diff options
Diffstat (limited to 'drivers/gpu/drm/i915')
85 files changed, 1905 insertions, 976 deletions
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 8f35be02b80e..a26edcdadc21 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -123,6 +123,7 @@ gt-y += \ gt/intel_ring.o \ gt/intel_ring_submission.o \ gt/intel_rps.o \ + gt/intel_sa_media.o \ gt/intel_sseu.o \ gt/intel_sseu_debugfs.o \ gt/intel_timeline.o \ diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index de3b4fb5d0a5..accf5311b664 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -675,7 +675,7 @@ bool intel_plane_uses_fence(const struct intel_plane_state *plane_state) return DISPLAY_VER(dev_priv) < 4 || (plane->fbc && - plane_state->view.gtt.type == I915_GGTT_VIEW_NORMAL); + plane_state->view.gtt.type == I915_GTT_VIEW_NORMAL); } /* diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index ae06153210b7..884e8e67b17c 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -45,7 +45,7 @@ struct drm_modeset_acquire_ctx; struct drm_plane; struct drm_plane_state; struct i915_address_space; -struct i915_ggtt_view; +struct i915_gtt_view; struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 0d502003228b..298d00a11f47 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -105,7 +105,7 @@ struct intel_fb_view { * In the normal view the FB object's backing store sg list is used * directly and hence the remap information here is not used. */ - struct i915_ggtt_view gtt; + struct i915_gtt_view gtt; /* * The GTT view (gtt.type) specific information for each FB color diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c index b191915ab351..eefa33c555ac 100644 --- a/drivers/gpu/drm/i915/display/intel_fb.c +++ b/drivers/gpu/drm/i915/display/intel_fb.c @@ -1395,7 +1395,7 @@ static u32 calc_plane_remap_info(const struct intel_framebuffer *fb, int color_p plane_view_height_tiles(fb, color_plane, dims, y)); } - if (view->gtt.type == I915_GGTT_VIEW_ROTATED) { + if (view->gtt.type == I915_GTT_VIEW_ROTATED) { drm_WARN_ON(&i915->drm, remap_info->linear); check_array_bounds(i915, view->gtt.rotated.plane, color_plane); @@ -1420,7 +1420,7 @@ static u32 calc_plane_remap_info(const struct intel_framebuffer *fb, int color_p /* rotate the tile dimensions to match the GTT view */ swap(tile_width, tile_height); } else { - drm_WARN_ON(&i915->drm, view->gtt.type != I915_GGTT_VIEW_REMAPPED); + drm_WARN_ON(&i915->drm, view->gtt.type != I915_GTT_VIEW_REMAPPED); check_array_bounds(i915, view->gtt.remapped.plane, color_plane); @@ -1503,12 +1503,12 @@ calc_plane_normal_size(const struct intel_framebuffer *fb, int color_plane, } static void intel_fb_view_init(struct drm_i915_private *i915, struct intel_fb_view *view, - enum i915_ggtt_view_type view_type) + enum i915_gtt_view_type view_type) { memset(view, 0, sizeof(*view)); view->gtt.type = view_type; - if (view_type == I915_GGTT_VIEW_REMAPPED && IS_ALDERLAKE_P(i915)) + if (view_type == I915_GTT_VIEW_REMAPPED && IS_ALDERLAKE_P(i915)) view->gtt.remapped.plane_alignment = SZ_2M / PAGE_SIZE; } @@ -1530,16 +1530,16 @@ int intel_fill_fb_info(struct drm_i915_private *i915, struct intel_framebuffer * int i, num_planes = fb->base.format->num_planes; unsigned int tile_size = intel_tile_size(i915); - intel_fb_view_init(i915, &fb->normal_view, I915_GGTT_VIEW_NORMAL); + intel_fb_view_init(i915, &fb->normal_view, I915_GTT_VIEW_NORMAL); drm_WARN_ON(&i915->drm, intel_fb_supports_90_270_rotation(fb) && intel_fb_needs_pot_stride_remap(fb)); if (intel_fb_supports_90_270_rotation(fb)) - intel_fb_view_init(i915, &fb->rotated_view, I915_GGTT_VIEW_ROTATED); + intel_fb_view_init(i915, &fb->rotated_view, I915_GTT_VIEW_ROTATED); if (intel_fb_needs_pot_stride_remap(fb)) - intel_fb_view_init(i915, &fb->remapped_view, I915_GGTT_VIEW_REMAPPED); + intel_fb_view_init(i915, &fb->remapped_view, I915_GTT_VIEW_REMAPPED); for (i = 0; i < num_planes; i++) { struct fb_plane_view_dims view_dims; @@ -1620,8 +1620,8 @@ static void intel_plane_remap_gtt(struct intel_plane_state *plane_state) u32 gtt_offset = 0; intel_fb_view_init(i915, &plane_state->view, - drm_rotation_90_or_270(rotation) ? I915_GGTT_VIEW_ROTATED : - I915_GGTT_VIEW_REMAPPED); + drm_rotation_90_or_270(rotation) ? I915_GTT_VIEW_ROTATED : + I915_GTT_VIEW_REMAPPED); src_x = plane_state->uapi.src.x1 >> 16; src_y = plane_state->uapi.src.y1 >> 16; diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c index bd6e7c98e751..c86e5d4ee016 100644 --- a/drivers/gpu/drm/i915/display/intel_fb_pin.c +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c @@ -18,7 +18,7 @@ static struct i915_vma * intel_pin_fb_obj_dpt(struct drm_framebuffer *fb, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, bool uses_fence, unsigned long *out_flags, struct i915_address_space *vm) @@ -79,7 +79,7 @@ err: struct i915_vma * intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb, bool phys_cursor, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, bool uses_fence, unsigned long *out_flags) { diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.h b/drivers/gpu/drm/i915/display/intel_fb_pin.h index e4fcd0218d9d..de0efaa25905 100644 --- a/drivers/gpu/drm/i915/display/intel_fb_pin.h +++ b/drivers/gpu/drm/i915/display/intel_fb_pin.h @@ -11,12 +11,12 @@ struct drm_framebuffer; struct i915_vma; struct intel_plane_state; -struct i915_ggtt_view; +struct i915_gtt_view; struct i915_vma * intel_pin_and_fence_fb_obj(struct drm_framebuffer *fb, bool phys_cursor, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, bool uses_fence, unsigned long *out_flags); diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c index b3a254c34fc6..112aa0447a0d 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev.c @@ -198,8 +198,8 @@ static int intelfb_create(struct drm_fb_helper *helper, struct drm_i915_private *dev_priv = to_i915(dev); struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); struct i915_ggtt *ggtt = to_gt(dev_priv)->ggtt; - const struct i915_ggtt_view view = { - .type = I915_GGTT_VIEW_NORMAL, + const struct i915_gtt_view view = { + .type = I915_GTT_VIEW_NORMAL, }; intel_wakeref_t wakeref; struct fb_info *info; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index 1674b0c5802b..d44a152ce680 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -397,7 +397,7 @@ struct i915_vma * i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, struct i915_gem_ww_ctx *ww, u32 alignment, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, unsigned int flags) { struct drm_i915_private *i915 = to_i915(obj->base.dev); @@ -434,7 +434,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, */ vma = ERR_PTR(-ENOSPC); if ((flags & PIN_MAPPABLE) == 0 && - (!view || view->type == I915_GGTT_VIEW_NORMAL)) + (!view || view->type == I915_GTT_VIEW_NORMAL)) vma = i915_gem_object_ggtt_pin_ww(obj, ww, view, 0, alignment, flags | PIN_MAPPABLE | PIN_NONBLOCK); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_lmem.h b/drivers/gpu/drm/i915/gem/i915_gem_lmem.h index 1b88ea13435c..5a7a14e85c3f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_lmem.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_lmem.h @@ -12,8 +12,6 @@ struct drm_i915_private; struct drm_i915_gem_object; struct intel_memory_region; -extern const struct drm_i915_gem_object_ops i915_gem_lmem_obj_ops; - void __iomem * i915_gem_object_lmem_io_map(struct drm_i915_gem_object *obj, unsigned long n, diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 0c5c43852e24..73d9eda1d6b7 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -194,17 +194,17 @@ int i915_gem_mmap_gtt_version(void) return 4; } -static inline struct i915_ggtt_view +static inline struct i915_gtt_view compute_partial_view(const struct drm_i915_gem_object *obj, pgoff_t page_offset, unsigned int chunk) { - struct i915_ggtt_view view; + struct i915_gtt_view view; if (i915_gem_object_is_tiled(obj)) chunk = roundup(chunk, tile_row_pages(obj) ?: 1); - view.type = I915_GGTT_VIEW_PARTIAL; + view.type = I915_GTT_VIEW_PARTIAL; view.partial.offset = rounddown(page_offset, chunk); view.partial.size = min_t(unsigned int, chunk, @@ -212,7 +212,7 @@ compute_partial_view(const struct drm_i915_gem_object *obj, /* If the partial covers the entire object, just create a normal VMA. */ if (chunk >= obj->base.size >> PAGE_SHIFT) - view.type = I915_GGTT_VIEW_NORMAL; + view.type = I915_GTT_VIEW_NORMAL; return view; } @@ -341,12 +341,12 @@ retry: PIN_NOEVICT); if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK)) { /* Use a partial view if it is bigger than available space */ - struct i915_ggtt_view view = + struct i915_gtt_view view = compute_partial_view(obj, page_offset, MIN_CHUNK_PAGES); unsigned int flags; flags = PIN_MAPPABLE | PIN_NOSEARCH; - if (view.type == I915_GGTT_VIEW_NORMAL) + if (view.type == I915_GTT_VIEW_NORMAL) flags |= PIN_NONBLOCK; /* avoid warnings for pinned */ /* @@ -357,7 +357,7 @@ retry: vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); if (IS_ERR(vma) && vma != ERR_PTR(-EDEADLK)) { flags = PIN_MAPPABLE; - view.type = I915_GGTT_VIEW_PARTIAL; + view.type = I915_GTT_VIEW_PARTIAL; vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); } @@ -394,7 +394,7 @@ retry: /* Finally, remap it using the new GTT offset */ ret = remap_io_mapping(area, - area->vm_start + (vma->ggtt_view.partial.offset << PAGE_SHIFT), + area->vm_start + (vma->gtt_view.partial.offset << PAGE_SHIFT), (ggtt->gmadr.start + vma->node.start) >> PAGE_SHIFT, min_t(u64, vma->size, area->vm_end - area->vm_start), &ggtt->iomap); @@ -413,7 +413,7 @@ retry: vma->mmo = mmo; if (CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND) - intel_wakeref_auto(&to_gt(i915)->ggtt->userfault_wakeref, + intel_wakeref_auto(&to_gt(i915)->userfault_wakeref, msecs_to_jiffies_timeout(CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND)); if (write) { @@ -550,6 +550,20 @@ out: intel_runtime_pm_put(&i915->runtime_pm, wakeref); } +void i915_gem_object_runtime_pm_release_mmap_offset(struct drm_i915_gem_object *obj) +{ + struct ttm_buffer_object *bo = i915_gem_to_ttm(obj); + struct ttm_device *bdev = bo->bdev; + + drm_vma_node_unmap(&bo->base.vma_node, bdev->dev_mapping); + + if (obj->userfault_count) { + /* rpm wakeref provide exclusive access */ + list_del(&obj->userfault_link); + obj->userfault_count = 0; + } +} + void i915_gem_object_release_mmap_offset(struct drm_i915_gem_object *obj) { struct i915_mmap_offset *mmo, *mn; @@ -573,6 +587,13 @@ void i915_gem_object_release_mmap_offset(struct drm_i915_gem_object *obj) spin_lock(&obj->mmo.lock); } spin_unlock(&obj->mmo.lock); + + if (obj->userfault_count) { + mutex_lock(&to_gt(to_i915(obj->base.dev))->lmem_userfault_lock); + list_del(&obj->userfault_link); + mutex_unlock(&to_gt(to_i915(obj->base.dev))->lmem_userfault_lock); + obj->userfault_count = 0; + } } static struct i915_mmap_offset * diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.h b/drivers/gpu/drm/i915/gem/i915_gem_mman.h index efee9e0d2508..1fa91b3033b3 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.h @@ -27,6 +27,7 @@ int i915_gem_dumb_mmap_offset(struct drm_file *file_priv, void __i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj); void i915_gem_object_release_mmap_gtt(struct drm_i915_gem_object *obj); +void i915_gem_object_runtime_pm_release_mmap_offset(struct drm_i915_gem_object *obj); void i915_gem_object_release_mmap_offset(struct drm_i915_gem_object *obj); #endif diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 389e9f157ca5..7ff9c7877bec 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -238,7 +238,7 @@ static void __i915_gem_object_free_mmaps(struct drm_i915_gem_object *obj) { /* Skip serialisation and waking the device if known to be not used. */ - if (obj->userfault_count) + if (obj->userfault_count && !IS_DGFX(to_i915(obj->base.dev))) i915_gem_object_release_mmap_gtt(obj); if (!RB_EMPTY_ROOT(&obj->mmo.offsets)) { @@ -723,6 +723,9 @@ bool i915_gem_object_needs_ccs_pages(struct drm_i915_gem_object *obj) bool lmem_placement = false; int i; + if (!HAS_FLAT_CCS(to_i915(obj->base.dev))) + return false; + for (i = 0; i < obj->mm.n_placements; i++) { /* Compression is not allowed for the objects with smem placement */ if (obj->mm.placements[i]->type == INTEL_MEMORY_SYSTEM) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 6f0a3ce35567..7317d4102955 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -543,7 +543,7 @@ struct i915_vma * __must_check i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, struct i915_gem_ww_ctx *ww, u32 alignment, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, unsigned int flags); void i915_gem_object_make_unshrinkable(struct drm_i915_gem_object *obj); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index 9f6b14ec189a..40305e2bcd49 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -298,7 +298,8 @@ struct drm_i915_gem_object { }; /** - * Whether the object is currently in the GGTT mmap. + * Whether the object is currently in the GGTT or any other supported + * fake offset mmap backed by lmem. */ unsigned int userfault_count; struct list_head userfault_link; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c index 00359ec9d58b..3428f735e786 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c @@ -24,7 +24,7 @@ void i915_gem_suspend(struct drm_i915_private *i915) { GEM_TRACE("%s\n", dev_name(i915->drm.dev)); - intel_wakeref_auto(&to_gt(i915)->ggtt->userfault_wakeref, 0); + intel_wakeref_auto(&to_gt(i915)->userfault_wakeref, 0); flush_workqueue(i915->wq); /* diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index 4f4c9461a23b..acc561c0f0aa 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -430,48 +430,29 @@ static int i915_gem_init_stolen(struct intel_memory_region *mem) reserved_base = stolen_top; reserved_size = 0; - switch (GRAPHICS_VER(i915)) { - case 2: - case 3: - break; - case 4: - if (!IS_G4X(i915)) - break; - fallthrough; - case 5: - g4x_get_stolen_reserved(i915, uncore, + if (GRAPHICS_VER(i915) >= 11) { + icl_get_stolen_reserved(i915, uncore, &reserved_base, &reserved_size); - break; - case 6: - gen6_get_stolen_reserved(i915, uncore, - &reserved_base, &reserved_size); - break; - case 7: - if (IS_VALLEYVIEW(i915)) - vlv_get_stolen_reserved(i915, uncore, - &reserved_base, &reserved_size); - else - gen7_get_stolen_reserved(i915, uncore, - &reserved_base, &reserved_size); - break; - case 8: - case 9: + } else if (GRAPHICS_VER(i915) >= 8) { if (IS_LP(i915)) chv_get_stolen_reserved(i915, uncore, &reserved_base, &reserved_size); else bdw_get_stolen_reserved(i915, uncore, &reserved_base, &reserved_size); - break; - default: - MISSING_CASE(GRAPHICS_VER(i915)); - fallthrough; - case 11: - case 12: - icl_get_stolen_reserved(i915, uncore, - &reserved_base, - &reserved_size); - break; + } else if (GRAPHICS_VER(i915) >= 7) { + if (IS_VALLEYVIEW(i915)) + vlv_get_stolen_reserved(i915, uncore, + &reserved_base, &reserved_size); + else + gen7_get_stolen_reserved(i915, uncore, + &reserved_base, &reserved_size); + } else if (GRAPHICS_VER(i915) >= 6) { + gen6_get_stolen_reserved(i915, uncore, + &reserved_base, &reserved_size); + } else if (GRAPHICS_VER(i915) >= 5 || IS_G4X(i915)) { + g4x_get_stolen_reserved(i915, uncore, + &reserved_base, &reserved_size); } /* diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index bc9c432edffe..0544b0a4a43a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -297,7 +297,7 @@ static struct ttm_tt *i915_ttm_tt_create(struct ttm_buffer_object *bo, i915_tt->is_shmem = true; } - if (HAS_FLAT_CCS(i915) && i915_gem_object_needs_ccs_pages(obj)) + if (i915_gem_object_needs_ccs_pages(obj)) ccs_pages = DIV_ROUND_UP(DIV_ROUND_UP(bo->base.size, NUM_BYTES_PER_CCS_BYTE), PAGE_SIZE); @@ -509,9 +509,18 @@ static int i915_ttm_shrink(struct drm_i915_gem_object *obj, unsigned int flags) static void i915_ttm_delete_mem_notify(struct ttm_buffer_object *bo) { struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo); + intel_wakeref_t wakeref = 0; if (likely(obj)) { + /* ttm_bo_release() already has dma_resv_lock */ + if (i915_ttm_cpu_maps_iomem(bo->resource)) + wakeref = intel_runtime_pm_get(&to_i915(obj->base.dev)->runtime_pm); + __i915_gem_object_pages_fini(obj); + + if (wakeref) + intel_runtime_pm_put(&to_i915(obj->base.dev)->runtime_pm, wakeref); + i915_ttm_free_cached_io_rsgt(obj); } } @@ -981,6 +990,7 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) struct ttm_buffer_object *bo = area->vm_private_data; struct drm_device *dev = bo->base.dev; struct drm_i915_gem_object *obj; + intel_wakeref_t wakeref = 0; vm_fault_t ret; int idx; @@ -1002,6 +1012,9 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) return VM_FAULT_SIGBUS; } + if (i915_ttm_cpu_maps_iomem(bo->resource)) + wakeref = intel_runtime_pm_get(&to_i915(obj->base.dev)->runtime_pm); + if (!i915_ttm_resource_mappable(bo->resource)) { int err = -ENODEV; int i; @@ -1023,7 +1036,8 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) if (err) { drm_dbg(dev, "Unable to make resource CPU accessible\n"); dma_resv_unlock(bo->base.resv); - return VM_FAULT_SIGBUS; + ret = VM_FAULT_SIGBUS; + goto out_rpm; } } @@ -1034,12 +1048,30 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) } else { ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot); } + if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) - return ret; + goto out_rpm; + + /* ttm_bo_vm_reserve() already has dma_resv_lock */ + if (ret == VM_FAULT_NOPAGE && wakeref && !obj->userfault_count) { + obj->userfault_count = 1; + mutex_lock(&to_gt(to_i915(obj->base.dev))->lmem_userfault_lock); + list_add(&obj->userfault_link, &to_gt(to_i915(obj->base.dev))->lmem_userfault_list); + mutex_unlock(&to_gt(to_i915(obj->base.dev))->lmem_userfault_lock); + } + + if (wakeref & CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND) + intel_wakeref_auto(&to_gt(to_i915(obj->base.dev))->userfault_wakeref, + msecs_to_jiffies_timeout(CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND)); i915_ttm_adjust_lru(obj); dma_resv_unlock(bo->base.resv); + +out_rpm: + if (wakeref) + intel_runtime_pm_put(&to_i915(obj->base.dev)->runtime_pm, wakeref); + return ret; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm_pm.c index 9aad84059d56..07e49f22f2de 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm_pm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm_pm.c @@ -79,7 +79,12 @@ static int i915_ttm_backup(struct i915_gem_apply_to_region *apply, goto out_no_populate; err = i915_gem_obj_copy_ttm(backup, obj, pm_apply->allow_gpu, false); - GEM_WARN_ON(err); + if (err) { + drm_err(&i915->drm, + "Unable to copy from device to system memory, err:%pe\n", + ERR_PTR(err)); + goto out_no_populate; + } ttm_bo_wait_ctx(backup_bo, &ctx); obj->ttm.backup = backup; diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c index 13b088cc787e..a666d7e610f5 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_coherency.c @@ -434,5 +434,5 @@ int i915_gem_coherency_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_gem_coherency), }; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c index 62c61af77a42..51ed824b020c 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c @@ -476,5 +476,5 @@ int i915_gem_dmabuf_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_dmabuf_import_same_driver_lmem_smem), }; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 25bb3bd4c3fe..b73c91aa5450 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -93,7 +93,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, { const unsigned long npages = obj->base.size / PAGE_SIZE; struct drm_i915_private *i915 = to_i915(obj->base.dev); - struct i915_ggtt_view view; + struct i915_gtt_view view; struct i915_vma *vma; unsigned long page; u32 __iomem *io; @@ -210,7 +210,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj, } for_each_prime_number_from(page, 1, npages) { - struct i915_ggtt_view view = + struct i915_gtt_view view = compute_partial_view(obj, page, MIN_CHUNK_PAGES); u32 __iomem *io; struct page *p; @@ -1844,5 +1844,5 @@ int i915_gem_mman_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_mmap_gpu), }; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c index fe0a890775e2..bdf5bb40ccf1 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_object.c @@ -95,5 +95,5 @@ int i915_gem_object_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_gem_huge), }; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c index 98645797962f..e49fa6fa6aee 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c @@ -165,10 +165,12 @@ static u32 preparser_disable(bool state) return MI_ARB_CHECK | 1 << 8 | state; } -u32 *gen12_emit_aux_table_inv(u32 *cs, const i915_reg_t inv_reg) +u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv_reg) { + u32 gsi_offset = gt->uncore->gsi_offset; + *cs++ = MI_LOAD_REGISTER_IMM(1) | MI_LRI_MMIO_REMAP_EN; - *cs++ = i915_mmio_reg_offset(inv_reg); + *cs++ = i915_mmio_reg_offset(inv_reg) + gsi_offset; *cs++ = AUX_INV; *cs++ = MI_NOOP; @@ -254,7 +256,8 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) if (!HAS_FLAT_CCS(rq->engine->i915)) { /* hsdes: 1809175790 */ - cs = gen12_emit_aux_table_inv(cs, GEN12_GFX_CCS_AUX_NV); + cs = gen12_emit_aux_table_inv(rq->engine->gt, + cs, GEN12_GFX_CCS_AUX_NV); } *cs++ = preparser_disable(false); @@ -313,9 +316,11 @@ int gen12_emit_flush_xcs(struct i915_request *rq, u32 mode) if (aux_inv) { /* hsdes: 1809175790 */ if (rq->engine->class == VIDEO_DECODE_CLASS) - cs = gen12_emit_aux_table_inv(cs, GEN12_VD0_AUX_NV); + cs = gen12_emit_aux_table_inv(rq->engine->gt, + cs, GEN12_VD0_AUX_NV); else - cs = gen12_emit_aux_table_inv(cs, GEN12_VE0_AUX_NV); + cs = gen12_emit_aux_table_inv(rq->engine->gt, + cs, GEN12_VE0_AUX_NV); } if (mode & EMIT_INVALIDATE) diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.h b/drivers/gpu/drm/i915/gt/gen8_engine_cs.h index 32e3d2b831bb..e4d24c811dd6 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.h +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.h @@ -13,6 +13,7 @@ #include "intel_gt_regs.h" #include "intel_gpu_commands.h" +struct intel_gt; struct i915_request; int gen8_emit_flush_rcs(struct i915_request *rq, u32 mode); @@ -45,7 +46,7 @@ u32 *gen8_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); u32 *gen11_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); -u32 *gen12_emit_aux_table_inv(u32 *cs, const i915_reg_t inv_reg); +u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv_reg); static inline u32 * __gen8_emit_pipe_control(u32 *batch, u32 flags0, u32 flags1, u32 offset) diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index ca483984d780..1f7188129cd1 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -654,16 +654,83 @@ bool gen11_vdbox_has_sfc(struct intel_gt *gt, */ if ((gt->info.sfc_mask & BIT(physical_vdbox / 2)) == 0) return false; - else if (GRAPHICS_VER(i915) == 12) + else if (MEDIA_VER(i915) >= 12) return (physical_vdbox % 2 == 0) || !(BIT(physical_vdbox - 1) & vdbox_mask); - else if (GRAPHICS_VER(i915) == 11) + else if (MEDIA_VER(i915) == 11) return logical_vdbox % 2 == 0; - MISSING_CASE(GRAPHICS_VER(i915)); return false; } +static void engine_mask_apply_media_fuses(struct intel_gt *gt) +{ + struct drm_i915_private *i915 = gt->i915; + unsigned int logical_vdbox = 0; + unsigned int i; + u32 media_fuse, fuse1; + u16 vdbox_mask; + u16 vebox_mask; + + if (MEDIA_VER(gt->i915) < 11) + return; + + /* + * On newer platforms the fusing register is called 'enable' and has + * enable semantics, while on older platforms it is called 'disable' + * and bits have disable semantices. + */ + media_fuse = intel_uncore_read(gt->uncore, GEN11_GT_VEBOX_VDBOX_DISABLE); + if (MEDIA_VER_FULL(i915) < IP_VER(12, 50)) + media_fuse = ~media_fuse; + + vdbox_mask = media_fuse & GEN11_GT_VDBOX_DISABLE_MASK; + vebox_mask = (media_fuse & GEN11_GT_VEBOX_DISABLE_MASK) >> + GEN11_GT_VEBOX_DISABLE_SHIFT; + + if (MEDIA_VER_FULL(i915) >= IP_VER(12, 50)) { + fuse1 = intel_uncore_read(gt->uncore, HSW_PAVP_FUSE1); + gt->info.sfc_mask = REG_FIELD_GET(XEHP_SFC_ENABLE_MASK, fuse1); + } else { + gt->info.sfc_mask = ~0; + } + + for (i = 0; i < I915_MAX_VCS; i++) { + if (!HAS_ENGINE(gt, _VCS(i))) { + vdbox_mask &= ~BIT(i); + continue; + } + + if (!(BIT(i) & vdbox_mask)) { + gt->info.engine_mask &= ~BIT(_VCS(i)); + drm_dbg(&i915->drm, "vcs%u fused off\n", i); + continue; + } + + if (gen11_vdbox_has_sfc(gt, i, logical_vdbox, vdbox_mask)) + gt->info.vdbox_sfc_access |= BIT(i); + logical_vdbox++; + } + drm_dbg(&i915->drm, "vdbox enable: %04x, instances: %04lx\n", + vdbox_mask, VDBOX_MASK(gt)); + GEM_BUG_ON(vdbox_mask != VDBOX_MASK(gt)); + + for (i = 0; i < I915_MAX_VECS; i++) { + if (!HAS_ENGINE(gt, _VECS(i))) { + vebox_mask &= ~BIT(i); + continue; + } + + if (!(BIT(i) & vebox_mask)) { + gt->info.engine_mask &= ~BIT(_VECS(i)); + drm_dbg(&i915->drm, "vecs%u fused off\n", i); + } + } + drm_dbg(&i915->drm, "vebox enable: %04x, instances: %04lx\n", + vebox_mask, VEBOX_MASK(gt)); + GEM_BUG_ON(vebox_mask != VEBOX_MASK(gt)); +} + static void engine_mask_apply_compute_fuses(struct intel_gt *gt) { struct drm_i915_private *i915 = gt->i915; @@ -672,7 +739,10 @@ static void engine_mask_apply_compute_fuses(struct intel_gt *gt) unsigned long ccs_mask; unsigned int i; - if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 50)) + if (GRAPHICS_VER(i915) < 11) + return; + + if (hweight32(CCS_MASK(gt)) <= 1) return; ccs_mask = intel_slicemask_from_xehp_dssmask(info->sseu.compute_subslice_mask, @@ -694,6 +764,10 @@ static void engine_mask_apply_copy_fuses(struct intel_gt *gt) unsigned long meml3_mask; unsigned long quad; + if (!(GRAPHICS_VER_FULL(i915) >= IP_VER(12, 60) && + GRAPHICS_VER_FULL(i915) < IP_VER(12, 70))) + return; + meml3_mask = intel_uncore_read(gt->uncore, GEN10_MIRROR_FUSE3); meml3_mask = REG_FIELD_GET(GEN12_MEML3_EN_MASK, meml3_mask); @@ -727,75 +801,11 @@ static void engine_mask_apply_copy_fuses(struct intel_gt *gt) */ static intel_engine_mask_t init_engine_mask(struct intel_gt *gt) { - struct drm_i915_private *i915 = gt->i915; struct intel_gt_info *info = >->info; - struct intel_uncore *uncore = gt->uncore; - unsigned int logical_vdbox = 0; - unsigned int i; - u32 media_fuse, fuse1; - u16 vdbox_mask; - u16 vebox_mask; - - info->engine_mask = RUNTIME_INFO(i915)->platform_engine_mask; - - if (GRAPHICS_VER(i915) < 11) - return info->engine_mask; - /* - * On newer platforms the fusing register is called 'enable' and has - * enable semantics, while on older platforms it is called 'disable' - * and bits have disable semantices. - */ - media_fuse = intel_uncore_read(uncore, GEN11_GT_VEBOX_VDBOX_DISABLE); - if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 50)) - media_fuse = ~media_fuse; - - vdbox_mask = media_fuse & GEN11_GT_VDBOX_DISABLE_MASK; - vebox_mask = (media_fuse & GEN11_GT_VEBOX_DISABLE_MASK) >> - GEN11_GT_VEBOX_DISABLE_SHIFT; - - if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 50)) { - fuse1 = intel_uncore_read(uncore, HSW_PAVP_FUSE1); - gt->info.sfc_mask = REG_FIELD_GET(XEHP_SFC_ENABLE_MASK, fuse1); - } else { - gt->info.sfc_mask = ~0; - } - - for (i = 0; i < I915_MAX_VCS; i++) { - if (!HAS_ENGINE(gt, _VCS(i))) { - vdbox_mask &= ~BIT(i); - continue; - } - - if (!(BIT(i) & vdbox_mask)) { - info->engine_mask &= ~BIT(_VCS(i)); - drm_dbg(&i915->drm, "vcs%u fused off\n", i); - continue; - } - - if (gen11_vdbox_has_sfc(gt, i, logical_vdbox, vdbox_mask)) - gt->info.vdbox_sfc_access |= BIT(i); - logical_vdbox++; - } - drm_dbg(&i915->drm, "vdbox enable: %04x, instances: %04lx\n", - vdbox_mask, VDBOX_MASK(gt)); - GEM_BUG_ON(vdbox_mask != VDBOX_MASK(gt)); - - for (i = 0; i < I915_MAX_VECS; i++) { - if (!HAS_ENGINE(gt, _VECS(i))) { - vebox_mask &= ~BIT(i); - continue; - } - - if (!(BIT(i) & vebox_mask)) { - info->engine_mask &= ~BIT(_VECS(i)); - drm_dbg(&i915->drm, "vecs%u fused off\n", i); - } - } - drm_dbg(&i915->drm, "vebox enable: %04x, instances: %04lx\n", - vebox_mask, VEBOX_MASK(gt)); - GEM_BUG_ON(vebox_mask != VEBOX_MASK(gt)); + GEM_BUG_ON(!info->engine_mask); + engine_mask_apply_media_fuses(gt); engine_mask_apply_compute_fuses(gt); engine_mask_apply_copy_fuses(gt); @@ -1688,9 +1698,9 @@ bool intel_engine_irq_enable(struct intel_engine_cs *engine) return false; /* Caller disables interrupts */ - spin_lock(&engine->gt->irq_lock); + spin_lock(engine->gt->irq_lock); engine->irq_enable(engine); - spin_unlock(&engine->gt->irq_lock); + spin_unlock(engine->gt->irq_lock); return true; } @@ -1701,9 +1711,9 @@ void intel_engine_irq_disable(struct intel_engine_cs *engine) return; /* Caller disables interrupts */ - spin_lock(&engine->gt->irq_lock); + spin_lock(engine->gt->irq_lock); engine->irq_disable(engine); - spin_unlock(&engine->gt->irq_lock); + spin_unlock(engine->gt->irq_lock); } void intel_engines_reset_default_submission(struct intel_gt *gt) diff --git a/drivers/gpu/drm/i915/gt/intel_engine_regs.h b/drivers/gpu/drm/i915/gt/intel_engine_regs.h index 889f0df3940b..fe1a0d5fd4b1 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_engine_regs.h @@ -110,6 +110,7 @@ #define RING_SBBSTATE(base) _MMIO((base) + 0x118) /* hsw+ */ #define RING_SBBADDR_UDW(base) _MMIO((base) + 0x11c) /* gen8+ */ #define RING_BBADDR(base) _MMIO((base) + 0x140) +#define RING_BB_OFFSET(base) _MMIO((base) + 0x158) #define RING_BBADDR_UDW(base) _MMIO((base) + 0x168) /* gen8+ */ #define CCID(base) _MMIO((base) + 0x180) #define CCID_EN BIT(0) diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c index cf4a326f5f48..ea775e601686 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c @@ -842,7 +842,6 @@ void intel_ggtt_init_fences(struct i915_ggtt *ggtt) INIT_LIST_HEAD(&ggtt->fence_list); INIT_LIST_HEAD(&ggtt->userfault_list); - intel_wakeref_auto_init(&ggtt->userfault_wakeref, uncore->rpm); detect_bit_6_swizzle(ggtt); diff --git a/drivers/gpu/drm/i915/gt/intel_gsc.c b/drivers/gpu/drm/i915/gt/intel_gsc.c index 0e494028b81d..7af6db3194dd 100644 --- a/drivers/gpu/drm/i915/gt/intel_gsc.c +++ b/drivers/gpu/drm/i915/gt/intel_gsc.c @@ -7,6 +7,7 @@ #include <linux/mei_aux.h> #include "i915_drv.h" #include "i915_reg.h" +#include "gem/i915_gem_region.h" #include "gt/intel_gsc.h" #include "gt/intel_gt.h" @@ -36,10 +37,56 @@ static int gsc_irq_init(int irq) return irq_set_chip_data(irq, NULL); } +static int +gsc_ext_om_alloc(struct intel_gsc *gsc, struct intel_gsc_intf *intf, size_t size) +{ + struct intel_gt *gt = gsc_to_gt(gsc); + struct drm_i915_gem_object *obj; + int err; + + obj = i915_gem_object_create_lmem(gt->i915, size, + I915_BO_ALLOC_CONTIGUOUS | + I915_BO_ALLOC_CPU_CLEAR); + if (IS_ERR(obj)) { + drm_err(>->i915->drm, "Failed to allocate gsc memory\n"); + return PTR_ERR(obj); + } + + err = i915_gem_object_pin_pages_unlocked(obj); + if (err) { + drm_err(>->i915->drm, "Failed to pin pages for gsc memory\n"); + goto out_put; + } + + intf->gem_obj = obj; + + return 0; + +out_put: + i915_gem_object_put(obj); + return err; +} + +static void gsc_ext_om_destroy(struct intel_gsc_intf *intf) +{ + struct drm_i915_gem_object *obj = fetch_and_zero(&intf->gem_obj); + + if (!obj) + return; + + if (i915_gem_object_has_pinned_pages(obj)) + i915_gem_object_unpin_pages(obj); + + i915_gem_object_put(obj); +} + struct gsc_def { const char *name; unsigned long bar; size_t bar_size; + bool use_polling; + bool slow_firmware; + size_t lmem_size; }; /* gsc resources and definitions (HECI1 and HECI2) */ @@ -54,11 +101,25 @@ static const struct gsc_def gsc_def_dg1[] = { } }; +static const struct gsc_def gsc_def_xehpsdv[] = { + { + /* HECI1 not enabled on the device. */ + }, + { + .name = "mei-gscfi", + .bar = DG1_GSC_HECI2_BASE, + .bar_size = GSC_BAR_LENGTH, + .use_polling = true, + .slow_firmware = true, + } +}; + static const struct gsc_def gsc_def_dg2[] = { { .name = "mei-gsc", .bar = DG2_GSC_HECI1_BASE, .bar_size = GSC_BAR_LENGTH, + .lmem_size = SZ_4M, }, { .name = "mei-gscfi", @@ -75,26 +136,32 @@ static void gsc_release_dev(struct device *dev) kfree(adev); } -static void gsc_destroy_one(struct intel_gsc_intf *intf) +static void gsc_destroy_one(struct drm_i915_private *i915, + struct intel_gsc *gsc, unsigned int intf_id) { + struct intel_gsc_intf *intf = &gsc->intf[intf_id]; + if (intf->adev) { auxiliary_device_delete(&intf->adev->aux_dev); auxiliary_device_uninit(&intf->adev->aux_dev); intf->adev = NULL; } + if (intf->irq >= 0) irq_free_desc(intf->irq); intf->irq = -1; + + gsc_ext_om_destroy(intf); } -static void gsc_init_one(struct drm_i915_private *i915, - struct intel_gsc_intf *intf, +static void gsc_init_one(struct drm_i915_private *i915, struct intel_gsc *gsc, unsigned int intf_id) { struct pci_dev *pdev = to_pci_dev(i915->drm.dev); struct mei_aux_device *adev; struct auxiliary_device *aux_dev; const struct gsc_def *def; + struct intel_gsc_intf *intf = &gsc->intf[intf_id]; int ret; intf->irq = -1; @@ -105,6 +172,8 @@ static void gsc_init_one(struct drm_i915_private *i915, if (IS_DG1(i915)) { def = &gsc_def_dg1[intf_id]; + } else if (IS_XEHPSDV(i915)) { + def = &gsc_def_xehpsdv[intf_id]; } else if (IS_DG2(i915)) { def = &gsc_def_dg2[intf_id]; } else { @@ -117,10 +186,14 @@ static void gsc_init_one(struct drm_i915_private *i915, return; } + /* skip irq initialization */ + if (def->use_polling) + goto add_device; + intf->irq = irq_alloc_desc(0); if (intf->irq < 0) { drm_err(&i915->drm, "gsc irq error %d\n", intf->irq); - return; + goto fail; } ret = gsc_irq_init(intf->irq); @@ -129,16 +202,31 @@ static void gsc_init_one(struct drm_i915_private *i915, goto fail; } +add_device: adev = kzalloc(sizeof(*adev), GFP_KERNEL); if (!adev) goto fail; + if (def->lmem_size) { + drm_dbg(&i915->drm, "setting up GSC lmem\n"); + + if (gsc_ext_om_alloc(gsc, intf, def->lmem_size)) { + drm_err(&i915->drm, "setting up gsc extended operational memory failed\n"); + kfree(adev); + goto fail; + } + + adev->ext_op_mem.start = i915_gem_object_get_dma_address(intf->gem_obj, 0); + adev->ext_op_mem.end = adev->ext_op_mem.start + def->lmem_size; + } + adev->irq = intf->irq; adev->bar.parent = &pdev->resource[0]; adev->bar.start = def->bar + pdev->resource[0].start; adev->bar.end = adev->bar.start + def->bar_size - 1; adev->bar.flags = IORESOURCE_MEM; adev->bar.desc = IORES_DESC_NONE; + adev->slow_firmware = def->slow_firmware; aux_dev = &adev->aux_dev; aux_dev->name = def->name; @@ -165,7 +253,7 @@ static void gsc_init_one(struct drm_i915_private *i915, return; fail: - gsc_destroy_one(intf); + gsc_destroy_one(i915, gsc, intf->id); } static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id) @@ -182,10 +270,8 @@ static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id) return; } - if (gt->gsc.intf[intf_id].irq < 0) { - drm_err_ratelimited(>->i915->drm, "GSC irq: irq not set"); + if (gt->gsc.intf[intf_id].irq < 0) return; - } ret = generic_handle_irq(gt->gsc.intf[intf_id].irq); if (ret) @@ -208,7 +294,7 @@ void intel_gsc_init(struct intel_gsc *gsc, struct drm_i915_private *i915) return; for (i = 0; i < INTEL_GSC_NUM_INTERFACES; i++) - gsc_init_one(i915, &gsc->intf[i], i); + gsc_init_one(i915, gsc, i); } void intel_gsc_fini(struct intel_gsc *gsc) @@ -220,5 +306,5 @@ void intel_gsc_fini(struct intel_gsc *gsc) return; for (i = 0; i < INTEL_GSC_NUM_INTERFACES; i++) - gsc_destroy_one(&gsc->intf[i]); + gsc_destroy_one(gt->i915, gsc, i); } diff --git a/drivers/gpu/drm/i915/gt/intel_gsc.h b/drivers/gpu/drm/i915/gt/intel_gsc.h index 68582f912b21..fcac1775e9c3 100644 --- a/drivers/gpu/drm/i915/gt/intel_gsc.h +++ b/drivers/gpu/drm/i915/gt/intel_gsc.h @@ -20,11 +20,14 @@ struct mei_aux_device; /** * struct intel_gsc - graphics security controller + * + * @gem_obj: scratch memory GSC operations * @intf : gsc interface */ struct intel_gsc { struct intel_gsc_intf { struct mei_aux_device *adev; + struct drm_i915_gem_object *gem_obj; int irq; unsigned int id; } intf[INTEL_GSC_NUM_INTERFACES]; diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index e4bac2431e41..d0b03a928b9a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -31,14 +31,17 @@ #include "intel_rc6.h" #include "intel_renderstate.h" #include "intel_rps.h" +#include "intel_sa_media.h" #include "intel_gt_sysfs.h" #include "intel_uncore.h" #include "shmem_utils.h" -static void __intel_gt_init_early(struct intel_gt *gt) +void intel_gt_common_init_early(struct intel_gt *gt) { - spin_lock_init(>->irq_lock); + spin_lock_init(gt->irq_lock); + INIT_LIST_HEAD(>->lmem_userfault_list); + mutex_init(>->lmem_userfault_lock); INIT_LIST_HEAD(>->closed_vma); spin_lock_init(>->closed_lock); @@ -58,14 +61,19 @@ static void __intel_gt_init_early(struct intel_gt *gt) } /* Preliminary initialization of Tile 0 */ -void intel_root_gt_init_early(struct drm_i915_private *i915) +int intel_root_gt_init_early(struct drm_i915_private *i915) { struct intel_gt *gt = to_gt(i915); gt->i915 = i915; gt->uncore = &i915->uncore; + gt->irq_lock = drmm_kzalloc(&i915->drm, sizeof(*gt->irq_lock), GFP_KERNEL); + if (!gt->irq_lock) + return -ENOMEM; - __intel_gt_init_early(gt); + intel_gt_common_init_early(gt); + + return 0; } static int intel_gt_probe_lmem(struct intel_gt *gt) @@ -781,26 +789,25 @@ static int intel_gt_tile_setup(struct intel_gt *gt, phys_addr_t phys_addr) int ret; if (!gt_is_root(gt)) { - struct intel_uncore_mmio_debug *mmio_debug; struct intel_uncore *uncore; + spinlock_t *irq_lock; - uncore = kzalloc(sizeof(*uncore), GFP_KERNEL); + uncore = drmm_kzalloc(>->i915->drm, sizeof(*uncore), GFP_KERNEL); if (!uncore) return -ENOMEM; - mmio_debug = kzalloc(sizeof(*mmio_debug), GFP_KERNEL); - if (!mmio_debug) { - kfree(uncore); + irq_lock = drmm_kzalloc(>->i915->drm, sizeof(*irq_lock), GFP_KERNEL); + if (!irq_lock) return -ENOMEM; - } gt->uncore = uncore; - gt->uncore->debug = mmio_debug; + gt->irq_lock = irq_lock; - __intel_gt_init_early(gt); + intel_gt_common_init_early(gt); } intel_uncore_init_early(gt->uncore, gt); + intel_wakeref_auto_init(>->userfault_wakeref, gt->uncore->rpm); ret = intel_uncore_setup_mmio(gt->uncore, phys_addr); if (ret) @@ -811,24 +818,14 @@ static int intel_gt_tile_setup(struct intel_gt *gt, phys_addr_t phys_addr) return 0; } -static void -intel_gt_tile_cleanup(struct intel_gt *gt) -{ - intel_uncore_cleanup_mmio(gt->uncore); - - if (!gt_is_root(gt)) { - kfree(gt->uncore->debug); - kfree(gt->uncore); - kfree(gt); - } -} - int intel_gt_probe_all(struct drm_i915_private *i915) { struct pci_dev *pdev = to_pci_dev(i915->drm.dev); struct intel_gt *gt = &i915->gt0; + const struct intel_gt_definition *gtdef; phys_addr_t phys_addr; unsigned int mmio_bar; + unsigned int i; int ret; mmio_bar = GRAPHICS_VER(i915) == 2 ? GEN2_GTTMMADR_BAR : GTTMMADR_BAR; @@ -839,14 +836,74 @@ int intel_gt_probe_all(struct drm_i915_private *i915) * and it has been already initialized early during probe * in i915_driver_probe() */ + gt->i915 = i915; + gt->name = "Primary GT"; + gt->info.engine_mask = RUNTIME_INFO(i915)->platform_engine_mask; + + drm_dbg(&i915->drm, "Setting up %s\n", gt->name); ret = intel_gt_tile_setup(gt, phys_addr); if (ret) return ret; i915->gt[0] = gt; - /* TODO: add more tiles */ + if (!HAS_EXTRA_GT_LIST(i915)) + return 0; + + for (i = 1, gtdef = &INTEL_INFO(i915)->extra_gt_list[i - 1]; + gtdef->name != NULL; + i++, gtdef = &INTEL_INFO(i915)->extra_gt_list[i - 1]) { + gt = drmm_kzalloc(&i915->drm, sizeof(*gt), GFP_KERNEL); + if (!gt) { + ret = -ENOMEM; + goto err; + } + + gt->i915 = i915; + gt->name = gtdef->name; + gt->type = gtdef->type; + gt->info.engine_mask = gtdef->engine_mask; + gt->info.id = i; + + drm_dbg(&i915->drm, "Setting up %s\n", gt->name); + if (GEM_WARN_ON(range_overflows_t(resource_size_t, + gtdef->mapping_base, + SZ_16M, + pci_resource_len(pdev, mmio_bar)))) { + ret = -ENODEV; + goto err; + } + + switch (gtdef->type) { + case GT_TILE: + ret = intel_gt_tile_setup(gt, phys_addr + gtdef->mapping_base); + break; + + case GT_MEDIA: + ret = intel_sa_mediagt_setup(gt, phys_addr + gtdef->mapping_base, + gtdef->gsi_offset); + break; + + case GT_PRIMARY: + /* Primary GT should not appear in extra GT list */ + default: + MISSING_CASE(gtdef->type); + ret = -ENODEV; + } + + if (ret) + goto err; + + i915->gt[i] = gt; + } + return 0; + +err: + i915_probe_error(i915, "Failed to initialize %s! (%d)\n", gtdef->name, ret); + intel_gt_release_all(i915); + + return ret; } int intel_gt_tiles_init(struct drm_i915_private *i915) @@ -869,10 +926,8 @@ void intel_gt_release_all(struct drm_i915_private *i915) struct intel_gt *gt; unsigned int id; - for_each_gt(gt, i915, id) { - intel_gt_tile_cleanup(gt); + for_each_gt(gt, i915, id) i915->gt[id] = NULL; - } } void intel_gt_info_print(const struct intel_gt_info *info, diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index 40b06adf509a..2ee582e287c8 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -44,7 +44,8 @@ static inline struct intel_gt *gsc_to_gt(struct intel_gsc *gsc) return container_of(gsc, struct intel_gt, gsc); } -void intel_root_gt_init_early(struct drm_i915_private *i915); +void intel_gt_common_init_early(struct intel_gt *gt); +int intel_root_gt_init_early(struct drm_i915_private *i915); int intel_gt_assign_ggtt(struct intel_gt *gt); int intel_gt_init_mmio(struct intel_gt *gt); int __must_check intel_gt_init_hw(struct intel_gt *gt); @@ -54,7 +55,6 @@ void intel_gt_driver_register(struct intel_gt *gt); void intel_gt_driver_unregister(struct intel_gt *gt); void intel_gt_driver_remove(struct intel_gt *gt); void intel_gt_driver_release(struct intel_gt *gt); - void intel_gt_driver_late_release_all(struct drm_i915_private *i915); int intel_gt_wait_for_idle(struct intel_gt *gt, long timeout); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c index d5d1b04dbcad..3f656d3dba9a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c @@ -26,26 +26,6 @@ static u32 read_reference_ts_freq(struct intel_uncore *uncore) return base_freq + frac_freq; } -static u32 gen9_get_crystal_clock_freq(struct intel_uncore *uncore, - u32 rpm_config_reg) -{ - u32 f19_2_mhz = 19200000; - u32 f24_mhz = 24000000; - u32 crystal_clock = - (rpm_config_reg & GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_MASK) >> - GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT; - - switch (crystal_clock) { - case GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_19_2_MHZ: - return f19_2_mhz; - case GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_24_MHZ: - return f24_mhz; - default: - MISSING_CASE(crystal_clock); - return 0; - } -} - static u32 gen11_get_crystal_clock_freq(struct intel_uncore *uncore, u32 rpm_config_reg) { @@ -72,98 +52,106 @@ static u32 gen11_get_crystal_clock_freq(struct intel_uncore *uncore, } } -static u32 read_clock_frequency(struct intel_uncore *uncore) +static u32 gen11_read_clock_frequency(struct intel_uncore *uncore) { - u32 f12_5_mhz = 12500000; - u32 f19_2_mhz = 19200000; - u32 f24_mhz = 24000000; + u32 ctc_reg = intel_uncore_read(uncore, CTC_MODE); + u32 freq = 0; + + /* + * Note that on gen11+, the clock frequency may be reconfigured. + * We do not, and we assume nobody else does. + * + * First figure out the reference frequency. There are 2 ways + * we can compute the frequency, either through the + * TIMESTAMP_OVERRIDE register or through RPM_CONFIG. CTC_MODE + * tells us which one we should use. + */ + if ((ctc_reg & CTC_SOURCE_PARAMETER_MASK) == CTC_SOURCE_DIVIDE_LOGIC) { + freq = read_reference_ts_freq(uncore); + } else { + u32 c0 = intel_uncore_read(uncore, RPM_CONFIG0); + + freq = gen11_get_crystal_clock_freq(uncore, c0); - if (GRAPHICS_VER(uncore->i915) <= 4) { - /* - * PRMs say: - * - * "The value in this register increments once every 16 - * hclks." (through the “Clocking Configuration” - * (“CLKCFG”) MCHBAR register) - */ - return RUNTIME_INFO(uncore->i915)->rawclk_freq * 1000 / 16; - } else if (GRAPHICS_VER(uncore->i915) <= 8) { /* - * PRMs say: - * - * "The PCU TSC counts 10ns increments; this timestamp - * reflects bits 38:3 of the TSC (i.e. 80ns granularity, - * rolling over every 1.5 hours). + * Now figure out how the command stream's timestamp + * register increments from this frequency (it might + * increment only every few clock cycle). */ - return f12_5_mhz; - } else if (GRAPHICS_VER(uncore->i915) <= 9) { - u32 ctc_reg = intel_uncore_read(uncore, CTC_MODE); - u32 freq = 0; - - if ((ctc_reg & CTC_SOURCE_PARAMETER_MASK) == CTC_SOURCE_DIVIDE_LOGIC) { - freq = read_reference_ts_freq(uncore); - } else { - freq = IS_GEN9_LP(uncore->i915) ? f19_2_mhz : f24_mhz; - - /* - * Now figure out how the command stream's timestamp - * register increments from this frequency (it might - * increment only every few clock cycle). - */ - freq >>= 3 - ((ctc_reg & CTC_SHIFT_PARAMETER_MASK) >> - CTC_SHIFT_PARAMETER_SHIFT); - } - - return freq; - } else if (GRAPHICS_VER(uncore->i915) <= 12) { - u32 ctc_reg = intel_uncore_read(uncore, CTC_MODE); - u32 freq = 0; + freq >>= 3 - ((c0 & GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_MASK) >> + GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT); + } + + return freq; +} + +static u32 gen9_read_clock_frequency(struct intel_uncore *uncore) +{ + u32 ctc_reg = intel_uncore_read(uncore, CTC_MODE); + u32 freq = 0; + + if ((ctc_reg & CTC_SOURCE_PARAMETER_MASK) == CTC_SOURCE_DIVIDE_LOGIC) { + freq = read_reference_ts_freq(uncore); + } else { + freq = IS_GEN9_LP(uncore->i915) ? 