LMMP 4.3.0
Lamina High-Precision Arithmetic Library
载入中...
搜索中...
未找到
mul_toom42.c 文件参考
#include "impl/mul_hard.h"
#include "mparam.h"
#include "lmmpn/toom_interp.h"
#include "lmmp/lmmpn.h"
mul_toom42.c 的引用(Include)关系图:

浏览该文件的源代码.

结构体

struct  toom42_cache_t

宏定义

#define _bm1   (cache->_bm1)
#define _bm1   (cache->_bm1)
#define _bp1   (cache->_bp1)
#define _bp1   (cache->_bp1)
#define a0   numa
#define a0   numa
#define a0   numa
#define a1   (numa + n)
#define a1   (numa + n)
#define a1   (numa + n)
#define a13   (tp + n + 1)
#define a13   (tp + n + 1)
#define a13   bp1
#define a2   (numa + 2 * n)
#define a2   (numa + 2 * n)
#define a2   (numa + 2 * n)
#define a3   (numa + 3 * n)
#define a3   (numa + 3 * n)
#define a3   (numa + 3 * n)
#define am1   (dst + n)
#define am1   (dst + n)
#define am1   (dst + n)
#define ap1   tp
#define ap1   tp
#define ap1   tp
#define ap2   ap1
#define ap2   ap1
#define ap2   ap1
#define b0   numb
#define b0   numb
#define b0   numb
#define b1   (numb + n)
#define b1   (numb + n)
#define b1   (numb + n)
#define bm1   _bm1
#define bm1   _bm1
#define bm1   dst
#define bp1   _bp1
#define bp1   _bp1
#define bp1   (tp + n + 1)
#define bp2   (tp + n + 1)
#define bp2   (tp + n + 1)
#define bp2   bp1
#define lmmp_mul_n_(dst, numa, numb, n)
 Copyright (C) 2026 HJimmyK (Jericho Knox)
#define n   (cache->n)
#define n   (cache->n)
#define numb   (cache->numb)
#define numb   (cache->numb)
#define s   (cache->s)
#define s   (cache->s)
#define t   (cache->t)
#define t   (cache->t)
#define tp   (cache->tp)
#define tp   (cache->tp)
#define v0   dst
#define v0   dst
#define v0   dst
#define v1   (dst + 2 * n)
#define v1   (dst + 2 * n)
#define v1   (dst + 2 * n)
#define v2   (tp + 2 * n + 2)
#define v2   (tp + 2 * n + 2)
#define v2   (tp + 2 * n + 2)
#define vinf   (dst + 4 * n)
#define vinf   (dst + 4 * n)
#define vinf   (dst + 4 * n)
#define vm1   tp
#define vm1   tp
#define vm1   tp

函数

void lmmp_mul_toom42_ (mp_ptr restrict dst, mp_srcptr restrict numa, mp_size_t na, mp_srcptr restrict numb, mp_size_t nb)
static void lmmp_mul_toom42_cache_ (mp_ptr restrict dst, mp_srcptr restrict numa, const toom42_cache_t *cache, int flag)
static int lmmp_mul_toom42_cache_init_ (mp_ptr restrict dst, mp_srcptr restrict numa, toom42_cache_t *cache)
void lmmp_mul_toom42_unbalance_ (mp_ptr restrict dst, mp_srcptr restrict numa, mp_size_t na, mp_srcptr restrict numb, mp_size_t nb)

结构体说明

◆ toom42_cache_t

struct toom42_cache_t

在文件 mul_toom42.c 第 177 行定义.

toom42_cache_t 的协作图:
成员变量
mp_ptr restrict _bm1
mp_ptr restrict _bp1
mp_size_t n
mp_srcptr restrict numb
mp_size_t s
mp_size_t t
mp_ptr restrict tp

宏定义说明

◆ _bm1 [1/2]

#define _bm1   (cache->_bm1)

◆ _bm1 [2/2]

#define _bm1   (cache->_bm1)

◆ _bp1 [1/2]

#define _bp1   (cache->_bp1)

◆ _bp1 [2/2]

#define _bp1   (cache->_bp1)

◆ a0 [1/3]

#define a0   numa

◆ a0 [2/3]

#define a0   numa

◆ a0 [3/3]

#define a0   numa

◆ a1 [1/3]

#define a1   (numa + n)

◆ a1 [2/3]

#define a1   (numa + n)

◆ a1 [3/3]

#define a1   (numa + n)

◆ a13 [1/3]

#define a13   (tp + n + 1)

◆ a13 [2/3]

#define a13   (tp + n + 1)

◆ a13 [3/3]

#define a13   bp1

◆ a2 [1/3]

#define a2   (numa + 2 * n)

◆ a2 [2/3]

#define a2   (numa + 2 * n)

◆ a2 [3/3]

#define a2   (numa + 2 * n)

◆ a3 [1/3]

#define a3   (numa + 3 * n)

◆ a3 [2/3]

#define a3   (numa + 3 * n)

◆ a3 [3/3]

◆ am1 [1/3]

#define am1   (dst + n)

◆ am1 [2/3]

#define am1   (dst + n)

◆ am1 [3/3]

#define am1   (dst + n)

◆ ap1 [1/3]

#define ap1   tp

◆ ap1 [2/3]

#define ap1   tp

◆ ap1 [3/3]

#define ap1   tp

◆ ap2 [1/3]

#define ap2   ap1

◆ ap2 [2/3]

#define ap2   ap1

◆ ap2 [3/3]

#define ap2   ap1

◆ b0 [1/3]

#define b0   numb

◆ b0 [2/3]

#define b0   numb

◆ b0 [3/3]

#define b0   numb

◆ b1 [1/3]

#define b1   (numb + n)

◆ b1 [2/3]

#define b1   (numb + n)

◆ b1 [3/3]

#define b1   (numb + n)

◆ bm1 [1/3]

#define bm1   _bm1

◆ bm1 [2/3]

#define bm1   _bm1

◆ bm1 [3/3]

#define bm1   dst

◆ bp1 [1/3]

#define bp1   _bp1

◆ bp1 [2/3]

#define bp1   _bp1

◆ bp1 [3/3]

#define bp1   (tp + n + 1)

◆ bp2 [1/3]

#define bp2   (tp + n + 1)

◆ bp2 [2/3]

#define bp2   (tp + n + 1)

◆ bp2 [3/3]

#define bp2   bp1

◆ lmmp_mul_n_

#define lmmp_mul_n_ ( dst,
numa,
numb,
n )
值:
lmmp_mul_hard_n_((dst), (numa), (numb), (n)); \
else if ((n) < MUL_TOOM33_THRESHOLD) \
lmmp_mul_toom22_((dst), (numa), (n), (numb), (n)); \
else \
lmmp_mul_toom33_((dst), (numa), (n), (numb), (n))
#define MUL_TOOM22_THRESHOLD
#define MUL_TOOM33_THRESHOLD
#define numb
#define n

Copyright (C) 2026 HJimmyK (Jericho Knox)

This file is part of LMMP.

LMMP is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 3 of the License, or (at your option) any later version.

This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.

You should have received a copy of the GNU Lesser General Public License along with this library. If not, see https://www.gnu.org/licenses/.

SPDX-License-Identifier: LGPL-3.0-or-later

在文件 mul_toom42.c 第 31 行定义.

◆ n [1/2]

#define n   (cache->n)

◆ n [2/2]

#define n   (cache->n)

被这些函数引用 __lmmp_leading_zeros_(), __lmmp_mul_hard_n_(), __lmmp_mul_n_(), __lmmp_sqr_hard_n_(), __lmmp_tailing_zeros_(), _odd_nPr_(), add_small_inplace(), alloc_limbs(), alloc_limbs(), bench_add_n(), bench_add_nc(), bench_addmul_1(), bench_addshl1_n(), bench_aggregate(), bench_div_1(), bench_div_1_s(), bench_div_2(), bench_div_2_s(), bench_factorial(), bench_fib(), bench_gcd_hgcd(), bench_gcd_lehmer(), bench_is_prime_n(), bench_is_prime_n_notrial(), bench_is_strong_lucas(), bench_mod_1(), bench_mod_2(), bench_mul_1(), bench_mul_mersenne_dc(), bench_mul_mersenne_fft(), bench_mul_n(), bench_mullo(), bench_mullo_basecase(), bench_mullo_dc(), bench_mullo_fft(), bench_not(), bench_powlo(), bench_powmod_even(), bench_powmod_odd(), bench_shl(), bench_shlnot(), bench_shr(), bench_shr1add_n(), bench_shr1sub_n(), bench_sqr(), bench_sqr_basecase(), bench_sqrt(), bench_sub_n(), bench_sub_nc(), bench_submul_1(), bench_subshl1_n(), binvert_mulhi_(), cbrt_floor_ok(), check_addmul_2(), check_decimal_one(), check_hex_one(), check_mul_n(), check_value(), count_2facodd_factors(), count_alloc(), count_bits(), count_bits(), count_bits(), count_bits(), count_factors(), count_factors(), count_factors(), count_factors(), count_hyperfac_factors(), count_realloc(), count_superfac_factors(), fac_size_bigger(), fac_size_lower(), factor_size_int(), factor_size_int(), factorial_buffer(), fib2_core_(), fill_deterministic(), fill_odd_mod(), fill_pattern(), fill_pattern(), fill_random(), ipn_mont_half_(), is2_notrial_i_(), is_prime_small(), is_square_2(), jacobi_D(), k2t_limbs(), limbs_norm(), limbs_same(), limbs_same(), ll_small_is_prime_mr(), ll_small_preinv(), lmmp_10pow_1_(), lmmp_12pow_1_(), lmmp_14pow_1_(), lmmp_2factorial_(), lmmp_2factorial_(), lmmp_2factorial_size_(), lmmp_2factorial_size_(), lmmp_6pow_1_(), lmmp_add_n_(), lmmp_add_n_sub_n_(), lmmp_add_nc_(), lmmp_addmul_1_(), lmmp_addmul_2_(), lmmp_addshl1_n_(), lmmp_arith_seqprod_(), lmmp_arith_seqprod_(), lmmp_arith_seqprod_size_(), lmmp_b36_to_chars_(), lmmp_binvert_(), lmmp_binvert_(), lmmp_binvert_n_dc_(), lmmp_binvert_n_dc_(), lmmp_binvert_unbalance_(), lmmp_binvert_unbalance_(), lmmp_binvert_unbalance_1_(), lmmp_binvert_unbalance_1_(), lmmp_binvert_unbalance_2_(), lmmp_binvert_unbalance_2_(), lmmp_cbrt_6_est_(), lmmp_cbrt_chebyshev_(), lmmp_cbrt_estimate(), lmmp_cbrt_shapply_(), lmmp_cbrt_ulong_(), lmmp_cmp_(), lmmp_dec_to_chars_(), lmmp_div3_(), lmmp_div_divide_n_(), lmmp_elem_mul_ulong_(), lmmp_elem_mul_ulong_(), lmmp_extract_bits_(), lmmp_extract_bits_(), lmmp_factorial_(), lmmp_factorial_(), lmmp_factorial_size_(), lmmp_factorial_size_(), lmmp_fft_best_k_(), lmmp_fft_next_size_(), lmmp_fibonacci2_(), lmmp_fibonacci_(), lmmp_fibonacci_size_(), lmmp_gcd_norm_(), lmmp_gcdext_hgcd_(), lmmp_gcdext_smag_(), lmmp_hgcd_(), lmmp_hgcd_adjust_(), lmmp_hgcd_apply_mod_(), lmmp_hgcd_core_(), lmmp_hgcd_lehmer2_(), lmmp_hgcd_salloc_(), lmmp_hgcd_scratch_size_(), lmmp_hgcd_step_(), lmmp_huff_tree_build_(), lmmp_hyperfac_(), lmmp_hyperfac_(), lmmp_hyperfac_size_(), lmmp_hyperfac_size_(), lmmp_invcbrt_newton_(), lmmp_invsqrt_newton_(), lmmp_is_prime_uint_(), lmmp_is_prime_uint_notrial_(), lmmp_is_prime_ulong_(), lmmp_is_prime_ulong_notrial_(), lmmp_is_prothprime_(), lmmp_leading_zeros_(), lmmp_mersenne_next_size_(), lmmp_mod_2p48sub1_(), lmmp_mont_fold_hi_(), lmmp_mont_init_(), lmmp_mont_need_(), lmmp_mont_redc_(), lmmp_mont_redc_bc_(), lmmp_mont_redcify_(), lmmp_mont_sp_init_(), lmmp_mont_sp_redc_(), lmmp_mul_fermat_(), lmmp_mul_fermat_recombine_(), lmmp_mul_fermat_recurse_(), lmmp_mul_mersenne_fft_(), lmmp_mul_n_(), lmmp_mul_n_(), lmmp_mul_toom22_(), lmmp_mul_toom32_(), lmmp_mul_toom33_(), lmmp_mul_toom42_(), lmmp_mul_toom42_cache_(), lmmp_mul_toom42_cache_init_(), lmmp_mul_toom43_(), lmmp_mul_toom44_(), lmmp_mul_toom52_(), lmmp_mul_toom53_(), lmmp_mul_toom62_(), lmmp_mul_toom62_cache_(), lmmp_mul_toom62_cache_init_(), lmmp_mul_toom62_unbalance_(), lmmp_mul_toom6h_(), lmmp_mul_toom8h_(), lmmp_mullh_(), lmmp_mullo_(), lmmp_mullo_(), lmmp_mullo_basecase_(), lmmp_mullo_basecase_(), lmmp_mullo_crt_(), lmmp_mullo_dc_(), lmmp_mullo_dc_(), lmmp_mullo_fft_(), lmmp_mullo_fft_cache_init_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_multinomial_(), lmmp_multinomial_(), lmmp_multinomial_size_(), lmmp_multinomial_size_(), lmmp_nCr_(), lmmp_nCr_(), lmmp_nCr_size_(), lmmp_nCr_size_(), lmmp_next_prime_ulong_(), lmmp_nPr_(), lmmp_nPr_(), lmmp_nPr_size_(), lmmp_nPr_size_(), lmmp_nthroot_ulong_(), lmmp_odd_2factorial_uint_(), lmmp_odd_2factorial_ushort_(), lmmp_odd_factorial_(), lmmp_odd_factorial_uint_(), lmmp_odd_factorial_uint_(), lmmp_odd_multinomial_uint_(), lmmp_odd_multinomial_ushort_(), lmmp_odd_nCr_uint_(), lmmp_odd_nCr_uint_(), lmmp_odd_nCr_ushort_(), lmmp_odd_nCr_ushort_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_uint_(), lmmp_odd_nPr_uint_(), lmmp_odd_nPr_ulong_(), lmmp_odd_nPr_ulong_(), lmmp_odd_nPr_ushort_(), lmmp_odd_nPr_ushort_(), lmmp_perfsqr_(), lmmp_perfsqr_filter_(), lmmp_pow_(), lmmp_pow_(), lmmp_pow_basecase_(), lmmp_pow_basecase_(), lmmp_pow_size_(), lmmp_pow_win2_(), lmmp_pow_win2_(), lmmp_powlo_(), lmmp_powlo_(), lmmp_powmod_(), lmmp_powmod_(), lmmp_powmod_2_(), lmmp_powmod_odd_(), lmmp_powmod_odd_(), lmmp_powmod_odd_mont_(), lmmp_powmod_odd_mont_need_(), lmmp_prev_prime_ulong_(), lmmp_prime_cache_init_(), lmmp_prime_cnt16_(), lmmp_prime_int_table_init_(), lmmp_prime_size_(), lmmp_primefac_(), lmmp_primefac_(), lmmp_primefac_size_(), lmmp_random_(), lmmp_random_(), lmmp_redc_(), lmmp_redc_(), lmmp_seed_random_(), lmmp_seed_random_(), lmmp_shlnot_(), lmmp_shlnot_(), lmmp_shr1add_n_(), lmmp_shr1add_nc_(), lmmp_shr1sub_n_(), lmmp_shr1sub_nc_(), lmmp_shr64_(), lmmp_shr64_(), lmmp_sqr_fermat_(), lmmp_sqr_fermat_recurse_(), lmmp_sqr_mersenne_fft_(), lmmp_sqr_toom2_(), lmmp_sqr_toom3_(), lmmp_sqr_toom4_(), lmmp_sqr_toom6_(), lmmp_sqr_toom8_(), lmmp_sqrlo_basecase_(), lmmp_sqrlo_basecase_(), lmmp_sqrlo_dc_(), lmmp_sqrlo_dc_(), lmmp_sqrlo_n_(), lmmp_strong_random_(), lmmp_strong_random_(), lmmp_sub_n_(), lmmp_sub_nc_(), lmmp_submul_1_(), lmmp_subshl1_n_(), lmmp_superfac_(), lmmp_superfac_(), lmmp_superfac_size_(), lmmp_superfac_size_(), lmmp_tailing_zeros_(), lmmp_to_str_basecase_(), lmmp_toom_couple_handling_(), lmmp_toom_eval_dgr3_pm1_(), lmmp_toom_eval_dgr3_pm2_(), lmmp_toom_eval_pm1_(), lmmp_toom_eval_pm2_(), lmmp_toom_eval_pm2exp_(), lmmp_toom_eval_pm2rexp_(), lmmp_toom_interp12_(), lmmp_toom_interp16_(), lmmp_toom_interp5_(), lmmp_toom_interp6_(), lmmp_toom_interp7_(), lmmp_zero_q_(), main(), make_odd_mod(), matrix_equal(), mersenne_limbs(), mod32(), mont2_dbl(), mont2_lucas_strong_(), mont2_mul_2(), mont2_mul_2(), mont2_neg(), mont2_powmod_2_(), mont2_preinv(), mont2_redc(), mont2_redc(), mont2_redc_mul_1(), mont2_sprp_stage2(), mont2_sqr(), mont2_sqr(), mont2_sub(), mont2_to_mont1(), mont2_to_mont2(), mrs_fermat_usable_(), mrs_mul_bnp1_(), mrs_mul_core_(), mrs_neg_(), mrs_norm_zero_(), mrs_sqr_bnp1_(), mrs_sqr_core_(), mulmod_limbs(), norm_len(), nthroot_add_small_(), nthroot_sub_small_(), overlap_buf_check_guard(), overlap_buf_create(), pcg64_le_seq_next(), pool_cnt(), pool_n(), pow_n(), pow_nPr_(), powmod2b_u32(), prime_table_words(), proth_limbs(), random_limbs(), random_limbs_with_top(), ref_digits(), ref_hex_string(), ref_is_prime32(), ref_is_prime64(), rem_sum_ok(), report_n(), run_even(), run_even2(), run_factorial(), run_fft(), run_fib(), run_hgcd(), run_is_prime_n(), run_is_prime_n_notrial(), run_is_prime_n_prime(), run_lehmer(), run_limb_sizes(), run_mersenne(), run_mul(), run_mul_n(), run_mullo(), run_odd(), run_powlo(), run_sqr(), run_sqrt(), run_strong_lucas(), run_toom(), shr1_inplace(), small_factor_q(), sprp_check_u32(), sqrt_floor_ok(), sub_small_inplace(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), trial_div16(), trial_div35711(), trialdiv_init_chunk_(), trialdiv_init_inv_(), urem32_preinv(), verify_div_relation(), verify_inverse() , 以及 wheel_leq_idx().

