Skip to content

F
ffmpeg.wasm-core
Switch branch/tag
Find file
Normal viewHistoryPermalink
vp9dsp_mips.h 13.5 KB
Newer Older
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 MC functions
012ba786
 
Shivraj Patil committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 
/*
 * Copyright (c) 2015 Shivraj Patil (Shivraj.Patil@imgtec.com)
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#ifndef AVCODEC_MIPS_VP9DSP_MIPS_H
#define AVCODEC_MIPS_VP9DSP_MIPS_H

#define VP9_8TAP_MIPS_MSA_FUNC(SIZE, type, type_idx)                         \
void ff_put_8tap_##type##_##SIZE##h_msa(uint8_t *dst, ptrdiff_t dststride,   \
                                        const uint8_t *src,                  \
                                        ptrdiff_t srcstride,                 \
                                        int h, int mx, int my);              \
                                                                             \
void ff_put_8tap_##type##_##SIZE##v_msa(uint8_t *dst, ptrdiff_t dststride,   \
                                        const uint8_t *src,                  \
                                        ptrdiff_t srcstride,                 \
                                        int h, int mx, int my);              \
                                                                             \
void ff_put_8tap_##type##_##SIZE##hv_msa(uint8_t *dst, ptrdiff_t dststride,  \
                                         const uint8_t *src,                 \
                                         ptrdiff_t srcstride,                \
                                         int h, int mx, int my);             \
                                                                             \
void ff_avg_8tap_##type##_##SIZE##h_msa(uint8_t *dst, ptrdiff_t dststride,   \
                                        const uint8_t *src,                  \
                                        ptrdiff_t srcstride,                 \
                                        int h, int mx, int my);              \
                                                                             \
void ff_avg_8tap_##type##_##SIZE##v_msa(uint8_t *dst, ptrdiff_t dststride,   \
                                        const uint8_t *src,                  \
                                        ptrdiff_t srcstride,                 \
                                        int h, int mx, int my);              \
                                                                             \
void ff_avg_8tap_##type##_##SIZE##hv_msa(uint8_t *dst, ptrdiff_t dststride,  \
                                         const uint8_t *src,                 \
                                         ptrdiff_t srcstride,                \
                                         int h, int mx, int my);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 bilinear functions
71aede3c
 
Shivraj Patil committed
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 
#define VP9_BILINEAR_MIPS_MSA_FUNC(SIZE)                                   \
void ff_put_bilin_##SIZE##h_msa(uint8_t *dst, ptrdiff_t dststride,         \
                                const uint8_t *src, ptrdiff_t srcstride,   \
                                int h, int mx, int my);                    \
                                                                           \
void ff_put_bilin_##SIZE##v_msa(uint8_t *dst, ptrdiff_t dststride,         \
                                const uint8_t *src, ptrdiff_t srcstride,   \
                                int h, int mx, int my);                    \
                                                                           \
void ff_put_bilin_##SIZE##hv_msa(uint8_t *dst, ptrdiff_t dststride,        \
                                 const uint8_t *src, ptrdiff_t srcstride,  \
                                 int h, int mx, int my);                   \
                                                                           \
void ff_avg_bilin_##SIZE##h_msa(uint8_t *dst, ptrdiff_t dststride,         \
                                const uint8_t *src, ptrdiff_t srcstride,   \
                                int h, int mx, int my);                    \
                                                                           \
void ff_avg_bilin_##SIZE##v_msa(uint8_t *dst, ptrdiff_t dststride,         \
                                const uint8_t *src, ptrdiff_t srcstride,   \
                                int h, int mx, int my);                    \
                                                                           \
void ff_avg_bilin_##SIZE##hv_msa(uint8_t *dst, ptrdiff_t dststride,        \
                                 const uint8_t *src, ptrdiff_t srcstride,  \
                                 int h, int mx, int my);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 MC functions
012ba786
 
