drm/amdgpu: use all SDMA instances for TTM clears and moves
Subsystems (5)
v4 cover drm/amdgpu: use all SDMA instances for TTM clears and moves
The drm/ttm patch modifies TTM to support multiple contexts for the pipelined moves.
Then amdgpu/ttm is updated to express dependencies between jobs explicitely, instead of relying on the ordering of execution guaranteed by the use of a single instance.
With all of this in place, we can use multiple entities, with each having access to the available SDMA instances.
This rework also gives the opportunity to merge the clear functions into a single one and to optimize a bit GART usage.
Since v3 some patches have been already reviewed and merged separately:
- https://lists.freedesktop.org/archives/amd-gfx/2026-January/137747.html
- https://gitlab.freedesktop.org/drm/kernel/-/commit/ddf055b80a544d6f36f77be5f0c6d3c80177d57c
This version depend on them.
v3: https://lists.freedesktop.org/archives/dri-devel/2025-November/537830.html
Pierre-Eric Pelloux-Prayer (12):
drm/amdgpu: allocate clear entities dynamically drm/amdgpu: allocate move entities dynamically drm/amdgpu: round robin through clear_entities in amdgpu_fill_buffer drm/amdgpu: use TTM_NUM_MOVE_FENCES when reserving fences drm/amdgpu: use multiple entities in amdgpu_move_blit drm/amdgpu: pass all the sdma scheds to amdgpu_mman drm/amdgpu: only use working sdma schedulers for ttm drm/amdgpu: create multiple clear/move ttm entities drm/amdgpu: give ttm entities access to all the sdma scheds drm/amdgpu: get rid of amdgpu_ttm_clear_buffer drm/amdgpu: rename amdgpu_fill_buffer as amdgpu_ttm_clear_buffer drm/amdgpu: split amdgpu_ttm_set_buffer_funcs_status in 2 funcs
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 4 +- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 17 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 329 ++++++++++-------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 29 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 6 +- drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 13 +- drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 15 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 12 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 11 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 14 +- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 12 +- drivers/gpu/drm/amd/amdgpu/si_dma.c | 12 +- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 5 +- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 +- .../amd/display/amdgpu_dm/amdgpu_dm_plane.c | 6 +- .../drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c | 6 +- 23 files changed, 300 insertions(+), 243 deletions(-)
Signature · 1 line
2.43.0
PATCH v4 1/12 drm/amdgpu: allocate clear entities dynamically
No functional change for now, as we always allocate a single entity and use it everywhere.
---
v4: stop using adev->sdma.num_instances
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 55 +++++++++++++++------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 3 +- 3 files changed, 42 insertions(+), 18 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +1 -1
@@ -1325,7 +1325,7 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)if (r) goto out; - r =amdgpu_fill_buffer (&adev->mman.clear_entity, abo,0 , &bo->base._resv, + r =amdgpu_fill_buffer (&adev->mman.clear_entities[0 ], abo,0 , &bo->base._resv, &fence,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE );if (WARN_ON (r)) goto out;
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +39 -16
103 lines not on this page · raw message
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -1
@@ -72,8 +72,9 @@ struct amdgpu_mman {/* @default_entity: for workarounds, has no gart windows */ struct amdgpu_ttm_buffer_entity default_entity; -struct amdgpu_ttm_buffer_entity clear_entity;struct amdgpu_ttm_buffer_entity move_entity; +struct amdgpu_ttm_buffer_entity *clear_entities; +u32 num_clear_entities;struct amdgpu_vram_mgr vram_mgr;struct amdgpu_gtt_mgr gtt_mgr;
Signature · 1 line
2.43.0
PATCH v4 2/12 drm/amdgpu: allocate move entities dynamically
In reply to Pierre-Eric Pelloux-PrayerNo functional change for now, as we always allocate a single entity.
---
v4: stop using adev->sdma.num_instances
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 35 +++++++++++++++--------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 3 +- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 2 +- 3 files changed, 25 insertions(+), 15 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +22 -13
@@ -399,7 +399,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, dst.offset =0 ; r = amdgpu_ttm_copy_mem_to_mem(adev, - &adev->mman.move_entity, + &adev->mman.move_entities[0], &src, &dst, new_mem->size, amdgpu_bo_encrypted(abo),
@@ -412,7 +412,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, (abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) {struct dma_fence *wipe_fence =NULL ; - r =amdgpu_fill_buffer (&adev->mman.move_entity, + r =amdgpu_fill_buffer (&adev->mman.move_entities[0 ], abo,0 ,NULL , &wipe_fence,AMDGPU_KERNEL_JOB_ID_MOVE_BLIT );if (r) {
@@ -2346,7 +2346,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)void amdgpu_ttm_set_buffer_funcs_status (struct amdgpu_device *adev,bool enable) {struct ttm_resource_manager *man =ttm_manager_type (&adev->mman.bdev,TTM_PL_VRAM ); -u32 num_clear_entities; +u32 num_clear_entities, num_move_entities;uint64_t size;int r, i, j;
@@ -2364,6 +2364,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) } num_clear_entities =1 ; + num_move_entities =1 ; ring = adev->mman.buffer_funcs_ring; sched = &ring->sched; r = amdgpu_ttm_buffer_entity_init(&adev->mman.gtt_mgr,
@@ -2400,14 +2401,20 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) } } - r =amdgpu_ttm_buffer_entity_init (&adev->mman.gtt_mgr, - &adev->mman.move_entity, -DRM_SCHED_PRIORITY_NORMAL , - &sched,1 ,2 ); -if (r <0 ) { -dev_err (adev->dev, -"Failed setting up TTM BO move entity (%d)\n" , r); - goto error_free_clear_entities; + adev->mman.num_move_entities = num_move_entities; +for (i =0 ; i < num_move_entities; i++) { + r =amdgpu_ttm_buffer_entity_init ( + &adev->mman.gtt_mgr, + &adev->mman.move_entities[i], +DRM_SCHED_PRIORITY_NORMAL , &sched,1 ,2 ); + +if (r <0 ) { +for (j =0 ; j < i; j++) +amdgpu_ttm_buffer_entity_fini ( + &adev->mman.gtt_mgr, &adev->mman.move_entities[j]); + adev->mman.num_move_entities =0 ; + goto error_free_clear_entities; + } } }else { amdgpu_ttm_buffer_entity_fini(&adev->mman.gtt_mgr,
@@ -2415,8 +2422,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)for (i =0 ; i < adev->mman.num_clear_entities; i++)amdgpu_ttm_buffer_entity_fini (&adev->mman.gtt_mgr, &adev->mman.clear_entities[i]); -amdgpu_ttm_buffer_entity_fini (&adev->mman.gtt_mgr, - &adev->mman.move_entity); +for (i =0 ; i < adev->mman.num_move_entities; i++) +amdgpu_ttm_buffer_entity_fini (&adev->mman.gtt_mgr, + &adev->mman.move_entities[i]);/* Drop all the old fences since re-creating the scheduler entities * will allocate new contexts. */
@@ -2424,6 +2432,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)kfree (adev->mman.clear_entities); adev->mman.clear_entities =NULL ; adev->mman.num_clear_entities =0 ; + adev->mman.num_move_entities =0 ; }/* this just adjusts TTM size idea, which sets lpfn to the correct value */
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -1
@@ -72,9 +72,10 @@ struct amdgpu_mman {/* @default_entity: for workarounds, has no gart windows */ struct amdgpu_ttm_buffer_entity default_entity; -struct amdgpu_ttm_buffer_entity move_entity;struct amdgpu_ttm_buffer_entity *clear_entities;u32 num_clear_entities; +struct amdgpu_ttm_buffer_entity move_entities[TTM_NUM_MOVE_FENCES ]; +u32 num_move_entities;struct amdgpu_vram_mgr vram_mgr;struct amdgpu_gtt_mgr gtt_mgr;
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +1 -1
@@ -136,7 +136,7 @@ svm_migrate_copy_memory_gart(struct amdgpu_device *adev, dma_addr_t *sys,u64 size;int r; - entity = &adev->mman.move_entity; + entity = &adev->mman.move_entities[0 ];mutex_lock (&entity->lock);
Signature · 1 line
2.43.0
PATCH v4 3/12 drm/amdgpu: round robin through clear_entities in amdgpu_fill_buffer
In reply to Pierre-Eric Pelloux-PrayerThis makes clear of different BOs run in parallel. Partial jobs to clear a single BO still execute sequentially.