◆ numb [1/2]

#define numb   (cache->numb)

◆ numb [2/2]

#define numb   (cache->numb)

被这些函数引用 __lmmp_mul_hard_n_(), __lmmp_mul_n_(), bench_div(), lmmp_add_(), lmmp_add_n_(), lmmp_add_n_sub_n_(), lmmp_add_nc_(), lmmp_addmul_1_(), lmmp_addmul_2_(), lmmp_addshl1_n_(), lmmp_cmp_(), lmmp_div_(), lmmp_div_2_(), lmmp_div_2_s_(), lmmp_div_2_s_(), lmmp_div_3_2_(), lmmp_div_3_2_(), lmmp_div_basecase_(), lmmp_div_basecase_(), lmmp_div_divide_(), lmmp_div_divide_(), lmmp_div_divide_n_(), lmmp_div_mulinv_(), lmmp_div_mulinv_(), lmmp_div_s_(), lmmp_div_s_(), lmmp_fft_bfy_(), lmmp_fft_bfy_fused_(), lmmp_ifft_bfy_(), lmmp_ifft_bfy_fused_(), lmmp_mod_2_(), lmmp_mul_(), lmmp_mul_(), lmmp_mul_basecase_(), lmmp_mul_basecase_(), lmmp_mul_fermat_(), lmmp_mul_fermat_cache_init_(), lmmp_mul_fermat_recurse_(), lmmp_mul_fft_(), lmmp_mul_fft_cache_init_(), lmmp_mul_fft_mods_(), lmmp_mul_fft_unbalance_(), lmmp_mul_fft_unbalance_(), lmmp_mul_mersenne_(), lmmp_mul_mersenne_cache_init_(), lmmp_mul_mersenne_dc_(), lmmp_mul_mersenne_fft_(), lmmp_mul_n_(), lmmp_mul_n_(), lmmp_mul_toom22_(), lmmp_mul_toom22_(), lmmp_mul_toom32_(), lmmp_mul_toom32_(), lmmp_mul_toom33_(), lmmp_mul_toom33_(), lmmp_mul_toom42_(), lmmp_mul_toom42_(), lmmp_mul_toom42_unbalance_(), lmmp_mul_toom42_unbalance_(), lmmp_mul_toom43_(), lmmp_mul_toom43_(), lmmp_mul_toom44_(), lmmp_mul_toom44_(), lmmp_mul_toom52_(), lmmp_mul_toom52_(), lmmp_mul_toom53_(), lmmp_mul_toom53_(), lmmp_mul_toom62_(), lmmp_mul_toom62_(), lmmp_mul_toom62_unbalance_(), lmmp_mul_toom62_unbalance_(), lmmp_mul_toom6h_(), lmmp_mul_toom6h_(), lmmp_mul_toom8h_(), lmmp_mul_toom8h_(), lmmp_mullh_(), lmmp_mullo_(), lmmp_mullo_(), lmmp_mullo_basecase_(), lmmp_mullo_basecase_(), lmmp_mullo_dc_(), lmmp_mullo_dc_(), lmmp_mullo_fft_(), lmmp_mullo_fft_cache_init_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_shlnot_(), lmmp_shr1add_n_(), lmmp_shr1add_nc_(), lmmp_shr1sub_n_(), lmmp_shr1sub_nc_(), lmmp_sqr_hard_19_(), lmmp_sqrlo_dc_(), lmmp_sub_(), lmmp_sub_n_(), lmmp_sub_nc_(), lmmp_submul_1_(), lmmp_subshl1_n_(), TEST_CASE(), TEST_CASE() , 以及 try_div_().

◆ s [1/2]

#define s   (cache->s)

◆ s [2/2]

#define s   (cache->s)

被这些函数引用 _bpow_step_(), _umulhi128_sloppy_(), abs_coeffs(), bezout_identity_l(), check_gcdext(), coeff_bound_l(), ipn_jacobi_small_(), ipn_lucas_i_(), jacobi_small(), ll_small_sqrmod_sloppy(), lmmp_add_n_sub_n_(), lmmp_arith_seqprod_(), lmmp_binvert_unbalance_2_(), lmmp_cbrt_6_est_(), lmmp_cbrt_estimate(), lmmp_cbrt_knorm_(), lmmp_cbrt_shapply_(), lmmp_cbrt_snorm_(), lmmp_divexact_1_(), lmmp_divexact_2_(), lmmp_divexact_basecase_(), lmmp_divexact_by15_(), lmmp_divexact_by3_(), lmmp_divexact_by9_(), lmmp_fft_bfy_fused_(), lmmp_gcdext_11_(), lmmp_hgcd_core_(), lmmp_ifft_bfy_fused_(), lmmp_is_prime_table_(), lmmp_is_prothprime_(), lmmp_mod_4chunk_1_(), lmmp_mod_8chunk_1_(), lmmp_modinv_(), lmmp_modinv_ulong_(), lmmp_mont_sp_redc_(), lmmp_mul_toom22_(), lmmp_mul_toom32_(), lmmp_mul_toom33_(), lmmp_mul_toom42_(), lmmp_mul_toom42_cache_(), lmmp_mul_toom42_cache_init_(), lmmp_mul_toom43_(), lmmp_mul_toom44_(), lmmp_mul_toom52_(), lmmp_mul_toom53_(), lmmp_mul_toom62_(), lmmp_mul_toom62_cache_(), lmmp_mul_toom62_cache_init_(), lmmp_mul_toom6h_(), lmmp_mul_toom8h_(), lmmp_nthroot_(), lmmp_prime_int_table_init_(), lmmp_shr64_(), lmmp_shr64_(), lmmp_sqr_toom2_(), lmmp_sqr_toom3_(), lmmp_sqr_toom4_(), lmmp_sqr_toom6_(), lmmp_sqr_toom8_(), lmmp_sqrt_(), lmmp_sqrt_1_(), lmmp_sqrt_2_(), lmmp_toom_eval_pm2rexp_(), mat22_sqr_basecase_(), mont2_dbl(), mont2_lucas_strong_(), nthroot_top_limbs_(), pcg64_le_seq_init(), ref_is_prime64(), run_gcdext_pair(), run_is_prime_n_tier(), scalar_bezout(), shifted_pair(), sp_shape_limbs(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE() , 以及 xorshift64().

◆ t [1/2]

#define t   (cache->t)

◆ t [2/2]

#define t   (cache->t)

被这些函数引用 __lmmp_mulh_(), _udiv64_gen_internal_(), _udiv64by64_q_preinv(), _umul64to64hi_(), _umullo128x64_(), _usqr128to192_(), bench_div(), bench_is_prime_n_proth(), bench_is_prothprime(), bench_mul(), bench_mul_basecase(), bench_mul_fft(), bench_mul_toom22(), bench_mul_toom33(), bench_mul_toom44(), bezout_identity_l(), check_addmul_2(), check_gcdext(), exp2_fixed_64(), factor_2adic(), fill_pattern(), gcd_u64(), gcd_u64(), gcd_u64(), gcd_u64(), ipn_notrial_i_(), ipn_sprp_1_(), ipn_sprp_base2_(), is2_notrial_i_(), k2t_limbs(), ll_small_is_prime_mr(), ll_small_mul1_(), ll_small_reduce2_precise(), ll_small_sprp_stage2(), ll_small_sqrmod_sloppy(), lmmp_add_n_sub_n_(), lmmp_addmul_2_(), lmmp_arith_seqprod_(), lmmp_binvert_2_(), lmmp_binvert_unbalance_2_(), lmmp_cbrt_(), lmmp_cbrt_3_(), lmmp_cbrt_6_(), lmmp_cbrt_6_fix_(), lmmp_cube_3_(), lmmp_cube_3_(), lmmp_cube_6_(), lmmp_cube_6_(), lmmp_div_1_(), lmmp_div_1_s_(), lmmp_divexact_2_(), lmmp_divexact_basecase_(), lmmp_divexact_by15_(), lmmp_divexact_by9_(), lmmp_factors_mul_(), lmmp_factors_mul_ushort_(), lmmp_gcd_11_(), lmmp_gcd_22_(), lmmp_gcd_lehmer_step_(), lmmp_gcdext_11_(), lmmp_hgcd_lehmer2_(), lmmp_hgcd_matrix_mul_u_(), lmmp_hgcd_matrix_update_q_(), lmmp_hgcd_step_(), lmmp_ifft_bfy_fused_(), lmmp_is_prime_uint_notrial_(), lmmp_is_prime_ulong_notrial_(), lmmp_leak_tracker(), lmmp_mod_1_(), lmmp_mod_4chunk_1_(), lmmp_mod_8chunk_1_(), lmmp_mod_preinv_1_(), lmmp_mont_sp_redc_(), lmmp_mul_toom22_(), lmmp_mul_toom32_(), lmmp_mul_toom33_(), lmmp_mul_toom42_(), lmmp_mul_toom42_cache_(), lmmp_mul_toom42_cache_init_(), lmmp_mul_toom43_(), lmmp_mul_toom44_(), lmmp_mul_toom52_(), lmmp_mul_toom53_(), lmmp_mul_toom62_(), lmmp_mul_toom62_cache_(), lmmp_mul_toom62_cache_init_(), lmmp_mul_toom6h_(), lmmp_mul_toom8h_(), lmmp_mulh_(), lmmp_mullo_dc_(), lmmp_odd_multinomial_ushort_(), lmmp_odd_nCr_uint_(), lmmp_odd_nCr_ushort_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_ushort_(), lmmp_pow_size_(), lmmp_powmod_(), lmmp_primefac_(), lmmp_shlnot_(), lmmp_sqr_basecase_(), lmmp_sqrlo_dc_(), lmmp_sqrt_(), lmmp_sqrt_4_(), lmmp_toom_eval_pm2rexp_(), lmmp_xoshiro256pp_random(), mont2_sprp_stage2(), mont63_reduce(), mont63_reduce(), mont63_reduce(), mont64_reduce(), mont64_reduce(), mulmod64(), mulmod_limbs(), pool_proth(), powmod1_redc_(), powmod_ref_u64(), proth_cert_g(), proth_certified(), proth_limbs(), run_gcdext_pair(), run_proth(), scalar_bezout(), sprp_check63(), sprp_check64(), sprp_check_u32(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE(), urem32_preinv(), verify_div_relation() , 以及 verify_inverse().

