[DAGCombiner] Use generalized pattern match for visitFSUBForFMACombine.

[fakepaper56]

Summary

The patch makes visitFSUBForFMACombine serve vp.fsub too. It helps DAGCombiner
to fuse vp.fsub and vp.fmul patterns to vp.fma.

[luke]

LGTM

llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp

private:
     SDValue visitSIGN_EXTEND_INREG(SDNode *N);
     SDValue visitEXTEND_VECTOR_INREG(SDNode *N);
     SDValue visitTRUNCATE(SDNode *N);
     SDValue visitBITCAST(SDNode *N);
     SDValue visitFREEZE(SDNode *N);
     SDValue visitBUILD_PAIR(SDNode *N);
     SDValue visitFADD(SDNode *N);
     SDValue visitVP_FADD(SDNode *N);
+    SDValue visitVP_FSUB(SDNode *N);
     SDValue visitSTRICT_FADD(SDNode *N);
     SDValue visitFSUB(SDNode *N);
     SDValue visitFMUL(SDNode *N);
     SDValue visitFMA(SDNode *N);
     SDValue visitFDIV(SDNode *N);
     SDValue visitFREM(SDNode *N);
     SDValue visitFSQRT(SDNode *N);
     SDValue visitFCOPYSIGN(SDNode *N);
private:
     SDValue visitFP_TO_FP16(SDNode *N);
     SDValue visitFP16_TO_FP(SDNode *N);
     SDValue visitFP_TO_BF16(SDNode *N);
     SDValue visitVECREDUCE(SDNode *N);
     SDValue visitVPOp(SDNode *N);
 
     template <class MatchContextClass>
     SDValue visitFADDForFMACombine(SDNode *N);
+    template <class MatchContextClass>
     SDValue visitFSUBForFMACombine(SDNode *N);
     SDValue visitFMULForFMADistributiveCombine(SDNode *N);
 
     SDValue XformToShuffleWithZero(SDNode *N);
     bool reassociationCanBreakAddressingModePattern(unsigned Opc,
                                                     const SDLoc &DL,
                                                     SDNode *N,
                                                     SDValue N0,
if (N1.getOpcode() == ISD::FP_EXTEND) {
       }
     }
   }
 
   return SDValue();
 }
 
 /// Try to perform FMA combining on a given FSUB node.
+template <class MatchContextClass>
 SDValue DAGCombiner::visitFSUBForFMACombine(SDNode *N) {
   SDValue N0 = N->getOperand(0);
   SDValue N1 = N->getOperand(1);
   EVT VT = N->getValueType(0);
   SDLoc SL(N);
-
+  MatchContextClass matcher(DAG, TLI, N);
   const TargetOptions &Options = DAG.getTarget().Options;
+
+  bool UseVP = std::is_same_v<MatchContextClass, VPMatchContext>;
+
   // Floating-point multiply-add with intermediate rounding.
-  bool HasFMAD = (LegalOperations && TLI.isFMADLegal(DAG, N));
+  // FIXME: Make isFMADLegal have specific behavior when using VPMatchContext.
+  // FIXME: Add VP_FMAD opcode.
+  bool HasFMAD = !UseVP && (LegalOperations && TLI.isFMADLegal(DAG, N));
 
   // Floating-point multiply-add without intermediate rounding.
   bool HasFMA =
       TLI.isFMAFasterThanFMulAndFAdd(DAG.getMachineFunction(), VT) &&
-      (!LegalOperations || TLI.isOperationLegalOrCustom(ISD::FMA, VT));
+      (!LegalOperations || matcher.isOperationLegalOrCustom(ISD::FMA, VT));
 
   // No valid opcode, do not combine.
   if (!HasFMAD && !HasFMA)
     return SDValue();
 
   const SDNodeFlags Flags = N->getFlags();
   bool AllowFusionGlobally = (Options.AllowFPOpFusion == FPOpFusion::Fast ||
                               Options.UnsafeFPMath || HasFMAD);
 
   // If the subtraction is not contractable, do not combine.
   if (!AllowFusionGlobally && !N->getFlags().hasAllowContract())
     return SDValue();
 
   if (TLI.generateFMAsInMachineCombiner(VT, OptLevel))
     return SDValue();
 
