[openib-general] Re: [PATCH] IPOIB: initialization patch in ipoib

[Grant Grundler]

On Fri, Feb 18, 2005 at 05:45:08PM +0200, Michael S. Tsirkin wrote:
> > -        kfree(neigh);
> >         neigh->neighbour->ops->destructor = NULL;
> > +        kfree(neigh);
> > 
> 
> If the order of these operations really matters, wont you need
> some kind of memory barrier in between?

The original code was wrong even on a strongly ordered memory machine.

grant