Re: system/3802: xlock crashes X server

[Matthieu Herrb <matthieu.herrb@laas.fr>]

The following reply was made to PR system/3802; it has been noted by GNATS.

From: Matthieu Herrb <matthieu.herrb@laas.fr>
To: ntk@seas.upenn.edu
Cc: gnats@openbsd.org
Subject: Re: system/3802: xlock crashes X server
Date: Thu, 27 May 2004 23:45:22 +0200

 ntk@seas.upenn.edu wrote:
 
 >>Number:         3802
 >>Category:       system
 >>Synopsis:       xlock crashes X server
 >>Confidential:   yes
 >>Severity:       serious
 >>Priority:       medium
 >>Responsible:    bugs
 >>State:          open
 >>Quarter:        
 >>Keywords:       
 >>Date-Required:
 >>Class:          sw-bug
 >>Submitter-Id:   net
 >>Arrival-Date:   Thu May 27 19:30:02 GMT 2004
 >>Closed-Date:
 >>Last-Modified:
 >>Originator:     Nathan Kennedy
 >>Release:        3.5
 >>Organization:
 > 
 > net
 > 
 >>Environment:
 > 
 > 	
 > 	System      : OpenBSD 3.5
 > 	Architecture: OpenBSD.i386
 > 	Machine     : i386
 > 
 >>Description:
 > 
 > 	I first encountered this problem when I started a kernel compile
 > 	in an xterm, started xlock, and left.  When I came back, X had
 > 	crashed:
 > 
 > 	Fatal server error:
 > 	Caught signal 11.  Server aborting
 > 
 > 	I found that if I started xlock, hit a key and then clicked on
 > 	the "thumbnail" of the screensaver to lock the screen again
 > 	and repeated this a few times to cycle through the various modes,
 > 	it would soon crash the server again.  In particular this happened
 > 	quickly when I ran xlock -mode molecule although I doubt it is
 > 	a problem with this particular mode.
 > 
 > 	The system is a Pentium II running OpenBSD 3.5 GENERIC.
 > 
 
 A bit terse as a report. How do we guess which graphic card you have, 
 which X server you're using, and how it's configured?
 
 Having said that, the 'molecules' saver seems to be leeking data in the 
 X server. I'd bet that it doesn't free its GL display lists correctly, 
 but I'm not sure.
 -- 
 					Matthieu