From e5ae8d1eb08a3e27fff4ae264af4c8056d908639 Mon Sep 17 00:00:00 2001 From: Kenneth Graunke Date: Fri, 12 Sep 2025 15:31:45 -0700 Subject: [PATCH 01/11] drm/xe: Increase global invalidation timeout to 1000us The previous timeout of 500us seems to be too small; panning the map in the Roll20 VTT in Firefox on a KDE/Wayland desktop reliably triggered timeouts within a few seconds of usage, causing the monitor to freeze and the following to be printed to dmesg: [Jul30 13:44] xe 0000:03:00.0: [drm] *ERROR* GT0: Global invalidation timeout [Jul30 13:48] xe 0000:03:00.0: [drm] *ERROR* [CRTC:82:pipe A] flip_done timed out I haven't hit a single timeout since increasing it to 1000us even after several multi-hour testing sessions. Fixes: 0dd2dd0182bc ("drm/xe: Move DSB l2 flush to a more sensible place") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5710 Signed-off-by: Kenneth Graunke Cc: stable@vger.kernel.org Cc: Maarten Lankhorst Reviewed-by: Shuicheng Lin Link: https://lore.kernel.org/r/20250912223254.147940-1-kenneth@whitecape.org Signed-off-by: Lucas De Marchi (cherry picked from commit 146046907b56578263434107f5a7d5051847c459) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 2883b39c9b37..34d33965eac2 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -1070,7 +1070,7 @@ void xe_device_l2_flush(struct xe_device *xe) spin_lock(>->global_invl_lock); xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); - if (xe_mmio_wait32(>->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 500, NULL, true)) + if (xe_mmio_wait32(>->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 1000, NULL, true)) xe_gt_err_once(gt, "Global invalidation timeout\n"); spin_unlock(>->global_invl_lock); From 7ac74613e5f2ef3450f44fd2127198662c2563a9 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Thu, 9 Oct 2025 04:06:18 -0700 Subject: [PATCH 02/11] drm/xe: Don't allow evicting of BOs in same VM in array of VM binds MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit An array of VM binds can potentially evict other buffer objects (BOs) within the same VM under certain conditions, which may lead to NULL pointer dereferences later in the bind pipeline. To prevent this, clear the allow_res_evict flag in the xe_bo_validate call. v2: - Invert polarity of no_res_evict (Thomas) - Add comment in code explaining issue (Thomas) Cc: stable@vger.kernel.org Reported-by: Paulo Zanoni Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6268 Fixes: 774b5fa509a9 ("drm/xe: Avoid evicting object of the same vm in none fault mode") Fixes: 77f2ef3f16f5 ("drm/xe: Lock all gpuva ops during VM bind IOCTL") Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Signed-off-by: Matthew Brost Tested-by: Paulo Zanoni Reviewed-by: Thomas Hellström Link: https://lore.kernel.org/r/20251009110618.3481870-1-matthew.brost@intel.com (cherry picked from commit 8b9ba8d6d95fe75fed6b0480bb03da4b321bea08) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_vm.c | 32 +++++++++++++++++++++++--------- drivers/gpu/drm/xe/xe_vm_types.h | 1 + 2 files changed, 24 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 027e6ce648c5..f602b874e054 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -2832,7 +2832,7 @@ static void vm_bind_ioctl_ops_unwind(struct xe_vm *vm, } static int vma_lock_and_validate(struct drm_exec *exec, struct xe_vma *vma, - bool validate) + bool res_evict, bool validate) { struct xe_bo *bo = xe_vma_bo(vma); struct xe_vm *vm = xe_vma_vm(vma); @@ -2843,7 +2843,8 @@ static int vma_lock_and_validate(struct drm_exec *exec, struct xe_vma *vma, err = drm_exec_lock_obj(exec, &bo->ttm.base); if (!err && validate) err = xe_bo_validate(bo, vm, - !xe_vm_in_preempt_fence_mode(vm), exec); + !xe_vm_in_preempt_fence_mode(vm) && + res_evict, exec); } return err; @@ -2913,14 +2914,23 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op) } static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm, - struct xe_vma_op *op) + struct xe_vma_ops *vops, struct xe_vma_op *op) { int err = 0; + bool res_evict; + + /* + * We only allow evicting a BO within the VM if it is not part of an + * array of binds, as an array of binds can evict another BO within the + * bind. + */ + res_evict = !(vops->flags & XE_VMA_OPS_ARRAY_OF_BINDS); switch (op->base.op) { case DRM_GPUVA_OP_MAP: if (!op->map.invalidate_on_bind) err = vma_lock_and_validate(exec, op->map.vma, + res_evict, !xe_vm_in_fault_mode(vm) || op->map.immediate); break; @@ -2931,11 +2941,13 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm, err = vma_lock_and_validate(exec, gpuva_to_vma(op->base.remap.unmap->va), - false); + res_evict, false); if (!err && op->remap.prev) - err = vma_lock_and_validate(exec, op->remap.prev, true); + err = vma_lock_and_validate(exec, op->remap.prev, + res_evict, true); if (!err && op->remap.next) - err = vma_lock_and_validate(exec, op->remap.next, true); + err = vma_lock_and_validate(exec, op->remap.next, + res_evict, true); break; case DRM_GPUVA_OP_UNMAP: err = check_ufence(gpuva_to_vma(op->base.unmap.va)); @@ -2944,7 +2956,7 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm, err = vma_lock_and_validate(exec, gpuva_to_vma(op->base.unmap.va), - false); + res_evict, false); break; case DRM_GPUVA_OP_PREFETCH: { @@ -2959,7 +2971,7 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm, err = vma_lock_and_validate(exec, gpuva_to_vma(op->base.prefetch.va), - false); + res_evict, false); if (!err && !xe_vma_has_no_bo(vma)) err = xe_bo_migrate(xe_vma_bo(vma), region_to_mem_type[region], @@ -3005,7 +3017,7 @@ static int vm_bind_ioctl_ops_lock_and_prep(struct drm_exec *exec, return err; list_for_each_entry(op, &vops->list, link) { - err = op_lock_and_prep(exec, vm, op); + err = op_lock_and_prep(exec, vm, vops, op); if (err) return err; } @@ -3638,6 +3650,8 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) } xe_vma_ops_init(&vops, vm, q, syncs, num_syncs); + if (args->num_binds > 1) + vops.flags |= XE_VMA_OPS_ARRAY_OF_BINDS; for (i = 0; i < args->num_binds; ++i) { u64 range = bind_ops[i].range; u64 addr = bind_ops[i].addr; diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h index da39940501d8..413353e1c225 100644 --- a/drivers/gpu/drm/xe/xe_vm_types.h +++ b/drivers/gpu/drm/xe/xe_vm_types.h @@ -476,6 +476,7 @@ struct xe_vma_ops { /** @flag: signify the properties within xe_vma_ops*/ #define XE_VMA_OPS_FLAG_HAS_SVM_PREFETCH BIT(0) #define XE_VMA_OPS_FLAG_MADVISE BIT(1) +#define XE_VMA_OPS_ARRAY_OF_BINDS BIT(2) u32 flags; #ifdef TEST_VM_OPS_ERROR /** @inject_error: inject error to test error handling */ From d30203739be798d3de5c84db3060e96f00c54e82 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Thu, 18 Sep 2025 13:58:57 -0700 Subject: [PATCH 03/11] drm/xe: Move rebar to be done earlier MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There may be cases in which the BAR0 also needs to move to accommodate the bigger BAR2. However if it's not released, the BAR2 resize fails. During the vram probe it can't be released as it's already in use by xe_mmio for early register access. Add a new function in xe_vram and let xe_pci call it directly before even early device probe. This allows the BAR2 to resize in cases BAR0 also needs to move, assuming there aren't other reasons to hold that move: [] xe 0000:03:00.0: