diff --git a/Documentation/devicetree/bindings/display/msm/dsi.txt b/Documentation/devicetree/bindings/display/msm/dsi.txt
index f344b9e49198..e7423bea1424 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi.txt
+++ b/Documentation/devicetree/bindings/display/msm/dsi.txt
@@ -14,17 +14,20 @@ Required properties:
 - clocks: device clocks
   See Documentation/devicetree/bindings/clocks/clock-bindings.txt for details.
 - clock-names: the following clocks are required:
-  * "bus_clk"
-  * "byte_clk"
-  * "core_clk"
-  * "core_mmss_clk"
-  * "iface_clk"
   * "mdp_core_clk"
+  * "iface_clk"
+  * "bus_clk"
+  * "core_mmss_clk"
+  * "byte_clk"
   * "pixel_clk"
+  * "core_clk"
+  For DSIv2, we need an additional clock:
+   * "src_clk"
 - vdd-supply: phandle to vdd regulator device node
 - vddio-supply: phandle to vdd-io regulator device node
 - vdda-supply: phandle to vdda regulator device node
 - qcom,dsi-phy: phandle to DSI PHY device node
+- syscon-sfpb: A phandle to mmss_sfpb syscon node (only for DSIv2)
 
 Optional properties:
 - panel@0: Node of panel connected to this DSI controller.
@@ -51,6 +54,7 @@ Required properties:
   * "qcom,dsi-phy-28nm-hpm"
   * "qcom,dsi-phy-28nm-lp"
   * "qcom,dsi-phy-20nm"
+  * "qcom,dsi-phy-28nm-8960"
 - reg: Physical base address and length of the registers of PLL, PHY and PHY
   regulator
 - reg-names: The names of register regions. The following regions are required:
diff --git a/Documentation/devicetree/bindings/display/msm/mdp.txt b/Documentation/devicetree/bindings/display/msm/mdp.txt
index 0833edaba4c3..a214f6cd0363 100644
--- a/Documentation/devicetree/bindings/display/msm/mdp.txt
+++ b/Documentation/devicetree/bindings/display/msm/mdp.txt
@@ -2,18 +2,28 @@ Qualcomm adreno/snapdragon display controller
 
 Required properties:
 - compatible:
-  * "qcom,mdp" - mdp4
+  * "qcom,mdp4" - mdp4
+  * "qcom,mdp5" - mdp5
 - reg: Physical base address and length of the controller's registers.
 - interrupts: The interrupt signal from the display controller.
 - connectors: array of phandles for output device(s)
 - clocks: device clocks
   See ../clocks/clock-bindings.txt for details.
-- clock-names: the following clocks are required:
-  * "core_clk"
-  * "iface_clk"
-  * "src_clk"
-  * "hdmi_clk"
-  * "mpd_clk"
+- clock-names: the following clocks are required.
+  For MDP4:
+   * "core_clk"
+   * "iface_clk"
+   * "lut_clk"
+   * "src_clk"
+   * "hdmi_clk"
+   * "mdp_clk"
+  For MDP5:
+   * "bus_clk"
+   * "iface_clk"
+   * "core_clk_src"
+   * "core_clk"
+   * "lut_clk" (some MDP5 versions may not need this)
+   * "vsync_clk"
 
 Optional properties:
 - gpus: phandle for gpu device
@@ -26,7 +36,7 @@ Example:
 	...
 
 	mdp: qcom,mdp@5100000 {
-		compatible = "qcom,mdp";
+		compatible = "qcom,mdp4";
 		reg = <0x05100000 0xf0000>;
 		interrupts = <GIC_SPI 75 0>;
 		connectors = <&hdmi>;
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 84d3ec98e6b9..215495c2780c 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -54,3 +54,11 @@ config DRM_MSM_DSI_20NM_PHY
 	default y
 	help
 	  Choose this option if the 20nm DSI PHY is used on the platform.
+
+config DRM_MSM_DSI_28NM_8960_PHY
+	bool "Enable DSI 28nm 8960 PHY driver in MSM DRM"
+	depends on DRM_MSM_DSI
+	default y
+	help
+	  Choose this option if the 28nm DSI PHY 8960 variant is used on the
+	  platform.
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 1c90290be716..065ad4138799 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -54,6 +54,7 @@ msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o
 msm-$(CONFIG_COMMON_CLK) += mdp/mdp4/mdp4_lvds_pll.o
 
 msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \
+			mdp/mdp4/mdp4_dsi_encoder.o \
 			dsi/dsi_cfg.o \
 			dsi/dsi_host.o \
 			dsi/dsi_manager.o \
@@ -62,10 +63,12 @@ msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \
 
 msm-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/phy/dsi_phy_28nm.o
 msm-$(CONFIG_DRM_MSM_DSI_20NM_PHY) += dsi/phy/dsi_phy_20nm.o
+msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/phy/dsi_phy_28nm_8960.o
 
 ifeq ($(CONFIG_DRM_MSM_DSI_PLL),y)
 msm-y += dsi/pll/dsi_pll.o
 msm-$(CONFIG_DRM_MSM_DSI_28NM_PHY) += dsi/pll/dsi_pll_28nm.o
+msm-$(CONFIG_DRM_MSM_DSI_28NM_8960_PHY) += dsi/pll/dsi_pll_28nm_8960.o
 endif
 
 obj-$(CONFIG_DRM_MSM)	+= msm.o
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 1ea2df524fac..950d27d26b30 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -19,10 +19,6 @@
 
 #include "adreno_gpu.h"
 
-#if defined(DOWNSTREAM_CONFIG_MSM_BUS_SCALING) && !defined(CONFIG_OF)
-#  include <mach/kgsl.h>
-#endif
-
 #define ANY_ID 0xff
 
 bool hang_debug = false;
@@ -168,7 +164,6 @@ static void set_gpu_pdev(struct drm_device *dev,
 static int adreno_bind(struct device *dev, struct device *master, void *data)
 {
 	static struct adreno_platform_config config = {};
-#ifdef CONFIG_OF
 	struct device_node *child, *node = dev->of_node;
 	u32 val;
 	int ret;
@@ -205,53 +200,6 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
 		return -ENXIO;
 	}
 
-#else
-	struct kgsl_device_platform_data *pdata = dev->platform_data;
-	uint32_t version = socinfo_get_version();
-	if (cpu_is_apq8064ab()) {
-		config.fast_rate = 450000000;
-		config.slow_rate = 27000000;
-		config.bus_freq  = 4;
-		config.rev = ADRENO_REV(3, 2, 1, 0);
-	} else if (cpu_is_apq8064()) {
-		config.fast_rate = 400000000;
-		config.slow_rate = 27000000;
-		config.bus_freq  = 4;
-
-		if (SOCINFO_VERSION_MAJOR(version) == 2)
-			config.rev = ADRENO_REV(3, 2, 0, 2);
-		else if ((SOCINFO_VERSION_MAJOR(version) == 1) &&
-				(SOCINFO_VERSION_MINOR(version) == 1))
-			config.rev = ADRENO_REV(3, 2, 0, 1);
-		else
-			config.rev = ADRENO_REV(3, 2, 0, 0);
-
-	} else if (cpu_is_msm8960ab()) {
-		config.fast_rate = 400000000;
-		config.slow_rate = 320000000;
-		config.bus_freq  = 4;
-
-		if (SOCINFO_VERSION_MINOR(version) == 0)
-			config.rev = ADRENO_REV(3, 2, 1, 0);
-		else
-			config.rev = ADRENO_REV(3, 2, 1, 1);
-
-	} else if (cpu_is_msm8930()) {
-		config.fast_rate = 400000000;
-		config.slow_rate = 27000000;
-		config.bus_freq  = 3;
-
-		if ((SOCINFO_VERSION_MAJOR(version) == 1) &&
-			(SOCINFO_VERSION_MINOR(version) == 2))
-			config.rev = ADRENO_REV(3, 0, 5, 2);
-		else
-			config.rev = ADRENO_REV(3, 0, 5, 0);
-
-	}
-#  ifdef DOWNSTREAM_CONFIG_MSM_BUS_SCALING
-	config.bus_scale_table = pdata->bus_scale_table;
-#  endif
-#endif
 	dev->platform_data = &config;
 	set_gpu_pdev(dev_get_drvdata(master), to_platform_device(dev));
 	return 0;
diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h
index 5f5a3732cdf6..749fbb28ec3d 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.h
@@ -31,10 +31,12 @@ enum msm_dsi_phy_type {
 	MSM_DSI_PHY_28NM_HPM,
 	MSM_DSI_PHY_28NM_LP,
 	MSM_DSI_PHY_20NM,
+	MSM_DSI_PHY_28NM_8960,
 	MSM_DSI_PHY_MAX
 };
 
 #define DSI_DEV_REGULATOR_MAX	8
+#define DSI_BUS_CLK_MAX		4
 
 /* Regulators for DSI devices */
 struct dsi_reg_entry {
@@ -89,7 +91,7 @@ int msm_dsi_manager_phy_enable(int id,
 		u32 *clk_pre, u32 *clk_post);
 void msm_dsi_manager_phy_disable(int id);
 int msm_dsi_manager_cmd_xfer(int id, const struct mipi_dsi_msg *msg);
-bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len);
+bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 dma_base, u32 len);
 int msm_dsi_manager_register(struct msm_dsi *msm_dsi);
 void msm_dsi_manager_unregister(struct msm_dsi *msm_dsi);
 
@@ -143,7 +145,7 @@ int msm_dsi_host_cmd_tx(struct mipi_dsi_host *host,
 int msm_dsi_host_cmd_rx(struct mipi_dsi_host *host,
 					const struct mipi_dsi_msg *msg);
 void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host,
-					u32 iova, u32 len);
+					u32 dma_base, u32 len);
 int msm_dsi_host_enable(struct mipi_dsi_host *host);
 int msm_dsi_host_disable(struct mipi_dsi_host *host);
 int msm_dsi_host_power_on(struct mipi_dsi_host *host);
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
index 5872d5e5934f..2a827d8093a2 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
@@ -13,9 +13,26 @@
 
 #include "dsi_cfg.h"
 
-/* DSI v2 has not been supported by now */
-static const struct msm_dsi_config dsi_v2_cfg = {
+static const char * const dsi_v2_bus_clk_names[] = {
+	"core_mmss_clk", "iface_clk", "bus_clk",
+};
+
+static const struct msm_dsi_config apq8064_dsi_cfg = {
 	.io_offset = 0,
+	.reg_cfg = {
+		.num = 3,
+		.regs = {
+			{"vdda", 1200000, 1200000, 100000, 100},
+			{"avdd", 3000000, 3000000, 110000, 100},
+			{"vddio", 1800000, 1800000, 100000, 100},
+		},
+	},
+	.bus_clk_names = dsi_v2_bus_clk_names,
+	.num_bus_clks = ARRAY_SIZE(dsi_v2_bus_clk_names),
+};
+
+static const char * const dsi_6g_bus_clk_names[] = {
+	"mdp_core_clk", "iface_clk", "bus_clk", "core_mmss_clk",
 };
 
 static const struct msm_dsi_config msm8974_apq8084_dsi_cfg = {
@@ -29,6 +46,12 @@ static const struct msm_dsi_config msm8974_apq8084_dsi_cfg = {
 			{"vddio", 1800000, 1800000, 100000, 100},
 		},
 	},
+	.bus_clk_names = dsi_6g_bus_clk_names,
+	.num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names),
+};
+
+static const char * const dsi_8916_bus_clk_names[] = {
+	"mdp_core_clk", "iface_clk", "bus_clk",
 };
 
 static const struct msm_dsi_config msm8916_dsi_cfg = {
@@ -42,6 +65,8 @@ static const struct msm_dsi_config msm8916_dsi_cfg = {
 			{"vddio", 1800000, 1800000, 100000, 100},
 		},
 	},
+	.bus_clk_names = dsi_8916_bus_clk_names,
+	.num_bus_clks = ARRAY_SIZE(dsi_8916_bus_clk_names),
 };
 
 static const struct msm_dsi_config msm8994_dsi_cfg = {
@@ -57,11 +82,13 @@ static const struct msm_dsi_config msm8994_dsi_cfg = {
 			{"lab_reg", -1, -1, -1, -1},
 			{"ibb_reg", -1, -1, -1, -1},
 		},
-	}
+	},
+	.bus_clk_names = dsi_6g_bus_clk_names,
+	.num_bus_clks = ARRAY_SIZE(dsi_6g_bus_clk_names),
 };
 
 static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = {
-	{MSM_DSI_VER_MAJOR_V2, U32_MAX, &dsi_v2_cfg},
+	{MSM_DSI_VER_MAJOR_V2, MSM_DSI_V2_VER_MINOR_8064, &apq8064_dsi_cfg},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_0,
 						&msm8974_apq8084_dsi_cfg},
 	{MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V1_1,
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
index 4cf887240177..a68c836744a3 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h
@@ -25,11 +25,15 @@
 #define MSM_DSI_6G_VER_MINOR_V1_3	0x10030000
 #define MSM_DSI_6G_VER_MINOR_V1_3_1	0x10030001
 
+#define MSM_DSI_V2_VER_MINOR_8064	0x0
+
 #define DSI_6G_REG_SHIFT	4
 
 struct msm_dsi_config {
 	u32 io_offset;
 	struct dsi_reg_config reg_cfg;
+	const char * const *bus_clk_names;
+	const int num_bus_clks;
 };
 
 struct msm_dsi_cfg_handler {
diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 4c49868efcda..48f9967b4a1b 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -24,26 +24,36 @@
 #include <linux/of_graph.h>
 #include <linux/regulator/consumer.h>
 #include <linux/spinlock.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
 #include <video/mipi_display.h>
 
 #include "dsi.h"
 #include "dsi.xml.h"
+#include "sfpb.xml.h"
 #include "dsi_cfg.h"
 
 static int dsi_get_version(const void __iomem *base, u32 *major, u32 *minor)
 {
 	u32 ver;
-	u32 ver_6g;
 
 	if (!major || !minor)
 		return -EINVAL;
 