---
v4:
- check entity's validity
- use u32 for the index in the entities array
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 20 ++++++++++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 2 ++ 3 files changed, 20 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +2 -1
@@ -1325,7 +1325,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)if (r) goto out; - r =amdgpu_fill_buffer (&adev->mman.clear_entities[0 ], abo,0 , &bo->base._resv, + r =amdgpu_fill_buffer (amdgpu_ttm_next_clear_entity (adev), + abo,0 , &bo->base._resv, &fence,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE );if (WARN_ON (r)) goto out;
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +16 -4
@@ -2380,6 +2380,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) adev->mman.clear_entities =kcalloc (num_clear_entities,sizeof (struct amdgpu_ttm_buffer_entity ),GFP_KERNEL ); +atomic_set (&adev->mman.next_clear_entity,0 );if (!adev->mman.clear_entities) goto error_free_default_entity;
@@ -2653,11 +2654,8 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_buffer_entity *entity,struct amdgpu_res_cursor dst;int r; -if (!adev->mman.buffer_funcs_enabled) { -dev_err (adev->dev, -"Trying to clear memory with ring turned off.\n" ); +if (!entity)return -EINVAL ; - }amdgpu_res_first (bo->tbo.resource,0 ,amdgpu_bo_size (bo), &dst);
@@ -2693,6 +2691,20 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_buffer_entity *entity,return r; } +struct amdgpu_ttm_buffer_entity * +amdgpu_ttm_next_clear_entity (struct amdgpu_device *adev) +{ +struct amdgpu_mman *mman = &adev->mman; +u32 i; + +if (mman->num_clear_entities ==0 ) +return NULL ; + + i =atomic_inc_return (&mman->next_clear_entity) % + mman->num_clear_entities; +return &mman->clear_entities[i]; +} +/** * amdgpu_ttm_evict_resources - evict memory buffers * @adev: amdgpu device object
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -0
@@ -73,6 +73,7 @@ struct amdgpu_mman {/* @default_entity: for workarounds, has no gart windows */ struct amdgpu_ttm_buffer_entity default_entity;struct amdgpu_ttm_buffer_entity *clear_entities; +atomic_t next_clear_entity;u32 num_clear_entities;struct amdgpu_ttm_buffer_entity move_entities[TTM_NUM_MOVE_FENCES ];u32 num_move_entities;
@@ -193,6 +194,7 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_buffer_entity *entity,struct dma_resv *resv, struct dma_fence **f, u64 k_job_id); +struct amdgpu_ttm_buffer_entity *amdgpu_ttm_next_clear_entity (struct amdgpu_device *adev);int amdgpu_ttm_alloc_gart (struct ttm_buffer_object *bo);void amdgpu_ttm_recover_gart (struct ttm_buffer_object *tbo);
Signature · 1 line
2.43.0
PATCH v4 4/12 drm/amdgpu: use TTM_NUM_MOVE_FENCES when reserving fences
In reply to Pierre-Eric Pelloux-PrayerUse TTM_NUM_MOVE_FENCES as an upperbound of how many fences ttm might need to deal with moves/evictions.
---
v2: removed drm_err calls
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 5 ++--- drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 6 ++---- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 +-- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c | 6 ++---- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c | 6 ++---- 5 files changed, 9 insertions(+), 17 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +2 -3
@@ -916,9 +916,8 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, goto out_free_user_pages;amdgpu_bo_list_for_each_entry (e, p->bo_list) { -/* One fence for TTM and one for each CS job */ r =drm_exec_prepare_obj (&p->exec, &e->bo->tbo.base, -1 + p->gang_size); +TTM_NUM_MOVE_FENCES + p->gang_size);drm_exec_retry_on_contention (&p->exec);if (unlikely (r)) goto out_free_user_pages;
@@ -928,7 +927,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,if (p->uf_bo) { r =drm_exec_prepare_obj (&p->exec, &p->uf_bo->tbo.base, -1 + p->gang_size); +TTM_NUM_MOVE_FENCES + p->gang_size);drm_exec_retry_on_contention (&p->exec);if (unlikely (r)) goto out_free_user_pages;
drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +2 -4
@@ -328,11 +328,9 @@ static int amdgpu_vkms_prepare_fb(struct drm_plane *plane,return r; } - r =dma_resv_reserve_fences (rbo->tbo.base.resv,1 ); -if (r) { -dev_err (adev->dev,"allocating fence slot failed (%d)\n" , r); + r =dma_resv_reserve_fences (rbo->tbo.base.resv,TTM_NUM_MOVE_FENCES ); +if (r) goto error_unlock; - }if (plane->type !=DRM_PLANE_TYPE_CURSOR ) domain =amdgpu_display_supported_domains (adev, rbo->flags);
drivers/gpu/drm/amd/amdkfd/kfd_svm.c +1 -2
@@ -628,9 +628,8 @@ svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange, } } - r =dma_resv_reserve_fences (bo->tbo.base.resv,1 ); + r =dma_resv_reserve_fences (bo->tbo.base.resv,TTM_NUM_MOVE_FENCES );if (r) { -pr_debug ("failed %d to reserve bo\n" , r);amdgpu_bo_unreserve (bo); goto reserve_bo_failed; }
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +2 -4
@@ -952,11 +952,9 @@ static int amdgpu_dm_plane_helper_prepare_fb(struct drm_plane *plane,return r; } - r =dma_resv_reserve_fences (rbo->tbo.base.resv,1 ); -if (r) { -drm_err (adev_to_drm (adev),"reserving fence slot failed (%d)\n" , r); + r =dma_resv_reserve_fences (rbo->tbo.base.resv,TTM_NUM_MOVE_FENCES ); +if (r) goto error_unlock; - }if (plane->type !=DRM_PLANE_TYPE_CURSOR ) domain =amdgpu_display_supported_domains (adev, rbo->flags);
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c +2 -4
@@ -106,11 +106,9 @@ static int amdgpu_dm_wb_prepare_job(struct drm_writeback_connector *wb_connectorreturn r; } - r =dma_resv_reserve_fences (rbo->tbo.base.resv,1 ); -if (r) { -drm_err (adev_to_drm (adev),"reserving fence slot failed (%d)\n" , r); + r =dma_resv_reserve_fences (rbo->tbo.base.resv,TTM_NUM_MOVE_FENCES ); +if (r) goto error_unlock; - } domain =amdgpu_display_supported_domains (adev, rbo->flags);
Signature · 1 line
2.43.0
PATCH v4 5/12 drm/amdgpu: use multiple entities in amdgpu_move_blit
In reply to Pierre-Eric Pelloux-PrayerThanks to "drm/ttm: rework pipelined eviction fence handling", ttm can deal correctly with moves and evictions being executed from different contexts.