vgaarb: deactivate vga console [] xe 0000:03:00.0: [drm] Attempting to resize bar from 8192MiB -> 16384MiB [] xe 0000:03:00.0: BAR 0 [mem 0x83000000-0x83ffffff 64bit]: releasing [] xe 0000:03:00.0: BAR 2 [mem 0x4000000000-0x41ffffffff 64bit pref]: releasing [] pcieport 0000:02:01.0: bridge window [mem 0x4000000000-0x41ffffffff 64bit pref]: releasing [] pcieport 0000:01:00.0: bridge window [mem 0x4000000000-0x41ffffffff 64bit pref]: releasing [] pcieport 0000:01:00.0: bridge window [mem 0x4000000000-0x43ffffffff 64bit pref]: assigned [] pcieport 0000:02:01.0: bridge window [mem 0x4000000000-0x43ffffffff 64bit pref]: assigned [] xe 0000:03:00.0: BAR 2 [mem 0x4000000000-0x43ffffffff 64bit pref]: assigned [] xe 0000:03:00.0: BAR 0 [mem 0x83000000-0x83ffffff 64bit]: assigned [] pcieport 0000:00:01.0: PCI bridge to [bus 01-04] [] pcieport 0000:00:01.0: bridge window [mem 0x83000000-0x840fffff] [] pcieport 0000:00:01.0: bridge window [mem 0x4000000000-0x44007fffff 64bit pref] [] pcieport 0000:01:00.0: PCI bridge to [bus 02-04] [] pcieport 0000:01:00.0: bridge window [mem 0x83000000-0x840fffff] [] pcieport 0000:01:00.0: bridge window [mem 0x4000000000-0x43ffffffff 64bit pref] [] pcieport 0000:02:01.0: PCI bridge to [bus 03] [] pcieport 0000:02:01.0: bridge window [mem 0x83000000-0x83ffffff] [] pcieport 0000:02:01.0: bridge window [mem 0x4000000000-0x43ffffffff 64bit pref] [] xe 0000:03:00.0: [drm] BAR2 resized to 16384M [] xe 0000:03:00.0: [drm:xe_pci_probe [xe]] BATTLEMAGE e221:0000 dgfx:1 gfx:Xe2_HPG (20.02) ... For BMG there are additional fix needed in the PCI side, but this helps getting it to a working resize. All the rebar logic is more pci-specific than xe-specific and can be done very early in the probe sequence. In future it would be good to move it out of xe_vram.c, but this refactor is left for later. Cc: Ilpo Järvinen Cc: stable@vger.kernel.org # 6.12+ Link: https://lore.kernel.org/intel-xe/fafda2a3-fc63-ce97-d22b-803f771a4d19@linux.intel.com Reviewed-by: Ilpo Järvinen Link: https://lore.kernel.org/r/20250918-xe-pci-rebar-2-v1-2-6c094702a074@intel.com Signed-off-by: Lucas De Marchi (cherry picked from commit 45e33f220fd625492c11e15733d8e9b4f9db82a4) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_pci.c | 2 ++ drivers/gpu/drm/xe/xe_vram.c | 34 ++++++++++++++++++++++++++-------- drivers/gpu/drm/xe/xe_vram.h | 1 + 3 files changed, 29 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index be91343829dd..9a6df79fc5b6 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -867,6 +867,8 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (err) return err; + xe_vram_resize_bar(xe); + err = xe_device_probe_early(xe); /* * In Boot Survivability mode, no drm card is exposed and driver diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c index b44ebf50fedb..652df7a5f4f6 100644 --- a/drivers/gpu/drm/xe/xe_vram.c +++ b/drivers/gpu/drm/xe/xe_vram.c @@ -26,15 +26,35 @@ #define BAR_SIZE_SHIFT 20 -static void -_resize_bar(struct xe_device *xe, int resno, resource_size_t size) +/* + * Release all the BARs that could influence/block LMEMBAR resizing, i.e. + * assigned IORESOURCE_MEM_64 BARs + */ +static void release_bars(struct pci_dev *pdev) +{ + struct resource *res; + int i; + + pci_dev_for_each_resource(pdev, res, i) { + /* Resource already un-assigned, do not reset it */ + if (!res->parent) + continue; + + /* No need to release unrelated BARs */ + if (!