[llvm] [X86] LowerSelect - use BLENDV for scalar selection if not all operands are multi use (PR #125853)

via llvm-commits llvm-commits at lists.llvm.org
Wed Feb 5 13:56:51 PST 2025


goldsteinn wrote:

> Yes, removing the limit enables additional folds, some are actually beneficial but others cause extra moves.

Which tests?

https://github.com/llvm/llvm-project/pull/125853


More information about the llvm-commits mailing list