| 1 | /* |
| 2 | * Copyright (c) 2000, 2001 Fabrice Bellard |
| 3 | * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at> |
| 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 "libavutil/avassert.h" |
| 23 | |
| 24 | #include "bit_depth_template.c" |
| 25 | |
| 26 | #define H264_CHROMA_MC(OPNAME, OP)\ |
| 27 | static void FUNCC(OPNAME ## h264_chroma_mc1)(uint8_t *_dst/*align 8*/, uint8_t *_src/*align 1*/, int stride, int h, int x, int y){\ |
| 28 | pixel *dst = (pixel*)_dst;\ |
| 29 | pixel *src = (pixel*)_src;\ |
| 30 | const int A=(8-x)*(8-y);\ |
| 31 | const int B=( x)*(8-y);\ |
| 32 | const int C=(8-x)*( y);\ |
| 33 | const int D=( x)*( y);\ |
| 34 | int i;\ |
| 35 | stride >>= sizeof(pixel)-1;\ |
| 36 | \ |
| 37 | av_assert2(x<8 && y<8 && x>=0 && y>=0);\ |
| 38 | \ |
| 39 | if(D){\ |
| 40 | for(i=0; i<h; i++){\ |
| 41 | OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\ |
| 42 | dst+= stride;\ |
| 43 | src+= stride;\ |
| 44 | }\ |
| 45 | } else if (B + C) {\ |
| 46 | const int E= B+C;\ |
| 47 | const int step= C ? stride : 1;\ |
| 48 | for(i=0; i<h; i++){\ |
| 49 | OP(dst[0], (A*src[0] + E*src[step+0]));\ |
| 50 | dst+= stride;\ |
| 51 | src+= stride;\ |
| 52 | }\ |
| 53 | } else {\ |
| 54 | for(i=0; i<h; i++){\ |
| 55 | OP(dst[0], (A*src[0]));\ |
| 56 | dst+= stride;\ |
| 57 | src+= stride;\ |
| 58 | }\ |
| 59 | }\ |
| 60 | }\ |
| 61 | static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *_dst/*align 8*/, uint8_t *_src/*align 1*/, int stride, int h, int x, int y){\ |
| 62 | pixel *dst = (pixel*)_dst;\ |
| 63 | pixel *src = (pixel*)_src;\ |
| 64 | const int A=(8-x)*(8-y);\ |
| 65 | const int B=( x)*(8-y);\ |
| 66 | const int C=(8-x)*( y);\ |
| 67 | const int D=( x)*( y);\ |
| 68 | int i;\ |
| 69 | stride >>= sizeof(pixel)-1;\ |
| 70 | \ |
| 71 | av_assert2(x<8 && y<8 && x>=0 && y>=0);\ |
| 72 | \ |
| 73 | if(D){\ |
| 74 | for(i=0; i<h; i++){\ |
| 75 | OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\ |
| 76 | OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\ |
| 77 | dst+= stride;\ |
| 78 | src+= stride;\ |
| 79 | }\ |
| 80 | } else if (B + C) {\ |
| 81 | const int E= B+C;\ |
| 82 | const int step= C ? stride : 1;\ |
| 83 | for(i=0; i<h; i++){\ |
| 84 | OP(dst[0], (A*src[0] + E*src[step+0]));\ |
| 85 | OP(dst[1], (A*src[1] + E*src[step+1]));\ |
| 86 | dst+= stride;\ |
| 87 | src+= stride;\ |
| 88 | }\ |
| 89 | } else {\ |
| 90 | for ( i = 0; i < h; i++){\ |
| 91 | OP(dst[0], A * src[0]);\ |
| 92 | OP(dst[1], A * src[1]);\ |
| 93 | dst += stride;\ |
| 94 | src += stride;\ |
| 95 | }\ |
| 96 | }\ |
| 97 | }\ |
| 98 | \ |
| 99 | static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *_dst/*align 8*/, uint8_t *_src/*align 1*/, int stride, int h, int x, int y){\ |
| 100 | pixel *dst = (pixel*)_dst;\ |
| 101 | pixel *src = (pixel*)_src;\ |
| 102 | const int A=(8-x)*(8-y);\ |
| 103 | const int B=( x)*(8-y);\ |
| 104 | const int C=(8-x)*( y);\ |
| 105 | const int D=( x)*( y);\ |
| 106 | int i;\ |
| 107 | stride >>= sizeof(pixel)-1;\ |
| 108 | \ |
| 109 | av_assert2(x<8 && y<8 && x>=0 && y>=0);\ |
| 110 | \ |
| 111 | if(D){\ |
| 112 | for(i=0; i<h; i++){\ |
| 113 | OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\ |
| 114 | OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\ |
| 115 | OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\ |
