[Coco] mess HD problem
Bob Devries
devries.bob at gmail.com
Wed Aug 6 19:42:28 EDT 2014
Robert,
thanks for that information.
I looked at the drivers and descriptors in the current NitrOS9
repository, and found that the emudsk there is built using $FF80 as the
controller address.
I must somehow have mixed up the DD descriptor from an earlier/Alan
Dekok version.
The CRC of the emudsk module matches that of the current NitrOS9 driver,
so I assume I need to have $FF80 as the port address. I can't understand
how it worked at all!
Thanks for your input.
Regards, Bob Devries
Dalby, QLD, Australia
On 7/08/2014 9:09 AM, Robert Gault wrote:
> Bob Devries wrote:
>> Hi all,
>>
>> well, changing the port address to $FF80 appears to have fixed the
>> problem.
>>
>> I don't know how I came by a module with the incorrect hpa. Could it
>> be that a
>> previous version of emudsk used that address?
>>
>> Thanks for your input everyone.
>>
>> Regards, Bob Devries
>> Dalby, QLD, Australia
>>
>
> Bob,
>
> Here is some info.
>
> *Technical specs on Virtual Hard Disk interface Alan DeKok: JVC, MESS
> and VCC
>
> * Address Description
> * ------- -----------
> * FF80 Logical record number (high byte)
> * FF81 Logical record number (middle byte)
> * FF82 Logical record number (low byte)
> * FF83 Command/status register
> * FF84 Buffer address (high byte)
> * FF85 Buffer address (low byte)
> * FF86 DR# 0=first drive 1=second drive; MESS only
>
> Look at MESS/SRC/machine/coco_vhd.c
>
> -----------------------------------------------------------------------------
>
> * The JV Coco3x has 2 .vhd drives with the second address set as:
> * Address Description
> * ------- -----------
> * FFE0 Logical record number (high byte)
> * FFE1 Logical record number (middle byte)
> * FFE2 Logical record number (low byte)
> * FFE3 Command/status register
> * FFE4 Buffer address (high byte)
> * FFE5 Buffer address (low byte)
>
>
More information about the Coco
mailing list