From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg Thelen Subject: Re: [PATCH v6 8/9] memcg: check memcg dirty limits in page writeback Date: Mon, 14 Mar 2011 20:27:33 -0700 Message-ID: References: <1299869011-26152-1-git-send-email-gthelen@google.com> <1299869011-26152-9-git-send-email-gthelen@google.com> <20110314175408.GE31120@redhat.com> <20110314211002.GD4998@quack.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Vivek Goyal , Andrew Morton , linux-kernel@vger.kernel.org, linux-mm@kvack.org, containers@lists.osdl.org, linux-fsdevel@vger.kernel.org, Andrea Righi , Balbir Singh , KAMEZAWA Hiroyuki , Daisuke Nishimura , Minchan Kim , Johannes Weiner , Ciju Rajan K , David Rientjes , Wu Fengguang , Chad Talbott , Justin TerAvest To: Jan Kara Return-path: In-Reply-To: <20110314211002.GD4998@quack.suse.cz> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Mar 14, 2011 at 2:10 PM, Jan Kara wrote: > On Mon 14-03-11 13:54:08, Vivek Goyal wrote: >> On Fri, Mar 11, 2011 at 10:43:30AM -0800, Greg Thelen wrote: >> > If the current process is in a non-root memcg, then >> > balance_dirty_pages() will consider the memcg dirty limits as well= as >> > the system-wide limits. =A0This allows different cgroups to have d= istinct >> > dirty limits which trigger direct and background writeback at diff= erent >> > levels. >> > >> > If called with a mem_cgroup, then throttle_vm_writeout() queries t= he >> > given cgroup for its dirty memory usage limits. >> > >> > Signed-off-by: Andrea Righi >> > Signed-off-by: Greg Thelen >> > Acked-by: KAMEZAWA Hiroyuki >> > Acked-by: Wu Fengguang >> > --- >> > Changelog since v5: >> > - Simplified this change by using mem_cgroup_balance_dirty_pages()= rather than >> > =A0 cramming the somewhat different logic into balance_dirty_pages= (). =A0This means >> > =A0 the global (non-memcg) dirty limits are not passed around in t= he >> > =A0 struct dirty_info, so there's less change to existing code. >> >> Yes there is less change to existing code but now we also have a sep= arate >> throttlig logic for cgroups. >> >> I thought that we are moving in the direction of IO less throttling >> where bdi threads always do the IO and Jan Kara also implemented the >> logic to distribute the finished IO pages uniformly across the waiti= ng >> threads. > =A0Yes, we'd like to avoid doing IO from balance_dirty_pages(). But i= f the > logic in cgroups specific part won't get too fancy (which it doesn't = seem > to be the case currently), it shouldn't be too hard to convert it to = the new > approach. Handling memcg hierarchy was something that was not trivial to implemen= t in mem_cgroup_balance_dirty_pages. > We can talk about it at LSF but at least with my approach to IO-less > balance_dirty_pages() it would be easy to convert cgroups throttling = to > the new way. With Fengguang's approach it might be a bit harder since= he > computes a throughput and from that necessary delay for a throttled t= ask > but with cgroups that is impossible to compute so he'd have to add so= me > looping if we didn't write enough pages from the cgroup yet. But stil= l it > would be reasonable doable AFAICT. I am definitely interested in finding a way to merge these feature cleanly together. >> Keeping it separate for cgroups, reduces the complexity but also for= ks >> off the balancing logic for root and other cgroups. So if Jan Kara's >> changes go in, it automatically does not get used for memory cgroups= =2E >> >> Not sure how good a idea it is to use a separate throttling logic fo= r >> for non-root cgroups. > =A0Yeah, it looks a bit odd. I'd think that we could just cap > task_dirty_limit() by a value computed from a cgroup limit and be don= e > with that but I probably miss something... That is an interesting idea. When looking at upstream balance_dirty_pa= ges(), the result of task_dirty_limit() is compared per bdi_nr_reclaimable and bdi_nr_writeback. I think we should be comparing memcg usage to memcg = limits to catch cases where memcg usage is above memcg limits. Or am I missing something in your apporach? > Sure there is also a different > background limit but that's broken anyway because a flusher thread wi= ll > quickly stop doing writeback if global background limit is not exceed= ed. > But that's a separate topic so I'll reply with this to a more appropr= iate > email ;) ;) I am also interested in the this bg issue, but I should also try to stay on topic. > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Honza > -- > Jan Kara > SUSE Labs, CR >