From b42f8847437bc5e9a619db4dbb14137f588d661d Mon Sep 17 00:00:00 2001
From: Sirui Mu <msrlancern@gmail.com>
Date: Fri, 19 Jan 2024 08:23:11 +0800
Subject: [PATCH] [CIR][CIRGen] Support wide string literals (#399)

This commit supports the codegen of wide string literals, including
`wchar_t` string literals, `char16_t` string literals, and `char32_t`
string literals.

I'm not following the proposal in #374. The clang frontend doesn't
record the literal string. It only records the encoded code units for
wide string literals. So I believe that a dedicated string attribute
with an encoding tag as described in #374 may not be that helpful as I
thought.
---
 clang/lib/CIR/CodeGen/CIRGenModule.cpp | 33 ++++++++++++++++++++++++--
 clang/test/CIR/CodeGen/wide-string.cpp | 26 ++++++++++++++++++++
 2 files changed, 57 insertions(+), 2 deletions(-)
 create mode 100644 clang/test/CIR/CodeGen/wide-string.cpp

diff --git a/clang/lib/CIR/CodeGen/CIRGenModule.cpp b/clang/lib/CIR/CodeGen/CIRGenModule.cpp
index 7eab7323d70f..9b7f11d37ce2 100644
--- a/clang/lib/CIR/CodeGen/CIRGenModule.cpp
+++ b/clang/lib/CIR/CodeGen/CIRGenModule.cpp
@@ -1103,8 +1103,37 @@ CIRGenModule::getConstantArrayFromStringLiteral(const StringLiteral *E) {
     return builder.getString(Str, eltTy, finalSize);
   }
 
-  assert(0 && "not implemented");
-  return {};
+  auto arrayTy =
+      getTypes().ConvertType(E->getType()).dyn_cast<mlir::cir::ArrayType>();
+  assert(arrayTy && "string literals must be emitted as an array type");
+
+  auto arrayEltTy = arrayTy.getEltType().dyn_cast<mlir::cir::IntType>();
+  assert(arrayEltTy &&
+         "string literal elements must be emitted as integral type");
+
+  auto arraySize = arrayTy.getSize();
+  auto literalSize = E->getLength();
+
+  // Collect the code units.
+  SmallVector<uint32_t, 32> elementValues;
+  elementValues.reserve(arraySize);
+  for (unsigned i = 0; i < literalSize; ++i)
+    elementValues.push_back(E->getCodeUnit(i));
+  elementValues.resize(arraySize);
+
+  // If the string is full of null bytes, emit a #cir.zero instead.
+  if (std::all_of(elementValues.begin(), elementValues.end(),
+                  [](uint32_t x) { return x == 0; }))
+    return builder.getZeroAttr(arrayTy);
+
+  // Otherwise emit a constant array holding the characters.
+  SmallVector<mlir::Attribute, 32> elements;
+  elements.reserve(arraySize);
+  for (uint64_t i = 0; i < arraySize; ++i)
+    elements.push_back(mlir::cir::IntAttr::get(arrayEltTy, elementValues[i]));
+
+  auto elementsAttr = mlir::ArrayAttr::get(builder.getContext(), elements);
+  return builder.getConstArray(elementsAttr, arrayTy);
 }
 
 // TODO(cir): this could be a common AST helper for both CIR and LLVM codegen.
diff --git a/clang/test/CIR/CodeGen/wide-string.cpp b/clang/test/CIR/CodeGen/wide-string.cpp
new file mode 100644
index 000000000000..e7fc719647ab
--- /dev/null
+++ b/clang/test/CIR/CodeGen/wide-string.cpp
@@ -0,0 +1,26 @@
+// RUN: %clang_cc1 -std=c++17 -triple x86_64-unknown-linux-gnu -fclangir-enable -emit-cir %s -o %t.cir
+// RUN: FileCheck --input-file=%t.cir %s
+
+const char16_t *test_utf16() {
+  return u"你好世界";
+}
+
+// CHECK: cir.global "private" constant internal @{{.+}} = #cir.const_array<[#cir.int<20320> : !u16i, #cir.int<22909> : !u16i, #cir.int<19990> : !u16i, #cir.int<30028> : !u16i, #cir.int<0> : !u16i]> : !cir.array<!u16i x 5>
+
+const char32_t *test_utf32() {
+  return U"你好世界";
+}
+
+// CHECK: cir.global "private" constant internal @{{.+}} = #cir.const_array<[#cir.int<20320> : !u32i, #cir.int<22909> : !u32i, #cir.int<19990> : !u32i, #cir.int<30028> : !u32i, #cir.int<0> : !u32i]> : !cir.array<!u32i x 5>
+
+const char16_t *test_zero16() {
+  return u"\0\0\0\0";
+}
+
+// CHECK: cir.global "private" constant internal @{{.+}} = #cir.zero : !cir.array<!u16i x 5>
+
+const char32_t *test_zero32() {
+  return U"\0\0\0\0";
+}
+
+// CHECK: cir.global "private" constant internal @{{.+}} = #cir.zero : !cir.array<!u32i x 5>