[TIASM, Brass] Flexible Macros

driesguldolf · Post by **driesguldolf** » Sun 24 Jun, 2007 2:38 pm

Say you have a macro that negates a 16 bit register
you could make macros like this:

.deflong NEG_HL
   ld a, h
   cpl
   ld h, a
   ld a, l
   cpl
   ld l, a
   inc hl
.enddeflong

but what do you need to do if you want to make a more flexible macro: (pseudo)

Code: Select all

.macro NEGATE(reg as string):
   ld a, reg(0)
   cpl
   ld reg(0), a
   ld a, reg(1)
   cpl
   ld reg(1), a
   inc reg
.endmacro

Is this even possible in Brass? And if it is, how do you tell Brass that reg is a string and reg(0) is the first character and reg(1) is the 2nd?
Any ideas?

King Harold · Post by **King Harold** » Sun 24 Jun, 2007 3:26 pm

Well TASM is a bitch with this sort of thing(good thing you werent using it), but maybe in Brass..
I asked Ben about that too but I don't think he knew what I meant..
I don't know of any other way than this and I'm not even sure whether it works:

Code: Select all

.deflong NEGATE(regH, regL)
 ld a,regH
 cpl
 ld regH,a
 ld a,regL
 cpl
 ld regL,a
 inc regH regL ;not 100% sure what to do with the whitespace..
.enddeflong

sadly there is no subtring..
You could also hardcode every case with {curly brackets} which means the macro with matching input will be chosen. http://benryves.com/bin/brass/directives/define.htm

driesguldolf · Post by **driesguldolf** » Sun 24 Jun, 2007 3:33 pm

hehe, thanks King Harold, it works

I tried to compile that along with a "inc hl" and "inc h l"
it gave "23" as opcode in both cases! yay

King Harold · Post by **King Harold** » Sun 24 Jun, 2007 3:48 pm

This will probably be completely different in the new Brass (any news on that?)

Post by **benryves** » Mon 25 Jun, 2007 10:08 am

Yes, a quirk in Brass is that it ignored spaces when assembling (apart from inside strings) so inc h l will work. The "new" Brass isn't current in development, sorry.

driesguldolf · Post by **driesguldolf** » Fri 29 Jun, 2007 7:19 pm

I know it's a bit off topic (King Harold made a topic about it somewhere...)
But ben there is a serious problem with the .align directive, as King Harold said they appear correctly in the listing file but not in the intel hex file...
Strange, I tried the alternative (.org ($ + $00FF) & $FF00) but that give EXACT the same result as .align, it probabely is the fault of the intel hex writer...

Post by **benryves** » Mon 02 Jul, 2007 10:12 am

It probably is a bug in the hex file writer, then. I'll look into it - sorry for the inconvenience!

Post by **tr1p1ea** » Mon 02 Jul, 2007 10:33 am

I think i told you about this problem too ben a while ago on IRC. Though i did manage to overcome it by placing the table i needed at the start of an APP page. Brass is still ftw btw

.