FLINT3.7.0-dev
/

_d1_vec_exp, _d2_vec_exp, _d3_vec_exp, _d4_vec_exp, _d1_vec_sin, _d2_vec_sin, _d3_vec_sin, _d4_vec_sin, _d1_vec_cos, _d2_vec_cos, _d3_vec_cos, _d4_vec_cos, _d1_vec_sin_cos, _d2_vec_sin_cos, _d3_vec_sin_cos, _d4_vec_sin_cos, _d1_vec_log, _d2_vec_log, _d3_vec_log, _d4_vec_log, _d1_vec_atan, _d2_vec_atan, _d3_vec_atan, _d4_vec_atan

Synopsis#include <flint/dfloat.h>
void _d1_vec_exp(d1_ptr res, d1_srcptr x, slong len)
void _d2_vec_exp(d2_ptr res, d2_srcptr x, slong len)
void _d3_vec_exp(d3_ptr res, d3_srcptr x, slong len)
void _d4_vec_exp(d4_ptr res, d4_srcptr x, slong len)
void _d1_vec_sin(d1_ptr res, d1_srcptr x, slong len)
void _d2_vec_sin(d2_ptr res, d2_srcptr x, slong len)
void _d3_vec_sin(d3_ptr res, d3_srcptr x, slong len)
void _d4_vec_sin(d4_ptr res, d4_srcptr x, slong len)
void _d1_vec_cos(d1_ptr res, d1_srcptr x, slong len)
void _d2_vec_cos(d2_ptr res, d2_srcptr x, slong len)
void _d3_vec_cos(d3_ptr res, d3_srcptr x, slong len)
void _d4_vec_cos(d4_ptr res, d4_srcptr x, slong len)
void _d1_vec_sin_cos(d1_ptr sres, d1_ptr cres, d1_srcptr x, slong len)
void _d2_vec_sin_cos(d2_ptr sres, d2_ptr cres, d2_srcptr x, slong len)
void _d3_vec_sin_cos(d3_ptr sres, d3_ptr cres, d3_srcptr x, slong len)
void _d4_vec_sin_cos(d4_ptr sres, d4_ptr cres, d4_srcptr x, slong len)
void _d1_vec_log(d1_ptr res, d1_srcptr x, slong len)
void _d2_vec_log(d2_ptr res, d2_srcptr x, slong len)
void _d3_vec_log(d3_ptr res, d3_srcptr x, slong len)
void _d4_vec_log(d4_ptr res, d4_srcptr x, slong len)
void _d1_vec_atan(d1_ptr res, d1_srcptr x, slong len)
void _d2_vec_atan(d2_ptr res, d2_srcptr x, slong len)
void _d3_vec_atan(d3_ptr res, d3_srcptr x, slong len)
void _d4_vec_atan(d4_ptr res, d4_srcptr x, slong len)

Elementwise elementary functions with the same static error bounds as the scalar kernels, which they run on four arguments at a time (3 to 5 times faster per element than scalar calls). Elements that hit a special case are redone with the scalar code.

Cite this page

The FLINT team, “_d1_vec_exp, _d2_vec_exp, _d3_vec_exp, _d4_vec_exp, _d1_vec_sin, _d2_vec_sin, _d3_vec_sin, _d4_vec_sin, _d1_vec_cos, _d2_vec_cos, _d3_vec_cos, _d4_vec_cos, _d1_vec_sin_cos, _d2_vec_sin_cos, _d3_vec_sin_cos, _d4_vec_sin_cos, _d1_vec_log, _d2_vec_log, _d3_vec_log, _d4_vec_log, _d1_vec_atan, _d2_vec_atan, _d3_vec_atan, _d4_vec_atan (dfloat.h),” FLINT: Fast Library for Number Theory, documentation, version 3.7.0-dev, 2026. https://flintlib.org/doc/dfloat/_d1_vec_exp.html (generated 2026-10-20).