◆ tp [1/2]

#define tp   (cache->tp)

◆ tp [2/2]

#define tp   (cache->tp)

被这些函数引用 bench_mullo_dc(), bench_mullo_fft(), binvert_mulhi_(), fib2_core_(), lmmp_9pow_1_(), lmmp_binvert_(), lmmp_binvert_n_dc_(), lmmp_binvert_n_dc_(), lmmp_binvert_unbalance_(), lmmp_binvert_unbalance_(), lmmp_cbrt_(), lmmp_cbrt_6_(), lmmp_cbrt_6_fast_(), lmmp_cbrt_6_fast_(), lmmp_cbrt_divide_(), lmmp_cbrt_divide_(), lmmp_div_(), lmmp_div_divide_(), lmmp_div_divide_n_(), lmmp_div_mulinv_(), lmmp_div_s_(), lmmp_divexact_(), lmmp_divexact_divide_(), lmmp_divexact_mu_run_(), lmmp_divexact_unbalance_(), lmmp_elem_mul_ulong_(), lmmp_elem_mul_ulong_(), lmmp_factors_mul_(), lmmp_factors_mul_ushort_(), lmmp_fibonacci2_(), lmmp_fibonacci_(), lmmp_from_decimal_(), lmmp_from_str_(), lmmp_from_str_divide_(), lmmp_from_str_divide_(), lmmp_hgcd_matrix_mul_(), lmmp_huff_tree_mul_(), lmmp_huff_tree_mul_(), lmmp_inv_prediv_(), lmmp_invcbrt_newton_(), lmmp_invsqrt_newton_(), lmmp_mat22_mul_(), lmmp_mat22_sqr_(), lmmp_mont_redc_(), lmmp_mont_sp_redc_(), lmmp_mul_fft_(), lmmp_mul_fft_crt_(), lmmp_mul_fft_mods_(), lmmp_mul_mersenne_(), lmmp_mul_mersenne_dc_(), lmmp_mul_toom32_(), lmmp_mul_toom33_(), lmmp_mul_toom42_(), lmmp_mul_toom44_(), lmmp_mullo_(), lmmp_mullo_dc_(), lmmp_mullo_dc_(), lmmp_mullo_fft_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_mullo_n_(), lmmp_nthroot_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_product_(), lmmp_odd_nPr_ulong_(), lmmp_odd_nPr_ushort_(), lmmp_perfsqr_(), lmmp_powlo_(), lmmp_primefac_(), lmmp_redc_(), lmmp_redc_(), lmmp_sqr_fft_(), lmmp_sqr_mersenne_dc_(), lmmp_sqr_toom3_(), lmmp_sqr_toom4_(), lmmp_sqrlo_dc_(), lmmp_sqrlo_dc_(), lmmp_sqrlo_n_(), lmmp_sqrt_(), lmmp_sqrt_divide_(), lmmp_sqrt_divide_(), lmmp_sqrt_newton_(), lmmp_to_decimal_(), lmmp_to_str_(), lmmp_to_str_basecase_(), lmmp_toom_eval_dgr3_pm1_(), lmmp_toom_eval_dgr3_pm2_(), lmmp_toom_eval_pm1_(), lmmp_toom_eval_pm2_(), lmmp_toom_eval_pm2exp_(), lmmp_toom_interp12_(), lmmp_toom_interp16_(), lmmp_toom_interp7_(), mat22_mul_basecase_(), mat22_mul_strassen_(), mat22_sqr_basecase_(), mat22_sqr_strassen_(), mrs_mul_bnp1_(), mrs_mul_core_(), mrs_sqr_bnp1_(), mrs_sqr_core_(), pow_nPr_(), swap_huff_node(), TEST_CASE(), TEST_CASE(), TEST_CASE(), TEST_CASE() , 以及 TEST_CASE().

◆ v0 [1/3]

#define v0   dst

◆ v0 [2/3]

#define v0   dst

◆ v0 [3/3]

#define v0   dst

◆ v1 [1/3]

#define v1   (dst + 2 * n)

◆ v1 [2/3]

#define v1   (dst + 2 * n)

◆ v1 [3/3]

#define v1   (dst + 2 * n)

◆ v2 [1/3]

#define v2   (tp + 2 * n + 2)

◆ v2 [2/3]

#define v2   (tp + 2 * n + 2)

◆ v2 [3/3]

#define v2   (tp + 2 * n + 2)

◆ vinf [1/3]

#define vinf   (dst + 4 * n)

◆ vinf [2/3]

#define vinf   (dst + 4 * n)

◆ vinf [3/3]

#define vinf   (dst + 4 * n)

◆ vm1 [1/3]

#define vm1   tp

◆ vm1 [2/3]

#define vm1   tp

◆ vm1 [3/3]

#define vm1   tp

函数说明

◆ lmmp_mul_toom42_()

void lmmp_mul_toom42_ ( mp_ptr restrict dst,
mp_srcptr restrict numa,
mp_size_t na,
mp_srcptr restrict numb,
mp_size_t nb )

在文件 mul_toom42.c 第 54 行定义.

