diff options
Diffstat (limited to 'drivers/gpu')
96 files changed, 833 insertions, 453 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 12e71bbfd222..103635ab784c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -76,7 +76,7 @@ extern int amdgpu_modeset; extern int amdgpu_vram_limit; extern int amdgpu_vis_vram_limit; -extern unsigned amdgpu_gart_size; +extern int amdgpu_gart_size; extern int amdgpu_gtt_size; extern int amdgpu_moverate; extern int amdgpu_benchmarking; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index fb6e5dbd5a03..309f2419c6d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -155,7 +155,6 @@ static const struct kfd2kgd_calls kfd2kgd = { struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void) { return (struct kfd2kgd_calls *)&kfd2kgd; - return (struct kfd2kgd_calls *)&kfd2kgd; } static inline struct amdgpu_device *get_amdgpu_device(struct kgd_dev *kgd) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 269b835571eb..60d8bedb694d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -1079,6 +1079,9 @@ static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p, GFP_KERNEL); p->num_post_dep_syncobjs = 0; + if (!p->post_dep_syncobjs) + return -ENOMEM; + for (i = 0; i < num_deps; ++i) { p->post_dep_syncobjs[i] = drm_syncobj_find(p->filp, deps[i].handle); if (!p->post_dep_syncobjs[i]) @@ -1150,7 +1153,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, cs->out.handle = amdgpu_ctx_add_fence(p->ctx, ring, p->fence); job->uf_sequence = cs->out.handle; amdgpu_job_free_resources(job); - amdgpu_cs_parser_fini(p, 0, true); trace_amdgpu_cs_ioctl(job); amd_sched_entity_push_job(&job->base); @@ -1208,10 +1210,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) goto out; r = amdgpu_cs_submit(&parser, cs); - if (r) - goto out; - return 0; out: amdgpu_cs_parser_fini(&parser, r, reserved_buffers); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 1a459ac63df4..e630d918fefc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1062,11 +1062,11 @@ static void amdgpu_check_arguments(struct amdgpu_device *adev) amdgpu_sched_jobs = roundup_pow_of_two(amdgpu_sched_jobs); } - if (amdgpu_gart_size < 32) { + if (amdgpu_gart_size != -1 && amdgpu_gart_size < 32) { /* gart size must be greater or equal to 32M */ dev_warn(adev->dev, "gart size (%d) too small\n", amdgpu_gart_size); - amdgpu_gart_size = 32; + amdgpu_gart_size = -1; } if (amdgpu_gtt_size != -1 && amdgpu_gtt_size < 32) { @@ -2622,12 +2622,6 @@ static int amdgpu_recover_vram_from_shadow(struct amdgpu_device *adev, goto err; } - r = amdgpu_ttm_bind(&bo->shadow->tbo, &bo->shadow->tbo.mem); - if (r) { - DRM_ERROR("%p bind failed\n", bo->shadow); - goto err; - } - r = amdgpu_bo_restore_from_shadow(adev, ring, bo, NULL, fence, true); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index e39ec981b11c..0f16986ec5bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -76,7 +76,7 @@ int amdgpu_vram_limit = 0; int amdgpu_vis_vram_limit = 0; -unsigned amdgpu_gart_size = 256; +int amdgpu_gart_size = -1; /* auto */ int amdgpu_gtt_size = -1; /* auto */ int amdgpu_moverate = -1; /* auto */ int amdgpu_benchmarking = 0; @@ -128,7 +128,7 @@ module_param_named(vramlimit, amdgpu_vram_limit, int, 0600); MODULE_PARM_DESC(vis_vramlimit, "Restrict visible VRAM for testing, in megabytes"); module_param_named(vis_vramlimit, amdgpu_vis_vram_limit, int, 0444); -MODULE_PARM_DESC(gartsize, "Size of PCIE/IGP gart to setup in megabytes (32, 64, etc.)"); +MODULE_PARM_DESC(gartsize, "Size of GART to setup in megabytes (32, 64, etc., -1=auto)"); module_param_named(gartsize, amdgpu_gart_size, uint, 0600); MODULE_PARM_DESC(gttsize, "Size of the GTT domain in megabytes (-1 = auto)"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index 94c1e2e8e34c..f4370081f6e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -57,18 +57,6 @@ */ /** - * amdgpu_gart_set_defaults - set the default gart_size - * - * @adev: amdgpu_device pointer - * - * Set the default gart_size based on parameters and available VRAM. - */ -void amdgpu_gart_set_defaults(struct amdgpu_device *adev) -{ - adev->mc.gart_size = (uint64_t)amdgpu_gart_size << 20; -} - -/** * amdgpu_gart_table_ram_alloc - allocate system ram for gart page table * * @adev: amdgpu_device pointer diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h index d4cce6936200..afbe803b1a13 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h @@ -56,7 +56,6 @@ struct amdgpu_gart { const struct amdgpu_gart_funcs *gart_funcs; }; -void amdgpu_gart_set_defaults(struct amdgpu_device *adev); int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev); void amdgpu_gart_table_ram_free(struct amdgpu_device *adev); int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c index 9e05e257729f..0d15eb7d31d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c @@ -108,10 +108,10 @@ bool amdgpu_gtt_mgr_is_allocated(struct ttm_mem_reg *mem) * * Allocate the address space for a node. */ -int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man, - struct ttm_buffer_object *tbo, - const struct ttm_place *place, - struct ttm_mem_reg *mem) +static int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man, + struct ttm_buffer_object *tbo, + const struct ttm_place *place, + struct ttm_mem_reg *mem) { struct amdgpu_device *adev = amdgpu_ttm_adev(man->bdev); struct amdgpu_gtt_mgr *mgr = man->priv; @@ -143,12 +143,8 @@ int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man, fpfn, lpfn, mode); spin_unlock(&mgr->lock); - if (!r) { + if (!r) mem->start = node->start; - if (&tbo->mem == mem) - tbo->offset = (tbo->mem.start << PAGE_SHIFT) + - tbo->bdev->man[tbo->mem.mem_type].gpu_offset; - } return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 4bdd851f56d0..538e5f27d120 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -221,8 +221,9 @@ int amdgpu_irq_init(struct amdgpu_device *adev) spin_lock_init(&adev->irq.lock); - /* Disable vblank irqs aggressively for power-saving */ - adev->ddev->vblank_disable_immediate = true; + if (!adev->enable_virtual_display) + /* Disable vblank irqs aggressively for power-saving */ + adev->ddev->vblank_disable_immediate = true; r = drm_vblank_init(adev->ddev, adev->mode_info.num_crtc); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index e7e899190bef..9e495da0bb03 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -91,7 +91,10 @@ static void amdgpu_ttm_placement_init(struct amdgpu_device *adev, if (domain & AMDGPU_GEM_DOMAIN_GTT) { places[c].fpfn = 0; - places[c].lpfn = 0; + if (flags & AMDGPU_GEM_CREATE_SHADOW) + places[c].lpfn = adev->mc.gart_size >> PAGE_SHIFT; + else + places[c].lpfn = 0; places[c].flags = TTM_PL_FLAG_TT; if (flags & AMDGPU_GEM_CREATE_CPU_GTT_USWC) places[c].flags |= TTM_PL_FLAG_WC | @@ -446,17 +449,16 @@ static int amdgpu_bo_create_shadow(struct amdgpu_device *adev, if (bo->shadow) return 0; - bo->flags |= AMDGPU_GEM_CREATE_SHADOW; - memset(&placements, 0, - (AMDGPU_GEM_DOMAIN_MAX + 1) * sizeof(struct ttm_place)); - - amdgpu_ttm_placement_init(adev, &placement, - placements, AMDGPU_GEM_DOMAIN_GTT, - AMDGPU_GEM_CREATE_CPU_GTT_USWC); + memset(&placements, 0, sizeof(placements)); + amdgpu_ttm_placement_init(adev, &placement, placements, + AMDGPU_GEM_DOMAIN_GTT, + AMDGPU_GEM_CREATE_CPU_GTT_USWC | + AMDGPU_GEM_CREATE_SHADOW); r = amdgpu_bo_create_restricted(adev, size, byte_align, true, AMDGPU_GEM_DOMAIN_GTT, - AMDGPU_GEM_CREATE_CPU_GTT_USWC, + AMDGPU_GEM_CREATE_CPU_GTT_USWC | + AMDGPU_GEM_CREATE_SHADOW, NULL, &placement, bo->tbo.resv, 0, @@ -484,30 +486,28 @@ int amdgpu_bo_create(struct amdgpu_device *adev, { struct ttm_placement placement = {0}; struct ttm_place placements[AMDGPU_GEM_DOMAIN_MAX + 1]; + uint64_t parent_flags = flags & ~AMDGPU_GEM_CREATE_SHADOW; int r; - memset(&placements, 0, - (AMDGPU_GEM_DOMAIN_MAX + 1) * sizeof(struct ttm_place)); + memset(&placements, 0, sizeof(placements)); + amdgpu_ttm_placement_init(adev, &placement, placements, + domain, parent_flags); - amdgpu_ttm_placement_init(adev, &placement, - placements, domain, flags); - - r = amdgpu_bo_create_restricted(adev, size, byte_align, kernel, - domain, flags, sg, &placement, - resv, init_value, bo_ptr); + r = amdgpu_bo_create_restricted(adev, size, byte_align, kernel, domain, + parent_flags, sg, &placement, resv, + init_value, bo_ptr); if (r) return r; - if (amdgpu_need_backup(adev) && (flags & AMDGPU_GEM_CREATE_SHADOW)) { - if (!resv) { - r = ww_mutex_lock(&(*bo_ptr)->tbo.resv->lock, NULL); - WARN_ON(r != 0); - } + if ((flags & AMDGPU_GEM_CREATE_SHADOW) && amdgpu_need_backup(adev)) { + if (!resv) + WARN_ON(reservation_object_lock((*bo_ptr)->tbo.resv, + NULL)); r = amdgpu_bo_create_shadow(adev, size, byte_align, (*bo_ptr)); if (!resv) - ww_mutex_unlock(&(*bo_ptr)->tbo.resv->lock); + reservation_object_unlock((*bo_ptr)->tbo.resv); if (r) amdgpu_bo_unref(bo_ptr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 6c5646b48d1a..5ce65280b396 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -170,6 +170,16 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned irq_type) { int r; + int sched_hw_submission = amdgpu_sched_hw_submission; + + /* Set the hw submission limit higher for KIQ because + * it's used for a number of gfx/compute tasks by both + * KFD and KGD which may have outstanding fences and + * it doesn't really use the gpu scheduler anyway; + * KIQ tasks get submitted directly to the ring. + */ + if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ) + sched_hw_submission = max(sched_hw_submission, 256); if (ring->adev == NULL) { if (adev->num_rings >= AMDGPU_MAX_RINGS) @@ -178,8 +188,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, ring->adev = adev; ring->idx = adev->num_rings++; adev->rings[ring->idx] = ring; - r = amdgpu_fence_driver_init_ring(ring, - amdgpu_sched_hw_submission); + r = amdgpu_fence_driver_init_ring(ring, sched_hw_submission); if (r) return r; } @@ -218,8 +227,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, return r; } - ring->ring_size = roundup_pow_of_two(max_dw * 4 * - amdgpu_sched_hw_submission); + ring->ring_size = roundup_pow_of_two(max_dw * 4 * sched_hw_submission); ring->buf_mask = (ring->ring_size / 4) - 1; ring->ptr_mask = ring->funcs->support_64bit_ptrs ? diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 8b2c294f6f79..bc746131987f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -761,35 +761,11 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm) sg_free_table(ttm->sg); } -static int amdgpu_ttm_do_bind(struct ttm_tt *ttm, struct ttm_mem_reg *mem) -{ - struct amdgpu_ttm_tt *gtt = (void *)ttm; - uint64_t flags; - int r; - - spin_lock(>t->adev->gtt_list_lock); - flags = amdgpu_ttm_tt_pte_flags(gtt->adev, ttm, mem); - gtt->offset = (u64)mem->start << PAGE_SHIFT; - r = amdgpu_gart_bind(gtt->adev, gtt->offset, ttm->num_pages, - ttm->pages, gtt->ttm.dma_address, flags); - - if (r) { - DRM_ERROR("failed to bind %lu pages at 0x%08llX\n", - ttm->num_pages, gtt->offset); - goto error_gart_bind; - } - - list_add_tail(>t->list, >t->adev->gtt_list); -error_gart_bind: - spin_unlock(>t->adev->gtt_list_lock); - return r; - -} - static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, struct ttm_mem_reg *bo_mem) { struct amdgpu_ttm_tt *gtt = (void*)ttm; + uint64_t flags; int r = 0; if (gtt->userptr) { @@ -809,9 +785,24 @@ static int amdgpu_ttm_backend_bind(struct ttm_tt *ttm, bo_mem->mem_type == AMDGPU_PL_OA) return -EINVAL; - if (amdgpu_gtt_mgr_is_allocated(bo_mem)) - r = amdgpu_ttm_do_bind(ttm, bo_mem); + if (!amdgpu_gtt_mgr_is_allocated(bo_mem)) + return 0; + + spin_lock(>t->adev->gtt_list_lock); + flags = amdgpu_ttm_tt_pte_flags(gtt->adev, ttm, bo_mem); + gtt->offset = (u64)bo_mem->start << PAGE_SHIFT; + r = amdgpu_gart_bind(gtt->adev, gtt->offset, ttm->num_pages, + ttm->pages, gtt->ttm.dma_address, flags); + + if (r) { + DRM_ERROR("failed to bind %lu pages at 0x%08llX\n", + ttm->num_pages, gtt->offset); + goto error_gart_bind; + } + list_add_tail(>t->list, >t->adev->gtt_list); +error_gart_bind: + spin_unlock(>t->adev->gtt_list_lock); return