You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

113 lines
3.5KB

  1. /*
  2. * MMX optimized DSP utils
  3. * Copyright (c) 2007 Aurelien Jacobs <aurel@gnuage.org>
  4. *
  5. * This file is part of FFmpeg.
  6. *
  7. * FFmpeg is free software; you can redistribute it and/or
  8. * modify it under the terms of the GNU Lesser General Public
  9. * License as published by the Free Software Foundation; either
  10. * version 2.1 of the License, or (at your option) any later version.
  11. *
  12. * FFmpeg is distributed in the hope that it will be useful,
  13. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  15. * Lesser General Public License for more details.
  16. *
  17. * You should have received a copy of the GNU Lesser General Public
  18. * License along with FFmpeg; if not, write to the Free Software
  19. * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  20. */
  21. #ifndef FFMPEG_DSPUTIL_MMX_H
  22. #define FFMPEG_DSPUTIL_MMX_H
  23. #include <stdint.h>
  24. extern const uint64_t ff_bone;
  25. extern const uint64_t ff_wtwo;
  26. extern const uint64_t ff_pdw_80000000[2];
  27. extern const uint64_t ff_pw_3;
  28. extern const uint64_t ff_pw_4;
  29. extern const uint64_t ff_pw_5;
  30. extern const uint64_t ff_pw_8;
  31. extern const uint64_t ff_pw_15;
  32. extern const uint64_t ff_pw_16;
  33. extern const uint64_t ff_pw_20;
  34. extern const uint64_t ff_pw_32;
  35. extern const uint64_t ff_pw_42;
  36. extern const uint64_t ff_pw_64;
  37. extern const uint64_t ff_pw_96;
  38. extern const uint64_t ff_pw_128;
  39. extern const uint64_t ff_pb_1;
  40. extern const uint64_t ff_pb_3;
  41. extern const uint64_t ff_pb_7;
  42. extern const uint64_t ff_pb_3F;
  43. extern const uint64_t ff_pb_A1;
  44. extern const uint64_t ff_pb_FC;
  45. extern const double ff_pd_1[2];
  46. extern const double ff_pd_2[2];
  47. /* in/out: mma=mma+mmb, mmb=mmb-mma */
  48. #define SUMSUB_BA( a, b ) \
  49. "paddw "#b", "#a" \n\t"\
  50. "paddw "#b", "#b" \n\t"\
  51. "psubw "#a", "#b" \n\t"
  52. #define SBUTTERFLY(a,b,t,n,m)\
  53. "mov" #m " " #a ", " #t " \n\t" /* abcd */\
  54. "punpckl" #n " " #b ", " #a " \n\t" /* aebf */\
  55. "punpckh" #n " " #b ", " #t " \n\t" /* cgdh */\
  56. #define TRANSPOSE4(a,b,c,d,t)\
  57. SBUTTERFLY(a,b,t,wd,q) /* a=aebf t=cgdh */\
  58. SBUTTERFLY(c,d,b,wd,q) /* c=imjn b=kolp */\
  59. SBUTTERFLY(a,c,d,dq,q) /* a=aeim d=bfjn */\
  60. SBUTTERFLY(t,b,c,dq,q) /* t=cgko c=dhlp */
  61. #ifdef ARCH_X86_64
  62. // permutes 01234567 -> 05736421
  63. #define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\
  64. SBUTTERFLY(a,b,%%xmm8,wd,dqa)\
  65. SBUTTERFLY(c,d,b,wd,dqa)\
  66. SBUTTERFLY(e,f,d,wd,dqa)\
  67. SBUTTERFLY(g,h,f,wd,dqa)\
  68. SBUTTERFLY(a,c,h,dq,dqa)\
  69. SBUTTERFLY(%%xmm8,b,c,dq,dqa)\
  70. SBUTTERFLY(e,g,b,dq,dqa)\
  71. SBUTTERFLY(d,f,g,dq,dqa)\
  72. SBUTTERFLY(a,e,f,qdq,dqa)\
  73. SBUTTERFLY(%%xmm8,d,e,qdq,dqa)\
  74. SBUTTERFLY(h,b,d,qdq,dqa)\
  75. SBUTTERFLY(c,g,b,qdq,dqa)\
  76. "movdqa %%xmm8, "#g" \n\t"
  77. #else
  78. #define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\
  79. "movdqa "#h", "#t" \n\t"\
  80. SBUTTERFLY(a,b,h,wd,dqa)\
  81. "movdqa "#h", 16"#t" \n\t"\
  82. "movdqa "#t", "#h" \n\t"\
  83. SBUTTERFLY(c,d,b,wd,dqa)\
  84. SBUTTERFLY(e,f,d,wd,dqa)\
  85. SBUTTERFLY(g,h,f,wd,dqa)\
  86. SBUTTERFLY(a,c,h,dq,dqa)\
  87. "movdqa "#h", "#t" \n\t"\
  88. "movdqa 16"#t", "#h" \n\t"\
  89. SBUTTERFLY(h,b,c,dq,dqa)\
  90. SBUTTERFLY(e,g,b,dq,dqa)\
  91. SBUTTERFLY(d,f,g,dq,dqa)\
  92. SBUTTERFLY(a,e,f,qdq,dqa)\
  93. SBUTTERFLY(h,d,e,qdq,dqa)\
  94. "movdqa "#h", 16"#t" \n\t"\
  95. "movdqa "#t", "#h" \n\t"\
  96. SBUTTERFLY(h,b,d,qdq,dqa)\
  97. SBUTTERFLY(c,g,b,qdq,dqa)\
  98. "movdqa 16"#t", "#g" \n\t"
  99. #endif
  100. #endif /* FFMPEG_DSPUTIL_MMX_H */