[CIR] Upstream l-value emission for ExprWithCleanups #167938

andykaylor · 2025-11-13T19:51:53Z

This adds the necessary handler for emitting an l-value for an ExprWithCleanups expression.

llvmbot · 2025-11-13T19:52:26Z

@llvm/pr-subscribers-clangir

@llvm/pr-subscribers-clang

Author: Andy Kaylor (andykaylor)

Changes

This adds the necessary handler for emitting an l-value for an ExprWithCleanups expression.

Full diff: https://github.com/llvm/llvm-project/pull/167938.diff

2 Files Affected:

(modified) clang/lib/CIR/CodeGen/CIRGenFunction.cpp (+29)
(added) clang/test/CIR/CodeGen/temporary-materialization.cpp (+98)

diff --git a/clang/lib/CIR/CodeGen/CIRGenFunction.cpp b/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
index 5d5209b9ffb60..4a4e9e5d054ed 100644
--- a/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenFunction.cpp
@@ -912,6 +912,35 @@ LValue CIRGenFunction::emitLValue(const Expr *e) {
   case Expr::CXXOperatorCallExprClass:
   case Expr::UserDefinedLiteralClass:
     return emitCallExprLValue(cast<CallExpr>(e));
+  case Expr::ExprWithCleanupsClass: {
+    const auto *cleanups = cast<ExprWithCleanups>(e);
+    LValue lv;
+
+    mlir::Location scopeLoc = getLoc(e->getSourceRange());
+    [[maybe_unused]] auto scope = cir::ScopeOp::create(
+        builder, scopeLoc, /*scopeBuilder=*/
+        [&](mlir::OpBuilder &b, mlir::Location loc) {
+          CIRGenFunction::LexicalScope lexScope{*this, loc,
+                                                builder.getInsertionBlock()};
+
+          lv = emitLValue(cleanups->getSubExpr());
+          if (lv.isSimple()) {
+            // Defend against branches out of gnu statement expressions
+            // surrounded by cleanups.
+            Address addr = lv.getAddress();
+            mlir::Value v = addr.getPointer();
+            assert(!cir::MissingFeatures::addressIsKnownNonNull());
+            assert(!cir::MissingFeatures::opTBAA());
+            assert(!cir::MissingFeatures::objCGC());
+            lv = LValue::makeAddr(addr.withPointer(v), lv.getType(),
+                                  lv.getBaseInfo());
+          }
+        });
+
+    // FIXME: Is it possible to create an ExprWithCleanups that produces a
+    // bitfield lvalue or some other non-simple lvalue?
+    return lv;
+  }
   case Expr::ParenExprClass:
     return emitLValue(cast<ParenExpr>(e)->getSubExpr());
   case Expr::GenericSelectionExprClass:
diff --git a/clang/test/CIR/CodeGen/temporary-materialization.cpp b/clang/test/CIR/CodeGen/temporary-materialization.cpp
new file mode 100644
index 0000000000000..e7088f8bb4d9a
--- /dev/null
+++ b/clang/test/CIR/CodeGen/temporary-materialization.cpp
@@ -0,0 +1,98 @@
+// RUN: %clang_cc1 -std=c++17 -triple x86_64-unknown-linux-gnu -fclangir -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s --check-prefix=CIR
+// RUN: %clang_cc1 -std=c++17 -triple x86_64-unknown-linux-gnu -fclangir -emit-llvm %s -o %t-cir.ll
+// RUN: FileCheck --input-file=%t-cir.ll %s --check-prefix=LLVM
+// RUN: %clang_cc1 -std=c++17 -triple x86_64-unknown-linux-gnu -emit-llvm %s -o %t.ll
+// RUN: FileCheck --input-file=%t.ll %s --check-prefix=OGCG
+
+int make_int();
+
+int test() {
+  const int &x = make_int();
+  return x;
+}
+
+//      CIR: cir.func {{.*}} @_Z4testv()
+//      CIR:   %[[TEMP_SLOT:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["ref.tmp0", init]
+// CIR-NEXT:   %[[X:.*]] = cir.alloca !cir.ptr<!s32i>, !cir.ptr<!cir.ptr<!s32i>>, ["x", init, const]
+// CIR-NEXT:   cir.scope {