r; } @@ -824,20 +815,39 @@ bool amdgpu_ttm_is_bound(struct ttm_tt *ttm) int amdgpu_ttm_bind(struct ttm_buffer_object *bo, struct ttm_mem_reg *bo_mem) { + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev); struct ttm_tt *ttm = bo->ttm; + struct ttm_mem_reg tmp; + + struct ttm_placement placement; + struct ttm_place placements; int r; if (!ttm || amdgpu_ttm_is_bound(ttm)) return 0; - r = amdgpu_gtt_mgr_alloc(&bo->bdev->man[TTM_PL_TT], bo, - NULL, bo_mem); - if (r) { - DRM_ERROR("Failed to allocate GTT address space (%d)\n", r); + tmp = bo->mem; + tmp.mm_node = NULL; + placement.num_placement = 1; + placement.placement = &placements; + placement.num_busy_placement = 1; + placement.busy_placement = &placements; + placements.fpfn = 0; + placements.lpfn = adev->mc.gart_size >> PAGE_SHIFT; + placements.flags = bo->mem.placement | TTM_PL_FLAG_TT; + + r = ttm_bo_mem_space(bo, &placement, &tmp, true, false); + if (unlikely(r)) return r; - } - return amdgpu_ttm_do_bind(ttm, bo_mem); + r = ttm_bo_move_ttm(bo, true, false, &tmp); + if (unlikely(r)) + ttm_bo_mem_put(bo, &tmp); + else + bo->offset = (bo->mem.start << PAGE_SHIFT) + + bo->bdev->man[bo->mem.mem_type].gpu_offset; + + return r; } int amdgpu_ttm_recover_gart(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index f22a4758719d..43093bffa2cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -62,10 +62,6 @@ extern const struct ttm_mem_type_manager_func amdgpu_gtt_mgr_func; extern const struct ttm_mem_type_manager_func amdgpu_vram_mgr_func; bool amdgpu_gtt_mgr_is_allocated(struct ttm_mem_reg *mem); -int amdgpu_gtt_mgr_alloc(struct ttm_mem_type_manager *man, - struct ttm_buffer_object *tbo, - const struct ttm_place *place, - struct ttm_mem_reg *mem); uint64_t amdgpu_gtt_mgr_usage(struct ttm_mem_type_manager *man); uint64_t amdgpu_vram_mgr_usage(struct ttm_mem_type_manager *man); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b9a5a77eedaf..bd20ff018512 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -165,14 +165,6 @@ static int amdgpu_vm_validate_level(struct amdgpu_vm_pt *parent, unsigned i; int r; - if (parent->bo->shadow) { - struct amdgpu_bo *shadow = parent->bo->shadow; - - r = amdgpu_ttm_bind(&shadow->tbo, &shadow->tbo.mem); - if (r) - return r; - } - if (use_cpu_for_update) { r = amdgpu_bo_kmap(parent->bo, NULL); if (r) @@ -1277,7 +1269,7 @@ static void amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p, /* In the case of a mixed PT the PDE must point to it*/ if (p->adev->asic_type < CHIP_VEGA10 || nptes != AMDGPU_VM_PTE_COUNT(p->adev) || - p->func == amdgpu_vm_do_copy_ptes || + p->src || !(flags & AMDGPU_PTE_VALID)) { dst = amdgpu_bo_gpu_offset(entry->bo); @@ -1294,9 +1286,23 @@ static void amdgpu_vm_handle_huge_pages(struct amdgpu_pte_update_params *p, entry->addr = (dst | flags); if (use_cpu_update) { + /* In case a huge page is replaced with a system + * memory mapping, p->pages_addr != NULL and + * amdgpu_vm_cpu_set_ptes would try to translate dst + * through amdgpu_vm_map_gart. But dst is already a + * GPU address (of the page table). Disable + * amdgpu_vm_map_gart temporarily. + */ + dma_addr_t *tmp; + + tmp = p->pages_addr; + p->pages_addr = NULL; + pd_addr = (unsigned long)amdgpu_bo_kptr(parent->bo); pde = pd_addr + (entry - parent->entries) * 8; amdgpu_vm_cpu_set_ptes(p, pde, dst, 1, 0, flags); + + p->pages_addr = tmp; } else { if (parent->bo->shadow) { pd_addr = amdgpu_bo_gpu_offset(parent->bo->shadow); @@ -1610,7 +1616,6 @@ error_free: * * @adev: amdgpu_device pointer * @exclusive: fence we need to sync to - * @gtt_flags: flags as they are used for GTT * @pages_addr: DMA addresses to use for mapping * @vm: requested vm * @mapping: mapped range and flags to use for the update @@ -1624,7 +1629,6 @@ error_free: */ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, struct dma_fence *exclusive, - uint64_t gtt_flags, dma_addr_t *pages_addr, struct amdgpu_vm *vm, struct amdgpu_bo_va_mapping *mapping, @@ -1679,11 +1683,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev, } if (pages_addr) { - if (flags == gtt_flags) - src = adev->gart.table_addr + - (addr >> AMDGPU_GPU_PAGE_SHIFT) * 8; - else - max_entries = min(max_entries, 16ull * 1024ull); + max_entries = min(max_entries, 16ull * 1024ull); addr = 0; } else if (flags & AMDGPU_PTE_VALID) { addr += adev->vm_manager.vram_base_offset; @@ -1728,10 +1728,10 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_vm *vm = bo_va->base.vm; struct amdgpu_bo_va_mapping *mapping; dma_addr_t *pages_addr = NULL; - uint64_t gtt_flags, flags; struct ttm_mem_reg *mem; struct drm_mm_node *nodes; struct dma_fence *exclusive; + uint64_t flags; int r; if (clear || !bo_va->base.bo) { @@ -1751,15 +1751,10 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, exclusive = reservation_object_get_excl(bo->tbo.resv); } - if (bo) { + if (bo) flags = amdgpu_ttm_tt_pte_flags(adev, bo->tbo.ttm, mem); - gtt_flags = (amdgpu_ttm_is_bound(bo->tbo.ttm) && - adev == amdgpu_ttm_adev(bo->tbo.bdev)) ? - flags : 0; - } else { + else flags = 0x0; - gtt_flags = ~0x0; - } spin_lock(&vm->status_lock); if (!list_empty(&bo_va->base.vm_status)) @@ -1767,8 +1762,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, spin_unlock(&vm->status_lock); list_for_each_entry(mapping, &bo_va->invalids, list) { - r = amdgpu_vm_bo_split_mapping(adev, exclusive, - gtt_flags, pages_addr, vm, + r = amdgpu_vm_bo_split_mapping(adev, exclusive, pages_addr, vm, mapping, flags, nodes, &bo_va->last_pt_update); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index d228f5a99044..dbbe986f90f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -636,7 +636,194 @@ static void gfx_v6_0_tiling_mode_table_init(struct amdgpu_device *adev) NUM_BANKS(ADDR_SURF_2_BANK); for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) WREG32(mmGB_TILE_MODE0 + reg_offset, tilemode[reg_offset]); - } else if (adev->asic_type == CHIP_OLAND || adev->asic_type == CHIP_HAINAN) { + } else if (adev->asic_type == CHIP_OLAND) { + tilemode[0] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[1] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[2] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[3] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[4] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[5] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(split_equal_to_row_size) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[6] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(split_equal_to_row_size) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[7] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(split_equal_to_row_size) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[8] = MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) | + ARRAY_MODE(ARRAY_LINEAR_ALIGNED) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[9] = MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) | + ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[10] = MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[11] = MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[12] = MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[13] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_1D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[14] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[15] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[16] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[17] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P4_8x16) | + TILE_SPLIT(split_equal_to_row_size) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[21] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[22] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4); + tilemode[23] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[24] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) | + NUM_BANKS(ADDR_SURF_16_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2); + tilemode[25] = MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) | + ARRAY_MODE(ARRAY_2D_TILED_THIN1) | + PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) | + TILE_SPLIT(ADDR_SURF_TILE_SPLIT_1KB) | + NUM_BANKS(ADDR_SURF_8_BANK) | + BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) | + BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) | + MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1); + for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) + WREG32(mmGB_TILE_MODE0 + reg_offset, tilemode[reg_offset]); + } else if (adev->asic_type == CHIP_HAINAN) { tilemode[0] = MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) | ARRAY_MODE(ARRAY_2D_TILED_THIN1) | PIPE_CONFIG(ADDR_SURF_P2) | diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 832e592fcd07..fc260c13b1da 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -4579,9 +4579,9 @@ static int gfx_v8_0_mqd_init(struct amdgpu_ring *ring) mqd->compute_misc_reserved = 0x00000003; if (!(adev->flags & AMD_IS_APU)) { mqd->dynamic_cu_mask_addr_lo = lower_32_bits(ring->mqd_gpu_addr - + offsetof(struct vi_mqd_allocation, dyamic_cu_mask)); + + offsetof(struct vi_mqd_allocation, dynamic_cu_mask)); mqd->dynamic_cu_mask_addr_hi = upper_32_bits(ring->mqd_gpu_addr - + offsetof(struct vi_mqd_allocation, dyamic_cu_mask)); + + offsetof(struct vi_mqd_allocation, dynamic_cu_mask)); } eop_base_addr = ring->eop_gpu_addr >> 8; mqd->cp_hqd_eop_base_addr_lo = eop_base_addr; @@ -4768,8 +4768,8 @@ static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring) mutex_unlock(&adev->srbm_mutex); } else { memset((void *)mqd, 0, sizeof(struct vi_mqd_allocation)); - ((struct vi_mqd_allocation *)mqd)->dyamic_cu_mask = 0xFFFFFFFF; - ((struct vi_mqd_allocation *)mqd)->dyamic_rb_mask = 0xFFFFFFFF; + ((struct vi_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; + ((struct vi_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; mutex_lock(&adev->srbm_mutex); vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0); gfx_v8_0_mqd_init(ring); @@ -4792,8 +4792,8 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring) if (!adev->gfx.in_reset && !adev->gfx.in_suspend) { memset((void *)mqd, 0, sizeof(struct vi_mqd_allocation)); - ((struct vi_mqd_allocation *)mqd)->dyamic_cu_mask = 0xFFFFFFFF; - ((struct vi_mqd_allocation *)mqd)->dyamic_rb_mask = 0xFFFFFFFF; + ((struct vi_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF; + ((struct vi_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF; mutex_lock(&adev->srbm_mutex); vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0); gfx_v8_0_mqd_init(ring); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 4f2788b61a08..6c8040e616c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -124,7 +124,7 @@ static void gfxhub_v1_0_init_tlb_regs(struct amdgpu_device *adev) static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev) { - uint32_t tmp, field; + uint32_t tmp; /* Setup L2 cache */ tmp = RREG32_SOC15(GC, 0, mmVM_L2_CNTL); @@ -143,9 +143,8 @@ static void gfxhub_v1_0_init_cache_regs(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); WREG32_SOC15(GC, 0, mmVM_L2_CNTL2, tmp); - field = adev->vm_manager.fragment_size; tmp = mmVM_L2_CNTL3_DEFAULT; - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6); WREG32_SOC15(GC, 0, mmVM_L2_CNTL3, tmp); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index 12b0c4cd7a5a..5be9c83dfcf7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -332,7 +332,24 @@ static int gmc_v6_0_mc_init(struct amdgpu_device *adev) adev->mc.real_vram_size = RREG32(mmCONFIG_MEMSIZE) * 1024ULL * 1024ULL; adev->mc.visible_vram_size = adev->mc.aper_size; - amdgpu_gart_set_defaults(adev); + /* set the gart size */ + if (amdgpu_gart_size == -1) { + switch (adev->asic_type) { + case CHIP_HAINAN: /* no MM engines */ + default: + adev->mc.gart_size = 256ULL << 20; + break; + case CHIP_VERDE: /* UVD, VCE do not support GPUVM */ + case CHIP_TAHITI: /* UVD, VCE do not support GPUVM */ + case CHIP_PITCAIRN: /* UVD, VCE do not support GPUVM */ + case CHIP_OLAND: /* UVD, VCE do not support GPUVM */ + adev->mc.gart_size = 1024ULL << 