_dfloat_ball_mul_scaled, _dfloat_ball_div, _dfloat_ball_rsqrt, _dfloat_ball_sqrt
Synopsis
#include <flint/dfloat.h>void _dfloat_ball_mul_scaled(double * res, double * rrad, int n, const double * x, double rx, const double * y, double ry)sourcevoid _dfloat_ball_div(double * res, double * rrad, int n, const double * x, double rx, const double * y, double ry)sourceint _dfloat_ball_rsqrt(double * res, double * rrad, int n, const double * x, double rx)sourceint _dfloat_ball_sqrt(double * res, double * rrad, int n, const double * x, double rx)sourceRigorous product, quotient, reciprocal square root and square root of balls with midpoints of n components and radii rx, ry, for inputs of any magnitude, by scaling the heads to [1, 2), applying the fixed-N kernels and scaling back. Components that would become subnormal are flushed into the radius. The product is the slow path of dNb_mul; the others are the only implementations.
Cite this page
The FLINT team, “_dfloat_ball_mul_scaled, _dfloat_ball_div, _dfloat_ball_rsqrt, _dfloat_ball_sqrt (dfloat.h),” FLINT: Fast Library for Number Theory, documentation, version 3.7.0-dev, 2026. https://flintlib.org/doc/dfloat/_dfloat_ball_mul_scaled.html (generated 2026-10-20).
@misc{flint-doc-dfloat-dfloat-ball-mul-scaled,
author = {{The FLINT team}},
title = {{\_dfloat\_ball\_mul\_scaled, \_dfloat\_ball\_div, \_dfloat\_ball\_rsqrt, \_dfloat\_ball\_sqrt (dfloat.h)}},
howpublished = {FLINT: Fast Library for Number Theory, documentation, version 3.7.0-dev},
year = {2026},
url = {https://flintlib.org/doc/dfloat/_dfloat_ball_mul_scaled.html},
note = {Generated 2026-10-20}
}