[avr-libc-dev] fplib and MUL

[Peeter Vois]

Hi,

I have couple of ideas how to make fplib faster. Some headlines of the
methods to use are:

1) to use MUL commands for the CPU's that support it.

( the multiplication gets faster about 8 times).



2) to avoid division operations for log_2(x) calcultaion

( current algorithm has 10 multiplications and one division, it is
possible to get the same precision result with 9 multiplications ).




3) Instead of division with constant to use multiplication for example
by calculation of log_10(x),

( as multiplication is faster, the result comes faster)



Some more ideas will come as I do this optimisation. My question now is
that is someone doing something in this area? As first I will go and try
to make the multiplication and look if the device groups avr-gcc
entirely have or have not MUL instruction available.

Best Regards ...
Peeter