumerated on the latest hybrid platform. It will be deprecated on future platforms with Arch PEBS. Let's remove it from the existing hybrid platforms. Fixes: d9977c43bff8 ("perf/x86: Register hybrid PMUs") Signed-off-by: Kan Liang Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20250129154820.3755948-2-kan.liang@linux.intel.com --- arch/x86/events/intel/core.c | 10 ---------- arch/x86/events/intel/ds.c | 10 +++++++++- 2 files changed, 9 insertions(+), 11 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 7601196d1d18..966f7832497d 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -4941,11 +4941,6 @@ static void intel_pmu_check_hybrid_pmus(struct x86_hybrid_pmu *pmu) else pmu->intel_ctrl &= ~(1ULL << GLOBAL_CTRL_EN_PERF_METRICS); - if (pmu->intel_cap.pebs_output_pt_available) - pmu->pmu.capabilities |= PERF_PMU_CAP_AUX_OUTPUT; - else - pmu->pmu.capabilities &= ~PERF_PMU_CAP_AUX_OUTPUT; - intel_pmu_check_event_constraints(pmu->event_constraints, pmu->cntr_mask64, pmu->fixed_cntr_mask64, @@ -5023,9 +5018,6 @@ static bool init_hybrid_pmu(int cpu) pr_info("%s PMU driver: ", pmu->name); - if (pmu->intel_cap.pebs_output_pt_available) - pr_cont("PEBS-via-PT "); - pr_cont("\n"); x86_pmu_show_pmu_cap(&pmu->pmu); @@ -6370,11 +6362,9 @@ static __always_inline int intel_pmu_init_hybrid(enum hybrid_pmu_type pmus) pmu->intel_cap.capabilities = x86_pmu.intel_cap.capabilities; if (pmu->pmu_type & hybrid_small_tiny) { pmu->intel_cap.perf_metrics = 0; - pmu->intel_cap.pebs_output_pt_available = 1; pmu->mid_ack = true; } else if (pmu->pmu_type & hybrid_big) { pmu->intel_cap.perf_metrics = 1; - pmu->intel_cap.pebs_output_pt_available = 0; pmu->late_ack = true; } } diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index ba74e1198328..c2e2eae7309c 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -2578,7 +2578,15 @@ void __init intel_ds_init(void) } pr_cont("PEBS fmt4%c%s, ", pebs_type, pebs_qual); - if (!is_hybrid() && x86_pmu.intel_cap.pebs_output_pt_available) { + /* + * The PEBS-via-PT is not supported on hybrid platforms, + * because not all CPUs of a hybrid machine support it. + * The global x86_pmu.intel_cap, which only contains the + * common capabilities, is used to check the availability + * of the feature. The per-PMU pebs_output_pt_available + * in a hybrid machine should be ignored. + */ + if (x86_pmu.intel_cap.pebs_output_pt_available) { pr_cont("PEBS-via-PT, "); x86_get_pmu(smp_processor_id())->capabilities |= PERF_PMU_CAP_AUX_OUTPUT; } From 47a973fd75639fe80d59f9e1860113bb2a0b112b Mon Sep 17 00:00:00 2001 From: Kan Liang Date: Wed, 29 Jan 2025 07:48:19 -0800 Subject: [PATCH 0670/1443] perf/x86/intel: Fix ARCH_PERFMON_NUM_COUNTER_LEAF The EAX of the CPUID Leaf 023H enumerates the mask of valid sub-leaves. To tell the availability of the sub-leaf 1 (enumerate the counter mask), perf should check the bit 1 (0x2) of EAS, rather than bit 0 (0x1). The error is not user-visible on bare metal. Because the sub-leaf 0 and the sub-leaf 1 are always available. However, it may bring issues in a virtualization environment when a VMM only enumerates the sub-leaf 0. Introduce the cpuid35_e?x to replace the macros, which makes the implementation style consistent. Fixes: eb467aaac21e ("perf/x86/intel: Support Architectural PerfMon Extension leaf") Signed-off-by: Kan Liang Signed-off-by: Peter Zijlstra (Intel) Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20250129154820.3755948-3-kan.liang@linux.intel.com --- arch/x86/events/intel/core.c | 18 ++++++++++-------- arch/x86/include/asm/perf_event.h | 28 +++++++++++++++++++++++++--- 2 files changed, 35 insertions(+), 11 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 966f7832497d..f3d5b718f93f 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -4905,20 +4905,22 @@ static inline bool intel_pmu_broken_perf_cap(void) static void update_pmu_cap(struct x86_hybrid_pmu *pmu) { - unsigned int sub_bitmaps, eax, ebx, ecx, edx; + unsigned int cntr, fixed_cntr, ecx, edx; + union cpuid35_eax eax; + union cpuid35_ebx ebx; - cpuid(ARCH_PERFMON_EXT_LEAF, &sub_bitmaps, &ebx, &ecx, &edx); + cpuid(ARCH_PERFMON_EXT_LEAF, &eax.full, &ebx.full, &ecx, &edx); - if (ebx & ARCH_PERFMON_EXT_UMASK2) + if (ebx.split.umask2) pmu->config_mask |= ARCH_PERFMON_EVENTSEL_UMASK2; - if (ebx & ARCH_PERFMON_EXT_EQ) + if (ebx.split.eq) pmu->config_mask |= ARCH_PERFMON_EVENTSEL_EQ; - if (sub_bitmaps & ARCH_PERFMON_NUM_COUNTER_LEAF_BIT) { + if (eax.split.cntr_subleaf) { cpuid_count(ARCH_PERFMON_EXT_LEAF, ARCH_PERFMON_NUM_COUNTER_LEAF, - &eax, &ebx, &ecx, &edx); - pmu->cntr_mask64 = eax; - pmu->fixed_cntr_mask64 = ebx; + &cntr, &fixed_cntr, &ecx, &edx); + pmu->cntr_mask64 = cntr; + pmu->fixed_cntr_mask64 = fixed_cntr; } if (!intel_pmu_broken_perf_cap()) { diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index 1ac79f361645..0ba8d20f2d1d 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -188,11 +188,33 @@ union cpuid10_edx { * detection/enumeration details: */ #define ARCH_PERFMON_EXT_LEAF 0x00000023 -#define ARCH_PERFMON_EXT_UMASK2 0x1 -#define ARCH_PERFMON_EXT_EQ 0x2 -#define ARCH_PERFMON_NUM_COUNTER_LEAF_BIT 0x1 #define ARCH_PERFMON_NUM_COUNTER_LEAF 0x1 +union cpuid35_eax { + struct { + unsigned int leaf0:1; + /* Counters Sub-Leaf */ + unsigned int cntr_subleaf:1; + /* Auto Counter Reload Sub-Leaf */ + unsigned int acr_subleaf:1; + /* Events Sub-Leaf */ + unsigned int events_subleaf:1; + unsigned int reserved:28; + } split; + unsigned int full; +}; + +union cpuid35_ebx { + struct { + /* UnitMask2 Supported */ + unsigned int umask2:1; + /* EQ-bit Supported */ + unsigned int eq:1; + unsigned int reserved:30; + } split; + unsigned int full; +}; + /* * Intel Architectural LBR CPUID detection/enumeration details: */ From c631a2de7ae48d50434bdc205d901423f8577c65 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Thu, 30 Jan 2025 17:07:21 -0800 Subject: [PATCH 0671/1443] perf/x86/intel: Ensure LBRs are disabled when a CPU is starting Explicitly clear DEBUGCTL.LBR when a CPU is starting, prior to purging the LBR MSRs themselves, as at least one system has been found to transfer control to the kernel with LBRs enabled (it's unclear whether it's a BIOS flaw or a CPU goof). Because the kernel preserves the original DEBUGCTL, even when toggling LBRs, leaving DEBUGCTL.LBR as is results in running with LBRs enabled at all times. Closes: https://lore.kernel.org/all/c9d8269bff69f6359731d758e3b1135dedd7cc61.camel@redhat.com Reported-by: Maxim Levitsky Signed-off-by: Sean Christopherson Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Maxim Levitsky Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20250131010721.470503-1-seanjc@google.com --- arch/x86/events/intel/core.c | 5 ++++- arch/x86/include/asm/msr-index.h | 3 ++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index f3d5b718f93f..e86333eee266 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -5042,8 +5042,11 @@ static void intel_pmu_cpu_starting(int cpu) init_debug_store_on_cpu(cpu); /* - * Deal with CPUs that don't clear their LBRs on power-up. + * Deal with CPUs that don't clear their LBRs on power-up, and that may + * even boot with LBRs enabled. */ + if (!static_cpu_has(X86_FEATURE_ARCH_LBR) && x86_pmu.lbr_nr) + msr_clear_bit(MSR_IA32_DEBUGCTLMSR, DEBUGCTLMSR_LBR_BIT); intel_pmu_lbr_reset(); cpuc->lbr_sel = NULL; diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 9a71880eec07..72765b2fe0d8 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -395,7 +395,8 @@ #define MSR_IA32_PASID_VALID BIT_ULL(31) /* DEBUGCTLMSR bits (others vary by model): */ -#define DEBUGCTLMSR_LBR (1UL << 0) /* last branch recording */ +#define DEBUGCTLMSR_LBR_BIT 0 /* last branch recording */ +#define DEBUGCTLMSR_LBR (1UL << DEBUGCTLMSR_LBR_BIT) #define DEBUGCTLMSR_BTF_SHIFT 1 #define DEBUGCTLMSR_BTF (1UL << 1) /* single-step on branches */ #define DEBUGCTLMSR_BUS_LOCK_DETECT (1UL << 2) From 9ab127a18018fb06bd42a54ed38bb7b8c449d686 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 27 Jan 2025 08:10:02 +0100 Subject: [PATCH 0672/1443] drm/hisilicon/hibmc: select CONFIG_DRM_DISPLAY_DP_HELPER Without the DP helper code, the newly added displayport support causes a link failure: x86_64-linux-ld: drivers/gpu/drm/hisilicon/hibmc/dp/dp_aux.o: in function `hibmc_dp_aux_init': dp_aux.c:(.text+0x37e): undefined reference to `drm_dp_aux_init' x86_64-linux-ld: drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.o: in function `hibmc_dp_link_set_pattern': dp_link.c:(.text+0xae): undefined reference to `drm_dp_dpcd_write' x86_64-linux-ld: drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.o: in function `hibmc_dp_link_get_adjust_train': dp_link.c:(.text+0x121): undefined reference to `drm_dp_get_adjust_request_voltage' x86_64-linux-ld: dp_link.c:(.text+0x12e): undefined reference to `drm_dp_get_adjust_request_pre_emphasis' x86_64-linux-ld: drivers/gpu/drm/hisilicon/hibmc/dp/dp_link.o: in function `hibmc_dp_link_training': dp_link.c:(.text+0x2b0): undefined reference to `drm_dp_dpcd_write' x86_64-linux-ld: dp_link.c:(.text+0x2e3): undefined reference to `drm_dp_dpcd_write' Add both DRM_DISPLAY_DP_HELPER and DRM_DISPLAY_HELPER, which is in turn required by the former. Fixes: 0ab6ea261c1f ("drm/hisilicon/hibmc: add dp module in hibmc") Signed-off-by: Arnd Bergmann Reviewed-by: Dmitry Baryshkov Link: https://patchwork.freedesktop.org/patch/msgid/20250127071059.617567-1-arnd@kernel.org Signed-off-by: Dmitry Baryshkov --- drivers/gpu/drm/hisilicon/hibmc/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/hisilicon/hibmc/Kconfig b/drivers/gpu/drm/hisilicon/hibmc/Kconfig index 93b8d32e3be1..98d77d74999d 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/Kconfig +++ b/drivers/gpu/drm/hisilicon/hibmc/Kconfig @@ -4,6 +4,8 @@ config DRM_HISI_HIBMC depends on DRM && PCI depends on MMU select DRM_CLIENT_SELECTION + select DRM_DISPLAY_HELPER + select DRM_DISPLAY_DP_HELPER select DRM_KMS_HELPER select DRM_VRAM_HELPER select DRM_TTM From 2fa0fbeb69edd367b7c44f484e8dc5a5a1a311ef Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 7 Feb 2025 10:58:23 -1000 Subject: [PATCH 0673/1443] sched_ext: Implement auto local dispatching of migration disabled tasks Migration disabled tasks are special and pinned to their previous CPUs. They tripped up some unsuspecting BPF schedulers as their ->nr_cpus_allowed may not agree with the bits set in ->cpus_ptr. Make it easier for BPF schedulers by automatically dispatching them to the pinned local DSQs by default. If a BPF scheduler wants to handle migration disabled tasks explicitly, it can set SCX_OPS_ENQ_MIGRATION_DISABLED. Signed-off-by: Tejun Heo Acked-by: Andrea Righi --- kernel/sched/ext.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index a6d6d6dadde5..efdbf4d85a21 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -122,6 +122,19 @@ enum scx_ops_flags { */ SCX_OPS_SWITCH_PARTIAL = 1LLU << 3, + /* + * A migration disabled task can only execute on its current CPU. By + * default, such tasks are automatically put on the CPU's local DSQ with + * the default slice on enqueue. If this ops flag is set, they also go + * through ops.enqueue(). + * + * A migration disabled task never invokes ops.select_cpu() as it can + * only select the current CPU. Also, p->cpus_ptr will only contain its + * current CPU while p->nr_cpus_allowed keeps tracking p->user_cpus_ptr + * and thus may disagree with cpumask_weight(p->cpus_ptr). + */ + SCX_OPS_ENQ_MIGRATION_DISABLED = 1LLU << 4, + /* * CPU cgroup support flags */ @@ -130,6 +143,7 @@ enum scx_ops_flags { SCX_OPS_ALL_FLAGS = SCX_OPS_KEEP_BUILTIN_IDLE | SCX_OPS_ENQ_LAST | SCX_OPS_ENQ_EXITING | + SCX_OPS_ENQ_MIGRATION_DISABLED | SCX_OPS_SWITCH_PARTIAL | SCX_OPS_HAS_CGROUP_WEIGHT, }; @@ -882,6 +896,7 @@ static bool scx_warned_zero_slice; static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_last); static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_exiting); +static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_migration_disabled); static DEFINE_STATIC_KEY_FALSE(scx_ops_cpu_preempt); static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_enabled); @@ -2014,6 +2029,11 @@ static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, unlikely(p->flags & PF_EXITING)) goto local; + /* see %SCX_OPS_ENQ_MIGRATION_DISABLED */ + if (!static_branch_unlikely(&scx_ops_enq_migration_disabled) && + is_migration_disabled(p)) + goto local; + if (!SCX_HAS_OP(enqueue)) goto global; @@ -5052,6 +5072,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) static_branch_disable(&scx_has_op[i]); static_branch_disable(&scx_ops_enq_last); static_branch_disable(&scx_ops_enq_exiting); + static_branch_disable(&scx_ops_enq_migration_disabled); static_branch_disable(&scx_ops_cpu_preempt); static_branch_disable(&scx_builtin_idle_enabled); synchronize_rcu(); @@ -5661,6 +5682,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) if (ops->flags & SCX_OPS_ENQ_EXITING) static_branch_enable(&scx_ops_enq_exiting); + if (ops->flags & SCX_OPS_ENQ_MIGRATION_DISABLED) + static_branch_enable(&scx_ops_enq_migration_disabled); if (scx_ops.cpu_acquire || scx_ops.cpu_release) static_branch_enable(&scx_ops_cpu_preempt); From 32966821574cd2917bd60f2554f435fe527f4702 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 7 Feb 2025 10:59:06 -1000 Subject: [PATCH 0674/1443] sched_ext: Fix migration disabled handling in targeted dispatches A dispatch operation that can target a specific local DSQ - scx_bpf_dsq_move_to_local() or scx_bpf_dsq_move() - checks whether the task can be migrated to the target CPU using task_can_run_on_remote_rq(). If the task can't be migrated to the targeted CPU, it is bounced through a global DSQ. task_can_run_on_remote_rq() assumes that the task is on a CPU that's different from the targeted CPU but the callers doesn't uphold the assumption and may call the function when the task is already on the target CPU. When such task has migration disabled, task_can_run_on_remote_rq() ends up returning %false incorrectly unnecessarily bouncing the task to a global DSQ. Fix it by updating the callers to only call task_can_run_on_remote_rq() when the task is on a different CPU than the target CPU. As this is a bit subtle, for clarity and documentation: - Make task_can_run_on_remote_rq() trigger SCHED_WARN_ON() if the task is on the same CPU as the target CPU. - is_migration_disabled() test in task_can_run_on_remote_rq() cannot trigger if the task is on a different CPU than the target CPU as the preceding task_allowed_on_cpu() test should fail beforehand. Convert the test into SCHED_WARN_ON(). Signed-off-by: Tejun Heo Fixes: 4c30f5ce4f7a ("sched_ext: Implement scx_bpf_dispatch[_vtime]_from_dsq()") Fixes: 0366017e0973 ("sched_ext: Use task_can_run_on_remote_rq() test in dispatch_to_local_dsq()") Cc: stable@vger.kernel.org # v6.12+ --- kernel/sched/ext.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index efdbf4d85a21..e01144340d67 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -2333,12 +2333,16 @@ static void move_remote_task_to_local_dsq(struct task_struct *p, u64 enq_flags, * * - The BPF scheduler is bypassed while the rq is offline and we can always say * no to the BPF scheduler initiated migrations while offline. + * + * The caller must ensure that @p and @rq are on different CPUs. */ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, bool trigger_error) { int cpu = cpu_of(rq); + SCHED_WARN_ON(task_cpu(p) == cpu); + /* * We don't require the BPF scheduler to avoid dispatching to offline * CPUs mostly for convenience but also because CPUs can go offline @@ -2352,8 +2356,11 @@ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, return false; } - if (unlikely(is_migration_disabled(p))) - return false; + /* + * If @p has migration disabled, @p->cpus_ptr only contains its current + * CPU and the above task_allowed_on_cpu() test should have failed. + */ + SCHED_WARN_ON(is_migration_disabled(p)); if (!scx_rq_online(rq)) return false; @@ -2457,7 +2464,8 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, if (dst_dsq->id == SCX_DSQ_LOCAL) { dst_rq = container_of(dst_dsq, struct rq, scx.local_dsq); - if (!task_can_run_on_remote_rq(p, dst_rq, true)) { + if (src_rq != dst_rq && + unlikely(!task_can_run_on_remote_rq(p, dst_rq, true))) { dst_dsq = find_global_dsq(p); dst_rq = src_rq; } @@ -2611,7 +2619,8 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, } #ifdef CONFIG_SMP - if (unlikely(!task_can_run_on_remote_rq(p, dst_rq, true))) { + if (src_rq != dst_rq && + unlikely(!task_can_run_on_remote_rq(p, dst_rq, true))) { dispatch_enqueue(find_global_dsq(p), p, enq_flags | SCX_ENQ_CLEAR_OPSS); return; From c53fbdb60fb61fd6bda2bc0dc89837966625c5dc Mon Sep 17 00:00:00 2001 From: Quentin Perret Date: Fri, 7 Feb 2025 14:54:37 +0000 Subject: [PATCH 0675/1443] KVM: arm64: Improve error handling from check_host_shared_guest() The check_host_shared_guest() path expects to find a last-level valid PTE in the guest's stage-2 page-table. However, it checks the PTE's level before its validity, which makes it hard for callers to figure out what went wrong. To make error handling simpler, check the PTE's validity first. Signed-off-by: Quentin Perret Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20250207145438.1333475-2-qperret@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/mem_protect.