   // Always prefer FMAD to FMA for precision.
   unsigned PreferredFusedOpcode = HasFMAD ? ISD::FMAD : ISD::FMA;
   bool Aggressive = TLI.enableAggressiveFMAFusion(VT);
   bool NoSignedZero = Options.NoSignedZerosFPMath || Flags.hasNoSignedZeros();
 
   // Is the node an FMUL and contractable either due to global flags or
   // SDNodeFlags.
-  auto isContractableFMUL = [AllowFusionGlobally](SDValue N) {
-    if (N.getOpcode() != ISD::FMUL)
+  auto isContractableFMUL = [AllowFusionGlobally, &matcher](SDValue N) {
+    if (!matcher.match(N, ISD::FMUL))
       return false;
     return AllowFusionGlobally || N->getFlags().hasAllowContract();
   };
 
   // fold (fsub (fmul x, y), z) -> (fma x, y, (fneg z))
   auto tryToFoldXYSubZ = [&](SDValue XY, SDValue Z) {
     if (isContractableFMUL(XY) && (Aggressive || XY->hasOneUse())) {
-      return DAG.getNode(PreferredFusedOpcode, SL, VT, XY.getOperand(0),
-                         XY.getOperand(1), DAG.getNode(ISD::FNEG, SL, VT, Z));
+      return matcher.getNode(PreferredFusedOpcode, SL, VT, XY.getOperand(0),
+                             XY.getOperand(1),
+                             matcher.getNode(ISD::FNEG, SL, VT, Z));
     }
     return SDValue();
   };
 
   // fold (fsub x, (fmul y, z)) -> (fma (fneg y), z, x)
   // Note: Commutes FSUB operands.
   auto tryToFoldXSubYZ = [&](SDValue X, SDValue YZ) {
     if (isContractableFMUL(YZ) && (Aggressive || YZ->hasOneUse())) {
-      return DAG.getNode(PreferredFusedOpcode, SL, VT,
-                         DAG.getNode(ISD::FNEG, SL, VT, YZ.getOperand(0)),
-                         YZ.getOperand(1), X);
+      return matcher.getNode(
+          PreferredFusedOpcode, SL, VT,
+          matcher.getNode(ISD::FNEG, SL, VT, YZ.getOperand(0)),
     }
     return SDValue();
   };
 
   // If we have two choices trying to fold (fsub (fmul u, v), (fmul x, y)),
   // prefer to fold the multiply with fewer uses.
   if (isContractableFMUL(N0) && isContractableFMUL(N1) &&
       (N0->use_size() > N1->use_size())) {
     // fold (fsub (fmul a, b), (fmul c, d)) -> (fma (fneg c), d, (fmul a, b))
     if (SDValue V = tryToFoldXSubYZ(N0, N1))
       return V;
     // fold (fsub (fmul a, b), (fmul c, d)) -> (fma a, b, (fneg (fmul c, d)))
     if (SDValue V = tryToFoldXYSubZ(N0, N1))
       return V;
   } else {
     // fold (fsub (fmul x, y), z) -> (fma x, y, (fneg z))
     if (SDValue V = tryToFoldXYSubZ(N0, N1))
       return V;
     // fold (fsub x, (fmul y, z)) -> (fma (fneg y), z, x)
     if (SDValue V = tryToFoldXSubYZ(N0, N1))
       return V;
   }
 
   // fold (fsub (fneg (fmul, x, y)), z) -> (fma (fneg x), y, (fneg z))
-  if (N0.getOpcode() == ISD::FNEG && isContractableFMUL(N0.getOperand(0)) &&
+  if (matcher.match(N0, ISD::FNEG) && isContractableFMUL(N0.getOperand(0)) &&
       (Aggressive || (N0->hasOneUse() && N0.getOperand(0).hasOneUse()))) {
     SDValue N00 = N0.getOperand(0).getOperand(0);
     SDValue N01 = N0.getOperand(0).getOperand(1);
-    return DAG.getNode(PreferredFusedOpcode, SL, VT,
-                       DAG.getNode(ISD::FNEG, SL, VT, N00), N01,
-                       DAG.getNode(ISD::FNEG, SL, VT, N1));
+    return matcher.getNode(PreferredFusedOpcode, SL, VT,
+                           matcher.getNode(ISD::FNEG, SL, VT, N00), N01,
+                           matcher.getNode(ISD::FNEG, SL, VT, N1));
   }
 
   // Look through FP_EXTEND nodes to do more combining.
 
