reinterpret_cast vs strict aliasing
Asked Answered
E

3

14

I was reading about strict aliasing, but its still kinda foggy and I am never sure where is the line of defined / undefined behaviour. The most detailed post i found concentrates on C. So it would be nice if you could tell me if this is allowed and what has changed since C++98/11/...

#include <iostream>
#include <cstring>

template <typename T> T transform(T t);

struct my_buffer {
    char data[128];
    unsigned pos;
    my_buffer() : pos(0) {}
    void rewind() { pos = 0; }    
    template <typename T> void push_via_pointer_cast(const T& t) {
        *reinterpret_cast<T*>(&data[pos]) = transform(t);
        pos += sizeof(T);
    }
    template <typename T> void pop_via_pointer_cast(T& t) {
        t = transform( *reinterpret_cast<T*>(&data[pos]) );
        pos += sizeof(T);
    }            
};    
// actually do some real transformation here (and actually also needs an inverse)
// ie this restricts allowed types for T
template<> int transform<int>(int x) { return x; }
template<> double transform<double>(double x) { return x; }

int main() {
    my_buffer b;
    b.push_via_pointer_cast(1);
    b.push_via_pointer_cast(2.0);
    b.rewind();
    int x;
    double y;
    b.pop_via_pointer_cast(x);
    b.pop_via_pointer_cast(y);
    std::cout << x << " " << y << '\n';
}

Please dont pay too much attention to a possible out-of-bounds access and the fact that maybe there is no need to write something like that. I know that char* is allowed to point to anything, but I also have a T* that points to a char*. And maybe there is something else I am missing.

Here is a complete example also including push/pop via memcpy, which afaik isn't affected by strict aliasing.

TL;DR: Does the above code exhibit undefined behaviour (neglecting a out-of-bound acces for the moment), if yes, why? Did anything change with C++11 or one of the newer standards?

Enscroll answered 23/8, 2018 at 9:38 Comment(2)
My answer to "What is the strict aliasing rule?" covers C++ extensively and I believe covers your question as well. It is not uncommon for old answers to get new and much better answers over time, so it is important to look at all of the answers and not just the top ones. my most upvoted answer came 4 years after the original question was asked.Ruprecht
@ShafikYaghmour thanks, I will take a look. The problem with the other question is that it is tagged as C and C++faq but it does not have the C++ tag, so when I saw the top answers concentrating on C, I didnt consider it as a dupe. If you think it is one, feel free to flagEnscroll
S
11

I know that char* is allowed to point to anything, but I also have a T* that points to a char*.

Right, and that is a problem. While the pointer cast itself has defined behaviour, using it to access a non-existent object of type T is not.

Unlike C, C++ does not allow impromptu creation of objects*. You cannot simply assign to some memory location as type T and have an object of that type be created, you need an object of that type to be there already. This requires placement new. Previous standards were ambiguous on it, but currently, per [intro.object]:

1 [...] An object is created by a definition (6.1), by a new-expression (8.3.4), when implicitly changing the active member of a union (12.3), or when a temporary object is created (7.4, 15.2). [...]

Since you are not doing any of these things, no object is created.

Furthermore, C++ does not implicitly consider pointers to different object at the same address as equivalent. Your &data[pos] computes a pointer to a char object. Casting it to T* does not make it point to any T object residing at that address, and dereferencing that pointer has undefined behaviour. C++17 adds std::launder, which is a way to let the compiler know that you want to access a different object at that address than what you have a pointer to.

When you modify your code to use placement new and std::launder, and ensure you have no misaligned accesses (I presume you left that out for brevity), your code will have defined behaviour.

* There is discussion on allowing this in a future version of C++.

Scimitar answered 23/8, 2018 at 10:0 Comment(6)
i left out the memcpy solution for brevity. std::launder is completely new to me ;)Enscroll
after thinking about it, I am not so sure anymore if your argumentation convinces me. What about using memcpy instead of the casts? In that case I am also not really creating an instance of T in place of data, but afaik there is no problem when using memcpyEnscroll
@user463035818 Right, but in that case, by not accessing it as if it were a T, you avoid the need for having an object of type T there.Scimitar
that makes sense. So in the char array there is just some bits that happen to be memcopyable to a T (because I copied the bits of a T in that place) but not a T that I could retrive via a cast, because the language requires that if I treat some memory as if there was some object of type T I have to first create an object of type T at that place.Enscroll
@user463035818 Exactly.Scimitar
awesome, this was just the bit that was missing for me to understand why memcpy makes the differenceEnscroll
B
15

Aliasing is a situation when two entities refer to the same object. It may be either references or pointers.

int x;
int* p = &x;
int& r = x;
// aliases: x, r и *p  refer to same object.

It's important for compiler to expect that if a value was written using one name it would be accessible through another.

int foo(int* a, int* b) {
  *a = 0;
  *b = 1;
  return *a; 
  // *a might be 0, might be 1, if b points at same object. 
  // Compiler can't short-circuit this to "return 0;"
}

Now if pointers are of unrelated types, there is no reason for compiler to expect that they point at same address. This is the simplest UB:

int foo( float *f, int *i ) { 
    *i = 1;               
    *f = 0.f;            
   return *i;
}

int main() {
    int a = 0;

    std::cout << a << std::endl; 
    int x = foo(reinterpret_cast<float*>(&a), &a);
    std::cout << a << "\n"; 
    std::cout << x << "\n";   // Surprise?
}
// Output 0 0 0 or 0 0 1 , depending on optimization. 

Simply put, strict aliasing means that compiler expects names of unrelated types refer to object of different type, thus located in separate storage units. Because addresses used to access those storage units are de-facto same, result of accessing stored value is undefined and usually depends on optimization flags.

memcpy() circumvents that by taking the address, by pointer to char, and makes copy of data stored, within code of library function.

Strict aliasing applies to union members, which described separately, but reason is same: writing to one member of union doesn't guarantee the values of other members to change. That doesn't apply to shared fields in beginning of struct stored within union. Thus, type punning by union is prohibited. (Most compilers do not honor this for historical reasons and convenience of maintaining legacy code.)

From 2017 Standard: 6.10 Lvalues and rvalues

8 If a program attempts to access the stored value of an object through a glvalue of other than one of the following types the behavior is undefined

(8.1) — the dynamic type of the object,

(8.2) — a cv-qualified version of the dynamic type of the object,

(8.3) — a type similar (as defined in 7.5) to the dynamic type of the object,

(8.4) — a type that is the signed or unsigned type corresponding to the dynamic type of the object,

(8.5) — a type that is the signed or unsigned type corresponding to a cv-qualified version of the dynamic type of the object,

(8.6) — an aggregate or union type that includes one of the aforementioned types among its elements or nonstatic data members (including, recursively, an element or non-static data member of a subaggregate or contained union),

(8.7) — a type that is a (possibly cv-qualified) base class type of the dynamic type of the object,

(8.8) — a char, unsigned char, or std::byte type.

In 7.5

1 A cv-decomposition of a type T is a sequence of cvi and Pi such that T is “cv0 P0 cv1 P1 · · · cvn−1 Pn−1 cvn U” for n > 0, where each cvi is a set of cv-qualifiers (6.9.3), and each Pi is “pointer to” (11.3.1), “pointer to member of class Ci of type” (11.3.3), “array of Ni”, or “array of unknown bound of” (11.3.4). If Pi designates an array, the cv-qualifiers cvi+1 on the element type are also taken as the cv-qualifiers cvi of the array. [ Example: The type denoted by the type-id const int ** has two cv-decompositions, taking U as “int” and as “pointer to const int”. —end example ] The n-tuple of cv-qualifiers after the first one in the longest cv-decomposition of T, that is, cv1, cv2, . . . , cvn, is called the cv-qualification signature of T.

2 Two types T1 and T2 are similar if they have cv-decompositions with the same n such that corresponding Pi components are the same and the types denoted by U are the same.

Outcome is: while you can reinterpret_cast the pointer to a different, unrelated and not similar type, you can't use that pointer to access stored value:

char* pc = new char[100]{1,2,3,4,5,6,7,8,9,10}; // Note, initialized.
int* pi = reinterpret_cast<int*>(pc);  // no problem.
int i = *pi; // UB
char* pc2 = reinterpret_cast<char*>(pi+2);  // *(pi+2) would be UB
char c = *pc2; // no problem, unless increment didn't put us beyond array bound.
// c equals to 9

'reinterpret_cast' doesn't create objects. To dereference a pointer at a non-existing object is Undefined Behaviour, so you can't use dereferenced result of cast for writing if class it points to wasn't trivial.

Batchelor answered 23/8, 2018 at 10:5 Comment(10)
appreciate your exhaustive answer, but the case of a int foo( float *f, int *i ) is rather obvious while I dont see (maybe I will after studying the answers more carefully) what would be the benefit of strict aliasing for the compiler in my codeEnscroll
btw std::memcpy takes void* as parameter, but iirc there are the same excpetions for void* as for char* regarding strict aliasingEnscroll
I think the term "aliasing" normally refers to the situation where one pointers or reference are used to reference the same storage as another pointer or reference that will also be used to reference that storage in conflicting fashion, and the former isn't known to be derived from the latter. Given void foo(char *p) { *p^=1;} long long bar(long long *x) { if (*x) foo((char*)x); return *x;} I don't think the pointer p within foo would normally be regarded as "aliasing" pointer x from bar, because anything that knows that both exist would know that p is derived from x.Iila
@user463035818 True, it declared as 'void*' for sake of implicit cast of pointers, but by documentation it reinterprets pointers as 'unsigned char*'. If you cast one way, then back, it doesn't break anything. Your code, as is, got issue with inability to construct objects (all them were constructed before?) and absence of type safety. It looks like a miniature memory pool with stack interface and probably should be built as such, technically it doesn't involve breach of aliasing as long as you do not push one type and pop another.Batchelor
I'm not totally clear on this whole aliasing thing still. I ran your example (third code snippet), and the cout where you wrote surprise didn't surprise me at all actually. It printed out the 4 byte integer representation of the bits used to populate the address for initializing the float value at the same address. I was expecting the compiler would (this was the surprise I was expecting) optimize the float assignment out and instead return the int value of 1, but it didn't give me that - it gave me the float value interpreted as an int.Declinature
@Declinature change to -O2 and you get 1 instead of 0, if you're on gcc x86. Or even something odder on other platforms. It's Undefined Behavior, so result can be any.Batchelor
pi+2 this has undefined behavior. Did you mean reinterpret_cast<char*>(pi) + 2?Arola
@LanguageLawyer incrementing pointer here isn't UB and that memory location is within array's storage, it gets dereferenced as char* .Batchelor
incrementing pointer here isn't UB Then what is the behavior of pi+2?Arola
@Swift-FridayPie :char* pc2 = reinterpret_cast<char*>(pi+2); why is this UB ?Sublittoral
S
11

