float_dsp.c (4780B)
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 22 #include "config.h" 23 #include "attributes.h" 24 #include "float_dsp.h" 25 #include "mem.h" 26 27 static void vector_fmul_c(float *dst, const float *src0, const float *src1, 28 int len) 29 { 30 int i; 31 for (i = 0; i < len; i++) 32 dst[i] = src0[i] * src1[i]; 33 } 34 35 static void vector_dmul_c(double *dst, const double *src0, const double *src1, 36 int len) 37 { 38 int i; 39 for (i = 0; i < len; i++) 40 dst[i] = src0[i] * src1[i]; 41 } 42 43 static void vector_fmac_scalar_c(float *dst, const float *src, float mul, 44 int len) 45 { 46 int i; 47 for (i = 0; i < len; i++) 48 dst[i] += src[i] * mul; 49 } 50 51 static void vector_dmac_scalar_c(double *dst, const double *src, double mul, 52 int len) 53 { 54 int i; 55 for (i = 0; i < len; i++) 56 dst[i] += src[i] * mul; 57 } 58 59 static void vector_fmul_scalar_c(float *dst, const float *src, float mul, 60 int len) 61 { 62 int i; 63 for (i = 0; i < len; i++) 64 dst[i] = src[i] * mul; 65 } 66 67 static void vector_dmul_scalar_c(double *dst, const double *src, double mul, 68 int len) 69 { 70 int i; 71 for (i = 0; i < len; i++) 72 dst[i] = src[i] * mul; 73 } 74 75 static void vector_fmul_window_c(float *dst, const float *src0, 76 const float *src1, const float *win, int len) 77 { 78 int i, j; 79 80 dst += len; 81 win += len; 82 src0 += len; 83 84 for (i = -len, j = len - 1; i < 0; i++, j--) { 85 float s0 = src0[i]; 86 float s1 = src1[j]; 87 float wi = win[i]; 88 float wj = win[j]; 89 dst[i] = s0 * wj - s1 * wi; 90 dst[j] = s0 * wi + s1 * wj; 91 } 92 } 93 94 static void vector_fmul_add_c(float *dst, const float *src0, const float *src1, 95 const float *src2, int len){ 96 int i; 97 98 for (i = 0; i < len; i++) 99 dst[i] = src0[i] * src1[i] + src2[i]; 100 } 101 102 static void vector_fmul_reverse_c(float *dst, const float *src0, 103 const float *src1, int len) 104 { 105 int i; 106 107 src1 += len-1; 108 for (i = 0; i < len; i++) 109 dst[i] = src0[i] * src1[-i]; 110 } 111 112 static void butterflies_float_c(float *restrict v1, float *restrict v2, 113 int len) 114 { 115 int i; 116 117 for (i = 0; i < len; i++) { 118 float t = v1[i] - v2[i]; 119 v1[i] += v2[i]; 120 v2[i] = t; 121 } 122 } 123 124 float avpriv_scalarproduct_float_c(const float *v1, const float *v2, int len) 125 { 126 float p = 0.0; 127 int i; 128 129 for (i = 0; i < len; i++) 130 p += v1[i] * v2[i]; 131 132 return p; 133 } 134 135 double ff_scalarproduct_double_c(const double *v1, const double *v2, 136 size_t len) 137 { 138 double p = 0.0; 139 140 for (size_t i = 0; i < len; i++) 141 p += v1[i] * v2[i]; 142 143 return p; 144 } 145 146 av_cold AVFloatDSPContext *avpriv_float_dsp_alloc(int bit_exact) 147 { 148 AVFloatDSPContext *fdsp = av_mallocz(sizeof(AVFloatDSPContext)); 149 if (!fdsp) 150 return NULL; 151 152 fdsp->vector_fmul = vector_fmul_c; 153 fdsp->vector_dmul = vector_dmul_c; 154 fdsp->vector_fmac_scalar = vector_fmac_scalar_c; 155 fdsp->vector_fmul_scalar = vector_fmul_scalar_c; 156 fdsp->vector_dmac_scalar = vector_dmac_scalar_c; 157 fdsp->vector_dmul_scalar = vector_dmul_scalar_c; 158 fdsp->vector_fmul_window = vector_fmul_window_c; 159 fdsp->vector_fmul_add = vector_fmul_add_c; 160 fdsp->vector_fmul_reverse = vector_fmul_reverse_c; 161 fdsp->butterflies_float = butterflies_float_c; 162 fdsp->scalarproduct_float = avpriv_scalarproduct_float_c; 163 fdsp->scalarproduct_double = ff_scalarproduct_double_c; 164 165 #if ARCH_AARCH64 166 ff_float_dsp_init_aarch64(fdsp); 167 #elif ARCH_ARM 168 ff_float_dsp_init_arm(fdsp); 169 #elif ARCH_PPC 170 ff_float_dsp_init_ppc(fdsp, bit_exact); 171 #elif ARCH_RISCV 172 ff_float_dsp_init_riscv(fdsp); 173 #elif ARCH_X86 174 ff_float_dsp_init_x86(fdsp); 175 #elif ARCH_MIPS 176 ff_float_dsp_init_mips(fdsp); 177 #endif 178 return fdsp; 179 }