Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!rutgers!bellcore!dino!dana
From: dana@dino.bellcore.com (Dana A. Chee)
Newsgroups: comp.windows.x
Subject: Re: Fatal Sun R4 server bug
Message-ID: <22014@bellcore.bellcore.com>
Date: 11 Apr 90 21:33:35 GMT
References: <3078@goanna.oz.au>
Sender: news@bellcore.bellcore.com
Reply-To: dana@thumper.bellcore.com
Organization: Packet Communications Research Group (Bellcore)
Lines: 58


In article <3078@goanna.oz.au>, dmh@goanna.oz.au (Darren Hosking) writes:
|> Running the ULE (Utah raster toolkit) image display program getX11 causes
|> the following error and subsequent server crash on a Sun 3 (OS 3.5):
|> 
|> sunCG4CUpdateColormap: Invalid argument
|> 
|> Fatal server bug!
|> sunCG4CUpdateColormap: FBIOPUTCMAP failed
|> 
|> We are running X11R4 with fixes 1 to 9. Does anyone have any idea of what
|> the problem (and solution) is?
|> 
|> 	Thanks in advance, dmh
|> 
|> Darren Hosking				ACSnet: dmh@goanna.oz	 
|> Department Of Computer Science		ARPA:   dmh%goanna.oz.au@uunet.uu.net
|> RMIT					CSNET:  dmh%goanna.oz.au@australia
|> GPO Box 2476V				UUCP:   ...!uunet!goanna.oz.au!dmh
|> Melbourne Vic., Australia 3001

I know the cure, but not the complete disease.  The problems are many,
but here's a brief description:

1) getX11 is a very naughty boy trying to install its own colormap
without setting the window resource and letting the window manager do
it.

2) more important, getX11 creates a colormap, puts absolutely nothing
in it, and then installs it.  No telling what the server receives but
it makes it REALLY unhappy.  Below is the code:

color_map = XCreateColormap (dpy, root_window, visual, AllocNone );
if (color_map == NULL) {
    fprintf (stderr, "Could not create color map for visual\n");
    exit (EXIT_STATUS_FAILURE);
    }

XInstallColormap (dpy, color_map);

3) The sun server advertises that it supports all sorts of visuals
that it doesn't have.  getX11 likes to chose DirectColor if it can
find it.  I have no idea (haven't had time to look) what effect
allocating a colormap under DirectColor has on a PseudoColor device.

Anyway, my fix was to nuke the CreateColormap/InstallColormap stuff
for now (this was for someone else, and he needed it to run NOW), set:
color_map = DefaultColormap (dpy, screen);
and run it as pseudocolor.  It would be nicer if it was really fixed
but that's enought to hold it for now.

+*************************************************************************+
*  Dana Chee				(201) 829-4488			  *
*  Bellcore								  *
*  Room 2Q-250								  *
*  445 South Street			ARPA: dana@bellcore.com		  *
*  Morristown,  NJ  07960-1910		UUCP: {gateways}!bellcore!dana	  *
+*************************************************************************+