JM> executing prior to interrupt.  An interrupt places information in the fifo 
JM> (or queue, or pipe...) and a slow-speed function pulls it out when it is 
JM> required.  The problem I am having is:  What happens if I get interrupted 
JM> in a place where I am adjusting the size of the fifo?  I am unable to 
JM> disable interrupts at this time as I cannot afford to miss a single 
JM> interrupt.

Disabling interrupts doesn't mean you LOSE the interrupts, it means that
they queue up waiting for interrupts to clear.  Can't you afford a split
microsecond delay?  Failing that, make your queue circular, and have
separate read + write pointers, and assume that you never overflow.  These
are just suggestions.  Besides writing a comms program, I have no real time
experience.  BFN.  Paul. 
@EOT:

---