r.o.s.s wrote:
>>From: David Hopper <dhop@nwlink.com>
>>
>>I'd like to get further in debugging the X -scanpci code with an AlphaServer 
>>1200, which has dual PCI buses.  The code right now causes a hard reset, with 
>>no core file, no DDB, and an empty error log.
>>
>>Is there any way to create a sandbox that could catch this error in situ?
>>
>>Thanks,
>>Dave
> 
> You know, I would be .. like .. so tempted to simply design a less
> brain-damaged approach. I mean, scanning "the" pci bus looking for
> video thingies is such a hack. It should simply ask the console
> where the frame buffer is.  Just something to think about...

Ach.  You're absolutely correct, of course, but that's way beyond _this_ 
Anthropology-degree-holder...

You're posing a major architectural change to XFree86 itself, and based upon 
the glacial speed of that codebase, it might be easier to just put a 
propeller-beanie on the brain-damaged approach and call it 'good'.

Besides, don't all these different chipsets need to map video memory, texture 
memory, and address spaces differently?  Making a unified "ask the console" 
device would mean creating kernel module drivers for individual chipsets, and 
is the main reason why fb-dev in Linux is in such a quagmire...

> //

Dave