+// CIR-NEXT:     %[[TEMP_VALUE:.*]] = cir.call @_Z8make_intv() : () -> !s32i
+// CIR-NEXT:     cir.store{{.*}} %[[TEMP_VALUE]], %[[TEMP_SLOT]]
+// CIR-NEXT:   }
+// CIR-NEXT:   cir.store{{.*}} %[[TEMP_SLOT]], %[[X]]
+
+// LLVM: define {{.*}} i32 @_Z4testv()
+// LLVM:   %[[RETVAL:.*]] = alloca i32
+// LLVM:   %[[TEMP_SLOT:.*]] = alloca i32
+// LLVM:   %[[X:.*]] = alloca ptr
+// LLVM:   br label %[[SCOPE_LABEL:.*]]
+// LLVM: [[SCOPE_LABEL]]:
+// LLVM:   %[[TEMP_VALUE:.*]] = call i32 @_Z8make_intv()
+// LLVM:   store i32 %[[TEMP_VALUE]], ptr %[[TEMP_SLOT]]
+// LLVM:   br label %[[SCOPE_END_LABEL:.*]]
+// LLVM: [[SCOPE_END_LABEL]]:
+// LLVM:   store ptr %[[TEMP_SLOT]], ptr %[[X]]
+
+// OGCG: define {{.*}} i32 @_Z4testv()
+// OGCG:   %[[X:.*]] = alloca ptr
+// OGCG:   %[[TEMP_SLOT:.*]] = alloca i32
+// OGCG:   %[[TEMP_VALUE:.*]] = call noundef i32 @_Z8make_intv()
+// OGCG:   store i32 %[[TEMP_VALUE]], ptr %[[TEMP_SLOT]]
+// OGCG:   store ptr %[[TEMP_SLOT]], ptr %[[X]]
+
+int test_scoped() {
+  int x = make_int();
+  {
+    const int &y = make_int();
+    x = y;
+  }
+  return x;
+}
+
+//      CIR: cir.func {{.*}} @_Z11test_scopedv()
+//      CIR:   %[[X:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["x", init]
+//      CIR:   cir.scope {
+// CIR-NEXT:     %[[TEMP_SLOT:.*]] = cir.alloca !s32i, !cir.ptr<!s32i>, ["ref.tmp0", init]
+// CIR-NEXT:     %[[Y_ADDR:.*]] = cir.alloca !cir.ptr<!s32i>, !cir.ptr<!cir.ptr<!s32i>>, ["y", init, const]
+// CIR-NEXT:     cir.scope {
+// CIR-NEXT:       %[[TEMP_VALUE:.*]] = cir.call @_Z8make_intv() : () -> !s32i
+// CIR-NEXT:       cir.store{{.*}} %[[TEMP_VALUE]], %[[TEMP_SLOT]] : !s32i, !cir.ptr<!s32i>
+// CIR-NEXT:     }
+// CIR-NEXT:     cir.store{{.*}} %[[TEMP_SLOT]], %[[Y_ADDR]] : !cir.ptr<!s32i>, !cir.ptr<!cir.ptr<!s32i>>
+// CIR-NEXT:     %[[Y_REF:.*]] = cir.load %[[Y_ADDR]] : !cir.ptr<!cir.ptr<!s32i>>, !cir.ptr<!s32i>
+// CIR-NEXT:     %[[Y_VALUE:.*]] = cir.load{{.*}} %[[Y_REF]] : !cir.ptr<!s32i>, !s32i
+// CIR-NEXT:     cir.store{{.*}} %[[Y_VALUE]], %[[X]] : !s32i, !cir.ptr<!s32i>
+// CIR-NEXT:   }
+
+// LLVM: define {{.*}} i32 @_Z11test_scopedv()
+// LLVM:   %[[TEMP_SLOT:.*]] = alloca i32
+// LLVM:   %[[Y_ADDR:.*]] = alloca ptr
+// LLVM:   %[[RETVAL:.*]] = alloca i32
+// LLVM:   %[[X:.*]] = alloca i32
+// LLVM:   %[[TEMP_VALUE1:.*]] = call i32 @_Z8make_intv()
+// LLVM:   store i32 %[[TEMP_VALUE1]], ptr %[[X]]
+// LLVM:   br label %[[SCOPE_LABEL:.*]]
+// LLVM: [[SCOPE_LABEL]]:
+// LLVM:   br label %[[INNER_SCOPE_LABEL:.*]]
+// LLVM: [[INNER_SCOPE_LABEL]]:
+// LLVM:   %[[TEMP_VALUE2:.*]] = call i32 @_Z8make_intv()
+// LLVM:   store i32 %[[TEMP_VALUE2]], ptr %[[TEMP_SLOT]]
+// LLVM:   br label %[[INNER_SCOPE_END_LABEL:.*]]
+// LLVM: [[INNER_SCOPE_END_LABEL]]:
+// LLVM:   store ptr %[[TEMP_SLOT]], ptr %[[Y_ADDR]]
+// LLVM:   %[[Y_REF:.*]] = load ptr, ptr %[[Y_ADDR]]
+// LLVM:   %[[Y_VALUE:.*]] = load i32, ptr %[[Y_REF]]
+// LLVM:   store i32 %[[Y_VALUE]], ptr %[[X]]
+
+// OGCG: define {{.*}} i32 @_Z11test_scopedv()
+// OGCG:   %[[X:.*]] = alloca i32
+// OGCG:   %[[Y_ADDR:.*]] = alloca ptr
+// OGCG:   %[[TEMP_SLOT:.*]] = alloca i32
+// OGCG:   %[[TEMP_VALUE1:.*]] = call noundef i32 @_Z8make_intv()
+// OGCG:   store i32 %[[TEMP_VALUE1]], ptr %[[X]]
+// OGCG:   %[[TEMP_VALUE2:.*]] = call noundef i32 @_Z8make_intv()
+// OGCG:   store i32 %[[TEMP_VALUE2]], ptr %[[TEMP_SLOT]]
+// OGCG:   store ptr %[[TEMP_SLOT]], ptr %[[Y_ADDR]]
+// OGCG:   %[[Y_REF:.*]] = load ptr, ptr %[[Y_ADDR]]
+// OGCG:   %[[Y_VALUE:.*]] = load i32, ptr %[[Y_REF]]
+// OGCG:   store i32 %[[Y_VALUE]], ptr %[[X]]