---
v4: use u32 for the index in the entities array
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 13 +++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 1 + 2 files changed, 10 insertions(+), 4 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +9 -4
@@ -387,9 +387,11 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, {struct amdgpu_device *adev =amdgpu_ttm_adev (bo->bdev);struct amdgpu_bo *abo =ttm_to_amdgpu_bo (bo); +struct amdgpu_ttm_buffer_entity *entity;struct amdgpu_copy_mem src, dst;struct dma_fence *fence =NULL ;int r; +u32 e; src.bo = bo; dst.bo = bo;
@@ -398,8 +400,12 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, src.offset =0 ; dst.offset =0 ; + e =atomic_inc_return (&adev->mman.next_move_entity) % + adev->mman.num_move_entities; + entity = &adev->mman.move_entities[e]; + r = amdgpu_ttm_copy_mem_to_mem(adev, - &adev->mman.move_entities[0], + entity, &src, &dst, new_mem->size, amdgpu_bo_encrypted(abo),
@@ -411,9 +417,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,if (old_mem->mem_type ==TTM_PL_VRAM && (abo->flags &AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE )) {struct dma_fence *wipe_fence =NULL ; - - r =amdgpu_fill_buffer (&adev->mman.move_entities[0 ], - abo,0 ,NULL , &wipe_fence, + r =amdgpu_fill_buffer (entity, abo,0 ,NULL , &wipe_fence,AMDGPU_KERNEL_JOB_ID_MOVE_BLIT );if (r) { goto error;
@@ -2403,6 +2407,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) } adev->mman.num_move_entities = num_move_entities; +atomic_set (&adev->mman.next_move_entity,0 );for (i =0 ; i < num_move_entities; i++) { r = amdgpu_ttm_buffer_entity_init( &adev->mman.gtt_mgr,
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +1 -0
@@ -76,6 +76,7 @@ struct amdgpu_mman {atomic_t next_clear_entity;u32 num_clear_entities;struct amdgpu_ttm_buffer_entity move_entities[TTM_NUM_MOVE_FENCES ]; +atomic_t next_move_entity;u32 num_move_entities;struct amdgpu_vram_mgr vram_mgr;
Signature · 1 line
2.43.0
PATCH v4 6/12 drm/amdgpu: pass all the sdma scheds to amdgpu_mman
In reply to Pierre-Eric Pelloux-PrayerThis will allow the use of all of them for clear/fill buffer operations.
Since drm_sched_entity_init requires a scheduler array, we store schedulers rather than rings. For the few places that need access to a ring, we can get it from the sched using container_of.
Since the code is the same for all sdma versions, add a new helper amdgpu_sdma_set_buffer_funcs_scheds to set buffer_funcs_scheds based on the number of sdma instances.
Note: the new sched array is identical to the amdgpu_vm_manager one. These 2 could be merged.
--- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 4 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 32 ++++++++++++++++++---- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 3 +- drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 6 +--- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 6 +--- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 5 +--- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 5 +--- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 3 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 3 +- drivers/gpu/drm/amd/amdgpu/si_dma.c | 3 +- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 3 +- 17 files changed, 47 insertions(+), 42 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu.h +2 -0
@@ -1538,6 +1538,8 @@ ssize_t amdgpu_get_soft_full_reset_mask(struct amdgpu_ring *ring);ssize_t amdgpu_show_reset_mask (char *buf,uint32_t supported_reset);void amdgpu_sdma_set_vm_pte_scheds (struct amdgpu_device *adev,const struct amdgpu_vm_pte_funcs *vm_pte_funcs); +void amdgpu_sdma_set_buffer_funcs_scheds (struct amdgpu_device *adev, +const struct amdgpu_buffer_funcs *buffer_funcs);/* atpx handler */ #if defined(CONFIG_VGA_SWITCHEROO )
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +1 -1
@@ -4390,7 +4390,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, adev->num_rings =0 ;RCU_INIT_POINTER (adev->gang_submit,dma_fence_get_stub ()); adev->mman.buffer_funcs =NULL ; - adev->mman.buffer_funcs_ring =NULL ; + adev->mman.num_buffer_funcs_scheds =0 ; adev->vm_manager.vm_pte_funcs =NULL ; adev->vm_manager.vm_pte_num_scheds =0 ; adev->gmc.gmc_funcs =NULL ;
drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +3 -1
@@ -707,12 +707,14 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)void amdgpu_gmc_flush_gpu_tlb (struct amdgpu_device *adev,uint32_t vmid,uint32_t vmhub,uint32_t flush_type) { -struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; +struct amdgpu_ring *ring;struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];struct dma_fence *fence;struct amdgpu_job *job;int r; + ring =to_amdgpu_ring (adev->mman.buffer_funcs_scheds[0 ]); +if (!hub->sdma_invalidation_workaround || vmid || !adev->mman.buffer_funcs_enabled || !adev->ib_pool_ready || !ring->sched.ready) {
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +26 -6
@@ -168,7 +168,7 @@ amdgpu_ttm_job_submit(struct amdgpu_device *adev, struct amdgpu_ttm_buffer_entit {struct amdgpu_ring *ring; - ring = adev->mman.buffer_funcs_ring; + ring =to_amdgpu_ring (adev->mman.buffer_funcs_scheds[0 ]);amdgpu_ring_pad_ib (ring, &job->ibs[0 ]);WARN_ON (job->ibs[0 ].length_dw > num_dw);
@@ -2359,18 +2359,17 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)return ;if (enable) { -struct amdgpu_ring *ring;struct drm_gpu_scheduler *sched; -if (!adev->mman.buffer_funcs_ring || !adev->mman.buffer_funcs_ring->sched.ready) { +if (!adev->mman.num_buffer_funcs_scheds || + !adev->mman.buffer_funcs_scheds[0 ]->ready) {dev_warn (adev->dev,"Not enabling DMA transfers for in kernel use" );return ; } num_clear_entities =1 ; num_move_entities =1 ; - ring = adev->mman.buffer_funcs_ring; - sched = &ring->sched; + sched = adev->mman.buffer_funcs_scheds[0 ]; r = amdgpu_ttm_buffer_entity_init(&adev->mman.gtt_mgr, &adev->mman.default_entity, DRM_SCHED_PRIORITY_KERNEL,
@@ -2507,7 +2506,7 @@ int amdgpu_copy_buffer(struct amdgpu_device *adev,unsigned int i;int r; - ring = adev->mman.buffer_funcs_ring; + ring =to_amdgpu_ring (adev->mman.buffer_funcs_scheds[0 ]);if (!ring->sched.ready) { dev_err(adev->dev,
@@ -2740,6 +2739,27 @@ int amdgpu_ttm_evict_resources(struct amdgpu_device *adev, int mem_type)return ttm_resource_manager_evict_all (&adev->mman.bdev, man); } +void amdgpu_sdma_set_buffer_funcs_scheds (struct amdgpu_device *adev, +const struct amdgpu_buffer_funcs *buffer_funcs) +{ +struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB (0 )]; +struct drm_gpu_scheduler *sched; +int i; + + adev->mman.buffer_funcs = buffer_funcs; + +for (i =0 ; i < adev->sdma.num_instances; i++) { +if (adev->sdma.has_page_queue) + sched = &adev->sdma.instance[i].page.sched; +else + sched = &adev->sdma.instance[i].ring.sched; + adev->mman.buffer_funcs_scheds[i] = sched; + } + + adev->mman.num_buffer_funcs_scheds = hub->sdma_invalidation_workaround ? +1 : adev->sdma.num_instances; +} +#if defined(CONFIG_DEBUG_FS )static int amdgpu_ttm_page_pool_show (struct seq_file *m,void *unused)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -1
@@ -67,7 +67,8 @@ struct amdgpu_mman {/* buffer handling */ const struct amdgpu_buffer_funcs *buffer_funcs; -struct amdgpu_ring *buffer_funcs_ring; +struct drm_gpu_scheduler *buffer_funcs_scheds[AMDGPU_MAX_RINGS ]; +u32 num_buffer_funcs_scheds;bool buffer_funcs_enabled;/* @default_entity: for workarounds, has no gart windows */
drivers/gpu/drm/amd/amdgpu/cik_sdma.c +1 -2
@@ -1340,8 +1340,7 @@ static const struct amdgpu_buffer_funcs cik_sdma_buffer_funcs = {static void cik_sdma_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &cik_sdma_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &cik_sdma_buffer_funcs); } const struct amdgpu_ip_block_version cik_sdma_ip_block =
drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +1 -2
@@ -1235,8 +1235,7 @@ static const struct amdgpu_buffer_funcs sdma_v2_4_buffer_funcs = {static void sdma_v2_4_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v2_4_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v2_4_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v2_4_ip_block = {
drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +1 -2
@@ -1677,8 +1677,7 @@ static const struct amdgpu_buffer_funcs sdma_v3_0_buffer_funcs = {static void sdma_v3_0_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v3_0_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v3_0_buffer_funcs); } const struct amdgpu_ip_block_version sdma_v3_0_ip_block =
drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +1 -5
@@ -2615,11 +2615,7 @@ static const struct amdgpu_buffer_funcs sdma_v4_0_buffer_funcs = {static void sdma_v4_0_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v4_0_buffer_funcs; -if (adev->sdma.has_page_queue) - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].page; -else - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v4_0_buffer_funcs); } static void sdma_v4_0_get_ras_error_count(uint32_t value,
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +1 -5
@@ -2316,11 +2316,7 @@ static const struct amdgpu_buffer_funcs sdma_v4_4_2_buffer_funcs = {static void sdma_v4_4_2_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v4_4_2_buffer_funcs; -if (adev->sdma.has_page_queue) - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].page; -else - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v4_4_2_buffer_funcs); }/**
drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +1 -4
@@ -2061,10 +2061,7 @@ static const struct amdgpu_buffer_funcs sdma_v5_0_buffer_funcs = {static void sdma_v5_0_set_buffer_funcs (struct amdgpu_device *adev) { -if (adev->mman.buffer_funcs ==NULL ) { - adev->mman.buffer_funcs = &sdma_v5_0_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; - } +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v5_0_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v5_0_ip_block = {
drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +1 -4
@@ -2072,10 +2072,7 @@ static const struct amdgpu_buffer_funcs sdma_v5_2_buffer_funcs = {static void sdma_v5_2_set_buffer_funcs (struct amdgpu_device *adev) { -if (adev->mman.buffer_funcs ==NULL ) { - adev->mman.buffer_funcs = &sdma_v5_2_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; - } +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v5_2_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v5_2_ip_block = {
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +1 -2
@@ -1904,8 +1904,7 @@ static const struct amdgpu_buffer_funcs sdma_v6_0_buffer_funcs = {static void sdma_v6_0_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v6_0_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v6_0_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v6_0_ip_block = {
drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +1 -2
@@ -1845,8 +1845,7 @@ static const struct amdgpu_buffer_funcs sdma_v7_0_buffer_funcs = {static void sdma_v7_0_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v7_0_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v7_0_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v7_0_ip_block = {
drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c +1 -2
@@ -1755,8 +1755,7 @@ static const struct amdgpu_buffer_funcs sdma_v7_1_buffer_funcs = {static void sdma_v7_1_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &sdma_v7_1_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &sdma_v7_1_buffer_funcs); }const struct amdgpu_ip_block_version sdma_v7_1_ip_block = {
drivers/gpu/drm/amd/amdgpu/si_dma.c +1 -2
@@ -833,8 +833,7 @@ static const struct amdgpu_buffer_funcs si_dma_buffer_funcs = {static void si_dma_set_buffer_funcs (struct amdgpu_device *adev) { - adev->mman.buffer_funcs = &si_dma_buffer_funcs; - adev->mman.buffer_funcs_ring = &adev->sdma.instance[0 ].ring; +amdgpu_sdma_set_buffer_funcs_scheds (adev, &si_dma_buffer_funcs); } const struct amdgpu_ip_block_version si_dma_ip_block =
drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +2 -1
@@ -129,13 +129,14 @@ svm_migrate_copy_memory_gart(struct amdgpu_device *adev, dma_addr_t *sys, struct dma_fence **mfence) {const u64 GTT_MAX_PAGES =AMDGPU_GTT_MAX_TRANSFER_SIZE ; -struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; +struct amdgpu_ring *ring;struct amdgpu_ttm_buffer_entity *entity;u64 gart_s, gart_d;struct dma_fence *next;u64 size;int r; + ring =to_amdgpu_ring (adev->mman.buffer_funcs_scheds[0 ]); entity = &adev->mman.move_entities[0 ];mutex_lock (&entity->lock);
Signature · 1 line
2.43.0
PATCH v4 7/12 drm/amdgpu: only use working sdma schedulers for ttm
In reply to Pierre-Eric Pelloux-PrayerIt's possible that some sdma instances aren't working so we shouldn't try to use them from TTM. To achieve this, delay the call to amdgpu_sdma_set_buffer_funcs_scheds after the rings have been tested, and then use the 'ready' property to decide if a sched should be used or not.
Note that currently it's not doing much, because if the ring helper fails for any ring, the whole sdma block init fails.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 21 +++++++++++++++------ drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 10 ++++++++-- drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 5 +++-- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 5 +++-- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 9 +++++++-- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 6 ++++-- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 6 ++++-- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 9 +++++++-- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 9 +++++++-- drivers/gpu/drm/amd/amdgpu/si_dma.c | 9 +++++++-- 12 files changed, 67 insertions(+), 26 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +15 -6
@@ -2361,8 +2361,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)if (enable) {struct drm_gpu_scheduler *sched; -if (!adev->mman.num_buffer_funcs_scheds || - !adev->mman.buffer_funcs_scheds[0 ]->ready) { +if (!adev->mman.num_buffer_funcs_scheds) {dev_warn (adev->dev,"Not enabling DMA transfers for in kernel use" );return ; }
@@ -2744,20 +2743,30 @@ void amdgpu_sdma_set_buffer_funcs_scheds(struct amdgpu_device *adev, {struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB (0 )];struct drm_gpu_scheduler *sched; -int i; +int i, n; adev->mman.buffer_funcs = buffer_funcs; -for (i =0 ; i < adev->sdma.num_instances; i++) { +for (i =0 , n =0 ; i < adev->sdma.num_instances; i++) {if (adev->sdma.has_page_queue) sched = &adev->sdma.instance[i].page.sched;else sched = &adev->sdma.instance[i].ring.sched; - adev->mman.buffer_funcs_scheds[i] = sched; + +if (!sched->ready) +continue ; + + adev->mman.buffer_funcs_scheds[n++] = sched; + } + +if (n ==0 ) { + adev->mman.num_buffer_funcs_scheds =0 ; +drm_warn (&adev->ddev,"No working sdma ring available\n" ); +return ; } adev->mman.num_buffer_funcs_scheds = hub->sdma_invalidation_workaround ? -1 : adev->sdma.num_instances; +1 : n; }#if defined(CONFIG_DEBUG_FS )
drivers/gpu/drm/amd/amdgpu/cik_sdma.c +8 -2
@@ -939,7 +939,6 @@ static int cik_sdma_early_init(struct amdgpu_ip_block *ip_block)cik_sdma_set_ring_funcs (adev);cik_sdma_set_irq_funcs (adev); -cik_sdma_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &cik_sdma_vm_pte_funcs);return 0 ;
@@ -1000,8 +999,15 @@ static int cik_sdma_sw_fini(struct amdgpu_ip_block *ip_block)static int cik_sdma_hw_init (struct amdgpu_ip_block *ip_block) {struct amdgpu_device *adev = ip_block->adev; +int r; -return cik_sdma_start (adev); + r =cik_sdma_start (adev); +if (r) +return r; + +cik_sdma_set_buffer_funcs (adev); + +return 0 ; }static int cik_sdma_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +3 -2
@@ -828,7 +828,6 @@ static int sdma_v2_4_early_init(struct amdgpu_ip_block *ip_block)return r;sdma_v2_4_set_ring_funcs (adev); -sdma_v2_4_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v2_4_vm_pte_funcs);sdma_v2_4_set_irq_funcs (adev);
@@ -898,7 +897,9 @@ static int sdma_v2_4_hw_init(struct amdgpu_ip_block *ip_block)if (r)return r; -return r; +sdma_v2_4_set_buffer_funcs (adev); + +return 0 ; }static int sdma_v2_4_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +3 -2
@@ -1108,7 +1108,6 @@ static int sdma_v3_0_early_init(struct amdgpu_ip_block *ip_block)return r;sdma_v3_0_set_ring_funcs (adev); -sdma_v3_0_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v3_0_vm_pte_funcs);sdma_v3_0_set_irq_funcs (adev);
@@ -1184,7 +1183,9 @@ static int sdma_v3_0_hw_init(struct amdgpu_ip_block *ip_block)if (r)return r; -return r; +sdma_v3_0_set_buffer_funcs (adev); + +return 0 ; }static int sdma_v3_0_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +7 -2
@@ -1775,7 +1775,6 @@ static int sdma_v4_0_early_init(struct amdgpu_ip_block *ip_block) adev->sdma.has_page_queue = true;sdma_v4_0_set_ring_funcs (adev); -sdma_v4_0_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v4_0_vm_pte_funcs);sdma_v4_0_set_irq_funcs (adev);sdma_v4_0_set_ras_funcs (adev);
@@ -1961,6 +1960,7 @@ static int sdma_v4_0_sw_fini(struct amdgpu_ip_block *ip_block)static int sdma_v4_0_hw_init (struct amdgpu_ip_block *ip_block) {struct amdgpu_device *adev = ip_block->adev; +int r;if (adev->flags &AMD_IS_APU )amdgpu_dpm_set_powergating_by_smu (adev,AMD_IP_BLOCK_TYPE_SDMA , false,0 );
@@ -1968,7 +1968,12 @@ static int sdma_v4_0_hw_init(struct amdgpu_ip_block *ip_block)if (!amdgpu_sriov_vf (adev))sdma_v4_0_init_golden_registers (adev); -return sdma_v4_0_start (adev); + r =sdma_v4_0_start (adev); +if (r) +return r; +sdma_v4_0_set_buffer_funcs (adev); + +return 0 ; }static int sdma_v4_0_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +4 -2
@@ -1368,7 +1368,6 @@ static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block) adev->sdma.has_page_queue = true;sdma_v4_4_2_set_ring_funcs (adev); -sdma_v4_4_2_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v4_4_2_vm_pte_funcs);sdma_v4_4_2_set_irq_funcs (adev);sdma_v4_4_2_set_ras_funcs (adev);
@@ -1568,8 +1567,11 @@ static int sdma_v4_4_2_hw_init(struct amdgpu_ip_block *ip_block)sdma_v4_4_2_inst_init_golden_registers (adev, inst_mask); r =sdma_v4_4_2_inst_start (adev, inst_mask, false); +if (r) +return r; +sdma_v4_4_2_set_buffer_funcs (adev); -return r; +return 0 ; }static int sdma_v4_4_2_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +4 -2
@@ -1373,7 +1373,6 @@ static int sdma_v5_0_early_init(struct amdgpu_ip_block *ip_block)return r;sdma_v5_0_set_ring_funcs (adev); -sdma_v5_0_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v5_0_vm_pte_funcs);sdma_v5_0_set_irq_funcs (adev);sdma_v5_0_set_mqd_funcs (adev);
@@ -1481,8 +1480,11 @@ static int sdma_v5_0_hw_init(struct amdgpu_ip_block *ip_block)sdma_v5_0_init_golden_registers (adev); r =sdma_v5_0_start (adev); +if (r) +return r; +sdma_v5_0_set_buffer_funcs (adev); -return r; +return 0 ; }static int sdma_v5_0_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +7 -2
