Improve implementation of __masked_store_blend_64() for AVX target by doing two 8-wide 32-bit blends rather than serializing. Fixes issue #29

This commit is contained in:
Matt Pharr
2011-06-28 20:52:06 -07:00
parent ce7978ae74
commit 86de910ecd
2 changed files with 48 additions and 6 deletions

1
.gitignore vendored
View File

@@ -4,3 +4,4 @@ depend
ispc
ispc_test
objs
docs/doxygen