[PATCH] D20494: [CUDA] Add section to docs about controlling fp optimizations.

[Artem Belevich via llvm-commits]

tra added inline comments.

================
Comment at: docs/CompileCudaWithLLVM.rst:160
@@ +159,3 @@
+
+* ``-ffp-contract={on,off,fast}`` (default to ``fast`` for CUDA device code)
+  Controls whether the compiler emits fused multiply-add operations.
----------------
It's actually for both host and device. The idea was to keep host & device behavior as close to each other as we can.


http://reviews.llvm.org/D20494