I know that char* is allowed to point to anything, but I also have a T* that points to a char*.

Right, and that is a problem. While the pointer cast itself has defined behaviour, using it to access a non-existent object of type T is not.

Unlike C, C++ does not allow impromptu creation of objects*. You cannot simply assign to some memory location as type T and have an object of that type be created, you need an object of that type to be there already. This requires placement new. Previous standards were ambiguous on it, but currently, per [intro.object]:

1 [...] An object is created by a definition (6.1), by a new-expression (8.3.4), when implicitly changing the active member of a union (12.3), or when a temporary object is created (7.4, 15.2). [...]

Since you are not doing any of these things, no object is created.

Furthermore, C++ does not implicitly consider pointers to different object at the same address as equivalent. Your &data[pos] computes a pointer to a char object. Casting it to T* does not make it point to any T object residing at that address, and dereferencing that pointer has undefined behaviour. C++17 adds std::launder, which is a way to let the compiler know that you want to access a different object at that address than what you have a pointer to.

When you modify your code to use placement new and std::launder, and ensure you have no misaligned accesses (I presume you left that out for brevity), your code will have defined behaviour.

* There is discussion on allowing this in a future version of C++.

Scimitar answered 23/8, 2018 at 10:0 Comment(6)
i left out the memcpy solution for brevity. std::launder is completely new to me ;)Enscroll
after thinking about it, I am not so sure anymore if your argumentation convinces me. What about using memcpy instead of the casts? In that case I am also not really creating an instance of T in place of data, but afaik there is no problem when using memcpyEnscroll
@user463035818 Right, but in that case, by not accessing it as if it were a T, you avoid the need for having an object of type T there.Scimitar
that makes sense. So in the char array there is just some bits that happen to be memcopyable to a T (because I copied the bits of a T in that place) but not a T that I could retrive via a cast, because the language requires that if I treat some memory as if there was some object of type T I have to first create an object of type T at that place.Enscroll
@user463035818 Exactly.Scimitar
awesome, this was just the bit that was missing for me to understand why memcpy makes the differenceEnscroll
A
4

Short answer:

  1. You may not do this: *reinterpret_cast<T*>(&data[pos]) = until there has been an object of type T constructed at the pointed-to address. Which you can accomplish by placement new.

  2. Even then, you might need to use std::launder as for C++17 and later, since you access the created object (of type T) through a pointer &data[pos] of type char*.

"Direct" reinterpret_cast is allowed only in some special cases, e.g., when T is std::byte, char, or unsigned char.

Before C++17 I would use the memcpy-based solution. Compiler will likely optimize away any unnecessary copies.

Autoeroticism answered 23/8, 2018 at 9:58 Comment(7)
Whoa, 1. you can. What's the diference between: int *a = reinterpret_cast<int*>(malloc(sizoeof(int))); *a = 5; and char buf[sizeof(int)]; int *a = reinterpre_cast<int*>(buf); *a = 5; ?Bodycheck
@Bodycheck No, C++ is very explicit about when objects are created. Assignment does not implicitly create an object.Scimitar
Ok, thank you! Need to delve into that topic a bit more then.Bodycheck
@Bodycheck It's not a simple topic :). See, for instance, the bottom part of this answer: https://mcmap.net/q/92201/-what-is-the-purpose-of-std-launder. Or, even better, this one: https://mcmap.net/q/93394/-does-this-really-break-strict-aliasing-rules.Autoeroticism
@Bodycheck NB std::launder got own limitation if you try access aggregate types, e.g. arrays of arrays or structsBatchelor
@DanielLangr : are we allowed to do this : T val = *reinterpret_cast<T*>(&data[pos]) . If we are only reading and never modifying the data , is it allowed or is there still possibility of UB ?Sublittoral
@Sublittoral I need to admit that I am not sure whether this will work in general. Also, different standards may resolve this situation differently. But note that you cannot use memcpy even for just reading the binary representation of non-trivial objects. Which is effectively the same if you are only reading data through reinterpret_cast to char.Autoeroticism

© 2022 - 2024 — McMap. All rights reserved.