[PATCH] D51203: AMDGPU: Handle 32-bit address wraparounds for SMRD opcodes

[Nicolai Hähnle via Phabricator via llvm-commits]

nhaehnle added a comment.

That would be good, yeah. Note here it's not so much a case of preserving NUW as it is of deducing as much NUW as possible from getelementptrs.


Repository:
  rL LLVM

https://reviews.llvm.org/D51203