From: Paul Menage <paul@paulmenage.org> To: Glauber Costa <glommer@parallels.com> Cc: Greg Thelen <gthelen@google.com>, linux-kernel@vger.kernel.org, linux-mm@kvack.org, containers@lists.osdl.org, netdev@vger.kernel.org, xemul@parallels.com, "David S. Miller" <davem@davemloft.net>, Hiroyouki Kamezawa <kamezawa.hiroyu@jp.fujitsu.com>, "Eric W. Biederman" <ebiederm@xmission.com>, Suleiman Souhlal <suleiman@google.com>, Lennart Poettering <lennart@poettering.net> Subject: Re: [PATCH] per-cgroup tcp buffer limitation Date: Tue, 13 Sep 2011 11:09:03 -0700 [thread overview] Message-ID: <CALdu-PC7ESSUHuF4vfVoRFFfkaBt1V28rGW3-O5pT3WtegAh4g@mail.gmail.com> (raw) In-Reply-To: <4E6E39DD.2040102@parallels.com> On Mon, Sep 12, 2011 at 9:57 AM, Glauber Costa <glommer@parallels.com> wrote: > > I think at this point there is at least consensus that this could very well > live in memcg, right ? Yes, it looks that way. >> This is definitely an improvement, but I'd say it's not enough. I >> think we should consider something like: > > One step at a time =) Yes, as far as design and initial implementation goes - but the full plan has to be figured out before anything gets committed to mainline, given the stability guarantees that implies. >> - the 'active' control determines whether (all) child cgroups will >> have memory.{limit,usage}_in_bytes files, or >> memory.{kernel,user}_{limit,usage}_in_bytes files >> - kernel memory will be charged either against 'kernel' or 'total' >> depending on the value of unified > > You mean for display/pressure purposes, right? Internally, I think once we > have kernel memory, we always charge it to kernel memory, regardless of > anything else. The value in unified field will only take place when we need > to grab this value. > > I don't personally see a reason for not having all files present at all > times. There's pretty much only one reason - avoiding the overhead of maintaining multiple counters. Each set of counters (user, kernel, total) will have its own locks, contention and other overheads to keep up to date. If userspace doesn't care about one or two of the three, then that's mostly wasted. Now it might be that the accounting of all three can be done with little more overhead than that required to update just a split view or just a unified view, in which case there's much less argument against simplifying and tracking/charging/limiting all three. > > It is overly flexible if we're exposing these counters and expecting the > user to do anything with them. It is perfectly fine if a single file, when > read, displays this information as statistics. > When I proposed this, I guess I was envisioning that most of the counters (e.g. things like TCP buffers or general network buffers) would be primarily for stats, since the admin probably only cares about total memory usage. The main point of this was to allow people who want to do something like tracking/limiting TCP buffer usage specifically per-cgroup to do so, without having any performance impact on the regular users. Paul -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Paul Menage <paul@paulmenage.org> To: Glauber Costa <glommer@parallels.com> Cc: Greg Thelen <gthelen@google.com>, linux-kernel@vger.kernel.org, linux-mm@kvack.org, containers@lists.osdl.org, netdev@vger.kernel.org, xemul@parallels.com, "David S. Miller" <davem@davemloft.net>, Hiroyouki Kamezawa <kamezawa.hiroyu@jp.fujitsu.com>, "Eric W. Biederman" <ebiederm@xmission.com>, Suleiman Souhlal <suleiman@google.com>, Lennart Poettering <lennart@poettering.net> Subject: Re: [PATCH] per-cgroup tcp buffer limitation Date: Tue, 13 Sep 2011 11:09:03 -0700 [thread overview] Message-ID: <CALdu-PC7ESSUHuF4vfVoRFFfkaBt1V28rGW3-O5pT3WtegAh4g@mail.gmail.com> (raw) In-Reply-To: <4E6E39DD.2040102@parallels.com> On Mon, Sep 12, 2011 at 9:57 AM, Glauber Costa <glommer@parallels.com> wrote: > > I think at this point there is at least consensus that this could very well > live in memcg, right ? Yes, it looks that way. >> This is definitely an improvement, but I'd say it's not enough. I >> think we should consider something like: > > One step at a time =) Yes, as far as design and initial implementation goes - but the full plan has to be figured out before anything gets committed to mainline, given the stability guarantees that implies. >> - the 'active' control determines whether (all) child cgroups will >> have memory.{limit,usage}_in_bytes files, or >> memory.