[llvm] [AArch64][SVE] Lower unpredicated loads/stores as fixed LDR/STR with -msve-vector-bits=128. (PR #127500)

[Ricardo Jesus via llvm-commits]

================
@@ -23550,6 +23550,31 @@ static SDValue combineV3I8LoadExt(LoadSDNode *LD, SelectionDAG &DAG) {
   return DAG.getMergeValues({Extract, TokenFactor}, DL);
 }
 
+// Replace scalable loads with fixed loads when vscale_range(1, 1).
+// This enables further optimisations such as LDP folds.
+static SDValue combineVScale1Load(LoadSDNode *LD, SelectionDAG &DAG,
+                                  const AArch64Subtarget *Subtarget) {
+  EVT MemVT = LD->getMemoryVT();
+  if (!Subtarget->isNeonAvailable() || !MemVT.isScalableVector() ||
+      Subtarget->getMaxSVEVectorSizeInBits() != AArch64::SVEBitsPerBlock)
+    return SDValue();
+
+  // Skip unpacked types given their different layouts between Neon and SVE.
+  if (MemVT.getSizeInBits().getKnownMinValue() != AArch64::SVEBitsPerBlock)
+    return SDValue();
+
+  SDLoc DL(LD);
+  MVT NewVT = MVT::getVectorVT(MemVT.getVectorElementType().getSimpleVT(),
+                               MemVT.getVectorMinNumElements());
+  SDValue NewLoad = DAG.getLoad(
+      NewVT, DL, LD->getChain(), LD->getBasePtr(), LD->getPointerInfo(),
+      LD->getOriginalAlign(), LD->getMemOperand()->getFlags(), LD->getAAInfo());
+  SDValue Insert = convertToScalableVector(DAG, MemVT, NewLoad);
+  SDValue TokenFactor = DAG.getNode(ISD::TokenFactor, DL, MVT::Other,
+                                    {SDValue(cast<SDNode>(NewLoad), 1)});
----------------
rj-jesus wrote:

Thank you, you're right, it's not. I've removed it.

https://github.com/llvm/llvm-project/pull/127500