projects / profile / ivi / mesa.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 0ed8c56) | patch
gallivm: faster iround implementation for sse2
authorRoland Scheidegger <sroland@vmware.com>
Fri, 8 Oct 2010 16:38:25 +0000 (18:38 +0200)
committerRoland Scheidegger <sroland@vmware.com>
Fri, 8 Oct 2010 22:36:37 +0000 (00:36 +0200)
commitcb3af2b43439088fc0be0085df8b1ee1a91f33c7
treeba1562bd09e7539a9bfc4ac5bffbc62c0684d606tree | snapshot
parent0ed8c56bfe4ff5e3d451ec4791ee2cd64fb3daf2commit | diff
gallivm: faster iround implementation for sse2

sse2 supports round to nearest directly (or rather, assuming default nearest
rounding mode in MXCSR). Use intrinsic to use this rather than round (sse41)
or bit manipulation whenever possible.
src/gallium/auxiliary/gallivm/lp_bld_arit.c diff | blob | history
Domain: Graphics System / GL;