[CIR][CIRGen] emit cir.zero for constant string literals (llvm#373)

This PR addresses llvm#248 . Currently string literals are always lowered to a `cir.const_array` attribute even if the string literal only contains null bytes. This patch make the CodeGen emits `cir.zero` for these string literals.
shady-gang · Oct 1, 2024 · 9ca1936 · 9ca1936
1 parent acea016
commit 9ca1936
Show file tree

Hide file tree

Showing 3 changed files with 17 additions and 4 deletions.
diff --git a/clang/lib/CIR/CodeGen/CIRGenBuilder.h b/clang/lib/CIR/CodeGen/CIRGenBuilder.h
@@ -149,9 +149,17 @@ class CIRGenBuilderTy : public CIRBaseBuilderTy {
     return mlir::cir::ConstPtrAttr::get(getContext(), t, v);
   }
 
-  mlir::cir::ConstArrayAttr getString(llvm::StringRef str, mlir::Type eltTy,
-                                      unsigned size = 0) {
+  mlir::Attribute getString(llvm::StringRef str, mlir::Type eltTy,
+                            unsigned size = 0) {
     unsigned finalSize = size ? size : str.size();
+
+    // If the string is full of null bytes, emit a #cir.zero rather than
+    // a #cir.const_array.
+    if (str.count('\0') == str.size()) {
+      auto arrayTy = mlir::cir::ArrayType::get(getContext(), eltTy, finalSize);
+      return getZeroAttr(arrayTy);
+    }
+
     auto arrayTy = mlir::cir::ArrayType::get(getContext(), eltTy, finalSize);
     return getConstArray(mlir::StringAttr::get(str, arrayTy), arrayTy);
   }

diff --git a/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp b/clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp
@@ -1395,7 +1395,12 @@ mlir::cir::GlobalOp CIRGenItaniumRTTIBuilder::GetAddrOfTypeName(
   auto Align =
       CGM.getASTContext().getTypeAlignInChars(CGM.getASTContext().CharTy);
 
-  auto GV = CGM.createOrReplaceCXXRuntimeVariable(loc, Name, Init.getType(),
+  // builder.getString can return a #cir.zero if the string given to it only
+  // contains null bytes. However, type names cannot be full of null bytes.
+  // So cast Init to a ConstArrayAttr should be safe.
+  auto InitStr = cast<mlir::cir::ConstArrayAttr>(Init);
+
+  auto GV = CGM.createOrReplaceCXXRuntimeVariable(loc, Name, InitStr.getType(),
                                                   Linkage, Align);
   CIRGenModule::setInitializer(GV, Init);
   return GV;

diff --git a/clang/test/CIR/CodeGen/globals.c b/clang/test/CIR/CodeGen/globals.c
@@ -41,7 +41,7 @@ struct {
   char y[3];
   char z[3];
 } nestedString = {"1", "", "\0"};
-// CHECK: cir.global external @nestedString = #cir.const_struct<{#cir.const_array<"1\00\00" : !cir.array<!s8i x 3>> : !cir.array<!s8i x 3>, #cir.const_array<"\00\00\00" : !cir.array<!s8i x 3>> : !cir.array<!s8i x 3>, #cir.const_array<"\00\00\00" : !cir.array<!s8i x 3>> : !cir.array<!s8i x 3>}>
+// CHECK: cir.global external @nestedString = #cir.const_struct<{#cir.const_array<"1\00\00" : !cir.array<!s8i x 3>> : !cir.array<!s8i x 3>, #cir.zero : !cir.array<!s8i x 3>, #cir.zero : !cir.array<!s8i x 3>}>
 
 struct {
   char *name;