[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170901092108.lb3jla2hpczjvrh5@dhcp22.suse.cz>
Date: Fri, 1 Sep 2017 11:21:08 +0200
From: Michal Hocko <mhocko@...nel.org>
To: Konstantin Khlebnikov <khlebnikov@...dex-team.ru>
Cc: Kyeongdon Kim <kyeongdon.kim@....com>, akpm@...ux-foundation.org,
sfr@...b.auug.org.au, ying.huang@...el.com, vbabka@...e.cz,
hannes@...xchg.org, xieyisheng1@...wei.com, luto@...nel.org,
shli@...com, mgorman@...hsingularity.net, hillf.zj@...baba-inc.com,
kemi.wang@...el.com, rientjes@...gle.com, bigeasy@...utronix.de,
iamjoonsoo.kim@....com, bongkyu.kim@....com,
linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH] mm/vmstats: add counters for the page frag cache
On Fri 01-09-17 12:12:36, Konstantin Khlebnikov wrote:
> IMHO that's too much counters.
> Per-node NR_FRAGMENT_PAGES should be enough for guessing what's going on.
> Perf probes provides enough features for furhter debugging.
I would tend to agree. Adding a counter based on a single debugging
instance sounds like an overkill to me. Counters should be pretty cheep
but this is way too specialized API to export to the userspace.
We have other interfaces to debug memory leaks like page_owner.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists