Re: [RFC] some page can't be migrated

[Christoph Lameter]

On Sat, 26 Jan 2008, Andrew Morton wrote:

> > 
> > -	if (!page->mapping)
> > +	if (!page->mapping) {
> > +		if (!PageAnon(page) && PagePrivate(page))
> > +			try_to_release_page(page, GFP_KERNEL);
> >  		goto rcu_unlock;
> > +	}
> 
> We call something(GFP_KERNEL) under rcu_read_lock()?  I've lost track of
> the myriad flavours of rcu which we purport to support, but I don't think
> they'll all like us blocking under rcu_read_lock().

I still think your solution is good but even with the original patch we do 
not call try_to_release_page() with GFP_KERNEL under rcu.

   if (PageAnon(page)) {
                rcu_read_lock();
                rcu_locked = 1;
        }


rcu is only active if we have an anonymous page and in that case we do not
call try_to_release_page(). Just to make sure that Nick and I do not get 
dinged needlessly....

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/