[PATCH] D112760: Require 'contract' fast-math flag for FMA generation

[Andy Kaylor via Phabricator via llvm-commits]

andrew.w.kaylor added a comment.

In D112760#3101260 <https://reviews.llvm.org/D112760#3101260>, @arsenm wrote:

> This attribute has never been handled consistently in the backend. We have attributes approximately corresponding to each of the individual fast math flags, so the mirror of the IR controls would be to just set all of those attributes. I've never been sure how unsafe-fp-math fits in here, other than just a convenience to avoid setting all the others. I would also like to get rid of it, since you should be checking the individual properties anyway.

I'm glad to hear that. It looked like you (or someone else) had put a good bit of work into making various combinations of these options behave consistently with the AMDGPU backend, and I wasn't sure if you needed the global settings for some use case.

Can you add your feedback on my RFC thread (https://lists.llvm.org/pipermail/llvm-dev/2021-October/153460.html)? I think we're approaching consensus there, but your input would be very valuable.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D112760/new/

https://reviews.llvm.org/D112760