diff options
author | Dave Airlie <airlied@redhat.com> | 2022-11-24 04:05:43 +0300 |
---|---|---|
committer | Dave Airlie <airlied@redhat.com> | 2022-11-24 04:05:43 +0300 |
commit | d47f9580839eb6fe568e38b2084d94887fbf5ce0 (patch) | |
tree | 2493555057f8e5c850590d7438fdbf4472666201 /drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | |
parent | 3d335a523b938a445a674be24d1dd5c7a4c86fb6 (diff) | |
parent | eb7081409f94a9a8608593d0fb63a1aa3d6f95d8 (diff) | |
download | linux-d47f9580839eb6fe568e38b2084d94887fbf5ce0.tar.xz |
Backmerge tag 'v6.1-rc6' into drm-next
Linux 6.1-rc6
This is needed for drm-misc-next and tegra.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c')
-rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 18 |
1 files changed, 12 insertions, 6 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 459150460815..8516c814bc9b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -109,6 +109,7 @@ static int amdgpu_cs_p1_ib(struct amdgpu_cs_parser *p, return r; ++(num_ibs[r]); + p->gang_leader_idx = r; return 0; } @@ -298,7 +299,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, if (ret) goto free_all_kdata; } - p->gang_leader = p->jobs[p->gang_size - 1]; + p->gang_leader = p->jobs[p->gang_leader_idx]; if (p->ctx->vram_lost_counter != p->gang_leader->vram_lost_counter) { ret = -ECANCELED; @@ -1210,10 +1211,9 @@ static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p) return r; } - r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_size - 1]); + r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_leader_idx]); if (r && r != -ERESTARTSYS) DRM_ERROR("amdgpu_ctx_wait_prev_fence failed.\n"); - return r; } @@ -1247,9 +1247,12 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, for (i = 0; i < p->gang_size; ++i) drm_sched_job_arm(&p->jobs[i]->base); - for (i = 0; i < (p->gang_size - 1); ++i) { + for (i = 0; i < p->gang_size; ++i) { struct dma_fence *fence; + if (p->jobs[i] == leader) + continue; + fence = &p->jobs[i]->base.s_fence->scheduled; r = drm_sched_job_add_dependency(&leader->base, fence); if (r) @@ -1286,7 +1289,10 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, list_for_each_entry(e, &p->validated, tv.head) { /* Everybody except for the gang leader uses READ */ - for (i = 0; i < (p->gang_size - 1); ++i) { + for (i = 0; i < p->gang_size; ++i) { + if (p->jobs[i] == leader) + continue; + dma_resv_add_fence(e->tv.bo->base.resv, &p->jobs[i]->base.s_fence->finished, DMA_RESV_USAGE_READ); @@ -1296,7 +1302,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, e->tv.num_shared = 0; } - seq = amdgpu_ctx_add_fence(p->ctx, p->entities[p->gang_size - 1], + seq = amdgpu_ctx_add_fence(p->ctx, p->entities[p->gang_leader_idx], p->fence); amdgpu_cs_post_dependencies(p); |