Am 01.06.20 um 20:00 schrieb Alex Deucher:
From: Jack Xiao <Jack.Xiao@xxxxxxx>
As mes ring directly submits to hardwared,
it's no need to set up GPU scheduler for mes ring.
Signed-off-by: Jack Xiao <Jack.Xiao@xxxxxxx>
Acked-by: Alex Deucher <alexander.deucher@xxxxxxx>
Reviewed-by: Hawking Zhang <Hawking.Zhang@xxxxxxx>
Reviewed-by: Christian König <christian.koenig@xxxxxxx>
Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 3414e119f0cb..8712a2e1b869 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -469,8 +469,9 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
if (!ring->fence_drv.fences)
return -ENOMEM;
- /* No need to setup the GPU scheduler for KIQ ring */
- if (ring->funcs->type != AMDGPU_RING_TYPE_KIQ) {
+ /* No need to setup the GPU scheduler for KIQ and MES ring */
+ if (ring->funcs->type != AMDGPU_RING_TYPE_KIQ ||
+ ring->funcs->type != AMDGPU_RING_TYPE_MES) {
BTW: Making this a flag in the ring->funcs struct would probably be
cleaner than checking the two types here.
But not a must have right now.
Christian.
switch (ring->funcs->type) {
case AMDGPU_RING_TYPE_GFX:
timeout = adev->gfx_timeout;
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx