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
23#include <stddef.h>
24
25#include "libavutil/avassert.h"
26#include "bit_depth_template.c"
27
28#define H264_CHROMA_MC(OPNAME, OP)\
29static void FUNCC(OPNAME ## h264_chroma_mc1)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y){\
30    pixel *dst = (pixel*)_dst;\
31    pixel *src = (pixel*)_src;\
32    const int A=(8-x)*(8-y);\
33    const int B=(  x)*(8-y);\
34    const int C=(8-x)*(  y);\
35    const int D=(  x)*(  y);\
36    int i;\
37    stride >>= sizeof(pixel)-1;\
38    \
39    av_assert2(x<8 && y<8 && x>=0 && y>=0);\
40\
41    if(D){\
42        for(i=0; i<h; i++){\
43            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
44            dst+= stride;\
45            src+= stride;\
46        }\
47    } else if (B + C) {\
48        const int E= B+C;\
49        const int step= C ? stride : 1;\
50        for(i=0; i<h; i++){\
51            OP(dst[0], (A*src[0] + E*src[step+0]));\
52            dst+= stride;\
53            src+= stride;\
54        }\
55    } else {\
56        for(i=0; i<h; i++){\
57            OP(dst[0], (A*src[0]));\
58            dst+= stride;\
59            src+= stride;\
60        }\
61    }\
62}\
63static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
64{\
65    pixel *dst = (pixel*)_dst;\
66    pixel *src = (pixel*)_src;\
67    const int A=(8-x)*(8-y);\
68    const int B=(  x)*(8-y);\
69    const int C=(8-x)*(  y);\
70    const int D=(  x)*(  y);\
71    int i;\
72    stride >>= sizeof(pixel)-1;\
73    \
74    av_assert2(x<8 && y<8 && x>=0 && y>=0);\
75\
76    if(D){\
77        for(i=0; i<h; i++){\
78            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
79            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
80            dst+= stride;\
81            src+= stride;\
82        }\
83    } else if (B + C) {\
84        const int E= B+C;\
85        const ptrdiff_t step = C ? stride : 1;\
86        for(i=0; i<h; i++){\
87            OP(dst[0], (A*src[0] + E*src[step+0]));\
88            OP(dst[1], (A*src[1] + E*src[step+1]));\
89            dst+= stride;\
90            src+= stride;\
91        }\
92    } else {\
93        for ( i = 0; i < h; i++){\
94            OP(dst[0], A * src[0]);\
95            OP(dst[1], A * src[1]);\
96            dst += stride;\
97            src += stride;\
98        }\
99    }\
100}\
101\
102static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
103{\
104    pixel *dst = (pixel*)_dst;\
105    pixel *src = (pixel*)_src;\
106    const int A=(8-x)*(8-y);\
107    const int B=(  x)*(8-y);\
108    const int C=(8-x)*(  y);\
109    const int D=(  x)*(  y);\
110    int i;\
111    stride >>= sizeof(pixel)-1;\
112    \
113    av_assert2(x<8 && y<8 && x>=0 && y>=0);\
114\
115    if(D){\
116        for(i=0; i<h; i++){\
117            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
118            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
119            OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
120            OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
121            dst+= stride;\
122            src+= stride;\
123        }\
124    } else if (B + C) {\
125        const int E= B+C;\
126        const ptrdiff_t step = C ? stride : 1;\
127        for(i=0; i<h; i++){\
128            OP(dst[0], (A*src[0] + E*src[step+0]));\
129            OP(dst[1], (A*src[1] + E*src[step+1]));\
130            OP(dst[2], (A*src[2] + E*src[step+2]));\
131            OP(dst[3], (A*src[3] + E*src[step+3]));\
132            dst+= stride;\
133            src+= stride;\
134        }\
135    } else {\
136        for ( i = 0; i < h; i++){\
137            OP(dst[0], A * src[0]);\
138            OP(dst[1], A * src[1]);\
139            OP(dst[2], A * src[2]);\
140            OP(dst[3], A * src[3]);\
141            dst += stride;\
142            src += stride;\
143        }\
144    }\
145}\
146\
147static void FUNCC(OPNAME ## h264_chroma_mc8)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
148{\
149    pixel *dst = (pixel*)_dst;\
150    pixel *src = (pixel*)_src;\
151    const int A=(8-x)*(8-y);\
152    const int B=(  x)*(8-y);\
153    const int C=(8-x)*(  y);\
154    const int D=(  x)*(  y);\
155    int i;\
156    stride >>= sizeof(pixel)-1;\
157    \
158    av_assert2(x<8 && y<8 && x>=0 && y>=0);\
159\
160    if(D){\
161        for(i=0; i<h; i++){\
162            OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
163            OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
164            OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
165            OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
166            OP(dst[4], (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5]));\
167            OP(dst[5], (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6]));\
168            OP(dst[6], (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7]));\
169            OP(dst[7], (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8]));\
170            dst+= stride;\
171            src+= stride;\
172        }\
173    } else if (B + C) {\
174        const int E= B+C;\
175        const ptrdiff_t step = C ? stride : 1;\
176        for(i=0; i<h; i++){\
177            OP(dst[0], (A*src[0] + E*src[step+0]));\
178            OP(dst[1], (A*src[1] + E*src[step+1]));\
179            OP(dst[2], (A*src[2] + E*src[step+2]));\
180            OP(dst[3], (A*src[3] + E*src[step+3]));\
181            OP(dst[4], (A*src[4] + E*src[step+4]));\
182            OP(dst[5], (A*src[5] + E*src[step+5]));\
183            OP(dst[6], (A*src[6] + E*src[step+6]));\
184            OP(dst[7], (A*src[7] + E*src[step+7]));\
185            dst+= stride;\
186            src+= stride;\
187        }\
188    } else {\
189        for ( i = 0; i < h; i++){\
190            OP(dst[0], A * src[0]);\
191            OP(dst[1], A * src[1]);\
192            OP(dst[2], A * src[2]);\
193            OP(dst[3], A * src[3]);\
194            OP(dst[4], A * src[4]);\
195            OP(dst[5], A * src[5]);\
196            OP(dst[6], A * src[6]);\
197            OP(dst[7], A * src[7]);\
198            dst += stride;\
199            src += stride;\
200        }\
201    }\
202}
203
204#define op_avg(a, b) a = (((a)+(((b) + 32)>>6)+1)>>1)
205#define op_put(a, b) a = (((b) + 32)>>6)
206
207H264_CHROMA_MC(put_       , op_put)
208H264_CHROMA_MC(avg_       , op_avg)
209#undef op_avg
210#undef op_put
211