From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261944AbTEZSOc (ORCPT ); Mon, 26 May 2003 14:14:32 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261950AbTEZSOc (ORCPT ); Mon, 26 May 2003 14:14:32 -0400 Received: from vsat-148-64-8-86.c119.t7.mrt.starband.net ([148.64.8.86]:260 "EHLO chaos.mshome.net") by vger.kernel.org with ESMTP id S261944AbTEZSOa (ORCPT ); Mon, 26 May 2003 14:14:30 -0400 Date: Mon, 26 May 2003 12:26:55 -0600 From: Robert Creager To: Bruce Harada Cc: linux-kernel@vger.kernel.org Subject: Re: Problem with virtually no buffer usage in 2.4.21mdk kernel Message-Id: <20030526122655.0042c298.Robert_Creager@LogicalChaos.org> In-Reply-To: <20030527024141.5b168415.bharada@coral.ocn.ne.jp> References: <20030524223320.7c1ac413.Robert_Creager@LogicalChaos.org> <20030525151816.4a433953.bharada@coral.ocn.ne.jp> <20030525081253.05d4e5f3.Robert_Creager@LogicalChaos.org> <20030527024141.5b168415.bharada@coral.ocn.ne.jp> Organization: Starlight Vision, LLC. X-Mailer: Sylpheed version 0.8.11claws (GTK+ 1.2.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: multipart/signed; protocol="application/pgp-signature"; micalg="pgp-sha1"; boundary="=.ok:ag0eV+yI3GP" Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --=.ok:ag0eV+yI3GP Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit On Tue, 27 May 2003 02:41:41 +0900 Bruce Harada said something like: > On Sun, 25 May 2003 08:12:53 -0600 > Robert Creager wrote: > > > That's why I said virtually no buffer usage (not shared memory). I'm > > looking at the 88kb (90112). On my other two systems, both 2.4.19 with > > 640Mb and 128Mb memory, the buffer usage is 86Mb, and 44Mb respectivly. > > > > Other thoughts? > > Ah, my apologies. As for why your buffer usage is so low, it's likely a bit > difficult to tell whether it's indicative of a problem without knowing what > patches have been applied to the Mandrake kernel. Also, I believe the buffers > value these days isn't as useful a metric as it used to be, so personally I > wouldn't worry about it too much, unless you're seeing problems with the box. > I'm not seeing any problems that I'm aware of, just trying to eak out a little more performance out of the box, and the lack of disk buffer usage seems suspicious. I may try backing of to a known released kernel, and see what happens... Cheers, Rob -- O_ --=.ok:ag0eV+yI3GP Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iEYEARECAAYFAj7SXH8ACgkQgy51bQc2FFk2tgCcCEP8DyTqGOoSLeTOXsacVjlN qGkAn19MzucjjYir8aVlwfF6HVNtXgJz =uoZ3 -----END PGP SIGNATURE----- --=.ok:ag0eV+yI3GP--