[llvm] [X86] Fix redundant zeroing operations when casting ints to FP vectors (PR #212065)

Phoebe Wang via llvm-commits llvm-commits at lists.llvm.org
Mon Jul 27 20:35:24 PDT 2026


================
@@ -325,7 +325,7 @@ let Predicates = [UseAVX, OptForSize] in {
               (v4i32 (EXTRACT_SUBREG (v8i32 VR256:$src), sub_xmm)))), sub_xmm)>;
 }
 
-let Predicates = [UseSSE1, NoSSE41_Or_OptForSize] in {
+let Predicates = [UseSSE1] in {
----------------
phoebewang wrote:

Is the change related here? Any test shows it?

https://github.com/llvm/llvm-project/pull/212065


More information about the llvm-commits mailing list