1/*
2 * quarterpel DSP function templates
3 * Copyright (c) 2000, 2001 Fabrice Bellard
4 * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
5 *
6 * This file is part of FFmpeg.
7 *
8 * FFmpeg is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
12 *
13 * FFmpeg is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
16 * Lesser General Public License for more details.
17 *
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with FFmpeg; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 */
22
23/**
24 * @file
25 * quarterpel DSP function templates
26 */
27
28#define PIXOP2(OPNAME, OP)                                              \
29static inline void OPNAME ## _no_rnd_pixels8_l2_8(uint8_t *dst,         \
30                                                  const uint8_t *src1,  \
31                                                  const uint8_t *src2,  \
32                                                  int dst_stride,       \
33                                                  int src_stride1,      \
34                                                  int src_stride2,      \
35                                                  int h)                \
36{                                                                       \
37    int i;                                                              \
38                                                                        \
39    for (i = 0; i < h; i++) {                                           \
40        uint32_t a, b;                                                  \
41        a = AV_RN32(&src1[i * src_stride1]);                            \
42        b = AV_RN32(&src2[i * src_stride2]);                            \
43        OP(*((uint32_t *) &dst[i * dst_stride]),                        \
44           no_rnd_avg32(a, b));                                         \
45        a = AV_RN32(&src1[i * src_stride1 + 4]);                        \
46        b = AV_RN32(&src2[i * src_stride2 + 4]);                        \
47        OP(*((uint32_t *) &dst[i * dst_stride + 4]),                    \
48           no_rnd_avg32(a, b));                                         \
49    }                                                                   \
50}                                                                       \
51                                                                        \
52static inline void OPNAME ## _no_rnd_pixels16_l2_8(uint8_t *dst,        \
53                                                   const uint8_t *src1, \
54                                                   const uint8_t *src2, \
55                                                   int dst_stride,      \
56                                                   int src_stride1,     \
57                                                   int src_stride2,     \
58                                                   int h)               \
59{                                                                       \
60    OPNAME ## _no_rnd_pixels8_l2_8(dst, src1, src2, dst_stride,         \
61                                   src_stride1, src_stride2, h);        \
62    OPNAME ## _no_rnd_pixels8_l2_8(dst  + 8,                            \
63                                   src1 + 8,                            \
64                                   src2 + 8,                            \
65                                   dst_stride, src_stride1,             \
66                                   src_stride2, h);                     \
67}                                                                       \
68                                                                        \
69static inline void OPNAME ## _pixels8_l4_8(uint8_t *dst,                \
70                                           const uint8_t *src1,         \
71                                           const uint8_t *src2,         \
72                                           const uint8_t *src3,         \
73                                           const uint8_t *src4,         \
74                                           int dst_stride,              \
75                                           int src_stride1,             \
76                                           int src_stride2,             \
77                                           int src_stride3,             \
78                                           int src_stride4,             \
79                                           int h)                       \
80{                                                                       \
81    /* FIXME HIGH BIT DEPTH */                                          \
82    int i;                                                              \
83                                                                        \
84    for (i = 0; i < h; i++) {                                           \
85        uint32_t a, b, c, d, l0, l1, h0, h1;                            \
86        a  = AV_RN32(&src1[i * src_stride1]);                           \
87        b  = AV_RN32(&src2[i * src_stride2]);                           \
88        c  = AV_RN32(&src3[i * src_stride3]);                           \
89        d  = AV_RN32(&src4[i * src_stride4]);                           \
90        l0 = (a & 0x03030303UL) +                                       \
91             (b & 0x03030303UL) +                                       \
92                  0x02020202UL;                                         \
93        h0 = ((a & 0xFCFCFCFCUL) >> 2) +                                \
94             ((b & 0xFCFCFCFCUL) >> 2);                                 \
95        l1 = (c & 0x03030303UL) +                                       \
96             (d & 0x03030303UL);                                        \
97        h1 = ((c & 0xFCFCFCFCUL) >> 2) +                                \
98             ((d & 0xFCFCFCFCUL) >> 2);                                 \
99        OP(*((uint32_t *) &dst[i * dst_stride]),                        \
100           h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL));                \
101        a  = AV_RN32(&src1[i * src_stride1 + 4]);                       \
102        b  = AV_RN32(&src2[i * src_stride2 + 4]);                       \
103        c  = AV_RN32(&src3[i * src_stride3 + 4]);                       \
104        d  = AV_RN32(&src4[i * src_stride4 + 4]);                       \
105        l0 = (a & 0x03030303UL) +                                       \
106             (b & 0x03030303UL) +                                       \
107                  0x02020202UL;                                         \
108        h0 = ((a & 0xFCFCFCFCUL) >> 2) +                                \
109             ((b & 0xFCFCFCFCUL) >> 2);                                 \
110        l1 = (c & 0x03030303UL) +                                       \
111             (d & 0x03030303UL);                                        \
112        h1 = ((c & 0xFCFCFCFCUL) >> 2) +                                \
113             ((d & 0xFCFCFCFCUL) >> 2);                                 \
114        OP(*((uint32_t *) &dst[i * dst_stride + 4]),                    \
115           h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL));                \
116    }                                                                   \
117}                                                                       \
118                                                                        \
