On Fri, 29 Sep 1995 17:32:46 -0400 
 David Gilbert <dgilbert@pci.on.ca> wrote:

 > Program received signal SIGSEGV (11), Segmentation fault
 > 0x3e74 in CG2SaveScreen ()
 > (gdb) bt
 > #0  0x3e74 in CG2SaveScreen ()
 > #1  0x4660 in sunInitCommon ()
 > #2  0x3fd0 in sunCG2Init ()
 > #3  0x1d9a4 in AddScreen ()
 > #4  0x3530 in InitOutput ()
 > #5  0x1d0f4 in main ()
 > (gdb) 

I doubt that this is the same problem as the fellow with the mac is 
having.  He's puking when libc.so is closed.  Your executable is clearly 
getting _much_ further than that :-)

In your case, I suspect that the X server is mmap()'ing the device, and 
the driver isn't doing something that's expected, and so the server 
pukes.  Note, that's just a guess, not having looked at the Xsun code :-)

Are you able to get any output on the console?  I.e. if you use eeprom(8) 
to set the console device to `color' and then halt and power-cycle the 
system, does the box place console output there?  If the driver's buggy, 
I'd expect the PROM to place output there (before booting, etc) and the 
kernel to not.

Of course, it's entirely possible that the kernel would be able to print 
to the cgtwo console even with a busted cgtwo driver, given that the Sun 
4 uses PROM routines for that purpose.  (Rcons doesn't seem to work, 
which is why I made sure it didn't get attached on a Sun 4 when I sent 
those diffs to Paul.  Besides not having an OpenProm to get dimenstions 
from, it panic'd the kernel on my 4/260 ... I forget the exact panic 
message, and it wasn't really important to me at the time.)

In the event that you do set the console to `color' and the cgtwo driver 
is busted, you can reset the console by setting diag switch to `diag', 
going to the PROM prompt and:

	> q 1f
	(the prompt it displays) XX
	(the prompt it displays) .

Where "XX" is 10 for ttya and 11 for ttyb.  Sorry, I don't have a box in 
front of me at the moment, and forgot the prompt the binary editor uses 
... I think it's the location and the value in brackets ... not important :-)
Once you've made that incantation, you can put the diag switch back into 
`normal' mode.

Ciao.

--------------------------------------------------------------------------
Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939