diff options
| author | Andrey Grodzovsky <andrey.grodzovsky@amd.com> | 2018-10-25 21:47:02 +0200 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2018-10-26 20:27:30 +0200 |
| commit | 31e3aad62ab8039581d621403b7905aa19fb30af (patch) | |
| tree | ac2573dcf20c3c9b121b66c269fc69fb1f0af23b /drivers/gpu/drm | |
| parent | drm/amdgpu: fix VM leaf walking (diff) | |
| download | linux-31e3aad62ab8039581d621403b7905aa19fb30af.tar.xz linux-31e3aad62ab8039581d621403b7905aa19fb30af.zip | |
drm/amdgpu: Fix compute ring 1.0.0 failure after reset
Problem: After GPU reset on dGPUs with gfx8 compute ring
1.0.0 fails to pass the ring test. Ring registers inspection
shows that it's active and no hang is observed (rptr == wptr)
No significant diffs were observed between CP_HQD* registers
for the ring in good and bad shape.
Fix: No clear reason why but reversing the order of ring tests
fixes the problem.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 3d0f277a6523..617b0c8908a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -4815,8 +4815,10 @@ static int gfx_v8_0_kcq_resume(struct amdgpu_device *adev) if (r) goto done; - /* Test KCQs */ - for (i = 0; i < adev->gfx.num_compute_rings; i++) { + /* Test KCQs - reversing the order of rings seems to fix ring test failure + * after GPU reset + */ + for (i = adev->gfx.num_compute_rings - 1; i >= 0; i--) { ring = &adev->gfx.compute_ring[i]; ring->ready = true; r = amdgpu_ring_test_ring(ring); |