From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261405AbUDISYX (ORCPT ); Fri, 9 Apr 2004 14:24:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261582AbUDISYX (ORCPT ); Fri, 9 Apr 2004 14:24:23 -0400 Received: from adsl-67-65-232-1.dsl.lgvwtx.swbell.net ([67.65.232.1]:13963 "HELO rooker.dyndns.org") by vger.kernel.org with SMTP id S261405AbUDISYS (ORCPT ); Fri, 9 Apr 2004 14:24:18 -0400 Message-ID: <000901c41e5f$b7288930$6600a8c0@pixl> From: "Peter Maas" To: Subject: Re: 2.6.5-mm3 Date: Fri, 9 Apr 2004 13:23:10 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I had posted under the topic SMP + EXT3 + AACRAID yesterday about lockups occuring when I ran a 'dbench 32' on a dual pentium 4 with HT, on an Adaptec 2120S (aacraid) with an ext3 filesystem. So far with my testing on mm3 I have not had a lockup. 'dbench 32' also is also much faster, from about 60MB/s on 2.6.5 (the few times it did not lock) to 108MB/s on 2.6.5-mm3. I have SMT scheduler support enabled along with 4K stacks. 2.6.5-mm3+SMP+AACRAID+EXT3= completes 2.6.5+SMP+AACRAID+EXT3= lockup 2.6.5+SMP+AACRAID+EXT2= completes 2.4.21-9REL+SMP+AACRAID+EXT3= lockup 2.4.21-9REL+SMP+AACRAID+EXT2= completes 2.4.21-9REL+SingleProc+AACRAID+EXT3= lockup 2.4.21-9REL+SingleProc+AACRAID+EXT2=completes 2.6.5-mm3 is working for me!