20; + break; + } + } else { + adev->mc.gart_size = (u64)amdgpu_gart_size << 20; + } + gmc_v6_0_vram_gtt_location(adev, &adev->mc); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index e42c1ad3af5e..eace9e7182c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -386,7 +386,27 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev) if (adev->mc.visible_vram_size > adev->mc.real_vram_size) adev->mc.visible_vram_size = adev->mc.real_vram_size; - amdgpu_gart_set_defaults(adev); + /* set the gart size */ + if (amdgpu_gart_size == -1) { + switch (adev->asic_type) { + case CHIP_TOPAZ: /* no MM engines */ + default: + adev->mc.gart_size = 256ULL << 20; + break; +#ifdef CONFIG_DRM_AMDGPU_CIK + case CHIP_BONAIRE: /* UVD, VCE do not support GPUVM */ + case CHIP_HAWAII: /* UVD, VCE do not support GPUVM */ + case CHIP_KAVERI: /* UVD, VCE do not support GPUVM */ + case CHIP_KABINI: /* UVD, VCE do not support GPUVM */ + case CHIP_MULLINS: /* UVD, VCE do not support GPUVM */ + adev->mc.gart_size = 1024ULL << 20; + break; +#endif + } + } else { + adev->mc.gart_size = (u64)amdgpu_gart_size << 20; + } + gmc_v7_0_vram_gtt_location(adev, &adev->mc); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 7ca2dae8237a..3b3326daf32b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -562,7 +562,26 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev) if (adev->mc.visible_vram_size > adev->mc.real_vram_size) adev->mc.visible_vram_size = adev->mc.real_vram_size; - amdgpu_gart_set_defaults(adev); + /* set the gart size */ + if (amdgpu_gart_size == -1) { + switch (adev->asic_type) { + case CHIP_POLARIS11: /* all engines support GPUVM */ + case CHIP_POLARIS10: /* all engines support GPUVM */ + case CHIP_POLARIS12: /* all engines support GPUVM */ + default: + adev->mc.gart_size = 256ULL << 20; + break; + case CHIP_TONGA: /* UVD, VCE do not support GPUVM */ + case CHIP_FIJI: /* UVD, VCE do not support GPUVM */ + case CHIP_CARRIZO: /* UVD, VCE do not support GPUVM, DCE SG support */ + case CHIP_STONEY: /* UVD does not support GPUVM, DCE SG support */ + adev->mc.gart_size = 1024ULL << 20; + break; + } + } else { + adev->mc.gart_size = (u64)amdgpu_gart_size << 20; + } + gmc_v8_0_vram_gtt_location(adev, &adev->mc); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 2769c2b3b56e..d04d0b123212 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -499,7 +499,21 @@ static int gmc_v9_0_mc_init(struct amdgpu_device *adev) if (adev->mc.visible_vram_size > adev->mc.real_vram_size) adev->mc.visible_vram_size = adev->mc.real_vram_size; - amdgpu_gart_set_defaults(adev); + /* set the gart size */ + if (amdgpu_gart_size == -1) { + switch (adev->asic_type) { + case CHIP_VEGA10: /* all engines support GPUVM */ + default: + adev->mc.gart_size = 256ULL << 20; + break; + case CHIP_RAVEN: /* DCE SG support */ + adev->mc.gart_size = 1024ULL << 20; + break; + } + } else { + adev->mc.gart_size = (u64)amdgpu_gart_size << 20; + } + gmc_v9_0_vram_gtt_location(adev, &adev->mc); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 4395a4f12149..74cb647da30e 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -138,7 +138,7 @@ static void mmhub_v1_0_init_tlb_regs(struct amdgpu_device *adev) static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev) { - uint32_t tmp, field; + uint32_t tmp; /* Setup L2 cache */ tmp = RREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL); @@ -157,9 +157,8 @@ static void mmhub_v1_0_init_cache_regs(struct amdgpu_device *adev) tmp = REG_SET_FIELD(tmp, VM_L2_CNTL2, INVALIDATE_L2_CACHE, 1); WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL2, tmp); - field = adev->vm_manager.fragment_size; tmp = mmVM_L2_CNTL3_DEFAULT; - tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, field); + tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, BANK_SELECT, 9); tmp = REG_SET_FIELD(tmp, VM_L2_CNTL3, L2_CACHE_BIGK_FRAGMENT_SIZE, 6); WREG32_SOC15(MMHUB, 0, mmVM_L2_CNTL3, tmp); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index e4a8c2e52cb2..660b3fbade41 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -892,6 +892,8 @@ static int kfd_ioctl_get_tile_config(struct file *filep, int err = 0; dev = kfd_device_by_id(args->gpu_id); + if (!dev) + return -EINVAL; dev->kfd2kgd->get_tile_config(dev->kgd, &config); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 5979158c3f7b..944abfad39c1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -292,7 +292,10 @@ static int create_signal_event(struct file *devkfd, struct kfd_event *ev) { if (p->signal_event_count == KFD_SIGNAL_EVENT_LIMIT) { - pr_warn("Signal event wasn't created because limit was reached\n"); + if (!p->signal_event_limit_reached) { + pr_warn("Signal event wasn't created because limit was reached\n"); + p->signal_event_limit_reached = true; + } return -ENOMEM; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 681b639f5133..ed71ad40e8f7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -183,8 +183,8 @@ static void uninitialize(struct kernel_queue *kq) { if (kq->queue->properties.type == KFD_QUEUE_TYPE_HIQ) kq->mqd->destroy_mqd(kq->mqd, - NULL, - false, + kq->queue->mqd, + KFD_PREEMPT_TYPE_WAVEFRONT_RESET, QUEUE_PREEMPT_DEFAULT_TIMEOUT_MS, kq->queue->pipe, kq->queue->queue); @@ -210,6 +210,11 @@ static int acquire_packet_buffer(struct kernel_queue *kq, uint32_t wptr, rptr; unsigned int *queue_address; + /* When rptr == wptr, the buffer is empty. + * When rptr == wptr + 1, the buffer is full. + * It is always rptr that advances to the position of wptr, rather than + * the opposite. So we can only use up to queue_size_dwords - 1 dwords. + */ rptr = *kq->rptr_kernel; wptr = *kq->wptr_kernel; queue_address = (unsigned int *)kq->pq_kernel_addr; @@ -219,11 +224,10 @@ static int acquire_packet_buffer(struct kernel_queue *kq, pr_debug("wptr: %d\n", wptr); pr_debug("queue_address 0x%p\n", queue_address); - available_size = (rptr - 1 - wptr + queue_size_dwords) % + available_size = (rptr + queue_size_dwords - 1 - wptr) % queue_size_dwords; - if (packet_size_in_dwords >= queue_size_dwords || - packet_size_in_dwords >= available_size) { + if (packet_size_in_dwords > available_size) { /* * make sure calling functions know * acquire_packet_buffer() failed @@ -233,6 +237,14 @@ static int acquire_packet_buffer(struct kernel_queue *kq, } if (wptr + packet_size_in_dwords >= queue_size_dwords) { + /* make sure after rolling back to position 0, there is + * still enough space. + */ + if (packet_size_in_dwords >= rptr) { + *buffer_ptr = NULL; + return -ENOMEM; + } + /* fill nops, roll back and start at position 0 */ while (wptr > 0) { queue_address[wptr] = kq->nop_packet; wptr = (wptr + 1) % queue_size_dwords; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index b397ec726400..b87e96cee5fa 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -521,6 +521,7 @@ struct kfd_process { struct list_head signal_event_pages; u32 next_nonsignal_event_id; size_t signal_event_count; + bool signal_event_limit_reached; }; /** diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 1cae95e2b13a..03bec765b03d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -143,7 +143,6 @@ int pqm_create_queue(struct process_queue_manager *pqm, int num_queues = 0; struct queue *cur; - memset(&q_properties, 0, sizeof(struct queue_properties)); memcpy(&q_properties, properties, sizeof(struct queue_properties)); q = NULL; kq = NULL; diff --git a/drivers/gpu/drm/amd/include/vi_structs.h b/drivers/gpu/drm/amd/include/vi_structs.h index ca93b5160ba6..3e606a761d0e 100644 --- a/drivers/gpu/drm/amd/include/vi_structs.h +++ b/drivers/gpu/drm/amd/include/vi_structs.h @@ -419,8 +419,8 @@ struct vi_mqd_allocation { struct vi_mqd mqd; uint32_t wptr_poll_mem; uint32_t rptr_report_mem; - uint32_t dyamic_cu_mask; - uint32_t dyamic_rb_mask; + uint32_t dynamic_cu_mask; + uint32_t dynamic_rb_mask; }; struct cz_mqd { diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index 9d71a259d97d..f8f02e70b8bc 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -1558,7 +1558,8 @@ static int vega10_populate_smc_link_levels(struct pp_hwmgr *hwmgr) */ static int vega10_populate_single_gfx_level(struct pp_hwmgr *hwmgr, - uint32_t gfx_clock, PllSetting_t *current_gfxclk_level) + uint32_t gfx_clock, PllSetting_t *current_gfxclk_level, + uint32_t *acg_freq) { struct phm_ppt_v2_information *table_info = (struct phm_ppt_v2_information *)(hwmgr->pptable); @@ -1609,6 +1610,8 @@ static int vega10_populate_single_gfx_level(struct pp_hwmgr *hwmgr, cpu_to_le16(dividers.usPll_ss_slew_frac); current_gfxclk_level->Did = (uint8_t)(dividers.ulDid); + *acg_freq = gfx_clock / 100; /* 100 Khz to Mhz conversion */ + return 0; } @@ -1689,7 +1692,8 @@ static int vega10_populate_all_graphic_levels(struct pp_hwmgr *hwmgr) for (i = 0; i < dpm_table->count; i++) { result = vega10_populate_single_gfx_level(hwmgr, dpm_table->dpm_levels[i].value, - &(pp_table->GfxclkLevel[i])); + &(pp_table->GfxclkLevel[i]), + &(pp_table->AcgFreqTable[i])); if (result) return result; } @@ -1698,7 +1702,8 @@ static int vega10_populate_all_graphic_levels(struct pp_hwmgr *hwmgr) while (i < NUM_GFXCLK_DPM_LEVELS) { result = vega10_populate_single_gfx_level(hwmgr, dpm_table->dpm_levels[j].value, - &(pp_table->GfxclkLevel[i])); + &(pp_table->GfxclkLevel[i]), + &(pp_table->AcgFreqTable[i])); if (result) return result; i++; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h b/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h index f6d6c61f796a..2818c98ff5ca 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu9_driver_if.h @@ -315,10 +315,12 @@ typedef struct { uint8_t AcgEnable[NUM_GFXCLK_DPM_LEVELS]; GbVdroopTable_t AcgBtcGbVdroopTable; QuadraticInt_t AcgAvfsGb; - uint32_t Reserved[4]; + + /* ACG Frequency Table, in Mhz */ + uint32_t AcgFreqTable[NUM_GFXCLK_DPM_LEVELS]; /* Padding - ignore */ - uint32_t MmHubPadding[7]; /* SMU internal use */ + uint32_t MmHubPadding[3]; /* SMU internal use */ } PPTable_t; diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c index 76347ff6d655..c49a6f22002f 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c @@ -380,7 +380,8 @@ static int smu7_populate_single_firmware_entry(struct pp_smumgr *smumgr, entry->num_register_entries = 0; } - if (fw_type == UCODE_ID_RLC_G) + if ((fw_type == UCODE_ID_RLC_G) + || (fw_type == UCODE_ID_CP_MEC)) entry->flags = 1; else entry->flags = 0; diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 4e53aae9a1fb..0028591f3f95 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -2960,6 +2960,7 @@ out: drm_modeset_backoff(&ctx); } + drm_atomic_state_put(state); drm_modeset_drop_locks(&ctx); drm_modeset_acquire_fini(&ctx); diff --git a/drivers/gpu/drm/drm_blend.c b/drivers/gpu/drm/drm_blend.c index db6aeec50b82..2e5e089dd912 100644 --- a/drivers/gpu/drm/drm_blend.c +++ b/drivers/gpu/drm/drm_blend.c @@ -319,7 +319,7 @@ static int drm_atomic_helper_crtc_normalize_zpos(struct drm_crtc *crtc, DRM_DEBUG_ATOMIC("[CRTC:%d:%s] calculating normalized zpos values\n", crtc->base.id, crtc->name); - states = kmalloc_array(total_planes, sizeof(*states), GFP_TEMPORARY); + states = kmalloc_array(total_planes, sizeof(*states), GFP_KERNEL); if (!states) return -ENOMEM; diff --git a/drivers/gpu/drm/drm_dp_dual_mode_helper.c b/drivers/gpu/drm/drm_dp_dual_mode_helper.c index 80e62f669321..0ef9011a1856 100644 --- a/drivers/gpu/drm/drm_dp_dual_mode_helper.c +++ b/drivers/gpu/drm/drm_dp_dual_mode_helper.c @@ -111,7 +111,7 @@ ssize_t drm_dp_dual_mode_write(struct i2c_adapter *adapter, void *data; int ret; - data = kmalloc(msg.len, GFP_TEMPORARY); + data = kmalloc(msg.len, GFP_KERNEL); if (!data) return -ENOMEM; diff --git a/drivers/gpu/drm/drm_scdc_helper.c b/drivers/gpu/drm/drm_scdc_helper.c index 7d1b0f011d33..935653eb3616 100644 --- a/drivers/gpu/drm/drm_scdc_helper.c +++ b/drivers/gpu/drm/drm_scdc_helper.c @@ -102,7 +102,7 @@ ssize_t drm_scdc_write(struct i2c_adapter *adapter, u8 offset, void *data; int err; - data = kmalloc(1 + size, GFP_TEMPORARY); + data = kmalloc(1 + size, GFP_KERNEL); if (!data) return -ENOMEM; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index 5a634594a6ce..57881167ccd2 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -551,12 +551,15 @@ static const struct etnaviv_gem_ops etnaviv_gem_shmem_ops = { void etnaviv_gem_free_object(struct drm_gem_object *obj) { struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj); + struct etnaviv_drm_private *priv = obj->dev->dev_private; struct etnaviv_vram_mapping *mapping, *tmp; /* object should not be active */ WARN_ON(is_active(etnaviv_obj)); + mutex_lock(&priv->gem_lock); list_del(&etnaviv_obj->gem_node); + mutex_unlock(&priv->gem_lock); list_for_each_entry_safe(mapping, tmp, &etnaviv_obj->vram_list, obj_node) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c index a7ff2e4c00d2..46dfe0737f43 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c @@ -37,7 +37,7 @@ static struct etnaviv_gem_submit *submit_create(struct drm_device *dev, struct etnaviv_gem_submit *submit; size_t sz = size_vstruct(nr, sizeof(submit->bos[0]), sizeof(*submit)); - submit = kmalloc(sz, GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY); + submit = kmalloc(sz, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); if (submit) { submit->dev = dev; submit->gpu = gpu; @@ -445,8 +445,10 @@ int etnaviv_ioctl_gem_submit(struct drm_device *dev, void *data, cmdbuf->user_size = ALIGN(args->stream_size, 8); ret = etnaviv_gpu_submit(gpu, submit, cmdbuf); - if (ret == 0) - cmdbuf = NULL; + if (ret) + goto out; + + cmdbuf = NULL; if (args->flags & ETNA_SUBMIT_FENCE_FD_OUT) { /* diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index 730b8d9db187..6be5b53c3b27 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -14,6 +14,7 @@ #include <linux/clk.h> #include <linux/component.h> #include <linux/iopoll.h> +#include <linux/irq.h> #include <linux/mfd/syscon.h> #include <linux/of_device.h> #include <linux/of_gpio.h> diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index b1f7299600f0..82b72425a42f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -168,23 +168,21 @@ static struct drm_driver exynos_drm_driver = { static int exynos_drm_suspend(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - struct drm_connector *connector; - struct drm_connector_list_iter conn_iter; + struct exynos_drm_private *private; if (pm_runtime_suspended(dev) || !drm_dev) return 0; - drm_connector_list_iter_begin(drm_dev, &conn_iter); - drm_for_each_connector_iter(connector, &conn_iter) { - int old_dpms = connector->dpms; - - if (connector->funcs->dpms) - connector->funcs->dpms(connector, DRM_MODE_DPMS_OFF); + private = drm_dev->dev_private; - /* Set the old mode back to the connector for resume */ - connector->dpms = old_dpms; + drm_kms_helper_poll_disable(drm_dev); + exynos_drm_fbdev_suspend(drm_dev); + private->suspend_state = drm_atomic_helper_suspend(drm_dev); + if (IS_ERR(private->suspend_state)) { + exynos_drm_fbdev_resume(drm_dev); + drm_kms_helper_poll_enable(drm_dev); + return PTR_ERR(private->suspend_state); } - drm_connector_list_iter_end(&conn_iter); return 0; } @@ -192,22 +190,15 @@ static int exynos_drm_suspend(struct device *dev) static int exynos_drm_resume(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); - struct drm_connector *connector; - struct drm_connector_list_iter conn_iter; + struct exynos_drm_private *private; if (pm_runtime_suspended(dev) || !drm_dev) return 0; - drm_connector_list_iter_begin(drm_dev, &conn_iter); - drm_for_each_connector_iter(connector, &conn_iter) { - if (connector->funcs->dpms) { - int dpms = connector->dpms; - - connector->dpms = DRM_MODE_DPMS_OFF; - connector->funcs->dpms(connector, dpms); - } - } - drm_connector_list_iter_end(&conn_iter); + private = drm_dev->dev_private; + drm_atomic_helper_resume(drm_dev, private->suspend_state); + exynos_drm_fbdev_resume(drm_dev); + drm_kms_helper_poll_enable(drm_dev); return 0; } @@ -439,6 +430,7 @@ static void exynos_drm_unbind(struct device *dev) kfree(drm->dev_private); drm->dev_private = NULL; + dev_set_drvdata(dev, NULL); drm_dev_unref(drm); } diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index cf131c2aa23e..f8bae4cb4823 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -202,6 +202,7 @@ struct drm_exynos_file_private { */ struct exynos_drm_private { struct drm_fb_helper *fb_helper; + struct drm_atomic_state *suspend_state; struct device *dma_dev; void *mapping; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index c3a068409b48..dfb66ecf417b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -18,6 +18,8 @@ #include <drm/drm_crtc_helper.h> #include <drm/exynos_drm.h> +#include <linux/console.h> + #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" #include "exynos_drm_fbdev.h" @@ -285,3 +287,21 @@ void exynos_drm_output_poll_changed(struct drm_device *dev) drm_fb_helper_hotplug_event(fb_helper); } + +void exynos_drm_fbdev_suspend(struct drm_device *dev) +{ + struct exynos_drm_private *private = dev->dev_private; + + console_lock(); + drm_fb_helper_set_suspend(private->fb_helper, 1); + console_unlock(); +} + +void exynos_drm_fbdev_resume(struct drm_device *dev) +{ + struct exynos_drm_private *private = dev->dev_private; + + console_lock(); + drm_fb_helper_set_suspend(private->fb_helper, 0); + console_unlock(); +} diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.h b/drivers/gpu/drm/exynos/exynos_drm_fbdev.h index 330eef87f718..645d1bb7f665 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.h +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.h @@ -21,6 +21,8 @@ int exynos_drm_fbdev_init(struct drm_device *dev); void exynos_drm_fbdev_fini(struct drm_device *dev); void exynos_drm_fbdev_restore_mode(struct drm_device *dev); void exynos_drm_output_poll_changed(struct drm_device *dev); +void exynos_drm_fbdev_suspend(struct drm_device *drm); +void exynos_drm_fbdev_resume(struct drm_device *drm); #else @@ -39,6 +41,14 @@ static inline void exynos_drm_fbdev_restore_mode(struct drm_device *dev) #define exynos_drm_output_poll_changed (NULL) +static inline void exynos_drm_fbdev_suspend(struct drm_device *drm) +{ +} + +static inline void exynos_drm_fbdev_resume(struct drm_device *drm) +{ +} + #endif #endif diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 214fa5e51963..0109ff40b1db 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -944,22 +944,27 @@ static bool hdmi_mode_fixup(struct drm_encoder *encoder, struct drm_device *dev = encoder->dev; struct drm_connector *connector; struct drm_display_mode *m; + struct drm_connector_list_iter conn_iter; int mode_ok; drm_mode_set_crtcinfo(adjusted_mode, 0); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + drm_connector_list_iter_begin(dev, &conn_iter); + drm_for_each_connector_iter(connector, &conn_iter) { if (connector->encoder == encoder) break; } + if (connector) + drm_connector_get(connector); + drm_connector_list_iter_end(&conn_iter); - if (connector->encoder != encoder) + if (!connector) return true; mode_ok = hdmi_mode_valid(connector, adjusted_mode); if (mode_ok == MODE_OK) - return true; + goto cleanup; /* * Find the most suitable mode and copy it to adjusted_mode. @@ -979,6 +984,9 @@ static bool hdmi_mode_fixup(struct drm_encoder *encoder, } } +cleanup: + drm_connector_put(connector); + return true; } diff --git a/drivers/gpu/drm/i915/gvt/cfg_space.c b/drivers/gpu/drm/i915/gvt/cfg_space.c index 40af17ec6312..ff3154fe6588 100644 --- a/drivers/gpu/drm/i915/gvt/cfg_space.c +++ b/drivers/gpu/drm/i915/gvt/cfg_space.c @@ -197,78 +197,65 @@ static int emulate_pci_command_write(struct intel_vgpu *vgpu, static int emulate_pci_bar_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { - unsigned int bar_index = - (rounddown(offset, 8) % PCI_BASE_ADDRESS_0) / 8; u32 new = *(u32 *)(p_data); bool lo = IS_ALIGNED(offset, 8); u64 size; int ret = 0; bool mmio_enabled = vgpu_cfg_space(vgpu)[PCI_COMMAND] & PCI_COMMAND_MEMORY; + struct intel_vgpu_pci_bar *bars = vgpu->cfg_space.bar; - if (WARN_ON(bar_index >= INTEL_GVT_PCI_BAR_MAX)) - return -EINVAL; - + /* + * Power-up software can determine how much address + * space the device requires by writing a value of + * all 1's to the register and then reading the value + * back. The device will return 0's in all don't-care + * address bits. + */ if (new == 0xffffffff) { - /* - * Power-up software can determine how much address - * space the device requires by writing a value of - * all 1's to the register and then reading the value - * back. The device will return 0's in all don't-care - * address bits. - */ - size = vgpu->cfg_space.bar[bar_index].size; - if (lo) { - new = rounddown(new, size); - } else { - u32 val = vgpu_cfg_space(vgpu)[rounddown(offset, 8)]; - /* for 32bit mode bar it returns all-0 in upper 32 - * bit, for 64bit mode bar it will calculate the - * size with lower 32bit and return the corresponding - * value + switch (offset) { + case PCI_BASE_ADDRESS_0: + case PCI_BASE_ADDRESS_1: + size = ~(bars[INTEL_GVT_PCI_BAR_GTTMMIO].size -1); + intel_vgpu_write_pci_bar(vgpu, offset, + size >> (lo ? 0 : 32), lo); + /* + * Untrap the BAR, since guest hasn't configured a + * valid GPA */ - if (val & PCI_BASE_ADDRESS_MEM_TYPE_64) - new &= (~(size-1)) >> 32; - else - new = 0; - } - /* - * Unmapp & untrap the BAR, since guest hasn't configured a - * valid GPA - */ - switch (bar_index) { - case INTEL_GVT_PCI_BAR_GTTMMIO: ret = trap_gttmmio(vgpu, false); break; - case INTEL_GVT_PCI_BAR_APERTURE: + case PCI_BASE_ADDRESS_2: + case PCI_BASE_ADDRESS_3: + size = ~(bars[INTEL_GVT_PCI_BAR_APERTURE].size -1); + intel_vgpu_write_pci_bar(vgpu, offset, + size >> (lo ? 0 : 32), lo); ret = map_aperture(vgpu, false); break; + default: + /* Unimplemented BARs */ + intel_vgpu_write_pci_bar(vgpu, offset, 0x0, false); } - intel_vgpu_write_pci_bar(vgpu, offset, new, lo); } else { - /* - * Unmapp & untrap the old BAR first, since guest has - * re-configured the BAR - */ - switch (bar_index) { - case INTEL_GVT_PCI_BAR_GTTMMIO: - ret = trap_gttmmio(vgpu, false); + switch (offset) { + case PCI_BASE_ADDRESS_0: + case PCI_BASE_ADDRESS_1: + /* + * Untrap the old BAR first, since guest has + * re-configured the BAR + */ + trap_gttmmio(vgpu, false); + intel_vgpu_write_pci_bar(vgpu, offset, new, lo); + ret = trap_gttmmio(vgpu, mmio_enabled); break; - case INTEL_GVT_PCI_BAR_APERTURE: - ret = map_aperture(vgpu, false); + case PCI_BASE_ADDRESS_2: + case PCI_BASE_ADDRESS_3: + map_aperture(vgpu, false); + intel_vgpu_write_pci_bar(vgpu, offset, new, lo); + ret = map_aperture(vgpu, mmio_enabled); break; - } - intel_vgpu_write_pci_bar(vgpu, offset, new, lo); - /* Track the new BAR */ - if (mmio_enabled) { - switch (bar_index) { - case INTEL_GVT_PCI_BAR_GTTMMIO: - ret = trap_gttmmio(vgpu, true); - break; - case INTEL_GVT_PCI_BAR_APERTURE: - ret = map_aperture(vgpu, true); - break; - } + default: + intel_vgpu_write_pci_bar(vgpu, offset, new, lo); } } return ret; @@ -299,10 +286,7 @@ int intel_vgpu_emulate_cfg_write(struct intel_vgpu *vgpu, unsigned int offset, } switch (rounddown(offset, 4)) { - case PCI_BASE_ADDRESS_0: - case PCI_BASE_ADDRESS_1: - case PCI_BASE_ADDRESS_2: - case PCI_BASE_ADDRESS_3: + case PCI_BASE_ADDRESS_0 ... PCI_BASE_ADDRESS_5: if (WARN_ON(!IS_ALIGNED(offset, 4))) return -EINVAL; return emulate_pci_bar_write(vgpu, offset, p_data, bytes); @@ -344,7 +328,6 @@ void intel_vgpu_init_cfg_space(struct intel_vgpu *vgpu, struct intel_gvt *gvt = vgpu->gvt; const struct intel_gvt_device_info *info = &gvt->device_info; u16 *gmch_ctl; - int i; memcpy(vgpu_cfg_space(vgpu), gvt->firmware.cfg_space, info->cfg_space_size); @@ -371,13 +354,13 @@ void intel_vgpu_init_cfg_space(struct intel_vgpu *vgpu, */ memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_1, 0, 4); memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_3, 0, 4); + memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_4, 0, 8); memset(vgpu_cfg_space(vgpu) + INTEL_GVT_PCI_OPREGION, 0, 4); - for (i = 0; i < INTEL_GVT_MAX_BAR_NUM; i++) { - vgpu->cfg_space.bar[i].size = pci_resource_len( - gvt->dev_priv->drm.pdev, i * 2); - vgpu->cfg_space.bar[i].tracked = false; - } + vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].size = + pci_resource_len(gvt->dev_priv->drm.pdev, 0); + vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_APERTURE].size = + pci_resource_len(gvt->dev_priv->drm.pdev, 2); } /** diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c index 436377da41ba..03532dfc0cd5 100644 --- a/drivers/gpu/drm/i915/gvt/sched_policy.c +++ b/drivers/gpu/drm/i915/gvt/sched_policy.c @@ -308,20 +308,8 @@ static int tbs_sched_init_vgpu(struct intel_vgpu *vgpu) static void tbs_sched_clean_vgpu(struct intel_vgpu *vgpu) { - struct intel_gvt_workload_scheduler *scheduler = &vgpu->gvt->scheduler; - int ring_id; - kfree(vgpu->sched_data); vgpu->sched_data = NULL; - - spin_lock_bh(&scheduler->mmio_context_lock); - for (ring_id = 0; ring_id < I915_NUM_ENGINES; ring_id++) { - if (scheduler->engine_owner[ring_id] == vgpu) { - intel_gvt_switch_mmio(vgpu, NULL, ring_id); - scheduler->engine_owner[ring_id] = NULL; - } - } - spin_unlock_bh(&scheduler->mmio_context_lock); } static void tbs_sched_start_schedule(struct intel_vgpu *vgpu) @@ -388,6 +376,7 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu) { struct intel_gvt_workload_scheduler *scheduler = &vgpu->gvt->scheduler; + int ring_id; gvt_dbg_core("vgpu%d: stop schedule\n", vgpu->id); @@ -401,4 +390,13 @@ void intel_vgpu_stop_schedule(struct intel_vgpu *vgpu) scheduler->need_reschedule = true; scheduler->current_vgpu = NULL; } + + spin_lock_bh(&scheduler->mmio_context_lock); + for (ring_id = 0; ring_id < I915_NUM_ENGINES; ring_id++) { + if (scheduler->engine_owner[ring_id] == vgpu) { + intel_gvt_switch_mmio(vgpu, NULL, ring_id); + scheduler->engine_owner[ring_id] = NULL; + } + } + spin_unlock_bh(&scheduler->mmio_context_lock); } diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 57317715977f..32e857dc507c 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2540,7 +2540,7 @@ static void *i915_gem_object_map(const struct drm_i915_gem_object *obj, if (n_pages > ARRAY_SIZE(stack_pages)) { /* Too big for stack -- allocate temporary array instead */ - pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_TEMPORARY); + pages = kvmalloc_array(n_pages, sizeof(*pages), GFP_KERNEL); if (!pages) return NULL; } @@ -2657,6 +2657,9 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, if (READ_ONCE(obj->mm.pages)) return -ENODEV; + if (obj->mm.madv != I915_MADV_WILLNEED) + return -EFAULT; + /* Before the pages are instantiated the object is treated as being * in the CPU domain. The pages will be clflushed as required before * use, and we can freely write into the pages directly. If userspace @@ -3013,10 +3016,15 @@ void i915_gem_reset_finish(struct drm_i915_private *dev_priv) static void nop_submit_request(struct drm_i915_gem_request *request) { + unsigned long flags; + GEM_BUG_ON(!i915_terminally_wedged(&request->i915->gpu_error)); dma_fence_set_error(&request->fence, -EIO); - i915_gem_request_submit(request); + + spin_lock_irqsave(&request->engine->timeline->lock, flags); + __i915_gem_request_submit(request); intel_engine_init_global_seqno(request->engine, request->global_seqno); + spin_unlock_irqrestore(&request->engine->timeline->lock, flags); } static void engine_set_wedged(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index 4df039ef2ce3..e161d383b526 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -33,21 +33,20 @@ #include "intel_drv.h" #include "i915_trace.h" -static bool ggtt_is_idle(struct drm_i915_private *dev_priv) +static bool ggtt_is_idle(struct drm_i915_private *i915) { - struct i915_ggtt *ggtt = &dev_priv->ggtt; - struct intel_engine_cs *engine; - enum intel_engine_id id; + struct intel_engine_cs *engine; + enum intel_engine_id id; - for_each_engine(engine, dev_priv, id) { - struct intel_timeline *tl; + if (i915->gt.active_requests) + return false; - tl = &ggtt->base.timeline.engine[engine->id]; - if (i915_gem_active_isset(&tl->last_request)) - return false; - } + for_each_engine(engine, i915, id) { + if (engine->last_retired_context != i915->kernel_context) + return false; + } - return true; + return true; } static int ggtt_flush(struct drm_i915_private *i915) @@ -157,7 +156,8 @@ i915_gem_evict_something(struct i915_address_space *vm, min_size, alignment, cache_level, start, end, mode); - /* Retire before we search the active list. Although we have + /* + * Retire before we search the active list. Although we have * reasonable accuracy in our retirement lists, we may have * a stray pin (preventing eviction) that can only be resolved by * retiring. @@ -182,7 +182,8 @@ search_again: BUG_ON(ret); } - /* Can we unpin some objects such as idle hw contents, + /* + * Can we unpin some objects such as idle hw contents, * or pending flips? But since only the GGTT has global entries * such as scanouts, rinbuffers and contexts, we can skip the * purge when inspecting per-process local address spaces. @@ -190,19 +191,33 @@ search_again: if (!i915_is_ggtt(vm) || flags & PIN_NONBLOCK) return -ENOSPC; - if (ggtt_is_idle(dev_priv)) { - /* If we still have pending pageflip completions, drop - * back to userspace to give our workqueues time to - * acquire our locks and unpin the old scanouts. - */ - return intel_has_pending_fb_unpin(dev_priv) ? -EAGAIN : -ENOSPC; - } + /* + * Not everything in the GGTT is tracked via VMA using + * i915_vma_move_to_active(), otherwise we could evict as required + * with minimal stalling. Instead we are forced to idle the GPU and + * explicitly retire outstanding requests which will then remove + * the pinning for active objects such as contexts and ring, + * enabling us to evict them on the next iteration. + * + * To ensure that all user contexts are evictable, we perform + * a switch to the perma-pinned kernel context. This all also gives + * us a termination condition, when the last retired context is + * the kernel's there is no more we can evict. + */ + if (!ggtt_is_idle(dev_priv)) { + ret = ggtt_flush(dev_priv); + if (ret) + return ret; - ret = ggtt_flush(dev_priv); - if (ret) - return ret; + goto search_again; + } - goto search_again; + /* + * If we still have pending pageflip completions, drop + * back to userspace to give our workqueues time to + * acquire our locks and unpin the old scanouts. + */ + return intel_has_pending_fb_unpin(dev_priv) ? -EAGAIN : -ENOSPC; found: /* drm_mm doesn't allow any other other operations while diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 50d5e24f91a9..92437f455b43 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -293,7 +293,7 @@ static int eb_create(struct i915_execbuffer *eb) * as possible to perform the allocation and warn * if it fails. */ - flags = GFP_TEMPORARY; + flags = GFP_KERNEL; if (size > 1) flags |= __GFP_NORETRY | __GFP_NOWARN; @@ -1515,7 +1515,7 @@ static int eb_copy_relocations(const struct i915_execbuffer *eb) urelocs = u64_to_user_ptr(eb->exec[i].relocs_ptr); size = nreloc * sizeof(*relocs); - relocs = kvmalloc_array(size, 1, GFP_TEMPORARY); + relocs = kvmalloc_array(size, 1, GFP_KERNEL); if (!relocs) { kvfree(relocs); err = -ENOMEM; @@ -2077,7 +2077,7 @@ get_fence_array(struct drm_i915_gem_execbuffer2 *args, return ERR_PTR(-EFAULT); fences = kvmalloc_array(args->num_cliprects, sizeof(*fences), - __GFP_NOWARN | GFP_TEMPORARY); + __GFP_NOWARN | GFP_KERNEL); if (!fences) return ERR_PTR(-ENOMEM); @@ -2463,9 +2463,9 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, /* Copy in the exec list from userland */ exec_list = kvmalloc_array(args->buffer_count, sizeof(*exec_list), - __GFP_NOWARN | GFP_TEMPORARY); + __GFP_NOWARN | GFP_KERNEL); exec2_list = kvmalloc_array(args->buffer_count + 1, sz, - __GFP_NOWARN | GFP_TEMPORARY); + __GFP_NOWARN | GFP_KERNEL); if (exec_list == NULL || exec2_list == NULL) { DRM_DEBUG("Failed to allocate exec list for %d buffers\n", args->buffer_count); @@ -2543,7 +2543,7 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data, /* Allocate an extra slot for use by the command parser */ exec2_list = kvmalloc_array(args->buffer_count + 1, sz, - __GFP_NOWARN | GFP_TEMPORARY); + __GFP_NOWARN | GFP_KERNEL); if (exec2_list == NULL) { DRM_DEBUG("Failed to allocate exec list for %d buffers\n", args->buffer_count); diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 0d5a988b3867..e2410eb5d96e 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -3231,7 +3231,7 @@ intel_rotate_pages(struct intel_rotation_info *rot_info, /* Allocate a temporary list of source pages for random access. */ page_addr_list = kvmalloc_array(n_pages, sizeof(dma_addr_t), - GFP_TEMPORARY); + GFP_KERNEL); if (!page_addr_list) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index 23fd18bd1b56..709efe2357ea 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -507,7 +507,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) ret = -ENOMEM; pinned = 0; - pvec = kvmalloc_array(npages, sizeof(struct page *), GFP_TEMPORARY); + pvec = kvmalloc_array(npages, sizeof(struct page *), GFP_KERNEL); if (pvec != NULL) { struct mm_struct *mm = obj->userptr.mm->mm; unsigned int flags = 0; @@ -643,7 +643,7 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj) if (mm == current->mm) { pvec = kvmalloc_array(num_pages, sizeof(struct page *), - GFP_TEMPORARY | + GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); if (pvec) /* defer to worker if malloc fails */ diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index ed5a1eb839ad..0c779671fe2d 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -787,16 +787,16 @@ int i915_error_state_buf_init(struct drm_i915_error_state_buf *ebuf, */ ebuf->size = count + 1 > PAGE_SIZE ? count + 1 : PAGE_SIZE; ebuf->buf = kmalloc(ebuf->size, - GFP_TEMPORARY | __GFP_NORETRY | __GFP_NOWARN); + GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); if (ebuf->buf == NULL) { ebuf->size = PAGE_SIZE; - ebuf->buf = kmalloc(ebuf->size, GFP_TEMPORARY); + ebuf->buf = kmalloc(ebuf->size, GFP_KERNEL); } if (ebuf->buf == NULL) { ebuf->size = 128; - ebuf->buf = kmalloc(ebuf->size, GFP_TEMPORARY); + ebuf->buf = kmalloc(ebuf->size, GFP_KERNEL); } if (ebuf->buf == NULL) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index e21ce9c18b6e..b63893eeca73 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -839,7 +839,6 @@ static bool i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, pipe); int position; int vbl_start, vbl_end, hsync_start, htotal, vtotal; - bool in_vbl = true; unsigned long irqflags; if (WARN_ON(!mode->crtc_clock)) { @@ -922,8 +921,6 @@ static bool i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); - in_vbl = position >= vbl_start && position < vbl_end; - /* * While in vblank, position will be negative * counting up towards 0 at vbl_end. And outside diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index ed7cd9ee2c2a..c9bcc6c45012 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -6998,6 +6998,7 @@ enum { */ #define L3_GENERAL_PRIO_CREDITS(x) (((x) >> 1) << 19) #define L3_HIGH_PRIO_CREDITS(x) (((x) >> 1) << 14) +#define L3_PRIO_CREDITS_MASK ((0x1f << 19) | (0x1f << 14)) #define GEN7_L3CNTLREG1 _MMIO(0xB01C) #define GEN7_WA_FOR_GEN7_L3_CONTROL 0x3C47FF8C diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index d805b6e6fe71..27743be5b768 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -606,11 +606,6 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder, connector->encoder->base.id, connector->encoder->name); - /* ELD Conn_Type */ - connector->eld[5] &= ~(3 << 2); - if (intel_crtc_has_dp_encoder(crtc_state)) - connector->eld[5] |= (1 << 2); - connector->eld[6] = drm_av_sync_delay(connector, adjusted_mode) / 2; if (dev_priv->display.audio_codec_enable) diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index 183e87e8ea31..5d4cd3d00564 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -1163,6 +1163,13 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, is_hdmi = is_dvi && (child->common.device_type & DEVICE_TYPE_NOT_HDMI_OUTPUT) == 0; is_edp = is_dp && (child->common.device_type & DEVICE_TYPE_INTERNAL_CONNECTOR); + if (port == PORT_A && is_dvi) { + DRM_DEBUG_KMS("VBT claims port A supports DVI%s, ignoring\n", + is_hdmi ? "/HDMI" : ""); + is_dvi = false; + is_hdmi = false; + } + info->supports_dvi = is_dvi; info->supports_hdmi = is_hdmi; info->supports_dp = is_dp; @@ -1233,7 +1240,7 @@ static void parse_ddi_ports(struct drm_i915_private *dev_priv, { enum port port; - if (!HAS_DDI(dev_priv)) + if (!HAS_DDI(dev_priv) && !IS_CHERRYVIEW(dev_priv)) return; if (!dev_priv->vbt.child_dev_num) diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c index ff9ecd211abb..b8315bca852b 100644 --- a/drivers/gpu/drm/i915/intel_color.c +++ b/drivers/gpu/drm/i915/intel_color.c @@ -74,7 +74,7 @@ #define I9XX_CSC_COEFF_1_0 \ ((7 << 12) | I9XX_CSC_COEFF_FP(CTM_COEFF_1_0, 8)) -static bool crtc_state_is_legacy(struct drm_crtc_state *state) +static bool crtc_state_is_legacy_gamma(struct drm_crtc_state *state) { return !state->degamma_lut && !state->ctm && @@ -288,7 +288,7 @@ static void cherryview_load_csc_matrix(struct drm_crtc_state *state) } mode = (state->ctm ? CGM_PIPE_MODE_CSC : 0); - if (!crtc_state_is_legacy(state)) { + if (!crtc_state_is_legacy_gamma(state)) { mode |= (state->degamma_lut ? CGM_PIPE_MODE_DEGAMMA : 0) | (state->gamma_lut ? CGM_PIPE_MODE_GAMMA : 0); } @@ -469,7 +469,7 @@ static void broadwell_load_luts(struct drm_crtc_state *state) struct intel_crtc_state *intel_state = to_intel_crtc_state(state); enum pipe pipe = to_intel_crtc(state->crtc)->pipe; - if (crtc_state_is_legacy(state)) { + if (crtc_state_is_legacy_gamma(state)) { haswell_load_luts(state); return; } @@ -529,7 +529,7 @@ static void glk_load_luts(struct drm_crtc_state *state) glk_load_degamma_lut(state); - if (crtc_state_is_legacy(state)) { + if (crtc_state_is_legacy_gamma(state)) { haswell_load_luts(state); return; } @@ -551,7 +551,7 @@ static void cherryview_load_luts(struct drm_crtc_state *state) uint32_t i, lut_size; uint32_t word0, word1; - if (crtc_state_is_legacy(state)) { + if (crtc_state_is_legacy_gamma(state)) { /* Turn off degamma/gamma on CGM block. */ I915_WRITE(CGM_PIPE_MODE(pipe), (state->ctm ? CGM_PIPE_MODE_CSC : 0)); @@ -632,12 +632,10 @@ int intel_color_check(struct drm_crtc *crtc, return 0; /* - * We also allow no degamma lut and a gamma lut at the legacy + * We also allow no degamma lut/ctm and a gamma lut at the legacy * size (256 entries). */ - if (!crtc_state->degamma_lut && - crtc_state->gamma_lut && - crtc_state->gamma_lut->length == LEGACY_LUT_LENGTH) + if (crtc_state_is_legacy_gamma(crtc_state)) return 0; return -EINVAL; diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 965988f79a55..92c1f8e166dc 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -216,7 +216,7 @@ static void gen9_set_dc_state_debugmask(struct drm_i915_private *dev_priv) mask = DC_STATE_DEBUG_MASK_MEMORY_UP; - if (IS_BROXTON(dev_priv)) + if (IS_GEN9_LP(dev_priv)) mask |= DC_STATE_DEBUG_MASK_CORES; /* The below bit doesn't need to be cleared ever afterwards */ diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 4b4fd1f8110b..5e5fe03b638c 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -664,8 +664,8 @@ intel_ddi_get_buf_trans_fdi(struct drm_i915_private *dev_priv, int *n_entries) { if (IS_BROADWELL(dev_priv)) { - *n_entries = ARRAY_SIZE(hsw_ddi_translations_fdi); - return hsw_ddi_translations_fdi; + *n_entries = ARRAY_SIZE(bdw_ddi_translations_fdi); + return bdw_ddi_translations_fdi; } else if (IS_HASWELL(dev_priv)) { *n_entries = ARRAY_SIZE(hsw_ddi_translations_fdi); return hsw_ddi_translations_fdi; @@ -1655,7 +1655,8 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder, out: if (ret && IS_GEN9_LP(dev_priv)) { tmp = I915_READ(BXT_PHY_CTL(port)); - if ((tmp & (BXT_PHY_LANE_POWERDOWN_ACK | + if ((tmp & (BXT_PHY_CMNLANE_POWERDOWN_ACK | + BXT_PHY_LANE_POWERDOWN_ACK | BXT_PHY_LANE_ENABLED)) != BXT_PHY_LANE_ENABLED) DRM_ERROR("Port %c enabled but PHY powered down? " "(PHY_CTL %08x)\n", port_name(port), tmp); @@ -2101,8 +2102,7 @@ static void intel_ddi_clk_select(struct intel_encoder *encoder, * register writes. */ val = I915_READ(DPCLKA_CFGCR0); - val &= ~(DPCLKA_CFGCR0_DDI_CLK_OFF(port) | - DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port)); + val &= ~DPCLKA_CFGCR0_DDI_CLK_OFF(port); I915_WRITE(DPCLKA_CFGCR0, val); } else if (IS_GEN9_BC(dev_priv)) { /* DDI -> PLL mapping */ diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f17275519484..5c7828c52d12 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -10245,13 +10245,10 @@ struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev, { struct drm_i915_private *dev_priv = to_i915(dev); struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder; + enum transcoder cpu_transcoder; struct drm_display_mode *mode; struct intel_crtc_state *pipe_config; - int htot = I915_READ(HTOTAL(cpu_transcoder)); - int hsync = I915_READ(HSYNC(cpu_transcoder)); - int vtot = I915_READ(VTOTAL(cpu_transcoder)); - int vsync = I915_READ(VSYNC(cpu_transcoder)); + u32 htot, hsync, vtot, vsync; enum pipe pipe = intel_crtc->pipe; mode = kzalloc(sizeof(*mode), GFP_KERNEL); @@ -10279,6 +10276,13 @@ struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev, i9xx_crtc_clock_get(intel_crtc, pipe_config); mode->clock = pipe_config->port_clock / pipe_config->pixel_multiplier; + + cpu_transcoder = pipe_config->cpu_transcoder; + htot = I915_READ(HTOTAL(cpu_transcoder)); + hsync = I915_READ(HSYNC(cpu_transcoder)); + vtot = I915_READ(VTOTAL(cpu_transcoder)); + vsync = I915_READ(VSYNC(cpu_transcoder)); + mode->hdisplay = (htot & 0xffff) + 1; mode->htotal = ((htot & 0xffff0000) >> 16) + 1; mode->hsync_start = (hsync & 0xffff) + 1; @@ -12359,7 +12363,6 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) struct drm_crtc_state *old_crtc_state, *new_crtc_state; struct drm_crtc *crtc; struct intel_crtc_state *intel_cstate; - bool hw_check = intel_state->modeset; u64 put_domains[I915_MAX_PIPES] = {}; unsigned crtc_vblank_mask = 0; int i; @@ -12376,7 +12379,6 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) if (needs_modeset(new_crtc_state) || to_intel_crtc_state(new_crtc_state)->update_pipe) { - hw_check = true; put_domains[to_intel_crtc(crtc)->pipe] = modeset_get_crtc_power_domains(crtc, @@ -14030,7 +14032,7 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb, if (mode_cmd->handles[i] != mode_cmd->handles[0]) { DRM_DEBUG_KMS("bad plane %d handle\n", i); - return -EINVAL; + goto err; } stride_alignment = intel_fb_stride_alignment(fb, i); diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 64134947c0aa..203198659ab2 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -2307,8 +2307,8 @@ static void edp_panel_off(struct intel_dp *intel_dp) I915_WRITE(pp_ctrl_reg, pp); POSTING_READ(pp_ctrl_reg); - intel_dp->panel_power_off_time = ktime_get_boottime(); wait_panel_off(intel_dp); + intel_dp->panel_power_off_time = ktime_get_boottime(); /* We got a reference when we enabled the VDD. */ intel_display_power_put(dev_priv, intel_dp->aux_power_domain); @@ -5273,7 +5273,7 @@ intel_dp_init_panel_power_sequencer(struct drm_device *dev, * seems sufficient to avoid this problem. */ if (dev_priv->quirks & QUIRK_INCREASE_T12_DELAY) { - vbt.t11_t12 = max_t(u16, vbt.t11_t12, 900 * 10); + vbt.t11_t12 = max_t(u16, vbt.t11_t12, 1300 * 10); DRM_DEBUG_KMS("Increasing T12 panel delay as per the quirk to %d\n", vbt.t11_t12); } diff --git a/drivers/gpu/drm/i915/intel_dpio_phy.c b/drivers/gpu/drm/i915/intel_dpio_phy.c index 09b670929786..de38d014ed39 100644 --- a/drivers/gpu/drm/i915/intel_dpio_phy.c +++ b/drivers/gpu/drm/i915/intel_dpio_phy.c @@ -208,12 +208,6 @@ static const struct bxt_ddi_phy_info glk_ddi_phy_info[] = { }, }; -static u32 bxt_phy_port_mask(const struct bxt_ddi_phy_info *phy_info) -{ - return (phy_info->dual_channel * BIT(phy_info->channel[DPIO_CH1].port)) | - BIT(phy_info->channel[DPIO_CH0].port); -} - static const struct bxt_ddi_phy_info * bxt_get_phy_list(struct drm_i915_private *dev_priv, int *count) { @@ -313,7 +307,6 @@ bool bxt_ddi_phy_is_enabled(struct drm_i915_private *dev_priv, enum dpio_phy phy) { const struct bxt_ddi_phy_info *phy_info; - enum port port; phy_info = bxt_get_phy_info(dev_priv, phy); @@ -335,19 +328,6 @@ bool bxt_ddi_phy_is_enabled(struct drm_i915_private *dev_priv, return false; } - for_each_port_masked(port, bxt_phy_port_mask(phy_info)) { - u32 tmp = I915_READ(BXT_PHY_CTL(port)); - - if (tmp & BXT_PHY_CMNLANE_POWERDOWN_ACK) { - DRM_DEBUG_DRIVER("DDI PHY %d powered, but common lane " - "for port %c powered down " - "(PHY_CTL %08x)\n", - phy, port_name(port), tmp); - - return false; - } - } - return true; } diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c index a2a3d93d67bd..df808a94c511 100644 --- a/drivers/gpu/drm/i915/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c @@ -1996,7 +1996,7 @@ static void cnl_ddi_pll_enable(struct drm_i915_private *dev_priv, /* 3. Configure DPLL_CFGCR0 */ /* Avoid touch CFGCR1 if HDMI mode is not enabled */ - if (pll->state.hw_state.cfgcr0 & DPLL_CTRL1_HDMI_MODE(pll->id)) { + if (pll->state.hw_state.cfgcr0 & DPLL_CFGCR0_HDMI_MODE) { val = pll->state.hw_state.cfgcr1; I915_WRITE(CNL_DPLL_CFGCR1(pll->id), val); /* 4. Reab back to ensure writes completed */ diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c index f0c11aec5ea5..7442891762be 100644 --- a/drivers/gpu/drm/i915/intel_dsi.c +++ b/drivers/gpu/drm/i915/intel_dsi.c @@ -892,8 +892,6 @@ static void intel_dsi_disable(struct intel_encoder *encoder, struct intel_crtc_state *old_crtc_state, struct drm_connector_state *old_conn_state) { - struct drm_device *dev = encoder->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); enum port port; @@ -903,15 +901,6 @@ static void intel_dsi_disable(struct intel_encoder *encoder, intel_panel_disable_backlight(old_conn_state); /* - * Disable Device ready before the port shutdown in order - * to avoid split screen - */ - if (IS_BROXTON(dev_priv)) { - for_each_dsi_port(port, intel_dsi->ports) - I915_WRITE(MIPI_DEVICE_READY(port), 0); - } - - /* * According to the spec we should send SHUTDOWN before * MIPI_SEQ_DISPLAY_OFF only for v3+ VBTs, but field testing * has shown that the v3 sequence works for v2 VBTs too diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index 9ab596941372..3c2d9cf22ed5 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -1048,9 +1048,12 @@ static int bxt_init_workarounds(struct intel_engine_cs *engine) } /* WaProgramL3SqcReg1DefaultForPerf:bxt */ - if (IS_BXT_REVID(dev_priv, BXT_REVID_B0, REVID_FOREVER)) - I915_WRITE(GEN8_L3SQCREG1, L3_GENERAL_PRIO_CREDITS(62) | - L3_HIGH_PRIO_CREDITS(2)); + if (IS_BXT_REVID(dev_priv, BXT_REVID_B0, REVID_FOREVER)) { + u32 val = I915_READ(GEN8_L3SQCREG1); + val &= ~L3_PRIO_CREDITS_MASK; + val |= L3_GENERAL_PRIO_CREDITS(62) | L3_HIGH_PRIO_CREDITS(2); + I915_WRITE(GEN8_L3SQCREG1, val); + } /* WaToEnableHwFixForPushConstHWBug:bxt */ if (IS_BXT_REVID(dev_priv, BXT_REVID_C0, REVID_FOREVER)) diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c index 951e834dd274..28a778b785ac 100644 --- a/drivers/gpu/drm/i915/intel_modes.c +++ b/drivers/gpu/drm/i915/intel_modes.c @@ -30,6 +30,21 @@ #include "intel_drv.h" #include "i915_drv.h" +static void intel_connector_update_eld_conn_type(struct drm_connector *connector) +{ + u8 conn_type; + + if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort || + connector->connector_type == DRM_MODE_CONNECTOR_eDP) { + conn_type = DRM_ELD_CONN_TYPE_DP; + } else { + conn_type = DRM_ELD_CONN_TYPE_HDMI; + } + + connector->eld[DRM_ELD_SAD_COUNT_CONN_TYPE] &= ~DRM_ELD_CONN_TYPE_MASK; + connector->eld[DRM_ELD_SAD_COUNT_CONN_TYPE] |= conn_type; +} + /** * intel_connector_update_modes - update connector from edid * @connector: DRM connector device to use @@ -44,6 +59,8 @@ int intel_connector_update_modes(struct drm_connector *connector, ret = drm_add_edid_modes(connector, edid); drm_edid_to_eld(connector, edid); + intel_connector_update_eld_conn_type(connector); + return ret; } diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index a17b1de7d7e0..3b1c5d783ee7 