提交 2ef004d9 编写于 作者: C Christian König 提交者: Alex Deucher

drm/amdgpu: always signal all fences

A little fallout from "drm/amdgpu: sanitize fence numbers", we
sometimes need to signal all fences in the ring.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChunming Zhou <david1.zhou@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
上级 090b767e
......@@ -204,10 +204,13 @@ void amdgpu_fence_process(struct amdgpu_ring *ring)
if (seq != ring->fence_drv.sync_seq)
amdgpu_fence_schedule_fallback(ring);
if (unlikely(seq == last_seq))
return;
last_seq &= drv->num_fences_mask;
seq &= drv->num_fences_mask;
while (last_seq != seq) {
do {
struct fence *fence, **ptr;
++last_seq;
......@@ -228,7 +231,7 @@ void amdgpu_fence_process(struct amdgpu_ring *ring)
BUG();
fence_put(fence);
}
} while (last_seq != seq);
}
/**
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册