float_dsp.c 4.4 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164
  1. /*
  2. * Copyright 2005 Balatoni Denes
  3. * Copyright 2006 Loren Merritt
  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. #include "config.h"
  22. #include "attributes.h"
  23. #include "float_dsp.h"
  24. #include "mem.h"
  25. static void vector_fmul_c(float *dst, const float *src0, const float *src1,
  26. int len)
  27. {
  28. int i;
  29. for (i = 0; i < len; i++)
  30. dst[i] = src0[i] * src1[i];
  31. }
  32. static void vector_dmul_c(double *dst, const double *src0, const double *src1,
  33. int len)
  34. {
  35. int i;
  36. for (i = 0; i < len; i++)
  37. dst[i] = src0[i] * src1[i];
  38. }
  39. static void vector_fmac_scalar_c(float *dst, const float *src, float mul,
  40. int len)
  41. {
  42. int i;
  43. for (i = 0; i < len; i++)
  44. dst[i] += src[i] * mul;
  45. }
  46. static void vector_dmac_scalar_c(double *dst, const double *src, double mul,
  47. int len)
  48. {
  49. int i;
  50. for (i = 0; i < len; i++)
  51. dst[i] += src[i] * mul;
  52. }
  53. static void vector_fmul_scalar_c(float *dst, const float *src, float mul,
  54. int len)
  55. {
  56. int i;
  57. for (i = 0; i < len; i++)
  58. dst[i] = src[i] * mul;
  59. }
  60. static void vector_dmul_scalar_c(double *dst, const double *src, double mul,
  61. int len)
  62. {
  63. int i;
  64. for (i = 0; i < len; i++)
  65. dst[i] = src[i] * mul;
  66. }
  67. static void vector_fmul_window_c(float *dst, const float *src0,
  68. const float *src1, const float *win, int len)
  69. {
  70. int i, j;
  71. dst += len;
  72. win += len;
  73. src0 += len;
  74. for (i = -len, j = len - 1; i < 0; i++, j--) {
  75. float s0 = src0[i];
  76. float s1 = src1[j];
  77. float wi = win[i];
  78. float wj = win[j];
  79. dst[i] = s0 * wj - s1 * wi;
  80. dst[j] = s0 * wi + s1 * wj;
  81. }
  82. }
  83. static void vector_fmul_add_c(float *dst, const float *src0, const float *src1,
  84. const float *src2, int len){
  85. int i;
  86. for (i = 0; i < len; i++)
  87. dst[i] = src0[i] * src1[i] + src2[i];
  88. }
  89. static void vector_fmul_reverse_c(float *dst, const float *src0,
  90. const float *src1, int len)
  91. {
  92. int i;
  93. src1 += len-1;
  94. for (i = 0; i < len; i++)
  95. dst[i] = src0[i] * src1[-i];
  96. }
  97. static void butterflies_float_c(float *av_restrict v1, float *av_restrict v2,
  98. int len)
  99. {
  100. int i;
  101. for (i = 0; i < len; i++) {
  102. float t = v1[i] - v2[i];
  103. v1[i] += v2[i];
  104. v2[i] = t;
  105. }
  106. }
  107. float avpriv_scalarproduct_float_c(const float *v1, const float *v2, int len)
  108. {
  109. float p = 0.0;
  110. int i;
  111. for (i = 0; i < len; i++)
  112. p += v1[i] * v2[i];
  113. return p;
  114. }
  115. av_cold AVFloatDSPContext *avpriv_float_dsp_alloc(int bit_exact)
  116. {
  117. AVFloatDSPContext *fdsp = av_mallocz(sizeof(AVFloatDSPContext));
  118. if (!fdsp)
  119. return NULL;
  120. fdsp->vector_fmul = vector_fmul_c;
  121. fdsp->vector_dmul = vector_dmul_c;
  122. fdsp->vector_fmac_scalar = vector_fmac_scalar_c;
  123. fdsp->vector_fmul_scalar = vector_fmul_scalar_c;
  124. fdsp->vector_dmac_scalar = vector_dmac_scalar_c;
  125. fdsp->vector_dmul_scalar = vector_dmul_scalar_c;
  126. fdsp->vector_fmul_window = vector_fmul_window_c;
  127. fdsp->vector_fmul_add = vector_fmul_add_c;
  128. fdsp->vector_fmul_reverse = vector_fmul_reverse_c;
  129. fdsp->butterflies_float = butterflies_float_c;
  130. fdsp->scalarproduct_float = avpriv_scalarproduct_float_c;
  131. if (ARCH_AARCH64)
  132. ff_float_dsp_init_aarch64(fdsp);
  133. if (ARCH_ARM)
  134. ff_float_dsp_init_arm(fdsp);
  135. if (ARCH_PPC)
  136. ff_float_dsp_init_ppc(fdsp, bit_exact);
  137. if (ARCH_X86)
  138. ff_float_dsp_init_x86(fdsp);
  139. if (ARCH_MIPS)
  140. ff_float_dsp_init_mips(fdsp);
  141. return fdsp;
  142. }