[llvm] 7b2ac8f - [Matrix] Pass ShapeInfo to Visit* methods (NFC). (#142487)

[via llvm-commits]

Author: Jon Roelofs
Date: 2025-06-05T11:22:17-07:00
New Revision: 7b2ac8ff54fbc194fd639be3f4073733c1e3d05c

URL: https://github.com/llvm/llvm-project/commit/7b2ac8ff54fbc194fd639be3f4073733c1e3d05c
DIFF: https://github.com/llvm/llvm-project/commit/7b2ac8ff54fbc194fd639be3f4073733c1e3d05c.diff

LOG: [Matrix] Pass ShapeInfo to Visit* methods (NFC). (#142487)

They all require it now.

Added: 
    

Modified: 
    llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
index 38f92561a917d..20279bf69dd59 100644
--- a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
+++ b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
@@ -1056,19 +1056,20 @@ class LowerMatrixIntrinsics {
 
       IRBuilder<> Builder(Inst);
 
-      if (CallInst *CInst = dyn_cast<CallInst>(Inst))
-        Changed |= VisitCallInst(CInst);
+      const ShapeInfo &SI = ShapeMap.at(Inst);
 
       Value *Op1;
       Value *Op2;
       if (auto *BinOp = dyn_cast<BinaryOperator>(Inst))
-        VisitBinaryOperator(BinOp);
+        VisitBinaryOperator(BinOp, SI);
       else if (auto *UnOp = dyn_cast<UnaryOperator>(Inst))
-        VisitUnaryOperator(UnOp);
+        VisitUnaryOperator(UnOp, SI);
+      else if (CallInst *CInst = dyn_cast<CallInst>(Inst))
+        VisitCallInst(CInst);
       else if (match(Inst, m_Load(m_Value(Op1))))
-        VisitLoad(cast<LoadInst>(Inst), Op1, Builder);
+        VisitLoad(cast<LoadInst>(Inst), SI, Op1, Builder);
       else if (match(Inst, m_Store(m_Value(Op1), m_Value(Op2))))
-        VisitStore(cast<StoreInst>(Inst), Op1, Op2, Builder);
+        VisitStore(cast<StoreInst>(Inst), SI, Op1, Op2, Builder);
       else
         continue;
       Changed = true;
@@ -1109,10 +1110,10 @@ class LowerMatrixIntrinsics {
     return Changed;
   }
 
-  /// Replace intrinsic calls
-  bool VisitCallInst(CallInst *Inst) {
-    if (!Inst->getCalledFunction() || !Inst->getCalledFunction()->isIntrinsic())
-      return false;
+  /// Replace intrinsic calls.
+  void VisitCallInst(CallInst *Inst) {
+    assert(Inst->getCalledFunction() &&
+           Inst->getCalledFunction()->isIntrinsic());
 
     switch (Inst->getCalledFunction()->getIntrinsicID()) {
     case Intrinsic::matrix_multiply:
@@ -1128,9 +1129,9 @@ class LowerMatrixIntrinsics {
       LowerColumnMajorStore(Inst);
       break;
     default:
-      return false;
+      llvm_unreachable(
+          "only intrinsics supporting shape info should be seen here");
     }
-    return true;
   }
 
   /// Compute the alignment for a column/row \p Idx with \p Stride between them.
@@ -2107,48 +2108,36 @@ class LowerMatrixIntrinsics {
         Builder);
   }
 
-  /// Lower load instructions, if shape information is available.
-  void VisitLoad(LoadInst *Inst, Value *Ptr, IRBuilder<> &Builder) {
-    auto I = ShapeMap.find(Inst);
-    assert(I != ShapeMap.end() &&
-           "must only visit instructions with shape info");
-    LowerLoad(Inst, Ptr, Inst->getAlign(),
-              Builder.getInt64(I->second.getStride()), Inst->isVolatile(),
-              I->second);
+  /// Lower load instructions.
+  void VisitLoad(LoadInst *Inst, const ShapeInfo &SI, Value *Ptr,
+                 IRBuilder<> &Builder) {
+    LowerLoad(Inst, Ptr, Inst->getAlign(), Builder.getInt64(SI.getStride()),
+              Inst->isVolatile(), SI);
   }
 
-  void VisitStore(StoreInst *Inst, Value *StoredVal, Value *Ptr,
-                  IRBuilder<> &Builder) {
-    auto I = ShapeMap.find(Inst);
-    assert(I != ShapeMap.end() &&
-           "must only visit instructions with shape info");
+  void VisitStore(StoreInst *Inst, const ShapeInfo &SI, Value *StoredVal,
+                  Value *Ptr, IRBuilder<> &Builder) {
     LowerStore(Inst, StoredVal, Ptr, Inst->getAlign(),
-               Builder.getInt64(I->second.getStride()), Inst->isVolatile(),
-               I->second);
+               Builder.getInt64(SI.getStride()), Inst->isVolatile(), SI);
   }
 
-  /// Lower binary operators, if shape information is available.
-  void VisitBinaryOperator(BinaryOperator *Inst) {
-    auto I = ShapeMap.find(Inst);
-    assert(I != ShapeMap.end() &&
-           "must only visit instructions with shape info");
-
+  /// Lower binary operators.
+  void VisitBinaryOperator(BinaryOperator *Inst, const ShapeInfo &SI) {
     Value *Lhs = Inst->getOperand(0);
     Value *Rhs = Inst->getOperand(1);
 
     IRBuilder<> Builder(Inst);
-    ShapeInfo &Shape = I->second;
 
     MatrixTy Result;
-    MatrixTy A = getMatrix(Lhs, Shape, Builder);
-    MatrixTy B = getMatrix(Rhs, Shape, Builder);
+    MatrixTy A = getMatrix(Lhs, SI, Builder);
+    MatrixTy B = getMatrix(Rhs, SI, Builder);
     assert(A.isColumnMajor() == B.isColumnMajor() &&
            Result.isColumnMajor() == A.isColumnMajor() &&
            "operands must agree on matrix layout");
 
     Builder.setFastMathFlags(getFastMathFlags(Inst));
 
-    for (unsigned I = 0; I < Shape.getNumVectors(); ++I)
+    for (unsigned I = 0; I < SI.getNumVectors(); ++I)
       Result.addVector(Builder.CreateBinOp(Inst->getOpcode(), A.getVector(I),
                                            B.getVector(I)));
 
@@ -2158,19 +2147,14 @@ class LowerMatrixIntrinsics {
                      Builder);
   }
 
-  /// Lower unary operators, if shape information is available.
-  void VisitUnaryOperator(UnaryOperator *Inst) {
-    auto I = ShapeMap.find(Inst);
-    assert(I != ShapeMap.end() &&
-           "must only visit instructions with shape info");
-
+  /// Lower unary operators.
+  void VisitUnaryOperator(UnaryOperator *Inst, const ShapeInfo &SI) {
     Value *Op = Inst->getOperand(0);
 
     IRBuilder<> Builder(Inst);
-    ShapeInfo &Shape = I->second;
 
     MatrixTy Result;
-    MatrixTy M = getMatrix(Op, Shape, Builder);
+    MatrixTy M = getMatrix(Op, SI, Builder);
 
     Builder.setFastMathFlags(getFastMathFlags(Inst));
 
@@ -2184,7 +2168,7 @@ class LowerMatrixIntrinsics {
       }
     };
 
-    for (unsigned I = 0; I < Shape.getNumVectors(); ++I)
+    for (unsigned I = 0; I < SI.getNumVectors(); ++I)
       Result.addVector(BuildVectorOp(M.getVector(I)));
 
     finalizeLowering(Inst,