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.

144 lines
4.0KB

  1. /*
  2. * RV30/40 decoder common dsp functions
  3. * Copyright (c) 2007 Mike Melanson, Konstantin Shishkov
  4. * Copyright (c) 2011 Janne Grunau
  5. *
  6. * This file is part of FFmpeg.
  7. *
  8. * FFmpeg is free software; you can redistribute it and/or
  9. * modify it under the terms of the GNU Lesser General Public
  10. * License as published by the Free Software Foundation; either
  11. * version 2.1 of the License, or (at your option) any later version.
  12. *
  13. * FFmpeg is distributed in the hope that it will be useful,
  14. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  16. * Lesser General Public License for more details.
  17. *
  18. * You should have received a copy of the GNU Lesser General Public
  19. * License along with FFmpeg; if not, write to the Free Software
  20. * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  21. */
  22. /**
  23. * @file
  24. * RV30/40 decoder common dsp functions
  25. */
  26. #include "rv34dsp.h"
  27. #include "libavutil/common.h"
  28. /**
  29. * @name RV30/40 inverse transform functions
  30. * @{
  31. */
  32. static av_always_inline void rv34_row_transform(int temp[16], int16_t *block)
  33. {
  34. int i;
  35. for(i = 0; i < 4; i++){
  36. const int z0 = 13*(block[i+4*0] + block[i+4*2]);
  37. const int z1 = 13*(block[i+4*0] - block[i+4*2]);
  38. const int z2 = 7* block[i+4*1] - 17*block[i+4*3];
  39. const int z3 = 17* block[i+4*1] + 7*block[i+4*3];
  40. temp[4*i+0] = z0 + z3;
  41. temp[4*i+1] = z1 + z2;
  42. temp[4*i+2] = z1 - z2;
  43. temp[4*i+3] = z0 - z3;
  44. }
  45. }
  46. /**
  47. * Real Video 3.0/4.0 inverse transform + sample reconstruction
  48. * Code is almost the same as in SVQ3, only scaling is different.
  49. */
  50. static void rv34_idct_add_c(uint8_t *dst, ptrdiff_t stride, int16_t *block){
  51. int temp[16];
  52. int i;
  53. rv34_row_transform(temp, block);
  54. memset(block, 0, 16*sizeof(int16_t));
  55. for(i = 0; i < 4; i++){
  56. const int z0 = 13*(temp[4*0+i] + temp[4*2+i]) + 0x200;
  57. const int z1 = 13*(temp[4*0+i] - temp[4*2+i]) + 0x200;
  58. const int z2 = 7* temp[4*1+i] - 17*temp[4*3+i];
  59. const int z3 = 17* temp[4*1+i] + 7*temp[4*3+i];
  60. dst[0] = av_clip_uint8( dst[0] + ( (z0 + z3) >> 10 ) );
  61. dst[1] = av_clip_uint8( dst[1] + ( (z1 + z2) >> 10 ) );
  62. dst[2] = av_clip_uint8( dst[2] + ( (z1 - z2) >> 10 ) );
  63. dst[3] = av_clip_uint8( dst[3] + ( (z0 - z3) >> 10 ) );
  64. dst += stride;
  65. }
  66. }
  67. /**
  68. * RealVideo 3.0/4.0 inverse transform for DC block
  69. *
  70. * Code is almost the same as rv34_inv_transform()
  71. * but final coefficients are multiplied by 1.5 and have no rounding.
  72. */
  73. static void rv34_inv_transform_noround_c(int16_t *block){
  74. int temp[16];
  75. int i;
  76. rv34_row_transform(temp, block);
  77. for(i = 0; i < 4; i++){
  78. const int z0 = 39*(temp[4*0+i] + temp[4*2+i]);
  79. const int z1 = 39*(temp[4*0+i] - temp[4*2+i]);
  80. const int z2 = 21* temp[4*1+i] - 51*temp[4*3+i];
  81. const int z3 = 51* temp[4*1+i] + 21*temp[4*3+i];
  82. block[i*4+0] = (z0 + z3) >> 11;
  83. block[i*4+1] = (z1 + z2) >> 11;
  84. block[i*4+2] = (z1 - z2) >> 11;
  85. block[i*4+3] = (z0 - z3) >> 11;
  86. }
  87. }
  88. static void rv34_idct_dc_add_c(uint8_t *dst, ptrdiff_t stride, int dc)
  89. {
  90. int i, j;
  91. dc = (13*13*dc + 0x200) >> 10;
  92. for (i = 0; i < 4; i++)
  93. {
  94. for (j = 0; j < 4; j++)
  95. dst[j] = av_clip_uint8( dst[j] + dc );
  96. dst += stride;
  97. }
  98. }
  99. static void rv34_inv_transform_dc_noround_c(int16_t *block)
  100. {
  101. int16_t dc = (13 * 13 * 3 * block[0]) >> 11;
  102. int i, j;
  103. for (i = 0; i < 4; i++, block += 4)
  104. for (j = 0; j < 4; j++)
  105. block[j] = dc;
  106. }
  107. /** @} */ // transform
  108. av_cold void ff_rv34dsp_init(RV34DSPContext *c)
  109. {
  110. c->rv34_inv_transform = rv34_inv_transform_noround_c;
  111. c->rv34_inv_transform_dc = rv34_inv_transform_dc_noround_c;
  112. c->rv34_idct_add = rv34_idct_add_c;
  113. c->rv34_idct_dc_add = rv34_idct_dc_add_c;
  114. if (ARCH_ARM)
  115. ff_rv34dsp_init_arm(c);
  116. if (ARCH_X86)
  117. ff_rv34dsp_init_x86(c);
  118. }