   // fold (fsub (fpext (fmul x, y)), z)
   //   -> (fma (fpext x), (fpext y), (fneg z))
-  if (N0.getOpcode() == ISD::FP_EXTEND) {
+  if (matcher.match(N0, ISD::FP_EXTEND)) {
     SDValue N00 = N0.getOperand(0);
     if (isContractableFMUL(N00) &&
         TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                             N00.getValueType())) {
-      return DAG.getNode(PreferredFusedOpcode, SL, VT,
-                         DAG.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(0)),
-                         DAG.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(1)),
-                         DAG.getNode(ISD::FNEG, SL, VT, N1));
+      return matcher.getNode(
+          PreferredFusedOpcode, SL, VT,
+          matcher.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(0)),
+          matcher.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(1)),
+          matcher.getNode(ISD::FNEG, SL, VT, N1));
     }
   }
 
   // fold (fsub x, (fpext (fmul y, z)))
   //   -> (fma (fneg (fpext y)), (fpext z), x)
   // Note: Commutes FSUB operands.
-  if (N1.getOpcode() == ISD::FP_EXTEND) {
+  if (matcher.match(N1, ISD::FP_EXTEND)) {
     SDValue N10 = N1.getOperand(0);
     if (isContractableFMUL(N10) &&
         TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                             N10.getValueType())) {
-      return DAG.getNode(
+      return matcher.getNode(
           PreferredFusedOpcode, SL, VT,
-          DAG.getNode(ISD::FNEG, SL, VT,
-                      DAG.getNode(ISD::FP_EXTEND, SL, VT, N10.getOperand(0))),
-          DAG.getNode(ISD::FP_EXTEND, SL, VT, N10.getOperand(1)), N0);
+          matcher.getNode(
+              ISD::FNEG, SL, VT,
+              matcher.getNode(ISD::FP_EXTEND, SL, VT, N10.getOperand(0))),
+          matcher.getNode(ISD::FP_EXTEND, SL, VT, N10.getOperand(1)), N0);
     }
   }
 
   // fold (fsub (fpext (fneg (fmul, x, y))), z)
   //   -> (fneg (fma (fpext x), (fpext y), z))
   // Note: This could be removed with appropriate canonicalization of the
   // input expression into (fneg (fadd (fpext (fmul, x, y)), z). However, the
   // orthogonal flags -fp-contract=fast and -enable-unsafe-fp-math prevent
   // from implementing the canonicalization in visitFSUB.
-  if (N0.getOpcode() == ISD::FP_EXTEND) {
+  if (matcher.match(N0, ISD::FP_EXTEND)) {
     SDValue N00 = N0.getOperand(0);
-    if (N00.getOpcode() == ISD::FNEG) {
+    if (matcher.match(N00, ISD::FNEG)) {
       SDValue N000 = N00.getOperand(0);
       if (isContractableFMUL(N000) &&
           TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                               N00.getValueType())) {
-        return DAG.getNode(
+        return matcher.getNode(
             ISD::FNEG, SL, VT,
-            DAG.getNode(PreferredFusedOpcode, SL, VT,
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(0)),
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(1)),
-                        N1));
+            matcher.getNode(
+                PreferredFusedOpcode, SL, VT,
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(0)),
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(1)),
       }
     }
   }
 
