Phil Cross wrote:
> I must come clean. Having read all the replies, for which thanks, thought 
> about it, and tried a few things - macros DO work. I was in a muddle.
>   
.
I'm glad to hear they work -- I would have some programs that would be 
in a fix if they did not!!  :-)
.
> > My program was monolithic, and didn't use the linker. Then, when I 
> single-stepped through a program using MPLAB SIM and a macro was executed, 
> SIM showed a jump to the macro definition, and stepped through it. Hence I 
> could watch the execution of the macro by single-stepping.
>
> However, after modularising my program, and using the linker, SIM doesn't 
> step through the macro definitions - it just (apparently) jumps over the 
> macro instruction, although the macro is invisibly executed  This is why I 
> thought it wasn't working - I didn't see the execution of the macro when 
> single-stepping, as I did before I used the linker.
>
> So I guess my question now is why can't I single-step through a macro when I 
> use the linker

.
However, you STILL seem to be a little bit confused about macros!  The 
only thing that happens when you define a macro is that the assembler, 
when doing the first pass of assembling your program (there are one-pass 
assemblers, but most all of them are at least two-passes) , says:  "Ah 
ha!  I have a macro definition!  I am going to stick it in a special 
table somewhere, and whenever I come across  the place where this guy 
*uses* that macro on my next pass, then I will physically replace the 
place where he invokes the macro with the contents of this entry in the 
macro  table, replacing the parameter names with the arguments that he 
gives me."  The point being, the macro is not "called" -- the body of 
the macro physically replaces the line where it is invoked.  So, 
assuming you define the macro as  three lines of code, then your one 
invocation of the macro name will result in three lines of code being 
physically inserted there, and assembled, just as if you wrote it there 
in the first place.

What may possibly be happening, trying to figure out your difficulty, is 
that with most assemblers that I am familiar with (not sure about MPASM) 
you have the choice of either listing or not listing a macro expansion.  
That is, you have the choice of the output listing either showing the 
three lines of code (as per my previous example) or only showing the 
macro name.  Most people tend to not want to list macro expansions, but 
looking at the object code you will be able to see that the address is 
increment the proper number of instructions.

Hope this helps.  As someone else said, if you give a short example 
perhaps we can hone in on the exact difficulty better.

John
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist