Re: [PATCH] fix a race condition in cancelable mcs spinlocks

[Jason Low <jason.low2@hp.com>]

On Mon, 2014-06-02 at 10:14 -0400, Waiman Long wrote:
> On 06/01/2014 01:53 PM, Mikulas Patocka wrote:
> >   struct optimistic_spin_queue {
> > -	struct optimistic_spin_queue *next, *prev;
> > +	atomic_pointer_t next;
> > +	struct optimistic_spin_queue *prev;
> >   	int locked; /* 1 if lock acquired */
> >   };
> 
> Is there a way to do it without changing the pointer type? It will make 
> the code harder to read and understand.

I agree that it would be nice if there is a way to fix this without
changing the pointer type of "next". The change of the type to
atomic_pointer_t might make it less obvious what "next" is for. This is
then compounded with "prev" being kept as a pointer to
optimistic_spin_queue, which can further make it appear as if "next" may
potentially point to something different.