Hacker News new | past | comments | ask | show | jobs | submit login

Please check RLIBM project, it's building a collection of correctly rounded elementary functions for multiple representations ( 32-bit float, posits, bfloat16, tensorfloat32) for multiple rounding modes.

RLIBM: Rutgers Architecture and Programming Languages Lab's Correctly Rounded Libm:

https://people.cs.rutgers.edu/~sn349/rlibm/




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: