Skip to contentSkip to conversation

drm/amdgpu: disable gpu-sched load balance for uvd_enc

PATCH In review Nirmoy Das 2 participants 2 messages Sep 2, 2020

Subsystems (2)
lore
Share & tools
lore

Nirmoy Das

author
#0 lore raw

PATCH 1/1 drm/amdgpu: disable gpu-sched load balance for uvd_enc

On hardware with multiple uvd instances, dependent uvd_enc jobs may get scheduled to different uvd instances. Because uvd_enc jobs retain hw context, dependent jobs should always run on the same uvd instance. This patch disables GPU scheduler's load balancer for a context that binds jobs from same the context to a uvd instance.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +2 -0
index 7cd398d25498..c80d8339f58c 100644
@@ -114,8 +114,10 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 	scheds = adev->gpu_sched[hw_ip][hw_prio].sched;
 	num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds;
 
+	/* disable load balance if the hw engine retains context among dependent jobs */
 	if (hw_ip == AMDGPU_HW_IP_VCN_ENC ||
 	    hw_ip == AMDGPU_HW_IP_VCN_DEC ||
+	    hw_ip == AMDGPU_HW_IP_UVD_ENC ||
 	    hw_ip == AMDGPU_HW_IP_UVD) {
 		sched = drm_sched_pick_best(scheds, num_scheds);
 		scheds = &sched;

--
2.28.0

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

#1 lore raw

Please ignore this.

Quoted 6 lines · Nirmoy Das

On hardware with multiple uvd instances, dependent uvd_enc jobs may get scheduled to different uvd instances. Because uvd_enc jobs retain hw context, dependent jobs should always run on the same uvd instance. This patch disables GPU scheduler's load balancer for a context that binds jobs from same the context to a uvd instance.

from same the context --> from the same context

Quoted 22 lines
---
  drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 2 ++
  1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +2 -1
index 7cd398d25498..c80d8339f58c 100644
@@ -114,8 +114,10 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
  	scheds = adev->gpu_sched[hw_ip][hw_prio].sched;
  	num_scheds = adev->gpu_sched[hw_ip][hw_prio].num_scheds;
 
+	/* disable load balance if the hw engine retains context among dependent jobs */
  	if (hw_ip == AMDGPU_HW_IP_VCN_ENC ||
  	    hw_ip == AMDGPU_HW_IP_VCN_DEC ||
+	    hw_ip == AMDGPU_HW_IP_UVD_ENC ||
  	    hw_ip == AMDGPU_HW_IP_UVD) {
  		sched = drm_sched_pick_best(scheds, num_scheds);
  		scheds = &sched;
--
Quoted 2 lines

2.28.0

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx