[PATCH] D38533: [CGP]Restrict complex select/phi case for optimizeMemoryInst

[Serguei Katkov via Phabricator via llvm-commits]

skatkov added a comment.

In https://reviews.llvm.org/D38533#890238, @john.brawn wrote:

> There's no fundamental reason why a mixture of phi and select can't be handled, but this patch probably makes more sense in the context of later patches. I'll hold off reviewing it until then.


I tend to agree with you, right now I'm re-writing my main patch to support both Select and Phi at the same time. It seems it works. Most probably I will upload a first version of the patch tomorrow.


https://reviews.llvm.org/D38533