19200000 : 24000000; /* - * First figure out the reference frequency. There are 2 ways - * we can compute the frequency, either through the - * TIMESTAMP_OVERRIDE register or through RPM_CONFIG. CTC_MODE - * tells us which one we should use. + * Now figure out how the command stream's timestamp + * register increments from this frequency (it might + * increment only every few clock cycle). */ - if ((ctc_reg & CTC_SOURCE_PARAMETER_MASK) == CTC_SOURCE_DIVIDE_LOGIC) { - freq = read_reference_ts_freq(uncore); - } else { - u32 c0 = intel_uncore_read(uncore, RPM_CONFIG0); - - if (GRAPHICS_VER(uncore->i915) >= 11) - freq = gen11_get_crystal_clock_freq(uncore, c0); - else - freq = gen9_get_crystal_clock_freq(uncore, c0); - - /* - * Now figure out how the command stream's timestamp - * register increments from this frequency (it might - * increment only every few clock cycle). - */ - freq >>= 3 - ((c0 & GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_MASK) >> - GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT); - } - - return freq; + freq >>= 3 - ((ctc_reg & CTC_SHIFT_PARAMETER_MASK) >> + CTC_SHIFT_PARAMETER_SHIFT); } - MISSING_CASE("Unknown gen, unable to read command streamer timestamp frequency\n"); - return 0; + return freq; } -void intel_gt_init_clock_frequency(struct intel_gt *gt) +static u32 gen5_read_clock_frequency(struct intel_uncore *uncore) { /* - * Note that on gen11+, the clock frequency may be reconfigured. - * We do not, and we assume nobody else does. + * PRMs say: + * + * "The PCU TSC counts 10ns increments; this timestamp + * reflects bits 38:3 of the TSC (i.e. 80ns granularity, + * rolling over every 1.5 hours). + */ + return 12500000; +} + +static u32 gen2_read_clock_frequency(struct intel_uncore *uncore) +{ + /* + * PRMs say: + * + * "The value in this register increments once every 16 + * hclks." (through the “Clocking Configuration” + * (“CLKCFG”) MCHBAR register) */ + return RUNTIME_INFO(uncore->i915)->rawclk_freq * 1000 / 16; +} + +static u32 read_clock_frequency(struct intel_uncore *uncore) +{ + if (GRAPHICS_VER(uncore->i915) >= 11) + return gen11_read_clock_frequency(uncore); + else if (GRAPHICS_VER(uncore->i915) >= 9) + return gen9_read_clock_frequency(uncore); + else if (GRAPHICS_VER(uncore->i915) >= 5) + return gen5_read_clock_frequency(uncore); + else + return gen2_read_clock_frequency(uncore); +} + +void intel_gt_init_clock_frequency(struct intel_gt *gt) +{ gt->clock_frequency = read_clock_frequency(gt->uncore); - if (gt->clock_frequency) - gt->clock_period_ns = intel_gt_clock_interval_to_ns(gt, 1); /* Icelake appears to use another fixed frequency for CTX_TIMESTAMP */ if (GRAPHICS_VER(gt->i915) == 11) gt->clock_period_ns = NSEC_PER_SEC / 13750000; + else if (gt->clock_frequency) + gt->clock_period_ns = intel_gt_clock_interval_to_ns(gt, 1); GT_TRACE(gt, "Using clock frequency: %dkHz, period: %dns, wrap: %lldms\n", diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c index 3a72d4fd0214..f26882fdc24c 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c @@ -29,7 +29,7 @@ gen11_gt_engine_identity(struct intel_gt *gt, u32 timeout_ts; u32 ident; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); raw_reg_write(regs, GEN11_IIR_REG_SELECTOR(bank), BIT(bit)); @@ -59,11 +59,17 @@ static void gen11_other_irq_handler(struct intel_gt *gt, const u8 instance, const u16 iir) { + struct intel_gt *media_gt = gt->i915->media_gt; + if (instance == OTHER_GUC_INSTANCE) return guc_irq_handler(>->uc.guc, iir); + if (instance == OTHER_MEDIA_GUC_INSTANCE && media_gt) + return guc_irq_handler(&media_gt->uc.guc, iir); if (instance == OTHER_GTPM_INSTANCE) return gen11_rps_irq_handler(>->rps, iir); + if (instance == OTHER_MEDIA_GTPM_INSTANCE && media_gt) + return gen11_rps_irq_handler(&media_gt->rps, iir); if (instance == OTHER_KCR_INSTANCE) return intel_pxp_irq_handler(>->pxp, iir); @@ -81,6 +87,18 @@ gen11_engine_irq_handler(struct intel_gt *gt, const u8 class, { struct intel_engine_cs *engine; + /* + * Platforms with standalone media have their media engines in another + * GT. + */ + if (MEDIA_VER(gt->i915) >= 13 && + (class == VIDEO_DECODE_CLASS || class == VIDEO_ENHANCEMENT_CLASS)) { + if (!gt->i915->media_gt) + goto err; + + gt = gt->i915->media_gt; + } + if (instance <= MAX_ENGINE_INSTANCE) engine = gt->engine_class[class][instance]; else @@ -89,6 +107,7 @@ gen11_engine_irq_handler(struct intel_gt *gt, const u8 class, if (likely(engine)) return intel_engine_cs_irq(engine, iir); +err: WARN_ONCE(1, "unhandled engine interrupt class=0x%x, instance=0x%x\n", class, instance); } @@ -120,7 +139,7 @@ gen11_gt_bank_handler(struct intel_gt *gt, const unsigned int bank) unsigned long intr_dw; unsigned int bit; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); intr_dw = raw_reg_read(regs, GEN11_GT_INTR_DW(bank)); @@ -138,14 +157,14 @@ void gen11_gt_irq_handler(struct intel_gt *gt, const u32 master_ctl) { unsigned int bank; - spin_lock(>->irq_lock); + spin_lock(gt->irq_lock); for (bank = 0; bank < 2; bank++) { if (master_ctl & GEN11_GT_DW_IRQ(bank)) gen11_gt_bank_handler(gt, bank); } - spin_unlock(>->irq_lock); + spin_unlock(gt->irq_lock); } bool gen11_gt_reset_one_iir(struct intel_gt *gt, @@ -154,7 +173,7 @@ bool gen11_gt_reset_one_iir(struct intel_gt *gt, void __iomem * const regs = gt->uncore->regs; u32 dw; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); dw = raw_reg_read(regs, GEN11_GT_INTR_DW(bank)); if (dw & BIT(bit)) { @@ -310,9 +329,9 @@ static void gen7_parity_error_irq_handler(struct intel_gt *gt, u32 iir) if (!HAS_L3_DPF(gt->i915)) return; - spin_lock(>->irq_lock); + spin_lock(gt->irq_lock); gen5_gt_disable_irq(gt, GT_PARITY_ERROR(gt->i915)); - spin_unlock(>->irq_lock); + spin_unlock(gt->irq_lock); if (iir & GT_RENDER_L3_PARITY_ERROR_INTERRUPT_S1) gt->i915->l3_parity.which_slice |= 1 << 1; @@ -434,7 +453,7 @@ static void gen5_gt_update_irq(struct intel_gt *gt, u32 interrupt_mask, u32 enabled_irq_mask) { - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); GEM_BUG_ON(enabled_irq_mask & ~interrupt_mask); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c index 11060f5a4c89..52f2a28b2058 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c @@ -37,7 +37,7 @@ static void gen6_gt_pm_update_irq(struct intel_gt *gt, WARN_ON(enabled_irq_mask & ~interrupt_mask); - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); new_val = gt->pm_imr; new_val &= ~interrupt_mask; @@ -64,7 +64,7 @@ void gen6_gt_pm_reset_iir(struct intel_gt *gt, u32 reset_mask) struct intel_uncore *uncore = gt->uncore; i915_reg_t reg = GRAPHICS_VER(gt->i915) >= 8 ? GEN8_GT_IIR(2) : GEN6_PMIIR; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); intel_uncore_write(uncore, reg, reset_mask); intel_uncore_write(uncore, reg, reset_mask); @@ -92,7 +92,7 @@ static void write_pm_ier(struct intel_gt *gt) void gen6_gt_pm_enable_irq(struct intel_gt *gt, u32 enable_mask) { - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); gt->pm_ier |= enable_mask; write_pm_ier(gt); @@ -101,7 +101,7 @@ void gen6_gt_pm_enable_irq(struct intel_gt *gt, u32 enable_mask) void gen6_gt_pm_disable_irq(struct intel_gt *gt, u32 disable_mask) { - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); gt->pm_ier &= ~disable_mask; gen6_gt_pm_mask_irq(gt, disable_mask); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h index 94f9ddcfb3a5..2275ee47da95 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -1110,6 +1110,8 @@ #define GEN12_DISABLE_TDL_PUSH REG_BIT(9) #define GEN11_DIS_PICK_2ND_EU REG_BIT(7) #define GEN12_DISABLE_HDR_PAST_PAYLOAD_HOLD_FIX REG_BIT(4) +#define THREAD_EX_ARB_MODE REG_GENMASK(3, 2) +#define THREAD_EX_ARB_MODE_RR_AFTER_DEP REG_FIELD_PREP(THREAD_EX_ARB_MODE, 0x2) #define HSW_ROW_CHICKEN3 _MMIO(0xe49c) #define HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE (1 << 6) @@ -1552,6 +1554,8 @@ #define OTHER_GTPM_INSTANCE 1 #define OTHER_KCR_INSTANCE 4 #define OTHER_GSC_INSTANCE 6 +#define OTHER_MEDIA_GUC_INSTANCE 16 +#define OTHER_MEDIA_GTPM_INSTANCE 17 #define GEN11_IIR_REG_SELECTOR(x) _MMIO(0x190070 + ((x) * 4)) @@ -1576,4 +1580,12 @@ #define GEN12_SFC_DONE(n) _MMIO(0x1cc000 + (n) * 0x1000) +/* + * Standalone Media's non-engine GT registers are located at their regular GT + * offsets plus 0x380000. This extra offset is stored inside the intel_uncore + * structure so that the existing code can be used for both GTs without + * modification. + */ +#define MTL_MEDIA_GSI_BASE 0x380000 + #endif /* __INTEL_GT_REGS__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gt_types.h b/drivers/gpu/drm/i915/gt/intel_gt_types.h index 4d56f7d5a3be..f19c2de77ff6 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_types.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_types.h @@ -81,8 +81,17 @@ struct gt_defaults { u32 max_freq; }; +enum intel_gt_type { + GT_PRIMARY, + GT_TILE, + GT_MEDIA, +}; + struct intel_gt { struct drm_i915_private *i915; + const char *name; + enum intel_gt_type type; + struct intel_uncore *uncore; struct i915_ggtt *ggtt; @@ -132,6 +141,20 @@ struct intel_gt { struct intel_wakeref wakeref; atomic_t user_wakeref; + /** + * Protects access to lmem usefault list. + * It is required, if we are outside of the runtime suspend path, + * access to @lmem_userfault_list requires always first grabbing the + * runtime pm, to ensure we can't race against runtime suspend. + * Once we have that we also need to grab @lmem_userfault_lock, + * at which point we have exclusive access. + * The runtime suspend path is special since it doesn't really hold any locks, + * but instead has exclusive access by virtue of all other accesses requiring + * holding the runtime pm wakeref. + */ + struct mutex lmem_userfault_lock; + struct list_head lmem_userfault_list; + struct list_head closed_vma; spinlock_t closed_lock; /* guards the list of closed_vma */ @@ -147,6 +170,9 @@ struct intel_gt { */ intel_wakeref_t awake; + /* Manual runtime pm autosuspend delay for user GGTT/lmem mmaps */ + struct intel_wakeref_auto userfault_wakeref; + u32 clock_frequency; u32 clock_period_ns; @@ -154,7 +180,7 @@ struct intel_gt { struct intel_rc6 rc6; struct intel_rps rps; - spinlock_t irq_lock; + spinlock_t *irq_lock; u32 gt_imr; u32 pm_ier; u32 pm_imr; @@ -262,6 +288,14 @@ struct intel_gt { struct kobject *sysfs_defaults; }; +struct intel_gt_definition { + enum intel_gt_type type; + char *name; + u32 mapping_base; + u32 gsi_offset; + intel_engine_mask_t engine_mask; +}; + enum intel_gt_scratch_field { /* 8 bytes */ INTEL_GT_SCRATCH_FIELD_DEFAULT = 0, diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h index e639434e97fd..c0ca53cba9f0 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.h +++ b/drivers/gpu/drm/i915/gt/intel_gtt.h @@ -386,9 +386,6 @@ struct i915_ggtt { */ struct list_head userfault_list; - /* Manual runtime pm autosuspend delay for user GGTT mmaps */ - struct intel_wakeref_auto userfault_wakeref; - struct mutex error_mutex; struct drm_mm_node error_capture; struct drm_mm_node uc_fw; diff --git a/drivers/gpu/drm/i915/gt/intel_llc.c b/drivers/gpu/drm/i915/gt/intel_llc.c index 14fe65812e42..1d19c073ba2e 100644 --- a/drivers/gpu/drm/i915/gt/intel_llc.c +++ b/drivers/gpu/drm/i915/gt/intel_llc.c @@ -12,6 +12,7 @@ #include "intel_llc.h" #include "intel_mchbar_regs.h" #include "intel_pcode.h" +#include "intel_rps.h" struct ia_constants { unsigned int min_gpu_freq; @@ -55,9 +56,6 @@ static bool get_ia_constants(struct intel_llc *llc, if (!HAS_LLC(i915) || IS_DGFX(i915)) return false; - if (rps->max_freq <= rps->min_freq) - return false; - consts->max_ia_freq = cpu_max_MHz(); consts->min_ring_freq = @@ -65,13 +63,8 @@ static bool get_ia_constants(struct intel_llc *llc, /* convert DDR frequency from units of 266.6MHz to bandwidth */ consts->min_ring_freq = mult_frac(consts->min_ring_freq, 8, 3); - consts->min_gpu_freq = rps->min_freq; - consts->max_gpu_freq = rps->max_freq; - if (GRAPHICS_VER(i915) >= 9) { - /* Convert GT frequency to 50 HZ units */ - consts->min_gpu_freq /= GEN9_FREQ_SCALER; - consts->max_gpu_freq /= GEN9_FREQ_SCALER; - } + consts->min_gpu_freq = intel_rps_get_min_raw_freq(rps); + consts->max_gpu_freq = intel_rps_get_max_raw_freq(rps); return true; } @@ -131,6 +124,12 @@ static void gen6_update_ring_freq(struct intel_llc *llc) return; /* + * Although this is unlikely on any platform during initialization, + * let's ensure we don't get accidentally into infinite loop + */ + if (consts.max_gpu_freq <= consts.min_gpu_freq) + return; + /* * For each potential GPU frequency, load a ring frequency we'd like * to use for memory access. We do this by specifying the IA frequency * the PCU should use as a reference to determine the ring frequency. diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index eec73c66406c..3955292483a6 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -662,6 +662,21 @@ static int lrc_ring_mi_mode(const struct intel_engine_cs *engine) return -1; } +static int lrc_ring_bb_offset(const struct intel_engine_cs *engine) +{ + if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 50)) + return 0x80; + else if (GRAPHICS_VER(engine->i915) >= 12) + return 0x70; + else if (GRAPHICS_VER(engine->i915) >= 9) + return 0x64; + else if (GRAPHICS_VER(engine->i915) >= 8 && + engine->class == RENDER_CLASS) + return 0xc4; + else + return -1; +} + static int lrc_ring_gpr0(const struct intel_engine_cs *engine) { if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 50)) @@ -768,6 +783,7 @@ static void init_common_regs(u32 * const regs, bool inhibit) { u32 ctl; + int loc; ctl = _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH); ctl |= _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT); @@ -779,6 +795,10 @@ static void init_common_regs(u32 * const regs, regs[CTX_CONTEXT_CONTROL] = ctl; regs[CTX_TIMESTAMP] = ce->stats.runtime.last; + + loc = lrc_ring_bb_offset(engine); + if (loc != -1) + regs[loc + 1] = 0; } static void init_wa_bb_regs(u32 * const regs, @@ -1242,6 +1262,23 @@ dg2_emit_rcs_hang_wabb(const struct intel_context *ce, u32 *cs) return cs; } +/* + * The bspec's tuning guide asks us to program a vertical watermark value of + * 0x3FF. However this register is not saved/restored properly by the + * hardware, so we're required to apply the desired value via INDIRECT_CTX + * batch buffer to ensure the value takes effect properly. All other bits + * in this register should remain at 0 (the hardware default). + */ +static u32 * +dg2_emit_draw_watermark_setting(u32 *cs) +{ + *cs++ = MI_LOAD_REGISTER_IMM(1); + *cs++ = i915_mmio_reg_offset(DRAW_WATERMARK); + *cs++ = REG_FIELD_PREP(VERT_WM_VAL, 0x3FF); + + return cs; +} + static u32 * gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs) { @@ -1261,7 +1298,12 @@ gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs) /* hsdes: 1809175790 */ if (!HAS_FLAT_CCS(ce->engine->i915)) - cs = gen12_emit_aux_table_inv(cs, GEN12_GFX_CCS_AUX_NV); + cs = gen12_emit_aux_table_inv(ce->engine->gt, + cs, GEN12_GFX_CCS_AUX_NV); + + /* Wa_16014892111 */ + if (IS_DG2(ce->engine->i915)) + cs = dg2_emit_draw_watermark_setting(cs); return cs; } @@ -1283,9 +1325,11 @@ gen12_emit_indirect_ctx_xcs(const struct intel_context *ce, u32 *cs) /* hsdes: 1809175790 */ if (!HAS_FLAT_CCS(ce->engine->i915)) { if (ce->engine->class == VIDEO_DECODE_CLASS) - cs = gen12_emit_aux_table_inv(cs, GEN12_VD0_AUX_NV); + cs = gen12_emit_aux_table_inv(ce->engine->gt, + cs, GEN12_VD0_AUX_NV); else if (ce->engine->class == VIDEO_ENHANCEMENT_CLASS) - cs = gen12_emit_aux_table_inv(cs, GEN12_VE0_AUX_NV); + cs = gen12_emit_aux_table_inv(ce->engine->gt, + cs, GEN12_VE0_AUX_NV); } return cs; diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index 1211774e1d91..b36674356986 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -776,7 +776,7 @@ static void revoke_mmaps(struct intel_gt *gt) continue; node = &vma->mmo->vma_node; - vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT; + vma_offset = vma->gtt_view.partial.offset << PAGE_SHIFT; unmap_mapping_range(gt->i915->drm.anon_inode->i_mapping, drm_vma_node_offset_addr(node) + vma_offset, diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index c7d381ad90cf..6b86250c31ab 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -194,9 +194,9 @@ static void rps_enable_interrupts(struct intel_rps *rps) rps_reset_ei(rps); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen6_gt_pm_enable_irq(gt, rps->pm_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); intel_uncore_write(gt->uncore, GEN6_PMINTRMSK, rps_pm_mask(rps, rps->last_freq)); @@ -217,14 +217,14 @@ static void rps_reset_interrupts(struct intel_rps *rps) { struct intel_gt *gt = rps_to_gt(rps); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); if (GRAPHICS_VER(gt->i915) >= 11) gen11_rps_reset_interrupts(rps); else gen6_rps_reset_interrupts(rps); rps->pm_iir = 0; - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } static void rps_disable_interrupts(struct intel_rps *rps) @@ -234,9 +234,9 @@ static void rps_disable_interrupts(struct intel_rps *rps) intel_uncore_write(gt->uncore, GEN6_PMINTRMSK, rps_pm_sanitize_mask(rps, ~0u)); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen6_gt_pm_disable_irq(gt, GEN6_PM_RPS_EVENTS); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); intel_synchronize_irq(gt->i915); @@ -1107,7 +1107,12 @@ void gen6_rps_get_freq_caps(struct intel_rps *rps, struct intel_rps_freq_caps *c caps->min_freq = (rp_state_cap >> 0) & 0xff; } else { caps->rp0_freq = (rp_state_cap >> 0) & 0xff; - caps->rp1_freq = (rp_state_cap >> 8) & 0xff; + if (GRAPHICS_VER(i915) >= 10) + caps->rp1_freq = REG_FIELD_GET(RPE_MASK, + intel_uncore_read(to_gt(i915)->uncore, + GEN10_FREQ_INFO_REC)); + else + caps->rp1_freq = (rp_state_cap >> 8) & 0xff; caps->min_freq = (rp_state_cap >> 16) & 0xff; } @@ -1546,6 +1551,9 @@ void intel_rps_disable(struct intel_rps *rps) { struct drm_i915_private *i915 = rps_to_i915(rps); + if (!intel_rps_is_enabled(rps)) + return; + intel_rps_clear_enabled(rps); intel_rps_clear_interrupts(rps); intel_rps_clear_timer(rps); @@ -1789,10 +1797,10 @@ static void rps_work(struct work_struct *work) int new_freq, adj, min, max; u32 pm_iir = 0; - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); pm_iir = fetch_and_zero(&rps->pm_iir) & rps->pm_events; client_boost = atomic_read(&rps->num_waiters); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); /* Make sure we didn't queue anything we're not going to process. */ if (!pm_iir && !client_boost) @@ -1865,9 +1873,9 @@ static void rps_work(struct work_struct *work) mutex_unlock(&rps->lock); out: - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen6_gt_pm_unmask_irq(gt, rps->pm_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } void gen11_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) @@ -1875,7 +1883,7 @@ void gen11_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) struct intel_gt *gt = rps_to_gt(rps); const u32 events = rps->pm_events & pm_iir; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); if (unlikely(!events)) return; @@ -1895,7 +1903,7 @@ void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) events = pm_iir & rps->pm_events; if (events) { - spin_lock(>->irq_lock); + spin_lock(gt->irq_lock); GT_TRACE(gt, "irq events:%x\n", events); @@ -1903,7 +1911,7 @@ void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir) rps->pm_iir |= events; schedule_work(&rps->work); - spin_unlock(>->irq_lock); + spin_unlock(gt->irq_lock); } if (GRAPHICS_VER(gt->i915) >= 8) @@ -2128,6 +2136,31 @@ u32 intel_rps_get_max_frequency(struct intel_rps *rps) return intel_gpu_freq(rps, rps->max_freq_softlimit); } +/** + * intel_rps_get_max_raw_freq - returns the max frequency in some raw format. + * @rps: the intel_rps structure + * + * Returns the max frequency in a raw format. In newer platforms raw is in + * units of 50 MHz. + */ +u32 intel_rps_get_max_raw_freq(struct intel_rps *rps) +{ + struct intel_guc_slpc *slpc = rps_to_slpc(rps); + u32 freq; + + if (rps_uses_slpc(rps)) { + return DIV_ROUND_CLOSEST(slpc->rp0_freq, + GT_FREQUENCY_MULTIPLIER); + } else { + freq = rps->max_freq; + if (GRAPHICS_VER(rps_to_i915(rps)) >= 9) { + /* Convert GT frequency to 50 MHz units */ + freq /= GEN9_FREQ_SCALER; + } + return freq; + } +} + u32 intel_rps_get_rp0_frequency(struct intel_rps *rps) { struct intel_guc_slpc *slpc = rps_to_slpc(rps); @@ -2216,6 +2249,31 @@ u32 intel_rps_get_min_frequency(struct intel_rps *rps) return intel_gpu_freq(rps, rps->min_freq_softlimit); } +/** + * intel_rps_get_min_raw_freq - returns the min frequency in some raw format. + * @rps: the intel_rps structure + * + * Returns the min frequency in a raw format. In newer platforms raw is in + * units of 50 MHz. + */ +u32 intel_rps_get_min_raw_freq(struct intel_rps *rps) +{ + struct intel_guc_slpc *slpc = rps_to_slpc(rps); + u32 freq; + + if (rps_uses_slpc(rps)) { + return DIV_ROUND_CLOSEST(slpc->min_freq, + GT_FREQUENCY_MULTIPLIER); + } else { + freq = rps->min_freq; + if (GRAPHICS_VER(rps_to_i915(rps)) >= 9) { + /* Convert GT frequency to 50 MHz units */ + freq /= GEN9_FREQ_SCALER; + } + return freq; + } +} + static int set_min_freq(struct intel_rps *rps, u32 val) { int ret = 0; diff --git a/drivers/gpu/drm/i915/gt/intel_rps.h b/drivers/gpu/drm/i915/gt/intel_rps.h index 1e8d56491308..4509dfdc52e0 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.h +++ b/drivers/gpu/drm/i915/gt/intel_rps.h @@ -37,8 +37,10 @@ u32 intel_rps_get_cagf(struct intel_rps *rps, u32 rpstat1); u32 intel_rps_read_actual_frequency(struct intel_rps *rps); u32 intel_rps_get_requested_frequency(struct intel_rps *rps); u32 intel_rps_get_min_frequency(struct intel_rps *rps); +u32 intel_rps_get_min_raw_freq(struct intel_rps *rps); int intel_rps_set_min_frequency(struct intel_rps *rps, u32 val); u32 intel_rps_get_max_frequency(struct intel_rps *rps); +u32 intel_rps_get_max_raw_freq(struct intel_rps *rps); int intel_rps_set_max_frequency(struct intel_rps *rps, u32 val); u32 intel_rps_get_rp0_frequency(struct intel_rps *rps); u32 intel_rps_get_rp1_frequency(struct intel_rps *rps); diff --git a/drivers/gpu/drm/i915/gt/intel_sa_media.c b/drivers/gpu/drm/i915/gt/intel_sa_media.c new file mode 100644 index 000000000000..e8f3d18c12b8 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_sa_media.c @@ -0,0 +1,47 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2021 Intel Corporation + */ + +#include <drm/drm_managed.h> + +#include "i915_drv.h" +#include "gt/intel_gt.h" +#include "gt/intel_sa_media.h" + +int intel_sa_mediagt_setup(struct intel_gt *gt, phys_addr_t phys_addr, + u32 gsi_offset) +{ + struct drm_i915_private *i915 = gt->i915; + struct intel_uncore *uncore; + + uncore = drmm_kzalloc(&i915->drm, sizeof(*uncore), GFP_KERNEL); + if (!uncore) + return -ENOMEM; + + uncore->gsi_offset = gsi_offset; + + gt->irq_lock = to_gt(i915)->irq_lock; + intel_gt_common_init_early(gt); + intel_uncore_init_early(uncore, gt); + + /* + * Standalone media shares the general MMIO space with the primary + * GT. We'll re-use the primary GT's mapping. + */ + uncore->regs = i915->uncore.regs; + if (drm_WARN_ON(&i915->drm, uncore->regs == NULL)) + return -EIO; + + gt->uncore = uncore; + gt->phys_addr = phys_addr; + + /* + * For current platforms we can assume there's only a single + * media GT and cache it for quick lookup. + */ + drm_WARN_ON(&i915->drm, i915->media_gt); + i915->media_gt = gt; + + return 0; +} diff --git a/drivers/gpu/drm/i915/gt/intel_sa_media.h b/drivers/gpu/drm/i915/gt/intel_sa_media.h new file mode 100644 index 000000000000..3afb310de932 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_sa_media.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ +#ifndef __INTEL_SA_MEDIA__ +#define __INTEL_SA_MEDIA__ + +#include <linux/types.h> + +struct intel_gt; + +int intel_sa_mediagt_setup(struct intel_gt *gt, phys_addr_t phys_addr, + u32 gsi_offset); + +#endif /* __INTEL_SA_MEDIA_H__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 31e129329fb0..6d2003d598e6 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -2389,7 +2389,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) FF_DOP_CLOCK_GATE_DISABLE); } - if (HAS_PERCTX_PREEMPT_CTRL(i915)) { + if (IS_GRAPHICS_VER(i915, 9, 12)) { /* FtrPerCtxtPreemptionGranularityControl:skl,bxt,kbl,cfl,cnl,icl,tgl */ wa_masked_en(wal, GEN7_FF_SLICE_CS_CHICKEN1, @@ -2685,8 +2685,6 @@ add_render_compute_tuning_settings(struct drm_i915_private *i915, if (IS_DG2(i915)) { wa_write_or(wal, XEHP_L3SCQREG7, BLEND_FILL_CACHING_OPT_DIS); wa_write_clr_set(wal, RT_CTRL, STACKID_CTRL, STACKID_CTRL_512); - wa_write_clr_set(wal, DRAW_WATERMARK, VERT_WM_VAL, - REG_FIELD_PREP(VERT_WM_VAL, 0x3FF)); /* * This is also listed as Wa_22012654132 for certain DG2 @@ -2702,6 +2700,15 @@ add_render_compute_tuning_settings(struct drm_i915_private *i915, 0 /* write-only, so skip validation */, true); } + + /* + * This tuning setting proves beneficial only on ATS-M designs; the + * default "age based" setting is optimal on regular DG2 and other + * platforms. + */ + if (INTEL_INFO(i915)->tuning_thread_rr_after_dep) + wa_masked_field_set(wal, GEN9_ROW_CHICKEN4, THREAD_EX_ARB_MODE, + THREAD_EX_ARB_MODE_RR_AFTER_DEP); } /* diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c index 1109088fe8f6..82d3f8058995 100644 --- a/drivers/gpu/drm/i915/gt/selftest_lrc.c +++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c @@ -27,6 +27,9 @@ #define NUM_GPR 16 #define NUM_GPR_DW (NUM_GPR * 2) /* each GPR is 2 dwords */ +#define LRI_HEADER MI_INSTR(0x22, 0) +#define LRI_LENGTH_MASK GENMASK(7, 0) + static struct i915_vma *create_scratch(struct intel_gt *gt) { return __vm_create_scratch_for_read_pinned(>->ggtt->vm, PAGE_SIZE); @@ -202,7 +205,7 @@ static int live_lrc_layout(void *arg) continue; } - if ((lri & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + if ((lri & GENMASK(31, 23)) != LRI_HEADER) { pr_err("%s: Expected LRI command at dword %d, found %08x\n", engine->name, dw, lri); err = -EINVAL; @@ -357,6 +360,11 @@ static int live_lrc_fixed(void *arg) lrc_ring_cmd_buf_cctl(engine), "RING_CMD_BUF_CCTL" }, + { + i915_mmio_reg_offset(RING_BB_OFFSET(engine->mmio_base)), + lrc_ring_bb_offset(engine), + "RING_BB_OFFSET" + }, { }, }, *t; u32 *hw; @@ -987,18 +995,40 @@ store_context(struct intel_context *ce, struct i915_vma *scratch) hw = defaults; hw += LRC_STATE_OFFSET / sizeof(*hw); do { - u32 len = hw[dw] & 0x7f; + u32 len = hw[dw] & LRI_LENGTH_MASK; + + /* + * Keep it simple, skip parsing complex commands + * + * At present, there are no more MI_LOAD_REGISTER_IMM + * commands after the first 3D state command. Rather + * than include a table (see i915_cmd_parser.c) of all + * the possible commands and their instruction lengths + * (or mask for variable length instructions), assume + * we have gathered the complete list of registers and + * bail out. + */ + if ((hw[dw] >> INSTR_CLIENT_SHIFT) != INSTR_MI_CLIENT) + break; if (hw[dw] == 0) { dw++; continue; } - if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + if ((hw[dw] & GENMASK(31, 23)) != LRI_HEADER) { + /* Assume all other MI commands match LRI length mask */ dw += len + 2; continue; } + if (!len) { + pr_err("%s: invalid LRI found in context image\n", + ce->engine->name); + igt_hexdump(defaults, PAGE_SIZE); + break; + } + dw++; len = (len + 1) / 2; while (len--) { @@ -1150,18 +1180,29 @@ static struct i915_vma *load_context(struct intel_context *ce, u32 poison) hw = defaults; hw += LRC_STATE_OFFSET / sizeof(*hw); do { - u32 len = hw[dw] & 0x7f; + u32 len = hw[dw] & LRI_LENGTH_MASK; + + /* For simplicity, break parsing at the first complex command */ + if ((hw[dw] >> INSTR_CLIENT_SHIFT) != INSTR_MI_CLIENT) + break; if (hw[dw] == 0) { dw++; continue; } - if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + if ((hw[dw] & GENMASK(31, 23)) != LRI_HEADER) { dw += len + 2; continue; } + if (!len) { + pr_err("%s: invalid LRI found in context image\n", + ce->engine->name); + igt_hexdump(defaults, PAGE_SIZE); + break; + } + dw++; len = (len + 1) / 2; *cs++ = MI_LOAD_REGISTER_IMM(len); @@ -1292,18 +1333,29 @@ static int compare_isolation(struct intel_engine_cs *engine, hw = defaults; hw += LRC_STATE_OFFSET / sizeof(*hw); do { - u32 len = hw[dw] & 0x7f; + u32 len = hw[dw] & LRI_LENGTH_MASK; + + /* For simplicity, break parsing at the first complex command */ + if ((hw[dw] >> INSTR_CLIENT_SHIFT) != INSTR_MI_CLIENT) + break; if (hw[dw] == 0) { dw++; continue; } - if ((hw[dw] & GENMASK(31, 23)) != MI_INSTR(0x22, 0)) { + if ((hw[dw] & GENMASK(31, 23)) != LRI_HEADER) { dw += len + 2; continue; } + if (!len) { + pr_err("%s: invalid LRI found in context image\n", + engine->name); + igt_hexdump(defaults, PAGE_SIZE); + break; + } + dw++; len = (len + 1) / 2; while (len--) { @@ -1343,6 +1395,30 @@ err_A0: return err; } +static struct i915_vma * +create_result_vma(struct i915_address_space *vm, unsigned long sz) +{ + struct i915_vma *vma; + void *ptr; + + vma = create_user_vma(vm, sz); + if (IS_ERR(vma)) + return vma; + + /* Set the results to a known value distinct from the poison */ + ptr = i915_gem_object_pin_map_unlocked(vma->obj, I915_MAP_WC); + if (IS_ERR(ptr)) { + i915_vma_put(vma); + return ERR_CAST(ptr); + } + + memset(ptr, POISON_INUSE, vma->size); + i915_gem_object_flush_map(vma->obj); + i915_gem_object_unpin_map(vma->obj); + + return vma; +} + static int __lrc_isolation(struct intel_engine_cs *engine, u32 poison) { u32 *sema = memset32(engine->status_page.addr + 1000, 0, 1); @@ -1361,13 +1437,13 @@ static int __lrc_isolation(struct intel_engine_cs *engine, u32 poison) goto err_A; } - ref[0] = create_user_vma(A->vm, SZ_64K); + ref[0] = create_result_vma(A->vm, SZ_64K); if (IS_ERR(ref[0])) { err = PTR_ERR(ref[0]); goto err_B; } - ref[1] = create_user_vma(A->vm, SZ_64K); + ref[1] = create_result_vma(A->vm, SZ_64K); if (IS_ERR(ref[1])) { err = PTR_ERR(ref[1]); goto err_ref0; @@ -1389,13 +1465,13 @@ static int __lrc_isolation(struct intel_engine_cs *engine, u32 poison) } i915_request_put(rq); - result[0] = create_user_vma(A->vm, SZ_64K); + result[0] = create_result_vma(A->vm, SZ_64K); if (IS_ERR(result[0])) { err = PTR_ERR(result[0]); goto err_ref1; } - result[1] = create_user_vma(A->vm, SZ_64K); + result[1] = create_result_vma(A->vm, SZ_64K); if (IS_ERR(result[1])) { err = PTR_ERR(result[1]); goto err_result0; @@ -1408,18 +1484,17 @@ static int __lrc_isolation(struct intel_engine_cs *engine, u32 poison) } err = poison_registers(B, poison, sema); - if (err) { - WRITE_ONCE(*sema, -1); - i915_request_put(rq); - goto err_result1; - } - - if (i915_request_wait(rq, 0, HZ / 2) < 0) { - i915_request_put(rq); + if (err == 0 && i915_request_wait(rq, 0, HZ / 2) < 0) { + pr_err("%s(%s): wait for results timed out\n", + __func__, engine->name); err = -ETIME; - goto err_result1; } + + /* Always cancel the semaphore wait, just in case the GPU gets stuck */ + WRITE_ONCE(*sema, -1); i915_request_put(rq); + if (err) + goto err_result1; err = compare_isolation(engine, ref, result, A, poison); diff --git a/drivers/gpu/drm/i915/gt/selftest_slpc.c b/drivers/gpu/drm/i915/gt/selftest_slpc.c index ac29691e0b1a..f8a1d27df272 100644 --- a/drivers/gpu/drm/i915/gt/selftest_slpc.c +++ b/drivers/gpu/drm/i915/gt/selftest_slpc.c @@ -166,6 +166,15 @@ static int run_test(struct intel_gt *gt, int test_type) return -EIO; } + /* + * FIXME: With efficient frequency enabled, GuC can request + * frequencies higher than the SLPC max. While this is fixed + * in GuC, we level set these tests with RPn as min. + */ + err = slpc_set_min_freq(slpc, slpc->min_freq); + if (err) + return err; + if (slpc->min_freq == slpc->rp0_freq) { pr_err("Min/Max are fused to the same value\n"); return -EINVAL; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index 50af247e84c4..bac06e3d6f2c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -82,9 +82,9 @@ static void gen9_reset_guc_interrupts(struct intel_guc *guc) assert_rpm_wakelock_held(>->i915->runtime_pm); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen6_gt_pm_reset_iir(gt, gt->pm_guc_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } static void gen9_enable_guc_interrupts(struct intel_guc *guc) @@ -93,11 +93,11 @@ static void gen9_enable_guc_interrupts(struct intel_guc *guc) assert_rpm_wakelock_held(>->i915->runtime_pm); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); WARN_ON_ONCE(intel_uncore_read(gt->uncore, GEN8_GT_IIR(2)) & gt->pm_guc_events); gen6_gt_pm_enable_irq(gt, gt->pm_guc_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } static void gen9_disable_guc_interrupts(struct intel_guc *guc) @@ -106,11 +106,11 @@ static void gen9_disable_guc_interrupts(struct intel_guc *guc) assert_rpm_wakelock_held(>->i915->runtime_pm); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen6_gt_pm_disable_irq(gt, gt->pm_guc_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); intel_synchronize_irq(gt->i915); gen9_reset_guc_interrupts(guc); @@ -120,9 +120,9 @@ static void gen11_reset_guc_interrupts(struct intel_guc *guc) { struct intel_gt *gt = guc_to_gt(guc); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen11_gt_reset_one_iir(gt, 0, GEN11_GUC); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } static void gen11_enable_guc_interrupts(struct intel_guc *guc) @@ -130,25 +130,25 @@ static void gen11_enable_guc_interrupts(struct intel_guc *guc) struct intel_gt *gt = guc_to_gt(guc); u32 events = REG_FIELD_PREP(ENGINE1_MASK, GUC_INTR_GUC2HOST); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); WARN_ON_ONCE(gen11_gt_reset_one_iir(gt, 0, GEN11_GUC)); intel_uncore_write(gt->uncore, GEN11_GUC_SG_INTR_ENABLE, events); intel_uncore_write(gt->uncore, GEN11_GUC_SG_INTR_MASK, ~events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } static void gen11_disable_guc_interrupts(struct intel_guc *guc) { struct intel_gt *gt = guc_to_gt(guc); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); intel_uncore_write(gt->uncore, GEN11_GUC_SG_INTR_MASK, ~0); intel_uncore_write(gt->uncore, GEN11_GUC_SG_INTR_ENABLE, 0); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); intel_synchronize_irq(gt->i915); gen11_reset_guc_interrupts(guc); @@ -224,53 +224,22 @@ static u32 guc_ctl_feature_flags(struct intel_guc *guc) static u32 guc_ctl_log_params_flags(struct intel_guc *guc) { - u32 offset = intel_guc_ggtt_offset(guc, guc->log.vma) >> PAGE_SHIFT; - u32 flags; - - #if (((CRASH_BUFFER_SIZE) % SZ_1M) == 0) - #define LOG_UNIT SZ_1M - #define LOG_FLAG GUC_LOG_LOG_ALLOC_UNITS - #else - #define LOG_UNIT SZ_4K - #define LOG_FLAG 0 - #endif - - #if (((CAPTURE_BUFFER_SIZE) % SZ_1M) == 0) - #define CAPTURE_UNIT SZ_1M - #define CAPTURE_FLAG GUC_LOG_CAPTURE_ALLOC_UNITS - #else - #define CAPTURE_UNIT SZ_4K - #define CAPTURE_FLAG 0 - #endif - - BUILD_BUG_ON(!CRASH_BUFFER_SIZE); - BUILD_BUG_ON(!IS_ALIGNED(CRASH_BUFFER_SIZE, LOG_UNIT)); - BUILD_BUG_ON(!DEBUG_BUFFER_SIZE); - BUILD_BUG_ON(!IS_ALIGNED(DEBUG_BUFFER_SIZE, LOG_UNIT)); - BUILD_BUG_ON(!CAPTURE_BUFFER_SIZE); - BUILD_BUG_ON(!IS_ALIGNED(CAPTURE_BUFFER_SIZE, CAPTURE_UNIT)); - - BUILD_BUG_ON((CRASH_BUFFER_SIZE / LOG_UNIT - 1) > - (GUC_LOG_CRASH_MASK >> GUC_LOG_CRASH_SHIFT)); - BUILD_BUG_ON((DEBUG_BUFFER_SIZE / LOG_UNIT - 1) > - (GUC_LOG_DEBUG_MASK >> GUC_LOG_DEBUG_SHIFT)); - BUILD_BUG_ON((CAPTURE_BUFFER_SIZE / CAPTURE_UNIT - 1) > - (GUC_LOG_CAPTURE_MASK >> GUC_LOG_CAPTURE_SHIFT)); + struct intel_guc_log *log = &guc->log; + u32 offset, flags; + + GEM_BUG_ON(!log->sizes_initialised); + + offset = intel_guc_ggtt_offset(guc, log->vma) >> PAGE_SHIFT; flags = GUC_LOG_VALID | GUC_LOG_NOTIFY_ON_HALF_FULL | - CAPTURE_FLAG | - LOG_FLAG | - ((CRASH_BUFFER_SIZE / LOG_UNIT - 1) << GUC_LOG_CRASH_SHIFT) | - ((DEBUG_BUFFER_SIZE / LOG_UNIT - 1) << GUC_LOG_DEBUG_SHIFT) | - ((CAPTURE_BUFFER_SIZE / CAPTURE_UNIT - 1) << GUC_LOG_CAPTURE_SHIFT) | + log->sizes[GUC_LOG_SECTIONS_DEBUG].flag | + log->sizes[GUC_LOG_SECTIONS_CAPTURE].flag | + (log->sizes[GUC_LOG_SECTIONS_CRASH].count << GUC_LOG_CRASH_SHIFT) | + (log->sizes[GUC_LOG_SECTIONS_DEBUG].count << GUC_LOG_DEBUG_SHIFT) | + (log->sizes[GUC_LOG_SECTIONS_CAPTURE].count << GUC_LOG_CAPTURE_SHIFT) | (offset << GUC_LOG_BUF_ADDR_SHIFT); - #undef LOG_UNIT - #undef LOG_FLAG - #undef CAPTURE_UNIT - #undef CAPTURE_FLAG - return flags; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c index b54b7883320b..8f1165146013 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_capture.c @@ -656,16 +656,17 @@ static void check_guc_capture_size(struct intel_guc *guc) struct drm_i915_private *i915 = guc_to_gt(guc)->i915; int min_size = guc_capture_output_min_size_est(guc); int spare_size = min_size * GUC_CAPTURE_OVERBUFFER_MULTIPLIER; + u32 buffer_size = intel_guc_log_section_size_capture(&guc->log); if (min_size < 0) drm_warn(&i915->drm, "Failed to calculate GuC error state capture buffer minimum size: %d!\n", min_size); - else if (min_size > CAPTURE_BUFFER_SIZE) + else if (min_size > buffer_size) drm_warn(&i915->drm, "GuC error state capture buffer is too small: %d < %d\n", - CAPTURE_BUFFER_SIZE, min_size); - else if (spare_size > CAPTURE_BUFFER_SIZE) + buffer_size, min_size); + else if (spare_size > buffer_size) drm_notice(&i915->drm, "GuC error state capture buffer maybe too small: %d < %d (min = %d)\n", - CAPTURE_BUFFER_SIZE, spare_size, min_size); + buffer_size, spare_size, min_size); } /* @@ -1294,7 +1295,8 @@ static void __guc_capture_process_output(struct intel_guc *guc) log_buf_state = guc->log.buf_addr + (sizeof(struct guc_log_buffer_state) * GUC_CAPTURE_LOG_BUFFER); - src_data = guc->log.buf_addr + intel_guc_get_log_buffer_offset(GUC_CAPTURE_LOG_BUFFER); + src_data = guc->log.buf_addr + + intel_guc_get_log_buffer_offset(&guc->log, GUC_CAPTURE_LOG_BUFFER); /* * Make a copy of the state structure, inside GuC log buffer @@ -1302,7 +1304,7 @@ static void __guc_capture_process_output(struct intel_guc *guc) * from it multiple times. */ memcpy(&log_buf_state_local, log_buf_state, sizeof(struct guc_log_buffer_state)); - buffer_size = intel_guc_get_log_buffer_size(GUC_CAPTURE_LOG_BUFFER); + buffer_size = intel_guc_get_log_buffer_size(&guc->log, GUC_CAPTURE_LOG_BUFFER); read_offset = log_buf_state_local.read_ptr; write_offset = log_buf_state_local.sampled_write_ptr; full_count = log_buf_state_local.buffer_full_cnt; @@ -1381,33 +1383,22 @@ guc_capture_reg_to_str(const struct intel_guc *guc, u32 owner, u32 type, return NULL; } -#ifdef CONFIG_DRM_I915_DEBUG_GUC -#define __out(a, ...) \ - do { \ - drm_warn((&(a)->i915->drm), __VA_ARGS__); \ - i915_error_printf((a), __VA_ARGS__); \ - } while (0) -#else -#define __out(a, ...) \ - i915_error_printf(a, __VA_ARGS__) -#endif - #define GCAP_PRINT_INTEL_ENG_INFO(ebuf, eng) \ do { \ - __out(ebuf, " i915-Eng-Name: %s command stream\n", \ - (eng)->name); \ - __out(ebuf, " i915-Eng-Inst-Class: 0x%02x\n", (eng)->class); \ - __out(ebuf, " i915-Eng-Inst-Id: 0x%02x\n", (eng)->instance); \ - __out(ebuf, " i915-Eng-LogicalMask: 0x%08x\n", \ - (eng)->logical_mask); \ + i915_error_printf(ebuf, " i915-Eng-Name: %s command stream\n", \ + (eng)->name); \ + i915_error_printf(ebuf, " i915-Eng-Inst-Class: 0x%02x\n", (eng)->class); \ + i915_error_printf(ebuf, " i915-Eng-Inst-Id: 0x%02x\n", (eng)->instance); \ + i915_error_printf(ebuf, " i915-Eng-LogicalMask: 0x%08x\n", \ + (eng)->logical_mask); \ } while (0) #define GCAP_PRINT_GUC_INST_INFO(ebuf, node) \ do { \ - __out(ebuf, " GuC-Engine-Inst-Id: 0x%08x\n", \ - (node)->eng_inst); \ - __out(ebuf, " GuC-Context-Id: 0x%08x\n", (node)->guc_id); \ - __out(ebuf, " LRCA: 0x%08x\n", (node)->lrca); \ + i915_error_printf(ebuf, " GuC-Engine-Inst-Id: 0x%08x\n", \ + (node)->eng_inst); \ + i915_error_printf(ebuf, " GuC-Context-Id: 0x%08x\n", (node)->guc_id); \ + i915_error_printf(ebuf, " LRCA: 0x%08x\n", (node)->lrca); \ } while (0) int intel_guc_capture_print_engine_node(struct drm_i915_error_state_buf *ebuf, @@ -1439,57 +1430,57 @@ int intel_guc_capture_print_engine_node(struct drm_i915_error_state_buf *ebuf, guc = &ee->engine->gt->uc.guc; - __out(ebuf, "global --- GuC Error Capture on %s command stream:\n", - ee->engine->name); + i915_error_printf(ebuf, "global --- GuC Error Capture on %s command stream:\n", + ee->engine->name); node = ee->guc_capture_node; if (!node) { - __out(ebuf, " No matching ee-node\n"); + i915_error_printf(ebuf, " No matching ee-node\n"); return 0; } - __out(ebuf, "Coverage: %s\n", grptype[node->is_partial]); + i915_error_printf(ebuf, "Coverage: %s\n", grptype[node->is_partial]); for (i = GUC_CAPTURE_LIST_TYPE_GLOBAL; i < GUC_CAPTURE_LIST_TYPE_MAX; ++i) { - __out(ebuf, " RegListType: %s\n", - datatype[i % GUC_CAPTURE_LIST_TYPE_MAX]); - __out(ebuf, " Owner-Id: %d\n", node->reginfo[i].vfid); + i915_error_printf(ebuf, " RegListType: %s\n", + datatype[i % GUC_CAPTURE_LIST_TYPE_MAX]); + i915_error_printf(ebuf, " Owner-Id: %d\n", node->reginfo[i].vfid); switch (i) { case GUC_CAPTURE_LIST_TYPE_GLOBAL: default: break; case GUC_CAPTURE_LIST_TYPE_ENGINE_CLASS: - __out(ebuf, " GuC-Eng-Class: %d\n", node->eng_class); - __out(ebuf, " i915-Eng-Class: %d\n", - guc_class_to_engine_class(node->eng_class)); + i915_error_printf(ebuf, " GuC-Eng-Class: %d\n", node->eng_class); + i915_error_printf(ebuf, " i915-Eng-Class: %d\n", + guc_class_to_engine_class(node->eng_class)); break; case GUC_CAPTURE_LIST_TYPE_ENGINE_INSTANCE: eng = intel_guc_lookup_engine(guc, node->eng_class, node->eng_inst); if (eng) GCAP_PRINT_INTEL_ENG_INFO(ebuf, eng); else - __out(ebuf, " i915-Eng-Lookup Fail!\n"); + i915_error_printf(ebuf, " i915-Eng-Lookup Fail!\n"); GCAP_PRINT_GUC_INST_INFO(ebuf, node); break; } numregs = node->reginfo[i].num_regs; - __out(ebuf, " NumRegs: %d\n", numregs); + i915_error_printf(ebuf, " NumRegs: %d\n", numregs); j = 0; while (numregs--) { regs = node->reginfo[i].regs; str = guc_capture_reg_to_str(guc, GUC_CAPTURE_LIST_INDEX_PF, i, node->eng_class, 0, regs[j].offset, &is_ext); if (!str) - __out(ebuf, " REG-0x%08x", regs[j].offset); + i915_error_printf(ebuf, " REG-0x%08x", regs[j].offset); else - __out(ebuf, " %s", str); + i915_error_printf(ebuf, " %s", str); if (is_ext) - __out(ebuf, "[%ld][%ld]", - FIELD_GET(GUC_REGSET_STEERING_GROUP, regs[j].flags), - FIELD_GET(GUC_REGSET_STEERING_INSTANCE, regs[j].flags)); - __out(ebuf, ": 0x%08x\n", regs[j].value); + i915_error_printf(ebuf, "[%ld][%ld]", + FIELD_GET(GUC_REGSET_STEERING_GROUP, regs[j].flags), + FIELD_GET(GUC_REGSET_STEERING_INSTANCE, regs[j].flags)); + i915_error_printf(ebuf, ": 0x%08x\n", regs[j].value); ++j; } } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c index 4722d4b18ed1..b071973ac41c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c @@ -13,8 +13,157 @@ #include "intel_guc_capture.h" #include "intel_guc_log.h" +#if defined(CONFIG_DRM_I915_DEBUG_GUC) +#define GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE SZ_2M +#define GUC_LOG_DEFAULT_DEBUG_BUFFER_SIZE SZ_16M +#define GUC_LOG_DEFAULT_CAPTURE_BUFFER_SIZE SZ_4M +#elif defined(CONFIG_DRM_I915_DEBUG_GEM) +#define GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE SZ_1M +#define GUC_LOG_DEFAULT_DEBUG_BUFFER_SIZE SZ_2M +#define GUC_LOG_DEFAULT_CAPTURE_BUFFER_SIZE SZ_4M +#else +#define GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE SZ_8K +#define GUC_LOG_DEFAULT_DEBUG_BUFFER_SIZE SZ_64K +#define GUC_LOG_DEFAULT_CAPTURE_BUFFER_SIZE SZ_2M +#endif + static void guc_log_copy_debuglogs_for_relay(struct intel_guc_log *log); +struct guc_log_section { + u32 max; + u32 flag; + u32 default_val; + const char *name; +}; + +static s32 scale_log_param(struct intel_guc_log *log, const struct guc_log_section *section, + s32 param) +{ + /* -1 means default */ + if (param < 0) + return section->default_val; + + /* Check for 32-bit overflow */ + if (param >= SZ_4K) { + drm_err(&guc_to_gt(log_to_guc(log))->i915->drm, "Size too large for GuC %s log: %dMB!", + section->name, param); + return section->default_val; + } + + /* Param units are 1MB */ + return param * SZ_1M; +} + +static void _guc_log_init_sizes(struct intel_guc_log *log) +{ + struct intel_guc *guc = log_to_guc(log); + struct drm_i915_private *i915 = guc_to_gt(guc)->i915; + static const struct guc_log_section sections[GUC_LOG_SECTIONS_LIMIT] = { + { + GUC_LOG_CRASH_MASK >> GUC_LOG_CRASH_SHIFT, + GUC_LOG_LOG_ALLOC_UNITS, + GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE, + "crash dump" + }, + { + GUC_LOG_DEBUG_MASK >> GUC_LOG_DEBUG_SHIFT, + GUC_LOG_LOG_ALLOC_UNITS, + GUC_LOG_DEFAULT_DEBUG_BUFFER_SIZE, + "debug", + }, + { + GUC_LOG_CAPTURE_MASK >> GUC_LOG_CAPTURE_SHIFT, + GUC_LOG_CAPTURE_ALLOC_UNITS, + GUC_LOG_DEFAULT_CAPTURE_BUFFER_SIZE, + "capture", + } + }; + s32 params[GUC_LOG_SECTIONS_LIMIT] = { + GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE / SZ_1M, + GUC_LOG_DEFAULT_DEBUG_BUFFER_SIZE / SZ_1M, + GUC_LOG_DEFAULT_CAPTURE_BUFFER_SIZE / SZ_1M, + }; + int i; + + for (i = 0; i < GUC_LOG_SECTIONS_LIMIT; i++) + log->sizes[i].bytes = scale_log_param(log, sections + i, params[i]); + + /* If debug size > 1MB then bump default crash size to keep the same units */ + if (log->sizes[GUC_LOG_SECTIONS_DEBUG].bytes >= SZ_1M && + GUC_LOG_DEFAULT_CRASH_BUFFER_SIZE < SZ_1M) + log->sizes[GUC_LOG_SECTIONS_CRASH].bytes = SZ_1M; + + /* Prepare the GuC API structure fields: */ + for (i = 0; i < GUC_LOG_SECTIONS_LIMIT; i++) { + /* Convert to correct units */ + if ((log->sizes[i].bytes % SZ_1M) == 0) { + log->sizes[i].units = SZ_1M; + log->sizes[i].flag = sections[i].flag; + } else { + log->sizes[i].units = SZ_4K; + log->sizes[i].flag = 0; + } + + if (!IS_ALIGNED(log->sizes[i].bytes, log->sizes[i].units)) + drm_err(&i915->drm, "Mis-aligned GuC log %s size: 0x%X vs 0x%X!", + sections[i].name, log->sizes[i].bytes, log->sizes[i].units); + log->sizes[i].count = log->sizes[i].bytes / log->sizes[i].units; + + if (!log->sizes[i].count) { + drm_err(&i915->drm, "Zero GuC log %s size!", sections[i].name); + } else { + /* Size is +1 unit */ + log->sizes[i].count--; + } + + /* Clip to field size */ + if (log->sizes[i].count > sections[i].max) { + drm_err(&i915->drm, "GuC log %s size too large: %d vs %d!", + sections[i].name, log->sizes[i].count + 1, sections[i].max + 1); + log->sizes[i].count = sections[i].max; + } + } + + if (log->sizes[GUC_LOG_SECTIONS_CRASH].units != log->sizes[GUC_LOG_SECTIONS_DEBUG].units) { + drm_err(&i915->drm, "Unit mis-match for GuC log crash and debug sections: %d vs %d!", + log->sizes[GUC_LOG_SECTIONS_CRASH].units, + log->sizes[GUC_LOG_SECTIONS_DEBUG].units); + log->sizes[GUC_LOG_SECTIONS_CRASH].units = log->sizes[GUC_LOG_SECTIONS_DEBUG].units; + log->sizes[GUC_LOG_SECTIONS_CRASH].count = 0; + } + + log->sizes_initialised = true; +} + +static void guc_log_init_sizes(struct intel_guc_log *log) +{ + if (log->sizes_initialised) + return; + + _guc_log_init_sizes(log); +} + +static u32 intel_guc_log_section_size_crash(struct intel_guc_log *log) +{ + guc_log_init_sizes(log); + + return log->sizes[GUC_LOG_SECTIONS_CRASH].bytes; +} + +static u32 intel_guc_log_section_size_debug(struct intel_guc_log *log) +{ + guc_log_init_sizes(log); + + return log->sizes[GUC_LOG_SECTIONS_DEBUG].bytes; +} + +u32 intel_guc_log_section_size_capture(struct intel_guc_log *log) +{ + guc_log_init_sizes(log); + + return log->sizes[GUC_LOG_SECTIONS_CAPTURE].bytes; +} + static u32 intel_guc_log_size(struct intel_guc_log *log) { /* @@ -38,7 +187,10 @@ static u32 intel_guc_log_size(struct intel_guc_log *log) * | Capture logs | * +===============================+ + CAPTURE_SIZE */ - return PAGE_SIZE + CRASH_BUFFER_SIZE + DEBUG_BUFFER_SIZE + CAPTURE_BUFFER_SIZE; + return PAGE_SIZE + + intel_guc_log_section_size_crash(log) + + intel_guc_log_section_size_debug(log) + + intel_guc_log_section_size_capture(log); } /** @@ -165,7 +317,8 @@ static void guc_move_to_next_buf(struct intel_guc_log *log) smp_wmb(); /* All data has been written, so now move the offset of sub buffer. */ - relay_reserve(log->relay.channel, log->vma->obj->base.size - CAPTURE_BUFFER_SIZE); + relay_reserve(log->relay.channel, log->vma->obj->base.size - + intel_guc_log_section_size_capture(log)); /* Switch to the next sub buffer */ relay_flush(log->relay.channel); @@ -210,15 +363,16 @@ bool intel_guc_check_log_buf_overflow(struct intel_guc_log *log, return overflow; } -unsigned int intel_guc_get_log_buffer_size(enum guc_log_buffer_type type) +unsigned int intel_guc_get_log_buffer_size(struct intel_guc_log *log, + enum guc_log_buffer_type type) { switch (type) { case GUC_DEBUG_LOG_BUFFER: - return DEBUG_BUFFER_SIZE; + return intel_guc_log_section_size_debug(log); case GUC_CRASH_DUMP_LOG_BUFFER: - return CRASH_BUFFER_SIZE; + return intel_guc_log_section_size_crash(log); case GUC_CAPTURE_LOG_BUFFER: - return CAPTURE_BUFFER_SIZE; + return intel_guc_log_section_size_capture(log); default: MISSING_CASE(type); } @@ -226,7 +380,8 @@ unsigned int intel_guc_get_log_buffer_size(enum guc_log_buffer_type type) return 0; } -size_t intel_guc_get_log_buffer_offset(enum guc_log_buffer_type type) +size_t intel_guc_get_log_buffer_offset(struct intel_guc_log *log, + enum guc_log_buffer_type type) { enum guc_log_buffer_type i; size_t offset = PAGE_SIZE;/* for the log_buffer_states */ @@ -234,7 +389,7 @@ size_t intel_guc_get_log_buffer_offset(enum guc_log_buffer_type type) for (i = GUC_DEBUG_LOG_BUFFER; i < GUC_MAX_LOG_BUFFER; ++i) { if (i == type) break; - offset += intel_guc_get_log_buffer_size(i); + offset += intel_guc_get_log_buffer_size(log, i); } return offset; @@ -285,7 +440,7 @@ static void _guc_log_copy_debuglogs_for_relay(struct intel_guc_log *log) */ memcpy(&log_buf_state_local, log_buf_state, sizeof(struct guc_log_buffer_state)); - buffer_size = intel_guc_get_log_buffer_size(type); + buffer_size = intel_guc_get_log_buffer_size(log, type); read_offset = log_buf_state_local.read_ptr; write_offset = log_buf_state_local.sampled_write_ptr; full_cnt = log_buf_state_local.buffer_full_cnt; @@ -400,7 +555,7 @@ static int guc_log_relay_create(struct intel_guc_log *log) * Keep the size of sub buffers same as shared log buffer * but GuC log-events excludes the error-state-capture logs */ - subbuf_size = log->vma->size - CAPTURE_BUFFER_SIZE; + subbuf_size = log->vma->size - intel_guc_log_section_size_capture(log); /* * Store up to 8 snapshots, which is large enough to buffer sufficient diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h index dc9715411d62..02127703be80 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h @@ -15,20 +15,6 @@ struct intel_guc; -#if defined(CONFIG_DRM_I915_DEBUG_GUC) -#define CRASH_BUFFER_SIZE SZ_2M -#define DEBUG_BUFFER_SIZE SZ_16M -#define