From: Glauber Costa <glommer@parallels.com> To: Sha Zhengju <handai.szj@gmail.com> Cc: linux-mm@kvack.org, cgroups@vger.kernel.org, Tejun Heo <tj@kernel.org>, Andrew Morton <akpm@linux-foundation.org>, Michal Hocko <mhocko@suse.cz>, kamezawa.hiroyu@jp.fujitsu.com, anton.vorontsov@linaro.org, Johannes Weiner <hannes@cmpxchg.org>, Mel Gorman <mgorman@suse.de> Subject: Re: [PATCH v2 3/5] memcg: make it suck faster Date: Wed, 20 Mar 2013 11:13:24 +0400 [thread overview] Message-ID: <51496194.7020508@parallels.com> (raw) In-Reply-To: <CAFj3OHU6f3o5GmbFyUsqtSWqHruSS4Yyodx=s=Vh8mO7GfTE8w@mail.gmail.com> On 03/13/2013 12:08 PM, Sha Zhengju wrote: >> +static void memcg_update_root_statistics(void) >> > +{ >> > + int cpu; >> > + u64 pgin, pgout, faults, mjfaults; >> > + >> > + pgin = pgout = faults = mjfaults = 0; >> > + for_each_online_cpu(cpu) { >> > + struct vm_event_state *ev = &per_cpu(vm_event_states, cpu); >> > + struct mem_cgroup_stat_cpu *memcg_stat; >> > + >> > + memcg_stat = per_cpu_ptr(root_mem_cgroup->stat, cpu); >> > + >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGPGIN] = >> > + ev->event[PGPGIN]; >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGPGOUT] = >> > + ev->event[PGPGOUT]; > ev->event[PGPGIN/PGPGOUT] is counted in block layer(submit_bio()) and > represents the exactly number of pagein/pageout, but memcg > PGPGIN/PGPGOUT events only count it as an event and ignore the page > size. So here we can't straightforward take the ev->events for use. > You are right about that. Although I can't think of a straightforward way to handle this. Well, except for the obvious of adding another global statistic. >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGFAULT] = >> > + ev->event[PGFAULT]; >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGMAJFAULT] = >> > + ev->event[PGMAJFAULT]; >> > + >> > + memcg_stat->nr_page_events = ev->event[PGPGIN] + >> > + ev->event[PGPGOUT]; > There's no valid memcg->nr_page_events until now, so the threshold > notifier, but some people may use it even only root memcg exists. > Moreover, using PGPGIN + PGPGOUT(exactly number of pagein + pageout) > as nr_page_events is also inaccurate IMHO. > Humm, I believe I can zero out this. Looking at the code again, this is not imported to userspace. It is just used to activate the thresholds and the delta of nr_page_events is a lot more important than nr_page_events. -- 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>
WARNING: multiple messages have this Message-ID (diff)
From: Glauber Costa <glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org> To: Sha Zhengju <handai.szj-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Cc: linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>, Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>, Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>, kamezawa.hiroyu-+CUm20s59erQFUHtdCDX3A@public.gmane.org, anton.vorontsov-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org, Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>, Mel Gorman <mgorman-l3A5Bk7waGM@public.gmane.org> Subject: Re: [PATCH v2 3/5] memcg: make it suck faster Date: Wed, 20 Mar 2013 11:13:24 +0400 [thread overview] Message-ID: <51496194.7020508@parallels.com> (raw) In-Reply-To: <CAFj3OHU6f3o5GmbFyUsqtSWqHruSS4Yyodx=s=Vh8mO7GfTE8w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> On 03/13/2013 12:08 PM, Sha Zhengju wrote: >> +static void memcg_update_root_statistics(void) >> > +{ >> > + int cpu; >> > + u64 pgin, pgout, faults, mjfaults; >> > + >> > + pgin = pgout = faults = mjfaults = 0; >> > + for_each_online_cpu(cpu) { >> > + struct vm_event_state *ev = &per_cpu(vm_event_states, cpu); >> > + struct mem_cgroup_stat_cpu *memcg_stat; >> > + >> > + memcg_stat = per_cpu_ptr(root_mem_cgroup->stat, cpu); >> > + >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGPGIN] = >> > + ev->event[PGPGIN]; >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGPGOUT] = >> > + ev->event[PGPGOUT]; > ev->event[PGPGIN/PGPGOUT] is counted in block layer(submit_bio()) and > represents the exactly number of pagein/pageout, but memcg > PGPGIN/PGPGOUT events only count it as an event and ignore the page > size. So here we can't straightforward take the ev->events for use. > You are right about that. Although I can't think of a straightforward way to handle this. Well, except for the obvious of adding another global statistic. >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGFAULT] = >> > + ev->event[PGFAULT]; >> > + memcg_stat->events[MEM_CGROUP_EVENTS_PGMAJFAULT] = >> > + ev->event[PGMAJFAULT]; >> > + >> > + memcg_stat->nr_page_events = ev->event[PGPGIN] + >> > + ev->event[PGPGOUT]; > There's no valid memcg->nr_page_events until now, so the threshold > notifier, but some people may use it even only root memcg exists. > Moreover, using PGPGIN + PGPGOUT(exactly number of pagein + pageout) > as nr_page_events is also inaccurate IMHO. > Humm, I believe I can zero out this. Looking at the code again, this is not imported to userspace. It is just used to activate the thresholds and the delta of nr_page_events is a lot more important than nr_page_events.
