[llvm] [AArch64] Support wide interleaved store combine (PR #217856)

Kamlesh Kumar via llvm-commits llvm-commits at lists.llvm.org
Wed Sep 2 09:49:29 PDT 2026


================
@@ -28055,26 +28055,71 @@ performInterleavedStoreCombine(SDNode *N, TargetLowering::DAGCombinerInfo &DCI,
 
   const AArch64Subtarget &Subtarget = DAG.getSubtarget<AArch64Subtarget>();
   // Fixed length vector using NEON
-  if (!IsMasked && (SubBits == 64 || SubBits == 128) &&
-      Subtarget.isNeonAvailable()) {
+  if (!IsMasked && Subtarget.isNeonAvailable()) {
----------------
kamleshbhalui wrote:

I experimented with your suggestions, and It seems nice. I will update this PR.

https://github.com/llvm/llvm-project/pull/217856


More information about the llvm-commits mailing list