Module: Mesa Branch: main Commit: 32eb4a54dcb9bc518c9d74bfeda914dedfad0337 URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=32eb4a54dcb9bc518c9d74bfeda914dedfad0337
Author: Marek Olšák <marek.ol...@amd.com> Date: Sun Dec 3 17:19:41 2023 -0500 winsys/amdgpu: don't count memory usage because it's unused Acked-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimu...@amd.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-pra...@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26547> --- src/gallium/winsys/amdgpu/drm/amdgpu_cs.c | 41 ++++++------------------------- 1 file changed, 7 insertions(+), 34 deletions(-) diff --git a/src/gallium/winsys/amdgpu/drm/amdgpu_cs.c b/src/gallium/winsys/amdgpu/drm/amdgpu_cs.c index 1570b5f702e..54546088adb 100644 --- a/src/gallium/winsys/amdgpu/drm/amdgpu_cs.c +++ b/src/gallium/winsys/amdgpu/drm/amdgpu_cs.c @@ -638,8 +638,7 @@ amdgpu_do_add_real_buffer(struct amdgpu_cs_context *cs, } static int -amdgpu_lookup_or_add_real_buffer(struct radeon_cmdbuf *rcs, struct amdgpu_cs_context *cs, - struct amdgpu_winsys_bo *bo) +amdgpu_lookup_or_add_real_buffer(struct amdgpu_cs_context *cs, struct amdgpu_winsys_bo *bo) { unsigned hash; int idx = amdgpu_lookup_buffer(cs, bo, cs->real_buffers, cs->num_real_buffers); @@ -651,17 +650,10 @@ amdgpu_lookup_or_add_real_buffer(struct radeon_cmdbuf *rcs, struct amdgpu_cs_con hash = bo->unique_id & (BUFFER_HASHLIST_SIZE-1); cs->buffer_indices_hashlist[hash] = idx & 0x7fff; - - if (bo->base.placement & RADEON_DOMAIN_VRAM) - rcs->used_vram_kb += bo->base.size / 1024; - else if (bo->base.placement & RADEON_DOMAIN_GTT) - rcs->used_gart_kb += bo->base.size / 1024; - return idx; } -static int amdgpu_lookup_or_add_slab_buffer(struct radeon_cmdbuf *rcs, - struct amdgpu_cs_context *cs, +static int amdgpu_lookup_or_add_slab_buffer(struct amdgpu_cs_context *cs, struct amdgpu_winsys_bo *bo) { struct amdgpu_cs_buffer *buffer; @@ -672,7 +664,7 @@ static int amdgpu_lookup_or_add_slab_buffer(struct radeon_cmdbuf *rcs, if (idx >= 0) return idx; - real_idx = amdgpu_lookup_or_add_real_buffer(rcs, cs, &get_slab_bo(bo)->real->b); + real_idx = amdgpu_lookup_or_add_real_buffer(cs, &get_slab_bo(bo)->real->b); if (real_idx < 0) return -1; @@ -708,8 +700,7 @@ static int amdgpu_lookup_or_add_slab_buffer(struct radeon_cmdbuf *rcs, return idx; } -static int amdgpu_lookup_or_add_sparse_buffer(struct radeon_cmdbuf *rcs, - struct amdgpu_cs_context *cs, +static int amdgpu_lookup_or_add_sparse_buffer(struct amdgpu_cs_context *cs, struct amdgpu_winsys_bo *bo) { struct amdgpu_cs_buffer *buffer; @@ -746,21 +737,6 @@ static int amdgpu_lookup_or_add_sparse_buffer(struct radeon_cmdbuf *rcs, hash = bo->unique_id & (BUFFER_HASHLIST_SIZE-1); cs->buffer_indices_hashlist[hash] = idx & 0x7fff; - - /* We delay adding the backing buffers until we really have to. However, - * we cannot delay accounting for memory use. - */ - simple_mtx_lock(&get_sparse_bo(bo)->lock); - - list_for_each_entry(struct amdgpu_sparse_backing, backing, &get_sparse_bo(bo)->backing, list) { - if (bo->base.placement & RADEON_DOMAIN_VRAM) - rcs->used_vram_kb += backing->bo->b.base.size / 1024; - else if (bo->base.placement & RADEON_DOMAIN_GTT) - rcs->used_gart_kb += backing->bo->b.base.size / 1024; - } - - simple_mtx_unlock(&get_sparse_bo(bo)->lock); - return idx; } @@ -785,20 +761,20 @@ static unsigned amdgpu_cs_add_buffer(struct radeon_cmdbuf *rcs, return 0; if (bo->type == AMDGPU_BO_SLAB) { - int index = amdgpu_lookup_or_add_slab_buffer(rcs, cs, bo); + int index = amdgpu_lookup_or_add_slab_buffer(cs, bo); if (index < 0) return 0; buffer = &cs->slab_buffers[index]; cs->real_buffers[buffer->slab_real_idx].usage |= usage & ~RADEON_USAGE_SYNCHRONIZED; } else if (is_real_bo(bo)) { - int index = amdgpu_lookup_or_add_real_buffer(rcs, cs, bo); + int index = amdgpu_lookup_or_add_real_buffer(cs, bo); if (index < 0) return 0; buffer = &cs->real_buffers[index]; } else { - int index = amdgpu_lookup_or_add_sparse_buffer(rcs, cs, bo); + int index = amdgpu_lookup_or_add_sparse_buffer(cs, bo); if (index < 0) return 0; @@ -1866,9 +1842,6 @@ static int amdgpu_cs_flush(struct radeon_cmdbuf *rcs, RADEON_USAGE_READ | RADEON_PRIO_IB, 0); } - rcs->used_gart_kb = 0; - rcs->used_vram_kb = 0; - if (cs->ip_type == AMD_IP_GFX) ws->num_gfx_IBs++; else if (cs->ip_type == AMD_IP_SDMA)
