James Molloy ee0d992b07 Add support for fast-math flags to the FCmp instruction. ... FCmp behaves a lot like a floating-point binary operator in many ways, and can benefit from fast-math information. Flags such as nsz and nnan can affect if this fcmp (in combination with a select) can be treated as a fminnum/fmaxnum operation. This adds backwards-compatible bitcode support, IR parsing and writing, LangRef changes and IRBuilder changes. I'll need to audit InstSimplify and InstCombine in a followup to find places where flags should be copied. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241901 91177308-0d34-0410-b5e6-96231b3b80d8		2015-07-10 12:52:00 +00:00
..
llvm	Add support for fast-math flags to the FCmp instruction.	2015-07-10 12:52:00 +00:00
llvm-c	Revert the new EH instructions	2015-07-10 07:15:17 +00:00