[Public]

Ignore this.  Newer version sent.

Alex

________________________________
From: Deucher, Alexander <[email protected]>
Sent: Friday, February 21, 2025 10:32 AM
To: [email protected] <[email protected]>
Cc: Deucher, Alexander <[email protected]>; Zhang, Jesse(Jie) 
<[email protected]>
Subject: [PATCH] drm/amdgpu/job: move fix is_guilty logic change

Incrementing the gpu_reset counter needs to be
in the is_guilty block.

Fixes: f447ba2bbd48 ("drm/amdgpu: Update amdgpu_job_timedout to check if the 
ring is guilty")
Cc: Jesse Zhang <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index efba509e2b5d1..a82119b551dc3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -155,8 +155,8 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct 
drm_sched_job *s_job)
                 if (!r) {
                         if (amdgpu_ring_sched_ready(ring))
                                 drm_sched_stop(&ring->sched, s_job);
-                       atomic_inc(&ring->adev->gpu_reset_counter);
                         if (is_guilty) {
+                               atomic_inc(&ring->adev->gpu_reset_counter);
                                 amdgpu_fence_driver_force_completion(ring);
                                 dma_fence_set_error(&s_job->s_fence->finished, 
-ETIME);
                         }
--
2.48.1

Reply via email to