From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754546AbZDMDRm (ORCPT ); Sun, 12 Apr 2009 23:17:42 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753619AbZDMDRa (ORCPT ); Sun, 12 Apr 2009 23:17:30 -0400 Received: from mga09.intel.com ([134.134.136.24]:11642 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753404AbZDMDR3 (ORCPT ); Sun, 12 Apr 2009 23:17:29 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.40,178,1239001200"; d="scan'208";a="402549269" Subject: Re: [PATCH 0/13] Per-bdi writeback flusher threads #3 From: "Zhang, Yanmin" To: Jens Axboe Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, chris.mason@oracle.com, david@fromorbit.com, hch@infradead.org, akpm@linux-foundation.org, jack@suse.cz In-Reply-To: <20090410072109.GK5178@kernel.dk> References: <1239192016-19857-1-git-send-email-jens.axboe@oracle.com> <1239335199.2567.784.camel@ymzhang> <20090410072109.GK5178@kernel.dk> Content-Type: text/plain; charset=UTF-8 Date: Mon, 13 Apr 2009 11:18:03 +0800 Message-Id: <1239592683.2567.796.camel@ymzhang> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 (2.22.1-2.fc9) Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2009-04-10 at 09:21 +0200, Jens Axboe wrote: > On Fri, Apr 10 2009, Zhang, Yanmin wrote: > > On Wed, 2009-04-08 at 14:00 +0200, Jens Axboe wrote: > > > Hi, > > > > > > This is the third version of the patchset. Changes since v2: > > Jens, > > > > The patchset forgets kernel/sysctl.c part, so the compilation link fails. > > Yep, noticed that as well. Not really great timing with adding more > pdflush tunables... When I run a fio testing with a partition on one of my machines, I hit below bug checking. It's a  fio_mmap_randwrite_4k_preread testing. I'm not sure if it happens after the testing and before the next testing, or just happens when the testing is ongoing. When I see it, next testing stops at the beginning when mounting the partition. ----------------------sys log-------------------- kernel BUG at fs/jbd/journal.c:1166! invalid opcode: 0000 [#1] SMP last sysfs file: /sys/block/sdb/stat CPU 0 Modules linked in: igb Pid: 18329, comm: umount Not tainted 2.6.30-rc1-bdiflusherv3 #1 X8DTN RIP: 0010:[] [] journal_destroy+0x108/0x1b4 RSP: 0018:ffff8800bd95fe08 EFLAGS: 00010286 RAX: 000000000000c7c7 RBX: ffff8800bc93dc00 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000246 RDI: ffff8800bc93dd64 RBP: ffff8800bc93dd64 R08: ffff8800bc93dc88 R09: 00000000bd95fcc0 R10: 0000000000000246 R11: 00000000fffffffb R12: ffff8800bd95fe08 R13: ffff8800bd95fe20 R14: ffff8800bc93dc98 R15: ffff8800bc93dcc8 FS: 00007f3f26c717d0(0000) GS:ffffc20000000000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 00007f3f263fe000 CR3: 00000000bc18e000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process umount (pid: 18329, threadinfo ffff8800bd95e000, task ffff8800bcdcdca0) Stack: 0000000000000000 ffff8800bcdcdca0 ffffffff8024ddaa ffff8800bd95fe20 ffff8800bd95fe20 ffffffff802e53ac ffff8800bd95fe38 ffff8800bd0e3a00 ffff8800a1ff2400 ffff8800bc93b000 ffff8800bb8cb400 0000000000000000 Call Trace: [] ? autoremove_wake_function+0x0/0x2e [] ? mb_cache_shrink+0x22/0x9e [] ? ext3_put_super+0x29/0x204 [] ? generic_shutdown_super+0x70/0xde [] ? kill_block_super+0x22/0x3a [] ? deactivate_super+0x5f/0x77 [] ? sys_umount+0x2da/0x309 [] ? system_call_fastpath+0x16/0x1b Code: ff 48 8d ab 64 01 00 00 eb 0b fe 45 00 48 89 df e8 cc e8 ff ff 48 89 ef e8 2f e8 3d 00 48 83 7b 60 00 75 e6 48 83 7b 50 00 74 04 <0f> 0b eb fe 48 83 7b 58 00 74 04 0f 0b eb fe fe 83 64 01 00 00 RIP [] journal_destroy+0x108/0x1b4 RSP ---[ end trace 14f5ccbafee46eb4 ]--- > > There's also an issue with the SRCU init, so it will not even work > irregardless of the compilation problem. I'll have to update and repost > next week. >