[PATCH] [x86] set default reciprocal (division and square root) codegen to match GCC

[Akira Hatanaka]

I just want to make sure I understand where this is headed. The recip flag and refinement step are eventually going to be added to the IR fast-math flags and propagated to the DAG nodes, is that correct?


http://reviews.llvm.org/D10396

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/