[Intel-gfx] [PATCH] uxa: Do a little hack to use bo_pin instead of bo_map.
keithp at keithp.com
Thu Dec 18 00:00:18 CET 2008
On Wed, 2008-12-17 at 14:40 -0800, Eric Anholt wrote:
> Because of how fallbacky the uxa rendering core is, and our inability (without
> wfb in userland or page faulting in the kernel) to tell the kernel just where
> we're going to fall back, the clflush overhead can become outrageous, for
> example with emacs and xcompmgr. Instead of using drm_intel_bo_map, pin the
> buffer and do the fallback to the aperture mapping. This gets us the bad old
> performance that fb is designed for, instead of bad new performance.
This looks good to me; we'll be using something with equivalent
performance when we get the kernel GTT mapping stuff going, so it seems
reasonable to use this now.
This should clear up some of our big performance differences between uxa
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 189 bytes
Desc: This is a digitally signed message part
More information about the Intel-gfx