https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120268
--- Comment #5 from Kael Franco <kaelfandrew at gmail dot com> ---
(In reply to Kael Franco from comment #4)
> (In reply to Andrew Pinski from comment #1)
> > This is only valid with -fallow-store-data-races. Otherwise you can
> > introduce a race condition according to the C/C++11 memory model.
>
> I don't think -fallow-store-data-races is needed if the pointer was restrict
> or TYPE_QUAL_RESTRICT in GCC's tree:
>
> ```
> void f0(int * restrict arg0){
> if(arg0[0])
> arg0[0] = 0;
> }
> ```
Actually nevermind since restrict isn't a threading locking mechanism.