54 {
55 lmmp_param_assert(nb >= 20);
56 lmmp_param_assert(4 * nb > na);
57 lmmp_param_assert(3 * na > 2 * nb);
59 mp_size_t n = na >= 2 * nb ? (na + 3) >> 2 : (nb + 1) >> 1, s = na - 3 * n, t = nb - n;
62 int vm1_neg;
63 mp_limb_t cy, vinf0, am1h;
65
66#define a0 numa
67#define a1 (numa + n)
68#define a2 (numa + 2 * n)
69#define a3 (numa + 3 * n)
70#define b0 numb
71#define b1 (numb + n)
72
73#define v0 dst //[dst,2*n]
74#define v1 (dst + 2 * n) //[dst+2*n,2*n+1]
75#define vinf (dst + 4 * n) //[dst+4*n,s+t]
76#define vm1 tp //[tp,2*n+1]
77#define v2 (tp + 2 * n + 2) //[tp+2*n+2,2*n+1]
78
79#define bm1 dst //[dst,n]
80#define am1 (dst + n) //[dst+n,n+1]
81#define ap1 tp //[tp,n+1]
82#define bp1 (tp + n + 1) //[tp+n+1,n+1]
83#define ap2 ap1 // same space
84#define bp2 bp1 // same space
85#define a13 bp1 // temporary use
86
87 // ap1,am1
88 ap1[n] = lmmp_add_n_(ap1, a0, a2, n);
89 a13[n] = lmmp_add_(a13, a1, n, a3, s);
90 vm1_neg = lmmp_cmp_(ap1, a13, n + 1) < 0;
91 if (vm1_neg)
93 else
95 am1h = am1[n]; // overlap with v1
96
97 // bp1,bm1
98 if (t == n) {
99 if (lmmp_cmp_(b0, b1, n) < 0) {
100 bp1[n] = lmmp_add_n_sub_n_(bp1, bm1, b1, b0, n) >> 1;
101 vm1_neg ^= 1;
102 } else {
103 bp1[n] = lmmp_add_n_sub_n_(bp1, bm1, b0, b1, n) >> 1;
104 }
105 } else {
106 if (lmmp_zero_q_(b0 + t, n - t) && lmmp_cmp_(b0, b1, t) < 0) {
107 cy = lmmp_add_n_sub_n_(bp1, bm1, b1, b0, t);
108 lmmp_zero(bm1 + t, n - t);
109 vm1_neg ^= 1;
110 } else {
111 cy = lmmp_add_n_sub_n_(bp1, bm1, b0, b1, t);
112 lmmp_sub_1_(bm1 + t, b0 + t, n - t, cy & 1);
113 }
114 bp1[n] = lmmp_add_1_(bp1 + t, b0 + t, n - t, cy >> 1);
115 }
116
117 // vinf=a3*b1
118 if (s > t)
119 lmmp_mul_(vinf, a3, s, b1, t);
120 else
121 lmmp_mul_(vinf, b1, t, a3, s);
122 vinf0 = vinf[0]; // overlap with v1
123 cy = vinf[1]; // overlap with v1
124
125 // v1=ap1*bp1
126 lmmp_mul_n_(v1, ap1, bp1, n + 1);
127 vinf[1] = cy; // restore, since v1[2*n+1]==0.
128
129 // ap2
130 cy = lmmp_addshl1_n_(ap2, a2, a3, s);
131 if (s != n)
132 cy = lmmp_add_1_(ap2 + s, a2 + s, n - s, cy);
133 cy = 2 * cy + lmmp_addshl1_n_(ap2, a1, ap2, n);
134 cy = 2 * cy + lmmp_addshl1_n_(ap2, a0, ap2, n);
135 ap2[n] = cy;
136
137 // bp2=bp1+b1
138 lmmp_add_(bp2, bp1, n + 1, b1, t);
139
140 // v2=ap2*bp2
141 lmmp_mul_n_(v2, ap2, bp2, n + 1);
142
143 // vm1=am1*bm1
144 lmmp_mul_n_(vm1, am1, bm1, n);
145 if (am1h)
146 vm1[2 * n] = lmmp_add_n_(vm1 + n, vm1 + n, bm1, n);
147 else
148 vm1[2 * n] = 0;
149
150 // v0=a0*b0
151 lmmp_mul_n_(v0, a0, b0, n);
152
153 lmmp_toom_interp5_(dst, v2, vm1, n, s + t, vm1_neg, vinf0);
155#undef a0
156#undef a1
157#undef a2
158#undef a3
159#undef b0
160#undef b1
161
162#undef v0
163#undef v1
164#undef vinf
165#undef vm1
166#undef v2
167
168#undef bm1
169#undef am1
170#undef ap1
171#undef bp1
172#undef ap2
173#undef bp2
174#undef a13
175}
#define lmmp_mul_n_
#define lmmp_zero(dst, n)
定义 lmmp.h:476
uint64_t mp_size_t
定义 lmmp.h:93
uint64_t mp_limb_t
定义 lmmp.h:92
#define lmmp_param_assert(x)
定义 lmmp.h:509
static mp_limb_t lmmp_add_(mp_ptr dst, mp_srcptr numa, mp_size_t na, mp_srcptr numb, mp_size_t nb)
加法静态内联函数 [dst,na]=[numa,na]+[numb,nb]
static mp_limb_t lmmp_add_1_(mp_ptr dst, mp_srcptr numa, mp_size_t na, mp_limb_t x)
加单精度数静态内联函数 [dst,na]=[numa,na]+x
void lmmp_mul_(mp_ptr dst, mp_srcptr numa, mp_size_t na, mp_srcptr numb, mp_size_t nb)
不等长乘法操作 [dst,na+nb] = [numa,na] * [numb,nb]
mp_limb_t lmmp_addshl1_n_(mp_ptr dst, mp_srcptr numa, mp_srcptr numb, mp_size_t n)
加法结合左移1位操作 [dst,n] = [numa,n] + ([numb,n] << 1)
定义 shl.c:53
int lmmp_cmp_(mp_srcptr numa, mp_srcptr numb, mp_size_t n)
比较函数
定义 cmp.c:26
int lmmp_zero_q_(mp_srcptr p, mp_size_t n)
判零函数
定义 cmp.c:55
mp_limb_t lmmp_add_n_sub_n_(mp_ptr dsta, mp_ptr dstb, mp_srcptr numa, mp_srcptr numb, mp_size_t n)
同时执行n位加法和减法 ([dsta,n],[dstb,n]) = ([numa,n]+[numb,n],[numa,n]-[numb,n])
static mp_limb_t lmmp_sub_1_(mp_ptr dst, mp_srcptr numa, mp_size_t na, mp_limb_t x)
减单精度数静态内联函数 [dst,na]=[numa,na]-x
mp_limb_t lmmp_add_n_(mp_ptr dst, mp_srcptr numa, mp_srcptr numb, mp_size_t n)
无进位的n位加法 [dst,n] = [numa,n] + [numb,n]
定义 add_n.c:76
#define b0
#define b1
#define a0
#define a1
#define v0
#define vinf
#define am1
#define ap1
#define bp1
#define vm1
#define bm1
#define a2
#define v1
#define ap2
#define v2
#define bp2
#define t
#define a3
#define a13
#define tp
#define s
void lmmp_toom_interp5_(mp_ptr dst, mp_ptr v2, mp_ptr vm1, mp_size_t n, mp_size_t spt, int vm1_neg, mp_limb_t vinf0)
Copyright (C) 2026 HJimmyK (Jericho Knox)
#define restrict
Copyright (C) 2026 HJimmyK (Jericho Knox)
#define SALLOC_TYPE(n, type)
#define TEMP_S_DECL
#define TEMP_S_FREE

引用了 a0, a1, a13, a2, a3, am1, ap1, ap2, b0, b1, bm1, bp1, bp2, lmmp_add_(), lmmp_add_1_(), lmmp_add_n_(), lmmp_add_n_sub_n_(), lmmp_addshl1_n_(), lmmp_cmp_(), lmmp_mul_(), lmmp_mul_n_, lmmp_mul_toom42_(), lmmp_param_assert, lmmp_sub_1_(), lmmp_toom_interp5_(), lmmp_zero, lmmp_zero_q_(), n, numb, restrict, s, SALLOC_TYPE, t, TEMP_S_DECL, TEMP_S_FREE, tp, v0, v1, v2, vinf , 以及 vm1.

函数调用图:

◆ lmmp_mul_toom42_cache_()

void lmmp_mul_toom42_cache_ ( mp_ptr restrict dst,
mp_srcptr restrict numa,
const toom42_cache_t * cache,
int flag )
static

在文件 mul_toom42.c 第 323 行定义.

328 {
329#define numb (cache->numb)
330#define n (cache->n)
331#define s (cache->s)
332#define t (cache->t)
333#define _bp1 (cache->_bp1)
334#define _bm1 (cache->_bm1)
335#define tp (cache->tp)
336
337 int vm1_neg;
338 mp_limb_t cy, vinf0, am1h;
339
340#define a0 numa
341#define a1 (numa + n)
342#define a2 (numa + 2 * n)
343#define a3 (numa + 3 * n)
344#define b0 numb
345#define b1 (numb + n)
346
347#define v0 dst //[dst,2*n]
348#define v1 (dst + 2 * n) //[dst+2*n,2*n+1]
349#define vinf (dst + 4 * n) //[dst+4*n,s+t]
350#define vm1 tp //[tp,2*n+1]
351#define v2 (tp + 2 * n + 2) //[tp+2*n+2,2*n+1]
352
353#define bm1 _bm1 //[dst,n]
354#define am1 (dst + n) //[dst+n,n+1]
355#define ap1 tp //[tp,n+1]
356#define bp1 _bp1 //[TH._bp1,n+1]
357#define ap2 ap1 // same space
358#define bp2 (tp + n + 1) //[tp+n+1,n+1]
359#define a13 (tp + n + 1) // same space
360
361 // ap1,am1
362 ap1[n] = lmmp_add_n_(ap1, a0, a2, n);
363 a13[n] = lmmp_add_(a13, a1, n, a3, s);
364 vm1_neg = lmmp_cmp_(ap1, a13, n + 1) < 0;
365 if (vm1_neg)
366 lmmp_add_n_sub_n_(ap1, am1, a13, ap1, n + 1);
367 else
368 lmmp_add_n_sub_n_(ap1, am1, ap1, a13, n + 1);
369 am1h = am1[n]; // overlap with v1
370
371 if (flag)
372 vm1_neg ^= 1;
373
374 // vinf=a3*b1
375 if (s > t)
376 lmmp_mul_(vinf, a3, s, b1, t);
377 else
378 lmmp_mul_(vinf, b1, t, a3, s);
379 vinf0 = vinf[0]; // overlap with v1
380 cy = vinf[1]; // overlap with v1
381
382 // v1=ap1*bp1
383 lmmp_mul_n_(v1, ap1, bp1, n + 1);
384 vinf[1] = cy; // restore, since v1[2*n+1]==0.
385
386 // ap2
387 cy = lmmp_addshl1_n_(ap2, a2, a3, s);
388 if (s != n)
389 cy = lmmp_add_1_(ap2 + s, a2 + s, n - s, cy);
390 cy = 2 * cy + lmmp_addshl1_n_(ap2, a1, ap2, n);
391 cy = 2 * cy + lmmp_addshl1_n_(ap2, a0, ap2, n);
392 ap2[n] = cy;
393
394 // bp2=bp1+b1
395 lmmp_add_(bp2, bp1, n + 1, b1, t);
396
397 // v2=ap2*bp2
398 lmmp_mul_n_(v2, ap2, bp2, n + 1);
399
400 // vm1=am1*bm1
401 lmmp_mul_n_(vm1, am1, bm1, n);
402 if (am1h)
403 vm1[2 * n] = lmmp_add_n_(vm1 + n, vm1 + n, bm1, n);
404 else
405 vm1[2 * n] = 0;
406
407 // v0=a0*b0
408 lmmp_mul_n_(v0, a0, b0, n);
409
410 lmmp_toom_interp5_(dst, v2, vm1, n, s + t, vm1_neg, vinf0);
411
412#undef numb
413#undef n
414#undef s
415#undef t
416#undef _bp1
417#undef _bm1
418#undef tp
419}

引用了 a0, a1, a13, a2, a3, am1, ap1, ap2, b0, b1, bm1, bp1, bp2, lmmp_add_(), lmmp_add_1_(), lmmp_add_n_(), lmmp_add_n_sub_n_(), lmmp_addshl1_n_(), lmmp_cmp_(), lmmp_mul_(), lmmp_mul_n_, lmmp_mul_toom42_cache_(), lmmp_toom_interp5_(), n, restrict, s, t, v0, v1, v2, vinf , 以及 vm1.

被这些函数引用 lmmp_mul_toom42_cache_() , 以及 lmmp_mul_toom42_unbalance_().

函数调用图:
这是这个函数的调用关系图:

◆ lmmp_mul_toom42_cache_init_()

int lmmp_mul_toom42_cache_init_ ( mp_ptr restrict dst,
mp_srcptr restrict numa,
toom42_cache_t * cache )
static

在文件 mul_toom42.c 第 187 行定义.