-	/* From DSI6G(v3), addition of a 6G_HW_VERSION register at offset 0
+	/*
+	 * From DSI6G(v3), addition of a 6G_HW_VERSION register at offset 0
 	 * makes all other registers 4-byte shifted down.
+	 *
+	 * In order to identify between DSI6G(v3) and beyond, and DSIv2 and
+	 * older, we read the DSI_VERSION register without any shift(offset
+	 * 0x1f0). In the case of DSIv2, this hast to be a non-zero value. In
+	 * the case of DSI6G, this has to be zero (the offset points to a
+	 * scratch register which we never touch)
 	 */
-	ver_6g = msm_readl(base + REG_DSI_6G_HW_VERSION);
-	if (ver_6g == 0) {
-		ver = msm_readl(base + REG_DSI_VERSION);
+
+	ver = msm_readl(base + REG_DSI_VERSION);
+	if (ver) {
+		/* older dsi host, there is no register shift */
 		ver = FIELD(ver, DSI_VERSION_MAJOR);
 		if (ver <= MSM_DSI_VER_MAJOR_V2) {
 			/* old versions */
@@ -54,12 +64,17 @@ static int dsi_get_version(const void __iomem *base, u32 *major, u32 *minor)
 			return -EINVAL;
 		}
 	} else {
+		/*
+		 * newer host, offset 0 has 6G_HW_VERSION, the rest of the
+		 * registers are shifted down, read DSI_VERSION again with
+		 * the shifted offset
+		 */
 		ver = msm_readl(base + DSI_6G_REG_SHIFT + REG_DSI_VERSION);
 		ver = FIELD(ver, DSI_VERSION_MAJOR);
 		if (ver == MSM_DSI_VER_MAJOR_6G) {
 			/* 6G version */
 			*major = ver;
-			*minor = ver_6g;
+			*minor = msm_readl(base + REG_DSI_6G_HW_VERSION);
 			return 0;
 		} else {
 			return -EINVAL;
@@ -91,10 +106,9 @@ struct msm_dsi_host {
 
 	void __iomem *ctrl_base;
 	struct regulator_bulk_data supplies[DSI_DEV_REGULATOR_MAX];
-	struct clk *mdp_core_clk;
-	struct clk *ahb_clk;
-	struct clk *axi_clk;
-	struct clk *mmss_misc_ahb_clk;
+
+	struct clk *bus_clks[DSI_BUS_CLK_MAX];
+
 	struct clk *byte_clk;
 	struct clk *esc_clk;
 	struct clk *pixel_clk;
@@ -102,6 +116,14 @@ struct msm_dsi_host {
 	struct clk *pixel_clk_src;
 
 	u32 byte_clk_rate;
+	u32 esc_clk_rate;
+
+	/* DSI v2 specific clocks */
+	struct clk *src_clk;
+	struct clk *esc_clk_src;
+	struct clk *dsi_clk_src;
+
+	u32 src_clk_rate;
 
 	struct gpio_desc *disp_en_gpio;
 	struct gpio_desc *te_gpio;
@@ -119,9 +141,19 @@ struct msm_dsi_host {
 	struct work_struct err_work;
 	struct workqueue_struct *workqueue;
 
+	/* DSI 6G TX buffer*/
 	struct drm_gem_object *tx_gem_obj;
+
+	/* DSI v2 TX buffer */
+	void *tx_buf;
+	dma_addr_t tx_buf_paddr;
+
+	int tx_size;
+
 	u8 *rx_buf;
 
+	struct regmap *sfpb;
+
 	struct drm_display_mode *mode;
 
 	/* connected device info */
@@ -165,21 +197,31 @@ static const struct msm_dsi_cfg_handler *dsi_get_config(
 						struct msm_dsi_host *msm_host)
 {
 	const struct msm_dsi_cfg_handler *cfg_hnd = NULL;
+	struct device *dev = &msm_host->pdev->dev;
 	struct regulator *gdsc_reg;
+	struct clk *ahb_clk;
 	int ret;
 	u32 major = 0, minor = 0;
 
-	gdsc_reg = regulator_get(&msm_host->pdev->dev, "gdsc");
+	gdsc_reg = regulator_get(dev, "gdsc");
 	if (IS_ERR(gdsc_reg)) {
 		pr_err("%s: cannot get gdsc\n", __func__);
 		goto exit;
 	}
+
+	ahb_clk = clk_get(dev, "iface_clk");
+	if (IS_ERR(ahb_clk)) {
+		pr_err("%s: cannot get interface clock\n", __func__);
+		goto put_gdsc;
+	}
+
 	ret = regulator_enable(gdsc_reg);
 	if (ret) {
 		pr_err("%s: unable to enable gdsc\n", __func__);
-		goto put_gdsc;
+		goto put_clk;
 	}
-	ret = clk_prepare_enable(msm_host->ahb_clk);
+
+	ret = clk_prepare_enable(ahb_clk);
 	if (ret) {
 		pr_err("%s: unable to enable ahb_clk\n", __func__);
 		goto disable_gdsc;
@@ -196,9 +238,11 @@ static const struct msm_dsi_cfg_handler *dsi_get_config(
 	DBG("%s: Version %x:%x\n", __func__, major, minor);
 
 disable_clks:
-	clk_disable_unprepare(msm_host->ahb_clk);
+	clk_disable_unprepare(ahb_clk);
 disable_gdsc:
 	regulator_disable(gdsc_reg);
+put_clk:
+	clk_put(ahb_clk);
 put_gdsc:
 	regulator_put(gdsc_reg);
 exit:
@@ -295,40 +339,23 @@ static int dsi_regulator_init(struct msm_dsi_host *msm_host)
 static int dsi_clk_init(struct msm_dsi_host *msm_host)
 {
 	struct device *dev = &msm_host->pdev->dev;
-	int ret = 0;
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
+	const struct msm_dsi_config *cfg = cfg_hnd->cfg;
+	int i, ret = 0;
 
-	msm_host->mdp_core_clk = devm_clk_get(dev, "mdp_core_clk");
-	if (IS_ERR(msm_host->mdp_core_clk)) {
-		ret = PTR_ERR(msm_host->mdp_core_clk);
-		pr_err("%s: Unable to get mdp core clk. ret=%d\n",
-			__func__, ret);
-		goto exit;
-	}
-
-	msm_host->ahb_clk = devm_clk_get(dev, "iface_clk");
-	if (IS_ERR(msm_host->ahb_clk)) {
-		ret = PTR_ERR(msm_host->ahb_clk);
-		pr_err("%s: Unable to get mdss ahb clk. ret=%d\n",
-			__func__, ret);
-		goto exit;
-	}
-
-	msm_host->axi_clk = devm_clk_get(dev, "bus_clk");
-	if (IS_ERR(msm_host->axi_clk)) {
-		ret = PTR_ERR(msm_host->axi_clk);
-		pr_err("%s: Unable to get axi bus clk. ret=%d\n",
-			__func__, ret);
-		goto exit;
-	}
-
-	msm_host->mmss_misc_ahb_clk = devm_clk_get(dev, "core_mmss_clk");
-	if (IS_ERR(msm_host->mmss_misc_ahb_clk)) {
-		ret = PTR_ERR(msm_host->mmss_misc_ahb_clk);
-		pr_err("%s: Unable to get mmss misc ahb clk. ret=%d\n",
-			__func__, ret);
-		goto exit;
+	/* get bus clocks */
+	for (i = 0; i < cfg->num_bus_clks; i++) {
+		msm_host->bus_clks[i] = devm_clk_get(dev,
+						cfg->bus_clk_names[i]);
+		if (IS_ERR(msm_host->bus_clks[i])) {
+			ret = PTR_ERR(msm_host->bus_clks[i]);
+			pr_err("%s: Unable to get %s, ret = %d\n",
+				__func__, cfg->bus_clk_names[i], ret);
+			goto exit;
+		}
 	}
 
+	/* get link and source clocks */
 	msm_host->byte_clk = devm_clk_get(dev, "byte_clk");
 	if (IS_ERR(msm_host->byte_clk)) {
 		ret = PTR_ERR(msm_host->byte_clk);
@@ -356,80 +383,85 @@ static int dsi_clk_init(struct msm_dsi_host *msm_host)
 		goto exit;
 	}
 
-	msm_host->byte_clk_src = devm_clk_get(dev, "byte_clk_src");
-	if (IS_ERR(msm_host->byte_clk_src)) {
-		ret = PTR_ERR(msm_host->byte_clk_src);
+	msm_host->byte_clk_src = clk_get_parent(msm_host->byte_clk);
+	if (!msm_host->byte_clk_src) {
+		ret = -ENODEV;
 		pr_err("%s: can't find byte_clk_src. ret=%d\n", __func__, ret);
-		msm_host->byte_clk_src = NULL;
 		goto exit;
 	}
 
-	msm_host->pixel_clk_src = devm_clk_get(dev, "pixel_clk_src");
-	if (IS_ERR(msm_host->pixel_clk_src)) {
-		ret = PTR_ERR(msm_host->pixel_clk_src);
+	msm_host->pixel_clk_src = clk_get_parent(msm_host->pixel_clk);
+	if (!msm_host->pixel_clk_src) {
+		ret = -ENODEV;
 		pr_err("%s: can't find pixel_clk_src. ret=%d\n", __func__, ret);
-		msm_host->pixel_clk_src = NULL;
 		goto exit;
 	}
 
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) {
+		msm_host->src_clk = devm_clk_get(dev, "src_clk");
+		if (IS_ERR(msm_host->src_clk)) {
+			ret = PTR_ERR(msm_host->src_clk);
+			pr_err("%s: can't find dsi_src_clk. ret=%d\n",
+				__func__, ret);
+			msm_host->src_clk = NULL;
+			goto exit;
+		}
+
+		msm_host->esc_clk_src = clk_get_parent(msm_host->esc_clk);
+		if (!msm_host->esc_clk_src) {
+			ret = -ENODEV;
+			pr_err("%s: can't get esc_clk_src. ret=%d\n",
+				__func__, ret);
+			goto exit;
+		}
+
+		msm_host->dsi_clk_src = clk_get_parent(msm_host->src_clk);
+		if (!msm_host->dsi_clk_src) {
+			ret = -ENODEV;
+			pr_err("%s: can't get dsi_clk_src. ret=%d\n",
+				__func__, ret);
+		}
+	}
 exit:
 	return ret;
 }
 
 static int dsi_bus_clk_enable(struct msm_dsi_host *msm_host)
 {
-	int ret;
+	const struct msm_dsi_config *cfg = msm_host->cfg_hnd->cfg;
+	int i, ret;
 
 	DBG("id=%d", msm_host->id);
 
-	ret = clk_prepare_enable(msm_host->mdp_core_clk);
-	if (ret) {
-		pr_err("%s: failed to enable mdp_core_clock, %d\n",
-							 __func__, ret);
-		goto core_clk_err;
-	}
-
-	ret = clk_prepare_enable(msm_host->ahb_clk);
-	if (ret) {
-		pr_err("%s: failed to enable ahb clock, %d\n", __func__, ret);
-		goto ahb_clk_err;
-	}
-
-	ret = clk_prepare_enable(msm_host->axi_clk);
-	if (ret) {
-		pr_err("%s: failed to enable ahb clock, %d\n", __func__, ret);
-		goto axi_clk_err;
-	}
-
-	ret = clk_prepare_enable(msm_host->mmss_misc_ahb_clk);
-	if (ret) {
-		pr_err("%s: failed to enable mmss misc ahb clk, %d\n",
-			__func__, ret);
-		goto misc_ahb_clk_err;
+	for (i = 0; i < cfg->num_bus_clks; i++) {
+		ret = clk_prepare_enable(msm_host->bus_clks[i]);
+		if (ret) {
+			pr_err("%s: failed to enable bus clock %d ret %d\n",
+				__func__, i, ret);
+			goto err;
+		}
 	}
 
 	return 0;
+err:
+	for (; i > 0; i--)
+		clk_disable_unprepare(msm_host->bus_clks[i]);
 
-misc_ahb_clk_err:
-	clk_disable_unprepare(msm_host->axi_clk);
-axi_clk_err:
-	clk_disable_unprepare(msm_host->ahb_clk);
-ahb_clk_err:
-	clk_disable_unprepare(msm_host->mdp_core_clk);
-core_clk_err:
 	return ret;
 }
 
 static void dsi_bus_clk_disable(struct msm_dsi_host *msm_host)
 {
+	const struct msm_dsi_config *cfg = msm_host->cfg_hnd->cfg;
+	int i;
+
 	DBG("");
-	clk_disable_unprepare(msm_host->mmss_misc_ahb_clk);
-	clk_disable_unprepare(msm_host->axi_clk);
-	clk_disable_unprepare(msm_host->ahb_clk);
-	clk_disable_unprepare(msm_host->mdp_core_clk);
+
+	for (i = cfg->num_bus_clks - 1; i >= 0; i--)
+		clk_disable_unprepare(msm_host->bus_clks[i]);
 }
 
-static int dsi_link_clk_enable(struct msm_dsi_host *msm_host)
+static int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host)
 {
 	int ret;
 
@@ -476,11 +508,98 @@ static int dsi_link_clk_enable(struct msm_dsi_host *msm_host)
 	return ret;
 }
 
+static int dsi_link_clk_enable_v2(struct msm_dsi_host *msm_host)
+{
+	int ret;
+
+	DBG("Set clk rates: pclk=%d, byteclk=%d, esc_clk=%d, dsi_src_clk=%d",
+		msm_host->mode->clock, msm_host->byte_clk_rate,
+		msm_host->esc_clk_rate, msm_host->src_clk_rate);
+
+	ret = clk_set_rate(msm_host->byte_clk, msm_host->byte_clk_rate);
+	if (ret) {
+		pr_err("%s: Failed to set rate byte clk, %d\n", __func__, ret);
+		goto error;
+	}
+
+	ret = clk_set_rate(msm_host->esc_clk, msm_host->esc_clk_rate);
+	if (ret) {
+		pr_err("%s: Failed to set rate esc clk, %d\n", __func__, ret);
+		goto error;
+	}
+
+	ret = clk_set_rate(msm_host->src_clk, msm_host->src_clk_rate);
+	if (ret) {
+		pr_err("%s: Failed to set rate src clk, %d\n", __func__, ret);
+		goto error;
+	}
+
+	ret = clk_set_rate(msm_host->pixel_clk, msm_host->mode->clock * 1000);
+	if (ret) {
+		pr_err("%s: Failed to set rate pixel clk, %d\n", __func__, ret);
+		goto error;
+	}
+
+	ret = clk_prepare_enable(msm_host->byte_clk);
+	if (ret) {
+		pr_err("%s: Failed to enable dsi byte clk\n", __func__);
+		goto error;
+	}
+
+	ret = clk_prepare_enable(msm_host->esc_clk);
+	if (ret) {
+		pr_err("%s: Failed to enable dsi esc clk\n", __func__);
+		goto esc_clk_err;
+	}
+
+	ret = clk_prepare_enable(msm_host->src_clk);
+	if (ret) {
+		pr_err("%s: Failed to enable dsi src clk\n", __func__);
+		goto src_clk_err;
+	}
+
+	ret = clk_prepare_enable(msm_host->pixel_clk);
+	if (ret) {
+		pr_err("%s: Failed to enable dsi pixel clk\n", __func__);
+		goto pixel_clk_err;
+	}
+
+	return 0;
+
+pixel_clk_err:
+	clk_disable_unprepare(msm_host->src_clk);
+src_clk_err:
+	clk_disable_unprepare(msm_host->esc_clk);
+esc_clk_err:
+	clk_disable_unprepare(msm_host->byte_clk);
+error:
+	return ret;
+}
+
+static int dsi_link_clk_enable(struct msm_dsi_host *msm_host)
+{
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
+
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G)
+		return dsi_link_clk_enable_6g(msm_host);
+	else
+		return dsi_link_clk_enable_v2(msm_host);
+}
+
 static void dsi_link_clk_disable(struct msm_dsi_host *msm_host)
 {
-	clk_disable_unprepare(msm_host->esc_clk);
-	clk_disable_unprepare(msm_host->pixel_clk);
-	clk_disable_unprepare(msm_host->byte_clk);
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
+
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) {
+		clk_disable_unprepare(msm_host->esc_clk);
+		clk_disable_unprepare(msm_host->pixel_clk);
+		clk_disable_unprepare(msm_host->byte_clk);
+	} else {
+		clk_disable_unprepare(msm_host->pixel_clk);
+		clk_disable_unprepare(msm_host->src_clk);
+		clk_disable_unprepare(msm_host->esc_clk);
+		clk_disable_unprepare(msm_host->byte_clk);
+	}
 }
 
 static int dsi_clk_ctrl(struct msm_dsi_host *msm_host, bool enable)
@@ -515,6 +634,7 @@ static int dsi_clk_ctrl(struct msm_dsi_host *msm_host, bool enable)
 static int dsi_calc_clk_rate(struct msm_dsi_host *msm_host)
 {
 	struct drm_display_mode *mode = msm_host->mode;
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	u8 lanes = msm_host->lanes;
 	u32 bpp = dsi_get_bpp(msm_host->format);
 	u32 pclk_rate;
@@ -534,6 +654,47 @@ static int dsi_calc_clk_rate(struct msm_dsi_host *msm_host)
 
