commit | 2b918388ab0e15febfc5163e52ec48791bc25d47 | [log] [tgz] |
---|---|---|
author | Sanjay Patel <spatel@rotateright.com> | Wed Oct 01 21:20:06 2014 +0000 |
committer | Sanjay Patel <spatel@rotateright.com> | Wed Oct 01 21:20:06 2014 +0000 |
tree | 6d146e1b3e94c55357ffeb12b3e15e8c3af79cc7 | |
parent | 42f40dbbefe9d38407fb892d952bb05858647ad3 [diff] |
Lower FNEG ( FABS (x) ) -> FNABS (x) [X86 codegen] PR20578 Negative FABS of either a scalar or vector should be handled the same way on x86 with SSE/AVX: a single OR instruction of the FP operand with a constant to light up the sign bit(s). http://llvm.org/bugs/show_bug.cgi?id=20578 Differential Revision: http://reviews.llvm.org/D5201 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218822 91177308-0d34-0410-b5e6-96231b3b80d8