strlen.c 3.4 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109
  1. /* Copyright (C) 1991-2019 Free Software Foundation, Inc.
  2. This file is part of the GNU C Library.
  3. Written by Torbjorn Granlund (tege@sics.se),
  4. with help from Dan Sahlin (dan@sics.se);
  5. commentary by Jim Blandy (jimb@ai.mit.edu).
  6. The GNU C Library is free software; you can redistribute it and/or
  7. modify it under the terms of the GNU Lesser General Public
  8. License as published by the Free Software Foundation; either
  9. version 2.1 of the License, or (at your option) any later version.
  10. The GNU C Library is distributed in the hope that it will be useful,
  11. but WITHOUT ANY WARRANTY; without even the implied warranty of
  12. MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  13. Lesser General Public License for more details.
  14. You should have received a copy of the GNU Lesser General Public
  15. License along with the GNU C Library; if not, see
  16. <http://www.gnu.org/licenses/>. */
  17. #include <string.h>
  18. #include <stdlib.h>
  19. #undef strlen
  20. #ifndef STRLEN
  21. # define STRLEN strlen
  22. #endif
  23. /* Return the length of the null-terminated string STR. Scan for
  24. the null terminator quickly by testing four bytes at a time. */
  25. size_t
  26. STRLEN (const char *str)
  27. {
  28. const char *char_ptr;
  29. const unsigned long int *longword_ptr;
  30. unsigned long int longword, himagic, lomagic;
  31. /* Handle the first few characters by reading one character at a time.
  32. Do this until CHAR_PTR is aligned on a longword boundary. */
  33. for (char_ptr = str; ((unsigned long int) char_ptr
  34. & (sizeof (longword) - 1)) != 0;
  35. ++char_ptr)
  36. if (*char_ptr == '\0')
  37. return char_ptr - str;
  38. /* All these elucidatory comments refer to 4-byte longwords,
  39. but the theory applies equally well to 8-byte longwords. */
  40. longword_ptr = (unsigned long int *) char_ptr;
  41. /* Bits 31, 24, 16, and 8 of this number are zero. Call these bits
  42. the "holes." Note that there is a hole just to the left of
  43. each byte, with an extra at the end:
  44. bits: 01111110 11111110 11111110 11111111
  45. bytes: AAAAAAAA BBBBBBBB CCCCCCCC DDDDDDDD
  46. The 1-bits make sure that carries propagate to the next 0-bit.
  47. The 0-bits provide holes for carries to fall into. */
  48. himagic = 0x80808080L;
  49. lomagic = 0x01010101L;
  50. if (sizeof (longword) > 4)
  51. {
  52. /* 64-bit version of the magic. */
  53. /* Do the shift in two steps to avoid a warning if long has 32 bits. */
  54. himagic = ((himagic << 16) << 16) | himagic;
  55. lomagic = ((lomagic << 16) << 16) | lomagic;
  56. }
  57. if (sizeof (longword) > 8)
  58. abort ();
  59. /* Instead of the traditional loop which tests each character,
  60. we will test a longword at a time. The tricky part is testing
  61. if *any of the four* bytes in the longword in question are zero. */
  62. for (;;)
  63. {
  64. longword = *longword_ptr++;
  65. if (((longword - lomagic) & ~longword & himagic) != 0)
  66. {
  67. /* Which of the bytes was the zero? If none of them were, it was
  68. a misfire; continue the search. */
  69. const char *cp = (const char *) (longword_ptr - 1);
  70. if (cp[0] == 0)
  71. return cp - str;
  72. if (cp[1] == 0)
  73. return cp - str + 1;
  74. if (cp[2] == 0)
  75. return cp - str + 2;
  76. if (cp[3] == 0)
  77. return cp - str + 3;
  78. if (sizeof (longword) > 4)
  79. {
  80. if (cp[4] == 0)
  81. return cp - str + 4;
  82. if (cp[5] == 0)
  83. return cp - str + 5;
  84. if (cp[6] == 0)
  85. return cp - str + 6;
  86. if (cp[7] == 0)
  87. return cp - str + 7;
  88. }
  89. }
  90. }
  91. }
  92. libc_hidden_builtin_def (strlen)