From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnaldo Carvalho de Melo Subject: Re: BUG: Bad page map in process udevd (anon_vma: (null)) in 2.6.38-rc4 Date: Fri, 18 Feb 2011 17:01:28 -0200 Message-ID: <20110218190128.GF13211@ghostprotocols.net> References: <20110217090910.GA3781@tiehlicka.suse.cz> <20110217163531.GF14168@elte.hu> <20110218122938.GB26779@tiehlicka.suse.cz> <20110218162623.GD4862@tiehlicka.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Cc: "Eric W. Biederman" , Michal Hocko , Ingo Molnar , linux-mm@kvack.org, LKML , David Miller , Eric Dumazet , netdev@vger.kernel.org To: Linus Torvalds Return-path: Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org List-Id: netdev.vger.kernel.org Em Fri, Feb 18, 2011 at 10:48:18AM -0800, Linus Torvalds escreveu: > On Fri, Feb 18, 2011 at 10:08 AM, Eric W. Biederman > wrote: > > > > I am still getting programs segfaulting but that is happening on othe= r > > machines running on older kernels so I am going to chalk that up to a > > buggy test and a false positive. >=20 > Ok. >=20 > > I am have OOM problems getting my tests run to complete. =A0On a good > > day that happens about 1 time in 3 right now. =A0I'm guess I will hav= e > > to turn off DEBUG_PAGEALLOC to get everything to complete. > > DEBUG_PAGEALLOC causes us to use more memory doesn't it? >=20 > It does use a bit more memory, but it shouldn't be _that_ noticeable. > The real cost of DEBUG_PAGEALLOC is all the crazy page table > operations and TLB flushes we do for each allocation/deallocation. So > DEBUG_PAGEALLOC is very CPU-intensive, but it shouldn't have _that_ > much of a memory overhead - just some trivial overhead due to not > being able to use largepages for the normal kernel identity mappings. >=20 > But there might be some other interaction with OOM that I haven't thoug= ht about. >=20 > > The most interesting thing I have right now is a networking lockdep > > issue. =A0Does anyone know what is going on there? >=20 > This seems to be a fairly straightforward bug. >=20 > In net/ipv4/inet_timewait_sock.c we have this: >=20 > /* These are always called from BH context. See callers in > * tcp_input.c to verify this. > */ >=20 > /* This is for handling early-kills of TIME_WAIT sockets. */ > void inet_twsk_deschedule(struct inet_timewait_sock *tw, > struct inet_timewait_death_row *twdr) > { > spin_lock(&twdr->death_lock); > .. >=20 > and the intention is clearly that that spin_lock is BH-safe because > it's called from BH context. >=20 > Except that clearly isn't true. It's called from a worker thread: >=20 > > stack backtrace: > > Pid: 10833, comm: kworker/u:1 Not tainted 2.6.38-rc4-359399.2010Arora= KernelBeta.fc14.x86_64 #1 > > Call Trace: > > =A0[] ? inet_twsk_deschedule+0x29/0xa0 > > =A0[] ? inet_twsk_purge+0xf6/0x180 > > =A0[] ? inet_twsk_purge+0x30/0x180 > > =A0[] ? tcp_sk_exit_batch+0x1c/0x20 > > =A0[] ? ops_exit_list.clone.0+0x53/0x60 > > =A0[] ? cleanup_net+0x100/0x1b0 > > =A0[] ? process_one_work+0x187/0x4b0 > > =A0[] ? process_one_work+0x121/0x4b0 > > =A0[] ? cleanup_net+0x0/0x1b0 > > =A0[] ? worker_thread+0x15c/0x330 >=20 > so it can deadlock with a BH happening at the same time, afaik. >=20 > The code (and comment) is all from 2005, it looks like the BH->worker > thread has broken the code. But somebody who knows that code better > should take a deeper look at it. >=20 > Added acme to the cc, since the code is attributed to him back in 2005 > ;). Although I don't know how active he's been in networking lately > (seems to be all perf-related). Whatever, it can't hurt. Original code is ANK's, I just made it possible to use with DCCP, and yeah, the smiley is appropriate, something 6 years old and the world around it changing continually... well, thanks for the git blame ;-) - Arnaldo -- 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