Quoting Tvrtko Ursulin (2017-07-04 12:45:59)
>
> On 04/07/2017 11:06, Chris Wilson wrote:
> > Quoting Tvrtko Ursulin (2017-07-04 10:59:29)
> >>
> >> On 22/06/2017 17:02, Chris Wilson wrote:
> >>> Commit fabef825626d ("drm/i915: Drop struct_mutex around frontbuffer
> >>> flushes") adds a dependency to ifbdev->vma when flushing the framebufer,
> >>> but the checks are only against the existence of the ifbdev->fb and not
> >>> against ifbdev->vma. This leaves a window of opportunity where we may
> >>> try to operate on the fbdev prior to it being probed (thanks to
> >>> asynchronous booting).
> >>
> >> How about changing the intelfb_alloc to not write the fb to ifbdev but
> >> instead return it to the caller, so intelfb_create could then set the
> >> ifbdev->fb and ifbdev->vma atomically under the mutex?
> >
> > Midlayer mishap.
>
> After an IRC chat I see that the issue is intel_fbdev_init_bios also
> sets the fb and not vma, which I missed earlier.
>
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
I'll take that, and push with a cc: stable@
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
