Post-Increment Operator: Unexpected Behavior [duplicate]

Question

Possible Duplicate:
Could anyone explain these undefined behaviors (i = i++ + ++i , i = i++, etc…)

My code is as follows:

#include <stdio.h>
int main()
{
  int x = 10, y = 0;
  x = x++;
  printf("x: %d\n", x);
  y = x++;
  printf("y: %d\n", y);
}

Given the nature of post-increment, I would expect the following output:

x: 10
y: 10

My reasoning is that in line 5, x should be assigned to its initial value after the increment takes place.

Instead, however, I get this:

x: 11
y: 11

Digging into the assembly, this looks like a deliberate choice to me:

LCFI2:
        movl    $10, -4(%rbp)   // this is x
        movl    $0, -8(%rbp)    // this is y
        incl    -4(%rbp)        // x is simply incremented
        movl    -4(%rbp), %esi
        leaq    LC0(%rip), %rdi
        movl    $0, %eax
        call    _printf
        movl    -4(%rbp), %eax  // now x is saved in a register,
        movl    %eax, -8(%rbp)  // copied to y,
        incl    -4(%rbp)        // and finally incremented
        movl    -8(%rbp), %esi
        leaq    LC1(%rip), %rdi
        movl    $0, %eax
        call    _printf

What's going on here? Is GCC trying to save me from myself? I don't have a language reference handy but I would have thought that this breaks the intended semantics.

Answer 1

The behaviour is undefined as there is no intervening sequence point in x = x++, see e.g. the C FAQ.