@@ -1264,7 +1264,6 @@ static int sdma_v5_2_early_init(struct amdgpu_ip_block *ip_block)return r;sdma_v5_2_set_ring_funcs (adev); -sdma_v5_2_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v5_2_vm_pte_funcs);sdma_v5_2_set_irq_funcs (adev);sdma_v5_2_set_mqd_funcs (adev);
@@ -1401,8 +1400,14 @@ static int sdma_v5_2_sw_fini(struct amdgpu_ip_block *ip_block)static int sdma_v5_2_hw_init (struct amdgpu_ip_block *ip_block) {struct amdgpu_device *adev = ip_block->adev; +int r; -return sdma_v5_2_start (adev); + r =sdma_v5_2_start (adev); +if (r) +return r; +sdma_v5_2_set_buffer_funcs (adev); + +return 0 ; }static int sdma_v5_2_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +1 -1
@@ -1313,7 +1313,6 @@ static int sdma_v6_0_early_init(struct amdgpu_ip_block *ip_block)return r;sdma_v6_0_set_ring_funcs (adev); -sdma_v6_0_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v6_0_vm_pte_funcs);sdma_v6_0_set_irq_funcs (adev);sdma_v6_0_set_mqd_funcs (adev);
@@ -1486,6 +1485,7 @@ static int sdma_v6_0_hw_init(struct amdgpu_ip_block *ip_block) r =sdma_v6_0_start (adev);if (r)return r; +sdma_v6_0_set_buffer_funcs (adev);return sdma_v6_0_set_userq_trap_interrupts (adev, true); }
drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +1 -1
@@ -1299,7 +1299,6 @@ static int sdma_v7_0_early_init(struct amdgpu_ip_block *ip_block) }sdma_v7_0_set_ring_funcs (adev); -sdma_v7_0_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v7_0_vm_pte_funcs);sdma_v7_0_set_irq_funcs (adev);sdma_v7_0_set_mqd_funcs (adev);
@@ -1432,6 +1431,7 @@ static int sdma_v7_0_hw_init(struct amdgpu_ip_block *ip_block) r =sdma_v7_0_start (adev);if (r)return r; +sdma_v7_0_set_buffer_funcs (adev);return sdma_v7_0_set_userq_trap_interrupts (adev, true); }
drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c +7 -2
@@ -1266,7 +1266,6 @@ static int sdma_v7_1_early_init(struct amdgpu_ip_block *ip_block) }sdma_v7_1_set_ring_funcs (adev); -sdma_v7_1_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &sdma_v7_1_vm_pte_funcs);sdma_v7_1_set_irq_funcs (adev);sdma_v7_1_set_mqd_funcs (adev);
@@ -1365,10 +1364,16 @@ static int sdma_v7_1_hw_init(struct amdgpu_ip_block *ip_block) {struct amdgpu_device *adev = ip_block->adev;uint32_t inst_mask; +int r; inst_mask =GENMASK (adev->sdma.num_instances -1 ,0 ); -return sdma_v7_1_inst_start (adev, inst_mask); + r =sdma_v7_1_inst_start (adev, inst_mask); +if (r) +return r; +sdma_v7_1_set_buffer_funcs (adev); + +return 0 ; }static int sdma_v7_1_hw_fini (struct amdgpu_ip_block *ip_block)
drivers/gpu/drm/amd/amdgpu/si_dma.c +7 -2
@@ -487,7 +487,6 @@ static int si_dma_early_init(struct amdgpu_ip_block *ip_block) adev->sdma.num_instances =SDMA_MAX_INSTANCE ;si_dma_set_ring_funcs (adev); -si_dma_set_buffer_funcs (adev);amdgpu_sdma_set_vm_pte_scheds (adev, &si_dma_vm_pte_funcs);si_dma_set_irq_funcs (adev);
@@ -543,8 +542,14 @@ static int si_dma_sw_fini(struct amdgpu_ip_block *ip_block)static int si_dma_hw_init (struct amdgpu_ip_block *ip_block) {struct amdgpu_device *adev = ip_block->adev; +int r; -return si_dma_start (adev); + r =si_dma_start (adev); +if (r) +return r; +si_dma_set_buffer_funcs (adev); + +return 0 ; }static int si_dma_hw_fini (struct amdgpu_ip_block *ip_block)
Signature · 1 line
2.43.0
PATCH v4 8/12 drm/amdgpu: create multiple clear/move ttm entities
In reply to Pierre-Eric Pelloux-PrayerThis enables parallelism of operations.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +2 -2
@@ -2366,8 +2366,8 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)return ; } - num_clear_entities =1 ; - num_move_entities =1 ; + num_clear_entities =MIN (adev->mman.num_buffer_funcs_scheds,TTM_NUM_MOVE_FENCES ); + num_move_entities =MIN (adev->mman.num_buffer_funcs_scheds,TTM_NUM_MOVE_FENCES ); sched = adev->mman.buffer_funcs_scheds[0 ]; r = amdgpu_ttm_buffer_entity_init(&adev->mman.gtt_mgr, &adev->mman.default_entity,
Signature · 1 line
2.43.0
PATCH v4 9/12 drm/amdgpu: give ttm entities access to all the sdma scheds
In reply to Pierre-Eric Pelloux-PrayerWith this change we now have as many clear and move entities as we have sdma engines (limited to TTM_NUM_MOVE_FENCES).
To enable load-balancing this patch gives access to all sdma schedulers to all entities.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +9 -7
@@ -2359,8 +2359,6 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)return ;if (enable) { -struct drm_gpu_scheduler *sched; -if (!adev->mman.num_buffer_funcs_scheds) {dev_warn (adev->dev,"Not enabling DMA transfers for in kernel use" );return ;
@@ -2368,11 +2366,10 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) num_clear_entities =MIN (adev->mman.num_buffer_funcs_scheds,TTM_NUM_MOVE_FENCES ); num_move_entities =MIN (adev->mman.num_buffer_funcs_scheds,TTM_NUM_MOVE_FENCES ); - sched = adev->mman.buffer_funcs_scheds[0 ]; r =amdgpu_ttm_buffer_entity_init (&adev->mman.gtt_mgr, &adev->mman.default_entity,DRM_SCHED_PRIORITY_KERNEL , - &sched,1 ,0 ); + adev->mman.buffer_funcs_scheds,1 ,0 );if (r <0 ) {dev_err (adev->dev,"Failed setting up TTM entity (%d)\n" , r);
@@ -2390,8 +2387,11 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable)for (i =0 ; i < num_clear_entities; i++) { r =amdgpu_ttm_buffer_entity_init ( - &adev->mman.gtt_mgr, &adev->mman.clear_entities[i], -DRM_SCHED_PRIORITY_NORMAL , &sched,1 ,1 ); + &adev->mman.gtt_mgr, + &adev->mman.clear_entities[i], +DRM_SCHED_PRIORITY_NORMAL , + adev->mman.buffer_funcs_scheds, + adev->mman.num_buffer_funcs_scheds,1 );if (r <0 ) {for (j =0 ; j < i; j++)
@@ -2410,7 +2410,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) r =amdgpu_ttm_buffer_entity_init ( &adev->mman.gtt_mgr, &adev->mman.move_entities[i], -DRM_SCHED_PRIORITY_NORMAL , &sched,1 ,2 ); +DRM_SCHED_PRIORITY_NORMAL , + adev->mman.buffer_funcs_scheds, + adev->mman.num_buffer_funcs_scheds,2 );if (r <0 ) {for (j =0 ; j < i; j++)
Signature · 1 line
2.43.0
PATCH v4 10/12 drm/amdgpu: get rid of amdgpu_ttm_clear_buffer
In reply to Pierre-Eric Pelloux-PrayerIt's doing the same thing as amdgpu_fill_buffer(src_data=0), so drop it.
