[Coco] 6809 Push/Pull

[Robert Gault]

There is no reason to research such a question when you only need to run a simple test. After all, 
programs like EDTASM are readily available.

So here are the results of two tests. The sources are listed and the debug results for EDTASM.

100 START	LDX	#1
110		LDY	#2
120		PSHS	X,Y
130		PULS	X
140		SWI
150		END

A/IM		assembles the above in memory
Z		enter Debug
G START		run the above and stop automatically at SWI
R		display the register contents

X=0001 Y=0002

100 START 	LDX #1
110 		LDY #2
120 		PSHS X,Y
130 		PULS Y
140 		SWI
150 		END

gives
X=0001 Y=0001

100 START 	LDX #1
110 		LDY #2
120 		PSHS X,Y
130 		LDX #0
140		LDY #0
130 		PULS Y,X
140 		SWI
150 		END

X=0001 Y=0002

100 START 	LDX #1
110 		LDY #2
120 		PSHS X,Y
130 		LDX #0
140 		LDY #0
130 		PULS X,Y
140 		SWI
150 		END

X=0001 Y=0002

So, when EDTASM assembles the above, both PSH and PUL put the registers involved in a "post byte" 
where the registers become bits. That makes the post bytes for PULS X,Y and PULS Y,X identical and 
$30. It also means the 6809/6309 will ignore the order in which you write the involved registers and 
the stack order will be determined by the values in the post byte.

push/pull post byte
bit7                     bit0
PC, S/U, Y, X, DP, B, A, CC

That is easily seen by examining the stack using EDTASM's debug program.

Robert

Lee wrote:
> I did a quick Google search and this article seems to cover it pretty
> well.  https://silicon-heaven.com/howel/parts/6809_instruction_set.htm
>
> It looks like registers (on a single PSHS/PSHU) are pushed in the order PC,
> U/S, Y, X, DP, B, A, CC.  They are PULS/PULU'd in the opposite order.
>
> So if you did PSHS X,Y it would push Y first, then X.  If you did PULS X,Y
> it would pull in the opposite order, X first, then Y.
>
> -----
> Lee
>
>
> On Fri, Jun 5, 2020 at 9:44 PM Salvador Garcia via Coco <
> coco at maltedmedia.com> wrote:
>
>>   Adding to this thread...
>>
>> Is the developer allowed to do some thing like:
>>
>> PSHS    X
>>
>> PULS    Y
>>
>>
>> What if I try something like this:
>>
>> PSHS    X,Y
>>
>> PULS    X
>>
>>
>> Like others, just curious. :-) Thanks, Salvador
>>
>>
>>
>>
>>      On Friday, June 5, 2020, 03:10:58 PM CDT, Arthur Flexser <
>> flexser at fiu.edu> wrote:
>>
>>   The order of registers in PSHS and PULS is a fixed property of the
>> processor.  PSHS X,Y and PSHS Y,X generate identical opcodes, so it makes
>> no difference which you use.
>>
>> Art
>>
>> On Fri, Jun 5, 2020 at 4:06 PM Jim OKeefe <jimokeefe39 at gmail.com> wrote:
>>
>>> For all you ML gurus out there.
>>>
>>> PSHS    X,Y
>>>
>>> To return registers to their original values is the correct pull
>>>
>>> PULS      Y,X
>>>
>>> Or
>>>
>>> PULS      X,Y
>>>
>>> Thanks Jim
>>>
>>> --
>>> Coco mailing list
>>> Coco at maltedmedia.com
>>> https://pairlist5.pair.net/mailman/listinfo/coco
>>>
>>
>> --
>> Coco mailing list
>> Coco at maltedmedia.com
>> https://pairlist5.pair.net/mailman/listinfo/coco
>>
>>
>> --
>> Coco mailing list
>> Coco at maltedmedia.com
>> https://pairlist5.pair.net/mailman/listinfo/coco
>>
>