From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 12 Sep 2001 02:28:56 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 12 Sep 2001 02:28:47 -0400 Received: from finch-post-12.mail.demon.net ([194.217.242.41]:15372 "EHLO finch-post-12.mail.demon.net") by vger.kernel.org with ESMTP id ; Wed, 12 Sep 2001 02:28:33 -0400 Date: Wed, 12 Sep 2001 07:28:53 +0100 To: linux-kernel@vger.kernel.org Subject: Re: [GOLDMINE!!!] Athlon optimisation bug (was Re: Duron kernel crash) Message-ID: <20010912072853.A479@berserk.demon.co.uk> Mail-Followup-To: linux-kernel@vger.kernel.org In-Reply-To: <20010912082935.A391@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010912082935.A391@localhost>; from vdb@mail.ru on Wed, Sep 12, 2001 at 08:29:35AM +0400 From: Peter Horton Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Sep 12, 2001 at 08:29:35AM +0400, Dmitry Volkoff wrote: > >I don't have any problems with the kernel, but if I'm compiling (say) the > >kernel or any large program, user-mode apps like gcc, make, etc all > >randomly segfault on me. > > > >I'm running a Duron 800 on a Asus A7V133 mobo - and no bios version I've > >tried has made a bit of difference yet :p > > Same things here. I'm also experiencing random segfaults with kernels > 2.4.10pre[47]. My motherbord is Chaintech CT-7AIA2 (kt133A/686A), Duron 600. > I did'nt see such things with kernels 2.4.[78]. The problem persists even > after recompiling kernel with K6-III optimisation. Note, this motherboard > was rock solid and I never had any problems with athlon-optimised kernels > prior to 2.4.9. > A colleague purchased a number of A7V133s all of which exhibited seg faults in make and gcc. make reproducibly seg faulted in vfork(). This was with 2.2.18 kernels compiled for i586. We next tried a 2.4.8 kernel built for i686 and the problem persisted, though less reproducibly. The kernel reports 'Applying VIA southbridge workaround', but it looks like we need another fix. Most of the BIOS options had no effect, but changing the 'PCI latency' setting from 32 to 64 seems to have fixed it, fingers crossed. P. -- +--------------------------------------------------+ | Peter Horton | pdh@colonel-panic.org | | Software Engineer | http://www.colonel-panic.org | +--------------------------------------------------+