119static inline void OPNAME ## _no_rnd_pixels8_l4_8(uint8_t *dst,         \
120                                                  const uint8_t *src1,  \
121                                                  const uint8_t *src2,  \
122                                                  const uint8_t *src3,  \
123                                                  const uint8_t *src4,  \
124                                                  int dst_stride,       \
125                                                  int src_stride1,      \
126                                                  int src_stride2,      \
127                                                  int src_stride3,      \
128                                                  int src_stride4,      \
129                                                  int h)                \
130{                                                                       \
131    /* FIXME HIGH BIT DEPTH */                                          \
132    int i;                                                              \
133                                                                        \
134    for (i = 0; i < h; i++) {                                           \
135        uint32_t a, b, c, d, l0, l1, h0, h1;                            \
136        a  = AV_RN32(&src1[i * src_stride1]);                           \
137        b  = AV_RN32(&src2[i * src_stride2]);                           \
138        c  = AV_RN32(&src3[i * src_stride3]);                           \
139        d  = AV_RN32(&src4[i * src_stride4]);                           \
140        l0 = (a & 0x03030303UL) +                                       \
141             (b & 0x03030303UL) +                                       \
142                  0x01010101UL;                                         \
143        h0 = ((a & 0xFCFCFCFCUL) >> 2) +                                \
144             ((b & 0xFCFCFCFCUL) >> 2);                                 \
145        l1 = (c & 0x03030303UL) +                                       \
146             (d & 0x03030303UL);                                        \
147        h1 = ((c & 0xFCFCFCFCUL) >> 2) +                                \
148             ((d & 0xFCFCFCFCUL) >> 2);                                 \
149        OP(*((uint32_t *) &dst[i * dst_stride]),                        \
150           h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL));                \
151        a  = AV_RN32(&src1[i * src_stride1 + 4]);                       \
152        b  = AV_RN32(&src2[i * src_stride2 + 4]);                       \
153        c  = AV_RN32(&src3[i * src_stride3 + 4]);                       \
154        d  = AV_RN32(&src4[i * src_stride4 + 4]);                       \
155        l0 = (a & 0x03030303UL) +                                       \
156             (b & 0x03030303UL) +                                       \
157                  0x01010101UL;                                         \
158        h0 = ((a & 0xFCFCFCFCUL) >> 2) +                                \
159             ((b & 0xFCFCFCFCUL) >> 2);                                 \
160        l1 = (c & 0x03030303UL) +                                       \
161             (d & 0x03030303UL);                                        \
162        h1 = ((c & 0xFCFCFCFCUL) >> 2) +                                \
163             ((d & 0xFCFCFCFCUL) >> 2);                                 \
164        OP(*((uint32_t *) &dst[i * dst_stride + 4]),                    \
165           h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL));                \
166    }                                                                   \
167}                                                                       \
168                                                                        \
169static inline void OPNAME ## _pixels16_l4_8(uint8_t *dst,               \
170                                            const uint8_t *src1,        \
171                                            const uint8_t *src2,        \
172                                            const uint8_t *src3,        \
173                                            const uint8_t *src4,        \
174                                            int dst_stride,             \
175                                            int src_stride1,            \
176                                            int src_stride2,            \
177                                            int src_stride3,            \
178                                            int src_stride4,            \
179                                            int h)                      \
180{                                                                       \
181    OPNAME ## _pixels8_l4_8(dst, src1, src2, src3, src4, dst_stride,    \
182                            src_stride1, src_stride2, src_stride3,      \
183                            src_stride4, h);                            \
184    OPNAME ## _pixels8_l4_8(dst  + 8,                                   \
185                            src1 + 8, src2 + 8,                         \
186                            src3 + 8, src4 + 8,                         \
187                            dst_stride, src_stride1, src_stride2,       \
188                            src_stride3, src_stride4, h);               \
189}                                                                       \
190                                                                        \
191static inline void OPNAME ## _no_rnd_pixels16_l4_8(uint8_t *dst,        \
192                                                   const uint8_t *src1, \
193                                                   const uint8_t *src2, \
194                                                   const uint8_t *src3, \
195                                                   const uint8_t *src4, \
196                                                   int dst_stride,      \
197                                                   int src_stride1,     \
198                                                   int src_stride2,     \
199                                                   int src_stride3,     \
200                                                   int src_stride4,     \
201                                                   int h)               \
202{                                                                       \
203    OPNAME ## _no_rnd_pixels8_l4_8(dst, src1, src2, src3, src4,         \
204                                   dst_stride, src_stride1,             \
205                                   src_stride2, src_stride3,            \
206                                   src_stride4, h);                     \
207    OPNAME ## _no_rnd_pixels8_l4_8(dst  + 8,                            \
208                                   src1 + 8, src2 + 8,                  \
209                                   src3 + 8, src4 + 8,                  \
210                                   dst_stride, src_stride1,             \
211                                   src_stride2, src_stride3,            \
212                                   src_stride4, h);                     \
213}                                                                       \
214
215#define op_avg(a, b) a = rnd_avg32(a, b)
216#define op_put(a, b) a = b
217#define put_no_rnd_pixels8_8_c put_pixels8_8_c
218PIXOP2(avg, op_avg)
219PIXOP2(put, op_put)
220#undef op_avg
221#undef op_put
222