Craig Topper
|
79189e25c8
|
Don't use v16i32 for load pattern matching. All 512-bit loads are cated to v8i64.
llvm-svn: 188534
|
2013-08-16 06:07:34 +00:00 |
|
Craig Topper
|
1c614b247d
|
Revert r188449 as it turns out we're just missing the instructions that need the v16i32/v16f32 matching.
llvm-svn: 188454
|
2013-08-15 08:38:25 +00:00 |
|
Craig Topper
|
b1acbb9cab
|
Don't let isPermImmMask handle v16i32 since VPERMI doesn't match on that type. Remove 128-bit vector handling from isPermImmMask too, it's covered by isPSHUFDMask.
llvm-svn: 188449
|
2013-08-15 07:30:51 +00:00 |
|
Elena Demikhovsky
|
66a9e4f863
|
AVX-512: Added VPERM* instructons and MOV* zmm-to-zmm instructions.
Added a test for shuffles using VPERM.
llvm-svn: 188147
|
2013-08-11 07:55:09 +00:00 |
|