Re: GCC extension for atomic access to members

[Martin Sebor <msebor at gmail dot com>]

On 09/18/2017 06:19 AM, Florian Weimer wrote:
> I would like to see the GCC project to document that if the address of a
> member is taken, this does not constitute an access to the object as a
> whole.
>
> That is, in the following code:
>
> #include <stdatomic.h>
>
> struct S {
>   _Atomic int a;
>   int b;
> };
>
> int
> load_a (struct S *p)
> {
>   return atomic_load_explicit (&p->a, memory_order_relaxed);
> }
>
> int
> store_b (struct S *p, int b)
> {
>   p->b = b;
> }
>
> If one thread calls load_a and another thread calls store_b on the same
> struct S *, no data race happens.
>
> This is an extension over the C standard because of the way “->” is
> defined.  C requires that E1->E2 it is evaluated as (*(E1))->E2, and *E1
> is defined as an access to the entire struct, so there is a data race in
> load_a with the assignment in store_b.

It's true that E1->E2 is equivalent to (*(E1))->E2 (this is made
clear in C11 by Footnote 96) but C doesn't use the term "access"
in the definition of the * operator.  The * operator is defined
to result in an lvalue that designates an object, and operator .
to evaluate the first operand, which doesn't imply accessing it.

If E1->E2 implied accessing all members of *E1 then there would
be no way to assign an initial value to just the E2 member without
also accessing all other members of *E1, even those that haven't
yet been initialized.

Martin