erichkeane · 2025-11-13T19:59:01Z

clang/lib/CIR/CodeGen/CIRGenFunction.cpp

+
+          lv = emitLValue(cleanups->getSubExpr());
+          if (lv.isSimple()) {
+            // Defend against branches out of gnu statement expressions


I dont see any statement expression tests?

Oof! This is ugly. I was struggling to find a test case that would get us here in a situation that actually needed to do a cleanup. I just dug through the revision history to find the test case that was added when the corresponding code was added in classic codegen. It turns out the incubator (and hence this PR) doesn't handle that case correctly.

https://godbolt.org/z/36nEss113

This code should be forcing a cleanup with a function variant that accepts a list of values the needed to be re-loaded in the cleanup handler. That variant exists in the incubator but it doesn't handle the reloading of the variables. I'm not sure what it will take to implement it, but we're definitely not ready for that yet. It requires a level of cleanup branch handling that we aren't prepared for.

I think I need to just add a missing feature here.

I'm just glad you figured this out before I'd have to figure out how the heck to write a statement expression again :D

erichkeane · 2025-11-13T19:59:17Z

clang/lib/CIR/CodeGen/CIRGenFunction.cpp

+            assert(!cir::MissingFeatures::objCGC());
+            lv = LValue::makeAddr(addr.withPointer(v), lv.getType(),
+                                  lv.getBaseInfo());
+          }


Can we explain the 'else' here? It is odd to me that 'simple' is the only thing that would require work.

That's what the FIXME below is about (carried over from classic codegen). I can't think of a way to hit the else case. Nothing in the current clang lit tests goes to the 'else' case.

Hrmph, ok then. CodeGen constantly leaves me in a state of disgruntled wonder.

This adds the necessary handler for emitting an l-value for an ExprWithCleanups expression.

andykaylor requested review from AmrDeveloper, bcardosolopes, erichkeane, lanza, mmha and xlauko November 13, 2025 19:51

llvmbot added clang Clang issues not falling into any other category ClangIR Anything related to the ClangIR project labels Nov 13, 2025

erichkeane reviewed Nov 13, 2025

View reviewed changes

erichkeane approved these changes Nov 13, 2025

View reviewed changes

andykaylor added 2 commits November 13, 2025 15:29

[CIR] Upstream lvalue emission for ExprWithCleanups

99507ad

This adds the necessary handler for emitting an l-value for an ExprWithCleanups expression.

Replace non-working scope handling with a missing feature

fe28db8

andykaylor force-pushed the cir-exprwithcleanups-lvalue branch from ef5d9ba to fe28db8 Compare November 13, 2025 23:29

andykaylor merged commit cfad41c into llvm:main Nov 14, 2025
8 of 9 checks passed

andykaylor deleted the cir-exprwithcleanups-lvalue branch November 14, 2025 01:00

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[CIR] Upstream l-value emission for ExprWithCleanups #167938

[CIR] Upstream l-value emission for ExprWithCleanups #167938

Uh oh!

andykaylor commented Nov 13, 2025

Uh oh!

llvmbot commented Nov 13, 2025 •

edited

Loading

Uh oh!

erichkeane Nov 13, 2025

Uh oh!

andykaylor Nov 13, 2025

Uh oh!

erichkeane Nov 13, 2025

Uh oh!

erichkeane Nov 13, 2025

Uh oh!

andykaylor Nov 13, 2025

Uh oh!

erichkeane Nov 13, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

[CIR] Upstream l-value emission for ExprWithCleanups #167938

[CIR] Upstream l-value emission for ExprWithCleanups #167938

Uh oh!

Conversation

andykaylor commented Nov 13, 2025

Uh oh!

llvmbot commented Nov 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

erichkeane Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

andykaylor Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

erichkeane Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

erichkeane Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

andykaylor Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

erichkeane Nov 13, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

llvmbot commented Nov 13, 2025 •

edited

Loading