(res->flags & IORESOURCE_MEM_64)) + continue; + + pci_release_resource(pdev, i); + } +} + +static void resize_bar(struct xe_device *xe, int resno, resource_size_t size) { struct pci_dev *pdev = to_pci_dev(xe->drm.dev); int bar_size = pci_rebar_bytes_to_size(size); int ret; - if (pci_resource_len(pdev, resno)) - pci_release_resource(pdev, resno); + release_bars(pdev); ret = pci_resize_resource(pdev, resno, bar_size); if (ret) { @@ -50,7 +70,7 @@ _resize_bar(struct xe_device *xe, int resno, resource_size_t size) * if force_vram_bar_size is set, attempt to set to the requested size * else set to maximum possible size */ -static void resize_vram_bar(struct xe_device *xe) +void xe_vram_resize_bar(struct xe_device *xe) { int force_vram_bar_size = xe_modparam.force_vram_bar_size; struct pci_dev *pdev = to_pci_dev(xe->drm.dev); @@ -119,7 +139,7 @@ static void resize_vram_bar(struct xe_device *xe) pci_read_config_dword(pdev, PCI_COMMAND, &pci_cmd); pci_write_config_dword(pdev, PCI_COMMAND, pci_cmd & ~PCI_COMMAND_MEMORY); - _resize_bar(xe, LMEM_BAR, rebar_size); + resize_bar(xe, LMEM_BAR, rebar_size); pci_assign_unassigned_bus_resources(pdev->bus); pci_write_config_dword(pdev, PCI_COMMAND, pci_cmd); @@ -148,8 +168,6 @@ static int determine_lmem_bar_size(struct xe_device *xe, struct xe_vram_region * return -ENXIO; } - resize_vram_bar(xe); - lmem_bar->io_start = pci_resource_start(pdev, LMEM_BAR); lmem_bar->io_size = pci_resource_len(pdev, LMEM_BAR); if (!lmem_bar->io_size) diff --git a/drivers/gpu/drm/xe/xe_vram.h b/drivers/gpu/drm/xe/xe_vram.h index 72860f714fc6..13505cfb184d 100644 --- a/drivers/gpu/drm/xe/xe_vram.h +++ b/drivers/gpu/drm/xe/xe_vram.h @@ -11,6 +11,7 @@ struct xe_device; struct xe_vram_region; +void xe_vram_resize_bar(struct xe_device *xe); int xe_vram_probe(struct xe_device *xe); struct xe_vram_region *xe_vram_region_alloc(struct xe_device *xe, u8 id, u32 placement); From 1117e7d1e8e66bf7e40291178b829a8513f83a7a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Wed, 10 Sep 2025 18:09:39 +0200 Subject: [PATCH 04/11] drm/xe/migrate: Fix an error path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The exhaustive eviction accidently changed an error path goto to a return. Fix this. Fixes: 59eabff2a352 ("drm/xe: Convert xe_bo_create_pin_map() for exhaustive eviction") Cc: Matthew Brost Signed-off-by: Thomas Hellström Reviewed-by: Francois Dugast Link: https://lore.kernel.org/r/20250910160939.103473-1-thomas.hellstrom@linux.intel.com (cherry picked from commit 381f1ed15159c4b3f00dd37cc70924dedebeb111) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 1d667fa36cf3..569869a2b339 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -434,7 +434,7 @@ int xe_migrate_init(struct xe_migrate *m) err = xe_migrate_lock_prepare_vm(tile, m, vm); if (err) - return err; + goto err_out; if (xe->info.has_usm) { struct xe_hw_engine *hwe = xe_gt_hw_engine(primary_gt, From 7413e9f2be6b2b0caff9c517efa123d988914bba Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Thu, 9 Oct 2025 06:06:29 -0700 Subject: [PATCH 05/11] drm/xe: Handle mixed mappings and existing VRAM on atomic faults Moving to VRAM will fail if mixed mappings are present or if the page is already located in VRAM. Atomic faults that require a move to VRAM currently retry without attempting to evict mixed mappings or locate existing VRAM mappings. This patch fixes the issue by attempting to evict mixed mappings or find existing VRAM pages when a move to VRAM fails during atomic fault handling. Fixes: a9ac0fa455b0 ("drm/xe: Strict migration policy for atomic SVM faults") Signed-off-by: Matthew Brost Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20251009130629.3531962-1-matthew.brost@intel.com (cherry picked from commit 75188605c56d10c1bd3b1cd94f4872f349c3a9c8) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_svm.