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 7ad7b133b81a..41847c04b270 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -943,10 +943,10 @@ static int __check_host_shared_guest(struct pkvm_hyp_vm *vm, u64 *__phys, u64 ip ret = kvm_pgtable_get_leaf(&vm->pgt, ipa, &pte, &level); if (ret) return ret; - if (level != KVM_PGTABLE_LAST_LEVEL) - return -E2BIG; if (!kvm_pte_valid(pte)) return -ENOENT; + if (level != KVM_PGTABLE_LAST_LEVEL) + return -E2BIG; state = guest_get_page_state(pte, ipa); if (state != PKVM_PAGE_SHARED_BORROWED) From eabc7aaef7a553b64bf6e631ce04526af6c8d104 Mon Sep 17 00:00:00 2001 From: Quentin Perret Date: Fri, 7 Feb 2025 14:54:38 +0000 Subject: [PATCH 0676/1443] KVM: arm64: Simplify np-guest hypercalls When the handling of a guest stage-2 permission fault races with an MMU notifier, the faulting page might be gone from the guest's stage-2 by the point we attempt to call (p)kvm_pgtable_stage2_relax_perms(). In the normal KVM case, this leads to returning -EAGAIN which user_mem_abort() handles correctly by simply re-entering the guest. However, the pKVM hypercall implementation has additional logic to check the page state using __check_host_shared_guest() which gets confused with absence of a page mapped at the requested IPA and returns -ENOENT, hence breaking user_mem_abort() and hilarity ensues. Luckily, several of the hypercalls for managing the stage-2 page-table of NP guests have no effect on the pKVM ownership tracking (wrprotect, test_clear_young, mkyoung, and crucially relax_perms), so the extra state checking logic is in fact not strictly necessary. So, to fix the discrepancy between standard KVM and pKVM, let's just drop the superfluous __check_host_shared_guest() logic from those hypercalls and make the extra state checking a debug assertion dependent on CONFIG_NVHE_EL2_DEBUG as we already do for other transitions. Signed-off-by: Quentin Perret Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20250207145438.1333475-3-qperret@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/mem_protect.c | 69 +++++++++++++++------------ 1 file changed, 38 insertions(+), 31 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 41847c04b270..4c2f6a6a2efe 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -998,44 +998,57 @@ int __pkvm_host_unshare_guest(u64 gfn, struct pkvm_hyp_vm *vm) return ret; } -int __pkvm_host_relax_perms_guest(u64 gfn, struct pkvm_hyp_vcpu *vcpu, enum kvm_pgtable_prot prot) +static void assert_host_shared_guest(struct pkvm_hyp_vm *vm, u64 ipa) { - struct pkvm_hyp_vm *vm = pkvm_hyp_vcpu_to_hyp_vm(vcpu); - u64 ipa = hyp_pfn_to_phys(gfn); u64 phys; int ret; - if (prot & ~KVM_PGTABLE_PROT_RWX) - return -EINVAL; + if (!IS_ENABLED(CONFIG_NVHE_EL2_DEBUG)) + return; host_lock_component(); guest_lock_component(vm); ret = __check_host_shared_guest(vm, &phys, ipa); - if (!ret) - ret = kvm_pgtable_stage2_relax_perms(&vm->pgt, ipa, prot, 0); guest_unlock_component(vm); host_unlock_component(); + WARN_ON(ret && ret != -ENOENT); +} + +int __pkvm_host_relax_perms_guest(u64 gfn, struct pkvm_hyp_vcpu *vcpu, enum kvm_pgtable_prot prot) +{ + struct pkvm_hyp_vm *vm = pkvm_hyp_vcpu_to_hyp_vm(vcpu); + u64 ipa = hyp_pfn_to_phys(gfn); + int ret; + + if (pkvm_hyp_vm_is_protected(vm)) + return -EPERM; + + if (prot & ~KVM_PGTABLE_PROT_RWX) + return -EINVAL; + + assert_host_shared_guest(vm, ipa); + guest_lock_component(vm); + ret = kvm_pgtable_stage2_relax_perms(&vm->pgt, ipa, prot, 0); + guest_unlock_component(vm); + return ret; } int __pkvm_host_wrprotect_guest(u64 gfn, struct pkvm_hyp_vm *vm) { u64 ipa = hyp_pfn_to_phys(gfn); - u64 phys; int ret; - host_lock_component(); + if (pkvm_hyp_vm_is_protected(vm)) + return -EPERM; + + assert_host_shared_guest(vm, ipa); guest_lock_component(vm); - - ret = __check_host_shared_guest(vm, &phys, ipa); - if (!ret) - ret = kvm_pgtable_stage2_wrprotect(&vm->pgt, ipa, PAGE_SIZE); - + ret = kvm_pgtable_stage2_wrprotect(&vm->pgt, ipa, PAGE_SIZE); guest_unlock_component(vm); - host_unlock_component(); return ret; } @@ -1043,18 +1056,15 @@ int __pkvm_host_wrprotect_guest(u64 gfn, struct pkvm_hyp_vm *vm) int __pkvm_host_test_clear_young_guest(u64 gfn, bool mkold, struct pkvm_hyp_vm *vm) { u64 ipa = hyp_pfn_to_phys(gfn); - u64 phys; int ret; - host_lock_component(); + if (pkvm_hyp_vm_is_protected(vm)) + return -EPERM; + + assert_host_shared_guest(vm, ipa); guest_lock_component(vm); - - ret = __check_host_shared_guest(vm, &phys, ipa); - if (!ret) - ret = kvm_pgtable_stage2_test_clear_young(&vm->pgt, ipa, PAGE_SIZE, mkold); - + ret = kvm_pgtable_stage2_test_clear_young(&vm->pgt, ipa, PAGE_SIZE, mkold); guest_unlock_component(vm); - host_unlock_component(); return ret; } @@ -1063,18 +1073,15 @@ int __pkvm_host_mkyoung_guest(u64 gfn, struct pkvm_hyp_vcpu *vcpu) { struct pkvm_hyp_vm *vm = pkvm_hyp_vcpu_to_hyp_vm(vcpu); u64 ipa = hyp_pfn_to_phys(gfn); - u64 phys; int ret; - host_lock_component(); + if (pkvm_hyp_vm_is_protected(vm)) + return -EPERM; + + assert_host_shared_guest(vm, ipa); guest_lock_component(vm); - - ret = __check_host_shared_guest(vm, &phys, ipa); - if (!ret) - kvm_pgtable_stage2_mkyoung(&vm->pgt, ipa, 0); - + kvm_pgtable_stage2_mkyoung(&vm->pgt, ipa, 0); guest_unlock_component(vm); - host_unlock_component(); return ret; } From 7585946243d614bd2cd4e13377be2c711c9539e0 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 8 Feb 2025 18:54:28 +0100 Subject: [PATCH 0677/1443] PM: sleep: core: Restrict power.set_active propagation Commit 3775fc538f53 ("PM: sleep: core: Synchronize runtime PM status of parents and children") exposed an issue related to simple_pm_bus_pm_ops that uses pm_runtime_force_suspend() and pm_runtime_force_resume() as bus type PM callbacks for the noirq phases of system-wide suspend and resume. The problem is that pm_runtime_force_suspend() does not distinguish runtime-suspended devices from devices for which runtime PM has never been enabled, so if it sees a device with runtime PM status set to RPM_ACTIVE, it will assume that runtime PM is enabled for that device and so it will attempt to suspend it with the help of its runtime PM callbacks which may not be ready for that. As it turns out, this causes simple_pm_bus_runtime_suspend() to crash due to a NULL pointer dereference. Another problem related to the above commit and simple_pm_bus_pm_ops is that setting runtime PM status of a device handled by the latter to RPM_ACTIVE will actually prevent it from being resumed because pm_runtime_force_resume() only resumes devices with runtime PM status set to RPM_SUSPENDED. To mitigate these issues, do not allow power.set_active to propagate beyond the parent of the device with DPM_FLAG_SMART_SUSPEND set that will need to be resumed, which should be a sufficient stop-gap for the time being, but they will need to be properly addressed in the future because in general during system-wide resume it is necessary to resume all devices in a dependency chain in which at least one device is going to be resumed. Fixes: 3775fc538f53 ("PM: sleep: core: Synchronize runtime PM status of parents and children") Closes: https://lore.kernel.org/linux-pm/1c2433d4-7e0f-4395-b841-b8eac7c25651@nvidia.com/ Reported-by: Jon Hunter Tested-by: Johan Hovold Signed-off-by: Rafael J. Wysocki Link: https://patch.msgid.link/6137505.lOV4Wx5bFT@rjwysocki.net --- drivers/base/power/main.c | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index d497d448e4b2..40e1d8d8a589 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -1191,24 +1191,18 @@ static pm_message_t resume_event(pm_message_t sleep_state) return PMSG_ON; } -static void dpm_superior_set_must_resume(struct device *dev, bool set_active) +static void dpm_superior_set_must_resume(struct device *dev) { struct device_link *link; int idx; - if (dev->parent) { + if (dev->parent) dev->parent->power.must_resume = true; - if (set_active) - dev->parent->power.set_active = true; - } idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) { + list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) link->supplier->power.must_resume = true; - if (set_active) - link->supplier->power.set_active = true; - } device_links_read_unlock(idx); } @@ -1287,9 +1281,12 @@ static int device_suspend_noirq(struct device *dev, pm_message_t state, bool asy dev->power.must_resume = true; if (dev->power.must_resume) { - dev->power.set_active = dev->power.set_active || - dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND); - dpm_superior_set_must_resume(dev, dev->power.set_active); + if (dev_pm_test_driver_flags(dev, DPM_FLAG_SMART_SUSPEND)) { + dev->power.set_active = true; + if (dev->parent && !dev->parent->power.ignore_children) + dev->parent->power.set_active = true; + } + dpm_superior_set_must_resume(dev); } Complete: From 6f3d9d0dd335290fa364048de5b3a440f1cf259f Mon Sep 17 00:00:00 2001 From: Ryosuke Yasuoka Date: Thu, 6 Feb 2025 19:42:59 +0900 Subject: [PATCH 0678/1443] drm/virtio: Add drm_panic support Virtio gpu supports the drm_panic module, which displays a message to the screen when a kernel panic occurs. It is supported where it has vmapped shmem BO. Signed-off-by: Jocelyn Falempe Signed-off-by: Ryosuke Yasuoka Tested-by: Dmitry Osipenko Reviewed-by: Dmitry Osipenko Signed-off-by: Dmitry Osipenko Link: https://patchwork.freedesktop.org/patch/msgid/20250206104300.416014-1-ryasuoka@redhat.com --- drivers/gpu/drm/virtio/virtgpu_drv.h | 11 ++ drivers/gpu/drm/virtio/virtgpu_gem.c | 14 +++ drivers/gpu/drm/virtio/virtgpu_plane.c | 96 ++++++++++++++++ drivers/gpu/drm/virtio/virtgpu_vq.c | 151 +++++++++++++++++++++++++ 4 files changed, 272 insertions(+) diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h index f42ca9d8ed10..f17660a71a3e 100644 --- a/drivers/gpu/drm/virtio/virtgpu_drv.h +++ b/drivers/gpu/drm/virtio/virtgpu_drv.h @@ -310,6 +310,7 @@ int virtio_gpu_mode_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); +struct virtio_gpu_object_array *virtio_gpu_panic_array_alloc(void); struct virtio_gpu_object_array *virtio_gpu_array_alloc(u32 nents); struct virtio_gpu_object_array* virtio_gpu_array_from_handles(struct drm_file *drm_file, u32 *handles, u32 nents); @@ -334,12 +335,21 @@ void virtio_gpu_cmd_create_resource(struct virtio_gpu_device *vgdev, struct virtio_gpu_fence *fence); void virtio_gpu_cmd_unref_resource(struct virtio_gpu_device *vgdev, struct virtio_gpu_object *bo); +int virtio_gpu_panic_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, + uint64_t offset, + uint32_t width, uint32_t height, + uint32_t x, uint32_t y, + struct virtio_gpu_object_array *objs); void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, uint64_t offset, uint32_t width, uint32_t height, uint32_t x, uint32_t y, struct virtio_gpu_object_array *objs, struct virtio_gpu_fence *fence); +void virtio_gpu_panic_cmd_resource_flush(struct virtio_gpu_device *vgdev, + uint32_t resource_id, + uint32_t x, uint32_t y, + uint32_t width, uint32_t height); void virtio_gpu_cmd_resource_flush(struct virtio_gpu_device *vgdev, uint32_t resource_id, uint32_t x, uint32_t y, @@ -408,6 +418,7 @@ void virtio_gpu_ctrl_ack(struct virtqueue *vq); void virtio_gpu_cursor_ack(struct virtqueue *vq); void virtio_gpu_dequeue_ctrl_func(struct work_struct *work); void virtio_gpu_dequeue_cursor_func(struct work_struct *work); +void virtio_gpu_panic_notify(struct virtio_gpu_device *vgdev); void virtio_gpu_notify(struct virtio_gpu_device *vgdev); int diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c index 5aab588fc400..dde8fc1a3689 100644 --- a/drivers/gpu/drm/virtio/virtgpu_gem.c +++ b/drivers/gpu/drm/virtio/virtgpu_gem.c @@ -148,6 +148,20 @@ void virtio_gpu_gem_object_close(struct drm_gem_object *obj, virtio_gpu_notify(vgdev); } +/* For drm panic */ +struct virtio_gpu_object_array *virtio_gpu_panic_array_alloc(void) +{ + struct virtio_gpu_object_array *objs; + + objs = kmalloc(sizeof(struct virtio_gpu_object_array), GFP_ATOMIC); + if (!objs) + return NULL; + + objs->nents = 0; + objs->total = 1; + return objs; +} + struct virtio_gpu_object_array *virtio_gpu_array_alloc(u32 nents) { struct virtio_gpu_object_array *objs; diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c index 42aa554eca9f..a6f5a78f436a 100644 --- a/drivers/gpu/drm/virtio/virtgpu_plane.c +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c @@ -28,6 +28,8 @@ #include #include #include +#include +#include #include "virtgpu_drv.h" @@ -127,6 +129,30 @@ static int virtio_gpu_plane_atomic_check(struct drm_plane *plane, return ret; } +/* For drm panic */ +static int virtio_gpu_panic_update_dumb_bo(struct virtio_gpu_device *vgdev, + struct drm_plane_state *state, + struct drm_rect *rect) +{ + struct virtio_gpu_object *bo = + gem_to_virtio_gpu_obj(state->fb->obj[0]); + struct virtio_gpu_object_array *objs; + uint32_t w = rect->x2 - rect->x1; + uint32_t h = rect->y2 - rect->y1; + uint32_t x = rect->x1; + uint32_t y = rect->y1; + uint32_t off = x * state->fb->format->cpp[0] + + y * state->fb->pitches[0]; + + objs = virtio_gpu_panic_array_alloc(); + if (!objs) + return -ENOMEM; + virtio_gpu_array_add_obj(objs, &bo->base.base); + + return virtio_gpu_panic_cmd_transfer_to_host_2d(vgdev, off, w, h, x, y, + objs); +} + static void virtio_gpu_update_dumb_bo(struct virtio_gpu_device *vgdev, struct drm_plane_state *state, struct drm_rect *rect) @@ -150,6 +176,24 @@ static void virtio_gpu_update_dumb_bo(struct virtio_gpu_device *vgdev, objs, NULL); } +/* For drm_panic */ +static void virtio_gpu_panic_resource_flush(struct drm_plane *plane, + uint32_t x, uint32_t y, + uint32_t width, uint32_t height) +{ + struct drm_device *dev = plane->dev; + struct virtio_gpu_device *vgdev = dev->dev_private; + struct virtio_gpu_framebuffer *vgfb; + struct virtio_gpu_object *bo; + + vgfb = to_virtio_gpu_framebuffer(plane->state->fb); + bo = gem_to_virtio_gpu_obj(vgfb->base.obj[0]); + + virtio_gpu_panic_cmd_resource_flush(vgdev, bo->hw_res_handle, x, y, + width, height); + virtio_gpu_panic_notify(vgdev); +} + static void virtio_gpu_resource_flush(struct drm_plane *plane, uint32_t x, uint32_t y, uint32_t width, uint32_t height) @@ -446,11 +490,63 @@ static void virtio_gpu_cursor_plane_update(struct drm_plane *plane, virtio_gpu_cursor_ping(vgdev, output); } +static int virtio_drm_get_scanout_buffer(struct drm_plane *plane, + struct drm_scanout_buffer *sb) +{ + struct virtio_gpu_object *bo; + + if (!plane->state || !plane->state->fb || !plane->state->visible) + return -ENODEV; + + bo = gem_to_virtio_gpu_obj(plane->state->fb->obj[0]); + + /* Only support mapped shmem bo */ + if (virtio_gpu_is_vram(bo) || bo->base.base.import_attach || !bo->base.vaddr) + return -ENODEV; + + iosys_map_set_vaddr(&sb->map[0], bo->base.vaddr); + + sb->format = plane->state->fb->format; + sb->height = plane->state->fb->height; + sb->width = plane->state->fb->width; + sb->pitch[0] = plane->state->fb->pitches[0]; + return 0; +} + +static void virtio_panic_flush(struct drm_plane *plane) +{ + struct virtio_gpu_object *bo; + struct drm_device *dev = plane->dev; + struct virtio_gpu_device *vgdev = dev->dev_private; + struct drm_rect rect; + + rect.x1 = 0; + rect.y1 = 0; + rect.x2 = plane->state->fb->width; + rect.y2 = plane->state->fb->height; + + bo = gem_to_virtio_gpu_obj(plane->state->fb->obj[0]); + + if (bo->dumb) { + if (virtio_gpu_panic_update_dumb_bo(vgdev, plane->state, + &rect)) + return; + } + + virtio_gpu_panic_resource_flush(plane, + plane->state->src_x >> 16, + plane->state->src_y >> 16, + plane->state->src_w >> 16, + plane->state->src_h >> 16); +} + static const struct drm_plane_helper_funcs virtio_gpu_primary_helper_funcs = { .prepare_fb = virtio_gpu_plane_prepare_fb, .cleanup_fb = virtio_gpu_plane_cleanup_fb, .atomic_check = virtio_gpu_plane_atomic_check, .atomic_update = virtio_gpu_primary_plane_update, + .get_scanout_buffer = virtio_drm_get_scanout_buffer, + .panic_flush = virtio_panic_flush, }; static const struct drm_plane_helper_funcs virtio_gpu_cursor_helper_funcs = { diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c index 062639250a4e..55a15e247dd1 100644 --- a/drivers/gpu/drm/virtio/virtgpu_vq.c +++ b/drivers/gpu/drm/virtio/virtgpu_vq.c @@ -86,6 +86,22 @@ void virtio_gpu_free_vbufs(struct virtio_gpu_device *vgdev) vgdev->vbufs = NULL; } +/* For drm_panic */ +static struct virtio_gpu_vbuffer* +virtio_gpu_panic_get_vbuf(struct virtio_gpu_device *vgdev, int size) +{ + struct virtio_gpu_vbuffer *vbuf; + + vbuf = kmem_cache_zalloc(vgdev->vbufs, GFP_ATOMIC); + + vbuf->buf = (void *)vbuf + sizeof(*vbuf); + vbuf->size = size; + vbuf->resp_cb = NULL; + vbuf->resp_size = sizeof(struct virtio_gpu_ctrl_hdr); + vbuf->resp_buf = (void *)vbuf->buf + size; + return vbuf; +} + static struct virtio_gpu_vbuffer* virtio_gpu_get_vbuf(struct virtio_gpu_device *vgdev, int size, int resp_size, void *resp_buf, @@ -137,6 +153,18 @@ virtio_gpu_alloc_cursor(struct virtio_gpu_device *vgdev, return (struct virtio_gpu_update_cursor *)vbuf->buf; } +/* For drm_panic */ +static void *virtio_gpu_panic_alloc_cmd_resp(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer **vbuffer_p, + int cmd_size) +{ + struct virtio_gpu_vbuffer *vbuf; + + vbuf = virtio_gpu_panic_get_vbuf(vgdev, cmd_size); + *vbuffer_p = vbuf; + return (struct virtio_gpu_command *)vbuf->buf; +} + static void *virtio_gpu_alloc_cmd_resp(struct virtio_gpu_device *vgdev, virtio_gpu_resp_cb cb, struct virtio_gpu_vbuffer **vbuffer_p, @@ -311,6 +339,34 @@ static struct sg_table *vmalloc_to_sgt(char *data, uint32_t size, int *sg_ents) return sgt; } +/* For drm_panic */ +static int virtio_gpu_panic_queue_ctrl_sgs(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer *vbuf, + int elemcnt, + struct scatterlist **sgs, + int outcnt, + int incnt) +{ + struct virtqueue *vq = vgdev->ctrlq.vq; + int ret; + + if (vgdev->has_indirect) + elemcnt = 1; + + if (vq->num_free < elemcnt) + return -ENOMEM; + + ret = virtqueue_add_sgs(vq, sgs, outcnt, incnt, vbuf, GFP_ATOMIC); + WARN_ON(ret); + + vbuf->seqno = ++vgdev->ctrlq.seqno; + trace_virtio_gpu_cmd_queue(vq, virtio_gpu_vbuf_ctrl_hdr(vbuf), vbuf->seqno); + + atomic_inc(&vgdev->pending_commands); + + return 0; +} + static int virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev, struct virtio_gpu_vbuffer *vbuf, struct virtio_gpu_fence *fence, @@ -368,6 +424,32 @@ static int virtio_gpu_queue_ctrl_sgs(struct virtio_gpu_device *vgdev, return 0; } +/* For drm_panic */ +static int virtio_gpu_panic_queue_ctrl_buffer(struct virtio_gpu_device *vgdev, + struct virtio_gpu_vbuffer *vbuf) +{ + struct scatterlist *sgs[3], vcmd, vresp; + int elemcnt = 0, outcnt = 0, incnt = 0; + + /* set up vcmd */ + sg_init_one(&vcmd, vbuf->buf, vbuf->size); + elemcnt++; + sgs[outcnt] = &vcmd; + outcnt++; + + /* set up vresp */ + if (vbuf->resp_size) { + sg_init_one(&vresp, vbuf->resp_buf, vbuf->resp_size); + elemcnt++; + sgs[outcnt + incnt] = &vresp; + incnt++; + } + + return virtio_gpu_panic_queue_ctrl_sgs(vgdev, vbuf, + elemcnt, sgs, + outcnt, incnt); +} + static int virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev, struct virtio_gpu_vbuffer *vbuf, struct virtio_gpu_fence *fence) @@ -422,6 +504,21 @@ static int virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev, return ret; } +/* For drm_panic */ +void virtio_gpu_panic_notify(struct virtio_gpu_device *vgdev) +{ + bool notify; + + if (!atomic_read(&vgdev->pending_commands)) + return; + + atomic_set(&vgdev->pending_commands, 0); + notify = virtqueue_kick_prepare(vgdev->ctrlq.vq); + + if (notify) + virtqueue_notify(vgdev->ctrlq.vq); +} + void virtio_gpu_notify(struct virtio_gpu_device *vgdev) { bool notify; @@ -567,6 +664,29 @@ void virtio_gpu_cmd_set_scanout(struct virtio_gpu_device *vgdev, virtio_gpu_queue_ctrl_buffer(vgdev, vbuf); } +/* For drm_panic */ +void virtio_gpu_panic_cmd_resource_flush(struct virtio_gpu_device *vgdev, + uint32_t resource_id, + uint32_t x, uint32_t y, + uint32_t width, uint32_t height) +{ + struct virtio_gpu_resource_flush *cmd_p; + struct virtio_gpu_vbuffer *vbuf; + + cmd_p = virtio_gpu_panic_alloc_cmd_resp(vgdev, &vbuf, sizeof(*cmd_p)); + memset(cmd_p, 0, sizeof(*cmd_p)); + vbuf->objs = NULL; + + cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_FLUSH); + cmd_p->resource_id = cpu_to_le32(resource_id); + cmd_p->r.width = cpu_to_le32(width); + cmd_p->r.height = cpu_to_le32(height); + cmd_p->r.x = cpu_to_le32(x); + cmd_p->r.y = cpu_to_le32(y); + + virtio_gpu_panic_queue_ctrl_buffer(vgdev, vbuf); +} + void virtio_gpu_cmd_resource_flush(struct virtio_gpu_device *vgdev, uint32_t resource_id, uint32_t x, uint32_t y, @@ -591,6 +711,37 @@ void virtio_gpu_cmd_resource_flush(struct virtio_gpu_device *vgdev, virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, fence); } +/* For drm_panic */ +int virtio_gpu_panic_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, + uint64_t offset, + uint32_t width, uint32_t height, + uint32_t x, uint32_t y, + struct virtio_gpu_object_array *objs) +{ + struct virtio_gpu_object *bo = gem_to_virtio_gpu_obj(objs->objs[0]); + struct virtio_gpu_transfer_to_host_2d *cmd_p; + struct virtio_gpu_vbuffer *vbuf; + bool use_dma_api = !virtio_has_dma_quirk(vgdev->vdev); + + if (virtio_gpu_is_shmem(bo) && use_dma_api) + dma_sync_sgtable_for_device(vgdev->vdev->dev.parent, + bo->base.sgt, DMA_TO_DEVICE); + + cmd_p = virtio_gpu_panic_alloc_cmd_resp(vgdev, &vbuf, sizeof(*cmd_p)); + memset(cmd_p, 0, sizeof(*cmd_p)); + vbuf->objs = objs; + + cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_TRANSFER_TO_HOST_2D); + cmd_p->resource_id = cpu_to_le32(bo->hw_res_handle); + cmd_p->offset = cpu_to_le64(offset); + cmd_p->r.width = cpu_to_le32(width); + cmd_p->r.height = cpu_to_le32(height); + cmd_p->r.x = cpu_to_le32(x); + cmd_p->r.y = cpu_to_le32(y); + + return virtio_gpu_panic_queue_ctrl_buffer(vgdev, vbuf); +} + void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev, uint64_t offset, uint32_t width, uint32_t height, From a64dcfb451e254085a7daee5fe51bf22959d52d3 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 9 Feb 2025 12:45:03 -0800 Subject: [PATCH 0679/1443] Linux 6.14-rc2 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 9e0d63d9d94b..89628e354ca7 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 14 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Baby Opossum Posse # *DOCUMENTATION* From dc9c5166c3cb044f8a001e397195242fd6796eee Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Mon, 3 Feb 2025 11:14:57 +0100 Subject: [PATCH 0680/1443] powerpc/code-patching: Disable KASAN report during patching via temporary mm Erhard reports the following KASAN hit on Talos II (power9) with kernel 6.13: [ 12.028126] ================================================================== [ 12.028198] BUG: KASAN: user-memory-access in copy_to_kernel_nofault+0x8c/0x1a0 [ 12.028260] Write of size 8 at addr 0000187e458f2000 by task systemd/1 [ 12.028346] CPU: 87 UID: 0 PID: 1 Comm: systemd Tainted: G T 6.13.0-P9-dirty #3 [ 12.028408] Tainted: [T]=RANDSTRUCT [ 12.028446] Hardware name: T2P9D01 REV 1.01 POWER9 0x4e1202 opal:skiboot-bc106a0 PowerNV [ 12.028500] Call Trace: [ 12.028536] [c000000008dbf3b0] [c000000001656a48] dump_stack_lvl+0xbc/0x110 (unreliable) [ 12.028609] [c000000008dbf3f0] [c0000000006e2fc8] print_report+0x6b0/0x708 [ 12.028666] [c000000008dbf4e0] [c0000000006e2454] kasan_report+0x164/0x300 [ 12.028725] [c000000008dbf600] [c0000000006e54d4] kasan_check_range+0x314/0x370 [ 12.028784] [c000000008dbf640] [c0000000006e6310] __kasan_check_write+0x20/0x40 [ 12.028842] [c000000008dbf660] [c000000000578e8c] copy_to_kernel_nofault+0x8c/0x1a0 [ 12.028902] [c000000008dbf6a0] [c0000000000acfe4] __patch_instructions+0x194/0x210 [ 12.028965] [c000000008dbf6e0] [c0000000000ade80] patch_instructions+0x150/0x590 [ 12.029026] [c000000008dbf7c0] [c0000000001159bc] bpf_arch_text_copy+0x6c/0xe0 [ 12.029085] [c000000008dbf800] [c000000000424250] bpf_jit_binary_pack_finalize+0x40/0xc0 [ 12.029147] [c000000008dbf830] [c000000000115dec] bpf_int_jit_compile+0x3bc/0x930 [ 12.029206] [c000000008dbf990] [c000000000423720] bpf_prog_select_runtime+0x1f0/0x280 [ 12.029266] [c000000008dbfa00] [c000000000434b18] bpf_prog_load+0xbb8/0x1370 [ 12.029324] [c000000008dbfb70] [c000000000436ebc] __sys_bpf+0x5ac/0x2e00 [ 12.029379] [c000000008dbfd00] [c00000000043a228] sys_bpf+0x28/0x40 [ 12.029435] [c000000008dbfd20] [c000000000038eb4] system_call_exception+0x334/0x610 [ 12.029497] [c000000008dbfe50] [c00000000000c270] system_call_vectored_common+0xf0/0x280 [ 12.029561] --- interrupt: 3000 at 0x3fff82f5cfa8 [ 12.029608] NIP: 00003fff82f5cfa8 LR: 00003fff82f5cfa8 CTR: 0000000000000000 [ 12.029660] REGS: c000000008dbfe80 TRAP: 3000 Tainted: G T (6.13.0-P9-dirty) [ 12.029735] MSR: 900000000280f032 CR: 42004848 XER: 00000000 [ 12.029855] IRQMASK: 0 GPR00: 0000000000000169 00003fffdcf789a0 00003fff83067100 0000000000000005 GPR04: 00003fffdcf78a98 0000000000000090 0000000000000000 0000000000000008 GPR08: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 GPR12: 0000000000000000 00003fff836ff7e0 c000000000010678 0000000000000000 GPR16: 0000000000000000 0000000000000000 00003fffdcf78f28 00003fffdcf78f90 GPR20: 0000000000000000 0000000000000000 0000000000000000 00003fffdcf78f80 GPR24: 00003fffdcf78f70 00003fffdcf78d10 00003fff835c7239 00003fffdcf78bd8 GPR28: 00003fffdcf78a98 0000000000000000 0000000000000000 000000011f547580 [ 12.030316] NIP [00003fff82f5cfa8] 0x3fff82f5cfa8 [ 12.030361] LR [00003fff82f5cfa8] 0x3fff82f5cfa8 [ 12.030405] --- interrupt: 3000 [ 12.030444] ================================================================== Commit c28c15b6d28a ("powerpc/code-patching: Use temporary mm for Radix MMU") is inspired from x86 but unlike x86 is doesn't disable KASAN reports during patching. This wasn't a problem at the begining because __patch_mem() is not instrumented. Commit 465cabc97b42 ("powerpc/code-patching: introduce patch_instructions()") use copy_to_kernel_nofault() to copy several instructions at once. But when using temporary mm the destination is not regular kernel memory but a kind of kernel-like memory located in user address space. Because it is not in kernel address space it is not covered by KASAN shadow memory. Since commit e4137f08816b ("mm, kasan, kmsan: instrument copy_from/to_kernel_nofault") KASAN reports bad accesses from copy_to_kernel_nofault(). Here a bad access to user memory is reported because KASAN detects the lack of shadow memory and the address is below TASK_SIZE. Do like x86 in commit b3fd8e83ada0 ("x86/alternatives: Use temporary mm for text poking") and disable KASAN reports during patching when using temporary mm. Reported-by: Erhard Furtner Close: https://lore.kernel.org/all/20250201151435.48400261@yea/ Fixes: 465cabc97b42 ("powerpc/code-patching: introduce patch_instructions()") Signed-off-by: Christophe Leroy Acked-by: Michael Ellerman Signed-off-by: Madhavan Srinivasan Link: https://patch.msgid.link/1c05b2a1b02ad75b981cfc45927e0b4a90441046.1738577687.git.christophe.leroy@csgroup.eu --- arch/powerpc/lib/code-patching.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/powerpc/lib/code-patching.c b/arch/powerpc/lib/code-patching.c index af97fbb3c257..81c0f673eb25 100644 --- a/arch/powerpc/lib/code-patching.c +++ b/arch/powerpc/lib/code-patching.c @@ -493,7 +493,9 @@ static int __do_patch_instructions_mm(u32 *addr, u32 *code, size_t len, bool rep orig_mm = start_using_temp_mm(patching_mm); + kasan_disable_current(); err = __patch_instructions(patch_addr, code, len, repeat_instr); + kasan_enable_current(); /* context synchronisation performed by __patch_instructions */ stop_using_temp_mm(patching_mm, orig_mm); From 61bcc752d1b81fde3cae454ff20c1d3c359df500 Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Sun, 12 Jan 2025 19:24:46 +0100 Subject: [PATCH 0681/1443] powerpc/64s: Rewrite __real_pte() and __rpte_to_hidx() as static inline Rewrite __real_pte() and __rpte_to_hidx() as static inline in order to avoid following warnings/errors when building with 4k page size: CC arch/powerpc/mm/book3s64/hash_tlb.o arch/powerpc/mm/book3s64/hash_tlb.c: In function 'hpte_need_flush': arch/powerpc/mm/book3s64/hash_tlb.c:49:16: error: variable 'offset' set but not used [-Werror=unused-but-set-variable] 49 | int i, offset; | ^~~~~~ CC arch/powerpc/mm/book3s64/hash_native.o arch/powerpc/mm/book3s64/hash_native.c: In function 'native_flush_hash_range': arch/powerpc/mm/book3s64/hash_native.c:782:29: error: variable 'index' set but not used [-Werror=unused-but-set-variable] 782 | unsigned long hash, index, hidx, shift, slot; | ^~~~~ Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202501081741.AYFwybsq-lkp@intel.com/ Fixes: ff31e105464d ("powerpc/mm/hash64: Store the slot information at the right offset for hugetlb") Signed-off-by: Christophe Leroy Reviewed-by: Ritesh Harjani (IBM) Signed-off-by: Madhavan Srinivasan Link: https://patch.msgid.link/e0d340a5b7bd478ecbf245d826e6ab2778b74e06.1736706263.git.christophe.leroy@csgroup.eu --- arch/powerpc/include/asm/book3s/64/hash-4k.h | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/include/asm/book3s/64/hash-4k.h b/arch/powerpc/include/asm/book3s/64/hash-4k.h index c3efacab4b94..aa90a048f319 100644 --- a/arch/powerpc/include/asm/book3s/64/hash-4k.h +++ b/arch/powerpc/include/asm/book3s/64/hash-4k.h @@ -77,9 +77,17 @@ /* * With 4K page size the real_pte machinery is all nops. */ -#define __real_pte(e, p, o) ((real_pte_t){(e)}) +static inline real_pte_t __real_pte(pte_t pte, pte_t *ptep, int offset) +{ + return (real_pte_t){pte}; +} + #define __rpte_to_pte(r) ((r).pte) -#define __rpte_to_hidx(r,index) (pte_val(__rpte_to_pte(r)) >> H_PAGE_F_GIX_SHIFT) + +static inline unsigned long __rpte_to_hidx(real_pte_t rpte, unsigned long index) +{ + return pte_val(__rpte_to_pte(rpte)) >> H_PAGE_F_GIX_SHIFT; +} #define pte_iterate_hashed_subpages(rpte, psize, va, index, shift) \ do { \ From a1f7b7ff0e10ae574d388131596390157222f986 Mon Sep 17 00:00:00 2001 From: Pierre-Louis Bossart Date: Mon, 10 Feb 2025 10:17:27 +0200 Subject: [PATCH 0682/1443] PCI: pci_ids: add INTEL_HDA_PTL_H Add Intel PTL-H audio Device ID. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Bard Liao Acked-by: Bjorn Helgaas Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20250210081730.22916-2-peter.ujfalusi@linux.intel.com --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index de5deb1a0118..1a2594a38199 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -3134,6 +3134,7 @@ #define PCI_DEVICE_ID_INTEL_HDA_LNL_P 0xa828 #define PCI_DEVICE_ID_INTEL_S21152BB 0xb152 #define PCI_DEVICE_ID_INTEL_HDA_BMG 0xe2f7 +#define PCI_DEVICE_ID_INTEL_HDA_PTL_H 0xe328 #define PCI_DEVICE_ID_INTEL_HDA_PTL 0xe428 #define PCI_DEVICE_ID_INTEL_HDA_CML_R 0xf0c8 #define PCI_DEVICE_ID_INTEL_HDA_RKL_S 0xf1c8 From 214e6be2d91d5d58f28d3a37630480077a1aafbd Mon Sep 17 00:00:00 2001 From: Pierre-Louis Bossart Date: Mon, 10 Feb 2025 10:17:28 +0200 Subject: [PATCH 0683/1443] ALSA: hda: intel-dsp-config: Add PTL-H support Use same recipes as PTL for PTL-H. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Bard Liao Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20250210081730.22916-3-peter.ujfalusi@linux.intel.com --- sound/hda/intel-dsp-config.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/sound/hda/intel-dsp-config.c b/sound/hda/intel-dsp-config.c index f564ec7af194..ce3ae2cba660 100644 --- a/sound/hda/intel-dsp-config.c +++ b/sound/hda/intel-dsp-config.c @@ -539,6 +539,11 @@ static const struct config_entry config_table[] = { .flags = FLAG_SOF | FLAG_SOF_ONLY_IF_DMIC_OR_SOUNDWIRE, .device = PCI_DEVICE_ID_INTEL_HDA_PTL, }, + { + .flags = FLAG_SOF | FLAG_SOF_ONLY_IF_DMIC_OR_SOUNDWIRE, + .device = PCI_DEVICE_ID_INTEL_HDA_PTL_H, + }, + #endif }; From 4e9c87cfcd0584f2a2e2f352a43ff003d688f3a4 Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Mon, 10 Feb 2025 10:17:29 +0200 Subject: [PATCH 0684/1443] ASoC: SOF: Intel: pci-ptl: Add support for PTL-H PTL-H uses the same configuration as PTL. Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Bard Liao Acked-by: Mark Brown Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20250210081730.22916-4-peter.ujfalusi@linux.intel.com --- sound/soc/sof/intel/pci-ptl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/sof/intel/pci-ptl.c b/sound/soc/sof/intel/pci-ptl.c index 0aacdfac9fb4..c4fb6a2441b7 100644 --- a/sound/soc/sof/intel/pci-ptl.c +++ b/sound/soc/sof/intel/pci-ptl.c @@ -50,6 +50,7 @@ static const struct sof_dev_desc ptl_desc = { /* PCI IDs */ static const struct pci_device_id sof_pci_ids[] = { { PCI_DEVICE_DATA(INTEL, HDA_PTL, &ptl_desc) }, /* PTL */ + { PCI_DEVICE_DATA(INTEL, HDA_PTL_H, &ptl_desc) }, /* PTL-H */ { 0, } }; MODULE_DEVICE_TABLE(pci, sof_pci_ids); From d7e2447a4d51de5c3c03e3b7892898e98ddd9769 Mon Sep 17 00:00:00 2001 From: Pierre-Louis Bossart Date: Mon, 10 Feb 2025 10:17:30 +0200 Subject: [PATCH 0685/1443] ALSA: hda: hda-intel: add Panther Lake-H support Add Intel PTL-H audio Device ID. Signed-off-by: Pierre-Louis Bossart Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Bard Liao Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20250210081730.22916-5-peter.ujfalusi@linux.intel.com --- sound/pci/hda/hda_intel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 7d7f9aac50a9..67540e037309 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -2496,6 +2496,8 @@ static const struct pci_device_id azx_ids[] = { { PCI_DEVICE_DATA(INTEL, HDA_ARL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_SKYLAKE) }, /* Panther Lake */ { PCI_DEVICE_DATA(INTEL, HDA_PTL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_LNL) }, + /* Panther Lake-H */ + { PCI_DEVICE_DATA(INTEL, HDA_PTL_H, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_LNL) }, /* Apollolake (Broxton-P) */ { PCI_DEVICE_DATA(INTEL, HDA_APL, AZX_DRIVER_SKL | AZX_DCAPS_INTEL_BROXTON) }, /* Gemini-Lake */ From 70e90680c2592c38c62e5716f1296a2d74bae7af Mon Sep 17 00:00:00 2001 From: Nam Cao Date: Wed, 5 Feb 2025 11:46:33 +0100 Subject: [PATCH 0686/1443] ALSA: Switch to use hrtimer_setup() hrtimer_setup() takes the callback function pointer as argument and initializes the timer completely. Replace hrtimer_init() and the open coded initialization of hrtimer::function with the new setup mechanism. Patch was created by using Coccinelle. Acked-by: Zack Rusin Signed-off-by: Nam Cao Cc: Takashi Iwai Link: https://patch.msgid.link/598031332ce738c82286a158cb66eb7e735b2e79.1738746904.git.namcao@linutronix.de Signed-off-by: Takashi Iwai --- sound/core/hrtimer.c | 3 +-- sound/drivers/dummy.c | 3 +-- sound/drivers/pcsp/pcsp.c | 3 +-- sound/sh/sh_dac_audio.c | 3 +-- 4 files changed, 4 insertions(+), 8 deletions(-) diff --git a/sound/core/hrtimer.c b/sound/core/hrtimer.c index 147c1fea4708..e9c60dce59fb 100644 --- a/sound/core/hrtimer.c +++ b/sound/core/hrtimer.c @@ -66,9 +66,8 @@ static int snd_hrtimer_open(struct snd_timer *t) stime = kzalloc(sizeof(*stime), GFP_KERNEL); if (!stime) return -ENOMEM; - hrtimer_init(&stime->hrt, CLOCK_MONOTONIC, HRTIMER_MODE_REL); stime->timer = t; - stime->hrt.function = snd_hrtimer_callback; + hrtimer_setup(&stime->hrt, snd_hrtimer_callback, CLOCK_MONOTONIC, HRTIMER_MODE_REL); t->private_data = stime; return 0; } diff --git a/sound/drivers/dummy.c b/sound/drivers/dummy.c index 8f5df9b3aaaa..c1a3efb633c5 100644 --- a/sound/drivers/dummy.c +++ b/sound/drivers/dummy.c @@ -457,8 +457,7 @@ static int dummy_hrtimer_create(struct snd_pcm_substream *substream) if (!dpcm) return -ENOMEM; substream->runtime->private_data = dpcm; - hrtimer_init(&dpcm->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT); - dpcm->timer.function = dummy_hrtimer_callback; + hrtimer_setup(&dpcm->timer, dummy_hrtimer_callback, CLOCK_MONOTONIC, HRTIMER_MODE_REL_SOFT); dpcm->substream = substream; atomic_set(&dpcm->running, 0); return 0; diff --git a/sound/drivers/pcsp/pcsp.c b/sound/drivers/pcsp/pcsp.c index 78c9b1c7590f..e8482c2290c3 100644 --- a/sound/drivers/pcsp/pcsp.c +++ b/sound/drivers/pcsp/pcsp.c @@ -103,8 +103,7 @@ static int snd_card_pcsp_probe(int devnum, struct device *dev) if (devnum != 0) return -EINVAL; - hrtimer_init(&pcsp_chip.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - pcsp_chip.timer.function = pcsp_do_timer; + hrtimer_setup(&pcsp_chip.timer, pcsp_do_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); err = snd_devm_card_new(dev, index, id, THIS_MODULE, 0, &card); if (err < 0) diff --git a/sound/sh/sh_dac_audio.c b/sound/sh/sh_dac_audio.c index 3f5422145c5e..84a4b17a0cc2 100644 --- a/sound/sh/sh_dac_audio.c +++ b/sound/sh/sh_dac_audio.c @@ -312,8 +312,7 @@ static int snd_sh_dac_create(struct snd_card *card, chip->card = card; - hrtimer_init(&chip->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - chip->hrtimer.function = sh_dac_audio_timer; + hrtimer_setup(&chip->hrtimer, sh_dac_audio_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); dac_audio_reset(chip); chip->rate = 8000; From 4720e0ad30bbe87821e516e5c6cdeb94e615c6dd Mon Sep 17 00:00:00 2001 From: Andrzej Kacprowski Date: Tue, 4 Feb 2025 09:46:17 +0100 Subject: [PATCH 0687/1443] accel/ivpu: Add missing locks around mmu queues Multiple threads were accessing mmu cmd queue simultaneously causing sporadic failures in ivpu_mmu_cmdq_sync() function. Protect critical code with mmu mutex. