> If you compile it with -fno-math-errno then pow gets declared as > > declare double @pow(double, double) nounwind readonly > > [I'm not sure why it is marked readonly rather than readnone].
PS: it seems that it is marked readonly because the result depends on the floating point rounding mode, i.e. pow can be considered to read to (global) rounding mode. If you compile with -ffast-math then pow is declared readnone. This sort of subtle distinction seems quite important to me, and I'm not sure LLVM was making it before. _______________________________________________ llvm-commits mailing list llvm-commits@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