   // fold (fsub (fneg (fpext (fmul, x, y))), z)
   //   -> (fneg (fma (fpext x)), (fpext y), z)
   // Note: This could be removed with appropriate canonicalization of the
   // input expression into (fneg (fadd (fpext (fmul, x, y)), z). However, the
   // orthogonal flags -fp-contract=fast and -enable-unsafe-fp-math prevent
   // from implementing the canonicalization in visitFSUB.
-  if (N0.getOpcode() == ISD::FNEG) {
+  if (matcher.match(N0, ISD::FNEG)) {
     SDValue N00 = N0.getOperand(0);
-    if (N00.getOpcode() == ISD::FP_EXTEND) {
+    if (matcher.match(N00, ISD::FP_EXTEND)) {
       SDValue N000 = N00.getOperand(0);
       if (isContractableFMUL(N000) &&
           TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                               N000.getValueType())) {
-        return DAG.getNode(
+        return matcher.getNode(
             ISD::FNEG, SL, VT,
-            DAG.getNode(PreferredFusedOpcode, SL, VT,
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(0)),
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(1)),
-                        N1));
+            matcher.getNode(
+                PreferredFusedOpcode, SL, VT,
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(0)),
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N000.getOperand(1)),
       }
     }
   }
 
   auto isReassociable = [Options](SDNode *N) {
     return Options.UnsafeFPMath || N->getFlags().hasAllowReassociation();
   };
 
   auto isContractableAndReassociableFMUL = [&isContractableFMUL,
                                             &isReassociable](SDValue N) {
     return isContractableFMUL(N) && isReassociable(N.getNode());
   };
 
   auto isFusedOp = [&](SDValue N) {
-    unsigned Opcode = N.getOpcode();
-    return Opcode == ISD::FMA || Opcode == ISD::FMAD;
+    return matcher.match(N, ISD::FMA) || matcher.match(N, ISD::FMAD);
   };
 
   // More folding opportunities when target permits.
   if (Aggressive && isReassociable(N)) {
     bool CanFuse = Options.UnsafeFPMath || N->getFlags().hasAllowContract();
     // fold (fsub (fma x, y, (fmul u, v)), z)
     //   -> (fma x, y (fma u, v, (fneg z)))
     if (CanFuse && isFusedOp(N0) &&
         isContractableAndReassociableFMUL(N0.getOperand(2)) &&
         N0->hasOneUse() && N0.getOperand(2)->hasOneUse()) {
-      return DAG.getNode(PreferredFusedOpcode, SL, VT, N0.getOperand(0),
-                         N0.getOperand(1),
-                         DAG.getNode(PreferredFusedOpcode, SL, VT,
-                                     N0.getOperand(2).getOperand(0),
-                                     N0.getOperand(2).getOperand(1),
-                                     DAG.getNode(ISD::FNEG, SL, VT, N1)));
+      return matcher.getNode(
+          PreferredFusedOpcode, SL, VT, N0.getOperand(0), N0.getOperand(1),
+          matcher.getNode(PreferredFusedOpcode, SL, VT,
+                          matcher.getNode(ISD::FNEG, SL, VT, N1)));
     }
 
     // fold (fsub x, (fma y, z, (fmul u, v)))
     //   -> (fma (fneg y), z, (fma (fneg u), v, x))
     if (CanFuse && isFusedOp(N1) &&
         isContractableAndReassociableFMUL(N1.getOperand(2)) &&
         N1->hasOneUse() && NoSignedZero) {
       SDValue N20 = N1.getOperand(2).getOperand(0);
       SDValue N21 = N1.getOperand(2).getOperand(1);
-      return DAG.getNode(
+      return matcher.getNode(
           PreferredFusedOpcode, SL, VT,
-          DAG.getNode(ISD::FNEG, SL, VT, N1.getOperand(0)), N1.getOperand(1),
-          DAG.getNode(PreferredFusedOpcode, SL, VT,
-                      DAG.getNode(ISD::FNEG, SL, VT, N20), N21, N0));
+          matcher.getNode(ISD::FNEG, SL, VT, N1.getOperand(0)),
+          N1.getOperand(1),
+          matcher.getNode(PreferredFusedOpcode, SL, VT,
+                          matcher.getNode(ISD::FNEG, SL, VT, N20), N21, N0));
     }
 
