[llvm] [Codegen][LegalizeIntegerTypes] Improve shift through stack (PR #96151)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Wed Jul 3 05:25:23 PDT 2024


https://github.com/arsenm commented:

I think the code change looks fine, but the AArch64 and RISCV cases look like regressions. Is there an explanation for those? Is the type transform loop excessively widening these now? 

https://github.com/llvm/llvm-project/pull/96151


More information about the llvm-commits mailing list