| 116 | OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\ |
| 117 | dst+= stride;\ |
| 118 | src+= stride;\ |
| 119 | }\ |
| 120 | } else if (B + C) {\ |
| 121 | const int E= B+C;\ |
| 122 | const int step= C ? stride : 1;\ |
| 123 | for(i=0; i<h; i++){\ |
| 124 | OP(dst[0], (A*src[0] + E*src[step+0]));\ |
| 125 | OP(dst[1], (A*src[1] + E*src[step+1]));\ |
| 126 | OP(dst[2], (A*src[2] + E*src[step+2]));\ |
| 127 | OP(dst[3], (A*src[3] + E*src[step+3]));\ |
| 128 | dst+= stride;\ |
| 129 | src+= stride;\ |
| 130 | }\ |
| 131 | } else {\ |
| 132 | for ( i = 0; i < h; i++){\ |
| 133 | OP(dst[0], A * src[0]);\ |
| 134 | OP(dst[1], A * src[1]);\ |
| 135 | OP(dst[2], A * src[2]);\ |
| 136 | OP(dst[3], A * src[3]);\ |
| 137 | dst += stride;\ |
| 138 | src += stride;\ |
| 139 | }\ |
| 140 | }\ |
| 141 | }\ |
| 142 | \ |
| 143 | static void FUNCC(OPNAME ## h264_chroma_mc8)(uint8_t *_dst/*align 8*/, uint8_t *_src/*align 1*/, int stride, int h, int x, int y){\ |
| 144 | pixel *dst = (pixel*)_dst;\ |
| 145 | pixel *src = (pixel*)_src;\ |
| 146 | const int A=(8-x)*(8-y);\ |
| 147 | const int B=( x)*(8-y);\ |
| 148 | const int C=(8-x)*( y);\ |
| 149 | const int D=( x)*( y);\ |
| 150 | int i;\ |
| 151 | stride >>= sizeof(pixel)-1;\ |
| 152 | \ |
| 153 | av_assert2(x<8 && y<8 && x>=0 && y>=0);\ |
| 154 | \ |
| 155 | if(D){\ |
| 156 | for(i=0; i<h; i++){\ |
| 157 | OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\ |
| 158 | OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\ |
| 159 | OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\ |
| 160 | OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\ |
| 161 | OP(dst[4], (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5]));\ |
| 162 | OP(dst[5], (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6]));\ |
| 163 | OP(dst[6], (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7]));\ |
| 164 | OP(dst[7], (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8]));\ |
| 165 | dst+= stride;\ |
| 166 | src+= stride;\ |
| 167 | }\ |
| 168 | } else if (B + C) {\ |
| 169 | const int E= B+C;\ |
| 170 | const int step= C ? stride : 1;\ |
| 171 | for(i=0; i<h; i++){\ |
| 172 | OP(dst[0], (A*src[0] + E*src[step+0]));\ |
| 173 | OP(dst[1], (A*src[1] + E*src[step+1]));\ |
| 174 | OP(dst[2], (A*src[2] + E*src[step+2]));\ |
| 175 | OP(dst[3], (A*src[3] + E*src[step+3]));\ |
| 176 | OP(dst[4], (A*src[4] + E*src[step+4]));\ |
| 177 | OP(dst[5], (A*src[5] + E*src[step+5]));\ |
| 178 | OP(dst[6], (A*src[6] + E*src[step+6]));\ |
| 179 | OP(dst[7], (A*src[7] + E*src[step+7]));\ |
| 180 | dst+= stride;\ |
| 181 | src+= stride;\ |
| 182 | }\ |
| 183 | } else {\ |
| 184 | for ( i = 0; i < h; i++){\ |
| 185 | OP(dst[0], A * src[0]);\ |
| 186 | OP(dst[1], A * src[1]);\ |
| 187 | OP(dst[2], A * src[2]);\ |
| 188 | OP(dst[3], A * src[3]);\ |
| 189 | OP(dst[4], A * src[4]);\ |
| 190 | OP(dst[5], A * src[5]);\ |
| 191 | OP(dst[6], A * src[6]);\ |
| 192 | OP(dst[7], A * src[7]);\ |
| 193 | dst += stride;\ |
| 194 | src += stride;\ |
| 195 | }\ |
| 196 | }\ |
| 197 | } |
| 198 | |
| 199 | #define op_avg(a, b) a = (((a)+(((b) + 32)>>6)+1)>>1) |
| 200 | #define op_put(a, b) a = (((b) + 32)>>6) |
| 201 | |
| 202 | H264_CHROMA_MC(put_ , op_put) |
| 203 | H264_CHROMA_MC(avg_ , op_avg) |
| 204 | #undef op_avg |
| 205 | #undef op_put |