Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S967215AbXEGXai (ORCPT ); Mon, 7 May 2007 19:30:38 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S967175AbXEGXaf (ORCPT ); Mon, 7 May 2007 19:30:35 -0400 Received: from proxima.lp0.eu ([85.158.45.36]:43935 "EHLO proxima.lp0.eu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S967133AbXEGXaf (ORCPT ); Mon, 7 May 2007 19:30:35 -0400 Message-ID: <463FB697.6060902@simon.arlott.org.uk> Date: Tue, 08 May 2007 00:30:31 +0100 From: Simon Arlott User-Agent: Thunderbird 1.5.0.5 (X11/20060819) MIME-Version: 1.0 To: Andrew Morton CC: linux-kernel@vger.kernel.org Subject: Re: sleeping function called from invalid context at block/cfq-iosched.c (Was: Re: 2.6.21-mm1) References: <20070505014955.8f3990b5.akpm@linux-foundation.org> <463E3C50.9080207@simon.arlott.org.uk> <20070506135403.a6785f7d.akpm@linux-foundation.org> <463FAF1C.8060500@simon.arlott.org.uk> <20070507162349.0790706f.akpm@linux-foundation.org> In-Reply-To: <20070507162349.0790706f.akpm@linux-foundation.org> X-Enigmail-Version: 0.94.1.2 OpenPGP: id=89C93563 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 8683 Lines: 157 On 08/05/07 00:23, Andrew Morton wrote: > On Mon, 07 May 2007 23:58:36 +0100 > Simon Arlott wrote: > >> I've just got this under mainline too now (0ec54aa8af5e6faa346aa55a1ad15ee6c25bb42d 2007-05-05 22:08:06): >> >> [84567.084000] BUG: scheduling while atomic: pdflush/0xeff84acf/186 >> [84567.084000] INFO: lockdep is turned off. >> [84567.084000] do_IRQ: stack overflow: 440 >> [84567.084000] [] show_trace_log_lvl+0x1a/0x30 >> [84567.084000] [] show_trace+0x12/0x20 >> [84567.084000] [] dump_stack+0x15/0x20 >> [84567.084000] [] do_IRQ+0xd8/0xe0 >> [84567.084000] [] common_interrupt+0x2e/0x34 >> [84567.084000] [] printk+0x18/0x20 >> [84567.084000] [] debug_show_held_locks+0x23/0x30 >> [84567.084000] [] schedule+0x52a/0x680 >> [84567.084000] [] schedule_timeout+0x8a/0x90 >> [84567.084000] [] xlog_state_sync+0x21e/0x260 >> [84567.084000] [] _xfs_log_force+0x44/0x70 >> [84567.084000] [] xfs_alloc_search_busy+0xd2/0xe0 >> [84567.084000] [] xfs_alloc_get_freelist+0xe6/0x100 >> [84567.084000] [] xfs_alloc_split+0x1f/0x4c0 >> [84567.084000] [] xfs_alloc_insrec+0x35d/0x3b0 >> [84567.084000] [] xfs_alloc_insert+0x59/0xd0 >> [84567.084000] [] xfs_free_ag_extent+0x297/0x640 >> [84567.084000] [] xfs_alloc_fix_freelist+0x1d1/0x4c0 >> [84567.084000] [] xfs_alloc_vextent+0x1eb/0x4b0 >> [84567.084000] [] xfs_bmap_btalloc+0x3bb/0x8f0 >> [84567.084000] [] xfs_bmap_alloc+0x21/0x40 >> [84567.084000] [] xfs_bmapi+0xdbf/0x1450 >> [84567.084000] [] xfs_iomap_write_allocate+0x2ba/0x520 >> [84567.084000] [] xfs_iomap+0x45b/0x470 >> [84567.084000] [] xfs_bmap+0x2c/0x40 >> [84567.084000] [] xfs_map_blocks+0x3c/0x80 >> [84567.084000] [] xfs_page_state_convert+0x2f3/0x670 >> [84567.084000] [] xfs_vm_writepage+0x5a/0xf0 >> [84567.084000] [] generic_writepages+0x226/0x360 >> [84567.084000] [] xfs_vm_writepages+0x24/0x60 >> [84567.084000] [] do_writepages+0x2c/0x50 >> [84567.084000] [] __sync_single_inode+0x59/0x1f0 >> [84567.084000] [] __writeback_single_inode+0x44/0x1d0 >> [84567.084000] [] sync_sb_inodes+0x1c9/0x2e0 >> [84567.084000] [] writeback_inodes+0xce/0xe0 >> [84567.084000] [] wb_kupdate+0x73/0xf0 >> [84567.084000] [] __pdflush+0xce/0x1a0 >> [84567.084000] [] pdflush+0x25/0x30 >> [84567.084000] [] kthread+0x94/0xd0 >> [84567.084000] [] kernel_thread_helper+0x7/0x1c >> [84567.084000] ======================= >> [84567.087000] [] show_trace_log_lvl+0x1a/0x30 >> [84567.087000] [] do_IRQ: stack overflow: 328 >> [84567.087000] [] show_trace_log_lvl+0x1a/0x30 >> [84567.087000] [] show_trace+0x12/0x20 >> [84567.087000] [] dump_stack+0x15/0x20 >> [84567.087000] [] do_IRQ+0xd8/0xe0 >> [84567.087000] [] common_interrupt+0x2e/0x34 >> [84567.087000] [] printk+0x18/0x20 >> [84567.087000] [] print_trace_address+0x1d/0x30 >> [84567.087000] [] dump_trace+0x48/0xb0 >> [84567.087000] [] show_trace_log_lvl+0x1a/0x30 >> [84567.087000] [] show_trace+0x12/0x20 >> [84567.087000] [] dump_stack+0x15/0x20 >> [84567.087000] [] schedule+0x53a/0x680 >> [84567.087000] [] schedule_timeout+0x8a/0x90 >> [84567.087000] [] xlog_state_sync+0x21e/0x260 >> [84567.087000] [] _xfs_log_force+0x44/0x70 >> [84567.087000] [] xfs_alloc_search_busy+0xd2/0xe0 >> [84567.087000] [] xfs_alloc_get_freelist+0xe6/0x100 >> [84567.087000] [] xfs_alloc_split+0x1f/0x4c0 >> [84567.087000] [] xfs_alloc_insrec+0x35d/0x3b0 >> [84567.087000] [] xfs_alloc_insert+0x59/0xd0 >> [84567.087000] [] xfs_free_ag_extent+0x297/0x640 >> [84567.087000] [] xfs_alloc_fix_freelist+0x1d1/0x4c0 >> [84567.087000] [] xfs_alloc_vextent+0x1eb/0x4b0 >> [84567.087000] [] xfs_bmap_btalloc+0x3bb/0x8f0 >> [84567.087000] [] xfs_bmap_alloc+0x21/0x40 >> [84567.087000] [] xfs_bmapi+0xdbf/0x1450 >> [84567.087000] [] xfs_iomap_write_allocate+0x2ba/0x520 >> [84567.087000] [] xfs_iomap+0x45b/0x470 >> [84567.087000] [] xfs_bmap+0x2c/0x40 >> [84567.087000] [] xfs_map_blocks+0x3c/0x80 >> [84567.087000] [] xfs_page_state_convert+0x2f3/0x670 >> [84567.087000] [] xfs_vm_writepage+0x5a/0xf0 >> [84567.087000] [] generic_writepages+0x226/0x360 >> [84567.087000] [] xfs_vm_writepages+0x24/0x60 >> [84567.087000] [] do_writepages+0x2c/0x50 >> [84567.087000] [] __sync_single_inode+0x59/0x1f0 >> [84567.087000] [] __writeback_single_inode+0x44/0x1d0 >> [84567.087000] [] sync_sb_inodes+0x1c9/0x2e0 >> [84567.087000] [] writeback_inodes+0xce/0xe0 >> [84567.087000] [] wb_kupdate+0x73/0xf0 >> [84567.087000] [] __pdflush+0xce/0x1a0 >> [84567.087000] [] pdflush+0x25/0x30 >> [84567.087000] [] kthread+0x94/0xd0 >> [84567.087000] [] kernel_thread_helper+0x7/0x1c >> [84567.087000] BUG: unable to handle kernel paging request at virtual address 00008034 >> [84567.087000] printing eip: >> [84567.087000] b0104e36 >> [84567.087000] *pde = 00000000 >> [84567.087000] Oops: 0000 [#1] >> [84567.087000] PREEMPT >> [84567.087000] Modules linked in: drbd mt352 saa7134_dvb dvb_pll video_buf_dvb dvb_core >> [84567.087000] CPU: 0 >> [84567.087000] EIP: 0060:[] Not tainted VLI >> [84567.087000] EFLAGS: 00010046 (2.6.21-git #197) >> [84567.087000] EIP is at dump_trace+0x66/0xb0 >> [84567.087000] eax: 00000000 ebx: b18f6fe0 ecx: b06323ec edx: b05afc71 >> [84567.087000] esi: 00008000 edi: 00008ffd ebp: b18f6138 esp: b18f6120 >> [84567.087000] ds: 007b es: 007b fs: 0000 gs: 0000 ss: 0068 >> [84567.087000] Process pdflush (pid: 186, ti=b18f5000 task=eff84ad0 task.ti=b18f6000) >> [84567.087000] Stack: b18f6130 b04ea9af 00000001 b05b9937 0000000a b06d656e b18f6158 b0104f3a >> [84567.087000] b06323ec b05b9937 00000000 b06356a8 b18f6160 b18f622a b18f6168 b0104f62 >> [84567.087000] b05b9937 b06d656e b18f6178 b0105075 b05afeac b18f6184 b18f61a8 b01069a8 >> [84567.087000] Call Trace: >> [84567.087000] BUG: unable to handle kernel NULL pointer dereference at virtual address 00000034 >> [84567.087000] printing eip: >> [84567.087000] b0104e36 >> [84567.087000] *pde = 00000000 >> [84567.087000] Oops: 0000 [#2] >> [84567.087000] PREEMPT >> [84567.087000] Modules linked in: drbd mt352 saa7134_dvb dvb_pll video_buf_dvb dvb_core >> [84567.087000] CPU: 0 >> [84567.087000] EIP: 0060:[] Not tainted VLI >> [84567.087000] EFLAGS: 00010046 (2.6.21-git #197) >> [84567.087000] EIP is at dump_trace+0x66/0xb0 >> [84567.087000] eax: 00000000 ebx: b18f5f60 ecx: b06323ec edx: b05afc71 >> [84567.087000] esi: 00000000 edi: 00000ffd ebp: b18f5f60 esp: b18f5f48 >> [84567.087000] ds: 007b es: 007b fs: 0000 gs: 0000 ss: 0068 >> [84567.087000] Process pdflush (pid: 186, ti=b18f5000 task=eff84ad0 task.ti=b18f6000) >> [84567.087000] Stack: 5b3e303c 36353438 38302e37 b05afcde 00000018 00000000 b18f5f80 b0104f3a >> [84567.087000] b06323ec b05afcde >> >> Then it just gave up doing anything sane, presumably because of the stack overflow in do_IRQ: >> http://redrum.lp0.eu/tmp/100_4037.jpg >> http://redrum.lp0.eu/tmp/100_4038.jpg >> > > ugly. And you're using 8k stacks? 4k stacks, I will try to reproduce it with 8k stacks - which shouldn't be hard (see below). > The way we'll fix this is to wait for someone to do a git-bisect on it. > Perhaps you can do that? It will take a while, I need to boot to a good kernel so unattended DVB-T recording works reliably, I'm assuming switching to a different IO scheduler will avoid any problems for that (and not using the XFS filesystem...). It happened again not 5 minutes later when I continued compiling the kernel after sending that email so reproducing it should be easy. -- Simon Arlott - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/