[PATCH 3/4] drm: introduce DRM_CAP_ATOMIC_ASYNC_PAGE_FLIP

Simon Ser contact at emersion.fr
Tue Aug 30 12:58:49 UTC 2022


(Oops, I replied to the wrong thread. Re-sending to the correct one.)

On Tuesday, August 30th, 2022 at 10:41, Michel Dänzer <michel.daenzer at mailbox.org> wrote:

> > For the atomic uAPI, we need to pick on of these two approaches:
> >
> > 1. Let the kernel fall back to a sync flip if async isn't possible. This
> >    simplifies user-space, but then user-space has no reliable way to figure out
> >    what really happened (sync or async?). That could be fixed with a new
> >    read-only CRTC prop indicating whether the last flip was async or not.
> >    However, maybe someone will come up in the future with user-space which
> >    needs to only apply an update if async flip is possible, in which case this
> >    approach falls short.
>
> The future is now. :)
>
> As I described in the documentation patch discussion, this approach would
> make it tricky for a Wayland compositor to decide if it should use an async
> commit (which needs to be done ASAP to serve the intended purpose) or not (in
> which case the compositor may want to delay the commit as long as possible
> for minimal latency).

Ah right. If an async page-flip is not possible, then a Wayland compositor
could want to wait the "last moment" before the next vblank to see if a more
up-to-date buffer is available.

With that + Xorg usage, I think we have a rather solid case for failing async
flips rather than silently falling back to vsync.


More information about the amd-gfx mailing list