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -1699,6 +1699,8 @@ bxt_setup_backlight(struct intel_connector *connector, enum pipe unused) if (!panel->backlight.max) return -ENODEV; + panel->backlight.min = get_backlight_min_vbt(connector); + val = bxt_get_backlight(connector); val = intel_panel_compute_brightness(connector, val); panel->backlight.level = clamp(val, panel->backlight.min, @@ -1735,6 +1737,8 @@ cnp_setup_backlight(struct intel_connector *connector, enum pipe unused) if (!panel->backlight.max) return -ENODEV; + panel->backlight.min = get_backlight_min_vbt(connector); + val = bxt_get_backlight(connector); val = intel_panel_compute_brightness(connector, val); panel->backlight.level = clamp(val, panel->backlight.min, diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index ed662937ec3c..0a09f8ff6aff 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -8245,14 +8245,17 @@ static void gen8_set_l3sqc_credits(struct drm_i915_private *dev_priv, int high_prio_credits) { u32 misccpctl; + u32 val; /* WaTempDisableDOPClkGating:bdw */ misccpctl = I915_READ(GEN7_MISCCPCTL); I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE); - I915_WRITE(GEN8_L3SQCREG1, - L3_GENERAL_PRIO_CREDITS(general_prio_credits) | - L3_HIGH_PRIO_CREDITS(high_prio_credits)); + val = I915_READ(GEN8_L3SQCREG1); + val &= ~L3_PRIO_CREDITS_MASK; + val |= L3_GENERAL_PRIO_CREDITS(general_prio_credits); + val |= L3_HIGH_PRIO_CREDITS(high_prio_credits); + I915_WRITE(GEN8_L3SQCREG1, val); /* * Wait at least 100 clocks before re-enabling clock gating. diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index b66d8e136aa3..49577eba8e7e 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -368,7 +368,7 @@ static void hsw_power_well_enable(struct drm_i915_private *dev_priv, { enum i915_power_well_id id = power_well->id; bool wait_fuses = power_well->hsw.has_fuses; - enum skl_power_gate pg; + enum skl_power_gate uninitialized_var(pg); u32 val; if (wait_fuses) { @@ -2782,6 +2782,9 @@ static void cnl_display_core_init(struct drm_i915_private *dev_priv, bool resume /* 6. Enable DBUF */ gen9_dbuf_enable(dev_priv); + + if (resume && dev_priv->csr.dmc_payload) + intel_csr_load_program(dev_priv); } #undef CNL_PROCMON_IDX diff --git a/drivers/gpu/drm/i915/selftests/i915_random.c b/drivers/gpu/drm/i915/selftests/i915_random.c index d044bf9a6feb..222c511bea49 100644 --- a/drivers/gpu/drm/i915/selftests/i915_random.c +++ b/drivers/gpu/drm/i915/selftests/i915_random.c @@ -62,7 +62,7 @@ unsigned int *i915_random_order(unsigned int count, struct rnd_state *state) { unsigned int *order, i; - order = kmalloc_array(count, sizeof(*order), GFP_TEMPORARY); + order = kmalloc_array(count, sizeof(*order), GFP_KERNEL); if (!order) return order; diff --git a/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c b/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c index 7276194c04f7..828904b7d468 100644 --- a/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/selftests/intel_breadcrumbs.c @@ -117,12 +117,12 @@ static int igt_random_insert_remove(void *arg) mock_engine_reset(engine); - waiters = kvmalloc_array(count, sizeof(*waiters), GFP_TEMPORARY); + waiters = kvmalloc_array(count, sizeof(*waiters), GFP_KERNEL); if (!waiters) goto out_engines; bitmap = kcalloc(DIV_ROUND_UP(count, BITS_PER_LONG), sizeof(*bitmap), - GFP_TEMPORARY); + GFP_KERNEL); if (!bitmap) goto out_waiters; @@ -187,12 +187,12 @@ static int igt_insert_complete(void *arg) mock_engine_reset(engine); - waiters = kvmalloc_array(count, sizeof(*waiters), GFP_TEMPORARY); + waiters = kvmalloc_array(count, sizeof(*waiters), GFP_KERNEL); if (!waiters) goto out_engines; bitmap = kcalloc(DIV_ROUND_UP(count, BITS_PER_LONG), sizeof(*bitmap), - GFP_TEMPORARY); + GFP_KERNEL); if (!bitmap) goto out_waiters; @@ -368,7 +368,7 @@ static int igt_wakeup(void *arg) mock_engine_reset(engine); - waiters = kvmalloc_array(count, sizeof(*waiters), GFP_TEMPORARY); + waiters = kvmalloc_array(count, sizeof(*waiters), GFP_KERNEL); if (!waiters) goto out_engines; diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c index 2d0fef2cfca6..3cac22eb47ce 100644 --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c @@ -127,7 +127,7 @@ static int intel_uncore_check_forcewake_domains(struct drm_i915_private *dev_pri return 0; valid = kzalloc(BITS_TO_LONGS(FW_RANGE) * sizeof(*valid), - GFP_TEMPORARY); + GFP_KERNEL); if (!valid) return -ENOMEM; diff --git a/drivers/gpu/drm/lib/drm_random.c b/drivers/gpu/drm/lib/drm_random.c index 7b12a68c3b54..a78c4b483e8d 100644 --- a/drivers/gpu/drm/lib/drm_random.c +++ b/drivers/gpu/drm/lib/drm_random.c @@ -28,7 +28,7 @@ unsigned int *drm_random_order(unsigned int count, struct rnd_state *state) { unsigned int *order, i; - order = kmalloc_array(count, sizeof(*order), GFP_TEMPORARY); + order = kmalloc_array(count, sizeof(*order), GFP_KERNEL); if (!order) return order; diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index dbb31a014419..deaf869374ea 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -248,7 +248,7 @@ disable_clks: clk_disable_unprepare(ahb_clk); disable_gdsc: regulator_disable(gdsc_reg); - pm_runtime_put_autosuspend(dev); + pm_runtime_put_sync(dev); put_clk: clk_put(ahb_clk); put_gdsc: diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c index c2bdad88447e..824067d2d427 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c @@ -83,6 +83,8 @@ const struct mdp5_cfg_hw msm8x74v1_config = { .caps = MDP_LM_CAP_WB }, }, .nb_stages = 5, + .max_width = 2048, + .max_height = 0xFFFF, }, .dspp = { .count = 3, diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c index 6fcb58ab718c..440977677001 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c @@ -804,8 +804,6 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, spin_unlock_irqrestore(&mdp5_crtc->cursor.lock, flags); - pm_runtime_put_autosuspend(&pdev->dev); - set_cursor: ret = mdp5_ctl_set_cursor(ctl, pipeline, 0, cursor_enable); if (ret) { diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index f15821a0d900..ea5bb0e1632c 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -610,17 +610,6 @@ int msm_gem_sync_object(struct drm_gem_object *obj, struct dma_fence *fence; int i, ret; - if (!exclusive) { - /* NOTE: _reserve_shared() must happen before _add_shared_fence(), - * which makes this a slightly strange place to call it. OTOH this - * is a convenient can-fail point to hook it in. (And similar to - * how etnaviv and nouveau handle this.) - */ - ret = reservation_object_reserve_shared(msm_obj->resv); - if (ret) - return ret; - } - fobj = reservation_object_get_list(msm_obj->resv); if (!fobj || (fobj->shared_count == 0)) { fence = reservation_object_get_excl(msm_obj->resv); @@ -1045,10 +1034,10 @@ static void *_msm_gem_kernel_new(struct drm_device *dev, uint32_t size, } vaddr = msm_gem_get_vaddr(obj); - if (!vaddr) { + if (IS_ERR(vaddr)) { msm_gem_put_iova(obj, aspace); drm_gem_object_unreference(obj); - return ERR_PTR(-ENOMEM); + return ERR_CAST(vaddr); } if (bo) diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 8a75c0bd8a78..93535cac0676 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -40,7 +40,7 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev, if (sz > SIZE_MAX) return NULL; - submit = kmalloc(sz, GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY); + submit = kmalloc(sz, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); if (!submit) return NULL; @@ -221,7 +221,7 @@ fail: return ret; } -static int submit_fence_sync(struct msm_gem_submit *submit) +static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit) { int i, ret = 0; @@ -229,6 +229,20 @@ static int submit_fence_sync(struct msm_gem_submit *submit) struct msm_gem_object *msm_obj = submit->bos[i].obj; bool write = submit->bos[i].flags & MSM_SUBMIT_BO_WRITE; + if (!write) { + /* NOTE: _reserve_shared() must happen before + * _add_shared_fence(), which makes this a slightly + * strange place to call it. OTOH this is a + * convenient can-fail point to hook it in. + */ + ret = reservation_object_reserve_shared(msm_obj->resv); + if (ret) + return ret; + } + + if (no_implicit) + continue; + ret = msm_gem_sync_object(&msm_obj->base, submit->gpu->fctx, write); if (ret) break; @@ -451,11 +465,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (ret) goto out; - if (!(args->flags & MSM_SUBMIT_NO_IMPLICIT)) { - ret = submit_fence_sync(submit); - if (ret) - goto out; - } + ret = submit_fence_sync(submit, !!(args->flags & MSM_SUBMIT_NO_IMPLICIT)); + if (ret) + goto out; ret = submit_pin_objects(submit); if (ret) diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index ffbff27600e0..6a887032c66a 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -718,7 +718,8 @@ void msm_gpu_cleanup(struct msm_gpu *gpu) msm_gem_put_iova(gpu->rb->bo, gpu->aspace); msm_ringbuffer_destroy(gpu->rb); } - if (gpu->aspace) { + + if (!IS_ERR_OR_NULL(gpu->aspace)) { gpu->aspace->mmu->funcs->detach(gpu->aspace->mmu, NULL, 0); msm_gem_address_space_put(gpu->aspace); diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 0366b8092f97..ec56794ad039 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -111,10 +111,14 @@ static void rd_write(struct msm_rd_state *rd, const void *buf, int sz) wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0); + /* Note that smp_load_acquire() is not strictly required + * as CIRC_SPACE_TO_END() does not access the tail more + * than once. + */ n = min(sz, circ_space_to_end(&rd->fifo)); memcpy(fptr, ptr, n); - fifo->head = (fifo->head + n) & (BUF_SZ - 1); + smp_store_release(&fifo->head, (fifo->head + n) & (BUF_SZ - 1)); sz -= n; ptr += n; @@ -145,13 +149,17 @@ static ssize_t rd_read(struct file *file, char __user *buf, if (ret) goto out; + /* Note that smp_load_acquire() is not strictly required + * as CIRC_CNT_TO_END() does not access the head more than + * once. + */ n = min_t(int, sz, circ_count_to_end(&rd->fifo)); if (copy_to_user(buf, fptr, n)) { ret = -EFAULT; goto out; } - fifo->tail = (fifo->tail + n) & (BUF_SZ - 1); + smp_store_release(&fifo->tail, (fifo->tail + n) & (BUF_SZ - 1)); *ppos += n; wake_up_all(&rd->fifo_event); diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c index f7707849bb53..2b12d82aac15 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c +++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c @@ -223,7 +223,7 @@ void nouveau_fbcon_accel_save_disable(struct drm_device *dev) { struct nouveau_drm *drm = nouveau_drm(dev); - if (drm->fbcon) { + if (drm->fbcon && drm->fbcon->helper.fbdev) { drm->fbcon->saved_flags = drm->fbcon->helper.fbdev->flags; drm->fbcon->helper.fbdev->flags |= FBINFO_HWACCEL_DISABLED; } @@ -233,7 +233,7 @@ void nouveau_fbcon_accel_restore(struct drm_device *dev) { struct nouveau_drm *drm = nouveau_drm(dev); - if (drm->fbcon) { + if (drm->fbcon && drm->fbcon->helper.fbdev) { drm->fbcon->helper.fbdev->flags = drm->fbcon->saved_flags; } } @@ -245,7 +245,8 @@ nouveau_fbcon_accel_fini(struct drm_device *dev) struct nouveau_fbdev *fbcon = drm->fbcon; if (fbcon && drm->channel) { console_lock(); - fbcon->helper.fbdev->flags |= FBINFO_HWACCEL_DISABLED; + if (fbcon->helper.fbdev) + fbcon->helper.fbdev->flags |= FBINFO_HWACCEL_DISABLED; console_unlock(); nouveau_channel_idle(drm->channel); nvif_object_fini(&fbcon->twod); diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index 2dbf62a2ac41..e4751f92b342 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -3265,11 +3265,14 @@ nv50_mstm = { void nv50_mstm_service(struct nv50_mstm *mstm) { - struct drm_dp_aux *aux = mstm->mgr.aux; + struct drm_dp_aux *aux = mstm ? mstm->mgr.aux : NULL; bool handled = true; int ret; u8 esi[8] = {}; + if (!aux) + return; + while (handled) { ret = drm_dp_dpcd_read(aux, DP_SINK_COUNT_ESI, esi, 8); if (ret != 8) { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/bsp/g84.c b/drivers/gpu/drm/nouveau/nvkm/engine/bsp/g84.c index 8e2e24a74774..44e116f7880d 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/bsp/g84.