Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1947119AbWKKGxF (ORCPT ); Sat, 11 Nov 2006 01:53:05 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1947120AbWKKGxF (ORCPT ); Sat, 11 Nov 2006 01:53:05 -0500 Received: from smtp.osdl.org ([65.172.181.4]:10392 "EHLO smtp.osdl.org") by vger.kernel.org with ESMTP id S1947119AbWKKGxC (ORCPT ); Sat, 11 Nov 2006 01:53:02 -0500 Date: Fri, 10 Nov 2006 22:52:57 -0800 From: Andrew Morton To: "Igor A. Valcov" Cc: linux-kernel@vger.kernel.org, xfs@oss.sgi.com Subject: Re: XFS filesystem performance drop in kernels 2.6.16+ Message-Id: <20061110225257.63f91851.akpm@osdl.org> In-Reply-To: References: <4553F3C6.2030807@sandeen.net> X-Mailer: Sylpheed version 2.2.7 (GTK+ 2.8.17; x86_64-unknown-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 943 Lines: 26 On Fri, 10 Nov 2006 16:16:27 +0300 "Igor A. Valcov" wrote: > Below is a simplified version of the test program, Boy, I hope not. The results of this test program are of very little interest. > for (i = 0; i < 262144; i++) { > /* Write data to a big file */ > write (nFiles [0], buf, __BYTES); > > /* Write data to small files */ > for (f = 1; f < __FILES; f++) > write (nFiles [f], &f, sizeof (f)); > } This sits in a loop doing write(fd, buf, 4). This is wildly inefficient - you'd get a 10x throughput benefit and maybe 100x reduction in CPU cost simply by switching to fwrite(). I suspect something went wrong here. - 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/