I have made a small utility that I run just before the assembler
(sort of a preprocessor) which does just what You want.

The utility which I call PICBIT (I know there is something else also
called PICBIT) scans the source file for my special declarations and
then produces a file which can be included in the sourcefile. It also
scans other includefiles (even includefiles in includefiles) in the
source and handles them correctly.

Here is the special declaration in my main sourcefile or any
includefile included in this:

;@BITNAME   REGISTER   BIT

Which creates the following lines in the produced includefile:

b_BITNAME   EQU BIT
m_BITNAME   EQU 1<<BIT
#define r_BITNAME       REGISTER
#define f_BITNAME       REGISTER,BIT

I can then use the b_, m_, r_ or f_ prefix in my code and get it to
point to the right bit and/or register. If I sometime during my
developmnet change one bit (either hardware or an internal flag bit)
I only have to worry about changing it at one place.


> G'day to all.
>
> I am reworking some code and would like to improve some of my register
> handling.
>
> I currently use #define statements where ever possible to identify
> bits.  They have the great benefit that I don't have to keep track of
> which bits belong with which registers and have eliminated many hours
> of tracking down stupid mistakes.
>
> Some while back, I came up with a series of macros that generate masks
> based upon those bit defines.  Again, they help minimize my mistakes.
>
> Those macros take the form of:
> movbw           MACRO   reg,bit         ;load w with selected bit
>      movlw       b'00000001' << bit
>      endm                                ;Z, C & DC unchanged
>
> Typical usage is:
>
>      #define     _TICK  AD1FLG,4        ;defined way up near the front
>      of the co
> de
>
>      movfw       AD1FLG          ;TMR0 tick bit different from FLAGS
>      tick bit? xorwf       TMR0,W          ; andbw       _TICK
>        ;mask all but bit of interest bz          WdogLoop
>      ;same? movbw       _TICK xorwf       AD1FLG,F        ;flip FLAGS
>      tick bit for next time thru
>
> The chunk of code above illustrates what my problem is now.  The bit
> defines themselves work just fine, but I want the register that the
> bit belongs to also to be automated.  In other words, if I manipulate
> a bit and then stick it back into its register, I want the code to be
> able figure out which register it belongs to.
>
> I could do it the same way as I did the bits (create macros for every
> usable instruction) but there just has to be a better way.
>
> One of my attempts looks like:
> bitreg          MACRO   reg,bit,dest    ;determine register belonging
> to bit
>      reg,dest
>      endm
>
> with usage as follows:
>      movbw       _DISPF2         ;generate flash at 1/2 packet rate
>      btfsc       _FLASH          ;
>        xorwf     bitreg,_DISPF2,F
>
> and resulting in the following messages:
> Error[108]   E:\PIC\PLC\CIS3P0A.ASM 3383 : Illegal character (,)
> Warning[225] E:\PIC\PLC\CIS3P0A.ASM 3383 : Invalid label in operand
> (bitreg)
>
> That little 3 line snippet above is supposed to assemble as:
>      movlw       b'01000000'
>      btfsc       DISPCNT,7
>        xorwf     FLAG,F
>
> where _DISPF2 is bit 6 of register FLAG
>
> Any thoughts?  Ideas?
>
> dwayne
>
>
> Dwayne Reid   <dwayner@planet.eon.net>
> Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
> (780) 489-3199 voice          (780) 487-6397 fax
>
> Celebrating 16 years of Engineering Innovation (1984 - 2000)
>
> * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
> Do NOT send unsolicited commercial email to this email address.
> This message neither grants consent to receive unsolicited
> commercial email nor is intended to solicit commercial email.


==============================
Ruben Jvnsson
AB Liros Elektronik
Box 9124, 200 39 Malmv, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
ruben@pp.sbbs.se
==============================