The only caveat is that amdgpu_res_cleared() return value is only valid right after allocation.
---
v2: introduce new "bool consider_clear_status" arg
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 16 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 88 +++++----------------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 6 +- 3 files changed, 32 insertions(+), 78 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +10 -6
@@ -717,13 +717,17 @@ int amdgpu_bo_create(struct amdgpu_device *adev, bo->tbo.resource->mem_type == TTM_PL_VRAM) {struct dma_fence *fence; - r =amdgpu_ttm_clear_buffer (bo, bo->tbo.base.resv, &fence); + r =amdgpu_fill_buffer (amdgpu_ttm_next_clear_entity (adev), + bo,0 ,NULL , &fence, + true,AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER );if (unlikely (r)) goto fail_unreserve; -dma_resv_add_fence (bo->tbo.base.resv, fence, -DMA_RESV_USAGE_KERNEL ); -dma_fence_put (fence); +if (fence) { +dma_resv_add_fence (bo->tbo.base.resv, fence, +DMA_RESV_USAGE_KERNEL ); +dma_fence_put (fence); + } }if (!bp->resv)amdgpu_bo_unreserve (bo);
@@ -1326,8 +1330,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo) goto out; r =amdgpu_fill_buffer (amdgpu_ttm_next_clear_entity (adev), - abo,0 , &bo->base._resv, - &fence,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE ); + abo,0 , &bo->base._resv, &fence, + false,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE );if (WARN_ON (r)) goto out;
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +20 -68
118 lines not on this page · raw message
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -4
@@ -187,14 +187,12 @@ int amdgpu_copy_buffer(struct amdgpu_device *adev,struct dma_resv *resv, struct dma_fence **fence, bool vm_needs_flush, uint32_t copy_flags); -int amdgpu_ttm_clear_buffer (struct amdgpu_bo *bo, -struct dma_resv *resv, -struct dma_fence **fence);int amdgpu_fill_buffer (struct amdgpu_ttm_buffer_entity *entity,struct amdgpu_bo *bo,uint32_t src_data,struct dma_resv *resv, -struct dma_fence **f, +struct dma_fence **out_fence, +bool consider_clear_status,u64 k_job_id);struct amdgpu_ttm_buffer_entity *amdgpu_ttm_next_clear_entity (struct amdgpu_device *adev);
Signature · 1 line
2.43.0
PATCH v4 11/12 drm/amdgpu: rename amdgpu_fill_buffer as amdgpu_ttm_clear_buffer
In reply to Pierre-Eric Pelloux-PrayerThis is the only use case for this function.
---
v2: amdgpu_ttm_clear_buffer instead of amdgpu_clear_buffer
---
--- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 12 +++++------ drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 23 ++++++++++------------ drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 13 ++++++------ 3 files changed, 22 insertions(+), 26 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +6 -6
@@ -717,9 +717,9 @@ int amdgpu_bo_create(struct amdgpu_device *adev, bo->tbo.resource->mem_type == TTM_PL_VRAM) {struct dma_fence *fence; - r =amdgpu_fill_buffer (amdgpu_ttm_next_clear_entity (adev), - bo,0 ,NULL , &fence, - true,AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER ); + r =amdgpu_ttm_clear_buffer (amdgpu_ttm_next_clear_entity (adev), + bo,NULL , &fence, + true,AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER );if (unlikely (r)) goto fail_unreserve;
@@ -1329,9 +1329,9 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)if (r) goto out; - r =amdgpu_fill_buffer (amdgpu_ttm_next_clear_entity (adev), - abo,0 , &bo->base._resv, &fence, - false,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE ); + r =amdgpu_ttm_clear_buffer (amdgpu_ttm_next_clear_entity (adev), + abo, &bo->base._resv, &fence, + false,AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE );if (WARN_ON (r)) goto out;
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +10 -13
@@ -417,8 +417,8 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,if (old_mem->mem_type ==TTM_PL_VRAM && (abo->flags &AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE )) {struct dma_fence *wipe_fence =NULL ; - r =amdgpu_fill_buffer (entity, abo,0 ,NULL , &wipe_fence, - false,AMDGPU_KERNEL_JOB_ID_MOVE_BLIT ); + r =amdgpu_ttm_clear_buffer (entity, abo,NULL , &wipe_fence, + false,AMDGPU_KERNEL_JOB_ID_MOVE_BLIT );if (r) { goto error; }else if (wipe_fence) {
@@ -2582,26 +2582,23 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_device *adev, }/** -* amdgpu_fill_buffer - fill a buffer with a given value +* amdgpu_ttm_clear_buffer - fill a buffer with 0 * @entity: entity to use * @bo: the bo to fill -* @src_data: the value to set * @resv: fences contained in this reservation will be used as dependencies. * @out_fence: the fence from the last clear will be stored here. It might be * NULL if no job was run. -* @dependency: optional input dependency fence. * @consider_clear_status: true if region reported as cleared by amdgpu_res_cleared() * are skipped. * @k_job_id: trace id * */ -int amdgpu_fill_buffer (struct amdgpu_ttm_buffer_entity *entity, -struct amdgpu_bo *bo, -uint32_t src_data, -struct dma_resv *resv, -struct dma_fence **out_fence, -bool consider_clear_status, -u64 k_job_id) +int amdgpu_ttm_clear_buffer (struct amdgpu_ttm_buffer_entity *entity, +struct amdgpu_bo *bo, +struct dma_resv *resv, +struct dma_fence **out_fence, +bool consider_clear_status, +u64 k_job_id) {struct amdgpu_device *adev =amdgpu_ttm_adev (bo->tbo.bdev);struct dma_fence *fence =NULL ;
@@ -2632,7 +2629,7 @@ int amdgpu_fill_buffer(struct amdgpu_ttm_buffer_entity *entity, goto error; r =amdgpu_ttm_fill_mem (adev, entity, - src_data, to, cur_size, resv, +0 , to, cur_size, resv, &next, true, k_job_id);if (r) goto error;
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +6 -7
@@ -187,13 +187,12 @@ int amdgpu_copy_buffer(struct amdgpu_device *adev,struct dma_resv *resv, struct dma_fence **fence, bool vm_needs_flush, uint32_t copy_flags); -int amdgpu_fill_buffer (struct amdgpu_ttm_buffer_entity *entity, -struct amdgpu_bo *bo, -uint32_t src_data, -struct dma_resv *resv, -struct dma_fence **out_fence, -bool consider_clear_status, -u64 k_job_id); +int amdgpu_ttm_clear_buffer (struct amdgpu_ttm_buffer_entity *entity, +struct amdgpu_bo *bo, +struct dma_resv *resv, +struct dma_fence **out_fence, +bool consider_clear_status, +u64 k_job_id);struct amdgpu_ttm_buffer_entity *amdgpu_ttm_next_clear_entity (struct amdgpu_device *adev);int amdgpu_ttm_alloc_gart (struct ttm_buffer_object *bo);