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Andrzej Kacprowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-2-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_mmu.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index b80bdded9fd7..5ea010568faa 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -895,6 +895,9 @@ static int ivpu_mmu_evtq_disable(struct ivpu_device *vdev) void ivpu_mmu_discard_events(struct ivpu_device *vdev) { + struct ivpu_mmu_info *mmu = vdev->mmu; + + mutex_lock(&mmu->lock); /* * Disable event queue (stop MMU from updating the producer) * to allow synchronization of consumer and producer indexes @@ -908,6 +911,8 @@ void ivpu_mmu_discard_events(struct ivpu_device *vdev) ivpu_mmu_evtq_enable(vdev); drm_WARN_ON_ONCE(&vdev->drm, vdev->mmu->evtq.cons != vdev->mmu->evtq.prod); + + mutex_unlock(&mmu->lock); } int ivpu_mmu_disable_ssid_events(struct ivpu_device *vdev, u32 ssid) @@ -920,6 +925,8 @@ int ivpu_mmu_disable_ssid_events(struct ivpu_device *vdev, u32 ssid) if (ssid > IVPU_MMU_CDTAB_ENT_COUNT) return -EINVAL; + mutex_lock(&mmu->lock); + entry = cdtab->base + (ssid * IVPU_MMU_CDTAB_ENT_SIZE); val = READ_ONCE(entry[0]); @@ -932,6 +939,8 @@ int ivpu_mmu_disable_ssid_events(struct ivpu_device *vdev, u32 ssid) ivpu_mmu_cmdq_write_cfgi_all(vdev); ivpu_mmu_cmdq_sync(vdev); + mutex_unlock(&mmu->lock); + return 0; } From 7806bad76ac397a767f0c369534133c71c73b157 Mon Sep 17 00:00:00 2001 From: Andrzej Kacprowski Date: Tue, 4 Feb 2025 09:46:18 +0100 Subject: [PATCH 0688/1443] accel/ivpu: Prevent runtime suspend during context abort work Increment the runtime PM counter when entering ivpu_context_abort_work_fn() to prevent the device from suspending while the function is executing. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Andrzej Kacprowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-3-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_job.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index c1013f511efa..004059e4f1e8 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -965,6 +966,9 @@ void ivpu_context_abort_work_fn(struct work_struct *work) unsigned long ctx_id; unsigned long id; + if (drm_WARN_ON(&vdev->drm, pm_runtime_get_if_active(vdev->drm.dev) <= 0)) + return; + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) ivpu_jsm_reset_engine(vdev, 0); @@ -987,7 +991,7 @@ void ivpu_context_abort_work_fn(struct work_struct *work) ivpu_mmu_discard_events(vdev); if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) - return; + goto runtime_put; ivpu_jsm_hws_resume_engine(vdev, 0); /* @@ -1000,4 +1004,8 @@ void ivpu_context_abort_work_fn(struct work_struct *work) if (job->file_priv->aborted) ivpu_job_signal_and_destroy(vdev, job->job_id, DRM_IVPU_JOB_STATUS_ABORTED); mutex_unlock(&vdev->submitted_jobs_lock); + +runtime_put: + pm_runtime_mark_last_busy(vdev->drm.dev); + pm_runtime_put_autosuspend(vdev->drm.dev); } From 320323d2e5456df9d6236ac1ce9c030b1a74aa5b Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 4 Feb 2025 09:46:19 +0100 Subject: [PATCH 0689/1443] accel/ivpu: Add debugfs interface for setting HWS priority bands Add debugfs interface to modify following priority bands properties: * grace period * process grace period * process quantum This allows for the adjustment of hardware scheduling algorithm parameters for each existing priority band, facilitating validation and fine-tuning. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-4-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_debugfs.c | 84 +++++++++++++++++++++++++++++++ drivers/accel/ivpu/ivpu_hw.c | 21 ++++++++ drivers/accel/ivpu/ivpu_hw.h | 5 ++ drivers/accel/ivpu/ivpu_jsm_msg.c | 29 ++++------- 4 files changed, 121 insertions(+), 18 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c index e79715c53f2a..0825851656a2 100644 --- a/drivers/accel/ivpu/ivpu_debugfs.c +++ b/drivers/accel/ivpu/ivpu_debugfs.c @@ -398,6 +398,88 @@ static int dct_active_set(void *data, u64 active_percent) DEFINE_DEBUGFS_ATTRIBUTE(ivpu_dct_fops, dct_active_get, dct_active_set, "%llu\n"); +static int priority_bands_show(struct seq_file *s, void *v) +{ + struct ivpu_device *vdev = s->private; + struct ivpu_hw_info *hw = vdev->hw; + + for (int band = VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE; + band < VPU_JOB_SCHEDULING_PRIORITY_BAND_COUNT; band++) { + switch (band) { + case VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE: + seq_puts(s, "Idle: "); + break; + + case VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL: + seq_puts(s, "Normal: "); + break; + + case VPU_JOB_SCHEDULING_PRIORITY_BAND_FOCUS: + seq_puts(s, "Focus: "); + break; + + case VPU_JOB_SCHEDULING_PRIORITY_BAND_REALTIME: + seq_puts(s, "Realtime: "); + break; + } + + seq_printf(s, "grace_period %9u process_grace_period %9u process_quantum %9u\n", + hw->hws.grace_period[band], hw->hws.process_grace_period[band], + hw->hws.process_quantum[band]); + } + + return 0; +} + +static int priority_bands_fops_open(struct inode *inode, struct file *file) +{ + return single_open(file, priority_bands_show, inode->i_private); +} + +static ssize_t +priority_bands_fops_write(struct file *file, const char __user *user_buf, size_t size, loff_t *pos) +{ + struct seq_file *s = file->private_data; + struct ivpu_device *vdev = s->private; + char buf[64]; + u32 grace_period; + u32 process_grace_period; + u32 process_quantum; + u32 band; + int ret; + + if (size >= sizeof(buf)) + return -EINVAL; + + ret = simple_write_to_buffer(buf, sizeof(buf) - 1, pos, user_buf, size); + if (ret < 0) + return ret; + + buf[size] = '\0'; + ret = sscanf(buf, "%u %u %u %u", &band, &grace_period, &process_grace_period, + &process_quantum); + if (ret != 4) + return -EINVAL; + + if (band >= VPU_JOB_SCHEDULING_PRIORITY_BAND_COUNT) + return -EINVAL; + + vdev->hw->hws.grace_period[band] = grace_period; + vdev->hw->hws.process_grace_period[band] = process_grace_period; + vdev->hw->hws.process_quantum[band] = process_quantum; + + return size; +} + +static const struct file_operations ivpu_hws_priority_bands_fops = { + .owner = THIS_MODULE, + .open = priority_bands_fops_open, + .write = priority_bands_fops_write, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + void ivpu_debugfs_init(struct ivpu_device *vdev) { struct dentry *debugfs_root = vdev->drm.debugfs_root; @@ -420,6 +502,8 @@ void ivpu_debugfs_init(struct ivpu_device *vdev) &fw_trace_hw_comp_mask_fops); debugfs_create_file("fw_trace_level", 0200, debugfs_root, vdev, &fw_trace_level_fops); + debugfs_create_file("hws_priority_bands", 0200, debugfs_root, vdev, + &ivpu_hws_priority_bands_fops); debugfs_create_file("reset_engine", 0200, debugfs_root, vdev, &ivpu_reset_engine_fops); diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 2057907c5aa6..f31702b3ecbd 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -113,6 +113,26 @@ static void timeouts_init(struct ivpu_device *vdev) } } +static void priority_bands_init(struct ivpu_device *vdev) +{ + /* Idle */ + vdev->hw->hws.grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE] = 0; + vdev->hw->hws.process_grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE] = 50000; + vdev->hw->hws.process_quantum[VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE] = 160000; + /* Normal */ + vdev->hw->hws.grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL] = 50000; + vdev->hw->hws.process_grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL] = 50000; + vdev->hw->hws.process_quantum[VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL] = 300000; + /* Focus */ + vdev->hw->hws.grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_FOCUS] = 50000; + vdev->hw->hws.process_grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_FOCUS] = 50000; + vdev->hw->hws.process_quantum[VPU_JOB_SCHEDULING_PRIORITY_BAND_FOCUS] = 200000; + /* Realtime */ + vdev->hw->hws.grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_REALTIME] = 0; + vdev->hw->hws.process_grace_period[VPU_JOB_SCHEDULING_PRIORITY_BAND_REALTIME] = 50000; + vdev->hw->hws.process_quantum[VPU_JOB_SCHEDULING_PRIORITY_BAND_REALTIME] = 200000; +} + static void memory_ranges_init(struct ivpu_device *vdev) { if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { @@ -251,6 +271,7 @@ int ivpu_hw_init(struct ivpu_device *vdev) { ivpu_hw_btrs_info_init(vdev); ivpu_hw_btrs_freq_ratios_init(vdev); + priority_bands_init(vdev); memory_ranges_init(vdev); platform_init(vdev); wa_init(vdev); diff --git a/drivers/accel/ivpu/ivpu_hw.h b/drivers/accel/ivpu/ivpu_hw.h index fbef9816b9d0..16435f2756d0 100644 --- a/drivers/accel/ivpu/ivpu_hw.h +++ b/drivers/accel/ivpu/ivpu_hw.h @@ -36,6 +36,11 @@ struct ivpu_hw_info { u8 pn_ratio; u32 profiling_freq; } pll; + struct { + u32 grace_period[VPU_HWS_NUM_PRIORITY_BANDS]; + u32 process_quantum[VPU_HWS_NUM_PRIORITY_BANDS]; + u32 process_grace_period[VPU_HWS_NUM_PRIORITY_BANDS]; + } hws; u32 tile_fuse; u32 sku; u16 config; diff --git a/drivers/accel/ivpu/ivpu_jsm_msg.c b/drivers/accel/ivpu/ivpu_jsm_msg.c index 30a40be76930..219ab8afefab 100644 --- a/drivers/accel/ivpu/ivpu_jsm_msg.c +++ b/drivers/accel/ivpu/ivpu_jsm_msg.c @@ -7,6 +7,7 @@ #include "ivpu_hw.h" #include "ivpu_ipc.h" #include "ivpu_jsm_msg.h" +#include "vpu_jsm_api.h" const char *ivpu_jsm_msg_type_to_str(enum vpu_ipc_msg_type type) { @@ -407,26 +408,18 @@ int ivpu_jsm_hws_setup_priority_bands(struct ivpu_device *vdev) { struct vpu_jsm_msg req = { .type = VPU_JSM_MSG_SET_PRIORITY_BAND_SETUP }; struct vpu_jsm_msg resp; + struct ivpu_hw_info *hw = vdev->hw; + struct vpu_ipc_msg_payload_hws_priority_band_setup *setup = + &req.payload.hws_priority_band_setup; int ret; - /* Idle */ - req.payload.hws_priority_band_setup.grace_period[0] = 0; - req.payload.hws_priority_band_setup.process_grace_period[0] = 50000; - req.payload.hws_priority_band_setup.process_quantum[0] = 160000; - /* Normal */ - req.payload.hws_priority_band_setup.grace_period[1] = 50000; - req.payload.hws_priority_band_setup.process_grace_period[1] = 50000; - req.payload.hws_priority_band_setup.process_quantum[1] = 300000; - /* Focus */ - req.payload.hws_priority_band_setup.grace_period[2] = 50000; - req.payload.hws_priority_band_setup.process_grace_period[2] = 50000; - req.payload.hws_priority_band_setup.process_quantum[2] = 200000; - /* Realtime */ - req.payload.hws_priority_band_setup.grace_period[3] = 0; - req.payload.hws_priority_band_setup.process_grace_period[3] = 50000; - req.payload.hws_priority_band_setup.process_quantum[3] = 200000; - - req.payload.hws_priority_band_setup.normal_band_percentage = 10; + for (int band = VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE; + band < VPU_JOB_SCHEDULING_PRIORITY_BAND_COUNT; band++) { + setup->grace_period[band] = hw->hws.grace_period[band]; + setup->process_grace_period[band] = hw->hws.process_grace_period[band]; + setup->process_quantum[band] = hw->hws.process_quantum[band]; + } + setup->normal_band_percentage = 10; ret = ivpu_ipc_send_receive_internal(vdev, &req, VPU_JSM_MSG_SET_PRIORITY_BAND_SETUP_RSP, &resp, VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); From 67725f5e8d844e27db4514a76699ec6a7000c627 Mon Sep 17 00:00:00 2001 From: Tomasz Rusinowicz Date: Tue, 4 Feb 2025 09:46:20 +0100 Subject: [PATCH 0690/1443] accel/ivpu: Allow to import single buffer into multiple contexts Use ivpu_gem_prime_import() based on drm_gem_prime_import_dev() for importing buffers, removing optimization for same device imports. This optimization reused the same ivpu_bo object in multiple contexts but a single buffer can be MMU-mapped only to a single context. Each import now creates a new instance of ivpu_bo object that shares the same sg_table but have separate MMU mappings. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Tomasz Rusinowicz Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-5-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.c | 2 +- drivers/accel/ivpu/ivpu_gem.c | 43 +++++++++++++++++++++++++++++++++++ drivers/accel/ivpu/ivpu_gem.h | 1 + 3 files changed, 45 insertions(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index 6a80d626d609..f23e3e8ea9d8 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -452,7 +452,7 @@ static const struct drm_driver driver = { .postclose = ivpu_postclose, .gem_create_object = ivpu_gem_create_object, - .gem_prime_import_sg_table = drm_gem_shmem_prime_import_sg_table, + .gem_prime_import = ivpu_gem_prime_import, .ioctls = ivpu_drm_ioctls, .num_ioctls = ARRAY_SIZE(ivpu_drm_ioctls), diff --git a/drivers/accel/ivpu/ivpu_gem.c b/drivers/accel/ivpu/ivpu_gem.c index 16178054e629..8741c73b92ce 100644 --- a/drivers/accel/ivpu/ivpu_gem.c +++ b/drivers/accel/ivpu/ivpu_gem.c @@ -20,6 +20,8 @@ #include "ivpu_mmu.h" #include "ivpu_mmu_context.h" +MODULE_IMPORT_NS("DMA_BUF"); + static const struct drm_gem_object_funcs ivpu_gem_funcs; static inline void ivpu_dbg_bo(struct ivpu_device *vdev, struct ivpu_bo *bo, const char *action) @@ -172,6 +174,47 @@ struct drm_gem_object *ivpu_gem_create_object(struct drm_device *dev, size_t siz return &bo->base.base; } +struct drm_gem_object *ivpu_gem_prime_import(struct drm_device *dev, + struct dma_buf *dma_buf) +{ + struct device *attach_dev = dev->dev; + struct dma_buf_attachment *attach; + struct sg_table *sgt; + struct drm_gem_object *obj; + int ret; + + attach = dma_buf_attach(dma_buf, attach_dev); + if (IS_ERR(attach)) + return ERR_CAST(attach); + + get_dma_buf(dma_buf); + + sgt = dma_buf_map_attachment_unlocked(attach, DMA_BIDIRECTIONAL); + if (IS_ERR(sgt)) { + ret = PTR_ERR(sgt); + goto fail_detach; + } + + obj = drm_gem_shmem_prime_import_sg_table(dev, attach, sgt); + if (IS_ERR(obj)) { + ret = PTR_ERR(obj); + goto fail_unmap; + } + + obj->import_attach = attach; + obj->resv = dma_buf->resv; + + return obj; + +fail_unmap: + dma_buf_unmap_attachment_unlocked(attach, sgt, DMA_BIDIRECTIONAL); +fail_detach: + dma_buf_detach(dma_buf, attach); + dma_buf_put(dma_buf); + + return ERR_PTR(ret); +} + static struct ivpu_bo *ivpu_bo_alloc(struct ivpu_device *vdev, u64 size, u32 flags) { struct drm_gem_shmem_object *shmem; diff --git a/drivers/accel/ivpu/ivpu_gem.h b/drivers/accel/ivpu/ivpu_gem.h index d975000abd78..a222a9ec9d61 100644 --- a/drivers/accel/ivpu/ivpu_gem.h +++ b/drivers/accel/ivpu/ivpu_gem.h @@ -28,6 +28,7 @@ int ivpu_bo_pin(struct ivpu_bo *bo); void ivpu_bo_unbind_all_bos_from_context(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx); struct drm_gem_object *ivpu_gem_create_object(struct drm_device *dev, size_t size); +struct drm_gem_object *ivpu_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf); struct ivpu_bo *ivpu_bo_create(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, struct ivpu_addr_range *range, u64 size, u32 flags); struct ivpu_bo *ivpu_bo_create_global(struct ivpu_device *vdev, u64 size, u32 flags); From 55e856c344b0473e78729e59e501b28d3c238b51 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 4 Feb 2025 09:46:21 +0100 Subject: [PATCH 0691/1443] accel/ivpu: Add test modes to toggle clock relinquish disable Add IVPU_TEST_MODE_CLK_RELINQ_[DISABLE|ENABLE] that overrides workaround for disabling clock relinquish for testing purposes. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-6-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.h | 2 ++ drivers/accel/ivpu/ivpu_hw.c | 6 ++++++ 2 files changed, 8 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index ca21102ca366..73d808068c8b 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -206,6 +206,8 @@ extern bool ivpu_force_snoop; #define IVPU_TEST_MODE_MIP_DISABLE BIT(6) #define IVPU_TEST_MODE_DISABLE_TIMEOUTS BIT(8) #define IVPU_TEST_MODE_TURBO BIT(9) +#define IVPU_TEST_MODE_CLK_RELINQ_DISABLE BIT(10) +#define IVPU_TEST_MODE_CLK_RELINQ_ENABLE BIT(11) extern int ivpu_test_mode; struct ivpu_file_priv *ivpu_file_priv_get(struct ivpu_file_priv *file_priv); diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index f31702b3ecbd..4d9f30e3b505 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -68,6 +68,12 @@ static void wa_init(struct ivpu_device *vdev) ivpu_revision(vdev) < IVPU_HW_IP_REV_LNL_B0) vdev->wa.disable_clock_relinquish = true; + if (ivpu_test_mode & IVPU_TEST_MODE_CLK_RELINQ_ENABLE) + vdev->wa.disable_clock_relinquish = false; + + if (ivpu_test_mode & IVPU_TEST_MODE_CLK_RELINQ_DISABLE) + vdev->wa.disable_clock_relinquish = true; + if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) vdev->wa.wp0_during_power_up = true; From 011529fe81121e2d343522ecbb149e9b2d1e5931 Mon Sep 17 00:00:00 2001 From: Karol Wachowski Date: Tue, 4 Feb 2025 09:46:22 +0100 Subject: [PATCH 0692/1443] accel/ivpu: Implement D0i2 disable test mode Add power_profile firmware boot param and set it to 0 by default which is default FW power profile. Implement IVPU_TEST_MODE_D0I2_DISABLE which is used for setting power profile boot param value to 1 which prevents NPU from entering d0i2 power state. Reviewed-by: Jacek Lawrynowicz Signed-off-by: Karol Wachowski Signed-off-by: Jacek Lawrynowicz Link: https://patchwork.freedesktop.org/patch/msgid/20250204084622.2422544-7-jacek.lawrynowicz@linux.intel.com --- drivers/accel/ivpu/ivpu_drv.h | 2 ++ drivers/accel/ivpu/ivpu_fw.c | 4 ++++ drivers/accel/ivpu/ivpu_hw.c | 4 ++++ 3 files changed, 10 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index 73d808068c8b..92753effb1c9 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -111,6 +111,7 @@ struct ivpu_wa_table { bool disable_clock_relinquish; bool disable_d0i3_msg; bool wp0_during_power_up; + bool disable_d0i2; }; struct ivpu_hw_info; @@ -208,6 +209,7 @@ extern bool ivpu_force_snoop; #define IVPU_TEST_MODE_TURBO BIT(9) #define IVPU_TEST_MODE_CLK_RELINQ_DISABLE BIT(10) #define IVPU_TEST_MODE_CLK_RELINQ_ENABLE BIT(11) +#define IVPU_TEST_MODE_D0I2_DISABLE BIT(12) extern int ivpu_test_mode; struct ivpu_file_priv *ivpu_file_priv_get(struct ivpu_file_priv *file_priv); diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index cfe8f79ec5a6..7a1bb92d8c81 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -534,6 +534,8 @@ static void ivpu_fw_boot_params_print(struct ivpu_device *vdev, struct vpu_boot_ boot_params->d0i3_entry_vpu_ts); ivpu_dbg(vdev, FW_BOOT, "boot_params.system_time_us = %llu\n", boot_params->system_time_us); + ivpu_dbg(vdev, FW_BOOT, "boot_params.power_profile = %u\n", + boot_params->power_profile); } void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params *boot_params) @@ -634,6 +636,8 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params boot_params->d0i3_delayed_entry = 1; boot_params->d0i3_residency_time_us = 0; boot_params->d0i3_entry_vpu_ts = 0; + if (IVPU_WA(disable_d0i2)) + boot_params->power_profile = 1; boot_params->system_time_us = ktime_to_us(ktime_get_real()); wmb(); /* Flush WC buffers after writing bootparams */ diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 4d9f30e3b505..ec9a3629da3a 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -77,11 +77,15 @@ static void wa_init(struct ivpu_device *vdev) if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) vdev->wa.wp0_during_power_up = true; + if (ivpu_test_mode & IVPU_TEST_MODE_D0I2_DISABLE) + vdev->wa.disable_d0i2 = true; + IVPU_PRINT_WA(punit_disabled); IVPU_PRINT_WA(clear_runtime_mem); IVPU_PRINT_WA(interrupt_clear_with_0); IVPU_PRINT_WA(disable_clock_relinquish); IVPU_PRINT_WA(wp0_during_power_up); + IVPU_PRINT_WA(disable_d0i2); } static void timeouts_init(struct ivpu_device *vdev) From 8dbccafce3c8ae026606f5c7bc6637667d9d5595 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 10 Feb 2025 09:17:58 +0000 Subject: [PATCH 0693/1443] KVM: arm64: Fix __pkvm_host_mkyoung_guest() return value Don't use an uninitialised stack variable, and just return 0 on the non-error path. