From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 13 Nov 2025 18:24:10 +0000 (-0500)
Subject: drm/amdgpu: don't reemit ring contents more than once
X-Git-Tag: v6.19-rc5~24^2^2~8
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=9fc27cbabee6d2e63b0268ca709ad3129b3ac50d;p=thirdparty%2Fkernel%2Flinux.git

drm/amdgpu: don't reemit ring contents more than once

If we cancel a bad job and reemit the ring contents, and
we get another timeout, cancel everything rather than reemitting.
The wptr markers are only relevant for the original emit.  If
we reemit, the wptr markers are no longer correct.

Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit fb62a2067ca4555a6572d911e05919a311c010aa)
---

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index c7843e336310..4f74a02a9a05 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -709,6 +709,7 @@ void amdgpu_fence_driver_guilty_force_completion(struct amdgpu_fence *af)
 	struct amdgpu_ring *ring = af->ring;
 	unsigned long flags;
 	u32 seq, last_seq;
+	bool reemitted = false;
 
 	last_seq = amdgpu_fence_read(ring) & ring->fence_drv.num_fences_mask;
 	seq = ring->fence_drv.sync_seq & ring->fence_drv.num_fences_mask;
@@ -726,7 +727,9 @@ void amdgpu_fence_driver_guilty_force_completion(struct amdgpu_fence *af)
 		if (unprocessed && !dma_fence_is_signaled_locked(unprocessed)) {
 			fence = container_of(unprocessed, struct amdgpu_fence, base);
 
-			if (fence == af)
+			if (fence->reemitted > 1)
+				reemitted = true;
+			else if (fence == af)
 				dma_fence_set_error(&fence->base, -ETIME);
 			else if (fence->context == af->context)
 				dma_fence_set_error(&fence->base, -ECANCELED);
@@ -734,9 +737,16 @@ void amdgpu_fence_driver_guilty_force_completion(struct amdgpu_fence *af)
 		rcu_read_unlock();
 	} while (last_seq != seq);
 	spin_unlock_irqrestore(&ring->fence_drv.lock, flags);
-	/* signal the guilty fence */
-	amdgpu_fence_write(ring, (u32)af->base.seqno);
-	amdgpu_fence_process(ring);
+
+	if (reemitted) {
+		/* if we've already reemitted once then just cancel everything */
+		amdgpu_fence_driver_force_completion(af->ring);
+		af->ring->ring_backup_entries_to_copy = 0;
+	} else {
+		/* signal the guilty fence */
+		amdgpu_fence_write(ring, (u32)af->base.seqno);
+		amdgpu_fence_process(ring);
+	}
 }
 
 void amdgpu_fence_save_wptr(struct amdgpu_fence *af)
@@ -784,10 +794,12 @@ void amdgpu_ring_backup_unprocessed_commands(struct amdgpu_ring *ring,
 			/* save everything if the ring is not guilty, otherwise
 			 * just save the content from other contexts.
 			 */
-			if (!guilty_fence || (fence->context != guilty_fence->context))
+			if (!fence->reemitted &&
+			    (!guilty_fence || (fence->context != guilty_fence->context)))
 				amdgpu_ring_backup_unprocessed_command(ring, wptr,
 								       fence->wptr);
 			wptr = fence->wptr;
+			fence->reemitted++;
 		}
 		rcu_read_unlock();
 	} while (last_seq != seq);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 7a27c6c4bb44..5044cf9e45fb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -148,6 +148,8 @@ struct amdgpu_fence {
 	u64				wptr;
 	/* fence context for resets */
 	u64				context;
+	/* has this fence been reemitted */
+	unsigned int			reemitted;
 };
 
 extern const struct drm_sched_backend_ops amdgpu_sched_ops;