*BSD News Article 18909


Return to BSD News archive

Newsgroups: comp.os.386bsd.bugs
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!howland.reston.ans.net!darwin.sura.net!news-feed-1.peachnet.edu!concert!sas!mozart.unx.sas.com!torpid.unx.sas.com!sastdr
From: sastdr@torpid.unx.sas.com (Thomas David Rivers)
Subject: Re: Problems with XFree86.1.3
Sender: news@unx.sas.com (Noter of Newsworthy Events)
Message-ID: <CAsu68.I8w@unx.sas.com>
Date: Tue, 27 Jul 1993 01:15:44 GMT
References: <CA71wq.BF2@cosy.sbg.ac.at> <1993Jul23.193714.3176@knobel.GUN.de>
Nntp-Posting-Host: torpid.unx.sas.com
Organization: SAS Institute Inc.
Lines: 57

In article <1993Jul23.193714.3176@knobel.GUN.de> andreas@knobel.GUN.de (Andreas Klemm) writes:
>In <CA71wq.BF2@cosy.sbg.ac.at> peter@cosy.sbg.ac.at  (Peter Burgstaller) writes:
>
>|Hi folks!
>
>|Last week I moved from XFree86.1.2 to the newer version.
>|I have some problems with iconifying windows.
>|Sometimes the X server is going down and sometimes not.
>|It may be related to my new graphic card *ATI Graphics Ultra pro 2MB*.
>|However it doesn't depend on the windowmanager. I tried 
>|olwm, olvwm, gwm and twm.
>|Systembackground:
>|486DX/2 66 EISA 100MB Conner + 1GB Dec HD's.
>
>No problems here with XFree86.1.3.
>I compiled the mit tree with the XFree86 diffs applied using gcc-2.4.5
>without problems and ... it runs fine. I tried iconifying several
>windows ... no problems.
>System: NetBSD-0.8 on a 
>486/33 EISA, VI-811 (ET4000 based local bus video card), 16 Mb.
>-- 
>----  Andreas Klemm  -------    /////// andreas@knobel.GUN.de \\\\\\\
>private : +49 2137 1 26 09      D-41469 Neuss, Germany
>at work : +49 2173 3964 161     Wiechers & Partner Datentechnik GmbH
>Telefax : +49 2173 3964 222     Abteilung Unix Support, D-40789 Monheim



There was, indeed a problem in XFree86-1.3, which was 
particularly tickled when using olwm and iconifying windows,
but could have occurred anywhere.

The problem doesn't show up on non-386BSD operating systems;
because of "less strict" memory checks.  

Basically, the code in the server which drew lines would
overwrite the end of an array. 

There is a new version of 1.3, 1.3u1 (update 1) which corrects
the problem.

Also, there is a problem with line drawing in the speedup mode
of the ET4000 code; the problem only occurs when your virtual
width is not 1024.  The update removes the particular speedup
which causes the difficulties.


All these items were revealed in the newsgroup which handles
XFree86 problems, and, I believe an announcement of this
was posted here earlier - so, for more details you may want
to look there.

	- Dave Rivers -
	(rivers@ponds.uucp (home))
	(sastdr@unx.sas.com (work))
-- 
UPDATE ALL INFORMATION AND POD INTO COSMOS - Federal Express