itmr rollover in delay loop

[Philipp Rumpf]

> >>       while(((cr16+loops)-mfctl(16))>0);

>         printf("0x%x  0x%x  0x%x  %s\n", a, b, (b-a), (0 < (int)(b-a)) ? "y" : "n");
>         b++;

Note that Frank is modifying b, and evaluating  0 < (int)(b-a), while in
the original example, the modified term was subtracted from the constant
(for one call) one, so it would have been (int)(a-b) > 0 using Frank's
variable names.  This explains why Frank is still seeing "n" for a == b
while I am already seeing false.

	Philipp Rumpf