From: bugzilla-daemon@bugzilla.kernel.org Subject: [Bug 15906] performance regression in "umount" of filesystems using barriers Date: Thu, 6 May 2010 03:49:03 GMT Message-ID: <201005060349.o463n3Nf004098@demeter.kernel.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" To: linux-ext4@vger.kernel.org Return-path: Received: from demeter.kernel.org ([140.211.167.39]:37777 "EHLO demeter.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752558Ab0EFDtG (ORCPT ); Wed, 5 May 2010 23:49:06 -0400 Received: from demeter.kernel.org (localhost.localdomain [127.0.0.1]) by demeter.kernel.org (8.14.3/8.14.3) with ESMTP id o463n3f6004099 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Thu, 6 May 2010 03:49:03 GMT In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: https://bugzilla.kernel.org/show_bug.cgi?id=15906 --- Comment #17 from Dmitry Monakhov 2010-05-06 03:48:58 --- (In reply to comment #15) > The summary for this patch should probably be changed, yes? It should happen > with filesystems other than ext4 that use barriers (i.e., btrfs, XFS), and it > should happen on a single disk just as easily as a LVM. The latter I can > confirm --- you can see the performance degradation quite easily on a single > disk, so the mention of LVM in the summary is probably misleading. Just a note, performance degradation is not that huge for XFS, as soon as i understand this is because it batch several inode writes in to one 'log' so less barriers required But nor than less this is generic writeback issue. -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.