     // fold (fsub (fma x, y, (fpext (fmul u, v))), z)
     //   -> (fma x, y (fma (fpext u), (fpext v), (fneg z)))
     if (isFusedOp(N0) && N0->hasOneUse()) {
       SDValue N02 = N0.getOperand(2);
-      if (N02.getOpcode() == ISD::FP_EXTEND) {
+      if (matcher.match(N02, ISD::FP_EXTEND)) {
         SDValue N020 = N02.getOperand(0);
         if (isContractableAndReassociableFMUL(N020) &&
             TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                                 N020.getValueType())) {
-          return DAG.getNode(
+          return matcher.getNode(
               PreferredFusedOpcode, SL, VT, N0.getOperand(0), N0.getOperand(1),
-              DAG.getNode(
+              matcher.getNode(
                   PreferredFusedOpcode, SL, VT,
-                  DAG.getNode(ISD::FP_EXTEND, SL, VT, N020.getOperand(0)),
-                  DAG.getNode(ISD::FP_EXTEND, SL, VT, N020.getOperand(1)),
-                  DAG.getNode(ISD::FNEG, SL, VT, N1)));
+                  matcher.getNode(ISD::FP_EXTEND, SL, VT, N020.getOperand(0)),
+                  matcher.getNode(ISD::FP_EXTEND, SL, VT, N020.getOperand(1)),
+                  matcher.getNode(ISD::FNEG, SL, VT, N1)));
         }
       }
     }
 
     // fold (fsub (fpext (fma x, y, (fmul u, v))), z)
     //   -> (fma (fpext x), (fpext y),
     //           (fma (fpext u), (fpext v), (fneg z)))
     // FIXME: This turns two single-precision and one double-precision
     // operation into two double-precision operations, which might not be
     // interesting for all targets, especially GPUs.
-    if (N0.getOpcode() == ISD::FP_EXTEND) {
+    if (matcher.match(N0, ISD::FP_EXTEND)) {
       SDValue N00 = N0.getOperand(0);
       if (isFusedOp(N00)) {
         SDValue N002 = N00.getOperand(2);
         if (isContractableAndReassociableFMUL(N002) &&
             TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                                 N00.getValueType())) {
-          return DAG.getNode(
+          return matcher.getNode(
               PreferredFusedOpcode, SL, VT,
-              DAG.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(0)),
-              DAG.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(1)),
-              DAG.getNode(
+              matcher.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(0)),
+              matcher.getNode(ISD::FP_EXTEND, SL, VT, N00.getOperand(1)),
+              matcher.getNode(
                   PreferredFusedOpcode, SL, VT,
-                  DAG.getNode(ISD::FP_EXTEND, SL, VT, N002.getOperand(0)),
-                  DAG.getNode(ISD::FP_EXTEND, SL, VT, N002.getOperand(1)),
-                  DAG.getNode(ISD::FNEG, SL, VT, N1)));
+                  matcher.getNode(ISD::FP_EXTEND, SL, VT, N002.getOperand(0)),
+                  matcher.getNode(ISD::FP_EXTEND, SL, VT, N002.getOperand(1)),
+                  matcher.getNode(ISD::FNEG, SL, VT, N1)));
         }
       }
     }
 
     // fold (fsub x, (fma y, z, (fpext (fmul u, v))))
     //   -> (fma (fneg y), z, (fma (fneg (fpext u)), (fpext v), x))
-    if (isFusedOp(N1) && N1.getOperand(2).getOpcode() == ISD::FP_EXTEND &&
+    if (isFusedOp(N1) && matcher.match(N1.getOperand(2), ISD::FP_EXTEND) &&
         N1->hasOneUse()) {
       SDValue N120 = N1.getOperand(2).getOperand(0);
       if (isContractableAndReassociableFMUL(N120) &&
           TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                               N120.getValueType())) {
         SDValue N1200 = N120.getOperand(0);
         SDValue N1201 = N120.getOperand(1);
-        return DAG.getNode(
+        return matcher.getNode(
+            PreferredFusedOpcode, SL, VT,
+            matcher.getNode(ISD::FNEG, SL, VT, N1.getOperand(0)),
+            N1.getOperand(1),
+            matcher.getNode(
                 PreferredFusedOpcode, SL, VT,
-            DAG.getNode(ISD::FNEG, SL, VT, N1.getOperand(0)), N1.getOperand(1),
-            DAG.getNode(PreferredFusedOpcode, SL, VT,
-                        DAG.getNode(ISD::FNEG, SL, VT,
-                                    DAG.getNode(ISD::FP_EXTEND, SL, VT, N1200)),
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N1201), N0));
+                matcher.getNode(ISD::FNEG, SL, VT,
+                                matcher.getNode(ISD::FP_EXTEND, SL, VT, N1200)),
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N1201), N0));
       }
     }
 
