mesa/src
Kenneth Graunke 2b6817c91c i965: Use the correct number of threads for compute shaders.
We were programming the number of threads per subslice, when we should
have been programming the total number of threads on the GPU as a whole.

Thanks to Curro and Jordan for helping track this down!

On Skylake GT3e:
- Improves performance in Unreal's Elemental Demo by roughly 1.5-1.7x.
- Improves performance in Synmark's Gl43CSDof by roughly 3.7x.
- Improves performance in Synmark's Gl43GSCloth by roughly 1.18x.

On Broadwell GT2:
- Improves performance in Unreal's Elemental Demo by roughly 1.2-1.5x.
- Improves performance in Synmark's Gl43CSDof by roughly 2.0x.
- Improves performance in Synmark's Gl43GSCloth by 1.47035% +/-
  0.255654% (n=25).

On Haswell GT3e:
- Improves performance in Unreal's Elemental Demo (in GL 4.3 mode)
  by roughly 1.10x.
- Improves performance in Synmark's Gl43CSDof by roughly 1.18x.
- Decreases performance in Synmark's Gl43CSCloth by -1.99484% +/-
  0.432771% (n=64).

On Ivybridge GT2:
- Improves performance in Unreal's Elemental Demo (in GL 4.2 mode)
  by roughly 1.03x.
- Improves performance in Synmark's G/43CSDof by roughly 1.25x.
- No change in Synmark's Gl43CSCloth (n=28).

Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
(cherry picked from commit 0fb85ac08d)
2016-06-15 09:29:11 +01:00
..
compiler glsl: geom shader max_vertices layout must match. 2016-06-14 15:48:38 +01:00
egl egl: android: don't add the image loader extension for !render_node 2016-06-07 12:35:40 +01:00
gallium gallium: Fix region overlap conditions for rectangles with a shared edge 2016-06-15 09:29:10 +01:00
gbm gbm: remove define _DEFAULT_SOURCE 2016-05-23 12:09:11 +01:00
getopt
glx glx: fix crash with bad fbconfig 2016-06-14 15:48:39 +01:00
gtest
hgl
intel i965: Fix shared local memory size for Gen9+. 2016-06-15 09:29:10 +01:00
loader loader: add a libdrm case for loader_get_device_name_for_fd 2016-04-21 14:41:41 +01:00
mapi mesa: add drawbuffer argument to ClearNamedFramebufferfi 2016-06-15 09:29:10 +01:00
mesa i965: Use the correct number of threads for compute shaders. 2016-06-15 09:29:11 +01:00
util mesa: Add .gitignore entries for make check binaries 2016-05-25 09:41:44 -07:00
Makefile.am mesa: automake: distclean git_sha1.h when building OOT 2016-06-07 12:35:53 +01:00
SConscript glsl: move to compiler/ 2016-01-26 16:08:33 +00:00