 	DBG("pclk=%d, bclk=%d", pclk_rate, msm_host->byte_clk_rate);
 
+	msm_host->esc_clk_rate = clk_get_rate(msm_host->esc_clk);
+
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) {
+		unsigned int esc_mhz, esc_div;
+		unsigned long byte_mhz;
+
+		msm_host->src_clk_rate = (pclk_rate * bpp) / 8;
+
+		/*
+		 * esc clock is byte clock followed by a 4 bit divider,
+		 * we need to find an escape clock frequency within the
+		 * mipi DSI spec range within the maximum divider limit
+		 * We iterate here between an escape clock frequencey
+		 * between 20 Mhz to 5 Mhz and pick up the first one
+		 * that can be supported by our divider
+		 */
+
+		byte_mhz = msm_host->byte_clk_rate / 1000000;
+
+		for (esc_mhz = 20; esc_mhz >= 5; esc_mhz--) {
+			esc_div = DIV_ROUND_UP(byte_mhz, esc_mhz);
+
+			/*
+			 * TODO: Ideally, we shouldn't know what sort of divider
+			 * is available in mmss_cc, we're just assuming that
+			 * it'll always be a 4 bit divider. Need to come up with
+			 * a better way here.
+			 */
+			if (esc_div >= 1 && esc_div <= 16)
+				break;
+		}
+
+		if (esc_mhz < 5)
+			return -EINVAL;
+
+		msm_host->esc_clk_rate = msm_host->byte_clk_rate / esc_div;
+
+		DBG("esc=%d, src=%d", msm_host->esc_clk_rate,
+			msm_host->src_clk_rate);
+	}
+
 	return 0;
 }
 
@@ -835,29 +996,46 @@ static void dsi_wait4video_eng_busy(struct msm_dsi_host *msm_host)
 static int dsi_tx_buf_alloc(struct msm_dsi_host *msm_host, int size)
 {
 	struct drm_device *dev = msm_host->dev;
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	int ret;
 	u32 iova;
 
-	mutex_lock(&dev->struct_mutex);
-	msm_host->tx_gem_obj = msm_gem_new(dev, size, MSM_BO_UNCACHED);
-	if (IS_ERR(msm_host->tx_gem_obj)) {
-		ret = PTR_ERR(msm_host->tx_gem_obj);
-		pr_err("%s: failed to allocate gem, %d\n", __func__, ret);
-		msm_host->tx_gem_obj = NULL;
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) {
+		mutex_lock(&dev->struct_mutex);
+		msm_host->tx_gem_obj = msm_gem_new(dev, size, MSM_BO_UNCACHED);
+		if (IS_ERR(msm_host->tx_gem_obj)) {
+			ret = PTR_ERR(msm_host->tx_gem_obj);
+			pr_err("%s: failed to allocate gem, %d\n",
+				__func__, ret);
+			msm_host->tx_gem_obj = NULL;
+			mutex_unlock(&dev->struct_mutex);
+			return ret;
+		}
+
+		ret = msm_gem_get_iova_locked(msm_host->tx_gem_obj, 0, &iova);
 		mutex_unlock(&dev->struct_mutex);
-		return ret;
-	}
+		if (ret) {
+			pr_err("%s: failed to get iova, %d\n", __func__, ret);
+			return ret;
+		}
 
-	ret = msm_gem_get_iova_locked(msm_host->tx_gem_obj, 0, &iova);
-	if (ret) {
-		pr_err("%s: failed to get iova, %d\n", __func__, ret);
-		return ret;
-	}
-	mutex_unlock(&dev->struct_mutex);
+		if (iova & 0x07) {
+			pr_err("%s: buf NOT 8 bytes aligned\n", __func__);
+			return -EINVAL;
+		}
 
-	if (iova & 0x07) {
-		pr_err("%s: buf NOT 8 bytes aligned\n", __func__);
-		return -EINVAL;
+		msm_host->tx_size = msm_host->tx_gem_obj->size;
+	} else {
+		msm_host->tx_buf = dma_alloc_coherent(dev->dev, size,
+					&msm_host->tx_buf_paddr, GFP_KERNEL);
+		if (!msm_host->tx_buf) {
+			ret = -ENOMEM;
+			pr_err("%s: failed to allocate tx buf, %d\n",
+				__func__, ret);
+			return ret;
+		}
+
+		msm_host->tx_size = size;
 	}
 
 	return 0;
@@ -874,14 +1052,19 @@ static void dsi_tx_buf_free(struct msm_dsi_host *msm_host)
 		msm_host->tx_gem_obj = NULL;
 		mutex_unlock(&dev->struct_mutex);
 	}
+
+	if (msm_host->tx_buf)
+		dma_free_coherent(dev->dev, msm_host->tx_size, msm_host->tx_buf,
+			msm_host->tx_buf_paddr);
 }
 
 /*
  * prepare cmd buffer to be txed
  */
-static int dsi_cmd_dma_add(struct drm_gem_object *tx_gem,
-			const struct mipi_dsi_msg *msg)
+static int dsi_cmd_dma_add(struct msm_dsi_host *msm_host,
+			   const struct mipi_dsi_msg *msg)
 {
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	struct mipi_dsi_packet packet;
 	int len;
 	int ret;
@@ -894,17 +1077,20 @@ static int dsi_cmd_dma_add(struct drm_gem_object *tx_gem,
 	}
 	len = (packet.size + 3) & (~0x3);
 
-	if (len > tx_gem->size) {
+	if (len > msm_host->tx_size) {
 		pr_err("%s: packet size is too big\n", __func__);
 		return -EINVAL;
 	}
 
-	data = msm_gem_vaddr(tx_gem);
-
-	if (IS_ERR(data)) {
-		ret = PTR_ERR(data);
-		pr_err("%s: get vaddr failed, %d\n", __func__, ret);
-		return ret;
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) {
+		data = msm_gem_vaddr(msm_host->tx_gem_obj);
+		if (IS_ERR(data)) {
+			ret = PTR_ERR(data);
+			pr_err("%s: get vaddr failed, %d\n", __func__, ret);
+			return ret;
+		}
+	} else {
+		data = msm_host->tx_buf;
 	}
 
 	/* MSM specific command format in memory */
@@ -970,17 +1156,21 @@ static int dsi_long_read_resp(u8 *buf, const struct mipi_dsi_msg *msg)
 	return msg->rx_len;
 }
 
-
 static int dsi_cmd_dma_tx(struct msm_dsi_host *msm_host, int len)
 {
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	int ret;
-	u32 iova;
+	u32 dma_base;
 	bool triggered;
 
-	ret = msm_gem_get_iova(msm_host->tx_gem_obj, 0, &iova);
-	if (ret) {
-		pr_err("%s: failed to get iova: %d\n", __func__, ret);
-		return ret;
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_6G) {
+		ret = msm_gem_get_iova(msm_host->tx_gem_obj, 0, &dma_base);
+		if (ret) {
+			pr_err("%s: failed to get iova: %d\n", __func__, ret);
+			return ret;
+		}
+	} else {
+		dma_base = msm_host->tx_buf_paddr;
 	}
 
 	reinit_completion(&msm_host->dma_comp);
@@ -988,7 +1178,7 @@ static int dsi_cmd_dma_tx(struct msm_dsi_host *msm_host, int len)
 	dsi_wait4video_eng_busy(msm_host);
 
 	triggered = msm_dsi_manager_cmd_xfer_trigger(
-						msm_host->id, iova, len);
+						msm_host->id, dma_base, len);
 	if (triggered) {
 		ret = wait_for_completion_timeout(&msm_host->dma_comp,
 					msecs_to_jiffies(200));
@@ -1060,7 +1250,7 @@ static int dsi_cmds2buf_tx(struct msm_dsi_host *msm_host,
 	int bllp_len = msm_host->mode->hdisplay *
 			dsi_get_bpp(msm_host->format) / 8;
 
-	len = dsi_cmd_dma_add(msm_host->tx_gem_obj, msg);
+	len = dsi_cmd_dma_add(msm_host, msg);
 	if (!len) {
 		pr_err("%s: failed to add cmd type = 0x%x\n",
 			__func__,  msg->type);
@@ -1383,6 +1573,16 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host)
 
 	msm_host->device_node = device_node;
 
+	if (of_property_read_bool(np, "syscon-sfpb")) {
+		msm_host->sfpb = syscon_regmap_lookup_by_phandle(np,
+					"syscon-sfpb");
+		if (IS_ERR(msm_host->sfpb)) {
+			dev_err(dev, "%s: failed to get sfpb regmap\n",
+				__func__);
+			return PTR_ERR(msm_host->sfpb);
+		}
+	}
+
 	return 0;
 }
 
@@ -1408,12 +1608,6 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 		goto fail;
 	}
 
-	ret = dsi_clk_init(msm_host);
-	if (ret) {
-		pr_err("%s: unable to initialize dsi clks\n", __func__);
-		goto fail;
-	}
-
 	msm_host->ctrl_base = msm_ioremap(pdev, "dsi_ctrl", "DSI CTRL");
 	if (IS_ERR(msm_host->ctrl_base)) {
 		pr_err("%s: unable to map Dsi ctrl base\n", __func__);
@@ -1437,6 +1631,12 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi)
 		goto fail;
 	}
 
+	ret = dsi_clk_init(msm_host);
+	if (ret) {
+		pr_err("%s: unable to initialize dsi clks\n", __func__);
+		goto fail;
+	}
+
 	msm_host->rx_buf = devm_kzalloc(&pdev->dev, SZ_4K, GFP_KERNEL);
 	if (!msm_host->rx_buf) {
 		pr_err("%s: alloc rx temp buf failed\n", __func__);
@@ -1750,11 +1950,12 @@ int msm_dsi_host_cmd_rx(struct mipi_dsi_host *host,
 	return ret;
 }
 
-void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, u32 iova, u32 len)
+void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host, u32 dma_base,
+				  u32 len)
 {
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
 
-	dsi_write(msm_host, REG_DSI_DMA_BASE, iova);
+	dsi_write(msm_host, REG_DSI_DMA_BASE, dma_base);
 	dsi_write(msm_host, REG_DSI_DMA_LEN, len);
 	dsi_write(msm_host, REG_DSI_TRIG_DMA, 1);
 
@@ -1766,6 +1967,7 @@ int msm_dsi_host_set_src_pll(struct mipi_dsi_host *host,
 	struct msm_dsi_pll *src_pll)
 {
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
+	const struct msm_dsi_cfg_handler *cfg_hnd = msm_host->cfg_hnd;
 	struct clk *byte_clk_provider, *pixel_clk_provider;
 	int ret;
 
@@ -1791,6 +1993,22 @@ int msm_dsi_host_set_src_pll(struct mipi_dsi_host *host,
 		goto exit;
 	}
 
+	if (cfg_hnd->major == MSM_DSI_VER_MAJOR_V2) {
+		ret = clk_set_parent(msm_host->dsi_clk_src, pixel_clk_provider);
+		if (ret) {
+			pr_err("%s: can't set parent to dsi_clk_src. ret=%d\n",
+				__func__, ret);
+			goto exit;
+		}
+
+		ret = clk_set_parent(msm_host->esc_clk_src, byte_clk_provider);
+		if (ret) {
+			pr_err("%s: can't set parent to esc_clk_src. ret=%d\n",
+				__func__, ret);
+			goto exit;
+		}
+	}
+
 exit:
 	return ret;
 }
@@ -1828,6 +2046,20 @@ int msm_dsi_host_disable(struct mipi_dsi_host *host)
 	return 0;
 }
 
+static void msm_dsi_sfpb_config(struct msm_dsi_host *msm_host, bool enable)
+{
+	enum sfpb_ahb_arb_master_port_en en;
+
+	if (!msm_host->sfpb)
+		return;
+
+	en = enable ? SFPB_MASTER_PORT_ENABLE : SFPB_MASTER_PORT_DISABLE;
+
+	regmap_update_bits(msm_host->sfpb, REG_SFPB_GPREG,
+			SFPB_GPREG_MASTER_PORT_EN__MASK,
+			SFPB_GPREG_MASTER_PORT_EN(en));
+}
+
 int msm_dsi_host_power_on(struct mipi_dsi_host *host)
 {
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
@@ -1840,6 +2072,8 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host)
 		goto unlock_ret;
 	}
 
+	msm_dsi_sfpb_config(msm_host, true);
+
 	ret = dsi_calc_clk_rate(msm_host);
 	if (ret) {
 		pr_err("%s: unable to calc clk rate, %d\n", __func__, ret);
@@ -1862,7 +2096,7 @@ int msm_dsi_host_power_on(struct mipi_dsi_host *host)
 	dsi_phy_sw_reset(msm_host);
 	ret = msm_dsi_manager_phy_enable(msm_host->id,
 					msm_host->byte_clk_rate * 8,
-					clk_get_rate(msm_host->esc_clk),
+					msm_host->esc_clk_rate,
 					&clk_pre, &clk_post);
 	dsi_bus_clk_disable(msm_host);
 	if (ret) {
@@ -1927,6 +2161,8 @@ int msm_dsi_host_power_off(struct mipi_dsi_host *host)
 
 	dsi_host_regulator_disable(msm_host);
 
+	msm_dsi_sfpb_config(msm_host, false);
+
 	DBG("-");
 
 	msm_host->power_on = false;
diff --git a/drivers/gpu/drm/msm/dsi/dsi_manager.c b/drivers/gpu/drm/msm/dsi/dsi_manager.c
index 0455ff75074a..58ba7ec17f51 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_manager.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_manager.c
@@ -774,7 +774,7 @@ int msm_dsi_manager_cmd_xfer(int id, const struct mipi_dsi_msg *msg)
 	return ret;
 }
 
-bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len)
+bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 dma_base, u32 len)
 {
 	struct msm_dsi *msm_dsi = dsi_mgr_get_dsi(id);
 	struct msm_dsi *msm_dsi0 = dsi_mgr_get_dsi(DSI_0);
@@ -784,9 +784,9 @@ bool msm_dsi_manager_cmd_xfer_trigger(int id, u32 iova, u32 len)
 		return false;
 
 	if (IS_SYNC_NEEDED() && msm_dsi0)
-		msm_dsi_host_cmd_xfer_commit(msm_dsi0->host, iova, len);
+		msm_dsi_host_cmd_xfer_commit(msm_dsi0->host, dma_base, len);
 
-	msm_dsi_host_cmd_xfer_commit(host, iova, len);
+	msm_dsi_host_cmd_xfer_commit(host, dma_base, len);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index f1f955f571fa..91a95fb04a4a 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -276,6 +276,10 @@ static const struct of_device_id dsi_phy_dt_match[] = {
 #ifdef CONFIG_DRM_MSM_DSI_20NM_PHY
 	{ .compatible = "qcom,dsi-phy-20nm",
 	  .data = &dsi_phy_20nm_cfgs },
+#endif
+#ifdef CONFIG_DRM_MSM_DSI_28NM_8960_PHY
+	{ .compatible = "qcom,dsi-phy-28nm-8960",
+	  .data = &dsi_phy_28nm_8960_cfgs },
 #endif
 	{}
 };
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
index 0456b253239f..0d54ed00386d 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
@@ -43,6 +43,7 @@ struct msm_dsi_phy_cfg {
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_hpm_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_28nm_lp_cfgs;
 extern const struct msm_dsi_phy_cfg dsi_phy_20nm_cfgs;
+extern const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs;
 
 struct msm_dsi_dphy_timing {
 	u32 clk_pre;
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c
new file mode 100644
index 000000000000..197b039ca1f1
--- /dev/null
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c
@@ -0,0 +1,195 @@
+/*
+ * Copyright (c) 2012-2015, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include "dsi_phy.h"
+#include "dsi.xml.h"
+
+static void dsi_28nm_dphy_set_timing(struct msm_dsi_phy *phy,
+		struct msm_dsi_dphy_timing *timing)
+{
+	void __iomem *base = phy->base;
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_0,
+		DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO(timing->clk_zero));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_1,
+		DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL(timing->clk_trail));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_2,
+		DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE(timing->clk_prepare));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_3, 0x0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_4,
+		DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT(timing->hs_exit));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_5,
+		DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO(timing->hs_zero));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_6,
+		DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE(timing->hs_prepare));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_7,
+		DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL(timing->hs_trail));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_8,
+		DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST(timing->hs_rqst));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_9,
+		DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO(timing->ta_go) |
+		DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE(timing->ta_sure));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_10,
+		DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET(timing->ta_get));
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_TIMING_CTRL_11,
+		DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD(0));
+}
+
+static void dsi_28nm_phy_regulator_init(struct msm_dsi_phy *phy)
+{
+	void __iomem *base = phy->reg_base;
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_0, 0x3);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_1, 1);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_2, 1);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_3, 0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_4,
+		0x100);
+}
+
+static void dsi_28nm_phy_regulator_ctrl(struct msm_dsi_phy *phy)
+{
+	void __iomem *base = phy->reg_base;
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_0, 0x3);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_1, 0xa);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_2, 0x4);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_3, 0x0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_4, 0x20);
+}
+
+static void dsi_28nm_phy_calibration(struct msm_dsi_phy *phy)
+{
+	void __iomem *base = phy->reg_base;
+	u32 status;
+	int i = 5000;
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CAL_PWR_CFG,
+			0x3);
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_SW_CFG_2, 0x0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_1, 0x5a);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_3, 0x10);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_4, 0x1);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_0, 0x1);
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_TRIGGER, 0x1);
+	usleep_range(5000, 6000);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_MISC_CAL_HW_TRIGGER, 0x0);
+
+	do {
+		status = dsi_phy_read(base +
+				REG_DSI_28nm_8960_PHY_MISC_CAL_STATUS);
+
+		if (!(status & DSI_28nm_8960_PHY_MISC_CAL_STATUS_CAL_BUSY))
+			break;
+
+		udelay(1);
+	} while (--i > 0);
+}
+
+static void dsi_28nm_phy_lane_config(struct msm_dsi_phy *phy)
+{
+	void __iomem *base = phy->base;
+	int i;
+
+	for (i = 0; i < 4; i++) {
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_0(i), 0x80);
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_1(i), 0x45);
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_CFG_2(i), 0x00);
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_DATAPATH(i),
+			0x00);
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_STR_0(i),
+			0x01);
+		dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LN_TEST_STR_1(i),
+			0x66);
+	}
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_0, 0x40);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_1, 0x67);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_CFG_2, 0x0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_DATAPATH, 0x0);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_STR0, 0x1);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LNCK_TEST_STR1, 0x88);
+}
+
+static int dsi_28nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id,
+		const unsigned long bit_rate, const unsigned long esc_rate)
+{
+	struct msm_dsi_dphy_timing *timing = &phy->timing;
+	void __iomem *base = phy->base;
+
+	DBG("");
+
+	if (msm_dsi_dphy_timing_calc(timing, bit_rate, esc_rate)) {
+		dev_err(&phy->pdev->dev,
+			"%s: D-PHY timing calculation failed\n", __func__);
+		return -EINVAL;
+	}
+
+	dsi_28nm_phy_regulator_init(phy);
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_LDO_CTRL, 0x04);
+
+	/* strength control */
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_0, 0xff);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_1, 0x00);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_STRENGTH_2, 0x06);
+
+	/* phy ctrl */
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_0, 0x5f);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_1, 0x00);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_2, 0x00);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_CTRL_3, 0x10);
+
+	dsi_28nm_phy_regulator_ctrl(phy);
+
+	dsi_28nm_phy_calibration(phy);
+
+	dsi_28nm_phy_lane_config(phy);
+
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_4, 0x0f);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_1, 0x03);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_0, 0x03);
+	dsi_phy_write(base + REG_DSI_28nm_8960_PHY_BIST_CTRL_4, 0x0);
+
+	dsi_28nm_dphy_set_timing(phy, timing);
+
+	return 0;
+}
+
+static void dsi_28nm_phy_disable(struct msm_dsi_phy *phy)
+{
+	dsi_phy_write(phy->base + REG_DSI_28nm_8960_PHY_CTRL_0, 0x0);
+
+	/*
+	 * Wait for the registers writes to complete in order to
+	 * ensure that the phy is completely disabled
+	 */
+	wmb();
+}
+
+const struct msm_dsi_phy_cfg dsi_phy_28nm_8960_cfgs = {
+	.type = MSM_DSI_PHY_28NM_8960,
+	.src_pll_truthtable = { {true, true}, {false, true} },
+	.reg_cfg = {
+		.num = 1,
+		.regs = {
+			{"vddio", 1800000, 1800000, 100000, 100},
+		},
+	},
+	.ops = {
+		.enable = dsi_28nm_phy_enable,
+		.disable = dsi_28nm_phy_disable,
+	},
+};
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
index 5104fc9f9a53..5cd438f91afe 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c
@@ -151,6 +151,9 @@ struct msm_dsi_pll *msm_dsi_pll_init(struct platform_device *pdev,
 	case MSM_DSI_PHY_28NM_LP:
 		pll = msm_dsi_pll_28nm_init(pdev, type, id);
 		break;
+	case MSM_DSI_PHY_28NM_8960:
+		pll = msm_dsi_pll_28nm_8960_init(pdev, id);
+		break;
 	default:
 		pll = ERR_PTR(-ENXIO);
 		break;
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
index 063caa2c5740..80b6038334a6 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
@@ -93,6 +93,16 @@ static inline struct msm_dsi_pll *msm_dsi_pll_28nm_init(
 	return ERR_PTR(-ENODEV);
 }
 #endif
+#ifdef CONFIG_DRM_MSM_DSI_28NM_8960_PHY
+struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev,
+					       int id);
+#else
+struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev,
+					       int id)
+{
+	return ERR_PTR(-ENODEV);
+}
+#endif
 