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202502100911.8c9DbtKD-lkp@intel.com/ Reviewed-by: Quentin Perret Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/mem_protect.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index 4c2f6a6a2efe..19c3c631708c 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -1073,7 +1073,6 @@ int __pkvm_host_mkyoung_guest(u64 gfn, struct pkvm_hyp_vcpu *vcpu) { struct pkvm_hyp_vm *vm = pkvm_hyp_vcpu_to_hyp_vm(vcpu); u64 ipa = hyp_pfn_to_phys(gfn); - int ret; if (pkvm_hyp_vm_is_protected(vm)) return -EPERM; @@ -1083,5 +1082,5 @@ int __pkvm_host_mkyoung_guest(u64 gfn, struct pkvm_hyp_vcpu *vcpu) kvm_pgtable_stage2_mkyoung(&vm->pgt, ipa, 0); guest_unlock_component(vm); - return ret; + return 0; } From cbc0a0ee34b5f108b184a6d59afd0f305564350a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= Date: Fri, 7 Feb 2025 12:31:11 +0100 Subject: [PATCH 0694/1443] drm/xe/pf: Use an explicit check to see if the device has LMTT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far, the main condition for using LMTT has been to check that the device is a discrete gfx. Let's add a dedicated function to check if the device supports LMTT as not all future discrete GPU platforms will require LMTT. v2: - use xe_has_device_lmtt only when necessary - leave IS_DGFX for other things related to LMEM provisioning v3: - remove IS_SRIOV_PF condition from xe_device_has_lmtt (Michal Wajdeczko) - keep IS_SRIOV_PF asserts in LMTT-related code (Michal Wajdeczko) v4: - update commit description Signed-off-by: Piotr Piórkowski Cc: Michal Wajdeczko Cc: Michał Winiarski Cc: Satyanarayana K V P Reviewed-by: Tejas Upadhyay Reviewed-by: Satyanarayana K V P Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250207113111.853821-2-piotr.piorkowski@intel.com --- drivers/gpu/drm/xe/xe_device.h | 5 +++++ drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 21 +++++++++++++-------- drivers/gpu/drm/xe/xe_lmtt.c | 4 ++-- 3 files changed, 20 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h index fc3c2af3fb7f..0bc3bc8e6803 100644 --- a/drivers/gpu/drm/xe/xe_device.h +++ b/drivers/gpu/drm/xe/xe_device.h @@ -170,6 +170,11 @@ static inline bool xe_device_uses_memirq(struct xe_device *xe) return xe_device_has_memirq(xe) && (IS_SRIOV_VF(xe) || xe_device_has_msix(xe)); } +static inline bool xe_device_has_lmtt(struct xe_device *xe) +{ + return IS_DGFX(xe); +} + u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size); void xe_device_snapshot_print(struct xe_device *xe, struct drm_printer *p); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index b1d994d65589..5c3e9e5bd051 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -1336,7 +1336,7 @@ static void pf_reset_vf_lmtt(struct xe_device *xe, unsigned int vfid) struct xe_tile *tile; unsigned int tid; - xe_assert(xe, IS_DGFX(xe)); + xe_assert(xe, xe_device_has_lmtt(xe)); xe_assert(xe, IS_SRIOV_PF(xe)); for_each_tile(tile, xe, tid) { @@ -1357,7 +1357,7 @@ static int pf_update_vf_lmtt(struct xe_device *xe, unsigned int vfid) unsigned int tid; int err; - xe_assert(xe, IS_DGFX(xe)); + xe_assert(xe, xe_device_has_lmtt(xe)); xe_assert(xe, IS_SRIOV_PF(xe)); total = 0; @@ -1434,7 +1434,8 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) if (unlikely(err)) return err; - pf_reset_vf_lmtt(xe, vfid); + if (xe_device_has_lmtt(xe)) + pf_reset_vf_lmtt(xe, vfid); pf_release_vf_config_lmem(gt, config); } xe_gt_assert(gt, !config->lmem_obj); @@ -1454,9 +1455,11 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) config->lmem_obj = bo; - err = pf_update_vf_lmtt(xe, vfid); - if (unlikely(err)) - goto release; + if (xe_device_has_lmtt(xe)) { + err = pf_update_vf_lmtt(xe, vfid); + if (unlikely(err)) + goto release; + } err = pf_push_vf_cfg_lmem(gt, vfid, bo->size); if (unlikely(err)) @@ -1467,7 +1470,8 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) return 0; reset_lmtt: - pf_reset_vf_lmtt(xe, vfid); + if (xe_device_has_lmtt(xe)) + pf_reset_vf_lmtt(xe, vfid); release: pf_release_vf_config_lmem(gt, config); return err; @@ -1981,7 +1985,8 @@ static void pf_release_vf_config(struct xe_gt *gt, unsigned int vfid) pf_release_vf_config_ggtt(gt, config); if (IS_DGFX(xe)) { pf_release_vf_config_lmem(gt, config); - pf_update_vf_lmtt(xe, vfid); + if (xe_device_has_lmtt(xe)) + pf_update_vf_lmtt(xe, vfid); } } pf_release_config_ctxs(gt, config); diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c index a60ceae4c6dd..89393dcb53d9 100644 --- a/drivers/gpu/drm/xe/xe_lmtt.c +++ b/drivers/gpu/drm/xe/xe_lmtt.c @@ -164,7 +164,7 @@ int xe_lmtt_init(struct xe_lmtt *lmtt) lmtt_assert(lmtt, IS_SRIOV_PF(xe)); lmtt_assert(lmtt, !lmtt->ops); - if (!IS_DGFX(xe)) + if (!xe_device_has_lmtt(xe)) return 0; if (xe_has_multi_level_lmtt(xe)) @@ -486,7 +486,7 @@ u64 xe_lmtt_estimate_pt_size(struct xe_lmtt *lmtt, u64 size) u64 pt_size; lmtt_assert(lmtt, IS_SRIOV_PF(lmtt_to_xe(lmtt))); - lmtt_assert(lmtt, IS_DGFX(lmtt_to_xe(lmtt))); + lmtt_assert(lmtt, xe_device_has_lmtt(lmtt_to_xe(lmtt))); lmtt_assert(lmtt, lmtt->ops); pt_size = PAGE_ALIGN(lmtt->ops->lmtt_pte_size(level) * From b921f66ccf5e8cf1b8a5052b35ceda454f19f5dd Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Mon, 3 Feb 2025 15:30:56 -0600 Subject: [PATCH 0695/1443] dt-bindings: rockchip: pmu: Ensure all properties are defined Device specific schemas should not allow undefined properties which is what 'additionalProperties: true' allows. Add the missing child nodes and fix this constraint. Signed-off-by: Rob Herring (Arm) Link: https://lore.kernel.org/r/20250203213056.13827-1-robh@kernel.org Signed-off-by: Heiko Stuebner --- Documentation/devicetree/bindings/arm/rockchip/pmu.yaml | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/arm/rockchip/pmu.yaml b/Documentation/devicetree/bindings/arm/rockchip/pmu.yaml index 932f981265cc..52016a141227 100644 --- a/Documentation/devicetree/bindings/arm/rockchip/pmu.yaml +++ b/Documentation/devicetree/bindings/arm/rockchip/pmu.yaml @@ -53,11 +53,17 @@ properties: reg: maxItems: 1 + power-controller: + type: object + + reboot-mode: + type: object + required: - compatible - reg -additionalProperties: true +additionalProperties: false examples: - | From 2c202e6c4f4dd19d2e8c1dfac9df05170aa3934f Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Sat, 8 Feb 2025 08:29:15 +0900 Subject: [PATCH 0696/1443] ata: libahci_platform: Do not set mask_port_map when not needed Commit 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port numbers") modified ahci_platform_get_resources() to allow identifying the ports of a controller that are defined as child nodes of the controller node in order to support non-consecutive port numbers (as defined by the platform device tree). However, this commit also erroneously sets bit 0 of hpriv->mask_port_map when the platform devices tree does not define port child nodes, to match the fact that the temporary default number of ports used in that case is 1 (which is also consistent with the fact that only index 0 of hpriv->phys[] is initialized with the call to ahci_platform_get_phy(). But doing so causes ahci_platform_init_host() to initialize and probe only the first port, even if this function determines that the controller has in fact multiple ports using the capability register of the controller (through a call to ahci_nr_ports()). This can be seen with the ahci_mvebu driver (Armada 385 SoC) with the second port declared as "dummy": ahci-mvebu f10a8000.sata: masking port_map 0x3 -> 0x1 ahci-mvebu f10a8000.sata: AHCI vers 0001.0000, 32 command slots, 6 Gbps, platform mode ahci-mvebu f10a8000.sata: 1/2 ports implemented (port mask 0x1) ahci-mvebu f10a8000.sata: flags: 64bit ncq sntf led only pmp fbs pio slum part sxs scsi host0: ahci-mvebu scsi host1: ahci-mvebu ata1: SATA max UDMA/133 mmio [mem 0xf10a8000-0xf10a9fff] port 0x100 irq 40 lpm-pol 0 ata2: DUMMY Fix this issue by removing setting bit 0 of hpriv->mask_port_map when the platform device tree does not define port child nodes. Reported-by: Klaus Kudielka Fixes: 8c87215dd3a2 ("ata: libahci_platform: support non-consecutive port numbers") Tested-by: Klaus Kudielka Signed-off-by: Damien Le Moal Acked-by: Josua Mayer Link: https://lore.kernel.org/r/20250207232915.1439174-1-dlemoal@kernel.org Signed-off-by: Niklas Cassel --- drivers/ata/libahci_platform.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/ata/libahci_platform.c b/drivers/ata/libahci_platform.c index 53b2c7719dc5..91d44302eac9 100644 --- a/drivers/ata/libahci_platform.c +++ b/drivers/ata/libahci_platform.c @@ -651,8 +651,6 @@ struct ahci_host_priv *ahci_platform_get_resources(struct platform_device *pdev, * If no sub-node was found, keep this for device tree * compatibility */ - hpriv->mask_port_map |= BIT(0); - rc = ahci_platform_get_phy(hpriv, 0, dev, dev->of_node); if (rc) goto err_out; From fc3a50c12ebd33c77b7e4a1e4f154a44315d3169 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= Date: Mon, 10 Feb 2025 09:15:10 +0100 Subject: [PATCH 0697/1443] drm/xe: Rename struct xe_mem_region to struct xe_vram_region MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The xe_mem_region structure has so far been used only in the context of VRAM regions. Also, the description of its fields clearly indicates that it was designed for VRAM regions. This struct is strictly related only to VRAM. So let's be clear on this point and rename it to xe_vram_region. Signed-off-by: Piotr Piórkowski Reviewed-by: Rodrigo Vivi Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250210081511.906452-2-piotr.piorkowski@intel.com --- drivers/gpu/drm/xe/xe_bo.c | 12 ++++++------ drivers/gpu/drm/xe/xe_device_types.h | 8 ++++---- drivers/gpu/drm/xe/xe_ttm_vram_mgr.c | 2 +- drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h | 4 ++-- 4 files changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 6812164e1470..a202d5525822 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -143,7 +143,7 @@ mem_type_to_migrate(struct xe_device *xe, u32 mem_type) return tile->migrate; } -static struct xe_mem_region *res_to_mem_region(struct ttm_resource *res) +static struct xe_vram_region *res_to_mem_region(struct ttm_resource *res) { struct xe_device *xe = ttm_to_xe_device(res->bo->bdev); struct ttm_resource_manager *mgr; @@ -179,7 +179,7 @@ static void add_vram(struct xe_device *xe, struct xe_bo *bo, struct ttm_place *places, u32 bo_flags, u32 mem_type, u32 *c) { struct ttm_place place = { .mem_type = mem_type }; - struct xe_mem_region *vram; + struct xe_vram_region *vram; u64 io_size; xe_assert(xe, *c < ARRAY_SIZE(bo->placements)); @@ -468,7 +468,7 @@ static int xe_ttm_io_mem_reserve(struct ttm_device *bdev, return 0; case XE_PL_VRAM0: case XE_PL_VRAM1: { - struct xe_mem_region *vram = res_to_mem_region(mem); + struct xe_vram_region *vram = res_to_mem_region(mem); if (!xe_ttm_resource_visible(mem)) return -EINVAL; @@ -815,7 +815,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict, /* Create a new VMAP once kernel BO back in VRAM */ if (!ret && resource_is_vram(new_mem)) { - struct xe_mem_region *vram = res_to_mem_region(new_mem); + struct xe_vram_region *vram = res_to_mem_region(new_mem); void __iomem *new_addr = vram->mapping + (new_mem->start << PAGE_SHIFT); @@ -1025,7 +1025,7 @@ static unsigned long xe_ttm_io_mem_pfn(struct ttm_buffer_object *ttm_bo, { struct xe_bo *bo = ttm_to_xe_bo(ttm_bo); struct xe_res_cursor cursor; - struct xe_mem_region *vram; + struct xe_vram_region *vram; if (ttm_bo->resource->mem_type == XE_PL_STOLEN) return xe_ttm_stolen_io_offset(bo, page_offset << PAGE_SHIFT) >> PAGE_SHIFT; @@ -1165,7 +1165,7 @@ static int xe_ttm_access_memory(struct ttm_buffer_object *ttm_bo, struct xe_device *xe = ttm_to_xe_device(ttm_bo->bdev); struct iosys_map vmap; struct xe_res_cursor cursor; - struct xe_mem_region *vram; + struct xe_vram_region *vram; int bytes_left = len; xe_bo_assert_held(bo); diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index c0e886bac183..1f27ed2c0448 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -70,11 +70,11 @@ struct xe_pxp; struct xe_tile * : (tile__)->xe) /** - * struct xe_mem_region - memory region structure + * struct xe_vram_region - memory region structure * This is used to describe a memory region in xe * device, such as HBM memory or CXL extension memory. */ -struct xe_mem_region { +struct xe_vram_region { /** @io_start: IO start address of this VRAM instance */ resource_size_t io_start; /** @@ -197,7 +197,7 @@ struct xe_tile { * Although VRAM is associated with a specific tile, it can * still be accessed by all tiles' GTs. */ - struct xe_mem_region vram; + struct xe_vram_region vram; /** @mem.vram_mgr: VRAM TTM manager */ struct xe_ttm_vram_mgr *vram_mgr; @@ -369,7 +369,7 @@ struct xe_device { /** @mem: memory info for device */ struct { /** @mem.vram: VRAM info for device */ - struct xe_mem_region vram; + struct xe_vram_region vram; /** @mem.sys_mgr: system TTM manager */ struct ttm_resource_manager sys_mgr; } mem; diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c index f4a16e5fa770..a8c37fb4fec0 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c @@ -340,7 +340,7 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr, int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr) { struct xe_device *xe = tile_to_xe(tile); - struct xe_mem_region *vram = &tile->mem.vram; + struct xe_vram_region *vram = &tile->mem.vram; mgr->vram = vram; return __xe_ttm_vram_mgr_init(xe, mgr, XE_PL_VRAM0 + tile->id, diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h b/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h index 2d75cf126289..4c52ced4ee44 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h @@ -9,7 +9,7 @@ #include #include -struct xe_mem_region; +struct xe_vram_region; /** * struct xe_ttm_vram_mgr - XE TTM VRAM manager @@ -22,7 +22,7 @@ struct xe_ttm_vram_mgr { /** @mm: DRM buddy allocator which manages the VRAM */ struct drm_buddy mm; /** @vram: ptr to details of associated VRAM region */ - struct xe_mem_region *vram; + struct xe_vram_region *vram; /** @visible_size: Proped size of the CPU visible portion */ u64 visible_size; /** @visible_avail: CPU visible portion still unallocated */ From 71163271dc227f8dd0d9446b8897a141c2eae957 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= Date: Mon, 10 Feb 2025 09:15:11 +0100 Subject: [PATCH 0698/1443] drm/xe: Move VRAM manager to struct xe_vram_region MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit VRAM manager is related directly to struct xe_vram_region so it should be inside this structure. Let's move the VRAM to struct xe_vram_region. v2: - remove xe_vram_region pointer from xe_ttm_vram_mgr - stop use dynamic alloaction for xe_ttm_vram_mgr in xe_vram_region - rename struct xe_ttm_vram_mgr vram_mgr to ttm v3: - fix "'ttm' not described in 'xe_vram_region'" Signed-off-by: Piotr Piórkowski Cc: Rodrigo Vivi Reviewed-by: Rodrigo Vivi Signed-off-by: Michal Wajdeczko Link: https://patchwork.freedesktop.org/patch/msgid/20250210081511.906452-3-piotr.piorkowski@intel.com --- drivers/gpu/drm/xe/xe_bo.c | 10 ++++++++-- drivers/gpu/drm/xe/xe_device_types.h | 6 +++--- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 2 +- drivers/gpu/drm/xe/xe_tile.c | 6 +----- drivers/gpu/drm/xe/xe_ttm_vram_mgr.c | 1 - drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h | 4 ---- 6 files changed, 13 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index a202d5525822..78d09c5ed26d 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -147,10 +147,13 @@ static struct xe_vram_region *res_to_mem_region(struct ttm_resource *res) { struct xe_device *xe = ttm_to_xe_device(res->bo->bdev); struct ttm_resource_manager *mgr; + struct xe_ttm_vram_mgr *vram_mgr; xe_assert(xe, resource_is_vram(res)); mgr = ttm_manager_type(&xe->ttm, res->mem_type); - return to_xe_ttm_vram_mgr(mgr)->vram; + vram_mgr = to_xe_ttm_vram_mgr(mgr); + + return container_of(vram_mgr, struct xe_vram_region, ttm); } static void try_add_system(struct xe_device *xe, struct xe_bo *bo, @@ -179,12 +182,15 @@ static void add_vram(struct xe_device *xe, struct xe_bo *bo, struct ttm_place *places, u32 bo_flags, u32 mem_type, u32 *c) { struct ttm_place place = { .mem_type = mem_type }; + struct ttm_resource_manager *mgr = ttm_manager_type(&xe->ttm, mem_type); + struct xe_ttm_vram_mgr *vram_mgr = to_xe_ttm_vram_mgr(mgr); + struct xe_vram_region *vram; u64 io_size; xe_assert(xe, *c < ARRAY_SIZE(bo->placements)); - vram = to_xe_ttm_vram_mgr(ttm_manager_type(&xe->ttm, mem_type))->vram; + vram = container_of(vram_mgr, struct xe_vram_region, ttm); xe_assert(xe, vram && vram->usable_size); io_size = vram->io_size; diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 1f27ed2c0448..28d10a1d7b64 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -23,6 +23,7 @@ #include "xe_sriov_types.h" #include "xe_step_types.h" #include "xe_survivability_mode_types.h" +#include "xe_ttm_vram_mgr_types.h" #if IS_ENABLED(CONFIG_DRM_XE_DEBUG) #define TEST_VM_OPS_ERROR @@ -105,6 +106,8 @@ struct xe_vram_region { resource_size_t actual_physical_size; /** @mapping: pointer to VRAM mappable space */ void __iomem *mapping; + /** @ttm: VRAM TTM manager */ + struct xe_ttm_vram_mgr ttm; }; /** @@ -199,9 +202,6 @@ struct xe_tile { */ struct xe_vram_region vram; - /** @mem.vram_mgr: VRAM TTM manager */ - struct xe_ttm_vram_mgr *vram_mgr; - /** @mem.ggtt: Global graphics translation table */ struct xe_ggtt *ggtt; diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 5c3e9e5bd051..88bd9d97ba5c 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -1564,7 +1564,7 @@ static u64 pf_query_free_lmem(struct xe_gt *gt) { struct xe_tile *tile = gt->tile; - return xe_ttm_vram_get_avail(&tile->mem.vram_mgr->manager); + return xe_ttm_vram_get_avail(&tile->mem.vram.ttm.manager); } static u64 pf_query_max_lmem(struct xe_gt *gt) diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c index 2825553b568f..d9a7a04ff652 100644 --- a/drivers/gpu/drm/xe/xe_tile.c +++ b/drivers/gpu/drm/xe/xe_tile.c @@ -94,10 +94,6 @@ static int xe_tile_alloc(struct xe_tile *tile) return -ENOMEM; tile->mem.ggtt->tile = tile; - tile->mem.vram_mgr = drmm_kzalloc(drm, sizeof(*tile->mem.vram_mgr), GFP_KERNEL); - if (!tile->mem.vram_mgr) - return -ENOMEM; - return 0; } @@ -139,7 +135,7 @@ static int tile_ttm_mgr_init(struct xe_tile *tile) int err; if (tile->mem.vram.usable_size) { - err = xe_ttm_vram_mgr_init(tile, tile->mem.vram_mgr); + err = xe_ttm_vram_mgr_init(tile, &tile->mem.vram.ttm); if (err) return err; xe->info.mem_region_mask |= BIT(tile->id) << 1; diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c index a8c37fb4fec0..9e375a40aee9 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c @@ -342,7 +342,6 @@ int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr) struct xe_device *xe = tile_to_xe(tile); struct xe_vram_region *vram = &tile->mem.vram; - mgr->vram = vram; return __xe_ttm_vram_mgr_init(xe, mgr, XE_PL_VRAM0 + tile->id, vram->usable_size, vram->io_size, PAGE_SIZE); diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h b/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h index 4c52ced4ee44..1144f9232ebb 100644 --- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h +++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr_types.h @@ -9,8 +9,6 @@ #include #include -struct xe_vram_region; - /** * struct xe_ttm_vram_mgr - XE TTM VRAM manager * @@ -21,8 +19,6 @@ struct xe_ttm_vram_mgr { struct ttm_resource_manager manager; /** @mm: DRM buddy allocator which manages the VRAM */ struct drm_buddy mm; - /** @vram: ptr to details of associated VRAM region */ - struct xe_vram_region *vram; /** @visible_size: Proped size of the CPU visible portion */ u64 visible_size; /** @visible_avail: CPU visible portion still unallocated */ From 0678c15b942770223ec339ed58c79303d1b991ec Mon Sep 17 00:00:00 2001 From: Zhaoyu Liu Date: Sun, 26 Jan 2025 17:32:56 +0800 Subject: [PATCH 0699/1443] drm/ttm: use ttm_resource_unevictable() to replace pin_count and swapped MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit TTM always uses pin_count and ttm_resource_is_swapped() together to determine whether a BO is unevictable. Now use ttm_resource_unevictable() to replace them. Signed-off-by: Zhaoyu Liu Reviewed-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20250126093256.GA688734@bytedance Signed-off-by: Christian König --- drivers/gpu/drm/ttm/ttm_resource.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c index 1644beb7a745..7e5a60c55813 100644 --- a/drivers/gpu/drm/ttm/ttm_resource.c +++ b/drivers/gpu/drm/ttm/ttm_resource.c @@ -269,11 +269,16 @@ static bool ttm_resource_is_swapped(struct ttm_resource *res, struct ttm_buffer_ return ttm_tt_is_swapped(bo->ttm); } +static bool ttm_resource_unevictable(struct ttm_resource *res, struct ttm_buffer_object *bo) +{ + return bo->pin_count || ttm_resource_is_swapped(res, bo); +} + /* Add the resource to a bulk move if the BO is configured for it */ void ttm_resource_add_bulk_move(struct ttm_resource *res, struct ttm_buffer_object *bo) { - if (bo->bulk_move && !bo->pin_count && !ttm_resource_is_swapped(res, bo)) + if (bo->bulk_move && !ttm_resource_unevictable(res, bo)) ttm_lru_bulk_move_add(bo->bulk_move, res); } @@ -281,7 +286,7 @@ void ttm_resource_add_bulk_move(struct ttm_resource *res, void ttm_resource_del_bulk_move(struct ttm_resource *res, struct ttm_buffer_object *bo) { - if (bo->bulk_move && !bo->pin_count && !ttm_resource_is_swapped(res, bo)) + if (bo->bulk_move && !ttm_resource_unevictable(res, bo)) ttm_lru_bulk_move_del(bo->bulk_move, res); } @@ -293,10 +298,10 @@ void ttm_resource_move_to_lru_tail(struct ttm_resource *res) lockdep_assert_held(&bo->bdev->lru_lock); - if (bo->pin_count || ttm_resource_is_swapped(res, bo)) { + if (ttm_resource_unevictable(res, bo)) { list_move_tail(&res->lru.link, &bdev->unevictable); - } else if (bo->bulk_move) { + } else if (bo->bulk_move) { struct ttm_lru_bulk_move_pos *pos = ttm_lru_bulk_move_pos(bo->bulk_move, res); @@ -335,7 +340,7 @@ void ttm_resource_init(struct ttm_buffer_object *bo, man = ttm_manager_type(bo->bdev, place->mem_type); spin_lock(&bo->bdev->lru_lock); - if (bo->pin_count || ttm_resource_is_swapped(res, bo)) + if (ttm_resource_unevictable(res, bo)) list_add_tail(&res->lru.link, &bo->bdev->unevictable); else list_add_tail(&res->lru.link, &man->lru[bo->priority]); From aaad45cfaa1bdb774e3755c93c0ab5ef6858917a Mon Sep 17 00:00:00 2001 From: Shixiong Ou Date: Thu, 6 Feb 2025 15:17:13 +0800 Subject: [PATCH 0700/1443] drm/bridge: convert to use devm_platform_ioremap_resource() convert to use devm_platform_ioremap_resource(). Signed-off-by: Shixiong Ou Reviewed-by: Maxime Ripard Acked-by: Manikandan Muralidharan Reviewed-and-tested-by: Dharma Balasubiramani Reviewed-by: Dharma Balasubiramani Tested-by: Dharma Balasubiramani Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206071713.180471-1-oushixiong1025@163.com --- drivers/gpu/drm/bridge/microchip-lvds.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/bridge/microchip-lvds.c b/drivers/gpu/drm/bridge/microchip-lvds.c index b8313dad6072..53dd140a1b8d 100644 --- a/drivers/gpu/drm/bridge/microchip-lvds.c +++ b/drivers/gpu/drm/bridge/microchip-lvds.c @@ -162,8 +162,7 @@ static int mchp_lvds_probe(struct platform_device *pdev) lvds->dev = dev; - lvds->regs = devm_ioremap_resource(lvds->dev, - platform_get_resource(pdev, IORESOURCE_MEM, 0)); + lvds->regs = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(lvds->regs)) return PTR_ERR(lvds->regs); From f74fd53ba34551b7626193fb70c17226f06e9bf1 Mon Sep 17 00:00:00 2001 From: Tejas Upadhyay Date: Wed, 5 Feb 2025 10:40:42 +0530 Subject: [PATCH 0701/1443] drm/xe/client: bo->client does not need bos_lock bos_lock is to protect list of bos used by client, it is not required to protect bo->client so bring it outside of bos_lock. Fixes: b27970f3e11c ("drm/xe: Add tracking support for bos per client") Signed-off-by: Tejas Upadhyay Reviewed-by: Himal Prasad Ghimiray Reviewed-by: Nirmoy Das Link: https://patchwork.freedesktop.org/patch/msgid/20250205051042.1991192-1-tejas.upadhyay@intel.com Signed-off-by: Nirmoy Das --- drivers/gpu/drm/xe/xe_drm_client.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_drm_client.c b/drivers/gpu/drm/xe/xe_drm_client.c index 63f30b6df70b..2d4874d2b922 100644 --- a/drivers/gpu/drm/xe/xe_drm_client.c +++ b/drivers/gpu/drm/xe/xe_drm_client.c @@ -135,8 +135,8 @@ void xe_drm_client_add_bo(struct xe_drm_client *client, XE_WARN_ON(bo->client); XE_WARN_ON(!list_empty(&bo->client_link)); - spin_lock(&client->bos_lock); bo->client = xe_drm_client_get(client); + spin_lock(&client->bos_lock); list_add_tail(&bo->client_link, &client->bos_list); spin_unlock(&client->bos_lock); } From 78ccf6a6bae11e451e20a52dd2bc2ab98f66326b Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Mon, 10 Feb 2025 11:19:53 +0800 Subject: [PATCH 0702/1443] ASoC: Intel: soc-acpi-intel-ptl-match: revise typo of rt712_vb + rt1320 support s/lnl/ptl Fixes: bd40d912728f ("ASoC: Intel: soc-acpi-intel-ptl-match: add rt712_vb + rt1320 support") Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Ranjani Sridharan Signed-off-by: Bard Liao Link: https://patch.msgid.link/20250210031954.6287-2-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/common/soc-acpi-intel-ptl-match.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c index 9eb4a43e3e7a..e487c4e1c034 100644 --- a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c +++ b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c @@ -270,7 +270,7 @@ static const struct snd_soc_acpi_link_adr lnl_sdw_rt713_vb_l2_rt1320_l13[] = { {} }; -static const struct snd_soc_acpi_link_adr lnl_sdw_rt712_vb_l2_rt1320_l1[] = { +static const struct snd_soc_acpi_link_adr ptl_sdw_rt712_vb_l2_rt1320_l1[] = { { .mask = BIT(2), .num_adr = ARRAY_SIZE(rt712_vb_2_group1_adr), @@ -337,10 +337,10 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_ptl_sdw_machines[] = { }, { .link_mask = BIT(1) | BIT(2), - .links = lnl_sdw_rt712_vb_l2_rt1320_l1, + .links = ptl_sdw_rt712_vb_l2_rt1320_l1, .drv_name = "sof_sdw", .machine_check = snd_soc_acpi_intel_sdca_is_device_rt712_vb, - .sof_tplg_filename = "sof-lnl-rt712-l2-rt1320-l1.tplg" + .sof_tplg_filename = "sof-ptl-rt712-l2-rt1320-l1.tplg" }, { .link_mask = BIT(1) | BIT(2) | BIT(3), From cb78b8dc7834066539253c039f276b3625fecd9f Mon Sep 17 00:00:00 2001 From: Peter Ujfalusi Date: Mon, 10 Feb 2025 11:19:54 +0800 Subject: [PATCH 0703/1443] ASoC: Intel: soc-acpi-intel-ptl-match: revise typo of rt713_vb_l2_rt1320_l13 s/lnl/ptl Fixes: a7ebb0255188 ("ASoC: Intel: soc-acpi-intel-ptl-match: add rt713_vb_l2_rt1320_l13 support") Signed-off-by: Peter Ujfalusi Reviewed-by: Kai Vehmanen Reviewed-by: Ranjani Sridharan Signed-off-by: Bard Liao Link: https://patch.msgid.link/20250210031954.6287-3-yung-chuan.liao@linux.intel.com Signed-off-by: Mark Brown --- sound/soc/intel/common/soc-acpi-intel-ptl-match.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c index e487c4e1c034..dd7993b76dee 100644 --- a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c +++ b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c @@ -251,7 +251,7 @@ static const struct snd_soc_acpi_link_adr ptl_rvp[] = { {} }; -static const struct snd_soc_acpi_link_adr lnl_sdw_rt713_vb_l2_rt1320_l13[] = { +static const struct snd_soc_acpi_link_adr ptl_sdw_rt713_vb_l2_rt1320_l13[] = { { .mask = BIT(2), .num_adr = ARRAY_SIZE(rt713_vb_2_adr), @@ -344,10 +344,10 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_ptl_sdw_machines[] = { }, { .link_mask = BIT(1) | BIT(2) | BIT(3), - .links = lnl_sdw_rt713_vb_l2_rt1320_l13, + .links = ptl_sdw_rt713_vb_l2_rt1320_l13, .drv_name = "sof_sdw", .machine_check = snd_soc_acpi_intel_sdca_is_device_rt712_vb, - .sof_tplg_filename = "sof-lnl-rt713-l2-rt1320-l13.tplg" + .sof_tplg_filename = "sof-ptl-rt713-l2-rt1320-l13.tplg" }, {}, }; From 91b98d5a6e8067c5226207487681a48f0d651e46 Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Fri, 7 Feb 2025 13:46:02 +0200 Subject: [PATCH 0704/1443] ASoC: SOF: amd: Add post_fw_run_delay ACP quirk Stress testing resume from suspend on Valve Steam Deck OLED (Galileo) revealed that the DSP firmware could enter an unrecoverable faulty state, where the kernel ring buffer is flooded with IPC related error messages: [ +0.017002] snd_sof_amd_vangogh 0000:04:00.5: acp_sof_ipc_send_msg: Failed to acquire HW lock [ +0.000054] snd_sof_amd_vangogh 0000:04:00.5: ipc3_tx_msg_unlocked: ipc message send for 0x30100000 failed: -22 [ +0.000005] snd_sof_amd_vangogh 0000:04:00.5: Failed to setup widget PIPELINE.6.ACPHS1.IN [ +0.000004] snd_sof_amd_vangogh 0000:04:00.5: Failed to restore pipeline after resume -22 [ +0.000003] snd_sof_amd_vangogh 0000:04:00.5: PM: dpm_run_callback(): pci_pm_resume returns -22 [ +0.000009] snd_sof_amd_vangogh 0000:04:00.5: PM: failed to resume async: error -22 [...] [ +0.002582] PM: suspend exit [ +0.065085] snd_sof_amd_vangogh 0000:04:00.5: ipc tx error for 0x30130000 (msg/reply size: 12/0): -22 [ +0.000499] snd_sof_amd_vangogh 0000:04:00.5: error: failed widget list set up for pcm 1 dir 0 [ +0.000011] snd_sof_amd_vangogh 0000:04:00.5: error: set pcm hw_params after resume [ +0.000006] snd_sof_amd_vangogh 0000:04:00.5: ASoC: error at snd_soc_pcm_component_prepare on 0000:04:00.5: -22 [...] A system reboot would be necessary to restore the speakers functionality. However, by delaying a bit any host to DSP transmission right after the firmware boot completed, the issue could not be reproduced anymore and sound continued to work flawlessly even after performing thousands of suspend/resume cycles. Introduce the post_fw_run_delay ACP quirk to allow providing the aforementioned delay via the snd_sof_dsp_ops->post_fw_run() callback for the affected devices. Signed-off-by: Cristian Ciocaltea Link: https://patch.msgid.link/20250207-sof-vangogh-fixes-v1-1-67824c1e4c9a@collabora.com Signed-off-by: Mark Brown --- sound/soc/sof/amd/acp.c | 1 + sound/soc/sof/amd/acp.h | 1 + sound/soc/sof/amd/vangogh.c | 18 ++++++++++++++++++ 3 files changed, 20 insertions(+) diff --git a/sound/soc/sof/amd/acp.c b/sound/soc/sof/amd/acp.c index 33648ff8b833..9e13c96528be 100644 --- a/sound/soc/sof/amd/acp.c +++ b/sound/soc/sof/amd/acp.c @@ -27,6 +27,7 @@ MODULE_PARM_DESC(enable_fw_debug, "Enable Firmware debug"); static struct acp_quirk_entry quirk_valve_galileo = { .signed_fw_image = true, .skip_iram_dram_size_mod = true, + .post_fw_run_delay = true, }; const struct dmi_system_id acp_sof_quirk_table[] = { diff --git a/sound/soc/sof/amd/acp.h b/sound/soc/sof/amd/acp.h index 800594440f73..2a19d82d6200 100644 --- a/sound/soc/sof/amd/acp.h +++ b/sound/soc/sof/amd/acp.h @@ -220,6 +220,7 @@ struct sof_amd_acp_desc { struct acp_quirk_entry { bool signed_fw_image; bool skip_iram_dram_size_mod; + bool post_fw_run_delay; }; /* Common device data struct for ACP devices */ diff --git a/sound/soc/sof/amd/vangogh.c b/sound/soc/sof/amd/vangogh.c index 8e2672106ac6..d5f1dddd43e7 100644 --- a/sound/soc/sof/amd/vangogh.c +++ b/sound/soc/sof/amd/vangogh.c @@ -11,6 +11,7 @@ * Hardware interface for Audio DSP on Vangogh platform */ +#include #include #include @@ -136,6 +137,20 @@ static struct snd_soc_dai_driver vangogh_sof_dai[] = { }, }; +static int sof_vangogh_post_fw_run_delay(struct snd_sof_dev *sdev) +{ + /* + * Resuming from suspend in some cases my cause the DSP firmware + * to enter an unrecoverable faulty state. Delaying a bit any host + * to DSP transmission right after firmware boot completion seems + * to resolve the issue. + */ + if (!sdev->first_boot) + usleep_range(100, 150); + + return 0; +} + /* Vangogh ops */ struct snd_sof_dsp_ops sof_vangogh_ops; EXPORT_SYMBOL_NS(sof_vangogh_ops, "SND_SOC_SOF_AMD_COMMON"); @@ -157,6 +172,9 @@ int sof_vangogh_ops_init(struct snd_sof_dev *sdev) if (quirks->signed_fw_image) sof_vangogh_ops.load_firmware = acp_sof_load_signed_firmware; + + if (quirks->post_fw_run_delay) + sof_vangogh_ops.post_fw_run = sof_vangogh_post_fw_run_delay; } return 0; From 2ecbc2e9f3b19e2199e8bc3ba603d299f1985f09 Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Fri, 7 Feb 2025 13:46:03 +0200 Subject: [PATCH 0705/1443] ASoC: SOF: amd: Drop unused includes from Vangogh driver Remove all the includes for headers which are not (directly) used from the Vangogh SOF driver sources. Signed-off-by: Cristian Ciocaltea Reviewed-by: Venkata Prasad Potturu Link: https://patch.msgid.link/20250207-sof-vangogh-fixes-v1-2-67824c1e4c9a@collabora.com Signed-off-by: Mark Brown --- sound/soc/sof/amd/pci-vangogh.c | 2 -- sound/soc/sof/amd/vangogh.c | 4 ---- 2 files changed, 6 deletions(-) diff --git a/sound/soc/sof/amd/pci-vangogh.c b/sound/soc/sof/amd/pci-vangogh.c index 53f64d6bc91b..28f2d4050a67 100644 --- a/sound/soc/sof/amd/pci-vangogh.c +++ b/sound/soc/sof/amd/pci-vangogh.c @@ -13,11 +13,9 @@ #include #include -#include #include #include -#include "../ops.h" #include "../sof-pci-dev.h" #include "../../amd/mach-config.h" #include "acp.h" diff --git a/sound/soc/sof/amd/vangogh.c b/sound/soc/sof/amd/vangogh.c index d5f1dddd43e7..6ed5f9aaa414 100644 --- a/sound/soc/sof/amd/vangogh.c +++ b/sound/soc/sof/amd/vangogh.c @@ -12,13 +12,9 @@ */ #include -#include #include -#include "../ops.h" -#include "../sof-audio.h" #include "acp.h" -#include "acp-dsp-offset.h" #define I2S_HS_INSTANCE 0 #define I2S_BT_INSTANCE 1 From ac84ca815adb4171a4276b1d44096b75f6a150b7 Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Fri, 7 Feb 2025 13:46:04 +0200 Subject: [PATCH 0706/1443] ASoC: SOF: amd: Handle IPC replies before FW_BOOT_COMPLETE In some cases, e.g. during resuming from suspend, there is a possibility that some IPC reply messages get received by the host while the DSP firmware has not yet reached the complete boot state. Detect when this happens and do not attempt to process the unexpected replies from DSP. Instead, provide proper debugging support. Signed-off-by: Cristian Ciocaltea Link: https://patch.msgid.link/20250207-sof-vangogh-fixes-v1-3-67824c1e4c9a@collabora.com Signed-off-by: Mark Brown --- sound/soc/sof/amd/acp-ipc.