[PATCH 3/3] drm/amdgpu: recovery hw jobs when gpu reset V3

[Chunming Zhou]

V3: directly use pd_addr.

Change-Id: If10da1e224d81a12fd4f8d760c48178adb9e82d0
Signed-off-by: Chunming Zhou <David1.Zhou at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 +++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c     | 3 ++-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 74f484d..0474b7b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1954,6 +1954,7 @@ int amdgpu_gpu_reset(struct amdgpu_device *adev)
 			continue;
 
 		kthread_park(ring->sched.thread);
+		amd_sched_hw_job_reset(&ring->sched);
 	}
 	/* after all hw jobs are reset, hw fence is meaningless, so force_completion */
 	amdgpu_fence_driver_force_completion(adev);
@@ -2001,8 +2002,9 @@ retry:
 			struct amdgpu_ring *ring = adev->rings[i];
 			if (!ring)
 				continue;
+			amd_sched_job_recovery(&ring->sched);
 			kthread_unpark(ring->sched.thread);
-			amdgpu_ring_restore(ring, ring_sizes[i], ring_data[i]);
+			kfree(ring_data[i]);
 			ring_sizes[i] = 0;
 			ring_data[i] = NULL;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index c8d9577..d43bced 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -378,7 +378,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job)
 	    amdgpu_vm_ring_has_compute_vm_bug(ring)))
 		amdgpu_ring_emit_pipeline_sync(ring);
 
-	if (ring->funcs->emit_vm_flush && job->vm_needs_flush) {
+	if (ring->funcs->emit_vm_flush && (job->vm_needs_flush ||
+	    amdgpu_vm_is_gpu_reset(adev, id))) {
 		struct fence *fence;
 
 		trace_amdgpu_vm_flush(job->vm_pd_addr, ring->idx, job->vm_id);
-- 
1.9.1