     // fold (fsub x, (fpext (fma y, z, (fmul u, v))))
     //   -> (fma (fneg (fpext y)), (fpext z),
     //           (fma (fneg (fpext u)), (fpext v), x))
     // FIXME: This turns two single-precision and one double-precision
     // operation into two double-precision operations, which might not be
     // interesting for all targets, especially GPUs.
-    if (N1.getOpcode() == ISD::FP_EXTEND && isFusedOp(N1.getOperand(0))) {
+    if (matcher.match(N1, ISD::FP_EXTEND) && isFusedOp(N1.getOperand(0))) {
       SDValue CvtSrc = N1.getOperand(0);
       SDValue N100 = CvtSrc.getOperand(0);
       SDValue N101 = CvtSrc.getOperand(1);
       SDValue N102 = CvtSrc.getOperand(2);
       if (isContractableAndReassociableFMUL(N102) &&
           TLI.isFPExtFoldable(DAG, PreferredFusedOpcode, VT,
                               CvtSrc.getValueType())) {
         SDValue N1020 = N102.getOperand(0);
         SDValue N1021 = N102.getOperand(1);
-        return DAG.getNode(
+        return matcher.getNode(
             PreferredFusedOpcode, SL, VT,
-            DAG.getNode(ISD::FNEG, SL, VT,
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N100)),
-            DAG.getNode(ISD::FP_EXTEND, SL, VT, N101),
-            DAG.getNode(PreferredFusedOpcode, SL, VT,
-                        DAG.getNode(ISD::FNEG, SL, VT,
-                                    DAG.getNode(ISD::FP_EXTEND, SL, VT, N1020)),
-                        DAG.getNode(ISD::FP_EXTEND, SL, VT, N1021), N0));
+            matcher.getNode(ISD::FNEG, SL, VT,
+                            matcher.getNode(ISD::FP_EXTEND, SL, VT, N100)),
+            matcher.getNode(ISD::FP_EXTEND, SL, VT, N101),
+            matcher.getNode(
+                PreferredFusedOpcode, SL, VT,
+                matcher.getNode(ISD::FNEG, SL, VT,
+                                matcher.getNode(ISD::FP_EXTEND, SL, VT, N1020)),
+                matcher.getNode(ISD::FP_EXTEND, SL, VT, N1021), N0));
       }
     }
   }
 
   return SDValue();
 }
 
 /// Try to perform FMA combining on a given FMUL node based on the distributive
SDValue DAGCombiner::visitFSUB(SDNode *N) {
   }
 
   // fold (fsub A, (fneg B)) -> (fadd A, B)
   if (SDValue NegN1 =
           TLI.getNegatedExpression(N1, DAG, LegalOperations, ForCodeSize))
     return DAG.getNode(ISD::FADD, DL, VT, N0, NegN1);
 
   // FSUB -> FMA combines:
-  if (SDValue Fused = visitFSUBForFMACombine(N)) {
+  if (SDValue Fused = visitFSUBForFMACombine<EmptyMatchContext>(N)) {
     AddToWorklist(Fused.getNode());
     return Fused;
   }
 
   return SDValue();
 }
 
 SDValue DAGCombiner::visitFMUL(SDNode *N) {
if ((Opcode == ISD::VECREDUCE_OR &&
         (Opcode == ISD::VECREDUCE_AND &&
          (N0.getOperand(0).isUndef() || isAllOnesOrAllOnesSplat(Vec))))
       return DAG.getNode(Opcode, SDLoc(N), N->getValueType(0), Subvec);
   }
 
