From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261649AbTJFT5n (ORCPT ); Mon, 6 Oct 2003 15:57:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261656AbTJFT5m (ORCPT ); Mon, 6 Oct 2003 15:57:42 -0400 Received: from mx3.evanzo-server.de ([81.209.142.20]:4747 "EHLO mx3.evanzo-server.de") by vger.kernel.org with ESMTP id S261649AbTJFT5l (ORCPT ); Mon, 6 Oct 2003 15:57:41 -0400 From: Markus Schoder To: linux-kernel@vger.kernel.org Subject: kernel panic with 2.6.0-test6-mm4 and nptl Date: Mon, 6 Oct 2003 21:57:31 +0200 User-Agent: KMail/1.5.4 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200310062157.31805.lists@gammarayburst.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org 2 out of 3 times I got a kernel panic when running the tst-eintr1 test from current glibc (CVS) and nptl 0.60. It also happens sometimes during normal usage. I used all features defined in sysdeps/unix/sysv/linux/kernel-features.h of glibc. The message is `Fatal exception in interrupt'. The stack trace looked like this (from memory) do_page_fault+0x0/0x50c do_page_fault+0x41/0x50c do_page_fault+0x0/0x50c error_code+0x2f/0x50c repeated 100s of times (infinite recursion?). Machine is an Athlon XP 2000+, 768MB RAM, no modules loaded. On one occasion I experienced a similar panic with 2.6.0-test6 plain when using the same glibc but I could not reproduce it. -- Markus