 #endif /* __DSI_PLL_H__ */
 
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c
new file mode 100644
index 000000000000..38c90e1eb002
--- /dev/null
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c
@@ -0,0 +1,533 @@
+/*
+ * Copyright (c) 2012-2015, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk-provider.h>
+
+#include "dsi_pll.h"
+#include "dsi.xml.h"
+
+/*
+ * DSI PLL 28nm (8960/A family) - clock diagram (eg: DSI1):
+ *
+ *
+ *                        +------+
+ *  dsi1vco_clk ----o-----| DIV1 |---dsi1pllbit (not exposed as clock)
+ *  F * byte_clk    |     +------+
+ *                  | bit clock divider (F / 8)
+ *                  |
+ *                  |     +------+
+ *                  o-----| DIV2 |---dsi0pllbyte---o---> To byte RCG
+ *                  |     +------+                 | (sets parent rate)
+ *                  | byte clock divider (F)       |
+ *                  |                              |
+ *                  |                              o---> To esc RCG
+ *                  |                                (doesn't set parent rate)
+ *                  |
+ *                  |     +------+
+ *                  o-----| DIV3 |----dsi0pll------o---> To dsi RCG
+ *                        +------+                 | (sets parent rate)
+ *                  dsi clock divider (F * magic)  |
+ *                                                 |
+ *                                                 o---> To pixel rcg
+ *                                                  (doesn't set parent rate)
+ */
+
+#define POLL_MAX_READS		8000
+#define POLL_TIMEOUT_US		1
+
+#define NUM_PROVIDED_CLKS	2
+
+#define VCO_REF_CLK_RATE	27000000
+#define VCO_MIN_RATE		600000000
+#define VCO_MAX_RATE		1200000000
+
+#define DSI_BYTE_PLL_CLK	0
+#define DSI_PIXEL_PLL_CLK	1
+
+#define VCO_PREF_DIV_RATIO	27
+
+struct pll_28nm_cached_state {
+	unsigned long vco_rate;
+	u8 postdiv3;
+	u8 postdiv2;
+	u8 postdiv1;
+};
+
+struct clk_bytediv {
+	struct clk_hw hw;
+	void __iomem *reg;
+};
+
+struct dsi_pll_28nm {
+	struct msm_dsi_pll base;
+
+	int id;
+	struct platform_device *pdev;
+	void __iomem *mmio;
+
+	/* custom byte clock divider */
+	struct clk_bytediv *bytediv;
+
+	/* private clocks: */
+	struct clk *clks[NUM_DSI_CLOCKS_MAX];
+	u32 num_clks;
+
+	/* clock-provider: */
+	struct clk *provided_clks[NUM_PROVIDED_CLKS];
+	struct clk_onecell_data clk_data;
+
+	struct pll_28nm_cached_state cached_state;
+};
+
+#define to_pll_28nm(x)	container_of(x, struct dsi_pll_28nm, base)
+
+static bool pll_28nm_poll_for_ready(struct dsi_pll_28nm *pll_28nm,
+				    int nb_tries, int timeout_us)
+{
+	bool pll_locked = false;
+	u32 val;
+
+	while (nb_tries--) {
+		val = pll_read(pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_RDY);
+		pll_locked = !!(val & DSI_28nm_8960_PHY_PLL_RDY_PLL_RDY);
+
+		if (pll_locked)
+			break;
+
+		udelay(timeout_us);
+	}
+	DBG("DSI PLL is %slocked", pll_locked ? "" : "*not* ");
+
+	return pll_locked;
+}
+
+/*
+ * Clock Callbacks
+ */
+static int dsi_pll_28nm_clk_set_rate(struct clk_hw *hw, unsigned long rate,
+				     unsigned long parent_rate)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+	void __iomem *base = pll_28nm->mmio;
+	u32 val, temp, fb_divider;
+
+	DBG("rate=%lu, parent's=%lu", rate, parent_rate);
+
+	temp = rate / 10;
+	val = VCO_REF_CLK_RATE / 10;
+	fb_divider = (temp * VCO_PREF_DIV_RATIO) / val;
+	fb_divider = fb_divider / 2 - 1;
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_1,
+			fb_divider & 0xff);
+
+	val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2);
+
+	val |= (fb_divider >> 8) & 0x07;
+
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2,
+			val);
+
+	val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3);
+
+	val |= (VCO_PREF_DIV_RATIO - 1) & 0x3f;
+
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3,
+			val);
+
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_6,
+			0xf);
+
+	val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8);
+	val |= 0x7 << 4;
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8,
+			val);
+
+	return 0;
+}
+
+static int dsi_pll_28nm_clk_is_enabled(struct clk_hw *hw)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+
+	return pll_28nm_poll_for_ready(pll_28nm, POLL_MAX_READS,
+					POLL_TIMEOUT_US);
+}
+
+static unsigned long dsi_pll_28nm_clk_recalc_rate(struct clk_hw *hw,
+						  unsigned long parent_rate)
+{
+	struct msm_dsi_pll *pll = hw_clk_to_pll(hw);
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+	void __iomem *base = pll_28nm->mmio;
+	unsigned long vco_rate;
+	u32 status, fb_divider, temp, ref_divider;
+
+	VERB("parent_rate=%lu", parent_rate);
+
+	status = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_0);
+
+	if (status & DSI_28nm_8960_PHY_PLL_CTRL_0_ENABLE) {
+		fb_divider = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_1);
+		fb_divider &= 0xff;
+		temp = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_2) & 0x07;
+		fb_divider = (temp << 8) | fb_divider;
+		fb_divider += 1;
+
+		ref_divider = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_3);
+		ref_divider &= 0x3f;
+		ref_divider += 1;
+
+		/* multiply by 2 */
+		vco_rate = (parent_rate / ref_divider) * fb_divider * 2;
+	} else {
+		vco_rate = 0;
+	}
+
+	DBG("returning vco rate = %lu", vco_rate);
+
+	return vco_rate;
+}
+
+static const struct clk_ops clk_ops_dsi_pll_28nm_vco = {
+	.round_rate = msm_dsi_pll_helper_clk_round_rate,
+	.set_rate = dsi_pll_28nm_clk_set_rate,
+	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
+	.prepare = msm_dsi_pll_helper_clk_prepare,
+	.unprepare = msm_dsi_pll_helper_clk_unprepare,
+	.is_enabled = dsi_pll_28nm_clk_is_enabled,
+};
+
+/*
+ * Custom byte clock divier clk_ops
+ *
+ * This clock is the entry point to configuring the PLL. The user (dsi host)
+ * will set this clock's rate to the desired byte clock rate. The VCO lock
+ * frequency is a multiple of the byte clock rate. The multiplication factor
+ * (shown as F in the diagram above) is a function of the byte clock rate.
+ *
+ * This custom divider clock ensures that its parent (VCO) is set to the
+ * desired rate, and that the byte clock postdivider (POSTDIV2) is configured
+ * accordingly
+ */
+#define to_clk_bytediv(_hw) container_of(_hw, struct clk_bytediv, hw)
+
+static unsigned long clk_bytediv_recalc_rate(struct clk_hw *hw,
+		unsigned long parent_rate)
+{
+	struct clk_bytediv *bytediv = to_clk_bytediv(hw);
+	unsigned int div;
+
+	div = pll_read(bytediv->reg) & 0xff;
+
+	return parent_rate / (div + 1);
+}
+
+/* find multiplication factor(wrt byte clock) at which the VCO should be set */
+static unsigned int get_vco_mul_factor(unsigned long byte_clk_rate)
+{
+	unsigned long bit_mhz;
+
+	/* convert to bit clock in Mhz */
+	bit_mhz = (byte_clk_rate * 8) / 1000000;
+
+	if (bit_mhz < 125)
+		return 64;
+	else if (bit_mhz < 250)
+		return 32;
+	else if (bit_mhz < 600)
+		return 16;
+	else
+		return 8;
+}
+
+static long clk_bytediv_round_rate(struct clk_hw *hw, unsigned long rate,
+				   unsigned long *prate)
+{
+	unsigned long best_parent;
+	unsigned int factor;
+
+	factor = get_vco_mul_factor(rate);
+
+	best_parent = rate * factor;
+	*prate = clk_hw_round_rate(clk_hw_get_parent(hw), best_parent);
+
+	return *prate / factor;
+}
+
+static int clk_bytediv_set_rate(struct clk_hw *hw, unsigned long rate,
+				unsigned long parent_rate)
+{
+	struct clk_bytediv *bytediv = to_clk_bytediv(hw);
+	u32 val;
+	unsigned int factor;
+
+	factor = get_vco_mul_factor(rate);
+
+	val = pll_read(bytediv->reg);
+	val |= (factor - 1) & 0xff;
+	pll_write(bytediv->reg, val);
+
+	return 0;
+}
+
+/* Our special byte clock divider ops */
+static const struct clk_ops clk_bytediv_ops = {
+	.round_rate = clk_bytediv_round_rate,
+	.set_rate = clk_bytediv_set_rate,
+	.recalc_rate = clk_bytediv_recalc_rate,
+};
+
+/*
+ * PLL Callbacks
+ */
+static int dsi_pll_28nm_enable_seq(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+	struct device *dev = &pll_28nm->pdev->dev;
+	void __iomem *base = pll_28nm->mmio;
+	bool locked;
+	unsigned int bit_div, byte_div;
+	int max_reads = 1000, timeout_us = 100;
+	u32 val;
+
+	DBG("id=%d", pll_28nm->id);
+
+	/*
+	 * before enabling the PLL, configure the bit clock divider since we
+	 * don't expose it as a clock to the outside world
+	 * 1: read back the byte clock divider that should already be set
+	 * 2: divide by 8 to get bit clock divider
+	 * 3: write it to POSTDIV1
+	 */
+	val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9);
+	byte_div = val + 1;
+	bit_div = byte_div / 8;
+
+	val = pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8);
+	val &= ~0xf;
+	val |= (bit_div - 1);
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8, val);
+
+	/* enable the PLL */
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_0,
+			DSI_28nm_8960_PHY_PLL_CTRL_0_ENABLE);
+
+	locked = pll_28nm_poll_for_ready(pll_28nm, max_reads, timeout_us);
+
+	if (unlikely(!locked))
+		dev_err(dev, "DSI PLL lock failed\n");
+	else
+		DBG("DSI PLL lock success");
+
+	return locked ? 0 : -EINVAL;
+}
+
+static void dsi_pll_28nm_disable_seq(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+
+	DBG("id=%d", pll_28nm->id);
+	pll_write(pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_CTRL_0, 0x00);
+}
+
+static void dsi_pll_28nm_save_state(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+	struct pll_28nm_cached_state *cached_state = &pll_28nm->cached_state;
+	void __iomem *base = pll_28nm->mmio;
+
+	cached_state->postdiv3 =
+			pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_10);
+	cached_state->postdiv2 =
+			pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9);
+	cached_state->postdiv1 =
+			pll_read(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8);
+
+	cached_state->vco_rate = clk_hw_get_rate(&pll->clk_hw);
+}
+
+static int dsi_pll_28nm_restore_state(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+	struct pll_28nm_cached_state *cached_state = &pll_28nm->cached_state;
+	void __iomem *base = pll_28nm->mmio;
+	int ret;
+
+	ret = dsi_pll_28nm_clk_set_rate(&pll->clk_hw,
+					cached_state->vco_rate, 0);
+	if (ret) {
+		dev_err(&pll_28nm->pdev->dev,
+			"restore vco rate failed. ret=%d\n", ret);
+		return ret;
+	}
+
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_10,
+			cached_state->postdiv3);
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_9,
+			cached_state->postdiv2);
+	pll_write(base + REG_DSI_28nm_8960_PHY_PLL_CTRL_8,
+			cached_state->postdiv1);
+
+	return 0;
+}
+
+static int dsi_pll_28nm_get_provider(struct msm_dsi_pll *pll,
+				struct clk **byte_clk_provider,
+				struct clk **pixel_clk_provider)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+
+	if (byte_clk_provider)
+		*byte_clk_provider = pll_28nm->provided_clks[DSI_BYTE_PLL_CLK];
+	if (pixel_clk_provider)
+		*pixel_clk_provider =
+				pll_28nm->provided_clks[DSI_PIXEL_PLL_CLK];
+
+	return 0;
+}
+
+static void dsi_pll_28nm_destroy(struct msm_dsi_pll *pll)
+{
+	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(pll);
+
+	msm_dsi_pll_helper_unregister_clks(pll_28nm->pdev,
+					pll_28nm->clks, pll_28nm->num_clks);
+}
+
+static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm)
+{
+	char *clk_name, *parent_name, *vco_name;
+	struct clk_init_data vco_init = {
+		.parent_names = (const char *[]){ "pxo" },
+		.num_parents = 1,
+		.ops = &clk_ops_dsi_pll_28nm_vco,
+	};
+	struct device *dev = &pll_28nm->pdev->dev;
+	struct clk **clks = pll_28nm->clks;
+	struct clk **provided_clks = pll_28nm->provided_clks;
+	struct clk_bytediv *bytediv;
+	struct clk_init_data bytediv_init = { };
+	int ret, num = 0;
+
+	DBG("%d", pll_28nm->id);
+
+	bytediv = devm_kzalloc(dev, sizeof(*bytediv), GFP_KERNEL);
+	if (!bytediv)
+		return -ENOMEM;
+
+	vco_name = devm_kzalloc(dev, 32, GFP_KERNEL);
+	if (!vco_name)
+		return -ENOMEM;
+
+	parent_name = devm_kzalloc(dev, 32, GFP_KERNEL);
+	if (!parent_name)
+		return -ENOMEM;
+
+	clk_name = devm_kzalloc(dev, 32, GFP_KERNEL);
+	if (!clk_name)
+		return -ENOMEM;
+
+	pll_28nm->bytediv = bytediv;
+
+	snprintf(vco_name, 32, "dsi%dvco_clk", pll_28nm->id);
+	vco_init.name = vco_name;
+
+	pll_28nm->base.clk_hw.init = &vco_init;
+
+	clks[num++] = clk_register(dev, &pll_28nm->base.clk_hw);
+
+	/* prepare and register bytediv */
+	bytediv->hw.init = &bytediv_init;
+	bytediv->reg = pll_28nm->mmio + REG_DSI_28nm_8960_PHY_PLL_CTRL_9;
+
+	snprintf(parent_name, 32, "dsi%dvco_clk", pll_28nm->id);
+	snprintf(clk_name, 32, "dsi%dpllbyte", pll_28nm->id);
+
+	bytediv_init.name = clk_name;
+	bytediv_init.ops = &clk_bytediv_ops;
+	bytediv_init.flags = CLK_SET_RATE_PARENT;
+	bytediv_init.parent_names = (const char * const *) &parent_name;
+	bytediv_init.num_parents = 1;
+
+	/* DIV2 */
+	clks[num++] = provided_clks[DSI_BYTE_PLL_CLK] =
+			clk_register(dev, &bytediv->hw);
+
+	snprintf(clk_name, 32, "dsi%dpll", pll_28nm->id);
+	/* DIV3 */
+	clks[num++] = provided_clks[DSI_PIXEL_PLL_CLK] =
+			clk_register_divider(dev, clk_name,
+				parent_name, 0, pll_28nm->mmio +
+				REG_DSI_28nm_8960_PHY_PLL_CTRL_10,
+				0, 8, 0, NULL);
+
+	pll_28nm->num_clks = num;
+
+	pll_28nm->clk_data.clk_num = NUM_PROVIDED_CLKS;
+	pll_28nm->clk_data.clks = provided_clks;
+
+	ret = of_clk_add_provider(dev->of_node,
+			of_clk_src_onecell_get, &pll_28nm->clk_data);
+	if (ret) {
+		dev_err(dev, "failed to register clk provider: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev,
+					       int id)
+{
+	struct dsi_pll_28nm *pll_28nm;
+	struct msm_dsi_pll *pll;
+	int ret;
+
+	if (!pdev)
+		return ERR_PTR(-ENODEV);
+
+	pll_28nm = devm_kzalloc(&pdev->dev, sizeof(*pll_28nm), GFP_KERNEL);
+	if (!pll_28nm)
+		return ERR_PTR(-ENOMEM);
+
+	pll_28nm->pdev = pdev;
+	pll_28nm->id = id + 1;
+
+	pll_28nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL");
+	if (IS_ERR_OR_NULL(pll_28nm->mmio)) {
+		dev_err(&pdev->dev, "%s: failed to map pll base\n", __func__);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	pll = &pll_28nm->base;
+	pll->min_rate = VCO_MIN_RATE;
+	pll->max_rate = VCO_MAX_RATE;
+	pll->get_provider = dsi_pll_28nm_get_provider;
+	pll->destroy = dsi_pll_28nm_destroy;
+	pll->disable_seq = dsi_pll_28nm_disable_seq;
+	pll->save_state = dsi_pll_28nm_save_state;
+	pll->restore_state = dsi_pll_28nm_restore_state;
+
+	pll->en_seq_cnt = 1;
+	pll->enable_seqs[0] = dsi_pll_28nm_enable_seq;
+
+	ret = pll_28nm_register(pll_28nm);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register PLL: %d\n", ret);
+		return ERR_PTR(ret);
+	}
+
+	return pll;
+}
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 1f4a95eeb348..9a0989c0b4de 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -17,6 +17,8 @@
  */
 
 #include <linux/of_irq.h>
+#include <linux/of_gpio.h>
+
 #include "hdmi.h"
 
 void hdmi_set_mode(struct hdmi *hdmi, bool power_on)
@@ -322,8 +324,6 @@ int hdmi_modeset_init(struct hdmi *hdmi,
  * The hdmi device:
  */
 
-#include <linux/of_gpio.h>
-
 #define HDMI_CFG(item, entry) \
 	.item ## _names = item ##_names_ ## entry, \
 	.item ## _cnt   = ARRAY_SIZE(item ## _names_ ## entry)
@@ -388,17 +388,6 @@ static struct hdmi_platform_config hdmi_tx_8996_config = {
 		.hpd_freq      = hpd_clk_freq_8x74,
 };
 
-static const struct of_device_id dt_match[] = {
-	{ .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config },
-	{ .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config },
-	{ .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config },
-	{ .compatible = "qcom,hdmi-tx-8974", .data = &hdmi_tx_8974_config },
-	{ .compatible = "qcom,hdmi-tx-8960", .data = &hdmi_tx_8960_config },
-	{ .compatible = "qcom,hdmi-tx-8660", .data = &hdmi_tx_8660_config },
-	{}
-};
-
-#ifdef CONFIG_OF
 static int get_gpio(struct device *dev, struct device_node *of_node, const char *name)
 {
 	int gpio = of_get_named_gpio(of_node, name, 0);
@@ -413,7 +402,6 @@ static int get_gpio(struct device *dev, struct device_node *of_node, const char
 	}
 	return gpio;
 }
-#endif
 
 static int hdmi_bind(struct device *dev, struct device *master, void *data)
 {
@@ -421,16 +409,12 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	struct msm_drm_private *priv = drm->dev_private;
 	static struct hdmi_platform_config *hdmi_cfg;
 	struct hdmi *hdmi;
-#ifdef CONFIG_OF
 	struct device_node *of_node = dev->of_node;
-	const struct of_device_id *match;
 
-	match = of_match_node(dt_match, of_node);
-	if (match && match->data) {
-		hdmi_cfg = (struct hdmi_platform_config *)match->data;
-		DBG("hdmi phy: %s", match->compatible);
-	} else {
-		dev_err(dev, "unknown phy: %s\n", of_node->name);
+	hdmi_cfg = (struct hdmi_platform_config *)
+			of_device_get_match_data(dev);
+	if (!hdmi_cfg) {
+		dev_err(dev, "unknown hdmi_cfg: %s\n", of_node->name);
 		return -ENXIO;
 	}
 
@@ -443,55 +427,6 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	hdmi_cfg->mux_sel_gpio  = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-sel");
 	hdmi_cfg->mux_lpm_gpio  = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-lpm");
 
