[PATCH 1/1][SRU][OEM-5.14] drm/amdgpu: force exit gfxoff on sdma resume for rmb s0ix

Alex Hung alex.hung at canonical.com
Fri Oct 15 19:58:01 UTC 2021

From: Prike Liang <Prike.Liang at amd.com>

BugLink: https://bugs.launchpad.net/bugs/1945348

In the s2idle stress test sdma resume fail occasionally,in the
failed case GPU is in the gfxoff state.This issue may introduce
by firmware miss handle doorbell S/R and now temporary fix the issue
by forcing exit gfxoff for sdma resume.

Signed-off-by: Prike Liang <Prike.Liang at amd.com>
Reviewed-by: Alex Deucher <alexander.deucher at amd.com>
Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
Cc: stable at vger.kernel.org
(cherry picked from commit 26db706a6d77b9e184feb11725e97e53b7a89519)
Signed-off-by: Alex Hung <alex.hung at canonical.com>
 drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
index 7486e5306786..27e0ca615edc 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c
@@ -883,6 +883,12 @@ static int sdma_v5_2_start(struct amdgpu_device *adev)
+	/* TODO: check whether can submit a doorbell request to raise
+	 * a doorbell fence to exit gfxoff.
+	 */
+	if (adev->in_s0ix)
+		amdgpu_gfx_off_ctrl(adev, false);
 	/* unhalt the MEs */
 	sdma_v5_2_enable(adev, true);
@@ -891,6 +897,8 @@ static int sdma_v5_2_start(struct amdgpu_device *adev)
 	/* start the gfx rings and rlc compute queues */
 	r = sdma_v5_2_gfx_resume(adev);
+	if (adev->in_s0ix)
+		amdgpu_gfx_off_ctrl(adev, true);
 	if (r)
 		return r;
 	r = sdma_v5_2_rlc_resume(adev);

More information about the kernel-team mailing list