strchr.S 2.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293
  1. /* Copyright (C) 1996-2026 Free Software Foundation, Inc.
  2. This file is part of the GNU C Library.
  3. The GNU C Library is free software; you can redistribute it and/or
  4. modify it under the terms of the GNU Lesser General Public
  5. License as published by the Free Software Foundation; either
  6. version 2.1 of the License, or (at your option) any later version.
  7. The GNU C Library is distributed in the hope that it will be useful,
  8. but WITHOUT ANY WARRANTY; without even the implied warranty of
  9. MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  10. Lesser General Public License for more details.
  11. You should have received a copy of the GNU Lesser General Public
  12. License along with the GNU C Library. If not, see
  13. <https://www.gnu.org/licenses/>. */
  14. /* Return the address of a given character within a null-terminated
  15. string, or null if it is not found.
  16. This is generally scheduled for the EV5 (got to look out for my own
  17. interests :-), but with EV4 needs in mind. There *should* be no more
  18. stalls for the EV4 than there are for the EV5.
  19. */
  20. #include <sysdep.h>
  21. .set noreorder
  22. .set noat
  23. ENTRY(strchr)
  24. #ifdef PROF
  25. ldgp gp, 0(pv)
  26. lda AT, _mcount
  27. jsr AT, (AT), _mcount
  28. .prologue 1
  29. #else
  30. .prologue 0
  31. #endif
  32. zapnot a1, 1, a1 # e0 : zero extend the search character
  33. ldq_u t0, 0(a0) # .. e1 : load first quadword
  34. sll a1, 8, t5 # e0 : replicate the search character
  35. andnot a0, 7, v0 # .. e1 : align our loop pointer
  36. or t5, a1, a1 # e0 :
  37. lda t4, -1 # .. e1 : build garbage mask
  38. sll a1, 16, t5 # e0 :
  39. cmpbge zero, t0, t2 # .. e1 : bits set iff byte == zero
  40. mskqh t4, a0, t4 # e0 :
  41. or t5, a1, a1 # .. e1 :
  42. sll a1, 32, t5 # e0 :
  43. cmpbge zero, t4, t4 # .. e1 : bits set iff byte is garbage
  44. or t5, a1, a1 # e0 :
  45. xor t0, a1, t1 # .. e1 : make bytes == c zero
  46. cmpbge zero, t1, t3 # e0 : bits set iff byte == c
  47. or t2, t3, t0 # e1 : bits set iff char match or zero match
  48. andnot t0, t4, t0 # e0 : clear garbage bits
  49. bne t0, $found # .. e1 (zdb)
  50. $loop: ldq t0, 8(v0) # e0 :
  51. addq v0, 8, v0 # .. e1 :
  52. nop # e0 :
  53. xor t0, a1, t1 # .. e1 (ev5 data stall)
  54. cmpbge zero, t0, t2 # e0 : bits set iff byte == 0
  55. cmpbge zero, t1, t3 # .. e1 : bits set iff byte == c
  56. or t2, t3, t0 # e0 :
  57. beq t0, $loop # .. e1 (zdb)
  58. $found: negq t0, t1 # e0 : clear all but least set bit
  59. and t0, t1, t0 # e1 (stall)
  60. and t0, t3, t1 # e0 : bit set iff byte was the char
  61. beq t1, $retnull # .. e1 (zdb)
  62. and t0, 0xf0, t2 # e0 : binary search for that set bit
  63. and t0, 0xcc, t3 # .. e1 :
  64. and t0, 0xaa, t4 # e0 :
  65. cmovne t2, 4, t2 # .. e1 :
  66. cmovne t3, 2, t3 # e0 :
  67. cmovne t4, 1, t4 # .. e1 :
  68. addq t2, t3, t2 # e0 :
  69. addq v0, t4, v0 # .. e1 :
  70. addq v0, t2, v0 # e0 :
  71. ret # .. e1 :
  72. $retnull:
  73. mov zero, v0 # e0 :
  74. ret # .. e1 :
  75. END(strchr)
  76. weak_alias (strchr, index)
  77. libc_hidden_builtin_def (strchr)