Matched some basic ISD::AVGFLOORU patterns

Author: Sh0g0-1758

llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp

@@ -2826,6 +2826,36 @@ SDValue DAGCombiner::visitADDLike(SDNode *N) {
   return SDValue();
 }
 
+// Attempt to form ext(avgflooru(A, B)) from add(and(A, B), lshr(xor(A, B), 1))
+static SDValue combineFixedwidthToAVG(SDNode *N, SelectionDAG &DAG) {
+  assert(N->getOpcode() == ISD::ADD && "ADD node is required here");
+  SDValue And = N->getOperand(0);
+  SDValue Lshr = N->getOperand(1);
+  if (And.getOpcode() != ISD::AND || Lshr.getOpcode() != ISD::SRL)
+    return SDValue();
+  SDValue Xor = Lshr.getOperand(0);
+  if (Xor.getOpcode() != ISD::XOR)
+    return SDValue();
+  SDValue And1 = And.getOperand(0);
+  SDValue And2 = And.getOperand(1);
+  SDValue Xor1 = Xor.getOperand(0);
+  SDValue Xor2 = Xor.getOperand(1);
+  if (Xor1 != And1 && Xor2 != And2)
+    return SDValue();
+  // Is the right shift using an immediate value of 1?
+  ConstantSDNode *N1C = isConstOrConstSplat(Lshr.getOperand(1));
+  if (!N1C || N1C->getAPIntValue() != 1)
+    return SDValue();
+  EVT VT = And.getValueType();
+  SDLoc DL(N);
+  const TargetLowering &TLI = DAG.getTargetLoweringInfo();
+  if (!TLI.isOperationLegalOrCustom(ISD::AVGFLOORU, VT))
+    return SDValue();
+  return DAG.getNode(ISD::AVGFLOORU, DL, VT,
+                     DAG.getExtOrTrunc(false, And1, DL, VT),
+                     DAG.getExtOrTrunc(false, And2, DL, VT));
+}
+
 SDValue DAGCombiner::visitADD(SDNode *N) {
   SDValue N0 = N->getOperand(0);
   SDValue N1 = N->getOperand(1);
@@ -2841,6 +2871,10 @@ SDValue DAGCombiner::visitADD(SDNode *N) {
   if (SDValue V = foldAddSubOfSignBit(N, DAG))
     return V;
 
+  // Try to match AVG fixedwidth pattern
+  if (SDValue V = combineFixedwidthToAVG(N, DAG))
+    return V;
+
   // fold (a+b) -> (a|b) iff a and b share no bits.
   if ((!LegalOperations || TLI.isOperationLegal(ISD::OR, VT)) &&
       DAG.haveNoCommonBitsSet(N0, N1))

llvm/test/CodeGen/AArch64/hadd-combine.ll

@@ -859,6 +859,21 @@ define <4 x i32> @urhadd_v4i32(<4 x i32> %x) {
   ret <4 x i32> %r
 }
 
+define i4 @fixedwidth(i4 %a0, i4 %a1)  {
+; CHECK-LABEL: fixedwidth:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    eor w8, w0, w1
+; CHECK-NEXT:    and w9, w0, w1
+; CHECK-NEXT:    and w8, w8, #0xe
+; CHECK-NEXT:    add w0, w9, w8, lsr #1
+; CHECK-NEXT:    ret
+  %and = and i4 %a0, %a1
+  %xor = xor i4 %a0, %a1
+  %srl = lshr i4 %xor, 1
+  %res = add i4 %and, %srl
+  ret i4 %res
+}
+
 declare <8 x i8> @llvm.aarch64.neon.shadd.v8i8(<8 x i8>, <8 x i8>)
 declare <4 x i16> @llvm.aarch64.neon.shadd.v4i16(<4 x i16>, <4 x i16>)
 declare <2 x i32> @llvm.aarch64.neon.shadd.v2i32(<2 x i32>, <2 x i32>)