search

cornell-c2s2 / c2s2_ip

A general repository for all of C2S2's IP, including testing
2 stars 1 forks source link

Floating Point Arithmetic Blocks #28

Open jjm469 opened 1 year ago

jjm469 commented 1 year ago

ADD DESCRIPTION

UnsignedByte commented 1 year ago

We would like to have a thoroughly tested (possibly using berkeley's hardfloat module) floating point library for at least the basic operations of +, -, *, and /. There are some helpful tutorials here that could help with implementation.

I would recommend starting with multiplication, as it is the easiest to implement in floating point (with division being pretty similar). I would also recommend first making combinational versions (possibly in a CombFloat IP) before looking into pipelined and latency insensitive designs.