From: Monk Liu <Monk.Liu@amd.com>
Date: Mon, 25 Dec 2017 15:14:58 +0800
Subject: drm/amdgpu: stop all rings before doing gpu recover
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Git-commit: 711826656bebb09b814349fac21cb13f88f92665
Patch-mainline: v4.17-rc1
References: FATE#326289 FATE#326079 FATE#326049 FATE#322398 FATE#326166

found recover_vram_from_shadow sometimes get executed
in paralle with SDMA scheduler, should stop all
schedulers before doing gpu reset/recover

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Petr Tesarik <ptesarik@suse.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |   40 ++++++++++-------------------
 1 file changed, 15 insertions(+), 25 deletions(-)

--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2651,22 +2651,23 @@ int amdgpu_device_gpu_recover(struct amd
 
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&adev->mman.bdev);
+
 	/* store modesetting */
 	if (amdgpu_device_has_dc_support(adev))
 		state = drm_atomic_helper_suspend(adev->ddev);
 
-	/* block scheduler */
+	/* block all schedulers and reset given job's ring */
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
 		struct amdgpu_ring *ring = adev->rings[i];
 
 		if (!ring || !ring->sched.thread)
 			continue;
 
-		/* only focus on the ring hit timeout if &job not NULL */
+		kthread_park(ring->sched.thread);
+
 		if (job && job->ring->idx != i)
 			continue;
 
-		kthread_park(ring->sched.thread);
 		drm_sched_hw_job_reset(&ring->sched, &job->base);
 
 		/* after all hw jobs are reset, hw fence is meaningless, so force_completion */
@@ -2709,33 +2710,22 @@ int amdgpu_device_gpu_recover(struct amd
 			}
 			dma_fence_put(fence);
 		}
+	}
 
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-			struct amdgpu_ring *ring = adev->rings[i];
-
-			if (!ring || !ring->sched.thread)
-				continue;
+	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		struct amdgpu_ring *ring = adev->rings[i];
 
-			/* only focus on the ring hit timeout if &job not NULL */
-			if (job && job->ring->idx != i)
-				continue;
+		if (!ring || !ring->sched.thread)
+			continue;
 
+		/* only need recovery sched of the given job's ring
+		 * or all rings (in the case @job is NULL)
+		 * after above amdgpu_reset accomplished
+		 */
+		if ((!job || job->ring->idx == i) && !r)
 			drm_sched_job_recovery(&ring->sched);
-			kthread_unpark(ring->sched.thread);
-		}
-	} else {
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-			struct amdgpu_ring *ring = adev->rings[i];
-
-			if (!ring || !ring->sched.thread)
-				continue;
-
-			/* only focus on the ring hit timeout if &job not NULL */
-			if (job && job->ring->idx != i)
-				continue;
 
-			kthread_unpark(adev->rings[i]->sched.thread);
-		}
+		kthread_unpark(ring->sched.thread);
 	}
 
 	if (amdgpu_device_has_dc_support(adev)) {