[Coco] NitrOS-9 newbie needs some help (More data)

[Boisy G. Pitre]

On Aug 11, 2004, at 4:59 AM, Nickolas Marentes wrote:

> Latest test includes running the 6809 version of NitrOS-9 and testing 
> the
> formatting.
>
> Without changing the 6309 back to a 6809, I booted the CoCo3 with the 
> 6809
> version of NitrOS-9 and began formating a disk in /D1 as 40 track 
> double
> sided and it worked flawlessly. No changes to the CoCo3, just using 
> the 6809
> version of NitrOS-9. I repeated the test several times, all successful.

Interesting.  The same CoCo, processor and "problematic" disk 
controller is used, and the results with NitrOS-9/6809 Level 2 are 
fine.

> I then booted the 6309 version of NitrOS-9 and proceeded to format a 
> disk in
> /D1 and it failed on every attempt.
>
> By failed, I mean that the CoCo3 simply freezes and remains locked up 
> until
> I do a Cold Reset to return to Extended Color Basic.
>
> So it would seem that NitrOS-9 is at fault but I'm not sure if it is 
> due to
> some timing issue with my CoCo3 or because of some issue with PAL 
> CoCo3's. I
> only have the one 6309 modified CoCo3.

It appears to be a software problem that is exacerbated by certain 
hardware.  I am hypothesizing that the problem code is in the SSWTrk 
portion of rb1773, which is where format calls into during the low 
level format.  There is timing critical code tied to the NMI there.  It 
is possible that the 6309, running in native mode and running slightly 
faster than the 6809, is magnifying a slight timing discrepancy in the 
code.

> I did dig up an old 128K 6809 CoCo3 but of course it doesn't boot 6309
> NitrOS-9 but it did boot up the 6809 version and it too formatted disks
> perfectly.
>
> I'm not sure where else to go except to try find another 512K ram 
> board and
> test to see if it is causing the timing fault. Using a Multi-crap 
> interface,
> it does seem to improve things a bit but it still fails almost half the
> times I try formatting.

I'll look more into this.

Boisy