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 7e2db71ff34e..b268ee0d2271 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -1073,7 +1073,17 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, drm_dbg(&vm->xe->drm, "VRAM allocation failed, falling back to retrying fault, asid=%u, errno=%pe\n", vm->usm.asid, ERR_PTR(err)); - goto retry; + + /* + * In the devmem-only case, mixed mappings may + * be found. The get_pages function will fix + * these up to a single location, allowing the + * page fault handler to make forward progress. + */ + if (ctx.devmem_only) + goto get_pages; + else + goto retry; } else { drm_err(&vm->xe->drm, "VRAM allocation failed, retry count exceeded, asid=%u, errno=%pe\n", @@ -1083,6 +1093,7 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, } } +get_pages: get_pages_start = xe_svm_stats_ktime_get(); range_debug(range, "GET PAGES"); From 1852d27aa998272696680607b65a2ceac966104e Mon Sep 17 00:00:00 2001 From: Vinay Belgaumkar Date: Thu, 9 Oct 2025 18:10:47 -0700 Subject: [PATCH 06/11] drm/xe: Enable media sampler power gating Where applicable, enable media sampler power gating. Also, add it to the powergate_info debugfs. v2: Remove the sampler powergate status since it is cleared quickly anyway. v3: Use vcs mask (Rodrigo) and fix the version check for media v4: Remove extra spaces v5: Media samplers are independent of vcs mask, use Media version 1255 (Matt Roper) Fixes: 38e8c4184ea0 ("drm/xe: Enable Coarse Power Gating") Cc: Rodrigo Vivi Cc: Matt Roper Reviewed-by: Rodrigo Vivi Signed-off-by: Vinay Belgaumkar Link: https://lore.kernel.org/r/20251010011047.2047584-1-vinay.belgaumkar@intel.com Signed-off-by: Rodrigo Vivi (cherry picked from commit 4cbc08649a54c3d533df9832342d52d409dfbbf0) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/regs/xe_gt_regs.h | 1 + drivers/gpu/drm/xe/xe_gt_idle.c | 8 ++++++++ 2 files changed, 9 insertions(+) diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h index 06cb6b02ec64..51f2a03847f9 100644 --- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h @@ -342,6 +342,7 @@ #define POWERGATE_ENABLE XE_REG(0xa210) #define RENDER_POWERGATE_ENABLE REG_BIT(0) #define MEDIA_POWERGATE_ENABLE REG_BIT(1) +#define MEDIA_SAMPLERS_POWERGATE_ENABLE REG_BIT(2) #define VDN_HCP_POWERGATE_ENABLE(n) REG_BIT(3 + 2 * (n)) #define VDN_MFXVDENC_POWERGATE_ENABLE(n) REG_BIT(4 + 2 * (n)) diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c index f8950a52d0a4..bdc9d9877ec4 100644 --- a/drivers/gpu/drm/xe/xe_gt_idle.c +++ b/drivers/gpu/drm/xe/xe_gt_idle.c @@ -124,6 +124,9 @@ void xe_gt_idle_enable_pg(struct xe_gt *gt) if (xe_gt_is_main_type(gt)) gtidle->powergate_enable |= RENDER_POWERGATE_ENABLE; + if (MEDIA_VERx100(xe) >= 1100 && MEDIA_VERx100(xe) < 1255) + gtidle->powergate_enable |= MEDIA_SAMPLERS_POWERGATE_ENABLE; + if (xe->info.platform != XE_DG1) { for (i = XE_HW_ENGINE_VCS0, j = 0; i <= XE_HW_ENGINE_VCS7; ++i, ++j) { if ((gt->info.engine_mask & BIT(i))) @@ -246,6 +249,11 @@ int xe_gt_idle_pg_print(struct xe_gt *gt, struct drm_printer *p) drm_printf(p, "Media Slice%d Power Gate Status: %s\n", n, str_up_down(pg_status & media_slices[n].status_bit)); } + + if (MEDIA_VERx100(xe) >= 1100 && MEDIA_VERx100(xe) < 1255) + drm_printf(p, "Media Samplers Power Gating Enabled: %s\n", + str_yes_no(pg_enabled & MEDIA_SAMPLERS_POWERGATE_ENABLE)); + return 0; } From 9f64b3cd051b825de0a2a9f145c8e003200cedd5 Mon Sep 17 00:00:00 2001 From: Shuicheng Lin Date: Fri, 10 Oct 2025 17:25:29 +0000 Subject: [PATCH 07/11] drm/xe/guc: Check GuC running state before deregistering exec queue In normal operation, a registered exec queue is disabled and deregistered through