On Mon, Nov 14, 2016 at 3:57 PM, Brian Starkey <brian.starkey@xxxxxxx> wrote: > I was just writing some internal docs, and it occurred to me that the > out-fence implementation here doesn't seem to match what we discussed > with Ville a few weeks back (which had completely slipped my mind). > > Did the idea of returning -1 fences for multiple commits within a > frame get dropped? I didn't see any discussion further than that > thread on v5 from October: > http://lkml.iu.edu/hypermail/linux/kernel/1610.2/04727.html Atm we only support a queue depth of 1, and not faster than vblank. This was just discussions to make sure we're not drawing ourselves into a corner with this uabi. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
