[libre-riscv-dev] [Bug 127] Transcendentals needed (SIN/COS/ATAN2/EXP/LOG/POW etc.)

[bugzilla-daemon at libre-riscv.org]

http://bugs.libre-riscv.org/show_bug.cgi?id=127

--- Comment #13 from Luke Kenneth Casson Leighton <lkcl at lkcl.net> ---
so CORDIC looks like it is the highest bang-per-buck algorithm that
will cover a large range of these.

initial thoughts: it's very similar to what's in div_core.py
however on second thoughts, CORDIC doesn't really lend itself
to higher RADIX so would need a separate pipeline anyway.
or... it does...
https://www.researchgate.net/publication/256970111_VLSI_architecture_for_parallel_radix-4_CORDIC/link/59ef33d7458515ec0c7b51a8/download

but *sigh* as usual it'll require "translation" of the paper...
or contating the authors.

-- 
You are receiving this mail because:
You are on the CC list for the bug.