-#else
-	static struct hdmi_platform_config config = {};
-	static const char *hpd_clk_names[] = {
-			"core_clk", "master_iface_clk", "slave_iface_clk",
-	};
-	if (cpu_is_apq8064()) {
-		static const char *hpd_reg_names[] = {"8921_hdmi_mvs"};
-		config.phy_init      = hdmi_phy_8960_init;
-		config.hpd_reg_names = hpd_reg_names;
-		config.hpd_reg_cnt   = ARRAY_SIZE(hpd_reg_names);
-		config.hpd_clk_names = hpd_clk_names;
-		config.hpd_clk_cnt   = ARRAY_SIZE(hpd_clk_names);
-		config.ddc_clk_gpio  = 70;
-		config.ddc_data_gpio = 71;
-		config.hpd_gpio      = 72;
-		config.mux_en_gpio   = -1;
-		config.mux_sel_gpio  = -1;
-	} else if (cpu_is_msm8960() || cpu_is_msm8960ab()) {
-		static const char *hpd_reg_names[] = {"8921_hdmi_mvs"};
-		config.phy_init      = hdmi_phy_8960_init;
-		config.hpd_reg_names = hpd_reg_names;
-		config.hpd_reg_cnt   = ARRAY_SIZE(hpd_reg_names);
-		config.hpd_clk_names = hpd_clk_names;
-		config.hpd_clk_cnt   = ARRAY_SIZE(hpd_clk_names);
-		config.ddc_clk_gpio  = 100;
-		config.ddc_data_gpio = 101;
-		config.hpd_gpio      = 102;
-		config.mux_en_gpio   = -1;
-		config.mux_sel_gpio  = -1;
-	} else if (cpu_is_msm8x60()) {
-		static const char *hpd_reg_names[] = {
-				"8901_hdmi_mvs", "8901_mpp0"
-		};
-		config.phy_init      = hdmi_phy_8x60_init;
-		config.hpd_reg_names = hpd_reg_names;
-		config.hpd_reg_cnt   = ARRAY_SIZE(hpd_reg_names);
-		config.hpd_clk_names = hpd_clk_names;
-		config.hpd_clk_cnt   = ARRAY_SIZE(hpd_clk_names);
-		config.ddc_clk_gpio  = 170;
-		config.ddc_data_gpio = 171;
-		config.hpd_gpio      = 172;
-		config.mux_en_gpio   = -1;
-		config.mux_sel_gpio  = -1;
-	}
-	config.mmio_name     = "hdmi_msm_hdmi_addr";
-	config.qfprom_mmio_name = "hdmi_msm_qfprom_addr";
-
-	hdmi_cfg = &config;
-#endif
 	dev->platform_data = hdmi_cfg;
 
 	hdmi = hdmi_init(to_platform_device(dev));
@@ -529,6 +464,16 @@ static int hdmi_dev_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id dt_match[] = {
+	{ .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config },
+	{ .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config },
+	{ .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config },
+	{ .compatible = "qcom,hdmi-tx-8974", .data = &hdmi_tx_8974_config },
+	{ .compatible = "qcom,hdmi-tx-8960", .data = &hdmi_tx_8960_config },
+	{ .compatible = "qcom,hdmi-tx-8660", .data = &hdmi_tx_8660_config },
+	{}
+};
+
 static struct platform_driver hdmi_driver = {
 	.probe = hdmi_dev_probe,
 	.remove = hdmi_dev_remove,
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c
new file mode 100644
index 000000000000..49052bb5d2e4
--- /dev/null
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_dsi_encoder.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright (c) 2015, The Linux Foundation. All rights reserved.
+ * Copyright (c) 2014, Inforce Computing. All rights reserved.
+ *
+ * Author: Vinay Simha <vinaysimha@inforcecomputing.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "mdp4_kms.h"
+
+#include "drm_crtc.h"
+#include "drm_crtc_helper.h"
+
+struct mdp4_dsi_encoder {
+	struct drm_encoder base;
+	struct drm_panel *panel;
+	bool enabled;
+};
+#define to_mdp4_dsi_encoder(x) container_of(x, struct mdp4_dsi_encoder, base)
+
+static struct mdp4_kms *get_kms(struct drm_encoder *encoder)
+{
+	struct msm_drm_private *priv = encoder->dev->dev_private;
+	return to_mdp4_kms(to_mdp_kms(priv->kms));
+}
+
+static void mdp4_dsi_encoder_destroy(struct drm_encoder *encoder)
+{
+	struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder);
+
+	drm_encoder_cleanup(encoder);
+	kfree(mdp4_dsi_encoder);
+}
+
+static const struct drm_encoder_funcs mdp4_dsi_encoder_funcs = {
+	.destroy = mdp4_dsi_encoder_destroy,
+};
+
+static bool mdp4_dsi_encoder_mode_fixup(struct drm_encoder *encoder,
+					const struct drm_display_mode *mode,
+					struct drm_display_mode *adjusted_mode)
+{
+	return true;
+}
+
+static void mdp4_dsi_encoder_mode_set(struct drm_encoder *encoder,
+				      struct drm_display_mode *mode,
+				      struct drm_display_mode *adjusted_mode)
+{
+	struct mdp4_kms *mdp4_kms = get_kms(encoder);
+	uint32_t dsi_hsync_skew, vsync_period, vsync_len, ctrl_pol;
+	uint32_t display_v_start, display_v_end;
+	uint32_t hsync_start_x, hsync_end_x;
+
+	mode = adjusted_mode;
+
+	DBG("set mode: %d:\"%s\" %d %d %d %d %d %d %d %d %d %d 0x%x 0x%x",
+			mode->base.id, mode->name,
+			mode->vrefresh, mode->clock,
+			mode->hdisplay, mode->hsync_start,
+			mode->hsync_end, mode->htotal,
+			mode->vdisplay, mode->vsync_start,
+			mode->vsync_end, mode->vtotal,
+			mode->type, mode->flags);
+
+	ctrl_pol = 0;
+	if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+		ctrl_pol |= MDP4_DSI_CTRL_POLARITY_HSYNC_LOW;
+	if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+		ctrl_pol |= MDP4_DSI_CTRL_POLARITY_VSYNC_LOW;
+	/* probably need to get DATA_EN polarity from panel.. */
+
+	dsi_hsync_skew = 0;  /* get this from panel? */
+
+	hsync_start_x = (mode->htotal - mode->hsync_start);
+	hsync_end_x = mode->htotal - (mode->hsync_start - mode->hdisplay) - 1;
+
+	vsync_period = mode->vtotal * mode->htotal;
+	vsync_len = (mode->vsync_end - mode->vsync_start) * mode->htotal;
+	display_v_start = (mode->vtotal - mode->vsync_start) * mode->htotal + dsi_hsync_skew;
+	display_v_end = vsync_period - ((mode->vsync_start - mode->vdisplay) * mode->htotal) + dsi_hsync_skew - 1;
+
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_CTRL,
+			MDP4_DSI_HSYNC_CTRL_PULSEW(mode->hsync_end - mode->hsync_start) |
+			MDP4_DSI_HSYNC_CTRL_PERIOD(mode->htotal));
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_PERIOD, vsync_period);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_VSYNC_LEN, vsync_len);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_HCTRL,
+			MDP4_DSI_DISPLAY_HCTRL_START(hsync_start_x) |
+			MDP4_DSI_DISPLAY_HCTRL_END(hsync_end_x));
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VSTART, display_v_start);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_DISPLAY_VEND, display_v_end);
+
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_CTRL_POLARITY, ctrl_pol);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_UNDERFLOW_CLR,
+			MDP4_DSI_UNDERFLOW_CLR_ENABLE_RECOVERY |
+			MDP4_DSI_UNDERFLOW_CLR_COLOR(0xff));
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_HCTL,
+			MDP4_DSI_ACTIVE_HCTL_START(0) |
+			MDP4_DSI_ACTIVE_HCTL_END(0));
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_HSYNC_SKEW, dsi_hsync_skew);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_BORDER_CLR, 0);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VSTART, 0);
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_ACTIVE_VEND, 0);
+}
+
+static void mdp4_dsi_encoder_disable(struct drm_encoder *encoder)
+{
+	struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder);
+	struct mdp4_kms *mdp4_kms = get_kms(encoder);
+
+	if (!mdp4_dsi_encoder->enabled)
+		return;
+
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 0);
+
+	/*
+	 * Wait for a vsync so we know the ENABLE=0 latched before
+	 * the (connector) source of the vsync's gets disabled,
+	 * otherwise we end up in a funny state if we re-enable
+	 * before the disable latches, which results that some of
+	 * the settings changes for the new modeset (like new
+	 * scanout buffer) don't latch properly..
+	 */
+	mdp_irq_wait(&mdp4_kms->base, MDP4_IRQ_PRIMARY_VSYNC);
+
+	mdp4_dsi_encoder->enabled = false;
+}
+
+static void mdp4_dsi_encoder_enable(struct drm_encoder *encoder)
+{
+	struct mdp4_dsi_encoder *mdp4_dsi_encoder = to_mdp4_dsi_encoder(encoder);
+	struct mdp4_kms *mdp4_kms = get_kms(encoder);
+
+	if (mdp4_dsi_encoder->enabled)
+		return;
+
+	 mdp4_crtc_set_config(encoder->crtc,
+			MDP4_DMA_CONFIG_PACK_ALIGN_MSB |
+			MDP4_DMA_CONFIG_DEFLKR_EN |
+			MDP4_DMA_CONFIG_DITHER_EN |
+			MDP4_DMA_CONFIG_R_BPC(BPC8) |
+			MDP4_DMA_CONFIG_G_BPC(BPC8) |
+			MDP4_DMA_CONFIG_B_BPC(BPC8) |
+			MDP4_DMA_CONFIG_PACK(0x21));
+
+	mdp4_crtc_set_intf(encoder->crtc, INTF_DSI_VIDEO, 0);
+
+	mdp4_write(mdp4_kms, REG_MDP4_DSI_ENABLE, 1);
+
+	mdp4_dsi_encoder->enabled = true;
+}
+
+static const struct drm_encoder_helper_funcs mdp4_dsi_encoder_helper_funcs = {
+	.mode_fixup = mdp4_dsi_encoder_mode_fixup,
+	.mode_set = mdp4_dsi_encoder_mode_set,
+	.disable = mdp4_dsi_encoder_disable,
+	.enable = mdp4_dsi_encoder_enable,
+};
+
+/* initialize encoder */
+struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev)
+{
+	struct drm_encoder *encoder = NULL;
+	struct mdp4_dsi_encoder *mdp4_dsi_encoder;
+	int ret;
+
+	mdp4_dsi_encoder = kzalloc(sizeof(*mdp4_dsi_encoder), GFP_KERNEL);
+	if (!mdp4_dsi_encoder) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	encoder = &mdp4_dsi_encoder->base;
+
+	drm_encoder_init(dev, encoder, &mdp4_dsi_encoder_funcs,
+			 DRM_MODE_ENCODER_DSI);
+	drm_encoder_helper_add(encoder, &mdp4_dsi_encoder_helper_funcs);
+
+	return encoder;
+
+fail:
+	if (encoder)
+		mdp4_dsi_encoder_destroy(encoder);
+
+	return ERR_PTR(ret);
+}
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c
index 5ed38cf548a1..a521207db8a1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_irq.c
@@ -29,7 +29,7 @@ void mdp4_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask,
 
 static void mdp4_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus)
 {
-	DRM_ERROR("errors: %08x\n", irqstatus);
+	DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus);
 }
 
 void mdp4_irq_preinstall(struct msm_kms *kms)
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
index 077f7521a971..5a8e3d6bcbff 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
@@ -169,7 +169,14 @@ static long mdp4_round_pixclk(struct msm_kms *kms, unsigned long rate,
 		struct drm_encoder *encoder)
 {
 	/* if we had >1 encoder, we'd need something more clever: */
-	return mdp4_dtv_round_pixclk(encoder, rate);
+	switch (encoder->encoder_type) {
+	case DRM_MODE_ENCODER_TMDS:
+		return mdp4_dtv_round_pixclk(encoder, rate);
+	case DRM_MODE_ENCODER_LVDS:
+	case DRM_MODE_ENCODER_DSI:
+	default:
+		return rate;
+	}
 }
 
 static void mdp4_preclose(struct msm_kms *kms, struct drm_file *file)
@@ -240,19 +247,18 @@ int mdp4_enable(struct mdp4_kms *mdp4_kms)
 	return 0;
 }
 
-#ifdef CONFIG_OF
-static struct drm_panel *detect_panel(struct drm_device *dev)
+static struct device_node *mdp4_detect_lcdc_panel(struct drm_device *dev)
 {
 	struct device_node *endpoint, *panel_node;
 	struct device_node *np = dev->dev->of_node;
-	struct drm_panel *panel = NULL;
 
 	endpoint = of_graph_get_next_endpoint(np, NULL);
 	if (!endpoint) {
-		dev_err(dev->dev, "no valid endpoint\n");
-		return ERR_PTR(-ENODEV);
+		DBG("no endpoint in MDP4 to fetch LVDS panel\n");
+		return NULL;
 	}
 
+	/* don't proceed if we have an endpoint but no panel_node tied to it */
 	panel_node = of_graph_get_remote_port_parent(endpoint);
 	if (!panel_node) {
 		dev_err(dev->dev, "no valid panel node\n");
@@ -262,20 +268,111 @@ static struct drm_panel *detect_panel(struct drm_device *dev)
 
 	of_node_put(endpoint);
 
-	panel = of_drm_find_panel(panel_node);
-	if (!panel) {
-		of_node_put(panel_node);
-		return ERR_PTR(-EPROBE_DEFER);
+	return panel_node;
+}
+
+static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms,
+				  int intf_type)
+{
+	struct drm_device *dev = mdp4_kms->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	struct device_node *panel_node;
+	struct drm_encoder *dsi_encs[MSM_DSI_ENCODER_NUM];
+	int i, dsi_id;
+	int ret;
+
+	switch (intf_type) {
+	case DRM_MODE_ENCODER_LVDS:
+		/*
+		 * bail out early if:
+		 * - there is no panel node (no need to initialize lcdc
+		 *   encoder and lvds connector), or
+		 * - panel node is a bad pointer
+		 */
+		panel_node = mdp4_detect_lcdc_panel(dev);
+		if (IS_ERR_OR_NULL(panel_node))
+			return PTR_ERR(panel_node);
+
+		encoder = mdp4_lcdc_encoder_init(dev, panel_node);
+		if (IS_ERR(encoder)) {
+			dev_err(dev->dev, "failed to construct LCDC encoder\n");
+			return PTR_ERR(encoder);
+		}
+
+		/* LCDC can be hooked to DMA_P (TODO: Add DMA_S later?) */
+		encoder->possible_crtcs = 1 << DMA_P;
+
+		connector = mdp4_lvds_connector_init(dev, panel_node, encoder);
+		if (IS_ERR(connector)) {
+			dev_err(dev->dev, "failed to initialize LVDS connector\n");
+			return PTR_ERR(connector);
+		}
+
+		priv->encoders[priv->num_encoders++] = encoder;
+		priv->connectors[priv->num_connectors++] = connector;
+
+		break;
+	case DRM_MODE_ENCODER_TMDS:
+		encoder = mdp4_dtv_encoder_init(dev);
+		if (IS_ERR(encoder)) {
+			dev_err(dev->dev, "failed to construct DTV encoder\n");
+			return PTR_ERR(encoder);
+		}
+
+		/* DTV can be hooked to DMA_E: */
+		encoder->possible_crtcs = 1 << 1;
+
+		if (priv->hdmi) {
+			/* Construct bridge/connector for HDMI: */
+			ret = hdmi_modeset_init(priv->hdmi, dev, encoder);
+			if (ret) {
+				dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret);
+				return ret;
+			}
+		}
+
+		priv->encoders[priv->num_encoders++] = encoder;
+
+		break;
+	case DRM_MODE_ENCODER_DSI:
+		/* only DSI1 supported for now */
+		dsi_id = 0;
+
+		if (!priv->dsi[dsi_id])
+			break;
+
+		for (i = 0; i < MSM_DSI_ENCODER_NUM; i++) {
+			dsi_encs[i] = mdp4_dsi_encoder_init(dev);
+			if (IS_ERR(dsi_encs[i])) {
+				ret = PTR_ERR(dsi_encs[i]);
+				dev_err(dev->dev,
+					"failed to construct DSI encoder: %d\n",
+					ret);
+				return ret;
+			}
+
+			/* TODO: Add DMA_S later? */
+			dsi_encs[i]->possible_crtcs = 1 << DMA_P;
+			priv->encoders[priv->num_encoders++] = dsi_encs[i];
+		}
+
+		ret = msm_dsi_modeset_init(priv->dsi[dsi_id], dev, dsi_encs);
+		if (ret) {
+			dev_err(dev->dev, "failed to initialize DSI: %d\n",
+				ret);
+			return ret;
+		}
+
+		break;
+	default:
+		dev_err(dev->dev, "Invalid or unsupported interface\n");
+		return -EINVAL;
 	}
 
