Floating point operations implemented in software way using only integers (32 bit).
- add
- sub
- mul
- div
- cos
- sin
Not very strict, just for education.
Negatives are supported. NAN and +=Inf are not implemented now.
Tests provided. May be useful for implementing FPU computaions for integer only processors.
The goal of the project is to write a C code for floating point fpu and then port it to Promela/SPIN language to be able to verify CPSs. See notes in http://spinroot.com/spin/Man/float.html. This project is to solve this.
Publication in SPIN 2019: https://link.springer.com/chapter/10.1007/978-3-030-30923-7_11