projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 00aecd9) | patch
[AVX-512] Add more patterns for masked and broadcasted logical operations where the...
authorCraig Topper <craig.topper@gmail.com>
Fri, 2 Sep 2016 05:29:13 +0000 (05:29 +0000)
committerCraig Topper <craig.topper@gmail.com>
Fri, 2 Sep 2016 05:29:13 +0000 (05:29 +0000)
commit45d65030899ad0664a83b7cdbc61d8b8bd5f8fa0
tree223c9abc6287be2b1d4dcc6b508671e7599cee5etree | snapshot
parent00aecd97bfa1ecee6bb548c2ba522f41ac86c674commit | diff
[AVX-512] Add more patterns for masked and broadcasted logical operations where the select or broadcast has a floating point type.

These are needed in order to remove the masked floating point logical operation intrinsics and use native IR.

llvm-svn: 280465
llvm/lib/Target/X86/X86InstrAVX512.td diff | blob | history
llvm/test/CodeGen/X86/avx512-arith.ll diff | blob | history
llvm/test/CodeGen/X86/vec_fabs.ll diff | blob | history
Domain: System / Toolchain;