]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
drm/amdgpu: Report individual reset error
authorLijo Lazar <lijo.lazar@amd.com>
Mon, 6 Oct 2025 05:09:03 +0000 (10:39 +0530)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 13 Nov 2025 20:37:39 +0000 (15:37 -0500)
commiteba3906b85ffcedc6b85aeaeec355b5ef7061daf
tree7c77b7a7192343a8b37203124f54b627951baf9e
parentf7569ef1cf978aa87aa81b5e9bf40a77497f3685
drm/amdgpu: Report individual reset error

[ Upstream commit 2e97663760e5fb7ee14f399c68e57b894f01e505 ]

If reinitialization of one of the GPUs fails after reset, it logs
failure on all subsequent GPUs eventhough they have resumed
successfully.

A sample log where only device at 0000:95:00.0 had a failure -

amdgpu 0000:15:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:65:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:75:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:85:00.0: amdgpu: GPU reset(19) succeeded!
amdgpu 0000:95:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:e5:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:f5:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:05:00.0: amdgpu: GPU reset(19) failed
amdgpu 0000:15:00.0: amdgpu: GPU reset end with ret = -5

To avoid confusion, report the error for each device
separately and return the first error as the overall result.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c