[libc++] Properly implement array cookies in the ARM ABI (llvm#160182)

When we implemented array cookie support for hardening std::unique_ptr,
the implementation was only done for the Itanium ABI. I did not
initially realize that ARM was using a different ABI for array cookies,
so unique_ptr should not have been hardened on ARM.

However, we were also incorrectly setting the ABI-detection macro: we
were pretending to be using a vanilla Itanium ABI when in reality the
(similar but different) ARM ABI was in use. As a result, unique_ptr was
using the wrong representation for array cookies on ARM, which
fortunately only mattered in the case of overaligned types.

This patch fixes that.

rdar://160852193
(cherry picked from commit b3a1994)

Author: ldionne

libcxx/include/__configuration/abi.h

@@ -30,8 +30,20 @@
 #elif _LIBCPP_ABI_FORCE_MICROSOFT
 #  define _LIBCPP_ABI_MICROSOFT
 #else
+// Windows uses the Microsoft ABI
 #  if defined(_WIN32) && defined(_MSC_VER)
 #    define _LIBCPP_ABI_MICROSOFT
+
+// 32-bit ARM uses the Itanium ABI with a few differences (array cookies, etc),
+// and so does 64-bit ARM on Apple platforms.
+#  elif defined(__arm__) || (defined(__APPLE__) && defined(__aarch64__))
+#    define _LIBCPP_ABI_ITANIUM_WITH_ARM_DIFFERENCES
+
+// Non-Apple 64-bit ARM uses the vanilla Itanium ABI
+#  elif defined(__aarch64__)
+#    define _LIBCPP_ABI_ITANIUM
+
+// We assume that other architectures also use the vanilla Itanium ABI too
 #  else
 #    define _LIBCPP_ABI_ITANIUM
 #  endif

libcxx/include/__memory/array_cookie.h

@@ -13,6 +13,7 @@
 #include <__config>
 #include <__configuration/abi.h>
 #include <__cstddef/size_t.h>
+#include <__memory/addressof.h>
 #include <__type_traits/integral_constant.h>
 #include <__type_traits/is_trivially_destructible.h>
 #include <__type_traits/negation.h>
@@ -26,28 +27,95 @@ _LIBCPP_BEGIN_NAMESPACE_STD
 // Trait representing whether a type requires an array cookie at the start of its allocation when
 // allocated as `new T[n]` and deallocated as `delete[] array`.
 //
-// Under the Itanium C++ ABI [1], we know that an array cookie is available unless `T` is trivially
-// destructible and the call to `operator delete[]` is not a sized operator delete. Under ABIs other
-// than the Itanium ABI, we assume there are no array cookies.
+// Under the Itanium C++ ABI [1] and the ARM ABI which derives from it, we know that an array cookie is available
+// unless `T` is trivially destructible and the call to `operator delete[]` is not a sized operator delete. Under
+// other ABIs, we assume there are no array cookies.
 //
 // [1]: https://itanium-cxx-abi.github.io/cxx-abi/abi.html#array-cookies
-#ifdef _LIBCPP_ABI_ITANIUM
+#if defined(_LIBCPP_ABI_ITANIUM) || defined(_LIBCPP_ABI_ITANIUM_WITH_ARM_DIFFERENCES)
 // TODO: Use a builtin instead
-// TODO: We should factor in the choice of the usual deallocation function in this determination.
+// TODO: We should factor in the choice of the usual deallocation function in this determination:
+//       a cookie may be available in more cases but we ignore those for now.
 template <class _Tp>
 struct __has_array_cookie : _Not<is_trivially_destructible<_Tp> > {};
 #else
 template <class _Tp>
 struct __has_array_cookie : false_type {};
 #endif
 