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/sound/soc/sof/amd/acp-ipc.c b/sound/soc/sof/amd/acp-ipc.c index 5f371d9263f3..12caefd08788 100644 --- a/sound/soc/sof/amd/acp-ipc.c +++ b/sound/soc/sof/amd/acp-ipc.c @@ -167,6 +167,7 @@ irqreturn_t acp_sof_ipc_irq_thread(int irq, void *context) if (sdev->first_boot && sdev->fw_state != SOF_FW_BOOT_COMPLETE) { acp_mailbox_read(sdev, sdev->dsp_box.offset, &status, sizeof(status)); + if ((status & SOF_IPC_PANIC_MAGIC_MASK) == SOF_IPC_PANIC_MAGIC) { snd_sof_dsp_panic(sdev, sdev->dsp_box.offset + sizeof(status), true); @@ -188,13 +189,21 @@ irqreturn_t acp_sof_ipc_irq_thread(int irq, void *context) dsp_ack = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SCRATCH_REG_0 + dsp_ack_write); if (dsp_ack) { - spin_lock_irq(&sdev->ipc_lock); - /* handle immediate reply from DSP core */ - acp_dsp_ipc_get_reply(sdev); - snd_sof_ipc_reply(sdev, 0); - /* set the done bit */ - acp_dsp_ipc_dsp_done(sdev); - spin_unlock_irq(&sdev->ipc_lock); + if (likely(sdev->fw_state == SOF_FW_BOOT_COMPLETE)) { + spin_lock_irq(&sdev->ipc_lock); + + /* handle immediate reply from DSP core */ + acp_dsp_ipc_get_reply(sdev); + snd_sof_ipc_reply(sdev, 0); + /* set the done bit */ + acp_dsp_ipc_dsp_done(sdev); + + spin_unlock_irq(&sdev->ipc_lock); + } else { + dev_dbg_ratelimited(sdev->dev, "IPC reply before FW_BOOT_COMPLETE: %#x\n", + dsp_ack); + } + ipc_irq = true; } From ccc8480d90e8cb60f06bd90e227f34784927e19f Mon Sep 17 00:00:00 2001 From: Cristian Ciocaltea Date: Fri, 7 Feb 2025 13:46:05 +0200 Subject: [PATCH 0707/1443] ASoC: SOF: amd: Add branch prediction hint in ACP IRQ handler The conditional involving sdev->first_boot in acp_sof_ipc_irq_thread() will succeed only once, i.e. during the very first run of the DSP firmware. Use the unlikely() annotation to help improve branch prediction accuracy. Signed-off-by: Cristian Ciocaltea Reviewed-by: Venkata Prasad Potturu Link: https://patch.msgid.link/20250207-sof-vangogh-fixes-v1-4-67824c1e4c9a@collabora.com Signed-off-by: Mark Brown --- sound/soc/sof/amd/acp-ipc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/sof/amd/acp-ipc.c b/sound/soc/sof/amd/acp-ipc.c index 12caefd08788..22d4b807e1bb 100644 --- a/sound/soc/sof/amd/acp-ipc.c +++ b/sound/soc/sof/amd/acp-ipc.c @@ -165,7 +165,7 @@ irqreturn_t acp_sof_ipc_irq_thread(int irq, void *context) int dsp_msg, dsp_ack; unsigned int status; - if (sdev->first_boot && sdev->fw_state != SOF_FW_BOOT_COMPLETE) { + if (unlikely(sdev->first_boot && sdev->fw_state != SOF_FW_BOOT_COMPLETE)) { acp_mailbox_read(sdev, sdev->dsp_box.offset, &status, sizeof(status)); if ((status & SOF_IPC_PANIC_MAGIC_MASK) == SOF_IPC_PANIC_MAGIC) { From 9759ae2cee7cd42b95f1c48aa3749bd02b5ddb08 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Fri, 17 Jan 2025 13:58:00 +0800 Subject: [PATCH 0708/1443] iommu: Fix potential memory leak in iopf_queue_remove_device() The iopf_queue_remove_device() helper removes a device from the per-iommu iopf queue when PRI is disabled on the device. It responds to all outstanding iopf's with an IOMMU_PAGE_RESP_INVALID code and detaches the device from the queue. However, it fails to release the group structure that represents a group of iopf's awaiting for a response after responding to the hardware. This can cause a memory leak if iopf_queue_remove_device() is called with pending iopf's. Fix it by calling iopf_free_group() after the iopf group is responded. Fixes: 199112327135 ("iommu: Track iopf group instead of last fault") Cc: stable@vger.kernel.org Suggested-by: Kevin Tian Signed-off-by: Lu Baolu Reviewed-by: Kevin Tian Reviewed-by: Jason Gunthorpe Link: https://lore.kernel.org/r/20250117055800.782462-1-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel --- drivers/iommu/io-pgfault.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/iommu/io-pgfault.c b/drivers/iommu/io-pgfault.c index 4674e618797c..8b5926c1452e 100644 --- a/drivers/iommu/io-pgfault.c +++ b/drivers/iommu/io-pgfault.c @@ -478,6 +478,7 @@ void iopf_queue_remove_device(struct iopf_queue *queue, struct device *dev) ops->page_response(dev, iopf, &resp); list_del_init(&group->pending_node); + iopf_free_group(group); } mutex_unlock(&fault_param->lock); From 1046cac109225eda0973b898e053aeb3d6c10e1d Mon Sep 17 00:00:00 2001 From: Sybil Isabel Dorsett Date: Mon, 3 Feb 2025 16:33:15 +0000 Subject: [PATCH 0709/1443] platform/x86: thinkpad_acpi: Fix invalid fan speed on ThinkPad X120e MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On ThinkPad X120e, fan speed is reported in ticks per revolution rather than RPM. Recalculate the fan speed value reported for ThinkPad X120e to RPM based on a 22.5 kHz clock. Based on the information on https://www.thinkwiki.org/wiki/How_to_control_fan_speed, the same problem is highly likely to be relevant to at least Edge11, but Edge11 is not addressed in this patch. Signed-off-by: Sybil Isabel Dorsett Reviewed-by: Ilpo Järvinen Link: https://lore.kernel.org/r/20250203163255.5525-1-sybdorsett@proton.me Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/thinkpad_acpi.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 1fcb0f99695a..e7778ea41478 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -7885,6 +7885,7 @@ static struct ibm_struct volume_driver_data = { #define FAN_NS_CTRL_STATUS BIT(2) /* Bit which determines control is enabled or not */ #define FAN_NS_CTRL BIT(4) /* Bit which determines control is by host or EC */ +#define FAN_CLOCK_TPM (22500*60) /* Ticks per minute for a 22.5 kHz clock */ enum { /* Fan control constants */ fan_status_offset = 0x2f, /* EC register 0x2f */ @@ -7940,6 +7941,7 @@ static int fan_watchdog_maxinterval; static bool fan_with_ns_addr; static bool ecfw_with_fan_dec_rpm; +static bool fan_speed_in_tpr; static struct mutex fan_mutex; @@ -8142,8 +8144,11 @@ static int fan_get_speed(unsigned int *speed) !acpi_ec_read(fan_rpm_offset + 1, &hi))) return -EIO; - if (likely(speed)) + if (likely(speed)) { *speed = (hi << 8) | lo; + if (fan_speed_in_tpr && *speed != 0) + *speed = FAN_CLOCK_TPM / *speed; + } break; case TPACPI_FAN_RD_TPEC_NS: if (!acpi_ec_read(fan_rpm_status_ns, &lo)) @@ -8176,8 +8181,11 @@ static int fan2_get_speed(unsigned int *speed) if (rc) return -EIO; - if (likely(speed)) + if (likely(speed)) { *speed = (hi << 8) | lo; + if (fan_speed_in_tpr && *speed != 0) + *speed = FAN_CLOCK_TPM / *speed; + } break; case TPACPI_FAN_RD_TPEC_NS: @@ -8788,6 +8796,7 @@ static const struct attribute_group fan_driver_attr_group = { #define TPACPI_FAN_NOFAN 0x0008 /* no fan available */ #define TPACPI_FAN_NS 0x0010 /* For EC with non-Standard register addresses */ #define TPACPI_FAN_DECRPM 0x0020 /* For ECFW's with RPM in register as decimal */ +#define TPACPI_FAN_TPR 0x0040 /* Fan speed is in Ticks Per Revolution */ static const struct tpacpi_quirk fan_quirk_table[] __initconst = { TPACPI_QEC_IBM('1', 'Y', TPACPI_FAN_Q1), @@ -8817,6 +8826,7 @@ static const struct tpacpi_quirk fan_quirk_table[] __initconst = { TPACPI_Q_LNV3('R', '0', 'V', TPACPI_FAN_NS), /* 11e Gen5 KL-Y */ TPACPI_Q_LNV3('N', '1', 'O', TPACPI_FAN_NOFAN), /* X1 Tablet (2nd gen) */ TPACPI_Q_LNV3('R', '0', 'Q', TPACPI_FAN_DECRPM),/* L480 */ + TPACPI_Q_LNV('8', 'F', TPACPI_FAN_TPR), /* ThinkPad x120e */ }; static int __init fan_init(struct ibm_init_struct *iibm) @@ -8887,6 +8897,8 @@ static int __init fan_init(struct ibm_init_struct *iibm) if (quirks & TPACPI_FAN_Q1) fan_quirk1_setup(); + if (quirks & TPACPI_FAN_TPR) + fan_speed_in_tpr = true; /* Try and probe the 2nd fan */ tp_features.second_fan = 1; /* needed for get_speed to work */ res = fan2_get_speed(&speed); From 9cff907cbf8c7fb5345918dbcc7b74a01656f34f Mon Sep 17 00:00:00 2001 From: Mark Pearson Date: Thu, 6 Feb 2025 14:39:41 -0500 Subject: [PATCH 0710/1443] platform/x86: thinkpad_acpi: Support for V9 DYTC platform profiles MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Newer Thinkpad AMD platforms are using V9 DYTC and this changes the profiles used for PSC mode. Add support for this update. Tested on P14s G5 AMD Signed-off-by: Mark Pearson Link: https://lore.kernel.org/r/20250206193953.58365-1-mpearson-lenovo@squebb.ca Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/thinkpad_acpi.c | 34 ++++++++++++++++++---------- 1 file changed, 22 insertions(+), 12 deletions(-) diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index e7778ea41478..99cdea723d32 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -10331,6 +10331,10 @@ static struct ibm_struct proxsensor_driver_data = { #define DYTC_MODE_PSC_BALANCE 5 /* Default mode aka balanced */ #define DYTC_MODE_PSC_PERFORM 7 /* High power mode aka performance */ +#define DYTC_MODE_PSCV9_LOWPOWER 1 /* Low power mode */ +#define DYTC_MODE_PSCV9_BALANCE 3 /* Default mode aka balanced */ +#define DYTC_MODE_PSCV9_PERFORM 4 /* High power mode aka performance */ + #define DYTC_ERR_MASK 0xF /* Bits 0-3 in cmd result are the error result */ #define DYTC_ERR_SUCCESS 1 /* CMD completed successful */ @@ -10351,6 +10355,10 @@ static int dytc_capabilities; static bool dytc_mmc_get_available; static int profile_force; +static int platform_psc_profile_lowpower = DYTC_MODE_PSC_LOWPOWER; +static int platform_psc_profile_balanced = DYTC_MODE_PSC_BALANCE; +static int platform_psc_profile_performance = DYTC_MODE_PSC_PERFORM; + static int convert_dytc_to_profile(int funcmode, int dytcmode, enum platform_profile_option *profile) { @@ -10372,19 +10380,15 @@ static int convert_dytc_to_profile(int funcmode, int dytcmode, } return 0; case DYTC_FUNCTION_PSC: - switch (dytcmode) { - case DYTC_MODE_PSC_LOWPOWER: + if (dytcmode == platform_psc_profile_lowpower) *profile = PLATFORM_PROFILE_LOW_POWER; - break; - case DYTC_MODE_PSC_BALANCE: + else if (dytcmode == platform_psc_profile_balanced) *profile = PLATFORM_PROFILE_BALANCED; - break; - case DYTC_MODE_PSC_PERFORM: + else if (dytcmode == platform_psc_profile_performance) *profile = PLATFORM_PROFILE_PERFORMANCE; - break; - default: /* Unknown mode */ + else return -EINVAL; - } + return 0; case DYTC_FUNCTION_AMT: /* For now return balanced. It's the closest we have to 'auto' */ @@ -10405,19 +10409,19 @@ static int convert_profile_to_dytc(enum platform_profile_option profile, int *pe if (dytc_capabilities & BIT(DYTC_FC_MMC)) *perfmode = DYTC_MODE_MMC_LOWPOWER; else if (dytc_capabilities & BIT(DYTC_FC_PSC)) - *perfmode = DYTC_MODE_PSC_LOWPOWER; + *perfmode = platform_psc_profile_lowpower; break; case PLATFORM_PROFILE_BALANCED: if (dytc_capabilities & BIT(DYTC_FC_MMC)) *perfmode = DYTC_MODE_MMC_BALANCE; else if (dytc_capabilities & BIT(DYTC_FC_PSC)) - *perfmode = DYTC_MODE_PSC_BALANCE; + *perfmode = platform_psc_profile_balanced; break; case PLATFORM_PROFILE_PERFORMANCE: if (dytc_capabilities & BIT(DYTC_FC_MMC)) *perfmode = DYTC_MODE_MMC_PERFORM; else if (dytc_capabilities & BIT(DYTC_FC_PSC)) - *perfmode = DYTC_MODE_PSC_PERFORM; + *perfmode = platform_psc_profile_performance; break; default: /* Unknown profile */ return -EOPNOTSUPP; @@ -10611,6 +10615,7 @@ static int tpacpi_dytc_profile_init(struct ibm_init_struct *iibm) if (output & BIT(DYTC_QUERY_ENABLE_BIT)) dytc_version = (output >> DYTC_QUERY_REV_BIT) & 0xF; + dbg_printk(TPACPI_DBG_INIT, "DYTC version %d\n", dytc_version); /* Check DYTC is enabled and supports mode setting */ if (dytc_version < 5) return -ENODEV; @@ -10649,6 +10654,11 @@ static int tpacpi_dytc_profile_init(struct ibm_init_struct *iibm) } } else if (dytc_capabilities & BIT(DYTC_FC_PSC)) { /* PSC MODE */ pr_debug("PSC is supported\n"); + if (dytc_version >= 9) { /* update profiles for DYTC 9 and up */ + platform_psc_profile_lowpower = DYTC_MODE_PSCV9_LOWPOWER; + platform_psc_profile_balanced = DYTC_MODE_PSCV9_BALANCE; + platform_psc_profile_performance = DYTC_MODE_PSCV9_PERFORM; + } } else { dbg_printk(TPACPI_DBG_INIT, "No DYTC support available\n"); return -ENODEV; From 2b9df00cded911e2ca2cfae5c45082166b24f8aa Mon Sep 17 00:00:00 2001 From: Niravkumar L Rabara Date: Mon, 10 Feb 2025 13:35:49 +0800 Subject: [PATCH 0711/1443] mtd: rawnand: cadence: fix error code in cadence_nand_init() Replace dma_request_channel() with dma_request_chan_by_mask() and use helper functions to return proper error code instead of fixed -EBUSY. Fixes: ec4ba01e894d ("mtd: rawnand: Add new Cadence NAND driver to MTD subsystem") Cc: stable@vger.kernel.org Signed-off-by: Niravkumar L Rabara Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/cadence-nand-controller.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/mtd/nand/raw/cadence-nand-controller.c b/drivers/mtd/nand/raw/cadence-nand-controller.c index 8d1d710e439d..fb5f671bdb7b 100644 --- a/drivers/mtd/nand/raw/cadence-nand-controller.c +++ b/drivers/mtd/nand/raw/cadence-nand-controller.c @@ -2904,11 +2904,10 @@ static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) dma_cap_set(DMA_MEMCPY, mask); if (cdns_ctrl->caps1->has_dma) { - cdns_ctrl->dmac = dma_request_channel(mask, NULL, NULL); - if (!cdns_ctrl->dmac) { - dev_err(cdns_ctrl->dev, - "Unable to get a DMA channel\n"); - ret = -EBUSY; + cdns_ctrl->dmac = dma_request_chan_by_mask(&mask); + if (IS_ERR(cdns_ctrl->dmac)) { + ret = dev_err_probe(cdns_ctrl->dev, PTR_ERR(cdns_ctrl->dmac), + "%d: Failed to get a DMA channel\n", ret); goto disable_irq; } } From d76d22b5096c5b05208fd982b153b3f182350b19 Mon Sep 17 00:00:00 2001 From: Niravkumar L Rabara Date: Mon, 10 Feb 2025 13:35:50 +0800 Subject: [PATCH 0712/1443] mtd: rawnand: cadence: use dma_map_resource for sdma address Remap the slave DMA I/O resources to enhance driver portability. Using a physical address causes DMA translation failure when the ARM SMMU is enabled. Fixes: ec4ba01e894d ("mtd: rawnand: Add new Cadence NAND driver to MTD subsystem") Cc: stable@vger.kernel.org Signed-off-by: Niravkumar L Rabara Signed-off-by: Miquel Raynal --- .../mtd/nand/raw/cadence-nand-controller.c | 29 ++++++++++++++++--- 1 file changed, 25 insertions(+), 4 deletions(-) diff --git a/drivers/mtd/nand/raw/cadence-nand-controller.c b/drivers/mtd/nand/raw/cadence-nand-controller.c index fb5f671bdb7b..47950a0ac6d2 100644 --- a/drivers/mtd/nand/raw/cadence-nand-controller.c +++ b/drivers/mtd/nand/raw/cadence-nand-controller.c @@ -471,6 +471,8 @@ struct cdns_nand_ctrl { struct { void __iomem *virt; dma_addr_t dma; + dma_addr_t iova_dma; + u32 size; } io; int irq; @@ -1835,11 +1837,11 @@ static int cadence_nand_slave_dma_transfer(struct cdns_nand_ctrl *cdns_ctrl, } if (dir == DMA_FROM_DEVICE) { - src_dma = cdns_ctrl->io.dma; + src_dma = cdns_ctrl->io.iova_dma; dst_dma = buf_dma; } else { src_dma = buf_dma; - dst_dma = cdns_ctrl->io.dma; + dst_dma = cdns_ctrl->io.iova_dma; } tx = dmaengine_prep_dma_memcpy(cdns_ctrl->dmac, dst_dma, src_dma, len, @@ -2869,6 +2871,7 @@ cadence_nand_irq_cleanup(int irqnum, struct cdns_nand_ctrl *cdns_ctrl) static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) { dma_cap_mask_t mask; + struct dma_device *dma_dev = cdns_ctrl->dmac->device; int ret; cdns_ctrl->cdma_desc = dma_alloc_coherent(cdns_ctrl->dev, @@ -2912,6 +2915,16 @@ static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) } } + cdns_ctrl->io.iova_dma = dma_map_resource(dma_dev->dev, cdns_ctrl->io.dma, + cdns_ctrl->io.size, + DMA_BIDIRECTIONAL, 0); + + ret = dma_mapping_error(dma_dev->dev, cdns_ctrl->io.iova_dma); + if (ret) { + dev_err(cdns_ctrl->dev, "Failed to map I/O resource to DMA\n"); + goto dma_release_chnl; + } + nand_controller_init(&cdns_ctrl->controller); INIT_LIST_HEAD(&cdns_ctrl->chips); @@ -2922,18 +2935,22 @@ static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) if (ret) { dev_err(cdns_ctrl->dev, "Failed to register MTD: %d\n", ret); - goto dma_release_chnl; + goto unmap_dma_resource; } kfree(cdns_ctrl->buf); cdns_ctrl->buf = kzalloc(cdns_ctrl->buf_size, GFP_KERNEL); if (!cdns_ctrl->buf) { ret = -ENOMEM; - goto dma_release_chnl; + goto unmap_dma_resource; } return 0; +unmap_dma_resource: + dma_unmap_resource(dma_dev->dev, cdns_ctrl->io.iova_dma, + cdns_ctrl->io.size, DMA_BIDIRECTIONAL, 0); + dma_release_chnl: if (cdns_ctrl->dmac) dma_release_channel(cdns_ctrl->dmac); @@ -2955,6 +2972,8 @@ static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) static void cadence_nand_remove(struct cdns_nand_ctrl *cdns_ctrl) { cadence_nand_chips_cleanup(cdns_ctrl); + dma_unmap_resource(cdns_ctrl->dmac->device->dev, cdns_ctrl->io.iova_dma, + cdns_ctrl->io.size, DMA_BIDIRECTIONAL, 0); cadence_nand_irq_cleanup(cdns_ctrl->irq, cdns_ctrl); kfree(cdns_ctrl->buf); dma_free_coherent(cdns_ctrl->dev, sizeof(struct cadence_nand_cdma_desc), @@ -3019,7 +3038,9 @@ static int cadence_nand_dt_probe(struct platform_device *ofdev) cdns_ctrl->io.virt = devm_platform_get_and_ioremap_resource(ofdev, 1, &res); if (IS_ERR(cdns_ctrl->io.virt)) return PTR_ERR(cdns_ctrl->io.virt); + cdns_ctrl->io.dma = res->start; + cdns_ctrl->io.size = resource_size(res); dt->clk = devm_clk_get(cdns_ctrl->dev, "nf_clk"); if (IS_ERR(dt->clk)) From f37d135b42cb484bdecee93f56b9f483214ede78 Mon Sep 17 00:00:00 2001 From: Niravkumar L Rabara Date: Mon, 10 Feb 2025 13:35:51 +0800 Subject: [PATCH 0713/1443] mtd: rawnand: cadence: fix incorrect device in dma_unmap_single dma_map_single is using physical/bus device (DMA) but dma_unmap_single is using framework device(NAND controller), which is incorrect. Fixed dma_unmap_single to use correct physical/bus device. Fixes: ec4ba01e894d ("mtd: rawnand: Add new Cadence NAND driver to MTD subsystem") Cc: stable@vger.kernel.org Signed-off-by: Niravkumar L Rabara Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/cadence-nand-controller.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/nand/raw/cadence-nand-controller.c b/drivers/mtd/nand/raw/cadence-nand-controller.c index 47950a0ac6d2..0b2db4173e72 100644 --- a/drivers/mtd/nand/raw/cadence-nand-controller.c +++ b/drivers/mtd/nand/raw/cadence-nand-controller.c @@ -1863,12 +1863,12 @@ static int cadence_nand_slave_dma_transfer(struct cdns_nand_ctrl *cdns_ctrl, dma_async_issue_pending(cdns_ctrl->dmac); wait_for_completion(&finished); - dma_unmap_single(cdns_ctrl->dev, buf_dma, len, dir); + dma_unmap_single(dma_dev->dev, buf_dma, len, dir); return 0; err_unmap: - dma_unmap_single(cdns_ctrl->dev, buf_dma, len, dir); + dma_unmap_single(dma_dev->dev, buf_dma, len, dir); err: dev_dbg(cdns_ctrl->dev, "Fall back to CPU I/O\n"); From a8505237066d054cd6ad63b52e7219ce41731753 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Sat, 8 Feb 2025 18:53:24 +0800 Subject: [PATCH 0714/1443] dt-bindings: display: panel: Add KD110N11-51IE and 2082109QFH040022-50E Add a new compatible for the panels KINGDISPLAY KD110N11-51IE and STARRY 2082109QFH040022-50E. Both panels use the HX83102 IC, so add the compatible to the hx83102 bindings file. Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Reviewed-by: Krzysztof Kozlowski Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250208105326.3850358-2-yelangyan@huaqin.corp-partner.google.com --- .../devicetree/bindings/display/panel/himax,hx83102.yaml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml b/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml index d876269e1fac..e4c1aa5deab9 100644 --- a/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml +++ b/Documentation/devicetree/bindings/display/panel/himax,hx83102.yaml @@ -22,6 +22,10 @@ properties: - csot,pna957qt1-1 # IVO t109nw41 11.0" WUXGA TFT LCD panel - ivo,t109nw41 + # KINGDISPLAY KD110N11-51IE 10.95" WUXGA TFT LCD panel + - kingdisplay,kd110n11-51ie + # STARRY 2082109QFH040022-50E 10.95" WUXGA TFT LCD panel + - starry,2082109qfh040022-50e # STARRY himax83102-j02 10.51" WUXGA TFT LCD panel - starry,himax83102-j02 - const: himax,hx83102 From 15d174fa85d936f3d0250b111e7d3b7f26a38b40 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Sat, 8 Feb 2025 18:53:25 +0800 Subject: [PATCH 0715/1443] drm/panel: panel-himax-hx83102: support for kingdisplay-kd110n11-51ie MIPI-DSI panel The kingdisplay-kd110n11-51ie is a 10.95" TFT panel. The MIPI controller on this panel is the same as the other panels here, so add this panel to this driver. Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250208105326.3850358-3-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-himax-hx83102.c | 115 ++++++++++++++++++++ 1 file changed, 115 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-himax-hx83102.c b/drivers/gpu/drm/panel/panel-himax-hx83102.c index c3f4f5ee175d..9a818dea653f 100644 --- a/drivers/gpu/drm/panel/panel-himax-hx83102.c +++ b/drivers/gpu/drm/panel/panel-himax-hx83102.c @@ -494,6 +494,96 @@ static int ivo_t109nw41_init(struct hx83102 *ctx) return dsi_ctx.accum_err; }; +static int kingdisplay_kd110n11_51ie_init(struct hx83102 *ctx) +{ + struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; + + msleep(50); + + hx83102_enable_extended_cmds(&dsi_ctx, true); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D9, 0xd1); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x2c, 0xb3, 0xb3, 0x31, 0xf1, + 0x33, 0xe0, 0x54, 0x36, 0x36, 0x3a, 0x3a, 0x32, 0x8b, + 0x11, 0xe5, 0x98); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xd9); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x8b, 0x33); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETDISP, 0x00, 0x47, 0xb0, 0x80, 0x00, 0x2c, + 0x80, 0x3c, 0x9f, 0x22, 0x20, 0x00, 0x00, 0x98, 0x51); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x64, 0x64, 0x64, 0x64, 0x64, 0x64, + 0x40, 0x84, 0x64, 0x84, 0x01, 0x9d, 0x01, 0x02, 0x01, 0x00, + 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETVDC, 0x1b, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_BE, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPTBA, 0xfc, 0xc4, 0x80, 0x9c, 0x36, 0x00, + 0x0d, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSTBA, 0x32, 0x32, 0x22, 0x11, 0x22, 0xa0, + 0x31, 0x08, 0xf5, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcc); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTCON, 0x80); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETRAMDMY, 0x97); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPWM, 0x00, 0x1e, 0x13, 0x88, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x08, 0x13, 0x07, 0x00, + 0x0f, 0x36); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPANEL, 0x02, 0x03, 0x44); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPCTRL, 0x07, 0x06, 0x00, 0x02, + 0x04, 0x2c, 0xff); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x06, 0x00, 0x00, 0x00, 0x40, 0x04, + 0x08, 0x04, 0x08, 0x37, 0x07, 0x44, 0x37, 0x2b, 0x2b, 0x03, + 0x03, 0x32, 0x10, 0x22, 0x00, 0x25, 0x32, 0x10, 0x29, 0x00, + 0x29, 0x32, 0x10, 0x08, 0x00, 0x08, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP1, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, + 0x18, 0x18, 0x18, 0x18, 0x07, 0x06, 0x07, 0x06, 0x05, 0x04, + 0x05, 0x04, 0x03, 0x02, 0x03, 0x02, 0x01, 0x00, 0x01, 0x00, + 0x18, 0x18, 0x25, 0x24, 0x25, 0x24, 0x1f, 0x1f, 0x1f, 0x1f, + 0x1e, 0x1e, 0x1e, 0x1e, 0x20, 0x20, 0x20, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0xe0, 0x10, 0x10, 0x0d, 0x1e, 0x9d, + 0x02, 0x52, 0x9d, 0x14, 0x14); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x01, 0x7f, 0x11, 0xfd); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x4f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x86); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D2, 0x64); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, 0x05, 0x15, 0x55, 0x45, + 0x55, 0x50, 0x05, 0x15, 0x55, 0x45, 0x55, 0x50); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0x02, 0x00, 0x24, 0x01, 0x7e, 0x0f, + 0x7c, 0x10, 0xa0, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x03, 0x07, 0x00, 0x10, 0x7b); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0f, 0x3f, 0xff, 0xcf, 0xff, 0xf0, + 0x0f, 0x3f, 0xff, 0xcf, 0xff, 0xf0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0xfe, 0x01, 0xfe, 0x01, 0xfe, 0x01, + 0x00, 0x00, 0x00, 0x23, 0x00, 0x23, 0x81, 0x02, 0x40, 0x00, + 0x20, 0x9d, 0x02, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x01, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETDISP, 0x66, 0x81); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x03, 0xff, 0xf8); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, + 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0, 0x0f, 0x2a, 0xaa, 0x8a, + 0xaa, 0xf0, 0x0f, 0x2a, 0xaa, 0x8a, 0xaa, 0xf0, 0x0a, 0x2a, + 0xaa, 0x8a, 0xaa, 0xa0, 0x0a, 0x2a, 0xaa, 0x8a, 0xaa, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x00); + hx83102_enable_extended_cmds(&dsi_ctx, false); + + return dsi_ctx.accum_err; +} + static const struct drm_display_mode starry_mode = { .clock = 162680, .hdisplay = 1200, @@ -582,6 +672,28 @@ static const struct hx83102_panel_desc ivo_t109nw41_desc = { .init = ivo_t109nw41_init, }; +static const struct drm_display_mode kingdisplay_kd110n11_51ie_default_mode = { + .clock = 182750, + .hdisplay = 1200, + .hsync_start = 1200 + 124, + .hsync_end = 1200 + 124 + 80, + .htotal = 1200 + 124 + 80 + 80, + .vdisplay = 1920, + .vsync_start = 1920 + 88, + .vsync_end = 1920 + 88 + 8, + .vtotal = 1920 + 88 + 8 + 38, + .type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED, +}; + +static const struct hx83102_panel_desc kingdisplay_kd110n11_51ie_desc = { + .modes = &kingdisplay_kd110n11_51ie_default_mode, + .size = { + .width_mm = 147, + .height_mm = 235, + }, + .init = kingdisplay_kd110n11_51ie_init, +}; + static int hx83102_enable(struct drm_panel *panel) { msleep(130); @@ -809,6 +921,9 @@ static const struct of_device_id hx83102_of_match[] = { { .compatible = "ivo,t109nw41", .data = &ivo_t109nw41_desc }, + { .compatible = "kingdisplay,kd110n11-51ie", + .data = &kingdisplay_kd110n11_51ie_desc + }, { .compatible = "starry,himax83102-j02", .data = &starry_desc }, From 1b80b2101d7b433718a21a9e7d413e066bf8aec6 Mon Sep 17 00:00:00 2001 From: Langyan Ye Date: Sat, 8 Feb 2025 18:53:26 +0800 Subject: [PATCH 0716/1443] drm/panel: panel-himax-hx83102: support for starry-2082109qfh040022-50e MIPI-DSI panel The starry-2082109qfh040022-50e is a 10.95" TFT panel. The MIPI controller on this panel is the same as the other panels here, so add this panel to this driver. Signed-off-by: Langyan Ye Reviewed-by: Douglas Anderson Signed-off-by: Douglas Anderson Link: https://patchwork.freedesktop.org/patch/msgid/20250208105326.3850358-4-yelangyan@huaqin.corp-partner.google.com --- drivers/gpu/drm/panel/panel-himax-hx83102.c | 142 ++++++++++++++++++++ 1 file changed, 142 insertions(+) diff --git a/drivers/gpu/drm/panel/panel-himax-hx83102.c b/drivers/gpu/drm/panel/panel-himax-hx83102.c index 9a818dea653f..66abfc44e424 100644 --- a/drivers/gpu/drm/panel/panel-himax-hx83102.c +++ b/drivers/gpu/drm/panel/panel-himax-hx83102.c @@ -24,6 +24,8 @@ #define HX83102_SETPOWER 0xb1 #define HX83102_SETDISP 0xb2 #define HX83102_SETCYC 0xb4 +#define HX83102_UNKNOWN_B6 0xb6 +#define HX83102_UNKNOWN_B8 0xb8 #define HX83102_SETEXTC 0xb9 #define HX83102_SETMIPI 0xba #define HX83102_SETVDC 0xbc @@ -584,6 +586,121 @@ static int kingdisplay_kd110n11_51ie_init(struct hx83102 *ctx) return dsi_ctx.accum_err; } +static int starry_2082109qfh040022_50e_init(struct hx83102 *ctx) +{ + struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; + + msleep(50); + + hx83102_enable_extended_cmds(&dsi_ctx, true); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D9, 0xd1); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x2c, 0xb5, 0xb5, 0x31, 0xf1, 0x33, + 0xc3, 0x57, 0x36, 0x36, 0x36, 0x36, 0x1a, 0x8b, 0x11, 0x65, + 0x00, 0x88, 0xfa, 0xff, 0xff, 0x8f, 0xff, 0x08, 0x3c, 0x33); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETDISP, 0x00, 0x47, 0xb0, 0x80, 0x00, 0x22, + 0x70, 0x3c, 0xa1, 0x22, 0x00, 0x00, 0x00, 0x88, 0xf4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x14, 0x16, 0x14, 0x50, 0x14, 0x50, + 0x0d, 0x6a, 0x0d, 0x6a, 0x01, 0x9e); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_B6, 0x34, 0x34, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_B8, 0x40); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcd); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x84); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETVDC, 0x1b, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_BE, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPTBA, 0xfc, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSTBA, 0x38, 0x38, 0x22, 0x11, 0x33, 0xa0, + 0x61, 0x08, 0xf5, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcc); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTCON, 0x80); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETRAMDMY, 0x97); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPWM, 0x00, 0x1e, 0x30, 0xd4, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x08, 0x13, 0x07, 0x00, 0x0f, + 0x16); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPANEL, 0x02, 0x03, 0x44); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCASCADE, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPCTRL, 0x37, 0x06, 0x00, 0x02, 0x04, + 0x2c, 0xff); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x06, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x3b, 0x03, 0x73, 0x3b, 0x21, 0x21, 0x03, + 0x03, 0x98, 0x10, 0x1d, 0x00, 0x1d, 0x32, 0x17, 0xa1, 0x07, + 0xa1, 0x43, 0x17, 0xa6, 0x07, 0xa6, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP1, 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, + 0x40, 0x40, 0x18, 0x18, 0x18, 0x18, 0x2a, 0x2b, 0x1f, 0x1f, + 0x1e, 0x1e, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2b, + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, + 0x0a, 0x0b, 0x20, 0x21, 0x18, 0x18, 0x18, 0x18); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x02, 0xaa, 0xea, 0xaa, 0xaa, 0x00, + 0x02, 0xaa, 0xea, 0xaa, 0xaa, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0x07, 0x10, 0x10, 0x2a, 0x32, 0x9f, + 0x01, 0x5a, 0x91, 0x14, 0x14, 0x00, 0x00, 0x00, 0x00, 0x12, + 0x05, 0x02, 0x02, 0x10, 0x33, 0x02, 0x04, 0x18, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPOWER, 0x01, 0x7f, 0x11, 0xfd); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x86); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D2, 0x3d); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x00, 0x00, 0x00, 0x80, 0x80, 0x0c, + 0xa1); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0x03, 0xff, 0xff, 0xff, 0xff, 0x00, + 0x03, 0xff, 0xff, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0x02, 0x00, 0x2d, 0x01, 0x7f, 0x0f, + 0x7c, 0x10, 0xa0, 0x00, 0x00, 0x77, 0x00, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETPTBA, 0xf2); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCLOCK, 0x02, 0x00, 0x00, 0x10, 0x58); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_D2, 0x0a, 0x0a, 0x05, 0x03, 0x0a, + 0x0a, 0x01, 0x03, 0x01, 0x01, 0x05, 0x0e); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcc); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP0, 0x03, 0x1f, 0xe0, 0x11, 0x70); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0xab, 0xff, 0xff, 0xff, 0xff, 0xa0, + 0xab, 0xff, 0xff, 0xff, 0xff, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETTP1, 0xfe, 0x01, 0xfe, 0x01, 0xfe, 0x01, + 0x00, 0x00, 0x00, 0x03, 0x00, 0x03, 0x81, 0x02, 0x40, 0x00, + 0x20, 0x9e, 0x02, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc6); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETCYC, 0x03, 0xff, 0xf8); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETGIP3, 0xaa, 0xab, 0xea, 0xaa, 0xaa, 0xa0, + 0xaa, 0xab, 0xea, 0xaa, 0xaa, 0xa0, 0xaa, 0xbf, 0xff, 0xff, + 0xfe, 0xa0, 0xaa, 0xbf, 0xff, 0xff, 0xfe, 0xa0, 0xaa, 0xaa, + 0xaa, 0xaa, 0xaa, 0xa0, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xa0); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_UNKNOWN_E1, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x96); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x4f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0xcc); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETMIPI, 0x84); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETSPCCMD, 0x3f); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83102_SETBANK, 0x00); + hx83102_enable_extended_cmds(&dsi_ctx, false); + + mipi_dsi_msleep(&dsi_ctx, 110); + + return dsi_ctx.accum_err; +} + static const struct drm_display_mode starry_mode = { .clock = 162680, .hdisplay = 1200, @@ -694,6 +811,28 @@ static const struct hx83102_panel_desc kingdisplay_kd110n11_51ie_desc = { .init = kingdisplay_kd110n11_51ie_init, }; +static const struct drm_display_mode starry_2082109qfh040022_50e_default_mode = { + .clock = 192050, + .hdisplay = 1200, + .hsync_start = 1200 + 160, + .hsync_end = 1200 + 160 + 66, + .htotal = 1200 + 160 + 66 + 120, + .vdisplay = 1920, + .vsync_start = 1920 + 115, + .vsync_end = 1920 + 115 + 8, + .vtotal = 1920 + 115 + 8 + 28, + .type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED, +}; + +static const struct hx83102_panel_desc starry_2082109qfh040022_50e_desc = { + .modes = &starry_2082109qfh040022_50e_default_mode, + .size = { + .width_mm = 147, + .height_mm = 235, + }, + .init = starry_2082109qfh040022_50e_init, +}; + static int hx83102_enable(struct drm_panel *panel) { msleep(130); @@ -924,6 +1063,9 @@ static const struct of_device_id hx83102_of_match[] = { { .compatible = "kingdisplay,kd110n11-51ie", .data = &kingdisplay_kd110n11_51ie_desc }, + { .compatible = "starry,2082109qfh040022-50e", + .data = &starry_2082109qfh040022_50e_desc + }, { .compatible = "starry,himax83102-j02", .data = &starry_desc }, From 66ac08bb7788a130f731614423a133c9c333c64e Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Thu, 6 Feb 2025 18:02:33 +0100 Subject: [PATCH 0717/1443] drm/panel: remove unnecessary forward declaration 'struct drm_device' is not used at all in this file since commit aa6c43644bc5 ("drm/panel: drop drm_device from drm_panel"). Signed-off-by: Luca Ceresoli Reviewed-by: Douglas Anderson Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206-drm-cleanups-v1-1-93df60faa756@bootlin.com --- include/drm/drm_panel.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/drm/drm_panel.h b/include/drm/drm_panel.h index 10015891b056..a9c042c8dea1 100644 --- a/include/drm/drm_panel.h +++ b/include/drm/drm_panel.h @@ -33,7 +33,6 @@ struct backlight_device; struct dentry; struct device_node; struct drm_connector; -struct drm_device; struct drm_panel_follower; struct drm_panel; struct display_timing; From f4e71f66036872e67a5eeb5800828541da5d12c9 Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Thu, 6 Feb 2025 18:02:34 +0100 Subject: [PATCH 0718/1443] drm/bridge: nxp-ptn3460: remove unused drm_panel.h include The file uses the panel_bridge APIs from drm_bridge.h, but no drm_panel APIs from drm_panel.h. Signed-off-by: Luca Ceresoli Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206-drm-cleanups-v1-2-93df60faa756@bootlin.com --- drivers/gpu/drm/bridge/nxp-ptn3460.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c index 44e36ae66db4..27261b2ac9c8 100644 --- a/drivers/gpu/drm/bridge/nxp-ptn3460.c +++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include From 202f3f60d1743cacb4924566daac846d33be5378 Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Thu, 6 Feb 2025 18:02:35 +0100 Subject: [PATCH 0719/1443] drm/bridge: parade-ps8622: remove unused drm_panel.h include The file uses the panel_bridge APIs from drm_bridge.h, but no drm_panel APIs from drm_panel.h. Signed-off-by: Luca Ceresoli Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206-drm-cleanups-v1-3-93df60faa756@bootlin.com --- drivers/gpu/drm/bridge/parade-ps8622.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c index ae3ab9262ef1..13ada42a5514 100644 --- a/drivers/gpu/drm/bridge/parade-ps8622.c +++ b/drivers/gpu/drm/bridge/parade-ps8622.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include From 10fab0675483617867e1f4f59f1b1fe9bd4d719d Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Thu, 6 Feb 2025 18:02:36 +0100 Subject: [PATCH 0720/1443] drm/bridge: parade-ps8640: remove unused drm_panel.h include The file uses the panel_bridge APIs from drm_bridge.h, but no drm_panel APIs from drm_panel.h. Signed-off-by: Luca Ceresoli Reviewed-by: Douglas Anderson Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206-drm-cleanups-v1-4-93df60faa756@bootlin.com --- drivers/gpu/drm/bridge/parade-ps8640.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c index 14d4dcf239da..b0e38177c1ec 100644 --- a/drivers/gpu/drm/bridge/parade-ps8640.c +++ b/drivers/gpu/drm/bridge/parade-ps8640.c @@ -20,7 +20,6 @@ #include #include #include -#include #include #define PAGE0_AUXCH_CFG3 0x76 From 38e092786e54ae102eafe07c938aa289895639f1 Mon Sep 17 00:00:00 2001 From: Luca Ceresoli Date: Thu, 6 Feb 2025 18:02:37 +0100 Subject: [PATCH 0721/1443] drm/bridge: tc358762: remove unused drm_panel.h include, add drm_bridge.h The file uses the panel_bridge APIs from drm_bridge.h, which is included only indirectly, and uses no drm_panel APIs from drm_panel.h. Signed-off-by: Luca Ceresoli Reviewed-by: Dmitry Baryshkov Signed-off-by: Robert Foss Link: https://patchwork.freedesktop.org/patch/msgid/20250206-drm-cleanups-v1-5-93df60faa756@bootlin.com --- drivers/gpu/drm/bridge/tc358762.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/tc358762.c b/drivers/gpu/drm/bridge/tc358762.c index 46198af9eebb..f92aff3d515a 100644 --- a/drivers/gpu/drm/bridge/tc358762.c +++ b/drivers/gpu/drm/bridge/tc358762.c @@ -20,10 +20,10 @@ #include