Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 12 Oct 2001 11:55:20 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 12 Oct 2001 11:55:11 -0400 Received: from soul.helsinki.fi ([128.214.3.1]:63748 "EHLO soul.helsinki.fi") by vger.kernel.org with ESMTP id ; Fri, 12 Oct 2001 11:54:55 -0400 Date: Fri, 12 Oct 2001 18:55:25 +0300 (EET DST) From: Mikael Johansson To: Subject: (memory?) bug between 2.4.9-ac10 and -ac14 In-Reply-To: <20011012083616.C9992@cpe-24-221-152-185.az.sprintbbd.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello All! A bug report which seems related to at least memory management: If becoming root and then su:ing some other, non-local, userid, an oops with segfault is the outcome. Non-local here means that the user info is located on another machine, our server. su:ing the dummy local user seems to work OK. I remember the new shell being the process that oopses, but can't unfortunately get the output (other than "segmentation fault") before Monday as the machine is behind locked doors :-/ Anyway, this occurs if I boot the machine with the full 1.5GB of memory, but _not_ if I specify mem=512M in lilo.conf, so there seems to be some sort of "large memory support" issue. The system is fine with 2.4.9-ac10, and unfine with these (all I've tested): ac-14, ac-15, ac-18. I can get more detailed info on Monday, but thought that maybe someone has come across this before. Brief system specs: Athlon 1.4GHz Abit KT7A (with Athlon/VIA-bug) 3x512MB SDRAM 2x40GB => RAID-0 Have a nice day, Mikael J. - 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/