Re: [PATCH 3/4][resend] Show kernel stack usage in /proc/meminfo and OOM log output

[KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>]

> On Mon, 13 Jul 2009 15:02:25 +0900 (JST)
> KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> wrote:
> 
> > ChangeLog
> >   Since v1
> >    - Rewrote the descriptin (Thanks Christoph!)
> > 
> > =====================
> > Subject: [PATCH] Show kernel stack usage in /proc/meminfo and OOM log output
> > 
> > The amount of memory allocated to kernel stacks can become significant and
> > cause OOM conditions. However, we do not display the amount of memory
> > consumed by stacks.
> > 
> > Add code to display the amount of memory used for stacks in /proc/meminfo.
> > 
> > ...
> >  
> > +static void account_kernel_stack(struct thread_info *ti, int account)
> > +{
> > +	struct zone *zone = page_zone(virt_to_page(ti));
> > +
> > +	mod_zone_page_state(zone, NR_KERNEL_STACK, account);
> > +}
> > +
> >  void free_task(struct task_struct *tsk)
> >  {
> >  	prop_local_destroy_single(&tsk->dirties);
> > +	account_kernel_stack(tsk->stack, -1);
> 
> But surely there are other less expensive ways of calculating this. 
> The number we want is small-known-constant * number-of-tasks.
> 
> number-of-tasks probably isn't tracked, but can be calculated along the
> lines of nr_running(), nr_uninterruptible() and nr_iowait().

But, nr_running() don't know zone information. we really need
per-zone tracking IMHO.


> number-of-tasks is also equal to number-of-task_structs and
> number-of_thread_infos which can be obtained from slab (if the arch
> implemented these via slab - uglier).

You know, Almost architecture doesn't use slab for kernel-stack.




--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>