I have modified my code in line with the suggestions regarding
relocatable code and using UDATA and let the compiler take care of the
allocation.

I think I am missing the point as it has become more difficult for me
to code.  This may be because I am used to the old approach.

Apart from loosing the ability to access 2 x 80 byte blocks of ram as
single 160 byte block of data by simply changing some bits in the
status register as I detailed in a previous post.

I now find that I must use banksel directives everywhere because I do
not know if the compiler has put the GPR I am referencing in the same
bank as a special purpose register I am referencing.

For example

	banksel		PORTB
	movfw		PORTB
	banksel		shadow
	movwf		shadow
	bsf			shadow,OVEN		;Curently ON at start should be off
	bsf			shadow,TC_CS       	;This starts conversions on max chip as
chip is not selected

	movfw		shadow
	banksel		PORTB
	movwf		PORTB


Before, I rarely needed a banksel directive, but now I found that I
need to use them whenever I am referencing any register.

I guess if I resign myself to this fact then I can move on.  I do
remember reading somewhere that PIC chips can release significant
speed performances as almost every instruction is executed in one
program clock cycle.  But if I have to use banksel everwhere then not
so efficient I think.  Perhaps I need to look at AVRs

I am doing something wrong and missing the point.

Cheers Justin
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.