191 {
192#define numb (cache->numb)
193#define n (cache->n)
194#define s (cache->s)
195#define t (cache->t)
196#define _bp1 (cache->_bp1)
197#define _bm1 (cache->_bm1)
198#define tp (cache->tp)
199
200 int vm1_neg, flag = 0;
201 mp_limb_t cy, vinf0, am1h;
202
203#define a0 numa
204#define a1 (numa + n)
205#define a2 (numa + 2 * n)
206#define a3 (numa + 3 * n)
207#define b0 numb
208#define b1 (numb + n)
209
210#define v0 dst //[dst,2*n]
211#define v1 (dst + 2 * n) //[dst+2*n,2*n+1]
212#define vinf (dst + 4 * n) //[dst+4*n,s+t]
213#define vm1 tp //[tp,2*n+1]
214#define v2 (tp + 2 * n + 2) //[tp+2*n+2,2*n+1]
215
216#define bm1 _bm1 //[dst,n]
217#define am1 (dst + n) //[dst+n,n+1]
218#define ap1 tp //[tp,n+1]
219#define bp1 _bp1 //[TH._bp1,n+1]
220#define ap2 ap1 // same space
221#define bp2 (tp + n + 1) //[tp+n+1,n+1]
222#define a13 (tp + n + 1) // same space
223
224 // ap1,am1
225 ap1[n] = lmmp_add_n_(ap1, a0, a2, n);
226 a13[n] = lmmp_add_(a13, a1, n, a3, s);
227 vm1_neg = lmmp_cmp_(ap1, a13, n + 1) < 0;
228 if (vm1_neg)
229 lmmp_add_n_sub_n_(ap1, am1, a13, ap1, n + 1);
230 else
231 lmmp_add_n_sub_n_(ap1, am1, ap1, a13, n + 1);
232 am1h = am1[n]; // overlap with v1
233
234 if (t == n) {
235 if (lmmp_cmp_(b0, b1, n) < 0) {
236 bp1[n] = lmmp_add_n_sub_n_(bp1, bm1, b1, b0, n) >> 1;
237 vm1_neg ^= 1;
238 flag = 1;
239 } else {
240 bp1[n] = lmmp_add_n_sub_n_(bp1, bm1, b0, b1, n) >> 1;
241 }
242 } else {
243 if (lmmp_zero_q_(b0 + t, n - t) && lmmp_cmp_(b0, b1, t) < 0) {
244 cy = lmmp_add_n_sub_n_(bp1, bm1, b1, b0, t);
245 lmmp_zero(bm1 + t, n - t);
246 vm1_neg ^= 1;
247 flag = 1;
248 } else {
249 cy = lmmp_add_n_sub_n_(bp1, bm1, b0, b1, t);
250 lmmp_sub_1_(bm1 + t, b0 + t, n - t, cy & 1);
251 }
252 bp1[n] = lmmp_add_1_(bp1 + t, b0 + t, n - t, cy >> 1);
253 }
254
255 // vinf=a3*b1
256 if (s > t)
257 lmmp_mul_(vinf, a3, s, b1, t);
258 else
259 lmmp_mul_(vinf, b1, t, a3, s);
260 vinf0 = vinf[0]; // overlap with v1
261 cy = vinf[1]; // overlap with v1
262
263 // v1=ap1*bp1
264 lmmp_mul_n_(v1, ap1, bp1, n + 1);
265 vinf[1] = cy; // restore, since v1[2*n+1]==0.
266
267 // ap2
268 cy = lmmp_addshl1_n_(ap2, a2, a3, s);
269 if (s != n)
270 cy = lmmp_add_1_(ap2 + s, a2 + s, n - s, cy);
271 cy = 2 * cy + lmmp_addshl1_n_(ap2, a1, ap2, n);
272 cy = 2 * cy + lmmp_addshl1_n_(ap2, a0, ap2, n);
273 ap2[n] = cy;
274
275 // bp2=bp1+b1
276 lmmp_add_(bp2, bp1, n + 1, b1, t);
277
278 // v2=ap2*bp2
279 lmmp_mul_n_(v2, ap2, bp2, n + 1);
280
281 // vm1=am1*bm1
282 lmmp_mul_n_(vm1, am1, bm1, n);
283 if (am1h)
284 vm1[2 * n] = lmmp_add_n_(vm1 + n, vm1 + n, bm1, n);
285 else
286 vm1[2 * n] = 0;
287
288 // v0=a0*b0
289 lmmp_mul_n_(v0, a0, b0, n);
290
291 lmmp_toom_interp5_(dst, v2, vm1, n, s + t, vm1_neg, vinf0);
292 return flag;
293#undef a0
294#undef a1
295#undef a2
296#undef a3
297#undef b0
298#undef b1
299
300#undef v0
301#undef v1
302#undef vinf
303#undef vm1
304#undef v2
305
306#undef bm1
307#undef am1
308#undef ap1
309#undef bp1
310#undef ap2
311#undef bp2
312#undef a13
313
314#undef numb
315#undef n
316#undef s
317#undef t
318#undef _bp1
319#undef _bm1
320#undef tp
321}

引用了 a0, a1, a13, a2, a3, am1, ap1, ap2, b0, b1, bm1, bp1, bp2, lmmp_add_(), lmmp_add_1_(), lmmp_add_n_(), lmmp_add_n_sub_n_(), lmmp_addshl1_n_(), lmmp_cmp_(), lmmp_mul_(), lmmp_mul_n_, lmmp_mul_toom42_cache_init_(), lmmp_sub_1_(), lmmp_toom_interp5_(), lmmp_zero, lmmp_zero_q_(), n, restrict, s, t, v0, v1, v2, vinf , 以及 vm1.

被这些函数引用 lmmp_mul_toom42_cache_init_() , 以及 lmmp_mul_toom42_unbalance_().

函数调用图:
这是这个函数的调用关系图:

◆ lmmp_mul_toom42_unbalance_()

void lmmp_mul_toom42_unbalance_ ( mp_ptr restrict dst,
mp_srcptr restrict numa,
mp_size_t na,
mp_srcptr restrict numb,
mp_size_t nb )

在文件 mul_toom42.c 第 421 行定义.

427 {
428 lmmp_param_assert(na >= 3 * nb);
429 lmmp_param_assert(nb >= 20);
432
433 toom42_cache_t cache;
434 cache.numb = numb;
435 cache.n = (2 * nb + 3) >> 2;
436 cache.s = 2 * nb - 3 * cache.n;
437 cache.t = nb - cache.n;
438 cache.tp = SALLOC_TYPE(4 * cache.n + 4, mp_limb_t);
439 cache._bp1 = SALLOC_TYPE(2 * cache.n + 1, mp_limb_t);
440 cache._bm1 = cache._bp1 + cache.n + 1;
441
442 int flag = lmmp_mul_toom42_cache_init_(dst, numa, &cache);
443 dst += 2 * nb;
444 numa += 2 * nb;
445 na -= 2 * nb;
446 lmmp_copy(ws, dst, nb);
447 while (2 * na >= 5 * nb) {
448 lmmp_mul_toom42_cache_(dst, numa, &cache, flag);
449 if (lmmp_add_n_(dst, dst, ws, nb))
450 lmmp_inc(dst + nb);
451 dst += 2 * nb;
452 numa += 2 * nb;
453 na -= 2 * nb;
454 lmmp_copy(ws, dst, nb);
455 }
456 // 0.5 nb <= na < 2.5 nb
457 if (na >= nb)
458 lmmp_mul_(dst, numa, na, numb, nb);
459 else
460 lmmp_mul_(dst, numb, nb, numa, na);
461 if (lmmp_add_n_(dst, dst, ws, nb))
462 lmmp_inc(dst + nb);
464}
#define lmmp_copy(dst, src, n)
定义 lmmp.h:474
#define lmmp_inc(p)
加1宏(预期无进位)
mp_ptr restrict tp
static void lmmp_mul_toom42_cache_(mp_ptr restrict dst, mp_srcptr restrict numa, const toom42_cache_t *cache, int flag)
static int lmmp_mul_toom42_cache_init_(mp_ptr restrict dst, mp_srcptr restrict numa, toom42_cache_t *cache)
mp_srcptr restrict numb
mp_ptr restrict _bp1
mp_ptr restrict _bm1

引用了 toom42_cache_t::_bm1, toom42_cache_t::_bp1, lmmp_add_n_(), lmmp_copy, lmmp_inc, lmmp_mul_(), lmmp_mul_toom42_cache_(), lmmp_mul_toom42_cache_init_(), lmmp_mul_toom42_unbalance_(), lmmp_param_assert, toom42_cache_t::n, numb, toom42_cache_t::numb, restrict, toom42_cache_t::s, SALLOC_TYPE, toom42_cache_t::t, TEMP_S_DECL, TEMP_S_FREE , 以及 toom42_cache_t::tp.

函数调用图: