[clang] [clang] Be careful when choosing "fast path" for initialization with #embed (PR #99023)

[Mariya Podchishchaeva via cfe-commits]

https://github.com/Fznamznon created https://github.com/llvm/llvm-project/pull/99023

When #embed appears in an initializer list, we may choose a "fast path"
- if the target declaration is a char array, we simply initialize it with string literal that contains embedded data. However we need to be careful when checking that we actually can use this "fast path" since char array may be nested in a struct.

>From 4a48f6ab0d26d5c83b29c866a79bd8cdcc9102b9 Mon Sep 17 00:00:00 2001
From: "Podchishchaeva, Mariya" <mariya.podchishchaeva at intel.com>
Date: Tue, 16 Jul 2024 04:45:57 -0700
Subject: [PATCH] [clang] Be careful when choosing "fast path" for
 initialization with #embed

When #embed appears in an initializer list, we may choose a "fast path"
- if the target declaration is a char array, we simply initialize it
 with string literal that contains embedded data. However we need to be
 careful when checking that we actually can use this "fast path" since
 char array may be nested in a struct.
---
 clang/lib/Sema/SemaInit.cpp                 | 17 +++++++++++++----
 clang/test/Preprocessor/embed_codegen.cpp   |  8 ++++++++
 clang/test/Preprocessor/embed_constexpr.cpp |  8 ++++++++
 3 files changed, 29 insertions(+), 4 deletions(-)

diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
index a27ed02fc73b8..d97a5c8988840 100644
--- a/clang/lib/Sema/SemaInit.cpp
+++ b/clang/lib/Sema/SemaInit.cpp
@@ -1993,9 +1993,18 @@ static bool checkDestructorReference(QualType ElementType, SourceLocation Loc,
   return SemaRef.DiagnoseUseOfDecl(Destructor, Loc);
 }
 
-static bool canInitializeArrayWithEmbedDataString(ArrayRef<Expr *> ExprList,
-                                                  QualType InitType,
-                                                  ASTContext &Context) {
+static bool
+canInitializeArrayWithEmbedDataString(ArrayRef<Expr *> ExprList,
+                                      const InitializedEntity &Entity,
+                                      ASTContext &Context) {
+  QualType InitType = Entity.getType();
+  const InitializedEntity *Parent = &Entity;
+
+  while (Parent) {
+    InitType = Parent->getType();
+    Parent = Parent->getParent();
+  }
+
   // Only one initializer, it's an embed and the types match;
   EmbedExpr *EE =
       ExprList.size() == 1
@@ -2034,7 +2043,7 @@ void InitListChecker::CheckArrayType(const InitializedEntity &Entity,
     }
   }
 
-  if (canInitializeArrayWithEmbedDataString(IList->inits(), DeclType,
+  if (canInitializeArrayWithEmbedDataString(IList->inits(), Entity,
                                             SemaRef.Context)) {
     EmbedExpr *Embed = cast<EmbedExpr>(IList->inits()[0]);
     IList->setInit(0, Embed->getDataStringLiteral());
diff --git a/clang/test/Preprocessor/embed_codegen.cpp b/clang/test/Preprocessor/embed_codegen.cpp
index 64110afc162d7..c73a265ac7494 100644
--- a/clang/test/Preprocessor/embed_codegen.cpp
+++ b/clang/test/Preprocessor/embed_codegen.cpp
@@ -3,6 +3,7 @@
 // CHECK: @__const._Z3fooi.ca = private unnamed_addr constant [3 x i32] [i32 0, i32 106, i32 107], align 4
 // CHECK: @__const._Z3fooi.sc = private unnamed_addr constant %struct.S1 { i32 106, i32 107, i32 0 }, align 4
 // CHECK: @__const._Z3fooi.t = private unnamed_addr constant [3 x %struct.T] [%struct.T { [2 x i32] [i32 48, i32 49], %struct.S1 { i32 50, i32 51, i32 52 } }, %struct.T { [2 x i32] [i32 53, i32 54], %struct.S1 { i32 55, i32 56, i32 57 } }, %struct.T { [2 x i32] [i32 10, i32 0], %struct.S1 zeroinitializer }], align 16
+// CHECK: @__const._Z3fooi.W = private unnamed_addr constant %struct.Wrapper { i32 48, %struct.HasCharArray { [10 x i8] c"123456789\0A" } }, align 4
 void foo(int a) {
 // CHECK: %a.addr = alloca i32, align 4
 // CHECK: store i32 %a, ptr %a.addr, align 4
@@ -81,4 +82,11 @@ struct T tnonc[] = {
 #embed <jk.txt> prefix(,)
 };
 
+
+struct HasCharArray { unsigned char h[10]; };
+struct Wrapper { int a; struct HasCharArray d;  };
+constexpr struct Wrapper W = {
+#embed "numbers.txt"
+};
+
 }
diff --git a/clang/test/Preprocessor/embed_constexpr.cpp b/clang/test/Preprocessor/embed_constexpr.cpp
index 1cadff76b4890..9b6470fcc7182 100644
--- a/clang/test/Preprocessor/embed_constexpr.cpp
+++ b/clang/test/Preprocessor/embed_constexpr.cpp
@@ -95,3 +95,11 @@ struct ST {};
 ST<
 #embed <jk.txt> limit(1)
 > st;
+
+struct HasCharArray { unsigned char h[10]; };
+struct Wrapper { int a; struct HasCharArray d;  };
+constexpr struct Wrapper W = {
+#embed "numbers.txt"
+};
+
+static_assert(W.d.h[2] == '3');