   return SDValue();
 }
 
+SDValue DAGCombiner::visitVP_FSUB(SDNode *N) {
+  SelectionDAG::FlagInserter FlagsInserter(DAG, N);
+
+  // FSUB -> FMA combines:
+  if (SDValue Fused = visitFSUBForFMACombine<VPMatchContext>(N)) {
+    AddToWorklist(Fused.getNode());
+    return Fused;
+  }
+  return SDValue();
+}
+
 SDValue DAGCombiner::visitVPOp(SDNode *N) {
 
   if (N->getOpcode() == ISD::VP_GATHER)
     if (SDValue SD = visitVPGATHER(N))
       return SD;
 
   if (N->getOpcode() == ISD::VP_SCATTER)
     if (SDValue SD = visitVPSCATTER(N))
if (auto MaskIdx = ISD::getVPMaskIdx(N->getOpcode()))
     AreAllEltsDisabled |=
         ISD::isConstantSplatVectorAllZeros(N->getOperand(*MaskIdx).getNode());
 
   // This is the only generic VP combine we support for now.
   if (!AreAllEltsDisabled) {
     switch (N->getOpcode()) {
     case ISD::VP_FADD:
       return visitVP_FADD(N);
+    case ISD::VP_FSUB:
+      return visitVP_FSUB(N);
     }
     return SDValue();
   }
 
   // Binary operations can be replaced by UNDEF.
   if (ISD::isVPBinaryOp(N->getOpcode()))
     return DAG.getUNDEF(N->getValueType(0));
 

llvm/test/CodeGen/RISCV/rvv/fold-vp-fsub-and-vp-fmul.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -mattr=+v -target-abi=lp64d -verify-machineinstrs < %s | FileCheck %s
+
+declare <vscale x 1 x double> @llvm.vp.fmul.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x i1> %m, i32 %vl)
+declare <vscale x 1 x double> @llvm.vp.fsub.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x i1> %m, i32 %vl)
+declare <vscale x 1 x double> @llvm.vp.fneg.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x i1> %m, i32 %vl)
+
+; (fsub (fmul x, y), z)) -> (fma x, y, (fneg z))
+define <vscale x 1 x double> @test1(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x double> %z, <vscale x 1 x i1> %m, i32 zeroext %vl) {
+; CHECK-LABEL: test1:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmsub.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %1 = call fast <vscale x 1 x double> @llvm.vp.fmul.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x i1> %m, i32 %vl)
+  %2 = call fast <vscale x 1 x double> @llvm.vp.fsub.nxv1f64(<vscale x 1 x double> %1, <vscale x 1 x double> %z, <vscale x 1 x i1> %m, i32 %vl)
+  ret <vscale x 1 x double> %2
+}
+
+; (fsub z, (fmul x, y))) -> (fma (fneg y), x, z)
+define <vscale x 1 x double> @test2(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x double> %z, <vscale x 1 x i1> %m, i32 zeroext %vl) {
+; CHECK-LABEL: test2:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfnmsub.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %1 = call fast <vscale x 1 x double> @llvm.vp.fmul.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x i1> %m, i32 %vl)
+  %2 = call fast <vscale x 1 x double> @llvm.vp.fsub.nxv1f64(<vscale x 1 x double> %z, <vscale x 1 x double> %1, <vscale x 1 x i1> %m, i32 %vl)
+  ret <vscale x 1 x double> %2
+}
+
+; (fsub (fneg (fmul x, y))), z) -> (fma (fneg x), y, (fneg z))
+define <vscale x 1 x double> @test3(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x double> %z, <vscale x 1 x i1> %m, i32 zeroext %vl) {
+; CHECK-LABEL: test3:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmsub.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %1 = call fast <vscale x 1 x double> @llvm.vp.fmul.nxv1f64(<vscale x 1 x double> %x, <vscale x 1 x double> %y, <vscale x 1 x i1> %m, i32 %vl)
+  %2 = call fast <vscale x 1 x double> @llvm.vp.fneg.nxv1f64(<vscale x 1 x double> %1, <vscale x 1 x i1> %m, i32 %vl)
+  %3 = call fast <vscale x 1 x double> @llvm.vp.fsub.nxv1f64(<vscale x 1 x double> %1, <vscale x 1 x double> %z, <vscale x 1 x i1> %m, i32 %vl)
+  ret <vscale x 1 x double> %3
+}