linux-kernel - Re: [patch] mm: memcontrol: use page lists for uncharge batching

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20140718150719.GH29639@cmpxchg.org>
Date:	Fri, 18 Jul 2014 11:07:19 -0400
From:	Johannes Weiner <hannes@...xchg.org>
To:	Michal Hocko <mhocko@...e.cz>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Hugh Dickins <hughd@...gle.com>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org
Subject: Re: [patch] mm: memcontrol: use page lists for uncharge batching

On Thu, Jul 17, 2014 at 05:29:36PM +0200, Michal Hocko wrote:
> On Mon 07-07-14 14:55:58, Johannes Weiner wrote:
> > Pages are now uncharged at release time, and all sources of batched
> > uncharges operate on lists of pages.  Directly use those lists, and
> > get rid of the per-task batching state.
> > 
> > This also batches statistics accounting, in addition to the res
> > counter charges, to reduce IRQ-disabling and re-enabling.
> 
> It is probably worth noticing that there is a higher chance of missing
> threshold events now when we can accumulate huge number of uncharges
> during munmaps. I do not think this is earth shattering and the overall
> improvement is worth it but changelog should mention it.

Does this actually matter, though?  We might deliver events a few
pages later than before, but as I read the threshold code, once
invoked it catches up from the last delivered threshold to the new
usage.  So we shouldn't *miss* any events.

> > Signed-off-by: Johannes Weiner <hannes@...xchg.org>
> 
> With the follow up fix from
> http://marc.info/?l=linux-mm&m=140552814228135&w=2
> 
> Acked-by: Michal Hocko <mhocko@...e.cz>

Thanks!

> > +static void uncharge_list(struct list_head *page_list)
> > +{
> > +	struct mem_cgroup *memcg = NULL;
> > +	unsigned long nr_memsw = 0;
> > +	unsigned long nr_anon = 0;
> > +	unsigned long nr_file = 0;
> > +	unsigned long nr_huge = 0;
> > +	unsigned long pgpgout = 0;
> > +	unsigned long nr_mem = 0;
> > +	struct list_head *next;
> > +	struct page *page;
> > +
> > +	next = page_list->next;
> > +	do {
> 
> I would use list_for_each_entry here which would also save list_empty
> check in mem_cgroup_uncharge_list

list_for_each_entry() wouldn't work for the singleton list where we
pass in page->lru.  That's why it's a do-while that always does the
first page before checking whether it looped back to the list head.

Do we need a comment for that?  I'm not convinced, there are only two
callsites, and the one that passes the singleton page->lru is right
below this function.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/