projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 28b6c23) | patch
AVX-512: Fixed a bug in OR/XOR operations for 512-bit FP values on KNL.
authorElena Demikhovsky <elena.demikhovsky@intel.com>
Sun, 13 Sep 2015 08:15:15 +0000 (08:15 +0000)
committerElena Demikhovsky <elena.demikhovsky@intel.com>
Sun, 13 Sep 2015 08:15:15 +0000 (08:15 +0000)
commitb21635658f4b9b7c7e28ff035e731903545646f0
tree74c437a00c10c4b8f99d4282db5a06528eca6ef6tree | snapshot
parent28b6c23c41301dfdbd55a998583503498d984a3fcommit | diff
AVX-512: Fixed a bug in OR/XOR operations for 512-bit FP values on KNL.
KNL does not have VXORPS, VORPS for 512-bit values.
I use integer VPXOR, VPOR that actually do the same.

X86ISD::FXOR/FOR are generated as a result of FSUB combining.

Differential Revision: http://reviews.llvm.org/D12753

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@247523 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/avx512-arith.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors