From: Alexander Beregalov Subject: Re: next-20090206: deadlock on ext4 Date: Mon, 9 Feb 2009 15:01:13 +0300 Message-ID: References: <498C7671.1050309@redhat.com> <498C9876.3060208@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: "linux-next@vger.kernel.org" , linux-ext4@vger.kernel.org, LKML To: Eric Sandeen Return-path: In-Reply-To: <498C9876.3060208@redhat.com> Sender: linux-next-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org 2009/2/6 Eric Sandeen : > Alexander Beregalov wrote: >> 2009/2/6 Eric Sandeen : >>> Alexander Beregalov wrote: >>>> Hi >>>> >>>> I run dbench on ext4 on loop device. >>>> >>>> EXT4-fs: barriers enabled >>>> kjournald2 starting: pid 2420, dev loop0:8, commit interval 5 seconds >>>> EXT4 FS on loop0, internal journal on loop0:8 >>>> EXT4-fs: delayed allocation enabled >>>> EXT4-fs: file extents enabled >>>> EXT4-fs: mballoc enabled >>>> EXT4-fs: mounted filesystem loop0 with ordered data mode >>>> JBD: barrier-based sync failed on loop0:8 - disabling barriers >>>> >>>> INFO: task pdflush:2339 blocked for more than 120 seconds. >>> Looks a lot like a bug I filed: >>> >>> http://bugzilla.kernel.org/show_bug.cgi?id=12579 >>> >>> but I'm having trouble reproducing it, now. I'll try dbench! >> Yes, I can reproduce it easy with dbench. > > Do you only hit it on loopback? What is the filesystem hosting the loop > file? I have reproduced it with 2.6.29-rc4. Loop file was on XFS. I can not reproduce it on ext4 on raw device. Let me know if I can do anything else to help resolving it.