Re: [RFC PATCH 5/5] refresh VM committed space after a task migration

[Andrea Righi]

Dave Hansen wrote:
> On Tue, 2008-06-10 at 01:33 +0200, Andrea Righi wrote:
> > +       preempt_disable();
> > +       committed = atomic_long_read(&p->vm_committed_space);
> > +       atomic_long_sub(committed, &old_mem->vmacct.vm_committed_space);
> > +       atomic_long_add(committed, &mem->vmacct.vm_committed_space);
> > +       preempt_enable();
> >  out:
> >         mmput(mm);
> >  }
>
> Why bother with the preempt stuff here?  What does the actually protect
> against?  I assume that you're trying to keep other tasks that might run
> on this CPU from seeing weird, inconsistent numbers in here.  Is there
> some other looks that keeps *other* cpus from seeing this?
>
> In any case, I think it needs a big, fat comment.

Yes, true, mem_cgroup_move_task() is called after the task->cgroups
pointer has been changed. So, even if task changes its committed space
between the atomic_long_sub() and atomic_long_add() it will be correctly
accounted in the new cgroup.

-Andrea
--
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/