search

unzvfu / cuda-fixnum

Extended-precision modular arithmetic library that targets CUDA.
MIT License
34 stars 8 forks source link

Use recurrency shallowing to remove carry dependencies #49

Open unzvfu opened 4 years ago

unzvfu commented 4 years ago

See p17 of https://gmplib.org/~tege/speed.pdf