Quoting Zhenyu Wang (2017-09-04 16:04:03)
> On 2017.09.04 11:01:41 +0100, Chris Wilson wrote:
> > Quoting changbin...@intel.com (2017-09-04 09:01:01)
> > > From: Changbin Du
> > >
> > > In the past, vGPU alloc fence registers by walking through mm.fence_list
> > > to find fence which pin_count =
On 2017.09.04 11:01:41 +0100, Chris Wilson wrote:
> Quoting changbin...@intel.com (2017-09-04 09:01:01)
> > From: Changbin Du
> >
> > In the past, vGPU alloc fence registers by walking through mm.fence_list
> > to find fence which pin_count = 0 and vma is empty. vGPU may not find
> > enough fence
Quoting changbin...@intel.com (2017-09-04 09:01:01)
> From: Changbin Du
>
> In the past, vGPU alloc fence registers by walking through mm.fence_list
> to find fence which pin_count = 0 and vma is empty. vGPU may not find
> enough fence registers this way. Because a fence can be bind to vma even
>
From: Changbin Du
In the past, vGPU alloc fence registers by walking through mm.fence_list
to find fence which pin_count = 0 and vma is empty. vGPU may not find
enough fence registers this way. Because a fence can be bind to vma even
though it is not in using. We have found such failure many time