| Line | Branch | Exec | Source |
|---|---|---|---|
| 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 | 534693 | static void vector_fmul_c(float *dst, const float *src0, const float *src1, | |
| 28 | int len) | ||
| 29 | { | ||
| 30 | int i; | ||
| 31 |
2/2✓ Branch 0 taken 61003904 times.
✓ Branch 1 taken 534693 times.
|
61538597 | for (i = 0; i < len; i++) |
| 32 | 61003904 | dst[i] = src0[i] * src1[i]; | |
| 33 | 534693 | } | |
| 34 | |||
| 35 | 3 | static void vector_dmul_c(double *dst, const double *src0, const double *src1, | |
| 36 | int len) | ||
| 37 | { | ||
| 38 | int i; | ||
| 39 |
2/2✓ Branch 0 taken 768 times.
✓ Branch 1 taken 3 times.
|
771 | for (i = 0; i < len; i++) |
| 40 | 768 | dst[i] = src0[i] * src1[i]; | |
| 41 | 3 | } | |
| 42 | |||
| 43 | 24577 | static void vector_fmac_scalar_c(float *dst, const float *src, float mul, | |
| 44 | int len) | ||
| 45 | { | ||
| 46 | int i; | ||
| 47 |
2/2✓ Branch 0 taken 11631008 times.
✓ Branch 1 taken 24577 times.
|
11655585 | for (i = 0; i < len; i++) |
| 48 | 11631008 | dst[i] += src[i] * mul; | |
| 49 | 24577 | } | |
| 50 | |||
| 51 | 3 | static void vector_dmac_scalar_c(double *dst, const double *src, double mul, | |
| 52 | int len) | ||
| 53 | { | ||
| 54 | int i; | ||
| 55 |
2/2✓ Branch 0 taken 768 times.
✓ Branch 1 taken 3 times.
|
771 | for (i = 0; i < len; i++) |
| 56 | 768 | dst[i] += src[i] * mul; | |
| 57 | 3 | } | |
| 58 | |||
| 59 | 859511 | static void vector_fmul_scalar_c(float *dst, const float *src, float mul, | |
| 60 | int len) | ||
| 61 | { | ||
| 62 | int i; | ||
| 63 |
2/2✓ Branch 0 taken 21713492 times.
✓ Branch 1 taken 859511 times.
|
22573003 | for (i = 0; i < len; i++) |
| 64 | 21713492 | dst[i] = src[i] * mul; | |
| 65 | 859511 | } | |
| 66 | |||
| 67 | 3 | static void vector_dmul_scalar_c(double *dst, const double *src, double mul, | |
| 68 | int len) | ||
| 69 | { | ||
| 70 | int i; | ||
| 71 |
2/2✓ Branch 0 taken 768 times.
✓ Branch 1 taken 3 times.
|
771 | for (i = 0; i < len; i++) |
| 72 | 768 | dst[i] = src[i] * mul; | |
| 73 | 3 | } | |
| 74 | |||
| 75 | 221265 | 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 | 221265 | dst += len; | |
| 81 | 221265 | win += len; | |
| 82 | 221265 | src0 += len; | |
| 83 | |||
| 84 |
2/2✓ Branch 0 taken 44055444 times.
✓ Branch 1 taken 221265 times.
|
44276709 | for (i = -len, j = len - 1; i < 0; i++, j--) { |
| 85 | 44055444 | float s0 = src0[i]; | |
| 86 | 44055444 | float s1 = src1[j]; | |
| 87 | 44055444 | float wi = win[i]; | |
| 88 | 44055444 | float wj = win[j]; | |
| 89 | 44055444 | dst[i] = s0 * wj - s1 * wi; | |
| 90 | 44055444 | dst[j] = s0 * wi + s1 * wj; | |
| 91 | } | ||
| 92 | 221265 | } | |
| 93 | |||
| 94 | 3865569 | 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 |
2/2✓ Branch 0 taken 231203968 times.
✓ Branch 1 taken 3865569 times.
|
235069537 | for (i = 0; i < len; i++) |
| 99 | 231203968 | dst[i] = src0[i] * src1[i] + src2[i]; | |
| 100 | 3865569 | } | |
| 101 | |||
| 102 | 406010 | static void vector_fmul_reverse_c(float *dst, const float *src0, | |
| 103 | const float *src1, int len) | ||
| 104 | { | ||
| 105 | int i; | ||
| 106 | |||
| 107 | 406010 | src1 += len-1; | |
| 108 |
2/2✓ Branch 0 taken 133347968 times.
✓ Branch 1 taken 406010 times.
|
133753978 | for (i = 0; i < len; i++) |
| 109 | 133347968 | dst[i] = src0[i] * src1[-i]; | |
| 110 | 406010 | } | |
| 111 | |||
| 112 | 590386 | static void butterflies_float_c(float *restrict v1, float *restrict v2, | |
| 113 | int len) | ||
| 114 | { | ||
| 115 | int i; | ||
| 116 | |||
| 117 |
2/2✓ Branch 0 taken 8699528 times.
✓ Branch 1 taken 590386 times.
|
9289914 | for (i = 0; i < len; i++) { |
| 118 | 8699528 | float t = v1[i] - v2[i]; | |
| 119 | 8699528 | v1[i] += v2[i]; | |
| 120 | 8699528 | v2[i] = t; | |
| 121 | } | ||
| 122 | 590386 | } | |
| 123 | |||
| 124 | 836465 | double ff_scalarproduct_double_c(const double *v1, const double *v2, | |
| 125 | size_t len) | ||
| 126 | { | ||
| 127 | 836465 | double p = 0.0; | |
| 128 | |||
| 129 |
2/2✓ Branch 0 taken 6692578 times.
✓ Branch 1 taken 836465 times.
|
7529043 | for (size_t i = 0; i < len; i++) |
| 130 | 6692578 | p += v1[i] * v2[i]; | |
| 131 | |||
| 132 | 836465 | return p; | |
| 133 | } | ||
| 134 | |||
| 135 | 1216 | av_cold AVFloatDSPContext *avpriv_float_dsp_alloc(int bit_exact) | |
| 136 | { | ||
| 137 | 1216 | AVFloatDSPContext *fdsp = av_mallocz(sizeof(AVFloatDSPContext)); | |
| 138 |
1/2✗ Branch 0 not taken.
✓ Branch 1 taken 1216 times.
|
1216 | if (!fdsp) |
| 139 | ✗ | return NULL; | |
| 140 | |||
| 141 | 1216 | fdsp->vector_fmul = vector_fmul_c; | |
| 142 | 1216 | fdsp->vector_dmul = vector_dmul_c; | |
| 143 | 1216 | fdsp->vector_fmac_scalar = vector_fmac_scalar_c; | |
| 144 | 1216 | fdsp->vector_fmul_scalar = vector_fmul_scalar_c; | |
| 145 | 1216 | fdsp->vector_dmac_scalar = vector_dmac_scalar_c; | |
| 146 | 1216 | fdsp->vector_dmul_scalar = vector_dmul_scalar_c; | |
| 147 | 1216 | fdsp->vector_fmul_window = vector_fmul_window_c; | |
| 148 | 1216 | fdsp->vector_fmul_add = vector_fmul_add_c; | |
| 149 | 1216 | fdsp->vector_fmul_reverse = vector_fmul_reverse_c; | |
| 150 | 1216 | fdsp->butterflies_float = butterflies_float_c; | |
| 151 | 1216 | fdsp->scalarproduct_float = ff_scalarproduct_float_c; | |
| 152 | 1216 | fdsp->scalarproduct_double = ff_scalarproduct_double_c; | |
| 153 | |||
| 154 | #if ARCH_AARCH64 | ||
| 155 | ff_float_dsp_init_aarch64(fdsp); | ||
| 156 | #elif ARCH_ARM | ||
| 157 | ff_float_dsp_init_arm(fdsp); | ||
| 158 | #elif ARCH_PPC | ||
| 159 | ff_float_dsp_init_ppc(fdsp, bit_exact); | ||
| 160 | #elif ARCH_RISCV | ||
| 161 | ff_float_dsp_init_riscv(fdsp); | ||
| 162 | #elif ARCH_X86 && HAVE_X86ASM | ||
| 163 | 1216 | ff_float_dsp_init_x86(fdsp); | |
| 164 | #elif ARCH_MIPS | ||
| 165 | ff_float_dsp_init_mips(fdsp); | ||
| 166 | #endif | ||
| 167 | 1216 | return fdsp; | |
| 168 | } | ||
| 169 |