Re: [PATCH] drm/i915/userptr: detect un-GUP-able pages early

2021-01-16 Thread Jinoh Kang
On 1/15/21 5:07 PM, Chris Wilson wrote: > Quoting Chris Wilson (2021-01-15 16:56:42) >> Quoting Jinoh Kang (2021-01-15 16:23:31) >>> If GUP-ineligible pages are passed to a GEM userptr object, -EFAULT is >>> returned only when the object is actually bound. >>> >>> The xf86-video-intel userspace dri

[PATCH] drm/i915/userptr: detect un-GUP-able pages early

2021-01-16 Thread Jinoh Kang
If GUP-ineligible pages are passed to a GEM userptr object, -EFAULT is returned only when the object is actually bound. The xf86-video-intel userspace driver cannot differentiate this condition, and marks the GPU as wedged. This not only disables graphics acceleration but may also cripple other f

Re: [PATCH] drm/i915/userptr: detect un-GUP-able pages early

2021-01-16 Thread Jinoh Kang
On 1/15/21 4:56 PM, Chris Wilson wrote: > Quoting Jinoh Kang (2021-01-15 16:23:31) >> If GUP-ineligible pages are passed to a GEM userptr object, -EFAULT is >> returned only when the object is actually bound. >> >> The xf86-video-intel userspace driver cannot differentiate this >> condition, and ma

Re: [PATCH] drm/i915/userptr: detect un-GUP-able pages early

2021-01-15 Thread Chris Wilson
Quoting Chris Wilson (2021-01-15 16:56:42) > Quoting Jinoh Kang (2021-01-15 16:23:31) > > If GUP-ineligible pages are passed to a GEM userptr object, -EFAULT is > > returned only when the object is actually bound. > > > > The xf86-video-intel userspace driver cannot differentiate this > > conditio

Re: [PATCH] drm/i915/userptr: detect un-GUP-able pages early

2021-01-15 Thread Chris Wilson
Quoting Jinoh Kang (2021-01-15 16:23:31) > If GUP-ineligible pages are passed to a GEM userptr object, -EFAULT is > returned only when the object is actually bound. > > The xf86-video-intel userspace driver cannot differentiate this > condition, and marks the GPU as wedged. The idea was to call g