[Coco] Trying to build new images

[Bill Gunshannon]

On 1/11/17 10:51 AM, Gene Heskett wrote:
> On Wednesday 11 January 2017 08:16:01 Bill Gunshannon wrote:
>
>> OK, next set of questions.
>>
>> Is rel a module or just a program?
>>
>> Why, you ask?
>> I saved the file used to build the boot track and ran ident
>> against it.  rel is listed as not a module and krn gets a bad
>> header message.
>>
>> Looking at the source for rel it appears to have everything in
>> it to be a module, but I'm no expert (yet!).
>>
>> Next, I tried to assemble rel.asm on my system.  It wouldn't
>> assemble.  Are the sources not in a format recognized by the
>> NitrOS9 asm command?
>>
>> Would the bad header in krn cause it to not execute?
>>
>> Anybody want to hazard a guess why these programs appear to be
>> broken?  Or, are the results from ident bogus?
>>
>> Oh yeah, I tried running ident against rel and krn in other disk
>> images with the same results.
>>
>> bill
>
> Ident is broken when run against the track 34 contents because that track
> is contaminated with a 6 byte rsdos header.  And a batch of trailing
> bytes.
>
> To see whats there, use my vfy with the -k option. The -k option turns it
> into showing the 6 byte preamble that that rsdos needs to identify it as
> an executable. vfy should then show the 3 modules that are in that
> track, with good crc's, and the trailing bytes, usually filled with hex
> 39's up to the irq vectors that bring it up to a full 4608 bytes to fill
> the track AND out those vectors at the correct locations in ram once rel
> has done its thing. If you're not familiar with vfy, get it from my web
> page. Put the vfy file in the archive in your cmds directory and run it
> with no arguments. It will spit out its own help screen. It has many
> options, and works on any file you want to change "insitu".
>
> Cheers, Gene Heskett
>

I think we already determined that you can  not do that on a
Level 1 systems and that is what I am currently working with.

bill