Extended-Precision Fixed-Point?

[Jim Osborn]

Could someone point me toward references, or share experiences in
performing multiplication and addition of signed extended-precision
fixed-point numbers? The integer part is never more than 16 bits,
but the fractional part can be arbitrarily long.

The obvious approach would seem to be to use whatever registers the
machine provides and work a register at a time in the local assembly
language. This is non-portable, but given the ease of propagating
carry bits, it's probably the speediest. I could imagine there's
some clever scheme that might take advantage of the local FPU, but I
can't seem to think of one. Any suggestions?

Is there advantage to be taken of the fact that some of the
multiplication is really squaring?

Sorry if this is FAQ.
--
pilchuck!jimo@phred Jim Osborn, Physio Control Corp
11811 Willows Rd, Redmond, WA, 98073
206-867-4704 direct to my desk