[PATCH] D26855: New unsafe-fp-math implementation for X86 target

Chandler Carruth via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Jan 18 19:03:12 PST 2017


chandlerc added a comment.

Unless the change is trivially and obviously an improvement on inspection of the result, I think you need data before making it. =] I've not looked at this particular change in enough detail to say whether it satisfies that. Maybe Gerolf or Craig could.

If there are compile time concerns (which actually came up) and they deal with floating point and unsafe floating point in particular, then I don't think a bootstrap is a useful way to measure anything one way or the other (there just isn't enough floating point).

In https://reviews.llvm.org/D26855#649578, @avt77 wrote:

> What is "Eigen project"? Could you point me to it?


http://eigen.tuxfamily.org/


https://reviews.llvm.org/D26855





More information about the llvm-commits mailing list