diff options
| author | Dave Airlie <airlied@redhat.com> | 2023-06-19 11:57:25 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2023-06-19 11:57:26 +1000 |
| commit | bcbede6fbeb0e1eb85ccbb532faf06d3b31f0e73 (patch) | |
| tree | 53062368790eb070705726f24a49beb34b8737d0 /drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | |
| parent | 4e237d84eec8783eb6dfbccb04a4e570b27a9b09 (diff) | |
| parent | 72f1de49ffb90b29748284f27f1d6b829ab1de95 (diff) | |
Merge tag 'amd-drm-next-6.5-2023-06-16' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.5-2023-06-16:
amdgpu:
- Misc display fixes
- W=1 fixes
- Improve scheduler naming
- DCN 3.1.4 fixes
- kdoc fixes
- Enable W=1
- VCN 4.0 fix
- xgmi fixes
- TOPDOWN fix for large BAR systems
- eDP fix
- PSR fixes
- SubVP fixes
- Freesync fix
- DPIA fix
- SMU 13.0.5 fixes
- vblflash fix
- RAS fixes
- SDMA 4 fix
- BO locking fix
- BO backing store fix
- NBIO 7.9 fixes
- GC 9.4.3 fixes
- GPU reset recovery fixes
- HMM fix
amdkfd:
- Fix NULL check
- Trap fixes
- Queue count fix
- Add event age tracking
radeon:
- fbdev client fix
scheduler:
- Avoid an infinite loop
UAPI:
- Add KFD event age tracking:
Proposed ROCT-Thunk-Interface:
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/efdbf6cfbc026bd68ac3c35d00dacf84370eb81e
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/1820ae0a2db85b6f584611dc0cde1a00e7c22915
Proposed ROCR-Runtime:
https://github.com/RadeonOpenCompute/ROCR-Runtime/compare/master...zhums:ROCR-Runtime:new_event_wait_review
https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/e1f5bdb88eb882ac798aeca2c00ea3fbb2dba459
https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/7d26afd14107b5c2a754c1a3f415d89f3aabb503
drm:
- DP MST fix
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230616163548.7706-1-alexander.deucher@amd.com
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_job.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 17 |
1 files changed, 14 insertions, 3 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index aca3a2bfe8d2..78476bc75b4e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -109,7 +109,7 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm, (*job)->vm = vm; amdgpu_sync_create(&(*job)->explicit_sync); - (*job)->vram_lost_counter = atomic_read(&adev->vram_lost_counter); + (*job)->generation = amdgpu_vm_generation(adev, vm); (*job)->vm_pd_addr = AMDGPU_BO_INVALID_OFFSET; if (!entity) @@ -258,16 +258,27 @@ amdgpu_job_prepare_job(struct drm_sched_job *sched_job, struct dma_fence *fence = NULL; int r; + /* Ignore soft recovered fences here */ + r = drm_sched_entity_error(s_entity); + if (r && r != -ENODATA) + goto error; + if (!fence && job->gang_submit) fence = amdgpu_device_switch_gang(ring->adev, job->gang_submit); while (!fence && job->vm && !job->vmid) { r = amdgpu_vmid_grab(job->vm, ring, job, &fence); - if (r) + if (r) { DRM_ERROR("Error getting VM ID (%d)\n", r); + goto error; + } } return fence; + +error: + dma_fence_set_error(&job->base.s_fence->finished, r); + return NULL; } static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job) @@ -284,7 +295,7 @@ static struct dma_fence *amdgpu_job_run(struct drm_sched_job *sched_job) trace_amdgpu_sched_run_job(job); /* Skip job if VRAM is lost and never resubmit gangs */ - if (job->vram_lost_counter != atomic_read(&adev->vram_lost_counter) || + if (job->generation != amdgpu_vm_generation(adev, job->vm) || (job->job_run_counter && job->gang_submit)) dma_fence_set_error(finished, -ECANCELED); |