Rollup merge of #122000 - erer1243:issue-121868, r=nikic

Fix 32-bit overflows in LLVM composite constants

Inspired by #121868. Fixes unsoundness created when constructing constant arrays, strings, and structs with 2^32 or more elements on x86_64. This introduces copies of a few LLVM functions that have their signatures updated to use size_t in place of unsigned int. Alternatively we could just add overflow checks and just disallow huge composite constants. That introduces less code, but maybe a huge static block of memory is useful in embedded/no-os situations?

Author: matthiaskrgr

compiler/rustc_codegen_llvm/src/common.rs

@@ -95,11 +95,13 @@ impl<'ll> BackendTypes for CodegenCx<'ll, '_> {
 
 impl<'ll> CodegenCx<'ll, '_> {
     pub fn const_array(&self, ty: &'ll Type, elts: &[&'ll Value]) -> &'ll Value {
-        unsafe { llvm::LLVMConstArray(ty, elts.as_ptr(), elts.len() as c_uint) }
+        let len = u64::try_from(elts.len()).expect("LLVMConstArray2 elements len overflow");
+        unsafe { llvm::LLVMConstArray2(ty, elts.as_ptr(), len) }
     }
 
     pub fn const_vector(&self, elts: &[&'ll Value]) -> &'ll Value {
-        unsafe { llvm::LLVMConstVector(elts.as_ptr(), elts.len() as c_uint) }
+        let len = c_uint::try_from(elts.len()).expect("LLVMConstVector elements len overflow");
+        unsafe { llvm::LLVMConstVector(elts.as_ptr(), len) }
     }
 
     pub fn const_bytes(&self, bytes: &[u8]) -> &'ll Value {
@@ -108,8 +110,8 @@ impl<'ll> CodegenCx<'ll, '_> {
 
     pub fn const_get_elt(&self, v: &'ll Value, idx: u64) -> &'ll Value {
         unsafe {
-            assert_eq!(idx as c_uint as u64, idx);
-            let r = llvm::LLVMGetAggregateElement(v, idx as c_uint).unwrap();
+            let idx = c_uint::try_from(idx).expect("LLVMGetAggregateElement index overflow");
+            let r = llvm::LLVMGetAggregateElement(v, idx).unwrap();
 
             debug!("const_get_elt(v={:?}, idx={}, r={:?})", v, idx, r);
 
@@ -329,7 +331,7 @@ pub fn val_ty(v: &Value) -> &Type {
 pub fn bytes_in_context<'ll>(llcx: &'ll llvm::Context, bytes: &[u8]) -> &'ll Value {
     unsafe {
         let ptr = bytes.as_ptr() as *const c_char;
-        llvm::LLVMConstStringInContext(llcx, ptr, bytes.len() as c_uint, True)
+        llvm::LLVMConstStringInContext2(llcx, ptr, bytes.len(), True)
     }
 }
 
@@ -338,9 +340,8 @@ pub fn struct_in_context<'ll>(
     elts: &[&'ll Value],
     packed: bool,
 ) -> &'ll Value {
-    unsafe {
-        llvm::LLVMConstStructInContext(llcx, elts.as_ptr(), elts.len() as c_uint, packed as Bool)
-    }
+    let len = c_uint::try_from(elts.len()).expect("LLVMConstStructInContext elements len overflow");
+    unsafe { llvm::LLVMConstStructInContext(llcx, elts.as_ptr(), len, packed as Bool) }
 }
 
 #[inline]

compiler/rustc_codegen_llvm/src/llvm/ffi.rs

@@ -936,10 +936,16 @@ extern "C" {
     pub fn LLVMConstReal(RealTy: &Type, N: f64) -> &Value;
 
     // Operations on composite constants
-    pub fn LLVMConstStringInContext(
+    pub fn LLVMConstArray2<'a>(
+        ElementTy: &'a Type,
+        ConstantVals: *const &'a Value,
+        Length: u64,
+    ) -> &'a Value;
+    pub fn LLVMArrayType2(ElementType: &Type, ElementCount: u64) -> &Type;
+    pub fn LLVMConstStringInContext2(
         C: &Context,
         Str: *const c_char,
-        Length: c_uint,
+        Length: size_t,
         DontNullTerminate: Bool,
     ) -> &Value;
     pub fn LLVMConstStructInContext<'a>(
@@ -948,14 +954,6 @@ extern "C" {
         Count: c_uint,
         Packed: Bool,
     ) -> &'a Value;
-
-    // FIXME: replace with LLVMConstArray2 when bumped minimal version to llvm-17
-    // https://github.com/llvm/llvm-project/commit/35276f16e5a2cae0dfb49c0fbf874d4d2f177acc
-    pub fn LLVMConstArray<'a>(
-        ElementTy: &'a Type,
-        ConstantVals: *const &'a Value,
-        Length: c_uint,
-    ) -> &'a Value;
     pub fn LLVMConstVector(ScalarConstantVals: *const &Value, Size: c_uint) -> &Value;
 