+struct __itanium_array_cookie {
+  size_t __element_count;
+};
+
+template <class _Tp>
+struct [[__gnu__::__aligned__(_LIBCPP_ALIGNOF(_Tp))]] __arm_array_cookie {
+  size_t __element_size;
+  size_t __element_count;
+};
+
+// Return the element count in the array cookie located before the given pointer.
+//
+// In the Itanium ABI [1]
+// ----------------------
+// The element count is stored immediately before the first element of the array. If the preferred alignment
+// of array elements (which is different from the ABI alignment) is more than that of size_t, additional
+// padding bytes exist before the array cookie. Assuming array elements of size and alignment 16 bytes, that
+// gives us the following layout:
+//
+// |ooooooooxxxxxxxxaaaaaaaaaaaaaaaabbbbbbbbbbbbbbbbccccccccccccccccdddddddddddddddd|
+//  ^^^^^^^^        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+//     |    ^^^^^^^^                               |
+//     |       |                              array elements
+//  padding    |
+//       element count
+//
+//
+// In the Itanium ABI with ARM differences [2]
+// -------------------------------------------
+// The array cookie is stored at the very start of the allocation and it has the following form:
+//
+//    struct array_cookie {
+//      std::size_t element_size; // element_size != 0
+//      std::size_t element_count;
+//    };
+//
+// Assuming elements of size and alignment 32 bytes, this gives us the following layout:
+//
+//  |xxxxxxxxXXXXXXXXooooooooooooooooaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb|
+//   ^^^^^^^^        ^^^^^^^^^^^^^^^^
+//      |    ^^^^^^^^        |       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+// element size  |        padding                                 |
+//         element count                                     array elements
+//
+// We must be careful to take into account the alignment of the array cookie, which may result in padding
+// bytes between the element count and the first element of the array. Note that for ARM, the compiler
+// aligns the array cookie using the ABI alignment, not the preferred alignment of array elements.
+//
+// [1]: https://itanium-cxx-abi.github.io/cxx-abi/abi.html#array-cookies
+// [2]: https://developer.apple.com/documentation/xcode/writing-arm64-code-for-apple-platforms#Handle-C++-differences
 template <class _Tp>
 // Avoid failures when -fsanitize-address-poison-custom-array-cookie is enabled
-_LIBCPP_HIDE_FROM_ABI _LIBCPP_NO_SANITIZE("address") size_t __get_array_cookie(_Tp const* __ptr) {
+_LIBCPP_HIDE_FROM_ABI _LIBCPP_NO_SANITIZE("address") size_t __get_array_cookie([[__maybe_unused__]] _Tp const* __ptr) {
   static_assert(
       __has_array_cookie<_Tp>::value, "Trying to access the array cookie of a type that is not guaranteed to have one");
-  size_t const* __cookie = reinterpret_cast<size_t const*>(__ptr) - 1; // TODO: Use a builtin instead
-  return *__cookie;
+
+#if defined(_LIBCPP_ABI_ITANIUM)
+  using _ArrayCookie = __itanium_array_cookie;
+#elif defined(_LIBCPP_ABI_ITANIUM_WITH_ARM_DIFFERENCES)
+  using _ArrayCookie = __arm_array_cookie<_Tp>;
+#else
+  static_assert(false, "The array cookie layout is unknown on this ABI");
+  struct _ArrayCookie { // dummy definition required to make the function parse
+    size_t element_count;
+  };
+#endif
+
+  char const* __array_cookie_start = reinterpret_cast<char const*>(__ptr) - sizeof(_ArrayCookie);
+  _ArrayCookie __cookie;
+  // This is necessary to avoid violating strict aliasing. It's valid because _ArrayCookie is an
+  // implicit lifetime type.
+  __builtin_memcpy(std::addressof(__cookie), __array_cookie_start, sizeof(_ArrayCookie));
+  return __cookie.__element_count;
 }
 
 _LIBCPP_END_NAMESPACE_STD

libcxx/test/std/utilities/smartptr/unique.ptr/unique.ptr.class/unique.ptr.observers/assert.subscript.pass.cpp

