Kevin Qin 79c6a4f347 [AArch64 NEON] Lower SELECT_CC with vector operand.
When the scalar compare is between floating point and operands are
vector, we custom lower SELECT_CC to use NEON SIMD compare for
generating less instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@200365 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-29 01:57:30 +00:00
..
2013-10-09 07:53:57 +00:00
2013-10-09 07:53:57 +00:00