Add an intrinsic and codegen support for fused multiply-accumulate. The intent
is to use this for architectures that have a native FMA instruction.

llvm-svn: 134742
20 files changed
tree: e0eed8377e924c9f1a203f2999cc9df109454645
  1. clang/
  2. compiler-rt/
  3. debuginfo-tests/
  4. libcxx/
  5. libcxxabi/
  6. lldb/
  7. llvm/
  8. polly/