[Coco] bad sysgo in nightly download?

[Tim Fadden]

On 1/7/2014 2:39 PM, Gene Heskett wrote:
> On Tuesday 07 January 2014 16:38:17 Tim Fadden did opine:
>
>> On 1/7/2014 12:47 PM, Gene Heskett wrote:
>>> On Tuesday 07 January 2014 14:39:03 Tim Fadden did opine:
>>>> Hi gang,
>>>>
>>>> I downloaded the nightly build 2 days ago, and have been trying to
>>>> build custom boot floppy's  and dw images, without success.
>>>>
>>>> I have narrowed it down that the files sysgo_dd and sysgo_h0 in the
>>>> nitros9 directory path are both bad, and much larger than working
>>>> sysgo's.
>>>>
>>>> I have no way of compiling these.
>>>>
>>>> Is it acceptable to use sysgo's from prior nitros images that are
>>>> known good?  I'm going to try, and expect it should work but.......
>>> Sysgo_dd is 483 bytes as built here, how big is yours?
>>>
>>> And size shouldn't be an issue since its a transient program that is
>>> executed once at bootup, and for size reasons, probably needn't be
>>> taking up room in the OS9Boot file, or in precious system ram.  It
>>> should load, execute, and go away.
>>>
>>> I would much rather believe that its failure is a bad descriptor for
>>> /h0 or /dd as the case might be. That would prevent its being found.
>>>
>>> Do you actually have your hard drive formatted and loaded with nitros9
>>> already?
>>>
>>>> Thanks.
>>>>
>>>> Tim
>>>>
>>>> --
>>>> Coco mailing list
>>>> Coco at maltedmedia.com
>>>> http://five.pairlist.net/mailman/listinfo/coco
>>> Cheers, Gene
>> Gene,
>>
>> The sysgo from the nightly download is 511 bytes.
>>
>> The reason I say it is bad is because ident say it is bad, and if I use
>> a previous sysgo, it works.
>>
>> This had nothing to do with the hard drive , I was trying to use the
>> /dd. Also, all of this is concerning making boot floppy's.
>>
>> How I caught on was that I was putting it into the os9boot file. When
>> booting from that image it failed with out of memory.
>>
>> After some more studying etc.  I took it out of the os9boot file, and on
>> the /dd device.  still no go.  Then I did an ident on it , and ident
>> says BAD.
>>
>> I then did Ident on the files on the nitros9 modules directory and both
>> the sysgo_dd and sysgo_h0 came up bad.
>>
>> I have succsesfully created literally hundreds of boot floppy's through
>> the years, so not a complete neophyte.  :-)
>>
>> This is the output from the sysgo that is on what I download from the
>> nightly link.
>>
>> Header for:  SysGo
>> Module size: $01FF    #511
>> Module CRC:  $221288 (Bad)
>> Hdr parity:  $D4
>> Exec. off:   $00F1    #241
>> Data Size:   $00FC    #252
>> Edition:     $05      #5
>> Ty/La At/Rv: $11 $83
>> Prog mod, 6809 obj, re-en, R/O
>>
>> Notice that it says (Bad).
>>
>> If the nightly link on the sourceforge page is not a good image, where
>> is the best latest image?
>>
>>
>> I may still not have everything correct,  but I can only fix one thing
>> at a time as I figure it out.
>> This is a problem.  Whether it is my overall problem is of no
>> importance. Ident should run clean on the sysgo modules.  After that is
>> sorted out, then I can move on.
>> Seeing I don't have a way to compile the source code, I am at a dead
>> end. By the way. using a known good sysgo module from a previous
>> version works.
>>
>> I brought this
>>
>> Ain't computers fun!
>> Tim
>>
>> p.s. when the sysgo issue is solved, I will start with the
>> superdriver/tc3 issue.  That does not work with the the nightly download
>> either.  Mark is also having problems with it, so it is not just me
>> doing something wrong.
> Humm, have you built that but used a known good descriptor?
>
> Cheers, Gene
Gene,

Please explain why you are stuck on the /dd desciptors.  I used the 
/x0_dd descriptor with the release, and the scsi descriptor I made. 
Neither of these work with the latest release.

The dw image DOES work. They both work just fine. I guess I could steal 
the sysgo from that and fix the modules/ sysmod files myslef, but that 
does not fix the issue for the distribution, just me.  They both work 
just fine.

Why does the sysgo module fail ident?  this is not normal is it? All the 
previous sysgo modules that I have pass an ident.

thx for the attention,  :-)

Tim