kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c, branch v6.19.12

drm/amdgpu: avoid a warning in timedout job handler

2026-03-04T12:19:58+00:00

[ Upstream commit c8cf9ddc549fb93cb5a35f3fe23487b1e6707e74 ] Only set an error on the fence if the fence is not signalled. We can end up with a warning if the per queue reset path signals the fence and sets an error as part of the reset, but fails to recover. Reviewed-by: Timur Kristóf Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdgpu: free hw_vm_fence when fail in amdgpu_job_alloc

2026-01-21T02:50:03+00:00

If drm_sched_job_init fails, hw_vm_fence is not freed currently, then cause memory leak. Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Link: https://lore.kernel.org/amd-gfx/a5a828cb-0e4a-41f0-94c3-df31e5ddad52@amd.com/T/#t Signed-off-by: Jiqian Chen Reviewed-by: Amos Kong Reviewed-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 5d42ee457ccd1fb5da4c7f817825b2806ec36956)

drm/amdgpu: free job fences on failure in amdgpu_job_alloc_with_ib

2025-11-26T16:27:24+00:00

Otherwise we're leaking memory. Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Signed-off-by: Pierre-Eric Pelloux-Prayer Reviewed-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: clear job on failure in amdgpu_job_alloc(_with_ib)

2025-11-26T16:25:29+00:00

If memory is freed we need to nullify the pointer or the caller might call kfree again (eg: amdgpu_cs_parser_fini calls kfree on all non-null job pointers). Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Signed-off-by: Pierre-Eric Pelloux-Prayer Reviewed-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: fix possible fence leaks from job structure

2025-11-04T16:53:59+00:00

If we don't end up initializing the fences, free them when we free the job. We can't set the hw_fence to NULL after emitting it because we need it in the cleanup path for the submit direct case. v2: take a reference to the fences if we emit them v3: handle non-job fence in error paths Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Reviewed-by: Jesse Zhang (v1) Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: move reset debug disable handling

2025-11-04T16:33:54+00:00

Move everything to the supported resets masks rather than having an explicit misc checks for this. Reviewed-by: Jesse Zhang Signed-off-by: Alex Deucher

drm/amdgpu: clean up and unify hw fence handling

2025-10-13T18:14:35+00:00

Decouple the amdgpu fence from the amdgpu_job structure. This lets us clean up the separate fence ops for the embedded fence and other fences. This also allows us to allocate the vm fence up front when we allocate the job. v2: Additional cleanup suggested by Christian v3: Additional cleanups suggested by Christian v4: Additional cleanups suggested by David and vm fence fix v5: cast seqno (David) Cc: David.Wu3@amd.com Cc: christian.koenig@amd.com Tested-by: David (Ming Qiang) Wu Reviewed-by: David (Ming Qiang) Wu Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: give each kernel job a unique id

2025-09-01T05:19:31+00:00

Userspace jobs have drm_file.client_id as a unique identifier as job's owners. For kernel jobs, we can allocate arbitrary values - the risk of overlap with userspace ids is small (given that it's a u64 value). In the unlikely case the overlap happens, it'll only impact trace events. Since this ID is traced in the gpu_scheduler trace events, this allows to determine the source of each job sent to the hardware. To make grepping easier, the IDs are defined as they will appear in the trace output. Signed-off-by: Pierre-Eric Pelloux-Prayer Acked-by: Alex Deucher Signed-off-by: Arunpravin Paneer Selvam Link: https://lore.kernel.org/r/20250604122827.2191-1-pierre-eric.pelloux-prayer@amd.com

drm/amdgpu: Retain job->vm in amdgpu_job_prepare_job

2025-08-04T19:41:09+00:00

The field job->vm is used in function amdgpu_job_run to get the page table re-generation counter and decide whether the job should be skipped. Specifically, function amdgpu_vm_generation checks if the VM is valid for this job to use. For instance, if a gfx job depends on a cancelled sdma job from entity vm->delayed, then the gfx job should be skipped. Fixes: 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in amdgpu_job_prepare") Signed-off-by: YuanShang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit ed76936c6b10b547c6df4ca75412331e9ef6d339) Cc: stable@vger.kernel.org

Merge tag 'amd-drm-next-6.17-2025-07-17' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

2025-07-21T01:57:43+00:00

amd-drm-next-6.17-2025-07-17: amdgpu: - Partition fixes - Reset fixes - RAS fixes - i2c fix - MPC updates - DSC cleanup - EDID fixes - Display idle D3 update - IPS updates - DMUB updates - Retimer fix - Replay fixes - Fix DC memory leak - Initial support for smartmux - DCN 4.0.1 degamma LUT fix - Per queue reset cleanups - Track ring state associated with a fence - SR-IOV fixes - SMU fixes - Per queue reset improvements for GC 9+ compute - Per queue reset improvements for GC 10+ gfx - Per queue reset improvements for SDMA 5+ - Per queue reset improvements for JPEG 2+ - Per queue reset improvements for VCN 2+ - GC 8 fix - ISP updates amdkfd: - Enable KFD on LoongArch radeon: - Drop console lock during suspend/resume UAPI: - Add userq slot info to INFO IOCTL Used for IGT userq validation tests (https://lists.freedesktop.org/archives/igt-dev/2025-July/093228.html) From: Alex Deucher Link: https://lore.kernel.org/r/20250717213827.2061581-1-alexander.deucher@amd.com Signed-off-by: Dave Airlie