-	return panel;
+	return 0;
 }
-#else
-static struct drm_panel *detect_panel(struct drm_device *dev)
-{
-	// ??? maybe use a module param to specify which panel is attached?
-}
-#endif
 
 static int modeset_init(struct mdp4_kms *mdp4_kms)
 {
@@ -283,111 +380,73 @@ static int modeset_init(struct mdp4_kms *mdp4_kms)
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_plane *plane;
 	struct drm_crtc *crtc;
-	struct drm_encoder *encoder;
-	struct drm_connector *connector;
-	struct drm_panel *panel;
-	int ret;
+	int i, ret;
+	static const enum mdp4_pipe rgb_planes[] = {
+		RGB1, RGB2,
+	};
+	static const enum mdp4_pipe vg_planes[] = {
+		VG1, VG2,
+	};
+	static const enum mdp4_dma mdp4_crtcs[] = {
+		DMA_P, DMA_E,
+	};
+	static const char * const mdp4_crtc_names[] = {
+		"DMA_P", "DMA_E",
+	};
+	static const int mdp4_intfs[] = {
+		DRM_MODE_ENCODER_LVDS,
+		DRM_MODE_ENCODER_DSI,
+		DRM_MODE_ENCODER_TMDS,
+	};
 
 	/* construct non-private planes: */
-	plane = mdp4_plane_init(dev, VG1, false);
-	if (IS_ERR(plane)) {
-		dev_err(dev->dev, "failed to construct plane for VG1\n");
-		ret = PTR_ERR(plane);
-		goto fail;
+	for (i = 0; i < ARRAY_SIZE(vg_planes); i++) {
+		plane = mdp4_plane_init(dev, vg_planes[i], false);
+		if (IS_ERR(plane)) {
+			dev_err(dev->dev,
+				"failed to construct plane for VG%d\n", i + 1);
+			ret = PTR_ERR(plane);
+			goto fail;
+		}
+		priv->planes[priv->num_planes++] = plane;
 	}
-	priv->planes[priv->num_planes++] = plane;
 
-	plane = mdp4_plane_init(dev, VG2, false);
-	if (IS_ERR(plane)) {
-		dev_err(dev->dev, "failed to construct plane for VG2\n");
-		ret = PTR_ERR(plane);
-		goto fail;
+	for (i = 0; i < ARRAY_SIZE(mdp4_crtcs); i++) {
+		plane = mdp4_plane_init(dev, rgb_planes[i], true);
+		if (IS_ERR(plane)) {
+			dev_err(dev->dev,
+				"failed to construct plane for RGB%d\n", i + 1);
+			ret = PTR_ERR(plane);
+			goto fail;
+		}
+
+		crtc  = mdp4_crtc_init(dev, plane, priv->num_crtcs, i,
+				mdp4_crtcs[i]);
+		if (IS_ERR(crtc)) {
+			dev_err(dev->dev, "failed to construct crtc for %s\n",
+				mdp4_crtc_names[i]);
+			ret = PTR_ERR(crtc);
+			goto fail;
+		}
+
+		priv->crtcs[priv->num_crtcs++] = crtc;
 	}
-	priv->planes[priv->num_planes++] = plane;
 
 	/*
-	 * Setup the LCDC/LVDS path: RGB2 -> DMA_P -> LCDC -> LVDS:
+	 * we currently set up two relatively fixed paths:
+	 *
+	 * LCDC/LVDS path: RGB1 -> DMA_P -> LCDC -> LVDS
+	 *			or
+	 * DSI path: RGB1 -> DMA_P -> DSI1 -> DSI Panel
+	 *
+	 * DTV/HDMI path: RGB2 -> DMA_E -> DTV -> HDMI
 	 */
 
-	panel = detect_panel(dev);
-	if (IS_ERR(panel)) {
-		ret = PTR_ERR(panel);
-		dev_err(dev->dev, "failed to detect LVDS panel: %d\n", ret);
-		goto fail;
-	}
-
-	plane = mdp4_plane_init(dev, RGB2, true);
-	if (IS_ERR(plane)) {
-		dev_err(dev->dev, "failed to construct plane for RGB2\n");
-		ret = PTR_ERR(plane);
-		goto fail;
-	}
-
-	crtc  = mdp4_crtc_init(dev, plane, priv->num_crtcs, 0, DMA_P);
-	if (IS_ERR(crtc)) {
-		dev_err(dev->dev, "failed to construct crtc for DMA_P\n");
-		ret = PTR_ERR(crtc);
-		goto fail;
-	}
-
-	encoder = mdp4_lcdc_encoder_init(dev, panel);
-	if (IS_ERR(encoder)) {
-		dev_err(dev->dev, "failed to construct LCDC encoder\n");
-		ret = PTR_ERR(encoder);
-		goto fail;
-	}
-
-	/* LCDC can be hooked to DMA_P: */
-	encoder->possible_crtcs = 1 << priv->num_crtcs;
-
-	priv->crtcs[priv->num_crtcs++] = crtc;
-	priv->encoders[priv->num_encoders++] = encoder;
-
-	connector = mdp4_lvds_connector_init(dev, panel, encoder);
-	if (IS_ERR(connector)) {
-		ret = PTR_ERR(connector);
-		dev_err(dev->dev, "failed to initialize LVDS connector: %d\n", ret);
-		goto fail;
-	}
-
-	priv->connectors[priv->num_connectors++] = connector;
-
-	/*
-	 * Setup DTV/HDMI path: RGB1 -> DMA_E -> DTV -> HDMI:
-	 */
-
-	plane = mdp4_plane_init(dev, RGB1, true);
-	if (IS_ERR(plane)) {
-		dev_err(dev->dev, "failed to construct plane for RGB1\n");
-		ret = PTR_ERR(plane);
-		goto fail;
-	}
-
-	crtc  = mdp4_crtc_init(dev, plane, priv->num_crtcs, 1, DMA_E);
-	if (IS_ERR(crtc)) {
-		dev_err(dev->dev, "failed to construct crtc for DMA_E\n");
-		ret = PTR_ERR(crtc);
-		goto fail;
-	}
-
-	encoder = mdp4_dtv_encoder_init(dev);
-	if (IS_ERR(encoder)) {
-		dev_err(dev->dev, "failed to construct DTV encoder\n");
-		ret = PTR_ERR(encoder);
-		goto fail;
-	}
-
-	/* DTV can be hooked to DMA_E: */
-	encoder->possible_crtcs = 1 << priv->num_crtcs;
-
-	priv->crtcs[priv->num_crtcs++] = crtc;
-	priv->encoders[priv->num_encoders++] = encoder;
-
-	if (priv->hdmi) {
-		/* Construct bridge/connector for HDMI: */
-		ret = hdmi_modeset_init(priv->hdmi, dev, encoder);
+	for (i = 0; i < ARRAY_SIZE(mdp4_intfs); i++) {
+		ret = mdp4_modeset_init_intf(mdp4_kms, mdp4_intfs[i]);
 		if (ret) {
-			dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret);
+			dev_err(dev->dev, "failed to initialize intf: %d, %d\n",
+				i, ret);
 			goto fail;
 		}
 	}
@@ -558,17 +617,10 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev)
 static struct mdp4_platform_config *mdp4_get_config(struct platform_device *dev)
 {
 	static struct mdp4_platform_config config = {};
-#ifdef CONFIG_OF
-	/* TODO */
+
+	/* TODO: Chips that aren't apq8064 have a 200 Mhz max_clk */
 	config.max_clk = 266667000;
 	config.iommu = iommu_domain_alloc(&platform_bus_type);
-#else
-	if (cpu_is_apq8064())
-		config.max_clk = 266667000;
-	else
-		config.max_clk = 200000000;
 
-	config.iommu = msm_get_iommu_domain(DISPLAY_READ_DOMAIN);
-#endif
 	return &config;
 }
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
index 8a7f6e1e2bca..d2c96ef431f4 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
@@ -157,7 +157,7 @@ static inline uint32_t mixercfg(uint32_t mixer_cfg, int mixer,
 			COND(mixer == 1, MDP4_LAYERMIXER_IN_CFG_PIPE6_MIXER1);
 		break;
 	default:
-		WARN_ON("invalid pipe");
+		WARN(1, "invalid pipe");
 		break;
 	}
 
@@ -212,10 +212,19 @@ struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev);
 
 long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate);
 struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
-		struct drm_panel *panel);
+		struct device_node *panel_node);
 
 struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
-		struct drm_panel *panel, struct drm_encoder *encoder);
+		struct device_node *panel_node, struct drm_encoder *encoder);
+
+#ifdef CONFIG_DRM_MSM_DSI
+struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev);
+#else
+static inline struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev)
+{
+	return ERR_PTR(-ENODEV);
+}
+#endif
 
 #ifdef CONFIG_COMMON_CLK
 struct clk *mpd4_lvds_pll_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
index 4cd6e721aa0a..1bda2de20334 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lcdc_encoder.c
@@ -23,6 +23,7 @@
 
 struct mdp4_lcdc_encoder {
 	struct drm_encoder base;
+	struct device_node *panel_node;
 	struct drm_panel *panel;
 	struct clk *lcdc_clk;
 	unsigned long int pixclock;
@@ -338,7 +339,7 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder)
 	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder =
 			to_mdp4_lcdc_encoder(encoder);
 	struct mdp4_kms *mdp4_kms = get_kms(encoder);
-	struct drm_panel *panel = mdp4_lcdc_encoder->panel;
+	struct drm_panel *panel;
 	int i, ret;
 
 	if (WARN_ON(!mdp4_lcdc_encoder->enabled))
@@ -346,6 +347,7 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder)
 
 	mdp4_write(mdp4_kms, REG_MDP4_LCDC_ENABLE, 0);
 
+	panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node);
 	if (panel) {
 		drm_panel_disable(panel);
 		drm_panel_unprepare(panel);
@@ -381,7 +383,7 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 			to_mdp4_lcdc_encoder(encoder);
 	unsigned long pc = mdp4_lcdc_encoder->pixclock;
 	struct mdp4_kms *mdp4_kms = get_kms(encoder);
-	struct drm_panel *panel = mdp4_lcdc_encoder->panel;
+	struct drm_panel *panel;
 	int i, ret;
 
 	if (WARN_ON(mdp4_lcdc_encoder->enabled))
@@ -414,6 +416,7 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder)
 	if (ret)
 		dev_err(dev->dev, "failed to enable lcdc_clk: %d\n", ret);
 
+	panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node);
 	if (panel) {
 		drm_panel_prepare(panel);
 		drm_panel_enable(panel);
@@ -442,7 +445,7 @@ long mdp4_lcdc_round_pixclk(struct drm_encoder *encoder, unsigned long rate)
 
 /* initialize encoder */
 struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
-		struct drm_panel *panel)
+		struct device_node *panel_node)
 {
 	struct drm_encoder *encoder = NULL;
 	struct mdp4_lcdc_encoder *mdp4_lcdc_encoder;
@@ -455,7 +458,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev,
 		goto fail;
 	}
 
-	mdp4_lcdc_encoder->panel = panel;
+	mdp4_lcdc_encoder->panel_node = panel_node;
 
 	encoder = &mdp4_lcdc_encoder->base;
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c
index 921185133d38..e73e1742b250 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_lvds_connector.c
@@ -23,6 +23,7 @@
 struct mdp4_lvds_connector {
 	struct drm_connector base;
 	struct drm_encoder *encoder;
+	struct device_node *panel_node;
 	struct drm_panel *panel;
 };
 #define to_mdp4_lvds_connector(x) container_of(x, struct mdp4_lvds_connector, base)
@@ -33,6 +34,10 @@ static enum drm_connector_status mdp4_lvds_connector_detect(
 	struct mdp4_lvds_connector *mdp4_lvds_connector =
 			to_mdp4_lvds_connector(connector);
 
+	if (!mdp4_lvds_connector->panel)
+		mdp4_lvds_connector->panel =
+			of_drm_find_panel(mdp4_lvds_connector->panel_node);
+
 	return mdp4_lvds_connector->panel ?
 			connector_status_connected :
 			connector_status_disconnected;
@@ -42,10 +47,6 @@ static void mdp4_lvds_connector_destroy(struct drm_connector *connector)
 {
 	struct mdp4_lvds_connector *mdp4_lvds_connector =
 			to_mdp4_lvds_connector(connector);
-	struct drm_panel *panel = mdp4_lvds_connector->panel;
-
-	if (panel)
-		drm_panel_detach(panel);
 
 	drm_connector_unregister(connector);
 	drm_connector_cleanup(connector);
@@ -60,9 +61,14 @@ static int mdp4_lvds_connector_get_modes(struct drm_connector *connector)
 	struct drm_panel *panel = mdp4_lvds_connector->panel;
 	int ret = 0;
 
-	if (panel)
+	if (panel) {
+		drm_panel_attach(panel, connector);
+
 		ret = panel->funcs->get_modes(panel);
 
+		drm_panel_detach(panel);
+	}
+
 	return ret;
 }
 
@@ -111,7 +117,7 @@ static const struct drm_connector_helper_funcs mdp4_lvds_connector_helper_funcs
 
 /* initialize connector */
 struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
-		struct drm_panel *panel, struct drm_encoder *encoder)
+		struct device_node *panel_node, struct drm_encoder *encoder)
 {
 	struct drm_connector *connector = NULL;
 	struct mdp4_lvds_connector *mdp4_lvds_connector;
@@ -124,7 +130,7 @@ struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
 	}
 
 	mdp4_lvds_connector->encoder = encoder;
-	mdp4_lvds_connector->panel = panel;
+	mdp4_lvds_connector->panel_node = panel_node;
 
 	connector = &mdp4_lvds_connector->base;
 