CAPTURE_BUFFER_SIZE SZ_4M -#elif defined(CONFIG_DRM_I915_DEBUG_GEM) -#define CRASH_BUFFER_SIZE SZ_1M -#define DEBUG_BUFFER_SIZE SZ_2M -#define CAPTURE_BUFFER_SIZE SZ_4M -#else -#define CRASH_BUFFER_SIZE SZ_8K -#define DEBUG_BUFFER_SIZE SZ_64K -#define CAPTURE_BUFFER_SIZE SZ_2M -#endif - /* * While we're using plain log level in i915, GuC controls are much more... * "elaborate"? We have a couple of bits for verbosity, separate bit for actual @@ -46,10 +32,30 @@ struct intel_guc; #define GUC_VERBOSITY_TO_LOG_LEVEL(x) ((x) + 2) #define GUC_LOG_LEVEL_MAX GUC_VERBOSITY_TO_LOG_LEVEL(GUC_LOG_VERBOSITY_MAX) +enum { + GUC_LOG_SECTIONS_CRASH, + GUC_LOG_SECTIONS_DEBUG, + GUC_LOG_SECTIONS_CAPTURE, + GUC_LOG_SECTIONS_LIMIT +}; + struct intel_guc_log { u32 level; + + /* Allocation settings */ + struct { + s32 bytes; /* Size in bytes */ + s32 units; /* GuC API units - 1MB or 4KB */ + s32 count; /* Number of API units */ + u32 flag; /* GuC API units flag */ + } sizes[GUC_LOG_SECTIONS_LIMIT]; + bool sizes_initialised; + + /* Combined buffer allocation */ struct i915_vma *vma; void *buf_addr; + + /* RelayFS support */ struct { bool buf_in_use; bool started; @@ -58,6 +64,7 @@ struct intel_guc_log { struct mutex lock; u32 full_count; } relay; + /* logging related stats */ struct { u32 sampled_overflow; @@ -69,8 +76,9 @@ struct intel_guc_log { void intel_guc_log_init_early(struct intel_guc_log *log); bool intel_guc_check_log_buf_overflow(struct intel_guc_log *log, enum guc_log_buffer_type type, unsigned int full_cnt); -unsigned int intel_guc_get_log_buffer_size(enum guc_log_buffer_type type); -size_t intel_guc_get_log_buffer_offset(enum guc_log_buffer_type type); +unsigned int intel_guc_get_log_buffer_size(struct intel_guc_log *log, + enum guc_log_buffer_type type); +size_t intel_guc_get_log_buffer_offset(struct intel_guc_log *log, enum guc_log_buffer_type type); int intel_guc_log_create(struct intel_guc_log *log); void intel_guc_log_destroy(struct intel_guc_log *log); @@ -92,4 +100,6 @@ void intel_guc_log_info(struct intel_guc_log *log, struct drm_printer *p); int intel_guc_log_dump(struct intel_guc_log *log, struct drm_printer *p, bool dump_load_err); +u32 intel_guc_log_section_size_capture(struct intel_guc_log *log); + #endif diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c index e1fa1f32f29e..fdd895f73f9f 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c @@ -137,17 +137,6 @@ static int guc_action_slpc_set_param(struct intel_guc *guc, u8 id, u32 value) return ret > 0 ? -EPROTO : ret; } -static int guc_action_slpc_unset_param(struct intel_guc *guc, u8 id) -{ - u32 request[] = { - GUC_ACTION_HOST2GUC_PC_SLPC_REQUEST, - SLPC_EVENT(SLPC_EVENT_PARAMETER_UNSET, 1), - id, - }; - - return intel_guc_send(guc, request, ARRAY_SIZE(request)); -} - static bool slpc_is_running(struct intel_guc_slpc *slpc) { return slpc_get_state(slpc) == SLPC_GLOBAL_STATE_RUNNING; @@ -201,16 +190,6 @@ static int slpc_set_param(struct intel_guc_slpc *slpc, u8 id, u32 value) return ret; } -static int slpc_unset_param(struct intel_guc_slpc *slpc, - u8 id) -{ - struct intel_guc *guc = slpc_to_guc(slpc); - - GEM_BUG_ON(id >= SLPC_MAX_PARAM); - - return guc_action_slpc_unset_param(guc, id); -} - static int slpc_force_min_freq(struct intel_guc_slpc *slpc, u32 freq) { struct drm_i915_private *i915 = slpc_to_i915(slpc); @@ -488,23 +467,33 @@ int intel_guc_slpc_set_min_freq(struct intel_guc_slpc *slpc, u32 val) /* Need a lock now since waitboost can be modifying min as well */ mutex_lock(&slpc->lock); - - with_intel_runtime_pm(&i915->runtime_pm, wakeref) { - - ret = slpc_set_param(slpc, - SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ, - val); - - /* Return standardized err code for sysfs calls */ - if (ret) - ret = -EIO; + wakeref = intel_runtime_pm_get(&i915->runtime_pm); + + /* Ignore efficient freq if lower min freq is requested */ + ret = slpc_set_param(slpc, + SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY, + val < slpc->rp1_freq); + if (ret) { + i915_probe_error(i915, "Failed to toggle efficient freq (%pe)\n", + ERR_PTR(ret)); + goto out; } + ret = slpc_set_param(slpc, + SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ, + val); + if (!ret) slpc->min_freq_softlimit = val; +out: + intel_runtime_pm_put(&i915->runtime_pm, wakeref); mutex_unlock(&slpc->lock); + /* Return standardized err code for sysfs calls */ + if (ret) + ret = -EIO; + return ret; } @@ -587,7 +576,9 @@ static int slpc_set_softlimits(struct intel_guc_slpc *slpc) return ret; if (!slpc->min_freq_softlimit) { - slpc->min_freq_softlimit = slpc->min_freq; + ret = intel_guc_slpc_get_min_freq(slpc, &slpc->min_freq_softlimit); + if (unlikely(ret)) + return ret; slpc_to_gt(slpc)->defaults.min_freq = slpc->min_freq_softlimit; } else if (slpc->min_freq_softlimit != slpc->min_freq) { return intel_guc_slpc_set_min_freq(slpc, @@ -597,29 +588,6 @@ static int slpc_set_softlimits(struct intel_guc_slpc *slpc) return 0; } -static int slpc_ignore_eff_freq(struct intel_guc_slpc *slpc, bool ignore) -{ - int ret = 0; - - if (ignore) { - ret = slpc_set_param(slpc, - SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY, - ignore); - if (!ret) - return slpc_set_param(slpc, - SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ, - slpc->min_freq); - } else { - ret = slpc_unset_param(slpc, - SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY); - if (!ret) - return slpc_unset_param(slpc, - SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ); - } - - return ret; -} - static int slpc_use_fused_rp0(struct intel_guc_slpc *slpc) { /* Force SLPC to used platform rp0 */ @@ -679,14 +647,6 @@ int intel_guc_slpc_enable(struct intel_guc_slpc *slpc) slpc_get_rp_values(slpc); - /* Ignore efficient freq and set min to platform min */ - ret = slpc_ignore_eff_freq(slpc, true); - if (unlikely(ret)) { - i915_probe_error(i915, "Failed to set SLPC min to RPn (%pe)\n", - ERR_PTR(ret)); - return ret; - } - /* Set SLPC max limit to RP0 */ ret = slpc_use_fused_rp0(slpc); if (unlikely(ret)) { diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 0d17da77e787..22ba66e48a9b 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -1438,7 +1438,12 @@ void intel_guc_busyness_park(struct intel_gt *gt) if (!guc_submission_initialized(guc)) return; - cancel_delayed_work(&guc->timestamp.work); + /* + * There is a race with suspend flow where the worker runs after suspend + * and causes an unclaimed register access warning. Cancel the worker + * synchronously here. + */ + cancel_delayed_work_sync(&guc->timestamp.work); /* * Before parking, we should sample engine busyness stats if we need to. @@ -1532,8 +1537,8 @@ void intel_guc_submission_reset_prepare(struct intel_guc *guc) __reset_guc_busyness_stats(guc); /* Flush IRQ handler */ - spin_lock_irq(&guc_to_gt(guc)->irq_lock); - spin_unlock_irq(&guc_to_gt(guc)->irq_lock); + spin_lock_irq(guc_to_gt(guc)->irq_lock); + spin_unlock_irq(guc_to_gt(guc)->irq_lock); guc_flush_submissions(guc); guc_flush_destroyed_contexts(guc); @@ -1868,7 +1873,7 @@ int intel_guc_submission_init(struct intel_guc *guc) if (guc->submission_initialized) return 0; - if (guc->fw.major_ver_found < 70) { + if (GET_UC_VER(guc) < MAKE_UC_VER(70, 0, 0)) { ret = guc_lrc_desc_pool_create_v69(guc); if (ret) return ret; @@ -2303,7 +2308,7 @@ static int register_context(struct intel_context *ce, bool loop) GEM_BUG_ON(intel_context_is_child(ce)); trace_intel_context_register(ce); - if (guc->fw.major_ver_found >= 70) + if (GET_UC_VER(guc) >= MAKE_UC_VER(70, 0, 0)) ret = register_context_v70(guc, ce, loop); else ret = register_context_v69(guc, ce, loop); @@ -2315,7 +2320,7 @@ static int register_context(struct intel_context *ce, bool loop) set_context_registered(ce); spin_unlock_irqrestore(&ce->guc_state.lock, flags); - if (guc->fw.major_ver_found >= 70) + if (GET_UC_VER(guc) >= MAKE_UC_VER(70, 0, 0)) guc_context_policy_init_v70(ce, loop); } @@ -2921,7 +2926,7 @@ static void __guc_context_set_preemption_timeout(struct intel_guc *guc, u16 guc_id, u32 preemption_timeout) { - if (guc->fw.major_ver_found >= 70) { + if (GET_UC_VER(guc) >= MAKE_UC_VER(70, 0, 0)) { struct context_policy policy; __guc_context_policy_start_klv(&policy, guc_id); @@ -3186,7 +3191,7 @@ static int guc_context_alloc(struct intel_context *ce) static void __guc_context_set_prio(struct intel_guc *guc, struct intel_context *ce) { - if (guc->fw.major_ver_found >= 70) { + if (GET_UC_VER(guc) >= MAKE_UC_VER(70, 0, 0)) { struct context_policy policy; __guc_context_policy_start_klv(&policy, ce->guc_id.id); @@ -4003,6 +4008,13 @@ static inline void guc_init_lrc_mapping(struct intel_guc *guc) xa_destroy(&guc->context_lookup); /* + * A reset might have occurred while we had a pending stalled request, + * so make sure we clean that up. + */ + guc->stalled_request = NULL; + guc->submission_stall_reason = STALL_NONE; + + /* * Some contexts might have been pinned before we enabled GuC * submission, so we need to add them to the GuC bookeeping. * Also, after a reset the of the GuC we want to make sure that the diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index f2e7c82985ef..dbd048b77e19 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -245,9 +245,9 @@ static int guc_enable_communication(struct intel_guc *guc) intel_guc_enable_interrupts(guc); /* check for CT messages received before we enabled interrupts */ - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); intel_guc_ct_event_handler(&guc->ct); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); drm_dbg(&i915->drm, "GuC communication enabled\n"); @@ -435,9 +435,11 @@ static void print_fw_ver(struct intel_uc *uc, struct intel_uc_fw *fw) { struct drm_i915_private *i915 = uc_to_gt(uc)->i915; - drm_info(&i915->drm, "%s firmware %s version %u.%u\n", - intel_uc_fw_type_repr(fw->type), fw->path, - fw->major_ver_found, fw->minor_ver_found); + drm_info(&i915->drm, "%s firmware %s version %u.%u.%u\n", + intel_uc_fw_type_repr(fw->type), fw->file_selected.path, + fw->file_selected.major_ver, + fw->file_selected.minor_ver, + fw->file_selected.patch_ver); } static int __uc_init_hw(struct intel_uc *uc) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index 58547292efa0..b91ad4aede1f 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -41,7 +41,7 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, "%s firmware -> %s\n", intel_uc_fw_type_repr(uc_fw->type), status == INTEL_UC_FIRMWARE_SELECTED ? - uc_fw->path : intel_uc_fw_status_repr(status)); + uc_fw->file_selected.path : intel_uc_fw_status_repr(status)); } #endif @@ -51,84 +51,153 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, * * Note that RKL and ADL-S have the same GuC/HuC device ID's and use the same * firmware as TGL. + * + * Version numbers: + * Originally, the driver required an exact match major/minor/patch furmware + * file and only supported that one version for any given platform. However, + * the new direction from upstream is to be backwards compatible with all + * prior releases and to be as flexible as possible as to what firmware is + * loaded. + * + * For GuC, the major version number signifies a backwards breaking API change. + * So, new format GuC firmware files are labelled by their major version only. + * For HuC, there is no KMD interaction, hence no version matching requirement. + * So, new format HuC firmware files have no version number at all. + * + * All of which means that the table below must keep all old format files with + * full three point version number. But newer files have reduced requirements. + * Having said that, the driver still needs to track the minor version number + * for GuC at least. As it is useful to report to the user that they are not + * running with a recent enough version for all KMD supported features, + * security fixes, etc. to be enabled. + */ +#define INTEL_GUC_FIRMWARE_DEFS(fw_def, guc_maj, guc_mmp) \ + fw_def(DG2, 0, guc_maj(dg2, 70, 5)) \ + fw_def(ALDERLAKE_P, 0, guc_maj(adlp, 70, 5)) \ + fw_def(ALDERLAKE_P, 0, guc_mmp(adlp, 70, 1, 1)) \ + fw_def(ALDERLAKE_P, 0, guc_mmp(adlp, 69, 0, 3)) \ + fw_def(ALDERLAKE_S, 0, guc_maj(tgl, 70, 5)) \ + fw_def(ALDERLAKE_S, 0, guc_mmp(tgl, 70, 1, 1)) \ + fw_def(ALDERLAKE_S, 0, guc_mmp(tgl, 69, 0, 3)) \ + fw_def(DG1, 0, guc_maj(dg1, 70, 5)) \ + fw_def(ROCKETLAKE, 0, guc_mmp(tgl, 70, 1, 1)) \ + fw_def(TIGERLAKE, 0, guc_mmp(tgl, 70, 1, 1)) \ + fw_def(JASPERLAKE, 0, guc_mmp(ehl, 70, 1, 1)) \ + fw_def(ELKHARTLAKE, 0, guc_mmp(ehl, 70, 1, 1)) \ + fw_def(ICELAKE, 0, guc_mmp(icl, 70, 1, 1)) \ + fw_def(COMETLAKE, 5, guc_mmp(cml, 70, 1, 1)) \ + fw_def(COMETLAKE, 0, guc_mmp(kbl, 70, 1, 1)) \ + fw_def(COFFEELAKE, 0, guc_mmp(kbl, 70, 1, 1)) \ + fw_def(GEMINILAKE, 0, guc_mmp(glk, 70, 1, 1)) \ + fw_def(KABYLAKE, 0, guc_mmp(kbl, 70, 1, 1)) \ + fw_def(BROXTON, 0, guc_mmp(bxt, 70, 1, 1)) \ + fw_def(SKYLAKE, 0, guc_mmp(skl, 70, 1, 1)) + +#define INTEL_HUC_FIRMWARE_DEFS(fw_def, huc_raw, huc_mmp) \ + fw_def(ALDERLAKE_P, 0, huc_raw(tgl)) \ + fw_def(ALDERLAKE_P, 0, huc_mmp(tgl, 7, 9, 3)) \ + fw_def(ALDERLAKE_S, 0, huc_raw(tgl)) \ + fw_def(ALDERLAKE_S, 0, huc_mmp(tgl, 7, 9, 3)) \ + fw_def(DG1, 0, huc_raw(dg1)) \ + fw_def(ROCKETLAKE, 0, huc_mmp(tgl, 7, 9, 3)) \ + fw_def(TIGERLAKE, 0, huc_mmp(tgl, 7, 9, 3)) \ + fw_def(JASPERLAKE, 0, huc_mmp(ehl, 9, 0, 0)) \ + fw_def(ELKHARTLAKE, 0, huc_mmp(ehl, 9, 0, 0)) \ + fw_def(ICELAKE, 0, huc_mmp(icl, 9, 0, 0)) \ + fw_def(COMETLAKE, 5, huc_mmp(cml, 4, 0, 0)) \ + fw_def(COMETLAKE, 0, huc_mmp(kbl, 4, 0, 0)) \ + fw_def(COFFEELAKE, 0, huc_mmp(kbl, 4, 0, 0)) \ + fw_def(GEMINILAKE, 0, huc_mmp(glk, 4, 0, 0)) \ + fw_def(KABYLAKE, 0, huc_mmp(kbl, 4, 0, 0)) \ + fw_def(BROXTON, 0, huc_mmp(bxt, 2, 0, 0)) \ + fw_def(SKYLAKE, 0, huc_mmp(skl, 2, 0, 0)) + +/* + * Set of macros for producing a list of filenames from the above table. */ -#define INTEL_GUC_FIRMWARE_DEFS(fw_def, guc_def) \ - fw_def(DG2, 0, guc_def(dg2, 70, 4, 1)) \ - fw_def(ALDERLAKE_P, 0, guc_def(adlp, 70, 1, 1)) \ - fw_def(ALDERLAKE_S, 0, guc_def(tgl, 70, 1, 1)) \ - fw_def(DG1, 0, guc_def(dg1, 70, 1, 1)) \ - fw_def(ROCKETLAKE, 0, guc_def(tgl, 70, 1, 1)) \ - fw_def(TIGERLAKE, 0, guc_def(tgl, 70, 1, 1)) \ - fw_def(JASPERLAKE, 0, guc_def(ehl, 70, 1, 1)) \ - fw_def(ELKHARTLAKE, 0, guc_def(ehl, 70, 1, 1)) \ - fw_def(ICELAKE, 0, guc_def(icl, 70, 1, 1)) \ - fw_def(COMETLAKE, 5, guc_def(cml, 70, 1, 1)) \ - fw_def(COMETLAKE, 0, guc_def(kbl, 70, 1, 1)) \ - fw_def(COFFEELAKE, 0, guc_def(kbl, 70, 1, 1)) \ - fw_def(GEMINILAKE, 0, guc_def(glk, 70, 1, 1)) \ - fw_def(KABYLAKE, 0, guc_def(kbl, 70, 1, 1)) \ - fw_def(BROXTON, 0, guc_def(bxt, 70, 1, 1)) \ - fw_def(SKYLAKE, 0, guc_def(skl, 70, 1, 1)) - -#define INTEL_GUC_FIRMWARE_DEFS_FALLBACK(fw_def, guc_def) \ - fw_def(ALDERLAKE_P, 0, guc_def(adlp, 69, 0, 3)) \ - fw_def(ALDERLAKE_S, 0, guc_def(tgl, 69, 0, 3)) - -#define INTEL_HUC_FIRMWARE_DEFS(fw_def, huc_def) \ - fw_def(ALDERLAKE_P, 0, huc_def(tgl, 7, 9, 3)) \ - fw_def(ALDERLAKE_S, 0, huc_def(tgl, 7, 9, 3)) \ - fw_def(DG1, 0, huc_def(dg1, 7, 9, 3)) \ - fw_def(ROCKETLAKE, 0, huc_def(tgl, 7, 9, 3)) \ - fw_def(TIGERLAKE, 0, huc_def(tgl, 7, 9, 3)) \ - fw_def(JASPERLAKE, 0, huc_def(ehl, 9, 0, 0)) \ - fw_def(ELKHARTLAKE, 0, huc_def(ehl, 9, 0, 0)) \ - fw_def(ICELAKE, 0, huc_def(icl, 9, 0, 0)) \ - fw_def(COMETLAKE, 5, huc_def(cml, 4, 0, 0)) \ - fw_def(COMETLAKE, 0, huc_def(kbl, 4, 0, 0)) \ - fw_def(COFFEELAKE, 0, huc_def(kbl, 4, 0, 0)) \ - fw_def(GEMINILAKE, 0, huc_def(glk, 4, 0, 0)) \ - fw_def(KABYLAKE, 0, huc_def(kbl, 4, 0, 0)) \ - fw_def(BROXTON, 0, huc_def(bxt, 2, 0, 0)) \ - fw_def(SKYLAKE, 0, huc_def(skl, 2, 0, 0)) - -#define __MAKE_UC_FW_PATH(prefix_, name_, major_, minor_, patch_) \ +#define __MAKE_UC_FW_PATH_BLANK(prefix_, name_) \ + "i915/" \ + __stringify(prefix_) name_ ".bin" + +#define __MAKE_UC_FW_PATH_MAJOR(prefix_, name_, major_) \ + "i915/" \ + __stringify(prefix_) name_ \ + __stringify(major_) ".bin" + +#define __MAKE_UC_FW_PATH_MMP(prefix_, name_, major_, minor_, patch_) \ "i915/" \ __stringify(prefix_) name_ \ __stringify(major_) "." \ __stringify(minor_) "." \ __stringify(patch_) ".bin" -#define MAKE_GUC_FW_PATH(prefix_, major_, minor_, patch_) \ - __MAKE_UC_FW_PATH(prefix_, "_guc_", major_, minor_, patch_) +/* Minor for internal driver use, not part of file name */ +#define MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_) \ + __MAKE_UC_FW_PATH_MAJOR(prefix_, "_guc_", major_) + +#define MAKE_GUC_FW_PATH_MMP(prefix_, major_, minor_, patch_) \ + __MAKE_UC_FW_PATH_MMP(prefix_, "_guc_", major_, minor_, patch_) -#define MAKE_HUC_FW_PATH(prefix_, major_, minor_, bld_num_) \ - __MAKE_UC_FW_PATH(prefix_, "_huc_", major_, minor_, bld_num_) +#define MAKE_HUC_FW_PATH_BLANK(prefix_) \ + __MAKE_UC_FW_PATH_BLANK(prefix_, "_huc") -/* All blobs need to be declared via MODULE_FIRMWARE() */ +#define MAKE_HUC_FW_PATH_MMP(prefix_, major_, minor_, patch_) \ + __MAKE_UC_FW_PATH_MMP(prefix_, "_huc_", major_, minor_, patch_) + +/* + * All blobs need to be declared via MODULE_FIRMWARE(). + * This first expansion of the table macros is solely to provide + * that declaration. + */ #define INTEL_UC_MODULE_FW(platform_, revid_, uc_) \ MODULE_FIRMWARE(uc_); -INTEL_GUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_GUC_FW_PATH) -INTEL_GUC_FIRMWARE_DEFS_FALLBACK(INTEL_UC_MODULE_FW, MAKE_GUC_FW_PATH) -INTEL_HUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_HUC_FW_PATH) +INTEL_GUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_GUC_FW_PATH_MAJOR, MAKE_GUC_FW_PATH_MMP) +INTEL_HUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_HUC_FW_PATH_BLANK, MAKE_HUC_FW_PATH_MMP) -/* The below structs and macros are used to iterate across the list of blobs */ +/* + * The next expansion of the table macros (in __uc_fw_auto_select below) provides + * actual data structures with both the filename and the version information. + * These structure arrays are then iterated over to the list of suitable files + * for the current platform and to then attempt to load those files, in the order + * listed, until one is successfully found. + */ struct __packed uc_fw_blob { + const char *path; + bool legacy; u8 major; u8 minor; - const char *path; + u8 patch; }; -#define UC_FW_BLOB(major_, minor_, path_) \ - { .major = major_, .minor = minor_, .path = path_ } +#define UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ + .major = major_, \ + .minor = minor_, \ + .patch = patch_, \ + .path = path_, + +#define UC_FW_BLOB_NEW(major_, minor_, patch_, path_) \ + { UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ + .legacy = false } -#define GUC_FW_BLOB(prefix_, major_, minor_, patch_) \ - UC_FW_BLOB(major_, minor_, \ - MAKE_GUC_FW_PATH(prefix_, major_, minor_, patch_)) +#define UC_FW_BLOB_OLD(major_, minor_, patch_, path_) \ + { UC_FW_BLOB_BASE(major_, minor_, patch_, path_) \ + .legacy = true } -#define HUC_FW_BLOB(prefix_, major_, minor_, bld_num_) \ - UC_FW_BLOB(major_, minor_, \ - MAKE_HUC_FW_PATH(prefix_, major_, minor_, bld_num_)) +#define GUC_FW_BLOB(prefix_, major_, minor_) \ + UC_FW_BLOB_NEW(major_, minor_, 0, \ + MAKE_GUC_FW_PATH_MAJOR(prefix_, major_, minor_)) + +#define GUC_FW_BLOB_MMP(prefix_, major_, minor_, patch_) \ + UC_FW_BLOB_OLD(major_, minor_, patch_, \ + MAKE_GUC_FW_PATH_MMP(prefix_, major_, minor_, patch_)) + +#define HUC_FW_BLOB(prefix_) \ + UC_FW_BLOB_NEW(0, 0, 0, MAKE_HUC_FW_PATH_BLANK(prefix_)) + +#define HUC_FW_BLOB_MMP(prefix_, major_, minor_, patch_) \ + UC_FW_BLOB_OLD(major_, minor_, patch_, \ + MAKE_HUC_FW_PATH_MMP(prefix_, major_, minor_, patch_)) struct __packed uc_fw_platform_requirement { enum intel_platform p; @@ -152,23 +221,22 @@ static void __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) { static const struct uc_fw_platform_requirement blobs_guc[] = { - INTEL_GUC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB) - }; - static const struct uc_fw_platform_requirement blobs_guc_fallback[] = { - INTEL_GUC_FIRMWARE_DEFS_FALLBACK(MAKE_FW_LIST, GUC_FW_BLOB) + INTEL_GUC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB, GUC_FW_BLOB_MMP) }; static const struct uc_fw_platform_requirement blobs_huc[] = { - INTEL_HUC_FIRMWARE_DEFS(MAKE_FW_LIST, HUC_FW_BLOB) + INTEL_HUC_FIRMWARE_DEFS(MAKE_FW_LIST, HUC_FW_BLOB, HUC_FW_BLOB_MMP) }; static const struct fw_blobs_by_type blobs_all[INTEL_UC_FW_NUM_TYPES] = { [INTEL_UC_FW_TYPE_GUC] = { blobs_guc, ARRAY_SIZE(blobs_guc) }, [INTEL_UC_FW_TYPE_HUC] = { blobs_huc, ARRAY_SIZE(blobs_huc) }, }; + static bool verified; const struct uc_fw_platform_requirement *fw_blobs; enum intel_platform p = INTEL_INFO(i915)->platform; u32 fw_count; u8 rev = INTEL_REVID(i915); int i; + bool found; /* * The only difference between the ADL GuC FWs is the HWConfig support. @@ -183,50 +251,102 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) fw_blobs = blobs_all[uc_fw->type].blobs; fw_count = blobs_all[uc_fw->type].count; + found = false; for (i = 0; i < fw_count && p <= fw_blobs[i].p; i++) { - if (p == fw_blobs[i].p && rev >= fw_blobs[i].rev) { - const struct uc_fw_blob *blob = &fw_blobs[i].blob; - uc_fw->path = blob->path; - uc_fw->wanted_path = blob->path; - uc_fw->major_ver_wanted = blob->major; - uc_fw->minor_ver_wanted = blob->minor; - break; - } - } + const struct uc_fw_blob *blob = &fw_blobs[i].blob; - if (uc_fw->type == INTEL_UC_FW_TYPE_GUC) { - const struct uc_fw_platform_requirement *blobs = blobs_guc_fallback; - u32 count = ARRAY_SIZE(blobs_guc_fallback); + if (p != fw_blobs[i].p) + continue; - for (i = 0; i < count && p <= blobs[i].p; i++) { - if (p == blobs[i].p && rev >= blobs[i].rev) { - const struct uc_fw_blob *blob = &blobs[i].blob; + if (rev < fw_blobs[i].rev) + continue; - uc_fw->fallback.path = blob->path; - uc_fw->fallback.major_ver = blob->major; - uc_fw->fallback.minor_ver = blob->minor; - break; - } + if (uc_fw->file_selected.path) { + if (uc_fw->file_selected.path == blob->path) + uc_fw->file_selected.path = NULL; + + continue; } + + uc_fw->file_selected.path = blob->path; + uc_fw->file_wanted.path = blob->path; + uc_fw->file_wanted.major_ver = blob->major; + uc_fw->file_wanted.minor_ver = blob->minor; + found = true; + break; + } + + if (!found && uc_fw->file_selected.path) { + /* Failed to find a match for the last attempt?! */ + uc_fw->file_selected.path = NULL; } /* make sure the list is ordered as expected */ - if (IS_ENABLED(CONFIG_DRM_I915_SELFTEST)) { + if (IS_ENABLED(CONFIG_DRM_I915_SELFTEST) && !verified) { + verified = true; + for (i = 1; i < fw_count; i++) { + /* Next platform is good: */ if (fw_blobs[i].p < fw_blobs[i - 1].p) continue; + /* Next platform revision is good: */ if (fw_blobs[i].p == fw_blobs[i - 1].p && fw_blobs[i].rev < fw_blobs[i - 1].rev) continue; - drm_err(&i915->drm, "Invalid FW blob order: %s r%u comes before %s r%u\n", - intel_platform_name(fw_blobs[i - 1].p), - fw_blobs[i - 1].rev, - intel_platform_name(fw_blobs[i].p), - fw_blobs[i].rev); + /* Platform/revision must be in order: */ + if (fw_blobs[i].p != fw_blobs[i - 1].p || + fw_blobs[i].rev != fw_blobs[i - 1].rev) + goto bad; + + /* Next major version is good: */ + if (fw_blobs[i].blob.major < fw_blobs[i - 1].blob.major) + continue; + + /* New must be before legacy: */ + if (!fw_blobs[i].blob.legacy && fw_blobs[i - 1].blob.legacy) + goto bad; + + /* New to legacy also means 0.0 to X.Y (HuC), or X.0 to X.Y (GuC) */ + if (fw_blobs[i].blob.legacy && !fw_blobs[i - 1].blob.legacy) { + if (!fw_blobs[i - 1].blob.major) + continue; + + if (fw_blobs[i].blob.major == fw_blobs[i - 1].blob.major) + continue; + } + + /* Major versions must be in order: */ + if (fw_blobs[i].blob.major != fw_blobs[i - 1].blob.major) + goto bad; + + /* Next minor version is good: */ + if (fw_blobs[i].blob.minor < fw_blobs[i - 1].blob.minor) + continue; + + /* Minor versions must be in order: */ + if (fw_blobs[i].blob.minor != fw_blobs[i - 1].blob.minor) + goto bad; + + /* Patch versions must be in order: */ + if (fw_blobs[i].blob.patch <= fw_blobs[i - 1].blob.patch) + continue; - uc_fw->path = NULL; +bad: + drm_err(&i915->drm, "Invalid FW blob order: %s r%u %s%d.%d.%d comes before %s r%u %s%d.%d.%d\n", + intel_platform_name(fw_blobs[i - 1].p), fw_blobs[i - 1].rev, + fw_blobs[i - 1].blob.legacy ? "L" : "v", + fw_blobs[i - 1].blob.major, + fw_blobs[i - 1].blob.minor, + fw_blobs[i - 1].blob.patch, + intel_platform_name(fw_blobs[i].p), fw_blobs[i].rev, + fw_blobs[i].blob.legacy ? "L" : "v", + fw_blobs[i].blob.major, + fw_blobs[i].blob.minor, + fw_blobs[i].blob.patch); + + uc_fw->file_selected.path = NULL; } } } @@ -259,7 +379,7 @@ static void __uc_fw_user_override(struct drm_i915_private *i915, struct intel_uc } if (unlikely(path)) { - uc_fw->path = path; + uc_fw->file_selected.path = path; uc_fw->user_overridden = true; } } @@ -283,7 +403,7 @@ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, */ BUILD_BUG_ON(INTEL_UC_FIRMWARE_UNINITIALIZED); GEM_BUG_ON(uc_fw->status); - GEM_BUG_ON(uc_fw->path); + GEM_BUG_ON(uc_fw->file_selected.path); uc_fw->type = type; @@ -292,7 +412,7 @@ void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, __uc_fw_user_override(i915, uc_fw); } - intel_uc_fw_change_status(uc_fw, uc_fw->path ? *uc_fw->path ? + intel_uc_fw_change_status(uc_fw, uc_fw->file_selected.path ? *uc_fw->file_selected.path ? INTEL_UC_FIRMWARE_SELECTED : INTEL_UC_FIRMWARE_DISABLED : INTEL_UC_FIRMWARE_NOT_SUPPORTED); @@ -305,32 +425,32 @@ static void __force_fw_fetch_failures(struct intel_uc_fw *uc_fw, int e) if (i915_inject_probe_error(i915, e)) { /* non-existing blob */ - uc_fw->path = "<invalid>"; + uc_fw->file_selected.path = "<invalid>"; uc_fw->user_overridden = user; } else if (i915_inject_probe_error(i915, e)) { /* require next major version */ - uc_fw->major_ver_wanted += 1; - uc_fw->minor_ver_wanted = 0; + uc_fw->file_wanted.major_ver += 1; + uc_fw->file_wanted.minor_ver = 0; uc_fw->user_overridden = user; } else if (i915_inject_probe_error(i915, e)) { /* require next minor version */ - uc_fw->minor_ver_wanted += 1; + uc_fw->file_wanted.minor_ver += 1; uc_fw->user_overridden = user; - } else if (uc_fw->major_ver_wanted && + } else if (uc_fw->file_wanted.major_ver && i915_inject_probe_error(i915, e)) { /* require prev major version */ - uc_fw->major_ver_wanted -= 1; - uc_fw->minor_ver_wanted = 0; + uc_fw->file_wanted.major_ver -= 1; + uc_fw->file_wanted.minor_ver = 0; uc_fw->user_overridden = user; - } else if (uc_fw->minor_ver_wanted && + } else if (uc_fw->file_wanted.minor_ver && i915_inject_probe_error(i915, e)) { /* require prev minor version - hey, this should work! */ - uc_fw->minor_ver_wanted -= 1; + uc_fw->file_wanted.minor_ver -= 1; uc_fw->user_overridden = user; } else if (user && i915_inject_probe_error(i915, e)) { /* officially unsupported platform */ - uc_fw->major_ver_wanted = 0; - uc_fw->minor_ver_wanted = 0; + uc_fw->file_wanted.major_ver = 0; + uc_fw->file_wanted.minor_ver = 0; uc_fw->user_overridden = true; } } @@ -339,10 +459,12 @@ static int check_gsc_manifest(const struct firmware *fw, struct intel_uc_fw *uc_fw) { u32 *dw = (u32 *)fw->data; - u32 version = dw[HUC_GSC_VERSION_DW]; + u32 version_hi = dw[HUC_GSC_VERSION_HI_DW]; + u32 version_lo = dw[HUC_GSC_VERSION_LO_DW]; - uc_fw->major_ver_found = FIELD_GET(HUC_GSC_MAJOR_VER_MASK, version); - uc_fw->minor_ver_found = FIELD_GET(HUC_GSC_MINOR_VER_MASK, version); + uc_fw->file_selected.major_ver = FIELD_GET(HUC_GSC_MAJOR_VER_HI_MASK, version_hi); + uc_fw->file_selected.minor_ver = FIELD_GET(HUC_GSC_MINOR_VER_HI_MASK, version_hi); + uc_fw->file_selected.patch_ver = FIELD_GET(HUC_GSC_PATCH_VER_LO_MASK, version_lo); return 0; } @@ -357,7 +479,7 @@ static int check_ccs_header(struct drm_i915_private *i915, /* Check the size of the blob before examining buffer contents */ if (unlikely(fw->size < sizeof(struct uc_css_header))) { drm_warn(&i915->drm, "%s firmware %s: invalid size: %zu < %zu\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, fw->size, sizeof(struct uc_css_header)); return -ENODATA; } @@ -370,7 +492,7 @@ static int check_ccs_header(struct drm_i915_private *i915, if (unlikely(size != sizeof(struct uc_css_header))) { drm_warn(&i915->drm, "%s firmware %s: unexpected header size: %zu != %zu\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, fw->size, sizeof(struct uc_css_header)); return -EPROTO; } @@ -385,7 +507,7 @@ static int check_ccs_header(struct drm_i915_private *i915, size = sizeof(struct uc_css_header) + uc_fw->ucode_size + uc_fw->rsa_size; if (unlikely(fw->size < size)) { drm_warn(&i915->drm, "%s firmware %s: invalid size: %zu < %zu\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, fw->size, size); return -ENOEXEC; } @@ -394,16 +516,18 @@ static int check_ccs_header(struct drm_i915_private *i915, size = __intel_uc_fw_get_upload_size(uc_fw); if (unlikely(size >= i915->wopcm.size)) { drm_warn(&i915->drm, "%s firmware %s: invalid size: %zu > %zu\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, size, (size_t)i915->wopcm.size); return -E2BIG; } /* Get version numbers from the CSS header */ - uc_fw->major_ver_found = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, - css->sw_version); - uc_fw->minor_ver_found = FIELD_GET(CSS_SW_VERSION_UC_MINOR, - css->sw_version); + uc_fw->file_selected.major_ver = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, + css->sw_version); + uc_fw->file_selected.minor_ver = FIELD_GET(CSS_SW_VERSION_UC_MINOR, + css->sw_version); + uc_fw->file_selected.patch_ver = FIELD_GET(CSS_SW_VERSION_UC_PATCH, + css->sw_version); if (uc_fw->type == INTEL_UC_FW_TYPE_GUC) uc_fw->private_data_size = css->private_data_size; @@ -422,9 +546,11 @@ static int check_ccs_header(struct drm_i915_private *i915, int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw) { struct drm_i915_private *i915 = __uc_fw_to_gt(uc_fw)->i915; + struct intel_uc_fw_file file_ideal; struct device *dev = i915->drm.dev; struct drm_i915_gem_object *obj; const struct firmware *fw = NULL; + bool old_ver = false; int err; GEM_BUG_ON(!i915->wopcm.size); @@ -437,24 +563,32 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw) __force_fw_fetch_failures(uc_fw, -EINVAL); __force_fw_fetch_failures(uc_fw, -ESTALE); - err = firmware_request_nowarn(&fw, uc_fw->path, dev); - if (err && !intel_uc_fw_is_overridden(uc_fw) && uc_fw->fallback.path) { - err = firmware_request_nowarn(&fw, uc_fw->fallback.path, dev); - if (!err) { - drm_notice(&i915->drm, - "%s firmware %s is recommended, but only %s was found\n", - intel_uc_fw_type_repr(uc_fw->type), - uc_fw->wanted_path, - uc_fw->fallback.path); - drm_info(&i915->drm, - "Consider updating your linux-firmware pkg or downloading from %s\n", - INTEL_UC_FIRMWARE_URL); - - uc_fw->path = uc_fw->fallback.path; - uc_fw->major_ver_wanted = uc_fw->fallback.major_ver; - uc_fw->minor_ver_wanted = uc_fw->fallback.minor_ver; + err = firmware_request_nowarn(&fw, uc_fw->file_selected.path, dev); + memcpy(&file_ideal, &uc_fw->file_wanted, sizeof(file_ideal)); + + /* Any error is terminal if overriding. Don't bother searching for older versions */ + if (err && intel_uc_fw_is_overridden(uc_fw)) + goto fail; + + while (err == -ENOENT) { + old_ver = true; + + __uc_fw_auto_select(i915, uc_fw); + if (!uc_fw->file_selected.path) { + /* + * No more options! But set the path back to something + * valid just in case it gets dereferenced. + */ + uc_fw->file_selected.path = file_ideal.path; + + /* Also, preserve the version that was really wanted */ + memcpy(&uc_fw->file_wanted, &file_ideal, sizeof(uc_fw->file_wanted)); + break; } + + err = firmware_request_nowarn(&fw, uc_fw->file_selected.path, dev); } + if (err) goto fail; @@ -465,18 +599,39 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw) if (err) goto fail; - if (uc_fw->major_ver_found != uc_fw->major_ver_wanted || - uc_fw->minor_ver_found < uc_fw->minor_ver_wanted) { - drm_notice(&i915->drm, "%s firmware %s: unexpected version: %u.%u != %u.%u\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, - uc_fw->major_ver_found, uc_fw->minor_ver_found, - uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted); - if (!intel_uc_fw_is_overridden(uc_fw)) { - err = -ENOEXEC; - goto fail; + if (uc_fw->file_wanted.major_ver) { + /* Check the file's major version was as it claimed */ + if (uc_fw->file_selected.major_ver != uc_fw->file_wanted.major_ver) { + drm_notice(&i915->drm, "%s firmware %s: unexpected version: %u.%u != %u.%u\n", + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, + uc_fw->file_selected.major_ver, uc_fw->file_selected.minor_ver, + uc_fw->file_wanted.major_ver, uc_fw->file_wanted.minor_ver); + if (!intel_uc_fw_is_overridden(uc_fw)) { + err = -ENOEXEC; + goto fail; + } + } else { + if (uc_fw->file_selected.minor_ver < uc_fw->file_wanted.minor_ver) + old_ver = true; } } + if (old_ver) { + /* Preserve the version that was really wanted */ + memcpy(&uc_fw->file_wanted, &file_ideal, sizeof(uc_fw->file_wanted)); + + drm_notice(&i915->drm, + "%s firmware %s (%d.%d) is recommended, but only %s (%d.%d) was found\n", + intel_uc_fw_type_repr(uc_fw->type), + uc_fw->file_wanted.path, + uc_fw->file_wanted.major_ver, uc_fw->file_wanted.minor_ver, + uc_fw->file_selected.path, + uc_fw->file_selected.major_ver, uc_fw->file_selected.minor_ver); + drm_info(&i915->drm, + "Consider updating your linux-firmware pkg or downloading from %s\n", + INTEL_UC_FIRMWARE_URL); + } + if (HAS_LMEM(i915)) { obj = i915_gem_object_create_lmem_from_data(i915, fw->data, fw->size); if (!IS_ERR(obj)) @@ -503,7 +658,7 @@ fail: INTEL_UC_FIRMWARE_ERROR); i915_probe_error(i915, "%s firmware %s: fetch failed with error %d\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, err); + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, err); drm_info(&i915->drm, "%s firmware(s) can be downloaded from %s\n", intel_uc_fw_type_repr(uc_fw->type), INTEL_UC_FIRMWARE_URL); @@ -645,7 +800,7 @@ int intel_uc_fw_upload(struct intel_uc_fw *uc_fw, u32 dst_offset, u32 dma_flags) fail: i915_probe_error(gt->i915, "Failed to load %s firmware %s (%d)\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, err); intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_LOAD_FAIL); return err; @@ -863,19 +1018,34 @@ size_t intel_uc_fw_copy_rsa(struct intel_uc_fw *uc_fw, void *dst, u32 max_len) */ void intel_uc_fw_dump(const struct intel_uc_fw *uc_fw, struct drm_printer *p) { + u32 ver_sel, ver_want; + drm_printf(p, "%s firmware: %s\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->wanted_path); - if (uc_fw->fallback.path) { - drm_printf(p, "%s firmware fallback: %s\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->fallback.path); - drm_printf(p, "fallback selected: %s\n", - str_yes_no(uc_fw->path == uc_fw->fallback.path)); - } + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path); + if (uc_fw->file_selected.path != uc_fw->file_wanted.path) + drm_printf(p, "%s firmware wanted: %s\n", + intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_wanted.path); drm_printf(p, "\tstatus: %s\n", intel_uc_fw_status_repr(uc_fw->status)); - drm_printf(p, "\tversion: wanted %u.%u, found %u.%u\n", - uc_fw->major_ver_wanted, uc_fw->minor_ver_wanted, - uc_fw->major_ver_found, uc_fw->minor_ver_found); + ver_sel = MAKE_UC_VER(uc_fw->file_selected.major_ver, + uc_fw->file_selected.minor_ver, + uc_fw->file_selected.patch_ver); + ver_want = MAKE_UC_VER(uc_fw->file_wanted.major_ver, + uc_fw->file_wanted.minor_ver, + uc_fw->file_wanted.patch_ver); + if (ver_sel < ver_want) + drm_printf(p, "\tversion: wanted %u.%u.%u, found %u.%u.%u\n", + uc_fw->file_wanted.major_ver, + uc_fw->file_wanted.minor_ver, + uc_fw->file_wanted.patch_ver, + uc_fw->file_selected.major_ver, + uc_fw->file_selected.minor_ver, + uc_fw->file_selected.patch_ver); + else + drm_printf(p, "\tversion: found %u.%u.%u\n", + uc_fw->file_selected.major_ver, + uc_fw->file_selected.minor_ver, + uc_fw->file_selected.patch_ver); drm_printf(p, "\tuCode: %u bytes\n", uc_fw->ucode_size); drm_printf(p, "\tRSA: %u bytes\n", uc_fw->rsa_size); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h index 7aa2644400b9..cb586f7df270 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h @@ -65,6 +65,18 @@ enum intel_uc_fw_type { #define INTEL_UC_FW_NUM_TYPES 2 /* + * The firmware build process will generate a version header file with major and + * minor version defined. The versions are built into CSS header of firmware. + * i915 kernel driver set the minimal firmware version required per platform. + */ +struct intel_uc_fw_file { + const char *path; + u16 major_ver; + u16 minor_ver; + u16 patch_ver; +}; + +/* * This structure encapsulates all the data needed during the process * of fetching, caching, and loading the firmware image into the uC. */ @@ -74,11 +86,12 @@ struct intel_uc_fw { const enum intel_uc_fw_status status; enum intel_uc_fw_status __status; /* no accidental overwrites */ }; - const char *wanted_path; - const char *path; + struct intel_uc_fw_file file_wanted; + struct intel_uc_fw_file file_selected; bool user_overridden; size_t size; struct drm_i915_gem_object *obj; + /** * @dummy: A vma used in binding the uc fw to ggtt. We can't define this * vma on the stack as it can lead to a stack overflow, so we define it @@ -89,30 +102,18 @@ struct intel_uc_fw { struct i915_vma_resource dummy; struct i915_vma *rsa_data; - /* - * The firmware build process will generate a version header file with major and - * minor version defined. The versions are built into CSS header of firmware. - * i915 kernel driver set the minimal firmware version required per platform. - */ - u16 major_ver_wanted; - u16 minor_ver_wanted; - u16 major_ver_found; - u16 minor_ver_found; - - struct { - const char *path; - u16 major_ver; - u16 minor_ver; - } fallback; - u32 rsa_size; u32 ucode_size; - u32 private_data_size; bool loaded_via_gsc; }; +#define MAKE_UC_VER(maj, min, pat) ((pat) | ((min) << 8) | ((maj) << 16)) +#define GET_UC_VER(uc) (MAKE_UC_VER((uc)->fw.file_selected.major_ver, \ + (uc)->fw.file_selected.minor_ver, \ + (uc)->fw.file_selected.patch_ver)) + #ifdef CONFIG_DRM_I915_DEBUG_GUC void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, enum intel_uc_fw_status status); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h index b05e0e35b734..7a411178bdbf 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw_abi.h @@ -83,8 +83,10 @@ struct uc_css_header { } __packed; static_assert(sizeof(struct uc_css_header) == 128); -#define HUC_GSC_VERSION_DW 44 -#define HUC_GSC_MAJOR_VER_MASK (0xFF << 0) -#define HUC_GSC_MINOR_VER_MASK (0xFF << 16) +#define HUC_GSC_VERSION_HI_DW 44 +#define HUC_GSC_MAJOR_VER_HI_MASK (0xFF << 0) +#define HUC_GSC_MINOR_VER_HI_MASK (0xFF << 16) +#define HUC_GSC_VERSION_LO_DW 45 +#define HUC_GSC_PATCH_VER_LO_MASK (0xFF << 0) #endif /* _INTEL_UC_FW_ABI_H */ diff --git a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c index 20e0c39259fb..e28518fe8b90 100644 --- a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c @@ -54,6 +54,9 @@ static int intel_guc_scrub_ctbs(void *arg) struct intel_engine_cs *engine; struct intel_context *ce; + if (!intel_has_gpu_reset(gt)) + return 0; + wakeref = intel_runtime_pm_get(gt->uncore->rpm); engine = intel_selftest_find_any_engine(gt); diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 3e99a6a88c3a..ae987e92251d 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -187,47 +187,47 @@ i915_debugfs_describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj) stringify_page_sizes(vma->resource->page_sizes_gtt, NULL, 0)); if (i915_vma_is_ggtt(vma) || i915_vma_is_dpt(vma)) { - switch (vma->ggtt_view.type) { - case I915_GGTT_VIEW_NORMAL: + switch (vma->gtt_view.type) { + case I915_GTT_VIEW_NORMAL: seq_puts(m, ", normal"); break; - case I915_GGTT_VIEW_PARTIAL: + case I915_GTT_VIEW_PARTIAL: seq_printf(m, ", partial [%08llx+%x]", - vma->ggtt_view.partial.offset << PAGE_SHIFT, - vma->ggtt_view.partial.size << PAGE_SHIFT); + vma->gtt_view.partial.offset << PAGE_SHIFT, + vma->gtt_view.partial.size << PAGE_SHIFT); break; - case I915_GGTT_VIEW_ROTATED: + case I915_GTT_VIEW_ROTATED: seq_printf(m, ", rotated [(%ux%u, src_stride=%u, dst_stride=%u, offset=%u), (%ux%u, src_stride=%u, dst_stride=%u, offset=%u)]", - vma->ggtt_view.rotated.plane[0].width, - vma->ggtt_view.rotated.plane[0].height, - vma->ggtt_view.rotated.plane[0].src_stride, - vma->ggtt_view.rotated.plane[0].dst_stride, - vma->ggtt_view.rotated.plane[0].offset, - vma->ggtt_view.rotated.plane[1].width, - vma->ggtt_view.rotated.plane[1].height, - vma->ggtt_view.rotated.plane[1].src_stride, - vma->ggtt_view.rotated.plane[1].dst_stride, - vma->ggtt_view.rotated.plane[1].offset); + vma->gtt_view.rotated.plane[0].width, + vma->gtt_view.rotated.plane[0].height, + vma->gtt_view.rotated.plane[0].src_stride, + vma->gtt_view.rotated.plane[0].dst_stride, + vma->gtt_view.rotated.plane[0].offset, + vma->gtt_view.rotated.plane[1].width, + vma->gtt_view.rotated.plane[1].height, + vma->gtt_view.rotated.plane[1].src_stride, + vma->gtt_view.rotated.plane[1].dst_stride, + vma->gtt_view.rotated.plane[1].offset); break; - case I915_GGTT_VIEW_REMAPPED: + case I915_GTT_VIEW_REMAPPED: seq_printf(m, ", remapped [(%ux%u, src_stride=%u, dst_stride=%u, offset=%u), (%ux%u, src_stride=%u, dst_stride=%u, offset=%u)]", - vma->ggtt_view.remapped.plane[0].width, - vma->ggtt_view.remapped.plane[0].height, - vma->ggtt_view.remapped.plane[0].src_stride, - vma->ggtt_view.remapped.plane[0].dst_stride, - vma->ggtt_view.remapped.plane[0].offset, - vma->ggtt_view.remapped.plane[1].width, - vma->ggtt_view.remapped.plane[1].height, - vma->ggtt_view.remapped.plane[1].src_stride, - vma->ggtt_view.remapped.plane[1].dst_stride, - vma->ggtt_view.remapped.plane[1].offset); + vma->gtt_view.remapped.plane[0].width, + vma->gtt_view.remapped.plane[0].height, + vma->gtt_view.remapped.plane[0].src_stride, + vma->gtt_view.remapped.plane[0].dst_stride, + vma->gtt_view.remapped.plane[0].offset, + vma->gtt_view.remapped.plane[1].width, + vma->gtt_view.remapped.plane[1].height, + vma->gtt_view.remapped.plane[1].src_stride, + vma->gtt_view.remapped.plane[1].dst_stride, + vma->gtt_view.remapped.plane[1].offset); break; default: - MISSING_CASE(vma->ggtt_view.type); + MISSING_CASE(vma->gtt_view.type); break; } } diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index 3e1d2a10e9da..c459eb362c47 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -106,6 +106,12 @@ static const char irst_name[] = "INT3392"; static const struct drm_driver i915_drm_driver; +static void i915_release_bridge_dev(struct drm_device *dev, + void *bridge) +{ + pci_dev_put(bridge); +} + static int i915_get_bridge_dev(struct drm_i915_private *dev_priv) { int domain = pci_domain_nr(to_pci_dev(dev_priv->drm.dev)->bus); @@ -116,7 +122,9 @@ static int i915_get_bridge_dev(struct drm_i915_private *dev_priv) drm_err(&dev_priv->drm, "bridge device not found\n"); return -EIO; } - return 0; + + return drmm_add_action_or_reset(&dev_priv->drm, i915_release_bridge_dev, + dev_priv->bridge_dev); } /* Allocate space for the MCH regs if needed, return nonzero on error */ @@ -303,8 +311,13 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv) static void sanitize_gpu(struct drm_i915_private *i915) { - if (!INTEL_INFO(i915)->gpu_reset_clobbers_display) - __intel_gt_reset(to_gt(i915), ALL_ENGINES); + if (!INTEL_INFO(i915)->gpu_reset_clobbers_display) { + struct intel_gt *gt; + unsigned int i; + + for_each_gt(gt, i915, i) + __intel_gt_reset(gt, ALL_ENGINES); + } } /** @@ -327,7 +340,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) intel_device_info_subplatform_init(dev_priv); intel_step_init(dev_priv); - intel_uncore_mmio_debug_init_early(&dev_priv->mmio_debug); + intel_uncore_mmio_debug_init_early(dev_priv); spin_lock_init(&dev_priv->irq_lock); spin_lock_init(&dev_priv->gpu_error.lock); @@ -358,7 +371,9 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) intel_wopcm_init_early(&dev_priv->wopcm); - intel_root_gt_init_early(dev_priv); + ret = intel_root_gt_init_early(dev_priv); + if (ret < 0) + goto err_rootgt; i915_drm_clients_init(&dev_priv->clients, dev_priv); @@ -383,6 +398,7 @@ err_gem: i915_gem_cleanup_early(dev_priv); intel_gt_driver_late_release_all(dev_priv); i915_drm_clients_fini(&dev_priv->clients); +err_rootgt: intel_region_ttm_device_fini(dev_priv); err_ttm: vlv_suspend_cleanup(dev_priv); @@ -424,7 +440,8 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv) */ static int i915_driver_mmio_probe(struct drm_i915_private *dev_priv) { - int ret; + struct intel_gt *gt; + int ret, i; if (i915_inject_probe_failure(dev_priv)) return -ENODEV; @@ -433,17 +450,27 @@ static int i915_driver_mmio_probe(struct drm_i915_private *dev_priv) if (ret < 0) return ret; - ret = intel_uncore_init_mmio(&dev_priv->uncore); - if (ret) - return ret; + for_each_gt(gt, dev_priv, i) { + ret = intel_uncore_init_mmio(gt->uncore); + if (ret) + return ret; + + ret = drmm_add_action_or_reset(&dev_priv->drm, + intel_uncore_fini_mmio, + gt->uncore); + if (ret) + return ret; + } /* Try to make sure MCHBAR is enabled before poking at it */ intel_setup_mchbar(dev_priv); intel_device_info_runtime_init(dev_priv); - ret = intel_gt_init_mmio(to_gt(dev_priv)); - if (ret) - goto err_uncore; + for_each_gt(gt, dev_priv, i) { + ret = intel_gt_init_mmio(gt); + if (ret) + goto err_uncore; + } /* As early as possible, scrub existing GPU state before clobbering */ sanitize_gpu(dev_priv); @@ -452,8 +479,6 @@ static int i915_driver_mmio_probe(struct drm_i915_private *dev_priv) err_uncore: intel_teardown_mchbar(dev_priv); - intel_uncore_fini_mmio(&dev_priv->uncore); - pci_dev_put(dev_priv->bridge_dev); return ret; } @@ -465,8 +490,6 @@ err_uncore: static void i915_driver_mmio_release(struct drm_i915_private *dev_priv) { intel_teardown_mchbar(dev_priv); - intel_uncore_fini_mmio(&dev_priv->uncore); - pci_dev_put(dev_priv->bridge_dev); } /** @@ -716,6 +739,8 @@ static void i915_driver_hw_remove(struct drm_i915_private *dev_priv) static void i915_driver_register(struct drm_i915_private *dev_priv) { struct drm_device *dev = &dev_priv->drm; + struct intel_gt *gt; + unsigned int i; i915_gem_driver_register(dev_priv); i915_pmu_register(dev_priv); @@ -735,7 +760,8 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) /* Depends on sysfs having been initialized */ i915_perf_register(dev_priv); - intel_gt_driver_register(to_gt(dev_priv)); + for_each_gt(gt, dev_priv, i) + intel_gt_driver_register(gt); intel_display_driver_register(dev_priv); @@ -754,6 +780,9 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) */ static void i915_driver_unregister(struct drm_i915_private *dev_priv) { + struct intel_gt *gt; + unsigned int i; + i915_switcheroo_unregister(dev_priv); intel_unregister_dsm_handler(); @@ -763,7 +792,8 @@ static void i915_driver_unregister(struct drm_i915_private *dev_priv) intel_display_driver_unregister(dev_priv); - intel_gt_driver_unregister(to_gt(dev_priv)); + for_each_gt(gt, dev_priv, i) + intel_gt_driver_unregister(gt); i915_perf_unregister(dev_priv); i915_pmu_unregister(dev_priv); @@ -785,6 +815,8 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv) { if (drm_debug_enabled(DRM_UT_DRIVER)) { struct drm_printer p = drm_debug_printer("i915 device info:"); + struct intel_gt *gt; + unsigned int i; drm_printf(&p, "pciid=0x%04x rev=0x%02x platform=%s (subplatform=0x%x) gen=%i\n", INTEL_DEVID(dev_priv), @@ -797,7 +829,8 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv) intel_device_info_print(INTEL_INFO(dev_priv), RUNTIME_INFO(dev_priv), &p); i915_print_iommu_status(dev_priv, &p); - intel_gt_info_print(&to_gt(dev_priv)->info, &p); + for_each_gt(gt, dev_priv, i) + intel_gt_info_print(>->info, &p); } if (IS_ENABLED(CONFIG_DRM_I915_DEBUG)) @@ -1216,13 +1249,15 @@ static int i915_drm_suspend_late(struct drm_device *dev, bool hibernation) struct drm_i915_private *dev_priv = to_i915(dev); struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; - int ret; + struct intel_gt *gt; + int