drm/amdgpu: use separate scheduler entity for UVD submissions
This allows us to remove the kernel context and use a better priority for the submissions. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:

committed by
Alex Deucher

parent
703297c1fe
commit
ead833eced
@@ -1644,6 +1644,7 @@ struct amdgpu_uvd {
|
||||
struct amdgpu_ring ring;
|
||||
struct amdgpu_irq_src irq;
|
||||
bool address_64_bit;
|
||||
struct amd_sched_entity entity;
|
||||
};
|
||||
|
||||
/*
|
||||
|
Reference in New Issue
Block a user