{kernel,user}_{limit,usage}_in_bytes files >> - kernel memory will be charged either against 'kernel' or 'total' >> depending on the value of unified > > You mean for display/pressure purposes, right? Internally, I think once we > have kernel memory, we always charge it to kernel memory, regardless of > anything else. The value in unified field will only take place when we need > to grab this value. > > I don't personally see a reason for not having all files present at all > times. There's pretty much only one reason - avoiding the overhead of maintaining multiple counters. Each set of counters (user, kernel, total) will have its own locks, contention and other overheads to keep up to date. If userspace doesn't care about one or two of the three, then that's mostly wasted. Now it might be that the accounting of all three can be done with little more overhead than that required to update just a split view or just a unified view, in which case there's much less argument against simplifying and tracking/charging/limiting all three. > > It is overly flexible if we're exposing these counters and expecting the > user to do anything with them. It is perfectly fine if a single file, when > read, displays this information as statistics. > When I proposed this, I guess I was envisioning that most of the counters (e.g. things like TCP buffers or general network buffers) would be primarily for stats, since the admin probably only cares about total memory usage. The main point of this was to allow people who want to do something like tracking/limiting TCP buffer usage specifically per-cgroup to do so, without having any performance impact on the regular users. Paul
next prev parent reply other threads:[~2011-09-13 18:09 UTC|newest] Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top 2011-09-06 2:35 [PATCH] per-cgroup tcp buffer limitation Glauber Costa 2011-09-06 2:35 ` Glauber Costa 2011-09-06 10:00 ` KAMEZAWA Hiroyuki 2011-09-06 10:00 ` KAMEZAWA Hiroyuki 2011-09-06 15:28 ` Glauber Costa 2011-09-06 15:28 ` Glauber Costa 2011-09-06 15:28 ` Glauber Costa 2011-09-06 22:52 ` Glauber Costa 2011-09-06 22:52 ` Glauber Costa 2011-09-06 22:52 ` Glauber Costa 2011-09-06 16:08 ` Greg Thelen 2011-09-06 16:08 ` Greg Thelen 2011-09-06 16:16 ` Glauber Costa 2011-09-06 16:16 ` Glauber Costa 2011-09-06 16:16 ` Glauber Costa 2011-09-06 22:12 ` Greg Thelen 2011-09-06 22:12 ` Greg Thelen 2011-09-06 22:37 ` Glauber Costa 2011-09-06 22:37 ` Glauber Costa 2011-09-06 22:37 ` Glauber Costa 2011-09-07 21:35 ` Greg Thelen 2011-09-07 21:35 ` Greg Thelen 2011-09-08 4:44 ` Glauber Costa 2011-09-08 4:44 ` Glauber Costa 2011-09-08 4:44 ` Glauber Costa 2011-09-08 21:53 ` Greg Thelen 2011-09-08 21:53 ` Greg Thelen 2011-09-09 4:17 ` Glauber Costa 2011-09-09 4:17 ` Glauber Costa 2011-09-09 4:17 ` Glauber Costa 2011-09-09 23:38 ` Greg Thelen 2011-09-09 23:38 ` Greg Thelen 2011-09-12 16:30 ` Glauber Costa 2011-09-12 16:30 ` Glauber Costa 2011-09-12 16:30 ` Glauber Costa 2011-09-12 5:03 ` Paul Menage 2011-09-12 5:03 ` Paul Menage 2011-09-12 16:57 ` Glauber Costa 2011-09-12 16:57 ` Glauber Costa 2011-09-12 16:57 ` Glauber Costa 2011-09-13 6:56 ` Greg Thelen 2011-09-13 6:56 ` Greg Thelen 2011-09-13 18:09 ` Paul Menage [this message] 2011-09-13 18:09 ` Paul Menage 2011-09-13 18:11 ` Glauber Costa 2011-09-13 18:11 ` Glauber Costa 2011-09-13 18:11 ` Glauber Costa 2011-09-13 18:46 ` Paul Menage 2011-09-13 18:46 ` Paul Menage 2011-09-13 19:08 ` Glauber Costa 2011-09-13 19:08 ` Glauber Costa 2011-09-13 19:08 ` Glauber Costa 2011-09-07 2:38 ` Eric Dumazet 2011-09-07 2:38 ` Eric Dumazet 2011-09-07 2:38 ` Eric Dumazet 2011-09-07 1:08 ` Paul Menage 2011-09-07 1:08 ` Paul Menage 2011-09-07 1:09 ` Glauber Costa 2011-09-07 1:09 ` Glauber Costa 2011-09-07 1:09 ` Glauber Costa 2011-09-07 1:29 ` Paul Menage 2011-09-07 1:29 ` Paul Menage 2011-09-07 1:32 ` Glauber Costa 2011-09-07 1:32 ` Glauber Costa 2011-09-07 1:32 ` Glauber Costa 2011-09-09 0:18 ` Rick Jones 2011-09-09 0:18 ` Rick Jones
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=CALdu-PC7ESSUHuF4vfVoRFFfkaBt1V28rGW3-O5pT3WtegAh4g@mail.gmail.com \ --to=paul@paulmenage.org \ --cc=containers@lists.osdl.org \ --cc=davem@davemloft.net \ --cc=ebiederm@xmission.com \ --cc=glommer@parallels.com \ --cc=gthelen@google.com \ --cc=kamezawa.hiroyu@jp.fujitsu.com \ --cc=lennart@poettering.net \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=netdev@vger.kernel.org \ --cc=suleiman@google.com \ --cc=xemul@parallels.com \ /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.