ret, i; disable_rpm_wakeref_asserts(rpm); i915_gem_suspend_late(dev_priv); - intel_uncore_suspend(&dev_priv->uncore); + for_each_gt(gt, dev_priv, i) + intel_uncore_suspend(gt->uncore); intel_power_domains_suspend(dev_priv, get_suspend_mode(dev_priv, hibernation)); @@ -1354,7 +1389,8 @@ static int i915_drm_resume_early(struct drm_device *dev) { struct drm_i915_private *dev_priv = to_i915(dev); struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); - int ret; + struct intel_gt *gt; + int ret, i; /* * We have a resume ordering issue with the snd-hda driver also @@ -1408,9 +1444,10 @@ static int i915_drm_resume_early(struct drm_device *dev) drm_err(&dev_priv->drm, "Resume prepare failed: %d, continuing anyway\n", ret); - intel_uncore_resume_early(&dev_priv->uncore); - - intel_gt_check_and_clear_faults(to_gt(dev_priv)); + for_each_gt(gt, dev_priv, i) { + intel_uncore_resume_early(gt->uncore); + intel_gt_check_and_clear_faults(gt); + } intel_display_power_resume_early(dev_priv); @@ -1590,7 +1627,8 @@ static int intel_runtime_suspend(struct device *kdev) { struct drm_i915_private *dev_priv = kdev_to_i915(kdev); struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; - int ret; + struct intel_gt *gt; + int ret, i; if (drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_RUNTIME_PM(dev_priv))) return -ENODEV; @@ -1605,11 +1643,13 @@ static int intel_runtime_suspend(struct device *kdev) */ i915_gem_runtime_suspend(dev_priv); - intel_gt_runtime_suspend(to_gt(dev_priv)); + for_each_gt(gt, dev_priv, i) + intel_gt_runtime_suspend(gt); intel_runtime_pm_disable_interrupts(dev_priv); - intel_uncore_suspend(&dev_priv->uncore); + for_each_gt(gt, dev_priv, i) + intel_uncore_suspend(gt->uncore); intel_display_power_suspend(dev_priv); @@ -1673,7 +1713,8 @@ static int intel_runtime_resume(struct device *kdev) { struct drm_i915_private *dev_priv = kdev_to_i915(kdev); struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; - int ret; + struct intel_gt *gt; + int ret, i; if (drm_WARN_ON_ONCE(&dev_priv->drm, !HAS_RUNTIME_PM(dev_priv))) return -ENODEV; @@ -1693,7 +1734,8 @@ static int intel_runtime_resume(struct device *kdev) ret = vlv_resume_prepare(dev_priv, true); - intel_uncore_runtime_resume(&dev_priv->uncore); + for_each_gt(gt, dev_priv, i) + intel_uncore_runtime_resume(gt->uncore); intel_runtime_pm_enable_interrupts(dev_priv); @@ -1701,7 +1743,8 @@ static int intel_runtime_resume(struct device *kdev) * No point of rolling back things in case of an error, as the best * we can do is to hope that things will still work (and disable RPM). */ - intel_gt_runtime_resume(to_gt(dev_priv)); + for_each_gt(gt, dev_priv, i) + intel_gt_runtime_resume(gt); /* * On VLV/CHV display interrupts are part of the display diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 48f1ce9362a5..9f9372931fd2 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -364,6 +364,9 @@ struct drm_i915_private { struct kobject *sysfs_gt; + /* Quick lookup of media GT (current platforms only have one) */ + struct intel_gt *media_gt; + struct { struct i915_gem_contexts { spinlock_t lock; /* locks list */ @@ -917,6 +920,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i)) #define HAS_LMEM(i915) HAS_REGION(i915, REGION_LMEM) +#define HAS_EXTRA_GT_LIST(dev_priv) (INTEL_INFO(dev_priv)->extra_gt_list) + /* * Platform has the dedicated compression control state for each lmem surfaces * stored in lmem to support the 3D and media compression formats. @@ -964,9 +969,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_GUC_DEPRIVILEGE(dev_priv) \ (INTEL_INFO(dev_priv)->has_guc_deprivilege) -#define HAS_PERCTX_PREEMPT_CTRL(i915) \ - ((GRAPHICS_VER(i915) >= 9) && GRAPHICS_VER_FULL(i915) < IP_VER(12, 55)) - #define HAS_D12_PLANE_MINIMIZATION(dev_priv) (IS_ROCKETLAKE(dev_priv) || \ IS_ALDERLAKE_S(dev_priv)) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 0f49ec9d494a..f18cc6270b2b 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -842,6 +842,10 @@ void i915_gem_runtime_suspend(struct drm_i915_private *i915) &to_gt(i915)->ggtt->userfault_list, userfault_link) __i915_gem_object_release_mmap_gtt(obj); + list_for_each_entry_safe(obj, on, + &to_gt(i915)->lmem_userfault_list, userfault_link) + i915_gem_object_runtime_pm_release_mmap_offset(obj); + /* * The fence will be lost when the device powers down. If any were * in use by hardware (i.e. they are pinned), we should not be powering @@ -885,7 +889,7 @@ static void discard_ggtt_vma(struct i915_vma *vma) struct i915_vma * i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj, struct i915_gem_ww_ctx *ww, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, u64 size, u64 alignment, u64 flags) { struct drm_i915_private *i915 = to_i915(obj->base.dev); @@ -896,7 +900,7 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj, GEM_WARN_ON(!ww); if (flags & PIN_MAPPABLE && - (!view || view->type == I915_GGTT_VIEW_NORMAL)) { + (!view || view->type == I915_GTT_VIEW_NORMAL)) { /* * If the required space is larger than the available * aperture, we will not able to find a slot for the @@ -987,7 +991,7 @@ new_vma: struct i915_vma * __must_check i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, u64 size, u64 alignment, u64 flags) { struct i915_gem_ww_ctx ww; @@ -1209,7 +1213,7 @@ void i915_gem_driver_unregister(struct drm_i915_private *i915) void i915_gem_driver_remove(struct drm_i915_private *dev_priv) { - intel_wakeref_auto_fini(&to_gt(dev_priv)->ggtt->userfault_wakeref); + intel_wakeref_auto_fini(&to_gt(dev_priv)->userfault_wakeref); i915_gem_suspend_late(dev_priv); intel_gt_driver_remove(to_gt(dev_priv)); diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h index 2fccb19ed9f7..a5cdf6662d01 100644 --- a/drivers/gpu/drm/i915/i915_gem.h +++ b/drivers/gpu/drm/i915/i915_gem.h @@ -36,7 +36,7 @@ struct drm_file; struct drm_i915_gem_object; struct drm_i915_private; struct i915_gem_ww_ctx; -struct i915_ggtt_view; +struct i915_gtt_view; struct i915_vma; void i915_gem_init_early(struct drm_i915_private *i915); @@ -48,12 +48,12 @@ void i915_gem_drain_workqueue(struct drm_i915_private *i915); struct i915_vma * __must_check i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj, struct i915_gem_ww_ctx *ww, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, u64 size, u64 alignment, u64 flags); struct i915_vma * __must_check i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, - const struct i915_ggtt_view *view, + const struct i915_gtt_view *view, u64 size, u64 alignment, u64 flags); int i915_gem_object_unbind(struct drm_i915_gem_object *obj, diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index b5fbc2252784..9ea2fe34e7d3 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -1022,8 +1022,10 @@ static void cleanup_params(struct i915_gpu_coredump *error) static void cleanup_uc(struct intel_uc_coredump *uc) { - kfree(uc->guc_fw.path); - kfree(uc->huc_fw.path); + kfree(uc->guc_fw.file_selected.path); + kfree(uc->huc_fw.file_selected.path); + kfree(uc->guc_fw.file_wanted.path); + kfree(uc->huc_fw.file_wanted.path); i915_vma_coredump_free(uc->guc.vma_log); i915_vma_coredump_free(uc->guc.vma_ctb); @@ -1705,12 +1707,10 @@ gt_record_uc(struct intel_gt_coredump *gt, memcpy(&error_uc->guc_fw, &uc->guc.fw, sizeof(uc->guc.fw)); memcpy(&error_uc->huc_fw, &uc->huc.fw, sizeof(uc->huc.fw)); - /* Non-default firmware paths will be specified by the modparam. - * As modparams are generally accesible from the userspace make - * explicit copies of the firmware paths. - */ - error_uc->guc_fw.path = kstrdup(uc->guc.fw.path, ALLOW_FAIL); - error_uc->huc_fw.path = kstrdup(uc->huc.fw.path, ALLOW_FAIL); + error_uc->guc_fw.file_selected.path = kstrdup(uc->guc.fw.file_selected.path, ALLOW_FAIL); + error_uc->huc_fw.file_selected.path = kstrdup(uc->huc.fw.file_selected.path, ALLOW_FAIL); + error_uc->guc_fw.file_wanted.path = kstrdup(uc->guc.fw.file_wanted.path, ALLOW_FAIL); + error_uc->huc_fw.file_wanted.path = kstrdup(uc->huc.fw.file_wanted.path, ALLOW_FAIL); /* * Save the GuC log and include a timestamp reference for converting the diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 515648cd1233..86a42d9e8041 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1104,9 +1104,9 @@ static void ivb_parity_work(struct work_struct *work) out: drm_WARN_ON(&dev_priv->drm, dev_priv->l3_parity.which_slice); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen5_gt_enable_irq(gt, GT_PARITY_ERROR(dev_priv)); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); mutex_unlock(&dev_priv->drm.struct_mutex); } diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 1576cbcf2254..77e7df21f539 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -26,6 +26,9 @@ #include <drm/drm_drv.h> #include <drm/i915_pciids.h> +#include "gt/intel_gt_regs.h" +#include "gt/intel_sa_media.h" + #include "i915_driver.h" #include "i915_drv.h" #include "i915_pci.h" @@ -1080,6 +1083,7 @@ static const struct intel_device_info ats_m_info = { DG2_FEATURES, .display = { 0 }, .require_force_probe = 1, + .tuning_thread_rr_after_dep = 1, }; #define XE_HPC_FEATURES \ @@ -1114,6 +1118,16 @@ static const struct intel_device_info pvc_info = { .display.has_cdclk_crawl = 1, \ .__runtime.fbc_mask = BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B) +static const struct intel_gt_definition xelpmp_extra_gt[] = { + { + .type = GT_MEDIA, + .name = "Standalone Media GT", + .gsi_offset = MTL_MEDIA_GSI_BASE, + .engine_mask = BIT(VECS0) | BIT(VCS0) | BIT(VCS2), + }, + {} +}; + __maybe_unused static const struct intel_device_info mtl_info = { XE_HP_FEATURES, @@ -1127,6 +1141,7 @@ static const struct intel_device_info mtl_info = { .__runtime.media.ip.ver = 13, PLATFORM(INTEL_METEORLAKE), .display.has_modular_fia = 1, + .extra_gt_list = xelpmp_extra_gt, .has_flat_ccs = 0, .has_snoop = 1, .__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM, diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 3053c175cf9b..1a9bd829fc7e 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -1798,14 +1798,14 @@ #define GT0_PERF_LIMIT_REASONS _MMIO(0x1381a8) #define GT0_PERF_LIMIT_REASONS_MASK 0xde3 -#define PROCHOT_MASK REG_BIT(1) -#define THERMAL_LIMIT_MASK REG_BIT(2) -#define RATL_MASK REG_BIT(6) -#define VR_THERMALERT_MASK REG_BIT(7) -#define VR_TDC_MASK REG_BIT(8) -#define POWER_LIMIT_4_MASK REG_BIT(9) -#define POWER_LIMIT_1_MASK REG_BIT(11) -#define POWER_LIMIT_2_MASK REG_BIT(12) +#define PROCHOT_MASK REG_BIT(0) +#define THERMAL_LIMIT_MASK REG_BIT(1) +#define RATL_MASK REG_BIT(5) +#define VR_THERMALERT_MASK REG_BIT(6) +#define VR_TDC_MASK REG_BIT(7) +#define POWER_LIMIT_4_MASK REG_BIT(8) +#define POWER_LIMIT_1_MASK REG_BIT(10) +#define POWER_LIMIT_2_MASK REG_BIT(11) #define CHV_CLK_CTL1 _MMIO(0x101100) #define VLV_CLK_CTL2 _MMIO(0x101104) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index 260371716490..f17c09ead7d7 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -109,7 +109,7 @@ static void __i915_vma_retire(struct i915_active *ref) static struct i915_vma * vma_create(struct drm_i915_gem_object *obj, struct i915_address_space *vm, - const struct i915_ggtt_view *view) + const struct i915_gtt_view *view) { struct i915_vma *pos = ERR_PTR(-E2BIG); struct i915_vma *vma; @@ -141,9 +141,9 @@ vma_create(struct drm_i915_gem_object *obj, INIT_LIST_HEAD(&vma->obj_link); RB_CLEAR_NODE(&vma->obj_node); - if (view && view->type != I915_GGTT_VIEW_NORMAL) { - vma->ggtt_view = *view; - if (view->type == I915_GGTT_VIEW_PARTIAL) { + if (view && view->type != I915_GTT_VIEW_NORMAL) { + vma->gtt_view = *view; + if (view->type == I915_GTT_VIEW_PARTIAL) { GEM_BUG_ON(range_overflows_t(u64, view->partial.offset, view->partial.size, @@ -151,10 +151,10 @@ vma_create(struct drm_i915_gem_object *obj, vma->size = view->partial.size; vma->size <<= PAGE_SHIFT; GEM_BUG_ON(vma->size > obj->base.size); - } else if (view->type == I915_GGTT_VIEW_ROTATED) { + } else if (view->type == I915_GTT_VIEW_ROTATED) { vma->size = intel_rotation_info_size(&view->rotated); vma->size <<= PAGE_SHIFT; - } else if (view->type == I915_GGTT_VIEW_REMAPPED) { + } else if (view->type == I915_GTT_VIEW_REMAPPED) { vma->size = intel_remapped_info_size(&view->remapped); vma->size <<= PAGE_SHIFT; } @@ -248,7 +248,7 @@ err_vma: static struct i915_vma * i915_vma_lookup(struct drm_i915_gem_object *obj, struct i915_address_space *vm, - const struct i915_ggtt_view *view) + const struct i915_gtt_view *view) { struct rb_node *rb; @@ -286,7 +286,7 @@ i915_vma_lookup(struct drm_i915_gem_object *obj, struct i915_vma * i915_vma_instance(struct drm_i915_gem_object *obj, struct i915_address_space *vm, - const struct i915_ggtt_view *view) + const struct i915_gtt_view *view) { struct i915_vma *vma; @@ -1203,7 +1203,7 @@ err_st_alloc: } static noinline struct sg_table * -intel_partial_pages(const struct i915_ggtt_view *view, +intel_partial_pages(const struct i915_gtt_view *view, struct drm_i915_gem_object *obj) { struct sg_table *st; @@ -1247,33 +1247,33 @@ __i915_vma_get_pages(struct i915_vma *vma) */ GEM_BUG_ON(!i915_gem_object_has_pinned_pages(vma->obj)); - switch (vma->ggtt_view.type) { + switch (vma->gtt_view.type) { default: - GEM_BUG_ON(vma->ggtt_view.type); + GEM_BUG_ON(vma->gtt_view.type); fallthrough; - case I915_GGTT_VIEW_NORMAL: + case I915_GTT_VIEW_NORMAL: pages = vma->obj->mm.pages; break; - case I915_GGTT_VIEW_ROTATED: + case I915_GTT_VIEW_ROTATED: pages = - intel_rotate_pages(&vma->ggtt_view.rotated, vma->obj); + intel_rotate_pages(&vma->gtt_view.rotated, vma->obj); break; - case I915_GGTT_VIEW_REMAPPED: + case I915_GTT_VIEW_REMAPPED: pages = - intel_remap_pages(&vma->ggtt_view.remapped, vma->obj); + intel_remap_pages(&vma->gtt_view.remapped, vma->obj); break; - case I915_GGTT_VIEW_PARTIAL: - pages = intel_partial_pages(&vma->ggtt_view, vma->obj); + case I915_GTT_VIEW_PARTIAL: + pages = intel_partial_pages(&vma->gtt_view, vma->obj); break; } if (IS_ERR(pages)) { drm_err(&vma->vm->i915->drm, "Failed to get pages for VMA view type %u (%ld)!\n", - vma->ggtt_view.type, PTR_ERR(pages)); + vma->gtt_view.type, PTR_ERR(pages)); return PTR_ERR(pages); } @@ -1806,7 +1806,7 @@ void i915_vma_revoke_mmap(struct i915_vma *vma) GEM_BUG_ON(!vma->obj->userfault_count); node = &vma->mmo->vma_node; - vma_offset = vma->ggtt_view.partial.offset << PAGE_SHIFT; + vma_offset = vma->gtt_view.partial.offset << PAGE_SHIFT; unmap_mapping_range(vma->vm->i915->drm.anon_inode->i_mapping, drm_vma_node_offset_addr(node) + vma_offset, vma->size, @@ -1882,12 +1882,13 @@ int _i915_vma_move_to_active(struct i915_vma *vma, enum dma_resv_usage usage; int idx; - obj->read_domains = 0; if (flags & EXEC_OBJECT_WRITE) { usage = DMA_RESV_USAGE_WRITE; obj->write_domain = I915_GEM_DOMAIN_RENDER; + obj->read_domains = 0; } else { usage = DMA_RESV_USAGE_READ; + obj->write_domain = 0; } dma_fence_array_for_each(curr, idx, fence) diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 33a58f605d75..aecd9c64486b 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -43,7 +43,7 @@ struct i915_vma * i915_vma_instance(struct drm_i915_gem_object *obj, struct i915_address_space *vm, - const struct i915_ggtt_view *view); + const struct i915_gtt_view *view); void i915_vma_unpin_and_release(struct i915_vma **p_vma, unsigned int flags); #define I915_VMA_RELEASE_MAP BIT(0) @@ -160,7 +160,7 @@ static inline void i915_vma_put(struct i915_vma *vma) static inline long i915_vma_compare(struct i915_vma *vma, struct i915_address_space *vm, - const struct i915_ggtt_view *view) + const struct i915_gtt_view *view) { ptrdiff_t cmp; @@ -170,8 +170,8 @@ i915_vma_compare(struct i915_vma *vma, if (cmp) return cmp; - BUILD_BUG_ON(I915_GGTT_VIEW_NORMAL != 0); - cmp = vma->ggtt_view.type; + BUILD_BUG_ON(I915_GTT_VIEW_NORMAL != 0); + cmp = vma->gtt_view.type; if (!view) return cmp; @@ -181,7 +181,7 @@ i915_vma_compare(struct i915_vma *vma, assert_i915_gem_gtt_types(); - /* ggtt_view.type also encodes its size so that we both distinguish + /* gtt_view.type also encodes its size so that we both distinguish * different views using it as a "type" and also use a compact (no * accessing of uninitialised padding bytes) memcmp without storing * an extra parameter or adding more code. @@ -191,14 +191,14 @@ i915_vma_compare(struct i915_vma *vma, * we assert above that all branches have the same address, and that * each branch has a unique type/size. */ - BUILD_BUG_ON(I915_GGTT_VIEW_NORMAL >= I915_GGTT_VIEW_PARTIAL); - BUILD_BUG_ON(I915_GGTT_VIEW_PARTIAL >= I915_GGTT_VIEW_ROTATED); - BUILD_BUG_ON(I915_GGTT_VIEW_ROTATED >= I915_GGTT_VIEW_REMAPPED); + BUILD_BUG_ON(I915_GTT_VIEW_NORMAL >= I915_GTT_VIEW_PARTIAL); + BUILD_BUG_ON(I915_GTT_VIEW_PARTIAL >= I915_GTT_VIEW_ROTATED); + BUILD_BUG_ON(I915_GTT_VIEW_ROTATED >= I915_GTT_VIEW_REMAPPED); BUILD_BUG_ON(offsetof(typeof(*view), rotated) != offsetof(typeof(*view), partial)); BUILD_BUG_ON(offsetof(typeof(*view), rotated) != offsetof(typeof(*view), remapped)); - return memcmp(&vma->ggtt_view.partial, &view->partial, view->type); + return memcmp(&vma->gtt_view.partial, &view->partial, view->type); } struct i915_vma_work *i915_vma_work(void); diff --git a/drivers/gpu/drm/i915/i915_vma_types.h b/drivers/gpu/drm/i915/i915_vma_types.h index be6e028c3b57..ec0f6c9f57d0 100644 --- a/drivers/gpu/drm/i915/i915_vma_types.h +++ b/drivers/gpu/drm/i915/i915_vma_types.h @@ -67,30 +67,30 @@ enum i915_cache_level; * Implementation and usage * * GGTT views are implemented using VMAs and are distinguished via enum - * i915_ggtt_view_type and struct i915_ggtt_view. + * i915_gtt_view_type and struct i915_gtt_view. * * A new flavour of core GEM functions which work with GGTT bound objects were * added with the _ggtt_ infix, and sometimes with _view postfix to avoid - * renaming in large amounts of code. They take the struct i915_ggtt_view + * renaming in large amounts of code. They take the struct i915_gtt_view * parameter encapsulating all metadata required to implement a view. * * As a helper for callers which are only interested in the normal view, - * globally const i915_ggtt_view_normal singleton instance exists. All old core + * globally const i915_gtt_view_normal singleton instance exists. All old core * GEM API functions, the ones not taking the view parameter, are operating on, * or with the normal GGTT view. * * Code wanting to add or use a new GGTT view needs to: * * 1. Add a new enum with a suitable name. - * 2. Extend the metadata in the i915_ggtt_view structure if required. + * 2. Extend the metadata in the i915_gtt_view structure if required. * 3. Add support to i915_get_vma_pages(). * * New views are required to build a scatter-gather table from within the - * i915_get_vma_pages function. This table is stored in the vma.ggtt_view and + * i915_get_vma_pages function. This table is stored in the vma.gtt_view and * exists for the lifetime of an VMA. * * Core API is designed to have copy semantics which means that passed in - * struct i915_ggtt_view does not need to be persistent (left around after + * struct i915_gtt_view does not need to be persistent (left around after * calling the core API functions). * */ @@ -130,11 +130,11 @@ struct intel_partial_info { unsigned int size; } __packed; -enum i915_ggtt_view_type { - I915_GGTT_VIEW_NORMAL = 0, - I915_GGTT_VIEW_ROTATED = sizeof(struct intel_rotation_info), - I915_GGTT_VIEW_PARTIAL = sizeof(struct intel_partial_info), - I915_GGTT_VIEW_REMAPPED = sizeof(struct intel_remapped_info), +enum i915_gtt_view_type { + I915_GTT_VIEW_NORMAL = 0, + I915_GTT_VIEW_ROTATED = sizeof(struct intel_rotation_info), + I915_GTT_VIEW_PARTIAL = sizeof(struct intel_partial_info), + I915_GTT_VIEW_REMAPPED = sizeof(struct intel_remapped_info), }; static inline void assert_i915_gem_gtt_types(void) @@ -152,18 +152,18 @@ static inline void assert_i915_gem_gtt_types(void) /* As we encode the size of each branch inside the union into its type, * we have to be careful that each branch has a unique size. */ - switch ((enum i915_ggtt_view_type)0) { - case I915_GGTT_VIEW_NORMAL: - case I915_GGTT_VIEW_PARTIAL: - case I915_GGTT_VIEW_ROTATED: - case I915_GGTT_VIEW_REMAPPED: + switch ((enum i915_gtt_view_type)0) { + case I915_GTT_VIEW_NORMAL: + case I915_GTT_VIEW_PARTIAL: + case I915_GTT_VIEW_ROTATED: + case I915_GTT_VIEW_REMAPPED: /* gcc complains if these are identical cases */ break; } } -struct i915_ggtt_view { - enum i915_ggtt_view_type type; +struct i915_gtt_view { + enum i915_gtt_view_type type; union { /* Members need to contain no holes/padding */ struct intel_partial_info partial; @@ -280,11 +280,11 @@ struct i915_vma { /** * Support different GGTT views into the same object. * This means there can be multiple VMA mappings per object and per VM. - * i915_ggtt_view_type is used to distinguish between those entries. - * The default one of zero (I915_GGTT_VIEW_NORMAL) is default and also + * i915_gtt_view_type is used to distinguish between those entries. + * The default one of zero (I915_GTT_VIEW_NORMAL) is default and also * assumed in GEM functions which take no ggtt view parameter. */ - struct i915_ggtt_view ggtt_view; + struct i915_gtt_view gtt_view; /** This object's place on the active/inactive lists */ struct list_head vm_link; diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index b6b971606727..d638235e1d26 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -37,6 +37,7 @@ struct drm_printer; struct drm_i915_private; +struct intel_gt_definition; /* Keep in gen based order, and chronological order within a gen */ enum intel_platform { @@ -171,6 +172,7 @@ enum intel_ppgtt_type { func(has_runtime_pm); \ func(has_snoop); \ func(has_coherent_ggtt); \ + func(tuning_thread_rr_after_dep); \ func(unfenced_needs_alignment); \ func(hws_needs_physical); @@ -257,6 +259,8 @@ struct intel_device_info { unsigned int dma_mask_size; /* available DMA address bits */ + const struct intel_gt_definition *extra_gt_list; + u8 gt; /* GT number, 0 if undefined */ #define DEFINE_FLAG(name) u8 name:1 diff --git a/drivers/gpu/drm/i915/intel_mchbar_regs.h b/drivers/gpu/drm/i915/intel_mchbar_regs.h index 2aad2f0cc8db..ffc702b79579 100644 --- a/drivers/gpu/drm/i915/intel_mchbar_regs.h +++ b/drivers/gpu/drm/i915/intel_mchbar_regs.h @@ -196,6 +196,9 @@ #define RP1_CAP_MASK REG_GENMASK(15, 8) #define RPN_CAP_MASK REG_GENMASK(23, 16) +#define GEN10_FREQ_INFO_REC _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5ef0) +#define RPE_MASK REG_GENMASK(15, 8) + /* snb MCH registers for priority tuning */ #define MCH_SSKPD _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5d10) #define SSKPD_NEW_WM0_MASK_HSW REG_GENMASK64(63, 56) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 7b454314ab85..8f86f56e7ca4 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -4356,9 +4356,8 @@ static void icl_init_clock_gating(struct drm_i915_private *dev_priv) static void gen12lp_init_clock_gating(struct drm_i915_private *dev_priv) { - /* Wa_1409120013:tgl,rkl,adl-s,dg1,dg2 */ - if (IS_TIGERLAKE(dev_priv) || IS_ROCKETLAKE(dev_priv) || - IS_ALDERLAKE_S(dev_priv) || IS_DG1(dev_priv) || IS_DG2(dev_priv)) + /* Wa_1409120013 */ + if (DISPLAY_VER(dev_priv) == 12) intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), DPFC_CHICKEN_COMP_DUMMY_PIXEL); diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index c96b5c762ae6..5cd423c7b646 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -21,6 +21,7 @@ * IN THE SOFTWARE. */ +#include <drm/drm_managed.h> #include <linux/pm_runtime.h> #include "gt/intel_engine_regs.h" @@ -44,29 +45,47 @@ fw_domains_get(struct intel_uncore *uncore, enum forcewake_domains fw_domains) } void -intel_uncore_mmio_debug_init_early(struct intel_uncore_mmio_debug *mmio_debug) +intel_uncore_mmio_debug_init_early(struct drm_i915_private *i915) { - spin_lock_init(&mmio_debug->lock); - mmio_debug->unclaimed_mmio_check = 1; + spin_lock_init(&i915->mmio_debug.lock); + i915->mmio_debug.unclaimed_mmio_check = 1; + + i915->uncore.debug = &i915->mmio_debug; } -static void mmio_debug_suspend(struct intel_uncore_mmio_debug *mmio_debug) +static void mmio_debug_suspend(struct intel_uncore *uncore) { - lockdep_assert_held(&mmio_debug->lock); + if (!uncore->debug) + return; + + spin_lock(&uncore->debug->lock); /* Save and disable mmio debugging for the user bypass */ - if (!mmio_debug->suspend_count++) { - mmio_debug->saved_mmio_check = mmio_debug->unclaimed_mmio_check; - mmio_debug->unclaimed_mmio_check = 0; + if (!uncore->debug->suspend_count++) { + uncore->debug->saved_mmio_check = uncore->debug->unclaimed_mmio_check; + uncore->debug->unclaimed_mmio_check = 0; } + + spin_unlock(&uncore->debug->lock); } -static void mmio_debug_resume(struct intel_uncore_mmio_debug *mmio_debug) +static bool check_for_unclaimed_mmio(struct intel_uncore *uncore); + +static void mmio_debug_resume(struct intel_uncore *uncore) { - lockdep_assert_held(&mmio_debug->lock); + if (!uncore->debug) + return; + + spin_lock(&uncore->debug->lock); + + if (!