Olin Lathrop wrote:
> Brent Brown wrote:
> > Consider the method...
> >
> > Initialise once only in mainline code:
> > watchcat = 0xAA
> >
> > During ISR:
> > if watchcat = 0xAA
> >   watchcat = !watchcat (invert the value)
> >
> > At end of mainline code:
> > if watchcat = 0x55
> >   watchcat = !watchcat
> >   reset watchdog
> 
> This is rather a convoluted way of saying the interrupt routine sets a flag
> and the foreground code kicks the dog based on the flag.

Yes, that is a good concise description, thanks.

Don't miss the other point though - a boolean flag (single bit, 0/1) could be used but 
if the flag itself was randomly corrupted (software bug, ESD, EMI, whatever) there 
would only be a 50/50 chance of detection.

Instead here an 8 bit token is used. It is initialised only once, thereafter modified by 
each section of code in a prescribed fashion. Somewhat analogous to a game of 
Chinese whispers. Corruption of the token itself now has a 255/256 chance of 
detection.

-- 
Brent Brown, Electronic Design Solutions
16 English Street, St Andrews,
Hamilton 3200, New Zealand
Ph: +64 7 849 0069
Fax: +64 7 849 0071
Cell: +64 27 433 4069
eMail:  brent.brown@clear.net.nz


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