[Coco] os9/drivewire driver problems

Aaron Wolfe aawolfe at gmail.com
Mon Nov 9 00:37:39 EST 2009


Just a note, my original code that pulled D and A from the stack at
the same time was a total mistake, due to my misunderstanding of..
well most everything :) I've got that sorted now.

On Mon, Nov 9, 2009 at 12:29 AM, Lothan <lothan at newsguy.com> wrote:

> Sheesh... I can't add today. You're right $04 + $02 is $06. I agree the

> other constructs don't make sense, but the result should be either logically

> OR'ing the register postbyte codes such that PULS A,B,D is the same as PULS

> D or PULS A,B (assuming it's just blindly picking through the

> comma-delimited register list) or raising an error that the list doesn't

> make any sense.

>

> --------------------------------------------------

> From: "Robert Gault" <robert.gault at worldnet.att.net>

> Sent: Sunday, November 08, 2009 11:24 PM

> To: "CoCoList for Color Computer Enthusiasts" <coco at maltedmedia.com>

> Subject: Re: [Coco] os9/drivewire driver problems

>

>> Lothan wrote:

>>>

>>> Given the layout of the register postbyte, PULS A,B is exactly the same

>>> as PULS D (e.g. the postbyte is $05 in either case), so PULS A,D and

>>> PULS B,D and PULS A,B,D should in theory all compile exactly the same as

>>> PULS A,B or PULS D.

>>>

>>> Someone feel free to smack me with a clue stick if I have that wrong.

>>

>> No smack, but you might consider this.

>>

>> PULS A,B and PULS D both give $35 $06. So, the post byte is 6 not 5. The

>> other constructs don't make any sense and a good assembler should report an

>> error. After all, you are asking for an ambiguous action. What values are

>> going to be put in regs A & B if you ask for PULS A,D?

>>

>> EDTASM gives Register Error for a multiple pull/push of the same register.

>>

>> --

>> Coco mailing list

>> Coco at maltedmedia.com

>> http://five.pairlist.net/mailman/listinfo/coco

>>

>

> --

> Coco mailing list

> Coco at maltedmedia.com

> http://five.pairlist.net/mailman/listinfo/coco

>




More information about the Coco mailing list