@@ -141,9 +147,6 @@ struct drm_connector *mdp4_lvds_connector_init(struct drm_device *dev,
 
 	drm_mode_connector_attach_encoder(connector, encoder);
 
-	if (panel)
-		drm_panel_attach(panel, connector);
-
 	return connector;
 
 fail:
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
index bb1225aa2f75..57f73f0c120d 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
@@ -553,9 +553,7 @@ struct mdp5_cfg_handler *mdp5_cfg_init(struct mdp5_kms *mdp5_kms,
 static struct mdp5_cfg_platform *mdp5_get_config(struct platform_device *dev)
 {
 	static struct mdp5_cfg_platform config = {};
-#ifdef CONFIG_OF
-	/* TODO */
-#endif
+
 	config.iommu = iommu_domain_alloc(&platform_bus_type);
 
 	return &config;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
index c9e32b08a7a0..a0196568bf5d 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_encoder.c
@@ -293,6 +293,24 @@ static const struct drm_encoder_helper_funcs mdp5_encoder_helper_funcs = {
 	.enable = mdp5_encoder_enable,
 };
 
+int mdp5_encoder_get_linecount(struct drm_encoder *encoder)
+{
+	struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder);
+	struct mdp5_kms *mdp5_kms = get_kms(encoder);
+	int intf = mdp5_encoder->intf.num;
+
+	return mdp5_read(mdp5_kms, REG_MDP5_INTF_LINE_COUNT(intf));
+}
+
+u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder)
+{
+	struct mdp5_encoder *mdp5_encoder = to_mdp5_encoder(encoder);
+	struct mdp5_kms *mdp5_kms = get_kms(encoder);
+	int intf = mdp5_encoder->intf.num;
+
+	return mdp5_read(mdp5_kms, REG_MDP5_INTF_FRAME_COUNT(intf));
+}
+
 int mdp5_encoder_set_split_display(struct drm_encoder *encoder,
 					struct drm_encoder *slave_encoder)
 {
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
index b0d4b53b97f4..73bc3e312fd4 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_irq.c
@@ -31,7 +31,7 @@ void mdp5_set_irqmask(struct mdp_kms *mdp_kms, uint32_t irqmask,
 
 static void mdp5_irq_error_handler(struct mdp_irq *irq, uint32_t irqstatus)
 {
-	DRM_ERROR("errors: %08x\n", irqstatus);
+	DRM_ERROR_RATELIMITED("errors: %08x\n", irqstatus);
 }
 
 void mdp5_irq_preinstall(struct msm_kms *kms)
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index b532faa8026d..e115318402bd 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -468,6 +468,127 @@ static int get_clk(struct platform_device *pdev, struct clk **clkp,
 	return 0;
 }
 
+static struct drm_encoder *get_encoder_from_crtc(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_encoder *encoder;
+
+	drm_for_each_encoder(encoder, dev)
+		if (encoder->crtc == crtc)
+			return encoder;
+
+	return NULL;
+}
+
+static int mdp5_get_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			       unsigned int flags, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime,
+			       const struct drm_display_mode *mode)
+{
+	struct msm_drm_private *priv = dev->dev_private;
+	struct drm_crtc *crtc;
+	struct drm_encoder *encoder;
+	int line, vsw, vbp, vactive_start, vactive_end, vfp_end;
+	int ret = 0;
+
+	crtc = priv->crtcs[pipe];
+	if (!crtc) {
+		DRM_ERROR("Invalid crtc %d\n", pipe);
+		return 0;
+	}
+
+	encoder = get_encoder_from_crtc(crtc);
+	if (!encoder) {
+		DRM_ERROR("no encoder found for crtc %d\n", pipe);
+		return 0;
+	}
+
+	ret |= DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_ACCURATE;
+
+	vsw = mode->crtc_vsync_end - mode->crtc_vsync_start;
+	vbp = mode->crtc_vtotal - mode->crtc_vsync_end;
+
+	/*
+	 * the line counter is 1 at the start of the VSYNC pulse and VTOTAL at
+	 * the end of VFP. Translate the porch values relative to the line
+	 * counter positions.
+	 */
+
+	vactive_start = vsw + vbp + 1;
+
+	vactive_end = vactive_start + mode->crtc_vdisplay;
+
+	/* last scan line before VSYNC */
+	vfp_end = mode->crtc_vtotal;
+
+	if (stime)
+		*stime = ktime_get();
+
+	line = mdp5_encoder_get_linecount(encoder);
+
+	if (line < vactive_start) {
+		line -= vactive_start;
+		ret |= DRM_SCANOUTPOS_IN_VBLANK;
+	} else if (line > vactive_end) {
+		line = line - vfp_end - vactive_start;
+		ret |= DRM_SCANOUTPOS_IN_VBLANK;
+	} else {
+		line -= vactive_start;
+	}
+
+	*vpos = line;
+	*hpos = 0;
+
+	if (etime)
+		*etime = ktime_get();
+
+	return ret;
+}
+
+static int mdp5_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe,
+				     int *max_error,
+				     struct timeval *vblank_time,
+				     unsigned flags)
+{
+	struct msm_drm_private *priv = dev->dev_private;
+	struct drm_crtc *crtc;
+
+	if (pipe < 0 || pipe >= priv->num_crtcs) {
+		DRM_ERROR("Invalid crtc %d\n", pipe);
+		return -EINVAL;
+	}
+
+	crtc = priv->crtcs[pipe];
+	if (!crtc) {
+		DRM_ERROR("Invalid crtc %d\n", pipe);
+		return -EINVAL;
+	}
+
+	return drm_calc_vbltimestamp_from_scanoutpos(dev, pipe, max_error,
+						     vblank_time, flags,
+						     &crtc->mode);
+}
+
+static u32 mdp5_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
+{
+	struct msm_drm_private *priv = dev->dev_private;
+	struct drm_crtc *crtc;
+	struct drm_encoder *encoder;
+
+	if (pipe < 0 || pipe >= priv->num_crtcs)
+		return 0;
+
+	crtc = priv->crtcs[pipe];
+	if (!crtc)
+		return 0;
+
+	encoder = get_encoder_from_crtc(crtc);
+	if (!encoder)
+		return 0;
+
+	return mdp5_encoder_get_framecount(encoder);
+}
+
 struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 {
 	struct platform_device *pdev = dev->platformdev;
@@ -590,6 +711,8 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 				!config->hw->intf.base[i])
 			continue;
 		mdp5_write(mdp5_kms, REG_MDP5_INTF_TIMING_ENGINE_EN(i), 0);
+
+		mdp5_write(mdp5_kms, REG_MDP5_INTF_FRAME_LINE_COUNT_EN(i), 0x3);
 	}
 	mdp5_disable(mdp5_kms);
 	mdelay(16);
@@ -635,6 +758,12 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 	dev->mode_config.max_width = config->hw->lm.max_width;
 	dev->mode_config.max_height = config->hw->lm.max_height;
 
+	dev->driver->get_vblank_timestamp = mdp5_get_vblank_timestamp;
+	dev->driver->get_scanout_position = mdp5_get_scanoutpos;
+	dev->driver->get_vblank_counter = mdp5_get_vblank_counter;
+	dev->max_vblank_count = 0xffffffff;
+	dev->vblank_disable_immediate = true;
+
 	return kms;
 
 fail:
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
index 84f65d415598..00730ba08a60 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
@@ -222,6 +222,8 @@ struct drm_encoder *mdp5_encoder_init(struct drm_device *dev,
 		struct mdp5_interface *intf, struct mdp5_ctl *ctl);
 int mdp5_encoder_set_split_display(struct drm_encoder *encoder,
 					struct drm_encoder *slave_encoder);
+int mdp5_encoder_get_linecount(struct drm_encoder *encoder);
+u32 mdp5_encoder_get_framecount(struct drm_encoder *encoder);
 
 #ifdef CONFIG_DRM_MSM_DSI
 struct drm_encoder *mdp5_cmd_encoder_init(struct drm_device *dev,
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index b88ce514eb8e..9a30807b900b 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -237,20 +237,9 @@ static int msm_unload(struct drm_device *dev)
 
 static int get_mdp_ver(struct platform_device *pdev)
 {
-#ifdef CONFIG_OF
-	static const struct of_device_id match_types[] = { {
-		.compatible = "qcom,mdss_mdp",
-		.data	= (void	*)5,
-	}, {
-		/* end node */
-	} };
 	struct device *dev = &pdev->dev;
-	const struct of_device_id *match;
-	match = of_match_node(match_types, dev->of_node);
-	if (match)
-		return (int)(unsigned long)match->data;
-#endif
-	return 4;
+
+	return (int) (unsigned long) of_device_get_match_data(dev);
 }
 
 #include <linux/of_address.h>
@@ -258,10 +247,10 @@ static int get_mdp_ver(struct platform_device *pdev)
 static int msm_init_vram(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
+	struct device_node *node;
 	unsigned long size = 0;
 	int ret = 0;
 
-#ifdef CONFIG_OF
 	/* In the device-tree world, we could have a 'memory-region'
 	 * phandle, which gives us a link to our "vram".  Allocating
 	 * is all nicely abstracted behind the dma api, but we need
@@ -278,7 +267,6 @@ static int msm_init_vram(struct drm_device *dev)
 	 *     as corruption on screen before we have a chance to
 	 *     load and do initial modeset)
 	 */
-	struct device_node *node;
 
 	node = of_parse_phandle(dev->dev->of_node, "memory-region", 0);
 	if (node) {
@@ -288,14 +276,12 @@ static int msm_init_vram(struct drm_device *dev)
 			return ret;
 		size = r.end - r.start;
 		DRM_INFO("using VRAM carveout: %lx@%pa\n", size, &r.start);
-	} else
-#endif
 
-	/* if we have no IOMMU, then we need to use carveout allocator.
-	 * Grab the entire CMA chunk carved out in early startup in
-	 * mach-msm:
-	 */
-	if (!iommu_present(&platform_bus_type)) {
+		/* if we have no IOMMU, then we need to use carveout allocator.
+		 * Grab the entire CMA chunk carved out in early startup in
+		 * mach-msm:
+		 */
+	} else if (!iommu_present(&platform_bus_type)) {
 		DRM_INFO("using %s VRAM carveout\n", vram);
 		size = memparse(vram, NULL);
 	}
@@ -1035,9 +1021,9 @@ static const struct dev_pm_ops msm_pm_ops = {
  * Componentized driver support:
  */
 
-#ifdef CONFIG_OF
-/* NOTE: the CONFIG_OF case duplicates the same code as exynos or imx
- * (or probably any other).. so probably some room for some helpers
+/*
+ * NOTE: duplication of the same code as exynos or imx (or probably any other).
+ * so probably some room for some helpers
  */
 static int compare_of(struct device *dev, void *data)
 {
@@ -1062,12 +1048,6 @@ static int add_components(struct device *dev, struct component_match **matchptr,
 
 	return 0;
 }
-#else
-static int compare_dev(struct device *dev, void *data)
-{
-	return dev == data;
-}
-#endif
 
 static int msm_drm_bind(struct device *dev)
 {
@@ -1091,35 +1071,9 @@ static const struct component_master_ops msm_drm_ops = {
 static int msm_pdev_probe(struct platform_device *pdev)
 {
 	struct component_match *match = NULL;
-#ifdef CONFIG_OF
+
 	add_components(&pdev->dev, &match, "connectors");
 	add_components(&pdev->dev, &match, "gpus");
-#else
-	/* For non-DT case, it kinda sucks.  We don't actually have a way
-	 * to know whether or not we are waiting for certain devices (or if
-	 * they are simply not present).  But for non-DT we only need to
-	 * care about apq8064/apq8060/etc (all mdp4/a3xx):
-	 */
-	static const char *devnames[] = {
-			"hdmi_msm.0", "kgsl-3d0.0",
-	};
-	int i;
-
-	DBG("Adding components..");
-
-	for (i = 0; i < ARRAY_SIZE(devnames); i++) {
-		struct device *dev;
-
-		dev = bus_find_device_by_name(&platform_bus_type,
-				NULL, devnames[i]);
-		if (!dev) {
-			dev_info(&pdev->dev, "still waiting for %s\n", devnames[i]);
-			return -EPROBE_DEFER;
-		}
-
-		component_match_add(&pdev->dev, &match, compare_dev, dev);
-	}
-#endif
 
 	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
 	return component_master_add_with_match(&pdev->dev, &msm_drm_ops, match);
@@ -1138,8 +1092,10 @@ static const struct platform_device_id msm_id[] = {
 };
 
 static const struct of_device_id dt_match[] = {
-	{ .compatible = "qcom,mdp" },      /* mdp4 */
-	{ .compatible = "qcom,mdss_mdp" }, /* mdp5 */
+	{ .compatible = "qcom,mdp4", .data = (void *) 4 },	/* mdp4 */
+	{ .compatible = "qcom,mdp5", .data = (void *) 5 },	/* mdp5 */
+	/* to support downstream DT files */
+	{ .compatible = "qcom,mdss_mdp", .data = (void *) 5 },  /* mdp5 */
 	{}
 };
 MODULE_DEVICE_TABLE(of, dt_match);
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 9a713b7a009d..c1e7bba2fdb7 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -31,14 +31,9 @@
 #include <linux/iommu.h>
 #include <linux/types.h>
 #include <linux/of_graph.h>
+#include <linux/of_device.h>
 #include <asm/sizes.h>
 
-#ifndef CONFIG_OF
-#include <mach/board.h>
-#include <mach/socinfo.h>
-#include <mach/iommu_domains.h>
-#endif
-
 #include <drm/drmP.h>
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index 3f6ec077b51d..d95af6eba602 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -121,7 +121,7 @@ static int msm_fbdev_create(struct drm_fb_helper *helper,
 		/* note: if fb creation failed, we can't rely on fb destroy
 		 * to unref the bo:
 		 */
-		drm_gem_object_unreference(fbdev->bo);
+		drm_gem_object_unreference_unlocked(fbdev->bo);
 		ret = PTR_ERR(fb);
 		goto fail;
 	}
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
index e995ffbcf86a..81e6e0d1d360 100644
--- a/include/uapi/drm/msm_drm.h
+++ b/include/uapi/drm/msm_drm.h
@@ -121,7 +121,7 @@ struct drm_msm_gem_cpu_fini {
 struct drm_msm_gem_submit_reloc {
 	__u32 submit_offset;  /* in, offset from submit_bo */
 	__u32 or;             /* in, value OR'd with result */
-	__s32  shift;          /* in, amount of left shift (can be negative) */
+	__s32 shift;          /* in, amount of left shift (can be negative) */
 	__u32 reloc_idx;      /* in, index of reloc_bo buffer */
 	__u64 reloc_offset;   /* in, offset from start of reloc_bo */
 };