gallivm,llvmpipe: Use 4-wide vectors on AMD Bulldozer.
authorJosé Fonseca <jfonseca@vmware.com>
Fri, 31 Aug 2012 16:01:50 +0000 (17:01 +0100)
committerJosé Fonseca <jfonseca@vmware.com>
Tue, 4 Sep 2012 07:49:00 +0000 (08:49 +0100)
commit7eb504019731368fd55f01e0264b195d4f99ae93
tree5e64eeb49422420d7b82356462bcc759bc27d270
parent9a31e090efb15ec34e7a1a5e707d600a11d74925
gallivm,llvmpipe: Use 4-wide vectors on AMD Bulldozer.

8-wide vectors is slower.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
src/gallium/auxiliary/gallivm/lp_bld_init.c
src/gallium/auxiliary/util/u_cpu_detect.c
src/gallium/auxiliary/util/u_cpu_detect.h