gnupic: Thread: gpasm question

    Hi,

while I was trying to add BANKSEL directives in SDCC I came
across the following question:

shouldn't function do_bankisel() at gpasm/directive.c contain
an extra
    && (state.device.class != PROC_CLASS_PIC16E)

to support BANKISEL for the 18F family?


Regards,
Vangelis Rokas

Doing that would allow gpasm to use bankisel for 18xx devices.  The FSRs
on 18xx devices are 12 bits wide (in two bytes) so a complete data
memory address can be stored.  So, I don't believe a bankisel is
required.

I will check on this.  I could be wrong.

MPASM 03.20.02 Released      BANKIS~4.ASM   1-18-2003  21:09:01        
PAGE  1


LOC  OBJECT CODE     LINE SOURCE TEXT
  VALUE

<snip>

000000 0E00           00022   movlw reg4
000002 6E00           00023   movwf FSR0
Message[312]: Page or Bank selection not needed for this device.  No
code generated.
                      00024   bankisel reg4
000004 6EEF           00025   movwf INDF0
                      00026 
000006 0E60           00027   movlw reg2
000008 6E00           00028   movwf FSR0
Message[312]: Page or Bank selection not needed for this device.  No
code generated.
                      00029   bankisel reg2
00000A 6EEF           00030   movwf INDF0

<snip>


On Fri, 2003-06-06 at 15:32, Vangelis Rokas wrote:
>     Hi,
> 
> while I was trying to add BANKSEL directives in SDCC I came
> across the following question:
> 
> shouldn't function do_bankisel() at gpasm/directive.c contain
> an extra
>     && (state.device.class != PROC_CLASS_PIC16E)
> 
> to support BANKISEL for the 18F family?
> 
> 
> Regards,
> Vangelis Rokas
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: ####@####.####
> For additional commands, e-mail: ####@####.####
> 

Hi again,
i need a better preprocessor/macro handling for gpasm.
I'm willing to implement it and for this, i need some
advice from the community how exactly it should be implemented.
Can you please help for this issue .
Thanks in advance
Chris