[clang] [clang][bytecode] Use in Expr::tryEvaluateStrLen() (PR #149677)

[Timm Baeder via cfe-commits]

Timm =?utf-8?q?Bäder?= <tbaeder at redhat.com>
Message-ID: <llvm.org/llvm/llvm-project/pull/149677 at github.com>
In-Reply-To:


https://github.com/tbaederr created https://github.com/llvm/llvm-project/pull/149677

None

>From 9db49be1f605ec039b984bd7df3e29c127df1866 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Sat, 19 Jul 2025 23:02:13 +0200
Subject: [PATCH 1/2] [clang][bytecode] Use in Expr::tryEvaluateStrLen()

---
 clang/lib/AST/ByteCode/Context.cpp    | 32 +++++++++++++++++++++++++++
 clang/lib/AST/ByteCode/Context.h      |  4 ++++
 clang/lib/AST/ExprConstant.cpp        |  4 ++++
 clang/test/Sema/warn-fortify-source.c |  5 +++++
 4 files changed, 45 insertions(+)

diff --git a/clang/lib/AST/ByteCode/Context.cpp b/clang/lib/AST/ByteCode/Context.cpp
index ead6e4af5d403..ebb809a33c1fc 100644
--- a/clang/lib/AST/ByteCode/Context.cpp
+++ b/clang/lib/AST/ByteCode/Context.cpp
@@ -235,6 +235,38 @@ bool Context::evaluateCharRange(State &Parent, const Expr *SizeExpr,
   return evaluateStringRepr(Parent, SizeExpr, PtrExpr, Result);
 }
 
+bool Context::evaluateStrlen(State &Parent, const Expr *E, uint64_t &Result) {
+  assert(Stk.empty());
+  Compiler<EvalEmitter> C(*this, *P, Parent, Stk);
+
+  auto PtrRes = C.interpretAsPointer(E, [&](const Pointer &Ptr) {
+    const Descriptor *FieldDesc = Ptr.getFieldDesc();
+    if (!FieldDesc->isPrimitiveArray())
+      return false;
+
+    unsigned N = Ptr.getNumElems();
+    PrimType ElemT = FieldDesc->getPrimType();
+    Result = 0;
+    for (unsigned I = Ptr.getIndex(); I != N; ++I) {
+      INT_TYPE_SWITCH(ElemT, {
+        auto Elem = Ptr.atIndex(I).deref<T>();
+        if (Elem.isZero())
+          return true;
+        ++Result;
+      });
+    }
+    // We didn't find a 0 byte.
+    return false;
+  });
+
+  if (PtrRes.isInvalid()) {
+    C.cleanup();
+    Stk.clear();
+    return false;
+  }
+  return true;
+}
+
 const LangOptions &Context::getLangOpts() const { return Ctx.getLangOpts(); }
 
 static PrimType integralTypeToPrimTypeS(unsigned BitWidth) {
diff --git a/clang/lib/AST/ByteCode/Context.h b/clang/lib/AST/ByteCode/Context.h
index acf750421f8af..6065d3e53f230 100644
--- a/clang/lib/AST/ByteCode/Context.h
+++ b/clang/lib/AST/ByteCode/Context.h
@@ -66,6 +66,10 @@ class Context final {
   bool evaluateCharRange(State &Parent, const Expr *SizeExpr,
                          const Expr *PtrExpr, std::string &Result);
 
+  /// Evalute \param E and if it can be evaluated to a string literal,
+  /// run strlen() on it.
+  bool evaluateStrlen(State &Parent, const Expr *E, uint64_t &Result);
+
   /// Returns the AST context.
   ASTContext &getASTContext() const { return Ctx; }
   /// Returns the language options.
diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp
index e57540511a9de..0d12161756467 100644
--- a/clang/lib/AST/ExprConstant.cpp
+++ b/clang/lib/AST/ExprConstant.cpp
@@ -18184,6 +18184,10 @@ bool Expr::EvaluateCharRangeAsString(APValue &Result,
 bool Expr::tryEvaluateStrLen(uint64_t &Result, ASTContext &Ctx) const {
   Expr::EvalStatus Status;
   EvalInfo Info(Ctx, Status, EvalInfo::EM_ConstantFold);
+
+  if (Info.EnableNewConstInterp)
+    return Info.Ctx.getInterpContext().evaluateStrlen(Info, this, Result);
+
   return EvaluateBuiltinStrLen(this, Result, Info);
 }
 
diff --git a/clang/test/Sema/warn-fortify-source.c b/clang/test/Sema/warn-fortify-source.c
index f48ea0907c657..216878c0836d8 100644
--- a/clang/test/Sema/warn-fortify-source.c
+++ b/clang/test/Sema/warn-fortify-source.c
@@ -3,6 +3,11 @@
 // RUN: %clang_cc1 -xc++ -triple x86_64-apple-macosx10.14.0 %s -verify
 // RUN: %clang_cc1 -xc++ -triple x86_64-apple-macosx10.14.0 %s -verify -DUSE_BUILTINS
 
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.14.0 %s -verify -fexperimental-new-constant-interpreter
+// RUN: %clang_cc1 -triple x86_64-apple-macosx10.14.0 %s -verify -DUSE_BUILTINS -fexperimental-new-constant-interpreter
+// RUN: %clang_cc1 -xc++ -triple x86_64-apple-macosx10.14.0 %s -verify -fexperimental-new-constant-interpreter
+// RUN: %clang_cc1 -xc++ -triple x86_64-apple-macosx10.14.0 %s -verify -DUSE_BUILTINS -fexperimental-new-constant-interpreter
+
 typedef unsigned long size_t;
 
 #ifdef __cplusplus

>From d80e62c13054c133ab12fe220a9febe3c62c284b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timm=20B=C3=A4der?= <tbaeder at redhat.com>
Date: Sat, 19 Jul 2025 23:09:05 +0200
Subject: [PATCH 2/2] Fast-path for single-byte chars

---
 clang/lib/AST/ByteCode/Context.cpp | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/clang/lib/AST/ByteCode/Context.cpp b/clang/lib/AST/ByteCode/Context.cpp
index ebb809a33c1fc..9e683d71f83bc 100644
--- a/clang/lib/AST/ByteCode/Context.cpp
+++ b/clang/lib/AST/ByteCode/Context.cpp
@@ -245,6 +245,11 @@ bool Context::evaluateStrlen(State &Parent, const Expr *E, uint64_t &Result) {
       return false;
 
     unsigned N = Ptr.getNumElems();
+    if (Ptr.elemSize() == 1) {
+      Result = strnlen(reinterpret_cast<const char *>(Ptr.getRawAddress()), N);
+      return Result != N;
+    }
+
     PrimType ElemT = FieldDesc->getPrimType();
     Result = 0;
     for (unsigned I = Ptr.getIndex(); I != N; ++I) {