2022-08-18 04:30:17

by Akhil P Oommen

[permalink] [raw]
Subject: [PATCH v4 7/7] drm/msm/a6xx: Handle GMU prepare-slumber hfi failure

When prepare-slumber hfi fails, we should follow a6xx_gmu_force_off()
sequence.

Signed-off-by: Akhil P Oommen <[email protected]>
---

(no changes since v1)

drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index db05942..3d00ef9 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1082,7 +1082,11 @@ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu)
a6xx_bus_clear_pending_transactions(adreno_gpu);

/* tell the GMU we want to slumber */
- a6xx_gmu_notify_slumber(gmu);
+ ret = a6xx_gmu_notify_slumber(gmu);
+ if (ret) {
+ a6xx_gmu_force_off(gmu);
+ return;
+ }

ret = gmu_poll_timeout(gmu,
REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS, val,
--
2.7.4