From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Paul E. McKenney" Subject: Re: 2.6.39-rc4+: Kernel leaking memory during FS scanning, regression? Date: Mon, 25 Apr 2011 10:29:14 -0700 Message-ID: <20110425172914.GB2468@linux.vnet.ibm.com> References: <20110424202158.45578f31@neptune.home> <20110424235928.71af51e0@neptune.home> <20110425114429.266A.A69D9226@jp.fujitsu.com> <20110425111705.786ef0c5@neptune.home> <20110425180450.1ede0845@neptune.home> <20110425190032.7904c95d@neptune.home> Reply-To: paulmck@linux.vnet.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Cc: Linus Torvalds , Mike Frysinger , KOSAKI Motohiro , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, "Paul E. McKenney" , Pekka Enberg To: Bruno =?iso-8859-1?Q?Pr=E9mont?= Return-path: Content-Disposition: inline In-Reply-To: <20110425190032.7904c95d@neptune.home> Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Mon, Apr 25, 2011 at 07:00:32PM +0200, Bruno Pr=E9mont wrote: > On Mon, 25 April 2011 Linus Torvalds wrote: > > 2011/4/25 Bruno Pr=E9mont : > > > > > > kmemleak reports 86681 new leaks between shortly after boot and -2 = state. > > > (and 2348 additional ones between -2 and -4). > >=20 > > I wouldn't necessarily trust kmemleak with the whole RCU-freeing > > thing. In your slubinfo reports, the kmemleak data itself also tends > > to overwhelm everything else - none of it looks unreasonable per se. > >=20 > > That said, you clearly have a *lot* of filp entries. I wouldn't > > consider it unreasonable, though, because depending on load those may > > well be fine. Perhaps you really do have some application(s) that hol= d > > thousands of files open. The default file limit is 1024 (I think), bu= t > > you can raise it, and some programs do end up opening tens of > > thousands of files for filesystem scanning purposes. > >=20 > > That said, I would suggest simply trying a saner kernel configuration= , > > and seeing if that makes a difference: > >=20 > > > Yes, it's uni-processor system, so SMP=3Dn. > > > TINY_RCU=3Dy, PREEMPT_VOLUNTARY=3Dy (whole /proc/config.gz attached= keeping > > > compression) > >=20 > > I'm not at all certain that TINY_RCU is appropriate for > > general-purpose loads. I'd call it more of a "embedded low-performanc= e > > option". >=20 > Well, TINY_RCU is the only option when doing PREEMPT_VOLUNTARY on > SMP=3Dn... You can either set SMP=3Dy and NR_CPUS=3D1 or you can handed-edit init/Kconfig to remove the dependency on SMP. Just change the depends on !PREEMPT && SMP to: depends on !PREEMPT This will work fine, especially for experimental purposes. > > The _real_ RCU implementation ("tree rcu") forces quiescent states > > every few jiffies and has logic to handle "I've got tons of RCU > > events, I really need to start handling them now". All of which I > > think tiny-rcu lacks. >=20 > Going to try it out (will take some time to compile), kmemleak disabled= . >=20 > > So right now I suspect that you have a situation where you just have = a > > simple load that just ends up never triggering any RCU cleanup, and > > the tiny-rcu thing just keeps on gathering events and delays freeing > > stuff almost arbitrarily long. >=20 > I hope tiny-rcu is not that broken... as it would mean driving any > PREEMPT_NONE or PREEMPT_VOLUNTARY system out of memory when compiling > packages (and probably also just unpacking larger tarballs or running > things like du). If it is broken, I will fix it. ;-) Thanx, Paul > And with system doing nothing (except monitoring itself) memory usage > goes increasing all the time until it starves (well it seems to keep > ~20M free, pushing processes it can to swap). Config is just being > make oldconfig from working 2.6.38 kernel (answering default for new > options) >=20 > Memory usage evolution graph in first message of this thread: > http://thread.gmane.org/gmane.linux.kernel.mm/61909/focus=3D1130480 >=20 > Attached graph matching numbers of previous mail. (dropping caches was = at > 17:55, system idle since then) >=20 > Bruno >=20 >=20 > > So try CONFIG_PREEMPT and CONFIG_TREE_PREEMPT_RCU to see if the > > behavior goes away. That would confirm the "it's just tinyrcu being > > too dang stupid" hypothesis. > >=20 > > Linus -- 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