     // Constant expressions
@@ -1530,9 +1528,6 @@ extern "C" {
     /// See llvm::LLVMTypeKind::getTypeID.
     pub fn LLVMRustGetTypeKind(Ty: &Type) -> TypeKind;
 
-    // Operations on array, pointer, and vector types (sequence types)
-    pub fn LLVMRustArrayType(ElementType: &Type, ElementCount: u64) -> &Type;
-
     // Operations on all values
     pub fn LLVMRustGlobalAddMetadata<'a>(Val: &'a Value, KindID: c_uint, Metadata: &'a Metadata);
     pub fn LLVMRustIsNonGVFunctionPointerTy(Val: &Value) -> bool;

compiler/rustc_codegen_llvm/src/type_.rs

@@ -233,7 +233,7 @@ impl<'ll, 'tcx> BaseTypeMethods<'tcx> for CodegenCx<'ll, 'tcx> {
     }
 
     fn type_array(&self, ty: &'ll Type, len: u64) -> &'ll Type {
-        unsafe { llvm::LLVMRustArrayType(ty, len) }
+        unsafe { llvm::LLVMArrayType2(ty, len) }
     }
 }
 

compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp

@@ -10,6 +10,7 @@
 #include "llvm/IR/IntrinsicsARM.h"
 #include "llvm/IR/LLVMRemarkStreamer.h"
 #include "llvm/IR/Mangler.h"
+#include "llvm/IR/Value.h"
 #include "llvm/Remarks/RemarkStreamer.h"
 #include "llvm/Remarks/RemarkSerializer.h"
 #include "llvm/Remarks/RemarkFormat.h"
@@ -1223,14 +1224,6 @@ extern "C" void LLVMRustWriteValueToString(LLVMValueRef V,
   }
 }
 
-// LLVMArrayType function does not support 64-bit ElementCount
-// FIXME: replace with LLVMArrayType2 when bumped minimal version to llvm-17
-// https://github.com/llvm/llvm-project/commit/35276f16e5a2cae0dfb49c0fbf874d4d2f177acc
-extern "C" LLVMTypeRef LLVMRustArrayType(LLVMTypeRef ElementTy,
-                                         uint64_t ElementCount) {
-  return wrap(ArrayType::get(unwrap(ElementTy), ElementCount));
-}
-
 DEFINE_SIMPLE_CONVERSION_FUNCTIONS(Twine, LLVMTwineRef)
 
 extern "C" void LLVMRustWriteTwineToString(LLVMTwineRef T, RustStringRef Str) {
@@ -2114,3 +2107,36 @@ extern "C" bool LLVMRustLLVMHasZlibCompressionForDebugSymbols() {
 extern "C" bool LLVMRustLLVMHasZstdCompressionForDebugSymbols() {
   return llvm::compression::zstd::isAvailable();
 }
+
+// Operations on composite constants.
+// These are clones of LLVM api functions that will become available in future releases.
+// They can be removed once Rust's minimum supported LLVM version supports them.
+// See https://github.com/rust-lang/rust/issues/121868
+// See https://llvm.org/doxygen/group__LLVMCCoreValueConstantComposite.html
+
+// FIXME: Remove when Rust's minimum supported LLVM version reaches 19.
+// https://github.com/llvm/llvm-project/commit/e1405e4f71c899420ebf8262d5e9745598419df8
+#if LLVM_VERSION_LT(19, 0)
+extern "C" LLVMValueRef LLVMConstStringInContext2(LLVMContextRef C,
+                                                  const char *Str,
+                                                  size_t Length,
+                                                  bool DontNullTerminate) {
+  return wrap(ConstantDataArray::getString(*unwrap(C), StringRef(Str, Length), !DontNullTerminate));
+}
+#endif
+
+// FIXME: Remove when Rust's minimum supported LLVM version reaches 17.
+// https://github.com/llvm/llvm-project/commit/35276f16e5a2cae0dfb49c0fbf874d4d2f177acc
+#if LLVM_VERSION_LT(17, 0)
+extern "C" LLVMValueRef LLVMConstArray2(LLVMTypeRef ElementTy,
+                                        LLVMValueRef *ConstantVals,
+                                        uint64_t Length) {
+  ArrayRef<Constant *> V(unwrap<Constant>(ConstantVals, Length), Length);
+  return wrap(ConstantArray::get(ArrayType::get(unwrap(ElementTy), Length), V));
+}
+
+extern "C" LLVMTypeRef LLVMArrayType2(LLVMTypeRef ElementTy,
+                                      uint64_t ElementCount) {
+  return wrap(ArrayType::get(unwrap(ElementTy), ElementCount));
+}
+#endif