Alan B. Pearce wrote:
> I get the impression, Peter, that you are setting out to do exactly
> what I did using Olins development environment.

I noticed some things you might want to look into:

> .bank#v(gbank) udata

I've got a nicer way of handling this since you grabbed the snapshot this
code was based on.  See the new DEFRAM macro.

>  if gbank != 0
> .bank0   udata
>    endif

DEFRAM gets around this messiness.

>        clrf    status          ; select direct and indirect bank 0
>        dbankis 0
>        ibankis 0
>        ...
>        dbank   intr_led_reg
>        bcf     intr_led_pin    ; show serial activity LED

I assume you are using DBANK only since this is intended to be temporary
debug code.  Otherwise DBANKIF would be better since it probably wouldn't
emit any instructions.  The bank is already known to be 0, and the port
register is probably in bank 0.

>  if save_fsr                   ; FSR needs to be saved ?
>        movf    fsr, w          ; save FSR
>        movwf   fsr_save
>    endif

This is a problem since the LED code stomped on the bank.  This code assumes
the bank is 0 since it was meant to be right after the CLRF STATUS.  In
general it is best to put app code after the standard interrupt entry is
complete.

>  if ncodepages > 1             ; multiple code pages may be in use ?
>        movf    pclath, w       ; save PCLATH
>        movwf   pclath_save
>        clrf    pclath
>    endif

Same bank dependency on PCLATH_SAVE here.

>        dbank   pir1            ; check for UART receive interrupt

Why DBANK and not DBANKIF?  I don't see any reason the assembler lost track
of the bank.  Anyway, I think it's better to state that separately with an
UNBANK, then have code that requires a particular bank use DBANKIF.  That
allows code snippets to be moved around without problems and redundant bank
setting cycles.

>        nop                     ; nop for ICD to stop on

Yeah, I wish the ICD didn't have that problem either.  However, I've been
surrounding these NOPs with IF DEBUG.  I can set one symbol in the main
include file to include or remove the ICD2 debug NOPs.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist