Austin Schuh | bb1338c | 2024-06-15 19:31:16 -0700 | [diff] [blame] | 1 | /* gmp_nextprime -- generate small primes reasonably efficiently for internal |
| 2 | GMP needs. |
| 3 | |
| 4 | Contributed to the GNU project by Torbjorn Granlund. Miscellaneous |
| 5 | improvements by Martin Boij. |
| 6 | |
| 7 | THE FUNCTIONS IN THIS FILE ARE INTERNAL WITH MUTABLE INTERFACES. IT IS ONLY |
| 8 | SAFE TO REACH THEM THROUGH DOCUMENTED INTERFACES. IN FACT, IT IS ALMOST |
| 9 | GUARANTEED THAT THEY WILL CHANGE OR DISAPPEAR IN A FUTURE GNU MP RELEASE. |
| 10 | |
| 11 | Copyright 2009 Free Software Foundation, Inc. |
| 12 | |
| 13 | This file is part of the GNU MP Library. |
| 14 | |
| 15 | The GNU MP Library is free software; you can redistribute it and/or modify |
| 16 | it under the terms of either: |
| 17 | |
| 18 | * the GNU Lesser General Public License as published by the Free |
| 19 | Software Foundation; either version 3 of the License, or (at your |
| 20 | option) any later version. |
| 21 | |
| 22 | or |
| 23 | |
| 24 | * the GNU General Public License as published by the Free Software |
| 25 | Foundation; either version 2 of the License, or (at your option) any |
| 26 | later version. |
| 27 | |
| 28 | or both in parallel, as here. |
| 29 | |
| 30 | The GNU MP Library is distributed in the hope that it will be useful, but |
| 31 | WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY |
| 32 | or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
| 33 | for more details. |
| 34 | |
| 35 | You should have received copies of the GNU General Public License and the |
| 36 | GNU Lesser General Public License along with the GNU MP Library. If not, |
| 37 | see https://www.gnu.org/licenses/. */ |
| 38 | |
| 39 | /* |
| 40 | Optimisation ideas: |
| 41 | |
| 42 | 1. Unroll the sieving loops. Should reach 1 write/cycle. That would be a 2x |
| 43 | improvement. |
| 44 | |
| 45 | 2. Separate sieving with primes p < SIEVESIZE and p >= SIEVESIZE. The latter |
| 46 | will need at most one write, and thus not need any inner loop. |
| 47 | |
| 48 | 3. For primes p >= SIEVESIZE, i.e., typically the majority of primes, we |
| 49 | perform more than one division per sieving write. That might dominate the |
| 50 | entire run time for the nextprime function. A incrementally initialised |
| 51 | remainder table of Pi(65536) = 6542 16-bit entries could replace that |
| 52 | division. |
| 53 | */ |
| 54 | |
| 55 | #include "gmp-impl.h" |
| 56 | #include <string.h> /* for memset */ |
| 57 | |
| 58 | |
| 59 | unsigned long int |
| 60 | gmp_nextprime (gmp_primesieve_t *ps) |
| 61 | { |
| 62 | unsigned long p, d, pi; |
| 63 | unsigned char *sp; |
| 64 | static unsigned char addtab[] = |
