[PATCH] D37989: InstCombine: Insert missing canonicalizes

[escha via Phabricator via llvm-commits]

escha added a comment.

Regardless of semantics, this patch almost surely causes a major problem for us.

I originally added fcanonicalize for the purpose of our numerics folks, who wanted to be able to force canonicalizations in numerics tests and so forth. It was meant to be a "dumb intrinsic" that would just blindly emit a no-op float instruction to force canonicalization of the input.

If we start emitting fcanonicalize all over the place in instcombine, it's going to destroy our codegen quality and significantly increase instruction count.

Do we really need to do this for the sake of float80 or something?


https://reviews.llvm.org/D37989