6.1-stable review patch. If anyone has any objections, please let me know. ------------------ From: Jesse Zhang [ Upstream commit fd65d1742992361fc2201ecb4e43411e6e417fcb ] Like the MES scheduler ring, the KIQ ring sets no_scheduler = true and uses a polling fence, so it is skipped by the force-completion loop in amdgpu_device_pre_asic_reset(). Its hw fence value lives in wb (GTT) memory and survives a MODE1 reset while fence_drv.sync_seq keeps advancing, so after a reset the first KIQ submission can poll forever on a seq that is never written back. Force complete the KIQ ring fences too so their hw fence is realigned to sync_seq. Cc: stable@vger.kernel.org Reviewed-by: Alex Deucher Suggested-by: Alex Deucher Signed-off-by: Jesse Zhang Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index fda32238cc6b3..87599141e9037 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -4956,6 +4956,14 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, amdgpu_fence_driver_force_completion(ring); } + /* + * KIQ rings are polling-fence/no_scheduler like MES, so realign their + * fence too (one ring per XCC), otherwise the first post-reset KIQ + * submission polls forever on a stale seq. + */ + if (adev->gfx.kiq.ring.fence_drv.initialized && adev->gfx.kiq.ring.sched.ready) + amdgpu_fence_driver_force_completion(&adev->gfx.kiq.ring); + amdgpu_fence_driver_isr_toggle(adev, false); if (job && job->vm) -- 2.53.0