[Coco] System-State Remote Debugging for NitrOS-9

[Boisy G. Pitre]

On Apr 4, 2005, at 11:00 AM, Robert Gault wrote:

> Should work and it would be interesting. However, if the main 
> objection to a pure Coco system is memory, why not reserve this for 
> 1Meg or bigger systems?
>
> I'd expect that a KrnP3 of the same size which would jump to a host 
> could be made to switch in extra memory and a debugging program. Even 
> though a NitrOS-9 Coco can use up to 2Meg, there are few cases where 
> the memory would actually be in use so the debugger could fit there.

True, this could be done.  However, debugging graphic/window based 
programs would be difficult because you would be debugging on the same 
screen that you're doing I/O on.  It could be done, however, but by 
using a dedicated serial port, we can see our debugging window on the 
host while watching the CoCo's screen change.

> Most likely there are more users with a PC and a 512K Coco than a 1Meg 
> or bigger Coco. But in that case, why not just use the debugging 
> functions of MESS or Collyer's JVC emulators than writing new code for 
> a debugger?

In my case, I will be testing NitrOS-9 drivers for the SuperBoard, and 
the emulator just doesn't give me the "true" environment I need.  Don't 
get me wrong, the emulator is a great tool, but it cannot replace the 
real hardware environment and all of the intricacies that it entails.

--
Boisy G. Pitre
E-Mail: boisy at boisypitre.com
Mobile: (337) 781-3997
Web: www.boisypitre.com