Use an integer constant for FABS / FNEG (x86).
authorSanjay Patel <spatel@rotateright.com>
Mon, 1 Sep 2014 19:01:47 +0000 (19:01 +0000)
committerSanjay Patel <spatel@rotateright.com>
Mon, 1 Sep 2014 19:01:47 +0000 (19:01 +0000)
commit3c7bd3fbefbe6385321befa4c9cc4e426e991a12
tree89aa9f9bf4d3f3deee68a6e9699e1b664d358827
parent73f8eff19e3b0d92ec41a4857b487e713b5e02b8
Use an integer constant for FABS / FNEG (x86).

This change will ease refactoring LowerFABS() and LowerFNEG()
since they have a lot of overlap.

Remove the creation of a floating point constant from an integer
because it's going to be used for a bitwise integer op anyway.

No change to codegen expected, but the verbose comment string
for asm output may change from float values to hex (integer),
depending on whether the constant already exists or not.

Differential Revision: http://reviews.llvm.org/D5052

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@216889 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp