FLINT3.7.0-dev

NMOD_VEC_SIMD_SCALAR_MAX_BITS

Synopsis#include <flint/nmod_vec.h>
NMOD_VEC_SIMD_SCALAR_MAX_BITS(add)source

The largest number of bits of nn for which _nmod_vec_scalar_mul_nmod (add = 0), _nmod_vec_scalar_addmul_nmod (add = 1) and their lazy variants run a vector kernel (Products by a scalar): 63 with AVX-512, 50 with AVX2, 50 and 32 with NEON, and 0 without vector kernels. Above it, and below NMOD_VEC_SIMD_MIN_LEN entries, they run their scalar code.

Cite this page

The FLINT team, “NMOD_VEC_SIMD_SCALAR_MAX_BITS (nmod_vec.h),” FLINT: Fast Library for Number Theory, documentation, version 3.7.0-dev, 2026. https://flintlib.org/doc/nmod_vec/NMOD_VEC_SIMD_SCALAR_MAX_BITS.html (generated 2026-11-20).