[AArch64] Combine getActiveLaneMask with vector_extract

... into a `whilelo` instruction with a pair of predicate registers.

Author: momchil-velikov

llvm/lib/Target/AArch64/AArch64ISelLowering.cpp

@@ -1847,8 +1847,8 @@ void AArch64TargetLowering::addTypeForNEON(MVT VT) {
 
 bool AArch64TargetLowering::shouldExpandGetActiveLaneMask(EVT ResVT,
                                                           EVT OpVT) const {
-  // Only SVE has a 1:1 mapping from intrinsic -> instruction (whilelo).
-  if (!Subtarget->hasSVE())
+  // Only SVE/SME has a 1:1 mapping from intrinsic -> instruction (whilelo).
+  if (!Subtarget->hasSVEorSME())
     return true;
 
   // We can only support legal predicate result types. We can use the SVE
@@ -20481,6 +20481,61 @@ static SDValue convertMergedOpToPredOp(SDNode *N, unsigned Opc,
   return SDValue();
 }
 
+static SDValue tryCombineWhileLo(SDNode *N,
+                                 TargetLowering::DAGCombinerInfo &DCI,
+                                 const AArch64Subtarget *Subtarget) {
+  if (DCI.isBeforeLegalize())
+    return SDValue();
+
+  if (!Subtarget->hasSVE2p1() && !Subtarget->hasSME2())
+    return SDValue();
+
+  if (!N->hasNUsesOfValue(2, 0))
+    return SDValue();
+
+  const uint64_t HalfSize = N->getValueType(0).getVectorMinNumElements() / 2;
+  if (HalfSize < 2)
+    return SDValue();
+
+  auto It = N->use_begin();
+  SDNode *Lo = *It++;
+  SDNode *Hi = *It;
+
+  uint64_t OffLo, OffHi;
+  if (Lo->getOpcode() != ISD::EXTRACT_SUBVECTOR ||
+      !isIntImmediate(Lo->getOperand(1).getNode(), OffLo) ||
+      Hi->getOpcode() != ISD::EXTRACT_SUBVECTOR ||
+      !isIntImmediate(Hi->getOperand(1).getNode(), OffHi))
+    return SDValue();
+
+  if (OffLo > OffHi) {
+    std::swap(Lo, Hi);
+    std::swap(OffLo, OffHi);
+  }
+
+  if (OffLo != 0 || OffHi != HalfSize)
+    return SDValue();
+
+  SelectionDAG &DAG = DCI.DAG;
+  SDLoc DL(N);
+  SDValue ID =
+      DAG.getTargetConstant(Intrinsic::aarch64_sve_whilelo_x2, DL, MVT::i64);
+  SDValue Idx = N->getOperand(1);
+  SDValue TC = N->getOperand(2);
+  if (Idx.getValueType() != MVT::i64) {
+    Idx = DAG.getZExtOrTrunc(Idx, DL, MVT::i64);
+    TC = DAG.getZExtOrTrunc(TC, DL, MVT::i64);
+  }
+  auto R =
+      DAG.getNode(ISD::INTRINSIC_WO_CHAIN, DL,
+                  {Lo->getValueType(0), Hi->getValueType(0)}, {ID, Idx, TC});
+
+  DCI.CombineTo(Lo, R.getValue(0));
+  DCI.CombineTo(Hi, R.getValue(1));
+
+  return SDValue(N, 0);
+}
+
 static SDValue performIntrinsicCombine(SDNode *N,
                                        TargetLowering::DAGCombinerInfo &DCI,
                                        const AArch64Subtarget *Subtarget) {
@@ -20811,6 +20866,8 @@ static SDValue performIntrinsicCombine(SDNode *N,
   case Intrinsic::aarch64_sve_ptest_last:
     return getPTest(DAG, N->getValueType(0), N->getOperand(1), N->getOperand(2),
                     AArch64CC::LAST_ACTIVE);
+  case Intrinsic::aarch64_sve_whilelo:
+    return tryCombineWhileLo(N, DCI, Subtarget);
   }
   return SDValue();
 }

llvm/test/CodeGen/AArch64/active_lane_mask.ll

@@ -1,5 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme < %s | FileCheck %s
 
 ; == Scalable ==