Eric Anholt c9e48e5b08 i965: Generalize VS compute-to-MRF for compute-to-another-GRF, too. ... No statistically significant performance difference on glbenchmark 2.7 (n=60). It reduces cycles spent in the vertex shader by 3.3% +/- 0.8% (n=5), but that's only about .3% of all cycles spent according to the fixed shader_time. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>		2012-12-14 16:06:35 -08:00
..
egl	egl/wayland: Dispatch the event queue before get_buffers	2012-11-30 17:05:50 -05:00
gallium	draw/llvmpipe: fix transform feedback position + enable other extensions	2012-12-14 11:34:40 +10:00
gbm	gbm: Use the kms dumb ioctls for cursor instead of libkms	2012-10-21 13:00:49 -04:00
getopt
glsl	glsl/linker: Pack between varyings.	2012-12-14 10:51:21 -08:00
glx	glx/dri2: set the __DRI2_FLUSH_DRAWABLE flag where it should be set	2012-12-07 14:19:28 +01:00
gtest	build: Fix gtest out-of-tree build	2012-08-14 10:54:39 -07:00
mapi	build: Fix GLES linkage without libglapi	2012-12-07 09:21:44 +01:00
mesa	i965: Generalize VS compute-to-MRF for compute-to-another-GRF, too.	2012-12-14 16:06:35 -08:00
.gitignore	automake: Convert src/Makefile to automake.	2012-06-11 09:28:00 -07:00
Makefile.am	automake: Convert src/Makefile to automake.	2012-06-11 09:28:00 -07:00
SConscript	Remove libGLU	2012-08-31 10:58:15 -07:00