next prev parent reply other threads:[~2013-03-20 7:13 UTC|newest] Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top 2013-03-05 13:10 [PATCH v2 0/5] bypass root memcg charges if no memcgs are possible Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-05 13:10 ` [PATCH v2 1/5] memcg: make nocpu_base available for non hotplug Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-06 0:04 ` Kamezawa Hiroyuki 2013-03-06 0:04 ` Kamezawa Hiroyuki 2013-03-19 11:07 ` Michal Hocko 2013-03-05 13:10 ` [PATCH v2 2/5] memcg: provide root figures from system totals Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-06 0:27 ` Kamezawa Hiroyuki 2013-03-06 8:30 ` Glauber Costa 2013-03-06 8:30 ` Glauber Costa 2013-03-06 10:45 ` Kamezawa Hiroyuki 2013-03-06 10:45 ` Kamezawa Hiroyuki 2013-03-06 10:52 ` Glauber Costa 2013-03-06 10:52 ` Glauber Costa 2013-03-06 10:59 ` Kamezawa Hiroyuki 2013-03-06 10:59 ` Kamezawa Hiroyuki 2013-03-13 6:58 ` Sha Zhengju 2013-03-13 6:58 ` Sha Zhengju 2013-03-13 9:15 ` Kamezawa Hiroyuki 2013-03-13 9:15 ` Kamezawa Hiroyuki 2013-03-13 9:59 ` Sha Zhengju 2013-03-14 0:03 ` Kamezawa Hiroyuki 2013-03-14 0:03 ` Kamezawa Hiroyuki 2013-03-06 10:50 ` Kamezawa Hiroyuki 2013-03-06 10:50 ` Kamezawa Hiroyuki 2013-03-19 12:46 ` Michal Hocko 2013-03-19 12:46 ` Michal Hocko 2013-03-19 12:55 ` Michal Hocko 2013-03-19 12:55 ` Michal Hocko 2013-03-20 7:03 ` Glauber Costa 2013-03-20 8:03 ` Michal Hocko 2013-03-20 8:03 ` Michal Hocko 2013-03-20 8:08 ` Glauber Costa 2013-03-20 8:18 ` Michal Hocko 2013-03-20 8:18 ` Michal Hocko 2013-03-20 8:34 ` Glauber Costa 2013-03-20 8:58 ` Michal Hocko 2013-03-20 9:30 ` Glauber Costa 2013-03-21 6:08 ` Kamezawa Hiroyuki 2013-03-21 6:08 ` Kamezawa Hiroyuki 2013-03-20 16:40 ` Anton Vorontsov 2013-03-20 16:40 ` Anton Vorontsov 2013-03-20 7:04 ` Glauber Costa 2013-03-05 13:10 ` [PATCH v2 3/5] memcg: make it suck faster Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-06 0:46 ` Kamezawa Hiroyuki 2013-03-06 0:46 ` Kamezawa Hiroyuki 2013-03-06 8:38 ` Glauber Costa 2013-03-06 10:54 ` Kamezawa Hiroyuki 2013-03-06 10:54 ` Kamezawa Hiroyuki 2013-03-13 8:08 ` Sha Zhengju 2013-03-13 8:08 ` Sha Zhengju 2013-03-20 7:13 ` Glauber Costa [this message] 2013-03-20 7:13 ` Glauber Costa 2013-03-19 13:58 ` Michal Hocko 2013-03-19 13:58 ` Michal Hocko 2013-03-20 7:00 ` Glauber Costa 2013-03-20 8:13 ` Michal Hocko 2013-03-20 8:13 ` Michal Hocko 2013-03-05 13:10 ` [PATCH v2 4/5] memcg: do not call page_cgroup_init at system_boot Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-06 1:07 ` Kamezawa Hiroyuki 2013-03-06 1:07 ` Kamezawa Hiroyuki 2013-03-06 8:22 ` Glauber Costa 2013-03-06 8:22 ` Glauber Costa 2013-03-19 14:06 ` Michal Hocko 2013-03-05 13:10 ` [PATCH v2 5/5] memcg: do not walk all the way to the root for memcg Glauber Costa 2013-03-05 13:10 ` Glauber Costa 2013-03-06 1:08 ` Kamezawa Hiroyuki 2013-03-06 1:08 ` Kamezawa Hiroyuki
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=51496194.7020508@parallels.com \ --to=glommer@parallels.com \ --cc=akpm@linux-foundation.org \ --cc=anton.vorontsov@linaro.org \ --cc=cgroups@vger.kernel.org \ --cc=handai.szj@gmail.com \ --cc=hannes@cmpxchg.org \ --cc=kamezawa.hiroyu@jp.fujitsu.com \ --cc=linux-mm@kvack.org \ --cc=mgorman@suse.de \ --cc=mhocko@suse.cz \ --cc=tj@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.