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/bsp/g84.c @@ -39,5 +39,5 @@ int g84_bsp_new(struct nvkm_device *device, int index, struct nvkm_engine **pengine) { return nvkm_xtensa_new_(&g84_bsp, device, index, - true, 0x103000, pengine); + device->chipset != 0x92, 0x103000, pengine); } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/base.c index 4a57defc99b3..1399d923d446 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/base.c @@ -171,7 +171,7 @@ nvkm_gpio_fini(struct nvkm_subdev *subdev, bool suspend) return 0; } -static struct dmi_system_id gpio_reset_ids[] = { +static const struct dmi_system_id gpio_reset_ids[] = { { .ident = "Apple Macbook 10,1", .matches = { diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/base.c index d06ad2c372bf..455da298227f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/base.c @@ -241,6 +241,8 @@ nvkm_vm_unmap_pgt(struct nvkm_vm *vm, int big, u32 fpde, u32 lpde) mmu->func->map_pgt(vpgd->obj, pde, vpgt->mem); } + mmu->func->flush(vm); + nvkm_memory_del(&pgt); } } diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 14c5613b4388..afbf50d0c08f 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -509,23 +509,25 @@ static void qxl_primary_atomic_update(struct drm_plane *plane, .y2 = qfb->base.height }; - if (!old_state->fb) { - qxl_io_log(qdev, - "create primary fb: %dx%d,%d,%d\n", - bo->surf.width, bo->surf.height, - bo->surf.stride, bo->surf.format); + if (old_state->fb) { + qfb_old = to_qxl_framebuffer(old_state->fb); + bo_old = gem_to_qxl_bo(qfb_old->obj); + } else { + bo_old = NULL; + } - qxl_io_create_primary(qdev, 0, bo); - bo->is_primary = true; + if (bo == bo_old) return; - } else { - qfb_old = to_qxl_framebuffer(old_state->fb); - bo_old = gem_to_qxl_bo(qfb_old->obj); + if (bo_old && bo_old->is_primary) { + qxl_io_destroy_primary(qdev); bo_old->is_primary = false; } - bo->is_primary = true; + if (!bo->is_primary) { + qxl_io_create_primary(qdev, 0, bo); + bo->is_primary = true; + } qxl_draw_dirty_fb(qdev, qfb, bo, 0, 0, &norect, 1, 1); } @@ -534,13 +536,15 @@ static void qxl_primary_atomic_disable(struct drm_plane *plane, { struct qxl_device *qdev = plane->dev->dev_private; - if (old_state->fb) - { struct qxl_framebuffer *qfb = + if (old_state->fb) { + struct qxl_framebuffer *qfb = to_qxl_framebuffer(old_state->fb); struct qxl_bo *bo = gem_to_qxl_bo(qfb->obj); - qxl_io_destroy_primary(qdev); - bo->is_primary = false; + if (bo->is_primary) { + qxl_io_destroy_primary(qdev); + bo->is_primary = false; + } } } @@ -698,14 +702,15 @@ static void qxl_plane_cleanup_fb(struct drm_plane *plane, struct drm_gem_object *obj; struct qxl_bo *user_bo; - if (!plane->state->fb) { - /* we never executed prepare_fb, so there's nothing to + if (!old_state->fb) { + /* + * we never executed prepare_fb, so there's nothing to * unpin. */ return; } - obj = to_qxl_framebuffer(plane->state->fb)->obj; + obj = to_qxl_framebuffer(old_state->fb)->obj; user_bo = gem_to_qxl_bo(obj); qxl_bo_unpin(user_bo); } diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 997131d58c7f..ffc10cadcf34 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1663,7 +1663,7 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, radeon_agp_suspend(rdev); pci_save_state(dev->pdev); - if (freeze && rdev->family >= CHIP_CEDAR) { + if (freeze && rdev->family >= CHIP_CEDAR && !(rdev->flags & RADEON_IS_IGP)) { rdev->asic->asic_reset(rdev, true); pci_restore_state(dev->pdev); } else if (suspend) { diff --git a/drivers/gpu/drm/selftests/test-drm_mm.c b/drivers/gpu/drm/selftests/test-drm_mm.c index dfdd858eda0a..86eb4c185a28 100644 --- a/drivers/gpu/drm/selftests/test-drm_mm.c +++ b/drivers/gpu/drm/selftests/test-drm_mm.c @@ -1627,7 +1627,7 @@ static int igt_topdown(void *ignored) goto err; bitmap = kzalloc(count / BITS_PER_LONG * sizeof(unsigned long), - GFP_TEMPORARY); + GFP_KERNEL); if (!bitmap) goto err_nodes; @@ -1741,7 +1741,7 @@ static int igt_bottomup(void *ignored) goto err; bitmap = kzalloc(count / BITS_PER_LONG * sizeof(unsigned long), - GFP_TEMPORARY); + GFP_KERNEL); if (!bitmap) goto err_nodes; diff --git a/drivers/gpu/drm/sun4i/Kconfig b/drivers/gpu/drm/sun4i/Kconfig index 06f05302ee75..882d85db9053 100644 --- a/drivers/gpu/drm/sun4i/Kconfig +++ b/drivers/gpu/drm/sun4i/Kconfig @@ -26,7 +26,7 @@ config DRM_SUN4I_HDMI_CEC bool "Allwinner A10 HDMI CEC Support" depends on DRM_SUN4I_HDMI select CEC_CORE - depends on CEC_PIN + select CEC_PIN help Choose this option if you have an Allwinner SoC with an HDMI controller and want to use CEC. diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi.h b/drivers/gpu/drm/sun4i/sun4i_hdmi.h index 1457750988da..a1f8cba251a2 100644 --- a/drivers/gpu/drm/sun4i/sun4i_hdmi.h +++ b/drivers/gpu/drm/sun4i/sun4i_hdmi.h @@ -15,7 +15,7 @@ #include <drm/drm_connector.h> #include <drm/drm_encoder.h> -#include <media/cec.h> +#include <media/cec-pin.h> #define SUN4I_HDMI_CTRL_REG 0x004 #define SUN4I_HDMI_CTRL_ENABLE BIT(31) diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c index 9ea6cd5a1370..3cf1a6932fac 100644 --- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c +++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c @@ -302,26 +302,29 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master, hdmi->mod_clk = devm_clk_get(dev, "mod"); if (IS_ERR(hdmi->mod_clk)) { dev_err(dev, "Couldn't get the HDMI mod clock\n"); - return PTR_ERR(hdmi->mod_clk); + ret = PTR_ERR(hdmi->mod_clk); + goto err_disable_bus_clk; } clk_prepare_enable(hdmi->mod_clk); hdmi->pll0_clk = devm_clk_get(dev, "pll-0"); if (IS_ERR(hdmi->pll0_clk)) { dev_err(dev, "Couldn't get the HDMI PLL 0 clock\n"); - return PTR_ERR(hdmi->pll0_clk); + ret = PTR_ERR(hdmi->pll0_clk); + goto err_disable_mod_clk; } hdmi->pll1_clk = devm_clk_get(dev, "pll-1"); if (IS_ERR(hdmi->pll1_clk)) { dev_err(dev, "Couldn't get the HDMI PLL 1 clock\n"); - return PTR_ERR(hdmi->pll1_clk); + ret = PTR_ERR(hdmi->pll1_clk); + goto err_disable_mod_clk; } ret = sun4i_tmds_create(hdmi); if (ret) { dev_err(dev, "Couldn't create the TMDS clock\n"); - return ret; + goto err_disable_mod_clk; } writel(SUN4I_HDMI_CTRL_ENABLE, hdmi->base + SUN4I_HDMI_CTRL_REG); @@ -362,7 +365,7 @@ static int sun4i_hdmi_bind(struct device *dev, struct device *master, ret = sun4i_hdmi_i2c_create(dev, hdmi); if (ret) { dev_err(dev, "Couldn't create the HDMI I2C adapter\n"); - return ret; + goto err_disable_mod_clk; } drm_encoder_helper_add(&hdmi->encoder, @@ -422,6 +425,10 @@ err_cleanup_connector: drm_encoder_cleanup(&hdmi->encoder); err_del_i2c_adapter: i2c_del_adapter(hdmi->i2c); +err_disable_mod_clk: + clk_disable_unprepare(hdmi->mod_clk); +err_disable_bus_clk: + clk_disable_unprepare(hdmi->bus_clk); return ret; } @@ -434,6 +441,8 @@ static void sun4i_hdmi_unbind(struct device *dev, struct device *master, drm_connector_cleanup(&hdmi->connector); drm_encoder_cleanup(&hdmi->encoder); i2c_del_adapter(hdmi->i2c); + clk_disable_unprepare(hdmi->mod_clk); + clk_disable_unprepare(hdmi->bus_clk); } static const struct component_ops sun4i_hdmi_ops = { diff --git a/drivers/gpu/drm/tegra/trace.h b/drivers/gpu/drm/tegra/trace.h index e9b7cdad5c4c..5a1ab4046e92 100644 --- a/drivers/gpu/drm/tegra/trace.h +++ b/drivers/gpu/drm/tegra/trace.h @@ -63,6 +63,6 @@ DEFINE_EVENT(register_access, sor_readl, /* This part must be outside protection */ #undef TRACE_INCLUDE_PATH -#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/tegra #define TRACE_INCLUDE_FILE trace #include <trace/define_trace.h> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index cba11f13d994..180ce6296416 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -109,8 +109,8 @@ static ssize_t ttm_bo_global_show(struct kobject *kobj, struct ttm_bo_global *glob = container_of(kobj, struct ttm_bo_global, kobj); - return snprintf(buffer, PAGE_SIZE, "%lu\n", - (unsigned long) atomic_read(&glob->bo_count)); + return snprintf(buffer, PAGE_SIZE, "%d\n", + atomic_read(&glob->bo_count)); } static struct attribute *ttm_bo_global_attrs[] = { diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index d0459b392e5e..c934ad5b3903 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -469,6 +469,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, * TODO: Explicit member copy would probably be better here. */ + atomic_inc(&bo->glob->bo_count); INIT_LIST_HEAD(&fbo->ddestroy); INIT_LIST_HEAD(&fbo->lru); INIT_LIST_HEAD(&fbo->swap); diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c index 6a573d21d3cc..658fa2d3e40c 100644 --- a/drivers/gpu/ipu-v3/ipu-common.c +++ b/drivers/gpu/ipu-v3/ipu-common.c @@ -405,6 +405,14 @@ int ipu_idmac_lock_enable(struct ipuv3_channel *channel, int num_bursts) return -EINVAL; } + /* + * IPUv3EX / i.MX51 has a different register layout, and on IPUv3M / + * i.MX53 channel arbitration locking doesn't seem to work properly. + * Allow enabling the lock feature on IPUv3H / i.MX6 only. + */ + if (bursts && ipu->ipu_type != IPUV3H) + return -EINVAL; + for (i = 0; i < ARRAY_SIZE(idmac_lock_en_info); i++) { if (channel->num == idmac_lock_en_info[i].chnum) break; diff --git a/drivers/gpu/ipu-v3/ipu-pre.c b/drivers/gpu/ipu-v3/ipu-pre.c index c35f74c83065..c860a7997cb5 100644 --- a/drivers/gpu/ipu-v3/ipu-pre.c +++ b/drivers/gpu/ipu-v3/ipu-pre.c @@ -73,6 +73,14 @@ #define IPU_PRE_STORE_ENG_CTRL_WR_NUM_BYTES(v) ((v & 0x7) << 1) #define IPU_PRE_STORE_ENG_CTRL_OUTPUT_ACTIVE_BPP(v) ((v & 0x3) << 4) +#define IPU_PRE_STORE_ENG_STATUS 0x120 +#define IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_X_MASK 0xffff +#define IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_X_SHIFT 0 +#define IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_Y_MASK 0x3fff +#define IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_Y_SHIFT 16 +#define IPU_PRE_STORE_ENG_STATUS_STORE_FIFO_FULL (1 << 30) +#define IPU_PRE_STORE_ENG_STATUS_STORE_FIELD (1 << 31) + #define IPU_PRE_STORE_ENG_SIZE 0x130 #define IPU_PRE_STORE_ENG_SIZE_INPUT_WIDTH(v) ((v & 0xffff) << 0) #define IPU_PRE_STORE_ENG_SIZE_INPUT_HEIGHT(v) ((v & 0xffff) << 16) @@ -93,6 +101,7 @@ struct ipu_pre { dma_addr_t buffer_paddr; void *buffer_virt; bool in_use; + unsigned int safe_window_end; }; static DEFINE_MUTEX(ipu_pre_list_mutex); @@ -160,6 +169,9 @@ void ipu_pre_configure(struct ipu_pre *pre, unsigned int width, u32 active_bpp = info->cpp[0] >> 1; u32 val; + /* calculate safe window for ctrl register updates */ + pre->safe_window_end = height - 2; + writel(bufaddr, pre->regs + IPU_PRE_CUR_BUF); writel(bufaddr, pre->regs + IPU_PRE_NEXT_BUF); @@ -199,7 +211,24 @@ void ipu_pre_configure(struct ipu_pre *pre, unsigned int width, void ipu_pre_update(struct ipu_pre *pre, unsigned int bufaddr) { + unsigned long timeout = jiffies + msecs_to_jiffies(5); + unsigned short current_yblock; + u32 val; + writel(bufaddr, pre->regs + IPU_PRE_NEXT_BUF); + + do { + if (time_after(jiffies, timeout)) { + dev_warn(pre->dev, "timeout waiting for PRE safe window\n"); + return; + } + + val = readl(pre->regs + IPU_PRE_STORE_ENG_STATUS); + current_yblock = + (val >> IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_Y_SHIFT) & + IPU_PRE_STORE_ENG_STATUS_STORE_BLOCK_Y_MASK; + } while (current_yblock == 0 || current_yblock >= pre->safe_window_end); + writel(IPU_PRE_CTRL_SDW_UPDATE, pre->regs + IPU_PRE_CTRL_SET); } diff --git a/drivers/gpu/ipu-v3/ipu-prg.c b/drivers/gpu/ipu-v3/ipu-prg.c index ecc9ea44dc50..0013ca9f72c8 100644 --- a/drivers/gpu/ipu-v3/ipu-prg.c +++ b/drivers/gpu/ipu-v3/ipu-prg.c @@ -14,6 +14,7 @@ #include <drm/drm_fourcc.h> #include <linux/clk.h> #include <linux/err.h> +#include <linux/iopoll.h> #include <linux/mfd/syscon.h> #include <linux/mfd/syscon/imx6q-iomuxc-gpr.h> #include <linux/module.h> @@ -329,6 +330,12 @@ int ipu_prg_channel_configure(struct ipuv3_channel *ipu_chan, val = IPU_PRG_REG_UPDATE_REG_UPDATE; writel(val, prg->regs + IPU_PRG_REG_UPDATE); + /* wait for both double buffers to be filled */ + readl_poll_timeout(prg->regs + IPU_PRG_STATUS, val, + (val & IPU_PRG_STATUS_BUFFER0_READY(prg_chan)) && + (val & IPU_PRG_STATUS_BUFFER1_READY(prg_chan)), + 5, 1000); + clk_disable_unprepare(prg->clk_ipg); chan->enabled = true; |
