markp@cannontech.com writes:
 > My question is about bank selection using the STATUS RPO/RP1 bits vs. the
 > BANKSEL command.  When I need to transition between banks, I simply change
 > the RP0/RP1 bits accordingly before and after the transition.  However, I
 > see many examples of code in the archives that use the BANKSEL command.  I
 > do not see this command in the spec sheet, well for the 16C773 & 16C715 I
 > use anyway.  Is it applicable to all PICs or just certain ones?  What are
 > the advantages of its use over the RP0/RP1 select method?

	The primary advantage of BANKSEL is that you don't have to
keep track, when writing code, of in which bank a variable (or SPFR)
is allocated.  This can be especially useful with relocatable code
that lets the linker assign variable locations, in which case you
don't, in general, know the bank of a variable at program writing
time.

	The disadvantage is that it always generates two extra
instructions, adjusting both RP0 and RP1 (for the processors that have
those bits), whether what you actually need is two, one, or no extra
instructions.  This can significantly expand code size and slow
execution.

	If you *know* the banks of your references as well as the
current settings of the bank select bits at the time that you are
writing the code, and you're beyond the beginner stage in which you
don't want to think about banks, then set and clear the RP bits
yourself.

	If you're writing a module that will be relocatably linked
into several projects, and you don't want to fix up bank selection
each time you use it, and you can't restrict where the using project
places your variables, use BANKSEL.

(I'm sorry if my anti-BANKSEL bias comes through to strongly.)

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