--uncore->debug->suspend_count) + uncore->debug->unclaimed_mmio_check = uncore->debug->saved_mmio_check; + + if (check_for_unclaimed_mmio(uncore)) + drm_info(&uncore->i915->drm, + "Invalid mmio detected during user access\n"); - if (!--mmio_debug->suspend_count) - mmio_debug->unclaimed_mmio_check = mmio_debug->saved_mmio_check; + spin_unlock(&uncore->debug->lock); } static const char * const forcewake_domain_names[] = { @@ -112,8 +131,11 @@ fw_domain_reset(const struct intel_uncore_forcewake_domain *d) * trying to reset here does exist at this point (engines could be fused * off in ICL+), so no waiting for acks */ - /* WaRsClearFWBitsAtReset:bdw,skl */ - fw_clear(d, 0xffff); + /* WaRsClearFWBitsAtReset */ + if (GRAPHICS_VER(d->uncore->i915) >= 12) + fw_clear(d, 0xefff); + else + fw_clear(d, 0xffff); } static inline void @@ -674,9 +696,7 @@ void intel_uncore_forcewake_user_get(struct intel_uncore *uncore) spin_lock_irq(&uncore->lock); if (!uncore->user_forcewake_count++) { intel_uncore_forcewake_get__locked(uncore, FORCEWAKE_ALL); - spin_lock(&uncore->debug->lock); - mmio_debug_suspend(uncore->debug); - spin_unlock(&uncore->debug->lock); + mmio_debug_suspend(uncore); } spin_unlock_irq(&uncore->lock); } @@ -692,14 +712,7 @@ void intel_uncore_forcewake_user_put(struct intel_uncore *uncore) { spin_lock_irq(&uncore->lock); if (!--uncore->user_forcewake_count) { - spin_lock(&uncore->debug->lock); - mmio_debug_resume(uncore->debug); - - if (check_for_unclaimed_mmio(uncore)) - drm_info(&uncore->i915->drm, - "Invalid mmio detected during user access\n"); - spin_unlock(&uncore->debug->lock); - + mmio_debug_resume(uncore); intel_uncore_forcewake_put__locked(uncore, FORCEWAKE_ALL); } spin_unlock_irq(&uncore->lock); @@ -915,6 +928,9 @@ find_fw_domain(struct intel_uncore *uncore, u32 offset) { const struct intel_forcewake_range *entry; + if (IS_GSI_REG(offset)) + offset += uncore->gsi_offset; + entry = BSEARCH(offset, uncore->fw_domains_table, uncore->fw_domains_table_entries, @@ -1130,6 +1146,9 @@ static bool is_shadowed(struct intel_uncore *uncore, u32 offset) if (drm_WARN_ON(&uncore->i915->drm, !uncore->shadowed_reg_table)) return false; + if (IS_GSI_REG(offset)) + offset += uncore->gsi_offset; + return BSEARCH(offset, uncore->shadowed_reg_table, uncore->shadowed_reg_table_entries, @@ -1701,7 +1720,7 @@ unclaimed_reg_debug(struct intel_uncore *uncore, const bool read, const bool before) { - if (likely(!uncore->i915->params.mmio_debug)) + if (likely(!uncore->i915->params.mmio_debug) || !uncore->debug) return; /* interrupts are disabled and re-enabled around uncore->lock usage */ @@ -1982,8 +2001,8 @@ static int __fw_domain_init(struct intel_uncore *uncore, d->uncore = uncore; d->wake_count = 0; - d->reg_set = uncore->regs + i915_mmio_reg_offset(reg_set); - d->reg_ack = uncore->regs + i915_mmio_reg_offset(reg_ack); + d->reg_set = uncore->regs + i915_mmio_reg_offset(reg_set) + uncore->gsi_offset; + d->reg_ack = uncore->regs + i915_mmio_reg_offset(reg_ack) + uncore->gsi_offset; d->id = domain_id; @@ -2220,6 +2239,11 @@ static int i915_pmic_bus_access_notifier(struct notifier_block *nb, return NOTIFY_OK; } +static void uncore_unmap_mmio(struct drm_device *drm, void *regs) +{ + iounmap(regs); +} + int intel_uncore_setup_mmio(struct intel_uncore *uncore, phys_addr_t phys_addr) { struct drm_i915_private *i915 = uncore->i915; @@ -2232,14 +2256,15 @@ int intel_uncore_setup_mmio(struct intel_uncore *uncore, phys_addr_t phys_addr) * clobbering the GTT which we want ioremap_wc instead. Fortunately, * the register BAR remains the same size for all the earlier * generations up to Ironlake. - * For dgfx chips register range is expanded to 4MB. + * For dgfx chips register range is expanded to 4MB, and this larger + * range is also used for integrated gpus beginning with Meteor Lake. */ - if (GRAPHICS_VER(i915) < 5) - mmio_size = 512 * 1024; - else if (IS_DGFX(i915)) + if (IS_DGFX(i915) || GRAPHICS_VER_FULL(i915) >= IP_VER(12, 70)) mmio_size = 4 * 1024 * 1024; - else + else if (GRAPHICS_VER(i915) >= 5) mmio_size = 2 * 1024 * 1024; + else + mmio_size = 512 * 1024; uncore->regs = ioremap(phys_addr, mmio_size); if (uncore->regs == NULL) { @@ -2247,12 +2272,7 @@ int intel_uncore_setup_mmio(struct intel_uncore *uncore, phys_addr_t phys_addr) return -EIO; } - return 0; -} - -void intel_uncore_cleanup_mmio(struct intel_uncore *uncore) -{ - iounmap(uncore->regs); + return drmm_add_action_or_reset(&i915->drm, uncore_unmap_mmio, uncore->regs); } void intel_uncore_init_early(struct intel_uncore *uncore, @@ -2262,7 +2282,6 @@ void intel_uncore_init_early(struct intel_uncore *uncore, uncore->i915 = gt->i915; uncore->gt = gt; uncore->rpm = >->i915->runtime_pm; - uncore->debug = >->i915->mmio_debug; } static void uncore_raw_init(struct intel_uncore *uncore) @@ -2442,8 +2461,11 @@ void intel_uncore_prune_engine_fw_domains(struct intel_uncore *uncore, } } -void intel_uncore_fini_mmio(struct intel_uncore *uncore) +/* Called via drm-managed action */ +void intel_uncore_fini_mmio(struct drm_device *dev, void *data) { + struct intel_uncore *uncore = data; + if (intel_uncore_has_forcewake(uncore)) { iosf_mbi_punit_acquire(); iosf_mbi_unregister_pmic_bus_access_notifier_unlocked( @@ -2573,6 +2595,9 @@ bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore) { bool ret; + if (!uncore->debug) + return false; + spin_lock_irq(&uncore->debug->lock); ret = check_for_unclaimed_mmio(uncore); spin_unlock_irq(&uncore->debug->lock); @@ -2585,6 +2610,9 @@ intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore) { bool ret = false; + if (drm_WARN_ON(&uncore->i915->drm, !uncore->debug)) + return false; + spin_lock_irq(&uncore->debug->lock); if (unlikely(uncore->debug->unclaimed_mmio_check <= 0)) diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index b1fa912a65e7..5022bac80b67 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -33,6 +33,7 @@ #include "i915_reg_defs.h" +struct drm_device; struct drm_i915_private; struct intel_runtime_pm; struct intel_uncore; @@ -135,6 +136,16 @@ struct intel_uncore { spinlock_t lock; /** lock is also taken in irq contexts. */ + /* + * Do we need to apply an additional offset to reach the beginning + * of the basic non-engine GT registers (referred to as "GSI" on + * newer platforms, or "GT block" on older platforms)? If so, we'll + * track that here and apply it transparently to registers in the + * appropriate range to maintain compatibility with our existing + * register definitions and GT code. + */ + u32 gsi_offset; + unsigned int flags; #define UNCORE_HAS_FORCEWAKE BIT(0) #define UNCORE_HAS_FPGA_DBG_UNCLAIMED BIT(1) @@ -210,8 +221,7 @@ intel_uncore_has_fifo(const struct intel_uncore *uncore) return uncore->flags & UNCORE_HAS_FIFO; } -void -intel_uncore_mmio_debug_init_early(struct intel_uncore_mmio_debug *mmio_debug); +void intel_uncore_mmio_debug_init_early(struct drm_i915_private *i915); void intel_uncore_init_early(struct intel_uncore *uncore, struct intel_gt *gt); int intel_uncore_setup_mmio(struct intel_uncore *uncore, phys_addr_t phys_addr); @@ -221,7 +231,7 @@ void intel_uncore_prune_engine_fw_domains(struct intel_uncore *uncore, bool intel_uncore_unclaimed_mmio(struct intel_uncore *uncore); bool intel_uncore_arm_unclaimed_mmio_detection(struct intel_uncore *uncore); void intel_uncore_cleanup_mmio(struct intel_uncore *uncore); -void intel_uncore_fini_mmio(struct intel_uncore *uncore); +void intel_uncore_fini_mmio(struct drm_device *dev, void *data); void intel_uncore_suspend(struct intel_uncore *uncore); void intel_uncore_resume_early(struct intel_uncore *uncore); void intel_uncore_runtime_resume(struct intel_uncore *uncore); @@ -294,19 +304,27 @@ intel_wait_for_register_fw(struct intel_uncore *uncore, 2, timeout_ms, NULL); } +#define IS_GSI_REG(reg) ((reg) < 0x40000) + /* register access functions */ #define __raw_read(x__, s__) \ static inline u##x__ __raw_uncore_read##x__(const struct intel_uncore *uncore, \ i915_reg_t reg) \ { \ - return read##s__(uncore->regs + i915_mmio_reg_offset(reg)); \ + u32 offset = i915_mmio_reg_offset(reg); \ + if (IS_GSI_REG(offset)) \ + offset += uncore->gsi_offset; \ + return read##s__(uncore->regs + offset); \ } #define __raw_write(x__, s__) \ static inline void __raw_uncore_write##x__(const struct intel_uncore *uncore, \ i915_reg_t reg, u##x__ val) \ { \ - write##s__(val, uncore->regs + i915_mmio_reg_offset(reg)); \ + u32 offset = i915_mmio_reg_offset(reg); \ + if (IS_GSI_REG(offset)) \ + offset += uncore->gsi_offset; \ + write##s__(val, uncore->regs + offset); \ } __raw_read(8, b) __raw_read(16, w) @@ -447,6 +465,18 @@ static inline int intel_uncore_write_and_verify(struct intel_uncore *uncore, return (reg_val & mask) != expected_val ? -EINVAL : 0; } +/* + * The raw_reg_{read,write} macros are intended as a micro-optimization for + * interrupt handlers so that the pointer indirection on uncore->regs can + * be computed once (and presumably cached in a register) instead of generating + * extra load instructions for each MMIO access. + * + * Given that these macros are only intended for non-GSI interrupt registers + * (and the goal is to avoid extra instructions generated by the compiler), + * these macros do not account for uncore->gsi_offset. Any caller that needs + * to use these macros on a GSI register is responsible for adding the + * appropriate GSI offset to the 'base' parameter. + */ #define raw_reg_read(base, reg) \ readl(base + i915_mmio_reg_offset(reg)) #define raw_reg_write(base, reg, value) \ diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c index 15311eaed848..69cdaaddc4a9 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c @@ -169,11 +169,23 @@ static void pxp_queue_termination(struct intel_pxp *pxp) * We want to get the same effect as if we received a termination * interrupt, so just pretend that we did. */ - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); intel_pxp_mark_termination_in_progress(pxp); pxp->session_events |= PXP_TERMINATION_REQUEST; queue_work(system_unbound_wq, &pxp->session_work); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); +} + +static bool pxp_component_bound(struct intel_pxp *pxp) +{ + bool bound = false; + + mutex_lock(&pxp->tee_mutex); + if (pxp->pxp_component) + bound = true; + mutex_unlock(&pxp->tee_mutex); + + return bound; } /* @@ -187,6 +199,9 @@ int intel_pxp_start(struct intel_pxp *pxp) if (!intel_pxp_is_enabled(pxp)) return -ENODEV; + if (wait_for(pxp_component_bound(pxp), 250)) + return -ENXIO; + mutex_lock(&pxp->arb_mutex); if (pxp->arb_is_valid) diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c index e888b5124a07..4359e8be4101 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c @@ -47,9 +47,9 @@ static int pxp_terminate_set(void *data, u64 val) return -ENODEV; /* simulate a termination interrupt */ - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); intel_pxp_irq_handler(pxp, GEN12_DISPLAY_PXP_STATE_TERMINATED_INTERRUPT); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); if (!wait_for_completion_timeout(&pxp->termination, msecs_to_jiffies(100))) diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c b/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c index 04745f914407..c28be430718a 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c @@ -25,7 +25,7 @@ void intel_pxp_irq_handler(struct intel_pxp *pxp, u16 iir) if (GEM_WARN_ON(!intel_pxp_is_enabled(pxp))) return; - lockdep_assert_held(>->irq_lock); + lockdep_assert_held(gt->irq_lock); if (unlikely(!iir)) return; @@ -55,16 +55,16 @@ static inline void __pxp_set_interrupts(struct intel_gt *gt, u32 interrupts) static inline void pxp_irq_reset(struct intel_gt *gt) { - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); gen11_gt_reset_one_iir(gt, 0, GEN11_KCR); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } void intel_pxp_irq_enable(struct intel_pxp *pxp) { struct intel_gt *gt = pxp_to_gt(pxp); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); if (!pxp->irq_enabled) WARN_ON_ONCE(gen11_gt_reset_one_iir(gt, 0, GEN11_KCR)); @@ -72,7 +72,7 @@ void intel_pxp_irq_enable(struct intel_pxp *pxp) __pxp_set_interrupts(gt, GEN12_PXP_INTERRUPTS); pxp->irq_enabled = true; - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); } void intel_pxp_irq_disable(struct intel_pxp *pxp) @@ -88,12 +88,12 @@ void intel_pxp_irq_disable(struct intel_pxp *pxp) */ GEM_WARN_ON(intel_pxp_is_active(pxp)); - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); pxp->irq_enabled = false; __pxp_set_interrupts(gt, 0); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); intel_synchronize_irq(gt->i915); pxp_irq_reset(gt); diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c index 92b00b4de240..1bb5b5249157 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_session.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_session.c @@ -144,9 +144,9 @@ void intel_pxp_session_work(struct work_struct *work) intel_wakeref_t wakeref; u32 events = 0; - spin_lock_irq(>->irq_lock); + spin_lock_irq(gt->irq_lock); events = fetch_and_zero(&pxp->session_events); - spin_unlock_irq(>->irq_lock); + spin_unlock_irq(gt->irq_lock); if (!events) return; diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index ab9f17fc85bc..e050a2de5fd1 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -1080,7 +1080,7 @@ static int misaligned_case(struct i915_address_space *vm, struct intel_memory_re bool is_stolen = mr->type == INTEL_MEMORY_STOLEN_SYSTEM || mr->type == INTEL_MEMORY_STOLEN_LOCAL; - obj = i915_gem_object_create_region(mr, size, 0, 0); + obj = i915_gem_object_create_region(mr, size, 0, I915_BO_ALLOC_GPU_ONLY); if (IS_ERR(obj)) { /* if iGVT-g or DMAR is active, stolen mem will be uninitialized */ if (PTR_ERR(obj) == -ENODEV && is_stolen) @@ -2324,5 +2324,5 @@ int i915_gem_gtt_live_selftests(struct drm_i915_private *i915) GEM_BUG_ON(offset_in_page(to_gt(i915)->ggtt->vm.total)); - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c index 88db2e3d81d0..429c6d73b159 100644 --- a/drivers/gpu/drm/i915/selftests/i915_perf.c +++ b/drivers/gpu/drm/i915/selftests/i915_perf.c @@ -431,7 +431,7 @@ int i915_perf_live_selftests(struct drm_i915_private *i915) if (err) return err; - err = i915_subtests(tests, i915); + err = i915_live_subtests(tests, i915); destroy_empty_config(&i915->perf); diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c index ec05f578a698..818a4909c1f3 100644 --- a/drivers/gpu/drm/i915/selftests/i915_request.c +++ b/drivers/gpu/drm/i915/selftests/i915_request.c @@ -1821,7 +1821,7 @@ int i915_request_live_selftests(struct drm_i915_private *i915) if (intel_gt_is_wedged(to_gt(i915))) return 0; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } static int switch_to_kernel_sync(struct intel_context *ce, int err) diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c index 6921ba128015..71b52d5efef4 100644 --- a/drivers/gpu/drm/i915/selftests/i915_vma.c +++ b/drivers/gpu/drm/i915/selftests/i915_vma.c @@ -51,9 +51,9 @@ static bool assert_vma(struct i915_vma *vma, ok = false; } - if (vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL) { + if (vma->gtt_view.type != I915_GTT_VIEW_NORMAL) { pr_err("VMA created with wrong type [%d]\n", - vma->ggtt_view.type); + vma->gtt_view.type); ok = false; } @@ -63,7 +63,7 @@ static bool assert_vma(struct i915_vma *vma, static struct i915_vma * checked_vma_instance(struct drm_i915_gem_object *obj, struct i915_address_space *vm, - const struct i915_ggtt_view *view) + const struct i915_gtt_view *view) { struct i915_vma *vma; bool ok = true; @@ -91,7 +91,7 @@ checked_vma_instance(struct drm_i915_gem_object *obj, } if (i915_vma_compare(vma, vma->vm, - i915_vma_is_ggtt(vma) ? &vma->ggtt_view : NULL)) { + i915_vma_is_ggtt(vma) ? &vma->gtt_view : NULL)) { pr_err("i915_vma_compare failed with itself\n"); return ERR_PTR(-EINVAL); } @@ -530,12 +530,12 @@ assert_remapped(struct drm_i915_gem_object *obj, return sg; } -static unsigned int remapped_size(enum i915_ggtt_view_type view_type, +static unsigned int remapped_size(enum i915_gtt_view_type view_type, const struct intel_remapped_plane_info *a, const struct intel_remapped_plane_info *b) { - if (view_type == I915_GGTT_VIEW_ROTATED) + if (view_type == I915_GTT_VIEW_ROTATED) return a->dst_stride * a->width + b->dst_stride * b->width; else return a->dst_stride * a->height + b->dst_stride * b->height; @@ -569,9 +569,9 @@ static int igt_vma_rotate_remap(void *arg) { } }, *a, *b; - enum i915_ggtt_view_type types[] = { - I915_GGTT_VIEW_ROTATED, - I915_GGTT_VIEW_REMAPPED, + enum i915_gtt_view_type types[] = { + I915_GTT_VIEW_ROTATED, + I915_GTT_VIEW_REMAPPED, 0, }, *t; const unsigned int max_pages = 64; @@ -588,7 +588,7 @@ static int igt_vma_rotate_remap(void *arg) for (t = types; *t; t++) { for (a = planes; a->width; a++) { for (b = planes + ARRAY_SIZE(planes); b-- != planes; ) { - struct i915_ggtt_view view = { + struct i915_gtt_view view = { .type = *t, .remapped.plane[0] = *a, .remapped.plane[1] = *b, @@ -602,11 +602,11 @@ static int igt_vma_rotate_remap(void *arg) max_offset = max_pages - max_offset; if (!plane_info[0].dst_stride) - plane_info[0].dst_stride = view.type == I915_GGTT_VIEW_ROTATED ? + plane_info[0].dst_stride = view.type == I915_GTT_VIEW_ROTATED ? plane_info[0].height : plane_info[0].width; if (!plane_info[1].dst_stride) - plane_info[1].dst_stride = view.type == I915_GGTT_VIEW_ROTATED ? + plane_info[1].dst_stride = view.type == I915_GTT_VIEW_ROTATED ? plane_info[1].height : plane_info[1].width; @@ -630,7 +630,7 @@ static int igt_vma_rotate_remap(void *arg) expected_pages = remapped_size(view.type, &plane_info[0], &plane_info[1]); - if (view.type == I915_GGTT_VIEW_ROTATED && + if (view.type == I915_GTT_VIEW_ROTATED && vma->size != expected_pages * PAGE_SIZE) { pr_err("VMA is wrong size, expected %lu, found %llu\n", PAGE_SIZE * expected_pages, vma->size); @@ -638,7 +638,7 @@ static int igt_vma_rotate_remap(void *arg) goto out_object; } - if (view.type == I915_GGTT_VIEW_REMAPPED && + if (view.type == I915_GTT_VIEW_REMAPPED && vma->size > expected_pages * PAGE_SIZE) { pr_err("VMA is wrong size, expected %lu, found %llu\n", PAGE_SIZE * expected_pages, vma->size); @@ -668,13 +668,13 @@ static int igt_vma_rotate_remap(void *arg) sg = vma->pages->sgl; for (n = 0; n < ARRAY_SIZE(view.rotated.plane); n++) { - if (view.type == I915_GGTT_VIEW_ROTATED) + if (view.type == I915_GTT_VIEW_ROTATED) sg = assert_rotated(obj, &view.rotated, n, sg); else sg = assert_remapped(obj, &view.remapped, n, sg); if (IS_ERR(sg)) { pr_err("Inconsistent %s VMA pages for plane %d: [(%d, %d, %d, %d, %d), (%d, %d, %d, %d, %d)]\n", - view.type == I915_GGTT_VIEW_ROTATED ? + view.type == I915_GTT_VIEW_ROTATED ? "rotated" : "remapped", n, plane_info[0].width, plane_info[0].height, @@ -741,7 +741,7 @@ static bool assert_partial(struct drm_i915_gem_object *obj, } static bool assert_pin(struct i915_vma *vma, - struct i915_ggtt_view *view, + struct i915_gtt_view *view, u64 size, const char *name) { @@ -759,8 +759,8 @@ static bool assert_pin(struct i915_vma *vma, ok = false; } - if (view && view->type != I915_GGTT_VIEW_NORMAL) { - if (memcmp(&vma->ggtt_view, view, sizeof(*view))) { + if (view && view->type != I915_GTT_VIEW_NORMAL) { + if (memcmp(&vma->gtt_view, view, sizeof(*view))) { pr_err("(%s) VMA mismatch upon creation!\n", name); ok = false; @@ -772,9 +772,9 @@ static bool assert_pin(struct i915_vma *vma, ok = false; } } else { - if (vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL) { + if (vma->gtt_view.type != I915_GTT_VIEW_NORMAL) { pr_err("Not the normal ggtt view! Found %d\n", - vma->ggtt_view.type); + vma->gtt_view.type); ok = false; } @@ -818,14 +818,14 @@ static int igt_vma_partial(void *arg) nvma = 0; for_each_prime_number_from(sz, 1, npages) { for_each_prime_number_from(offset, 0, npages - sz) { - struct i915_ggtt_view view; + struct i915_gtt_view view; - view.type = I915_GGTT_VIEW_PARTIAL; + view.type = I915_GTT_VIEW_PARTIAL; view.partial.offset = offset; view.partial.size = sz; if (sz == npages) - view.type = I915_GGTT_VIEW_NORMAL; + view.type = I915_GTT_VIEW_NORMAL; vma = checked_vma_instance(obj, vm, &view); if (IS_ERR(vma)) { @@ -976,9 +976,9 @@ static int igt_vma_remapped_gtt(void *arg) { } }, *p; - enum i915_ggtt_view_type types[] = { - I915_GGTT_VIEW_ROTATED, - I915_GGTT_VIEW_REMAPPED, + enum i915_gtt_view_type types[] = { + I915_GTT_VIEW_ROTATED, + I915_GTT_VIEW_REMAPPED, 0, }, *t; struct drm_i915_gem_object *obj; @@ -996,7 +996,7 @@ static int igt_vma_remapped_gtt(void *arg) for (t = types; *t; t++) { for (p = planes; p->width; p++) { - struct i915_ggtt_view view = { + struct i915_gtt_view view = { .type = *t, .rotated.plane[0] = *p, }; @@ -1012,7 +1012,7 @@ static int igt_vma_remapped_gtt(void *arg) goto out; if (!plane_info[0].dst_stride) - plane_info[0].dst_stride = *t == I915_GGTT_VIEW_ROTATED ? + plane_info[0].dst_stride = *t == I915_GTT_VIEW_ROTATED ? p->height : p->width; vma = i915_gem_object_ggtt_pin(obj, &view, 0, 0, PIN_MAPPABLE); @@ -1021,7 +1021,7 @@ static int igt_vma_remapped_gtt(void *arg) goto out; } - GEM_BUG_ON(vma->ggtt_view.type != *t); + GEM_BUG_ON(vma->gtt_view.type != *t); map = i915_vma_pin_iomap(vma); i915_vma_unpin(vma); @@ -1035,7 +1035,7 @@ static int igt_vma_remapped_gtt(void *arg) unsigned int offset; u32 val = y << 16 | x; - if (*t == I915_GGTT_VIEW_ROTATED) + if (*t == I915_GTT_VIEW_ROTATED) offset = (x * plane_info[0].dst_stride + y) * PAGE_SIZE; else offset = (y * plane_info[0].dst_stride + x) * PAGE_SIZE; @@ -1052,7 +1052,7 @@ static int igt_vma_remapped_gtt(void *arg) goto out; } - GEM_BUG_ON(vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL); + GEM_BUG_ON(vma->gtt_view.type != I915_GTT_VIEW_NORMAL); map = i915_vma_pin_iomap(vma); i915_vma_unpin(vma); @@ -1067,7 +1067,7 @@ static int igt_vma_remapped_gtt(void *arg) u32 exp = y << 16 | x; u32 val; - if (*t == I915_GGTT_VIEW_ROTATED) + if (*t == I915_GTT_VIEW_ROTATED) src_idx = rotated_index(&view.rotated, 0, x, y); else src_idx = remapped_index(&view.remapped, 0, x, y); @@ -1076,7 +1076,7 @@ static int igt_vma_remapped_gtt(void *arg) val = ioread32(&map[offset / sizeof(*map)]); if (val != exp) { pr_err("%s VMA write test failed, expected 0x%x, found 0x%x\n", - *t == I915_GGTT_VIEW_ROTATED ? "Rotated" : "Remapped", + *t == I915_GTT_VIEW_ROTATED ? "Rotated" : "Remapped", exp, val); i915_vma_unpin_iomap(vma); err = -EINVAL; @@ -1103,5 +1103,5 @@ int i915_vma_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_vma_remapped_gtt), }; - return i915_subtests(tests, i915); + return i915_live_subtests(tests, i915); } diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index 1fef5e233436..fff11c90f1fa 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -115,6 +115,7 @@ static struct dev_pm_domain pm_domain = { static void mock_gt_probe(struct drm_i915_private *i915) { i915->gt[0] = &i915->gt0; + i915->gt[0]->name = "Mock GT"; } struct drm_i915_private *mock_gem_device(void) |