[PATCH]: Fix PR tree-optimization/21407

[Daniel Berlin]

> Construct 2:
> 
>  struct Base
>  {
>    int a;
>    int b;
>  };
> 
>  struct Derived
>  {
>    int c;
>    int d;
>    struct Base b;
>    int e;
>    int f;
>  };
> 
>  void foo(struct Base *b)
>  {
>    struct Derived *d = 
>      (struct Derived *) ((char *)b) - offsetof(struct Derived, b);
> 
>    d->c = 23;
>  }
> 
>  void bar(struct Derived *d)
>  {
>    foo (&d->b);
>  }

> However, it is my opinion that construct 2 *is* allowed by the
> standard.  Obviously nothing fishy is happening up till the first
> statement of foo.  foo begins by converting b to a char*, which is an
> escape hatch - everything aliases an array of char (6.5p7), and the
> pointer is guaranteed to point to the first byte of Derived.b
> (6.3.2.3p7).

If i change "struct Base b" to "int b", and have foo take an "int *b" is
it still legal?
After all, you are still casting to a char * what appears to be a
pointer guaranteed to point to the beginning of Derived, by the same
logic.

I only ask this question because it has optimization implications.  if
it's only legal to do this when you pass a pointer to a structure that
is a member, and not a pointer to any old type member, i can simply
modify the code to take this into account.