Any way to do this in Pasmo?

[presh]

Suppose I have an unrolled loop, plus a table pointing to it, e.g.

Code: Select all

ldi_3:  
  LDI
ldi_2:  
  LDI
ldi_1:  
  LDI
  RET
  
ldi_table:
  DW ldi_1
  DW ldi_2
  DW ldi_3

Is there any way for Pasmo to automatically generate the labels? Pseudocode example of the kinda thing I'd like:

Code: Select all

%for i=3 to 1
ldi_{i}:
  LDI
%next i
  RET

ldi_table:  
%for i=1 to 3
  DW ldi_{i}
%next i

I can't tell how to do it from the docs

TIA

[Joefish]

I seem to remember trying something where I used '##' to concatenate a text label with the number of the loop, but I don't think I ever got it to work!
It's mentioned in the documentation under 'Parameter expansion'.

[presh]

Thanks @Joefish.

Yeah, I've just tried ## and & but they give me:

Code: Select all

ERROR: Unexpected '##' used as instruction
ERROR: Unexpected '&' used as instruction

Frustrating that there are no examples in the docs!

[bobs]

I’ve used that stuff in Melkhior’s, but not near the code right now to check just how. Will post something up later…

[bobs]

Sorry, but I can’t get this to work as you like. I think the problem is that ## only works within MACRO statements, as I use it, but IRP & REPT both effectively use a local DEFL to hold the iteration/parameter, which you can’t then pass to macros.

[RMartins]

...
I can't tell how to do it from the docs

TIA

As far as I remember you can do it, but you have to use a macro and the ## operator.

However, with the REPT and current documentation it will probably work too. I need to try though, to be sure.

[bobs]

You can do it with a combination of MACROs and IRPs…

Code: Select all

MACRO DEFLABEL_LDI, _idx
LDI_##_idx EQU $
ENDM

MACRO DWLABEL_LDI, _idx
DEFW LDI_##_idx
ENDM

IRP _idx, 3, 2, 1
DEFLABEL_LDI _idx
ldi
ENDM
ret

IRP _idx, 1, 2, 3
DWLABEL_LDI _idx
ENDM

However, there are much easier ways if you know that your labels are alway n bytes apart, as your look-up table is simply a numeric sequence based on a base address.

[RMartins]

Working sample

Code: Select all

IDXLABEL MACRO prefix, name, suffix
	prefix ## name ## suffix:
ENDM

DW_IDXLABEL MACRO prefix, name, suffix
	DW prefix ## name ## suffix
ENDM

main_start EQU $6000

ORG main_start

main:
	LD HL, $5800
	LD DE, $5820

IRP count, 3
IDXLABEL ldi_, count, _idx
  	LDI
ENDM
  	RET

ldi_table:  
IRP count, 3
	DW_IDXLABEL ldi_, count, _idx
ENDM

END main_start

Save it as "test.asm"

Run it with "-1" parameter, to see the macro expansion working.

Code: Select all

pasmo -1 --tapbas test.asm test.bin

NOTE:
It copies the first 3 Attributes from Attribute row 1 into first 3 Attributes in attribute row 2.

WARNING:
REPT can not be used since the manual states that it's not a macro parameter, so it can't be used as such, which is required in this case to concat the label name.
To concat a label into a single unit, we need to use the ## operator that is only available inside a Macro.

[Bubu]

I don't understand the real problem. Your code is:

Code: Select all

ldi_3:  
  LDI
ldi_2:  
  LDI
ldi_1:  
  LDI
  RET
  
ldi_table:
  DW ldi_1
  DW ldi_2
  DW ldi_3

Is this what you want? :

Code: Select all

address_1:
instructions...
address_2:
instructions...
address_3:
instructions...

table:
DW address_1
DW address_2
DW address_3

And what's the problem with PASMO?

[Stefan]

I don't understand the real problem.
<snip>

If you had first read the answers already given, then you may have had a better idea of the problem.

The question was how to use PASMO macros to automatically generate source code. So while the example is for [ 1, 2, 3 ], the actual use case is probably for a higher number, like [ 1, 2, ... 191, 192 ]

[Joefish]

Working sample...

Is that right? As I'm reading it for how IRP (rather than REPT) works, shouldn't the syntax be???:

IRP count, 1, 2, 3

not:

IRP count, 3

Still, this is gripping stuff! I could never figure it out.

[RMartins]

Is that right? As I'm reading it for how IRP (rather than REPT) works, shouldn't the syntax be???:

IRP count, 1, 2, 3

I haven't tried that, but according to the documentation, I think you would get 6 repetitions (1+2+3 = 6).

Or maybe you are confusing IRP with IRPC ?

But you can always take a look at the sample/test files that are in the Zip in the website.