[llvm] [X86] Prefer SIMD min/max/abs for scalars when staying in XMM domain (PR #210654)
Shaurya Srivastava via llvm-commits
llvm-commits at lists.llvm.org
Thu Jul 23 22:05:11 PDT 2026
Shaurya2k06 wrote:
i32 path is SSE4.1 , i64 is already VLX-gated . VLX-only would drop the #210569 SSE4.1 case.
https://github.com/llvm/llvm-project/pull/210654
More information about the llvm-commits
mailing list