[libre-riscv-dev] [Bug 132] SIMD-like nmigen signal for partitioning

[bugzilla-daemon at libre-riscv.org]

http://bugs.libre-riscv.org/show_bug.cgi?id=132

--- Comment #36 from Luke Kenneth Casson Leighton <lkcl at lkcl.net> ---
ok so except for fcvt and fclass, the removal of m.If/Elif/Else has been
done, replacing with Mux.  specialcases are exceptionally annoyingly weirdly
encoded, now: none of them are particularly obvious.  i used a cumulative
(unbalanced) mux-tree.

fpbase is going to become quite weird: it's where all the magic constants
exist, and where things like NaN and Inf, and denormalisation are detected.
i may just write a partitioned version of FPNumBaseRecord that has the
same API as FPNumBaseRecord.

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