Signature · 1 line
2.43.0
PATCH v4 12/12 drm/amdgpu: split amdgpu_ttm_set_buffer_funcs_status in 2 funcs
In reply to Pierre-Eric Pelloux-PrayerMakes a code slightly clearer and reduces indentation.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 14 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 186 +++++++++++---------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 4 +- 3 files changed, 108 insertions(+), 96 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +7 -7
@@ -3158,7 +3158,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)if (r) goto init_failed; -amdgpu_ttm_set_buffer_funcs_status (adev, true); +amdgpu_ttm_enable_buffer_funcs (adev);/* Don't init kfd if whole hive need to be reset during init */ if (adev->init_lvl->level !=AMDGPU_INIT_LEVEL_MINIMAL_XGMI ) {
@@ -3835,7 +3835,7 @@ static int amdgpu_device_ip_suspend(struct amdgpu_device *adev)amdgpu_virt_request_full_gpu (adev, false); } -amdgpu_ttm_set_buffer_funcs_status (adev, false); +amdgpu_ttm_disable_buffer_funcs (adev); r =amdgpu_device_ip_suspend_phase1 (adev);if (r)
@@ -4050,7 +4050,7 @@ static int amdgpu_device_ip_resume(struct amdgpu_device *adev) r =amdgpu_device_ip_resume_phase2 (adev); -amdgpu_ttm_set_buffer_funcs_status (adev, true); +amdgpu_ttm_enable_buffer_funcs (adev);if (r)return r;
@@ -4916,7 +4916,7 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev)/* disable ras feature must before hw fini */ amdgpu_ras_pre_fini (adev); -amdgpu_ttm_set_buffer_funcs_status (adev, false); +amdgpu_ttm_disable_buffer_funcs (adev);/* * device went through surprise hotplug; we need to destroy topology
@@ -5182,7 +5182,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)if (r) goto unwind_userq; -amdgpu_ttm_set_buffer_funcs_status (adev, false); +amdgpu_ttm_disable_buffer_funcs (adev);amdgpu_fence_driver_hw_fini (adev);
@@ -5196,7 +5196,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)return 0 ; unwind_evict: -amdgpu_ttm_set_buffer_funcs_status (adev, true); +amdgpu_ttm_enable_buffer_funcs (adev);amdgpu_fence_driver_hw_init (adev); unwind_userq:
@@ -5930,7 +5930,7 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context)if (r) goto out; -amdgpu_ttm_set_buffer_funcs_status (tmp_adev, true); +amdgpu_ttm_enable_buffer_funcs (tmp_adev); r =amdgpu_device_ip_resume_phase3 (tmp_adev);if (r)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +99 -87
227 lines not on this page · raw message
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +2 -2
@@ -178,8 +178,8 @@ bool amdgpu_res_cpu_visible(struct amdgpu_device *adev,int amdgpu_ttm_init (struct amdgpu_device *adev);void amdgpu_ttm_fini (struct amdgpu_device *adev); -void amdgpu_ttm_set_buffer_funcs_status (struct amdgpu_device *adev, -bool enable); +void amdgpu_ttm_enable_buffer_funcs (struct amdgpu_device *adev); +void amdgpu_ttm_disable_buffer_funcs (struct amdgpu_device *adev); int amdgpu_copy_buffer(struct amdgpu_device *adev, struct amdgpu_ttm_buffer_entity *entity, uint64_t src_offset,
Signature · 1 line
2.43.0
Christian König
maintainerFinally coming back to this patch set here.
Fell free to add Reviewed-by: Christian König <[email protected]> to the first two patches as well and then please start pushing the patches to amd-staging-drm-next.
I probably need to go over the last patches once more, but I think it would be better to have the first few upstream first.
Regards,
Christian.
Quoted 57 lines · Pierre-Eric Pelloux-Prayer
The drm/ttm patch modifies TTM to support multiple contexts for the pipelined moves.
Then amdgpu/ttm is updated to express dependencies between jobs explicitely, instead of relying on the ordering of execution guaranteed by the use of a single instance.
With all of this in place, we can use multiple entities, with each having access to the available SDMA instances.This rework also gives the opportunity to merge the clear functions into a single one and to optimize a bit GART usage.
Since v3 some patches have been already reviewed and merged separately:
- https://lists.freedesktop.org/archives/amd-gfx/2026-January/137747.html
- https://gitlab.freedesktop.org/drm/kernel/-/commit/ddf055b80a544d6f36f77be5f0c6d3c80177d57c
This version depend on them.v3: https://lists.freedesktop.org/archives/dri-devel/2025-November/537830.html
Pierre-Eric Pelloux-Prayer (12):
drm/amdgpu: allocate clear entities dynamically drm/amdgpu: allocate move entities dynamically drm/amdgpu: round robin throughclear_entitiesinamdgpu_fill_bufferdrm/amdgpu: useTTM_NUM_MOVE_FENCESwhen reserving fences drm/amdgpu: use multiple entities inamdgpu_move_blitdrm/amdgpu: pass all the sdma scheds toamdgpu_mmandrm/amdgpu: only use working sdma schedulers for ttm drm/amdgpu: create multiple clear/move ttm entities drm/amdgpu: give ttm entities access to all the sdma scheds drm/amdgpu: get rid ofamdgpu_ttm_clear_bufferdrm/amdgpu: renameamdgpu_fill_bufferasamdgpu_ttm_clear_bufferdrm/amdgpu: splitamdgpu_ttm_set_buffer_funcs_statusin 2 funcsdrivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 4 +- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 17 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 329 ++++++++++-------- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 29 +- drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c | 6 +- drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 13 +- drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 8 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 15 +- drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 12 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 11 +- drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 14 +- drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 5 +- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 12 +- drivers/gpu/drm/amd/amdgpu/si_dma.c | 12 +- drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 5 +- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 +- .../amd/display/amdgpu_dm/amdgpu_dm_plane.c | 6 +- .../drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c | 6 +- 23 files changed, 300 insertions(+), 243 deletions(-)