| 65 | { 2,4,2,4,6,2,6,4,2,4,6,6,2,6,4,2,6,4,6,8,4,2,4,2,4,8,6,4,6,2,4,6,2,6,6,4, |
| 66 | 2,4,6,2,6,4,2,4,2,10,2,10 }; |
| 67 | unsigned char *addp = addtab; |
| 68 | unsigned long ai; |
| 69 | |
| 70 | /* Look for already sieved primes. A sentinel at the end of the sieving |
| 71 | area allows us to use a very simple loop here. */ |
| 72 | d = ps->d; |
| 73 | sp = ps->s + d; |
| 74 | while (*sp != 0) |
| 75 | sp++; |
| 76 | if (sp != ps->s + SIEVESIZE) |
| 77 | { |
| 78 | d = sp - ps->s; |
| 79 | ps->d = d + 1; |
| 80 | return ps->s0 + 2 * d; |
| 81 | } |
| 82 | |
| 83 | /* Handle the number 2 separately. */ |
| 84 | if (ps->s0 < 3) |
| 85 | { |
| 86 | ps->s0 = 3 - 2 * SIEVESIZE; /* Tricky */ |
| 87 | return 2; |
| 88 | } |
| 89 | |
| 90 | /* Exhausted computed primes. Resieve, then call ourselves recursively. */ |
| 91 | |
| 92 | #if 0 |
| 93 | for (sp = ps->s; sp < ps->s + SIEVESIZE; sp++) |
| 94 | *sp = 0; |
| 95 | #else |
| 96 | memset (ps->s, 0, SIEVESIZE); |
| 97 | #endif |
| 98 | |
| 99 | ps->s0 += 2 * SIEVESIZE; |
| 100 | |
| 101 | /* Update sqrt_s0 as needed. */ |
| 102 | while ((ps->sqrt_s0 + 1) * (ps->sqrt_s0 + 1) <= ps->s0 + 2 * SIEVESIZE - 1) |
| 103 | ps->sqrt_s0++; |
| 104 | |
| 105 | pi = ((ps->s0 + 3) / 2) % 3; |
| 106 | if (pi > 0) |
| 107 | pi = 3 - pi; |
| 108 | if (ps->s0 + 2 * pi <= 3) |
| 109 | pi += 3; |
| 110 | sp = ps->s + pi; |
| 111 | while (sp < ps->s + SIEVESIZE) |
| 112 | { |
| 113 | *sp = 1, sp += 3; |
| 114 | } |
| 115 | |
| 116 | pi = ((ps->s0 + 5) / 2) % 5; |
| 117 | if (pi > 0) |
| 118 | pi = 5 - pi; |
| 119 | if (ps->s0 + 2 * pi <= 5) |
| 120 | pi += 5; |
| 121 | sp = ps->s + pi; |
| 122 | while (sp < ps->s + SIEVESIZE) |
| 123 | { |
| 124 | *sp = 1, sp += 5; |
| 125 | } |
| 126 | |
| 127 | pi = ((ps->s0 + 7) / 2) % 7; |
| 128 | if (pi > 0) |
| 129 | pi = 7 - pi; |
| 130 | if (ps->s0 + 2 * pi <= 7) |
| 131 | pi += 7; |
| 132 | sp = ps->s + pi; |
| 133 | while (sp < ps->s + SIEVESIZE) |
| 134 | { |
| 135 | *sp = 1, sp += 7; |
| 136 | } |
| 137 | |
| 138 | p = 11; |
| 139 | ai = 0; |
| 140 | while (p <= ps->sqrt_s0) |
| 141 | { |
| 142 | pi = ((ps->s0 + p) / 2) % p; |
| 143 | if (pi > 0) |
| 144 | pi = p - pi; |
| 145 | if (ps->s0 + 2 * pi <= p) |
| 146 | pi += p; |
| 147 | sp = ps->s + pi; |
| 148 | while (sp < ps->s + SIEVESIZE) |
| 149 | { |
| 150 | *sp = 1, sp += p; |
| 151 | } |
| 152 | p += addp[ai]; |
| 153 | ai = (ai + 1) % 48; |
| 154 | } |
| 155 | ps->d = 0; |
| 156 | return gmp_nextprime (ps); |
| 157 | } |
| 158 | |
| 159 | void |
| 160 | gmp_init_primesieve (gmp_primesieve_t *ps) |
| 161 | { |
| 162 | ps->s0 = 0; |
| 163 | ps->sqrt_s0 = 0; |
| 164 | ps->d = SIEVESIZE; |
| 165 | ps->s[SIEVESIZE] = 0; /* sentinel */ |
| 166 | } |