@@ -58,15 +58,18 @@ void test() {
   {
     {
       std::unique_ptr<WithCookie[]> ptr(new WithCookie[5]);
+      assert(&ptr[1] == ptr.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(ptr[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
     {
       std::unique_ptr<WithCookie[]> ptr = std::make_unique<WithCookie[]>(5);
+      assert(&ptr[1] == ptr.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(ptr[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
 #if TEST_STD_VER >= 20
     {
       std::unique_ptr<WithCookie[]> ptr = std::make_unique_for_overwrite<WithCookie[]>(5);
+      assert(&ptr[1] == ptr.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(ptr[6] = WithCookie(), "unique_ptr<T[]>::operator[](index): index out of range");
     }
 #endif
@@ -82,11 +85,13 @@ void test() {
   {
     {
       std::unique_ptr<NoCookie[]> ptr = std::make_unique<NoCookie[]>(5);
+      assert(&ptr[1] == ptr.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(ptr[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
 #  if TEST_STD_VER >= 20
     {
       std::unique_ptr<NoCookie[]> ptr = std::make_unique_for_overwrite<NoCookie[]>(5);
+      assert(&ptr[1] == ptr.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(ptr[6] = NoCookie(), "unique_ptr<T[]>::operator[](index): index out of range");
     }
 #  endif
@@ -101,6 +106,7 @@ void test() {
     {
       std::unique_ptr<T[]> ptr = std::make_unique<T[]>(5);
       std::unique_ptr<T[]> other(std::move(ptr));
+      assert(&other[1] == other.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(other[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
 
@@ -109,13 +115,15 @@ void test() {
       std::unique_ptr<T[]> ptr = std::make_unique<T[]>(5);
       std::unique_ptr<T[]> other;
       other = std::move(ptr);
+      assert(&other[1] == other.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(other[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
 
     // Bounds carried through converting move-constructor
     {
       std::unique_ptr<T[]> ptr = std::make_unique<T[]>(5);
       std::unique_ptr<T[], MyDeleter> other(std::move(ptr));
+      assert(&other[1] == other.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(other[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
 
@@ -124,6 +132,7 @@ void test() {
       std::unique_ptr<T[]> ptr = std::make_unique<T[]>(5);
       std::unique_ptr<T[], MyDeleter> other;
       other = std::move(ptr);
+      assert(&other[1] == other.get() + 1); // ensure no assertion
       TEST_LIBCPP_ASSERT_FAILURE(other[6], "unique_ptr<T[]>::operator[](index): index out of range");
     }
   });
@@ -144,6 +153,34 @@ struct WithCookie {
   char padding[Size];
 };
 
+template <std::size_t Size>
+struct alignas(128) OveralignedNoCookie {
+  char padding[Size];
+};
+
+template <std::size_t Size>
+struct alignas(128) OveralignedWithCookie {
+  OveralignedWithCookie() = default;
+  OveralignedWithCookie(OveralignedWithCookie const&) {}
+  OveralignedWithCookie& operator=(OveralignedWithCookie const&) { return *this; }
+  ~OveralignedWithCookie() {}
+  char padding[Size];
+};
+
+// These types have a different ABI alignment (alignof) and preferred alignment (__alignof) on some platforms.
+// Make sure things work with these types because array cookies can be sensitive to preferred alignment on some
+// platforms.
+struct WithCookiePreferredAlignment {
+  WithCookiePreferredAlignment() = default;
+  WithCookiePreferredAlignment(WithCookiePreferredAlignment const&) {}
+  WithCookiePreferredAlignment& operator=(WithCookiePreferredAlignment const&) { return *this; }
+  ~WithCookiePreferredAlignment() {}
+  long double data;
+};
+struct NoCookiePreferredAlignment {
+  long double data;
+};
+
 int main(int, char**) {
   test<WithCookie<1>, NoCookie<1>>();
   test<WithCookie<2>, NoCookie<2>>();
@@ -153,7 +190,18 @@ int main(int, char**) {
   test<WithCookie<16>, NoCookie<16>>();
   test<WithCookie<32>, NoCookie<32>>();
   test<WithCookie<256>, NoCookie<256>>();
+
+  test<OveralignedWithCookie<1>, OveralignedNoCookie<1>>();
+  test<OveralignedWithCookie<2>, OveralignedNoCookie<2>>();
+  test<OveralignedWithCookie<3>, OveralignedNoCookie<3>>();
+  test<OveralignedWithCookie<4>, OveralignedNoCookie<4>>();
+  test<OveralignedWithCookie<8>, OveralignedNoCookie<8>>();
+  test<OveralignedWithCookie<16>, OveralignedNoCookie<16>>();
+  test<OveralignedWithCookie<32>, OveralignedNoCookie<32>>();
+  test<OveralignedWithCookie<256>, OveralignedNoCookie<256>>();
+
   test<std::string, int>();
+  test<WithCookiePreferredAlignment, NoCookiePreferredAlignment>();
 
   return 0;
 }