The attached patch uses LLVM IR's fast-math flags when creating the IR. This allows users of clang's -ffast-math and -ffinite-math-only flags to take advantage of the new and upcoming fast-math optimizations done in LLVM. Test cases are also attached.
use-ir-fast-math.patch
Description: Binary data
fast-math.c
Description: Binary data
finite-math.c
Description: Binary data
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
