[PATCH 03/17] drm/amdgpu: detect timeout error when deactivating hqd

[Andres Rodriguez]

Handle HQD deactivation timeouts instead of ignoring them.

Reviewed-by: Edward O'Callaghan <funfunctor at folklore1984.net>
Acked-by: Christian König <christian.koenig at amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling at amd.com>
Signed-off-by: Andres Rodriguez <andresx7 at gmail.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index b670302..cd1af26 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4947,75 +4947,89 @@ static int gfx_v8_0_mqd_commit(struct amdgpu_ring *ring)
 
 	/* enable the doorbell if requested */
 	WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, mqd->cp_hqd_pq_doorbell_control);
 
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
 	WREG32(mmCP_HQD_PQ_WPTR, mqd->cp_hqd_pq_wptr);
 
 	/* set the vmid for the queue */
 	WREG32(mmCP_HQD_VMID, mqd->cp_hqd_vmid);
 
 	WREG32(mmCP_HQD_PERSISTENT_STATE, mqd->cp_hqd_persistent_state);
 
 	/* activate the queue */
 	WREG32(mmCP_HQD_ACTIVE, mqd->cp_hqd_active);
 
 	return 0;
 }
 
 static int gfx_v8_0_kiq_init_queue(struct amdgpu_ring *ring)
 {
+	int r = 0;
 	struct amdgpu_device *adev = ring->adev;
 	struct vi_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = AMDGPU_MAX_COMPUTE_RINGS;
 
 	gfx_v8_0_kiq_setting(ring);
 
 	if (adev->gfx.in_reset) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(*mqd));
 
 		/* reset ring buffer */
 		ring->wptr = 0;
 		amdgpu_ring_clear_ring(ring);
 
 		mutex_lock(&adev->srbm_mutex);
 		vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
-		gfx_v8_0_deactivate_hqd(adev, 1);
+		r = gfx_v8_0_deactivate_hqd(adev, 1);
+		if (r) {
+			dev_err(adev->dev, "failed to deactivate ring %s\n", ring->name);
+			goto out_unlock;
+		}
 		gfx_v8_0_mqd_commit(ring);
 		vi_srbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 	} else {
 		mutex_lock(&adev->srbm_mutex);
 		vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
 		gfx_v8_0_mqd_init(ring);
-		gfx_v8_0_deactivate_hqd(adev, 1);
+		r = gfx_v8_0_deactivate_hqd(adev, 1);
+		if (r) {
+			dev_err(adev->dev, "failed to deactivate ring %s\n", ring->name);
+			goto out_unlock;
+		}
 		gfx_v8_0_mqd_commit(ring);
 		vi_srbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
 	}
 
-	return 0;
+	return r;
+
+out_unlock:
+	vi_srbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+	return r;
 }
 
 static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct vi_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.compute_ring[0];
 
 	if (!adev->gfx.in_reset && !adev->gfx.in_suspend) {
 		mutex_lock(&adev->srbm_mutex);
 		vi_srbm_select(adev, ring->me, ring->pipe, ring->queue, 0);
 		gfx_v8_0_mqd_init(ring);
 		vi_srbm_select(adev, 0, 0, 0, 0);
 		mutex_unlock(&adev->srbm_mutex);
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(*mqd));
 	} else if (adev->gfx.in_reset) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
-- 
2.9.3