[clang] ffc6877 - [CIR] Implement CXXRewrittenBinaryOperator for AggregateExpr (#204667)

via cfe-commits cfe-commits at lists.llvm.org
Sat Jun 27 05:29:12 PDT 2026


Author: Amr Hesham
Date: 2026-06-27T12:29:08Z
New Revision: ffc6877d0294184fc6ccc44f698638a6d3636052

URL: https://github.com/llvm/llvm-project/commit/ffc6877d0294184fc6ccc44f698638a6d3636052
DIFF: https://github.com/llvm/llvm-project/commit/ffc6877d0294184fc6ccc44f698638a6d3636052.diff

LOG: [CIR] Implement CXXRewrittenBinaryOperator for AggregateExpr (#204667)

Implement CXXRewrittenBinaryOperator support for AggregateExpr

Added: 
    

Modified: 
    clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
    clang/test/CIR/CodeGen/cxx-rewritten-binary-operator.cpp

Removed: 
    


################################################################################
diff  --git a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
index c53453bd3e88b..4d54519e4d25f 100644
--- a/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenExprAggregate.cpp
@@ -443,8 +443,7 @@ class AggExprEmitter : public StmtVisitor<AggExprEmitter> {
   }
 
   void VisitCXXRewrittenBinaryOperator(CXXRewrittenBinaryOperator *e) {
-    cgf.cgm.errorNYI(e->getSourceRange(),
-                     "AggExprEmitter: VisitCXXRewrittenBinaryOperator");
+    Visit(e->getSemanticForm());
   }
   void VisitObjCMessageExpr(ObjCMessageExpr *e) {
     cgf.cgm.errorNYI(e->getSourceRange(),

diff  --git a/clang/test/CIR/CodeGen/cxx-rewritten-binary-operator.cpp b/clang/test/CIR/CodeGen/cxx-rewritten-binary-operator.cpp
index b841eba7ba4eb..69a8059b09484 100644
--- a/clang/test/CIR/CodeGen/cxx-rewritten-binary-operator.cpp
+++ b/clang/test/CIR/CodeGen/cxx-rewritten-binary-operator.cpp
@@ -89,3 +89,51 @@ void cxx_rewritten_binary_operator_complex_expr() {
 // OGCG: %[[R_IMAG_PTR:.*]] = getelementptr inbounds nuw { i32, i32 }, ptr %[[R_ADDR]], i32 0, i32 1
 // OGCG: store i32 %[[RESULT_REAL]], ptr %[[R_REAL_PTR]], align 4
 // OGCG: store i32 %[[RESULT_IMAG]], ptr %[[R_IMAG_PTR]], align 4
+
+struct Result {
+  int value;
+};
+
+struct SpaceshipResult {
+  Result operator<(int) const { return {}; }
+};
+
+struct Item {
+  SpaceshipResult operator<=>(const Item &) const { return {}; }
+};
+
+void cxx_rewritten_binary_operator_aggr_expr() {
+  Item a;
+  Item b;
+  Result r = a < b;
+}
+
+// CIR: %[[A_ADDR:.*]] = cir.alloca "a" {{.*}} : !cir.ptr<!rec_Item>
+// CIR: %[[B_ADDR:.*]] = cir.alloca "b" {{.*}} : !cir.ptr<!rec_Item>
+// CIR: %[[R_ADDR:.*]] = cir.alloca "r" {{.*}} init : !cir.ptr<!rec_Result>
+// CIR: %[[TMP_ADDR:.*]] = cir.alloca "ref.tmp0" {{.*}} : !cir.ptr<!rec_SpaceshipResult>
+// CIR: %[[OP_RESULT:.*]] = cir.call @_ZNK4ItemssERKS_(%[[A_ADDR]], %[[B_ADDR]]) : (!cir.ptr<!rec_Item> {llvm.align = 1 : i64, llvm.dereferenceable = 1 : i64, llvm.nonnull, llvm.noundef}, !cir.ptr<!rec_Item> {llvm.align = 1 : i64, llvm.dereferenceable = 1 : i64, llvm.nonnull, llvm.noundef}) -> !rec_SpaceshipResult
+// CIR: cir.store {{.*}} %[[OP_RESULT]], %[[TMP_ADDR]] : !rec_SpaceshipResult, !cir.ptr<!rec_SpaceshipResult>
+// CIR: %[[CONST_0:.*]] = cir.const #cir.int<0> : !s32i
+// CIR: %[[RESULT:.*]] = cir.call @_ZNK15SpaceshipResultltEi(%[[TMP_ADDR]], %[[CONST_0]]) : (!cir.ptr<!rec_SpaceshipResult> {llvm.align = 1 : i64, llvm.dereferenceable = 1 : i64, llvm.nonnull, llvm.noundef}, !s32i {llvm.noundef}) -> !rec_Result
+// CIR: cir.store {{.*}} %[[RESULT]], %[[R_ADDR]] : !rec_Result, !cir.ptr<!rec_Result>
+
+// The 
diff erence between LLVM and OGCG is due to missing ABI lowering.
+
+// LLVM: %[[A_ADDR:.*]] = alloca %struct.Item, i64 1, align 1
+// LLVM: %[[B_ADDR:.*]] = alloca %struct.Item, i64 1, align 1
+// LLVM: %[[R_ADDR:.*]] = alloca %struct.Result, i64 1, align 4
+// LLVM: %[[TMP_ADDR:.*]] = alloca %struct.SpaceshipResult, i64 1, align 1
+// LLVM: %[[OP_RESULT:.*]] = call %struct.SpaceshipResult @_ZNK4ItemssERKS_(ptr noundef nonnull align 1 dereferenceable(1) %[[A_ADDR]], ptr noundef nonnull align 1 dereferenceable(1) %[[B_ADDR]])
+// LLVM: store %struct.SpaceshipResult %[[OP_RESULT]], ptr %[[TMP_ADDR]], align 1
+// LLVM: %[[RESULT:.*]] = call %struct.Result @_ZNK15SpaceshipResultltEi(ptr noundef nonnull align 1 dereferenceable(1) %[[TMP_ADDR]], i32 noundef 0)
+// LLVM: store %struct.Result %[[RESULT]], ptr %[[R_ADDR]], align 4
+
+// OGCG: %[[A_ADDR:.*]] = alloca %struct.Item, align 1
+// OGCG: %[[B_ADDR:.*]] = alloca %struct.Item, align 1
+// OGCG: %[[R_ADDR:.*]] = alloca %struct.Result, align 4
+// OGCG: %[[TMP_ADDR:.*]] = alloca %struct.SpaceshipResult, align 1
+// OGCG: call void @_ZNK4ItemssERKS_(ptr noundef nonnull align 1 dereferenceable(1) %[[A_ADDR]], ptr noundef nonnull align 1 dereferenceable(1) %[[B_ADDR]])
+// OGCG: %[[RESULT:.*]] = call i32 @_ZNK15SpaceshipResultltEi(ptr noundef nonnull align 1 dereferenceable(1) %[[TMP_ADDR]], i32 noundef 0)
+// OGCG: %[[R_ADDR_PTR:.*]] = getelementptr inbounds nuw %struct.Result, ptr %[[R_ADDR:.*]], i32 0, i32 0
+// OGCG: store i32 %[[RESULT:.*]], ptr %[[R_ADDR_PTR]], align 4


        


More information about the cfe-commits mailing list