Shivraj Patil committed
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 
#define VP9_COPY_AVG_MIPS_MSA_FUNC(SIZE)                           \
void ff_copy##SIZE##_msa(uint8_t *dst, ptrdiff_t dststride,        \
                         const uint8_t *src, ptrdiff_t srcstride,  \
                         int h, int mx, int my);                   \
                                                                   \
void ff_avg##SIZE##_msa(uint8_t *dst, ptrdiff_t dststride,         \
                        const uint8_t *src, ptrdiff_t srcstride,   \
                        int h, int mx, int my);

VP9_8TAP_MIPS_MSA_FUNC(64, regular, FILTER_8TAP_REGULAR);
VP9_8TAP_MIPS_MSA_FUNC(32, regular, FILTER_8TAP_REGULAR);
VP9_8TAP_MIPS_MSA_FUNC(16, regular, FILTER_8TAP_REGULAR);
VP9_8TAP_MIPS_MSA_FUNC(8, regular, FILTER_8TAP_REGULAR);
VP9_8TAP_MIPS_MSA_FUNC(4, regular, FILTER_8TAP_REGULAR);

VP9_8TAP_MIPS_MSA_FUNC(64, sharp, FILTER_8TAP_SHARP);
VP9_8TAP_MIPS_MSA_FUNC(32, sharp, FILTER_8TAP_SHARP);
VP9_8TAP_MIPS_MSA_FUNC(16, sharp, FILTER_8TAP_SHARP);
VP9_8TAP_MIPS_MSA_FUNC(8, sharp, FILTER_8TAP_SHARP);
VP9_8TAP_MIPS_MSA_FUNC(4, sharp, FILTER_8TAP_SHARP);

VP9_8TAP_MIPS_MSA_FUNC(64, smooth, FILTER_8TAP_SMOOTH);
VP9_8TAP_MIPS_MSA_FUNC(32, smooth, FILTER_8TAP_SMOOTH);
VP9_8TAP_MIPS_MSA_FUNC(16, smooth, FILTER_8TAP_SMOOTH);
VP9_8TAP_MIPS_MSA_FUNC(8, smooth, FILTER_8TAP_SMOOTH);
VP9_8TAP_MIPS_MSA_FUNC(4, smooth, FILTER_8TAP_SMOOTH);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 bilinear functions
71aede3c
 
Shivraj Patil committed
107 108 109 110 111 112 
VP9_BILINEAR_MIPS_MSA_FUNC(64);
VP9_BILINEAR_MIPS_MSA_FUNC(32);
VP9_BILINEAR_MIPS_MSA_FUNC(16);
VP9_BILINEAR_MIPS_MSA_FUNC(8);
VP9_BILINEAR_MIPS_MSA_FUNC(4);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 MC functions
012ba786
 
Shivraj Patil committed
113 114 115 116 117 118 119 
VP9_COPY_AVG_MIPS_MSA_FUNC(64);
VP9_COPY_AVG_MIPS_MSA_FUNC(32);
VP9_COPY_AVG_MIPS_MSA_FUNC(16);
VP9_COPY_AVG_MIPS_MSA_FUNC(8);
VP9_COPY_AVG_MIPS_MSA_FUNC(4);

#undef VP9_8TAP_MIPS_MSA_FUNC
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 bilinear functions
71aede3c
 
Shivraj Patil committed
120 
#undef VP9_BILINEAR_MIPS_MSA_FUNC
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 MC functions
012ba786
 
Shivraj Patil committed
121 122 
#undef VP9_COPY_AVG_MIPS_MSA_FUNC
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 lpf functions
fd7eadd2
 
Shivraj Patil committed
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 
void ff_loop_filter_h_4_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                              int32_t i, int32_t h);
void ff_loop_filter_h_8_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                              int32_t i, int32_t h);
void ff_loop_filter_h_16_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                               int32_t i, int32_t h);
void ff_loop_filter_v_4_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                              int32_t i, int32_t h);
void ff_loop_filter_v_8_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                              int32_t i, int32_t h);
void ff_loop_filter_v_16_8_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                               int32_t i, int32_t h);
void ff_loop_filter_h_44_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_h_88_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_h_16_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_v_44_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_v_88_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_v_16_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_h_48_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_h_84_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_v_48_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
void ff_loop_filter_v_84_16_msa(uint8_t *dst, ptrdiff_t stride, int32_t e,
                                int32_t i, int32_t h);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 idct functions
c03800d5
 
Shivraj Patil committed
155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 
void ff_idct_idct_4x4_add_msa(uint8_t *dst, ptrdiff_t stride,
                              int16_t *block, int eob);
void ff_idct_idct_8x8_add_msa(uint8_t *dst, ptrdiff_t stride,
                              int16_t *block, int eob);
void ff_idct_idct_16x16_add_msa(uint8_t *dst, ptrdiff_t stride,
                                int16_t *block, int eob);
void ff_idct_idct_32x32_add_msa(uint8_t *dst, ptrdiff_t stride,
                                int16_t *block, int eob);
void ff_iadst_iadst_4x4_add_msa(uint8_t *dst, ptrdiff_t stride,
                                int16_t *block, int eob);
void ff_iadst_iadst_8x8_add_msa(uint8_t *dst, ptrdiff_t stride,
                                int16_t *block, int eob);
void ff_iadst_iadst_16x16_add_msa(uint8_t *dst, ptrdiff_t stride,
                                  int16_t *block, int eob);
void ff_iadst_idct_4x4_add_msa(uint8_t *dst, ptrdiff_t stride,
                               int16_t *block, int eob);
void ff_iadst_idct_8x8_add_msa(uint8_t *dst, ptrdiff_t stride,
                               int16_t *block, int eob);
void ff_iadst_idct_16x16_add_msa(uint8_t *dst, ptrdiff_t stride,
                                 int16_t *block, int eob);
void ff_idct_iadst_4x4_add_msa(uint8_t *pu8Dest, ptrdiff_t stride,
                               int16_t *block, int eob);
void ff_idct_iadst_8x8_add_msa(uint8_t *pu8Dest, ptrdiff_t stride,
                               int16_t *block, int eob);
void ff_idct_iadst_16x16_add_msa(uint8_t *pu8Dest, ptrdiff_t stride,
                                 int16_t *block, int eob);
void ff_iwht_iwht_4x4_add_msa(uint8_t *dst, ptrdiff_t stride,
                              int16_t *block, int eob);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 lpf functions
fd7eadd2
 
Shivraj Patil committed
183
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 intra functions
e21b090b
 
Shivraj Patil committed
184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 
void ff_vert_16x16_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                       const uint8_t *top);
void ff_vert_32x32_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                       const uint8_t *top);
void ff_hor_16x16_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                      const uint8_t *top);
void ff_hor_32x32_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                      const uint8_t *top);
void ff_dc_4x4_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                   const uint8_t *top);
void ff_dc_8x8_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                   const uint8_t *top);
void ff_dc_16x16_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                     const uint8_t *top);
void ff_dc_32x32_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                     const uint8_t *top);
void ff_dc_left_4x4_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                        const uint8_t *top);
void ff_dc_left_8x8_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                        const uint8_t *top);
void ff_dc_left_16x16_msa(uint8_t *dst, ptrdiff_t stride,
                          const uint8_t *left, const uint8_t *top);
void ff_dc_left_32x32_msa(uint8_t *dst, ptrdiff_t stride,
                          const uint8_t *left, const uint8_t *top);
void ff_dc_top_4x4_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                       const uint8_t *top);
void ff_dc_top_8x8_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                       const uint8_t *top);
void ff_dc_top_16x16_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_top_32x32_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_128_16x16_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_128_32x32_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_127_16x16_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_127_32x32_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_129_16x16_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_dc_129_32x32_msa(uint8_t *dst, ptrdiff_t stride,
                         const uint8_t *left, const uint8_t *top);
void ff_tm_4x4_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                   const uint8_t *top);
void ff_tm_8x8_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                   const uint8_t *top);
void ff_tm_16x16_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                     const uint8_t *top);
void ff_tm_32x32_msa(uint8_t *dst, ptrdiff_t stride, const uint8_t *left,
                     const uint8_t *top);
Shivraj Patil's avatar
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for VP9 MC functions
012ba786
 
Shivraj Patil committed
237 
#endif  // #ifndef AVCODEC_MIPS_VP9DSP_MIPS_H