An implementation of IEEE-754 / SMT-LIB floating-point
SymFPU is an implementation of the SMT-LIB / IEEE-754 operations in terms of bit-vector operations. It is templated in terms of the bit-vectors, propositions, floating-point formats and rounding mode types used. This allow the same code to be executed as an arbitrary precision "SoftFloat" library (although it's performance would not be good) or to be used to build symbolic representation of floating-point operations suitable for use in "bit-blasting" SMT solvers (you could also generate circuits from them but again, performance will likely not be good).
Release | Stable | Testing |
---|---|---|
Fedora Rawhide | 0-0.18.20190517gitc3acaf6.fc42 | - |
Fedora 42 | 0-0.18.20190517gitc3acaf6.fc42 | - |
Fedora 41 | 0-0.17.20190517gitc3acaf6.fc41 | - |
Fedora 40 | 0-0.16.20190517gitc3acaf6.fc40 | - |
You can contact the maintainers of this package via email at
symfpu dash maintainers at fedoraproject dot org
.