From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vivek Goyal Subject: Re: [PATCH v6 0/9] memcg: per cgroup dirty page accounting Date: Tue, 15 Mar 2011 17:23:39 -0400 Message-ID: <20110315212339.GC5740@redhat.com> References: <1299869011-26152-1-git-send-email-gthelen@google.com> <20110311171006.ec0d9c37.akpm@linux-foundation.org> <20110314202324.GG31120@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Cc: 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: Greg Thelen Return-path: Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Mar 14, 2011 at 07:41:13PM -0700, Greg Thelen wrote: > On Mon, Mar 14, 2011 at 1:23 PM, Vivek Goyal wrote: > > On Mon, Mar 14, 2011 at 11:29:17AM -0700, Greg Thelen wrote: > > > > [..] > >> > We could just crawl the memcg's page LRU and bring things under co= ntrol > >> > that way, couldn't we? =A0That would fix it. =A0What were the reas= ons for > >> > not doing this? > >> > >> My rational for pursuing bdi writeback was I/O locality. =A0I have h= eard that > >> per-page I/O has bad locality. =A0Per inode bdi-style writeback shou= ld have better > >> locality. > >> > >> My hunch is the best solution is a hybrid which uses a) bdi writebac= k with a > >> target memcg filter and b) using the memcg lru as a fallback to iden= tify the bdi > >> that needed writeback. =A0I think the part a) memcg filtering is lik= ely something > >> like: > >> =A0http://marc.info/?l=3Dlinux-kernel&m=3D129910424431837 > >> > >> The part b) bdi selection should not be too hard assuming that page-= to-mapping > >> locking is doable. > > > > Greg, > > > > IIUC, option b) seems to be going through pages of particular memcg a= nd > > mapping page to inode and start writeback on particular inode? >=20 > Yes. >=20 > > If yes, this might be reasonably good. In the case when cgroups are n= ot > > sharing inodes then it automatically maps one inode to one cgroup and > > once cgroup is over limit, it starts writebacks of its own inode. > > > > In case inode is shared, then we get the case of one cgroup writting > > back the pages of other cgroup. Well I guess that also can be handele= d > > by flusher thread where a bunch or group of pages can be compared wit= h > > the cgroup passed in writeback structure. I guess that might hurt us > > more than benefit us. >=20 > Agreed. For now just writing the entire inode is probably fine. >=20 > > IIUC how option b) works then we don't even need option a) where an N= level > > deep cache is maintained? >=20 > Originally I was thinking that bdi-wide writeback with memcg filter > was a good idea. But this may be unnecessarily complex. Now I am > agreeing with you that option (a) may not be needed. Memcg could > queue per-inode writeback using the memcg lru to locate inodes > (lru->page->inode) with something like this in > [mem_cgroup_]balance_dirty_pages(): >=20 > while (memcg_usage() >=3D memcg_fg_limit) { > inode =3D memcg_dirty_inode(cg); /* scan lru for a dirty page, the= n > grab mapping & inode */ > sync_inode(inode, &wbc); > } Is it possible to pass mem_cgroup in writeback_control structure or in work structure which in turn will be set in writeback_control. And modify writeback_inodes_wb() which will look that ->mem_cgroup is set. So instead of calling queue_io() it can call memcg_queue_io() and then memory cgroup can look at lru list and take its own decision on which inodes needs to be pushed for IO? Thanks Vivek -- 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: email@kvack.org