Re: mm/slab: What is cache_reap work for?

[Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>]

Christoph Lameter wrote:
> On Tue, 4 Jul 2017, Tetsuo Handa wrote:
> 
> > Thank you for explanation. What I observed is that it seems that
> > cache_reap work was not able to run because it used system_wq when
> > the system was unable to allocate memory for new worker thread due to
> > infinite too_many_isolated() loop in shrink_inactive_list().
> 
> Its ok for it not to run for awhile but that potentially traps memory. And
> you want more memory to be freed.
> 
> > I wondered whether cache_reap work qualifies as an mm_percpu_wq user
> > if cache_reap work does something like what vmstat_work work does (e.g.
> > update statistic counters which affect progress of memory allocation).
> > But "calls other functions that are used during regular slab allocation"
> > means cache_reap work cannot qualify as an mm_percpu_wq user...
> 
> Well if you audit the functions called then you may be able to get there.
> 

As far as I checked, it seems that the only operation that cache_reap work
might sleep is cond_resched(). Thus, I think that cache_reap work can
qualify as an mm_percpu_wq user.

--
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>