From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753481Ab1HPCqX (ORCPT ); Mon, 15 Aug 2011 22:46:23 -0400 Received: from mga01.intel.com ([192.55.52.88]:31698 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752398Ab1HPCqW (ORCPT ); Mon, 15 Aug 2011 22:46:22 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.67,378,1309762800"; d="scan'208";a="41784498" Subject: cfq-iosched: a regression From: Shaohua Li To: Jens Axboe Cc: teravest@google.com, lkml , "Shi, Alex" Content-Type: text/plain; charset="UTF-8" Date: Tue, 16 Aug 2011 10:47:06 +0800 Message-ID: <1313462826.27321.22.camel@sli10-conroe> Mime-Version: 1.0 X-Mailer: Evolution 2.32.2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org We have a kernel build regression since 3.1-rc1, which is about 10% regression. The kernel source is in an ext3 filesystem. Alex Shi bisect it to commit: commit a07405b7802691d29ab3b23bdc76ee6d006aad0b Author: Justin TerAvest Date: Sun Jul 10 22:09:19 2011 +0200 cfq: Remove special treatment for metadata rqs. Apparently this is caused by lack metadata preemption, where ext3/ext4 do use READ_META. I didn't see a way to fix the issue, so suggest reverting the patch. Reported-by: Alex Shi Reported-by: Shaohua Li