[Intel-gfx] [PATCH 00/13] RFC: pipelined fencing

Daniel Vetter daniel at ffwll.ch
Fri Feb 19 10:10:18 CET 2010


On Thu, Feb 18, 2010 at 09:14:02PM +0100, Daniel Vetter wrote:
> I've squeezed out the last (known) bug in this patch pile earlier this
> week and I'm now in benchmarking-mode. I'm trying to resubmit everything
> for proper review till weekend for I'll be offline next week. After all,
> this was just an rfc never intended to be merged.

Ok, this patch pile needs some more time. All else equal, cairo on xlib
can regress with pipelined fencing by up to 20%, and this shouldn't
happen.

I think the lazy request emit refactoring stuff is worth it on its own,
though. Depending upon what benchmarking reveals next week, I'll submit
that part alone when I'm back online.

-Daniel
-- 
Daniel Vetter
Mail: daniel at ffwll.ch
Mobile: +41 (0)79 365 57 48



More information about the Intel-gfx mailing list