From mboxrd@z Thu Jan 1 00:00:00 1970 Path: main.gmane.org!not-for-mail From: Robert Marshall Newsgroups: gmane.emacs.devel Subject: Re: Fix to long-standing crashes in GC Date: Sat, 15 May 2004 05:39:44 +0100 Organization: The first against the wall Sender: emacs-devel-bounces+emacs-devel=quimby.gnus.org@gnu.org Message-ID: References: <40A3BC23.8060000@math.ku.dk> NNTP-Posting-Host: deer.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 8bit X-Trace: sea.gmane.org 1084602217 8155 80.91.224.253 (15 May 2004 06:23:37 GMT) X-Complaints-To: usenet@sea.gmane.org NNTP-Posting-Date: Sat, 15 May 2004 06:23:37 +0000 (UTC) Original-X-From: emacs-devel-bounces+emacs-devel=quimby.gnus.org@gnu.org Sat May 15 08:23:29 2004 Return-path: Original-Received: from quimby.gnus.org ([80.91.224.244]) by deer.gmane.org with esmtp (Exim 3.35 #1 (Debian)) id 1BOsa1-00082v-00 for ; Sat, 15 May 2004 08:23:29 +0200 Original-Received: from monty-python.gnu.org ([199.232.76.173]) by quimby.gnus.org with esmtp (Exim 3.35 #1 (Debian)) id 1BOsa0-0001tc-00 for ; Sat, 15 May 2004 08:23:29 +0200 Original-Received: from localhost ([127.0.0.1] helo=monty-python.gnu.org) by monty-python.gnu.org with esmtp (Exim 4.34) id 1BOsYX-0003iz-Ed for emacs-devel@quimby.gnus.org; Sat, 15 May 2004 02:21:57 -0400 Original-Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.34) id 1BOsY3-0003bu-En for emacs-devel@gnu.org; Sat, 15 May 2004 02:21:27 -0400 Original-Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.34) id 1BOsXW-0003F3-RN for emacs-devel@gnu.org; Sat, 15 May 2004 02:21:25 -0400 Original-Received: from [80.91.224.249] (helo=main.gmane.org) by monty-python.gnu.org with esmtp (Exim 4.34) id 1BOsXW-0003Ey-B4 for emacs-devel@gnu.org; Sat, 15 May 2004 02:20:54 -0400 Original-Received: from root by main.gmane.org with local (Exim 3.35 #1 (Debian)) id 1BOsXV-0004uN-00 for ; Sat, 15 May 2004 08:20:53 +0200 Original-Received: from cpc1-macc1-6-0-cust223.bagu.cable.ntl.com ([81.97.79.223]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 15 May 2004 08:20:53 +0200 Original-Received: from robert by cpc1-macc1-6-0-cust223.bagu.cable.ntl.com with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 15 May 2004 08:20:53 +0200 X-Injected-Via-Gmane: http://gmane.org/ Original-To: emacs-devel@gnu.org Original-Lines: 44 Original-X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: cpc1-macc1-6-0-cust223.bagu.cable.ntl.com X-Home-Page: http://www.chezmarshall.freeserve.co.uk User-Agent: Gnus/5.1003 (Gnus v5.10.3) Emacs/21.3.50 (gnu/linux) Cancel-Lock: sha1:VxR8tgXLhkK3t+UBDmPD5sv+NLI= X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.4 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+emacs-devel=quimby.gnus.org@gnu.org Xref: main.gmane.org gmane.emacs.devel:23464 X-Report-Spam: http://spam.gmane.org/gmane.emacs.devel:23464 On Thu, 13 May 2004, Lars Hansen wrote: > > For a long time I have experienced occasional crashes of CVS Emacs, > usually during Tramp operations. I have four backtraces in case > someone is interested, but I have not been able produce a test > case. To me it looks as if GC is involved, but I am unexperienced > with gdb. When Kim told he had fixed a bug involving GC and process > output, it seemed at first very likely to be the cause of my > troubles. However: > > 1. In one of the crashes Tramp was not involved. > 2. CVS Emacs of today with Kim's fix, just crashed during a Tramp > operation. > > So now I don't know what to think. > Here is a backtrace from todays crash: > > (gdb) run > Starting program: /home/lh/cvsroot/emacs/src/emacs > > Program received signal SIGABRT, Aborted. > 0x402ec781 in kill () from /lib/libc.so.6 > (gdb) bt > #0 0x402ec781 in kill () from /lib/libc.so.6 > #1 0x080dfe0a in abort () at emacs.c:433 > #2 0x08128037 in mark_object (arg=1217041176) at alloc.c:5034 > #3 0x0812809a in mark_object (arg=-1463681024) at alloc.c:5051 > #4 0x08126b43 in mark_memory (start=0xbfffe070, end=0xbffff7e8) > at alloc.c:3700 > #5 0x08126f73 in mark_stack () at alloc.c:4055 6 0x08127549 in > #Fgarbage_collect () at alloc.c:4429 7 0x081627f5 in Fbyte_code I'm still getting what looks like this error with an old test case and the latest cvs C-x C-f /su:root@localhost:/var/log/dansguardian/Access.log it appears to be sensitive to the length of that file, it needs to be somewhere around (at least for this machine!) 1.6M - it's currently 1684352 Robert -- La grenouille songe..dans son château d'eau