the GuC, and freed only after the GuC confirms completion. However, if the driver is forced to unbind while the exec queue is still running, the user may call exec_destroy() after the GuC has already been stopped and CT communication disabled. In this case, the driver cannot receive a response from the GuC, preventing proper cleanup of exec queue resources. Fix this by directly releasing the resources when GuC is not running. Here is the failure dmesg log: " [ 468.089581] ---[ end trace 0000000000000000 ]--- [ 468.089608] pci 0000:03:00.0: [drm] *ERROR* GT0: GUC ID manager unclean (1/65535) [ 468.090558] pci 0000:03:00.0: [drm] GT0: total 65535 [ 468.090562] pci 0000:03:00.0: [drm] GT0: used 1 [ 468.090564] pci 0000:03:00.0: [drm] GT0: range 1..1 (1) [ 468.092716] ------------[ cut here ]------------ [ 468.092719] WARNING: CPU: 14 PID: 4775 at drivers/gpu/drm/xe/xe_ttm_vram_mgr.c:298 ttm_vram_mgr_fini+0xf8/0x130 [xe] " v2: use xe_uc_fw_is_running() instead of xe_guc_ct_enabled(). As CT may go down and come back during VF migration. Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Cc: stable@vger.kernel.org Cc: Matthew Brost Signed-off-by: Shuicheng Lin Reviewed-by: Matthew Brost Signed-off-by: Matthew Brost Link: https://lore.kernel.org/r/20251010172529.2967639-2-shuicheng.lin@intel.com (cherry picked from commit 9b42321a02c50a12b2beb6ae9469606257fbecea) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_guc_submit.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 53024eb5670b..94ed8159496f 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -44,6 +44,7 @@ #include "xe_ring_ops_types.h" #include "xe_sched_job.h" #include "xe_trace.h" +#include "xe_uc_fw.h" #include "xe_vm.h" static struct xe_guc * @@ -1489,7 +1490,17 @@ static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg) xe_gt_assert(guc_to_gt(guc), !(q->flags & EXEC_QUEUE_FLAG_PERMANENT)); trace_xe_exec_queue_cleanup_entity(q); - if (exec_queue_registered(q)) + /* + * Expected state transitions for cleanup: + * - If the exec queue is registered and GuC firmware is running, we must first + * disable scheduling and deregister the queue to ensure proper teardown and + * resource release in the GuC, then destroy the exec queue on driver side. + * - If the GuC is already stopped (e.g., during driver unload or GPU reset), + * we cannot expect a response for the deregister request. In this case, + * it is safe to directly destroy the exec queue on driver side, as the GuC + * will not process further requests and all resources must be cleaned up locally. + */ + if (exec_queue_registered(q) && xe_uc_fw_is_running(&guc->fw)) disable_scheduling_deregister(guc, q); else __guc_exec_queue_destroy(guc, q); From 7987b93e3a11a7a95ddf2b21563d3286661b999c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= Date: Fri, 10 Oct 2025 12:41:48 +0200 Subject: [PATCH 08/11] drm/xe/svm: Ensure data will be migrated to system if indicated by madvise. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If the location madvise() is set to DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM, the drm_pagemap in the SVM gpu fault handler will be set to NULL. However there is nothing that explicitly migrates the data to system if it is already present in device memory. In that case, set the device memory owner to NULL to ensure data gets properly migrated to system on page-fault. v2: - Remove redundant dpagemap assignment (Himal Prasad Ghimiray) Signed-off-by: Thomas Hellström Reviewed-by: Matthew Brost #v1 Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20251010104149.72783-2-thomas.hellstrom@linux.intel.com Fixes: 10aa5c806030 ("drm/gpusvm, drm/xe: Fix userptr to not allow device private pages") (cherry picked from commit 2cfcea7a745794f9b8e265a309717ca6ba335fc4) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_svm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index b268ee0d2271..da2a412f80c0 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -1034,6 +1034,9 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, if (err) return err; + dpagemap = xe_vma_resolve_pagemap(vma, tile); + if (!dpagemap && !ctx.devmem_only) + ctx.device_private_page_owner = NULL; range = xe_svm_range_find_or_insert(vm, fault_addr, vma, &ctx); if (IS_ERR(range)) @@ -1054,7 +1057,6 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, range_debug(range, "PAGE FAULT"); - dpagemap = xe_vma_resolve_pagemap(vma, tile); if (--migrate_try_count >= 0 && xe_svm_range_needs_migrate_to_vram(range, vma, !!dpagemap || ctx.devmem_only)) { ktime_t migrate_start = xe_svm_stats_ktime_get(); From 6d36f65ba551d28710c3e1aaceecacf19df0cd8f Mon Sep 17 00:00:00 2001 From: Matt Roper Date: Mon, 13 Oct 2025 08:30:15 -0700 Subject: [PATCH 09/11] drm/xe/kunit: Fix kerneldoc for parameterized tests Kunit's generate_params() was recently updated to take an additional test context parameter. Xe's IP and platform parameter generators were updated accordingly at the same time, but the new parameter was not added to the functions' kerneldoc, resulting in the following warnings: Warning: drivers/gpu/drm/xe/tests/xe_pci.c:78 function parameter 'test' not described in 'xe_pci_fake_data_gen_params' Warning: drivers/gpu/drm/xe/tests/xe_pci.c:254 function parameter 'test' not described in 'xe_pci_graphics_ip_gen_param' Warning: drivers/gpu/drm/xe/tests/xe_pci.c:278 function parameter 'test' not described in 'xe_pci_media_ip_gen_param' Warning: drivers/gpu/drm/xe/tests/xe_pci.c:302 function parameter 'test' not described in 'xe_pci_id_gen_param' Warning: drivers/gpu/drm/xe/tests/xe_pci.c:390 function parameter 'test' not described in 'xe_pci_live_device_gen_param' 5 warnings as errors Document the new parameter to eliminate the warnings and make CI happy. Fixes: b9a214b5f6aa ("kunit: Pass parameterized test context to generate_params()") Reviewed-by: Shuicheng Lin Link: https://lore.kernel.org/r/20251013153014.2362879-2-matthew.d.roper@intel.com Signed-off-by: Matt Roper (cherry picked from commit 89e347f8a70165d1e8d88a93d875da7742c902ce) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/tests/xe_pci.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c index 69e2840c7ef0..663a79ec960d 100644 --- a/drivers/gpu/drm/xe/tests/xe_pci.c +++ b/drivers/gpu/drm/xe/tests/xe_pci.c @@ -66,6 +66,7 @@ KUNIT_ARRAY_PARAM(platform, cases, xe_pci_fake_data_desc); /** * xe_pci_fake_data_gen_params - Generate struct xe_pci_fake_data parameters + * @test: test context object * @prev: the pointer to the previous parameter to iterate from or NULL * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE * @@ -242,6 +243,7 @@ KUNIT_ARRAY_PARAM(pci_id, pciidlist, xe_pci_id_kunit_desc); /** * xe_pci_graphics_ip_gen_param - Generate graphics struct xe_ip parameters + * @test: test context object * @prev: the pointer to the previous parameter to iterate from or NULL * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE * @@ -266,6 +268,7 @@ EXPORT_SYMBOL_IF_KUNIT(xe_pci_graphics_ip_gen_param); /** * xe_pci_media_ip_gen_param - Generate media struct xe_ip parameters + * @test: test context object * @prev: the pointer to the previous parameter to iterate from or NULL * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE * @@ -290,6 +293,7 @@ EXPORT_SYMBOL_IF_KUNIT(xe_pci_media_ip_gen_param); /** * xe_pci_id_gen_param - Generate struct pci_device_id parameters + * @test: test context object * @prev: the pointer to the previous parameter to iterate from or NULL * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE * @@ -376,6 +380,7 @@ EXPORT_SYMBOL_IF_KUNIT(xe_pci_fake_device_init); /** * xe_pci_live_device_gen_param - Helper to iterate Xe devices as KUnit parameters + * @test: test context object * @prev: the previously returned value, or NULL for the first iteration * @desc: the buffer for a parameter name * From 6a91af25cdbce2086d85cc4994cf791bda3a2c90 Mon Sep 17 00:00:00 2001 From: Matthew Auld Date: Fri, 10 Oct 2025 17:20:21 +0100 Subject: [PATCH 10/11] drm/xe/migrate: don't misalign current bytes If current bytes exceeds the max copy size, ensure the clamped size still accounts for the XE_CACHELINE_BYTES alignment, otherwise we trigger the assert in xe_migrate_vram with the size now being out of alignment. Fixes: 8c2d61e0e916 ("drm/xe/migrate: don't overflow max copy size") Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6212 Signed-off-by: Matthew Auld Cc: Stuart Summers Cc: Matthew Brost Reviewed-by: Matthew Brost Link: https://lore.kernel.org/r/20251010162020.190962-2-matthew.auld@intel.com (cherry picked from commit 641bcf8731d21b56760e3646a39a65f471e9efd1) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_migrate.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 569869a2b339..a36ce7dce8cc 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -2113,7 +2113,9 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, if (current_bytes & ~PAGE_MASK) { int pitch = 4; - current_bytes = min_t(int, current_bytes, S16_MAX * pitch); + current_bytes = min_t(int, current_bytes, + round_down(S16_MAX * pitch, + XE_CACHELINE_BYTES)); } __fence = xe_migrate_vram(m, current_bytes, From 225bc03d85427e7e3821d6f99f4f2d4a09350dda Mon Sep 17 00:00:00 2001 From: Matthew Auld Date: Fri, 10 Oct 2025 16:24:58 +0100 Subject: [PATCH 11/11] drm/xe/evict: drop bogus assert MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This assert can trigger here with non pin_map users that select LATE_RESTORE, since the vmap is allowed to be NULL given that save/restore can now use the blitter instead. The check here doesn't seem to have much value anymore given that we no longer move pinned memory, so any existing vmap is left well alone, and doesn't need to be recreated upon restore, so just drop the assert here. Fixes: 86f69c26113c ("drm/xe: use backup object for pinned save/restore") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6213 Signed-off-by: Matthew Auld Cc: Thomas Hellström Cc: Matthew Brost Reviewed-by: Thomas Hellström Link: https://lore.kernel.org/r/20251010152457.177884-2-matthew.auld@intel.com (cherry picked from commit a10b4a69c7f8f596d2c5218fbe84430734fab3b2) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_bo_evict.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo_evict.c b/drivers/gpu/drm/xe/xe_bo_evict.c index d5dbc51e8612..bc5b4c5fab81 100644 --- a/drivers/gpu/drm/xe/xe_bo_evict.c +++ b/drivers/gpu/drm/xe/xe_bo_evict.c @@ -182,7 +182,6 @@ int xe_bo_evict_all(struct xe_device *xe) static int xe_bo_restore_and_map_ggtt(struct xe_bo *bo) { - struct xe_device *xe = xe_bo_device(bo); int ret; ret = xe_bo_restore_pinned(bo); @@ -201,13 +200,6 @@ static int xe_bo_restore_and_map_ggtt(struct xe_bo *bo) } } - /* - * We expect validate to trigger a move VRAM and our move code - * should setup the iosys map. - */ - xe_assert(xe, !(bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE) || - !iosys_map_is_null(&bo->vmap)); - return 0; }