| Line | Branch | Exec | Source |
|---|---|---|---|
| 1 | /* | ||
| 2 | * This file is part of FFmpeg. | ||
| 3 | * | ||
| 4 | * FFmpeg is free software; you can redistribute it and/or | ||
| 5 | * modify it under the terms of the GNU Lesser General Public | ||
| 6 | * License as published by the Free Software Foundation; either | ||
| 7 | * version 2.1 of the License, or (at your option) any later version. | ||
| 8 | * | ||
| 9 | * FFmpeg is distributed in the hope that it will be useful, | ||
| 10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | ||
| 12 | * Lesser General Public License for more details. | ||
| 13 | * | ||
| 14 | * You should have received a copy of the GNU Lesser General Public | ||
| 15 | * License along with FFmpeg; if not, write to the Free Software | ||
| 16 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | ||
| 17 | */ | ||
| 18 | |||
| 19 | #include "config.h" | ||
| 20 | #include "libavutil/attributes.h" | ||
| 21 | #include "libavutil/common.h" | ||
| 22 | #include "idctdsp.h" | ||
| 23 | #include "mathops.h" | ||
| 24 | #include "wmv2dsp.h" | ||
| 25 | |||
| 26 | #define W0 2048 | ||
| 27 | #define W1 2841 /* 2048*sqrt (2)*cos (1*pi/16) */ | ||
| 28 | #define W2 2676 /* 2048*sqrt (2)*cos (2*pi/16) */ | ||
| 29 | #define W3 2408 /* 2048*sqrt (2)*cos (3*pi/16) */ | ||
| 30 | #define W4 2048 /* 2048*sqrt (2)*cos (4*pi/16) */ | ||
| 31 | #define W5 1609 /* 2048*sqrt (2)*cos (5*pi/16) */ | ||
| 32 | #define W6 1108 /* 2048*sqrt (2)*cos (6*pi/16) */ | ||
| 33 | #define W7 565 /* 2048*sqrt (2)*cos (7*pi/16) */ | ||
| 34 | |||
| 35 | 3017024 | static void wmv2_idct_row(short * b) | |
| 36 | { | ||
| 37 | int s1, s2; | ||
| 38 | int a0, a1, a2, a3, a4, a5, a6, a7; | ||
| 39 | |||
| 40 | /* step 1 */ | ||
| 41 | 3017024 | a1 = W1 * b[1] + W7 * b[7]; | |
| 42 | 3017024 | a7 = W7 * b[1] - W1 * b[7]; | |
| 43 | 3017024 | a5 = W5 * b[5] + W3 * b[3]; | |
| 44 | 3017024 | a3 = W3 * b[5] - W5 * b[3]; | |
| 45 | 3017024 | a2 = W2 * b[2] + W6 * b[6]; | |
| 46 | 3017024 | a6 = W6 * b[2] - W2 * b[6]; | |
| 47 | 3017024 | a0 = W0 * b[0] + W0 * b[4]; | |
| 48 | 3017024 | a4 = W0 * b[0] - W0 * b[4]; | |
| 49 | |||
| 50 | /* step 2 */ | ||
| 51 | 3017024 | s1 = (int)(181U * (a1 - a5 + a7 - a3) + 128) >> 8; // 1, 3, 5, 7 | |
| 52 | 3017024 | s2 = (int)(181U * (a1 - a5 - a7 + a3) + 128) >> 8; | |
| 53 | |||
| 54 | /* step 3 */ | ||
| 55 | 3017024 | b[0] = (a0 + a2 + a1 + a5 + (1 << 7)) >> 8; | |
| 56 | 3017024 | b[1] = (a4 + a6 + s1 + (1 << 7)) >> 8; | |
| 57 | 3017024 | b[2] = (a4 - a6 + s2 + (1 << 7)) >> 8; | |
| 58 | 3017024 | b[3] = (a0 - a2 + a7 + a3 + (1 << 7)) >> 8; | |
| 59 | 3017024 | b[4] = (a0 - a2 - a7 - a3 + (1 << 7)) >> 8; | |
| 60 | 3017024 | b[5] = (a4 - a6 - s2 + (1 << 7)) >> 8; | |
| 61 | 3017024 | b[6] = (a4 + a6 - s1 + (1 << 7)) >> 8; | |
| 62 | 3017024 | b[7] = (a0 + a2 - a1 - a5 + (1 << 7)) >> 8; | |
| 63 | 3017024 | } | |
| 64 | |||
| 65 | 3017024 | static void wmv2_idct_col(short * b) | |
| 66 | { | ||
| 67 | int s1, s2; | ||
| 68 | int a0, a1, a2, a3, a4, a5, a6, a7; | ||
| 69 | |||
| 70 | /* step 1, with extended precision */ | ||
| 71 | 3017024 | a1 = (W1 * b[8 * 1] + W7 * b[8 * 7] + 4) >> 3; | |
| 72 | 3017024 | a7 = (W7 * b[8 * 1] - W1 * b[8 * 7] + 4) >> 3; | |
| 73 | 3017024 | a5 = (W5 * b[8 * 5] + W3 * b[8 * 3] + 4) >> 3; | |
| 74 | 3017024 | a3 = (W3 * b[8 * 5] - W5 * b[8 * 3] + 4) >> 3; | |
| 75 | 3017024 | a2 = (W2 * b[8 * 2] + W6 * b[8 * 6] + 4) >> 3; | |
| 76 | 3017024 | a6 = (W6 * b[8 * 2] - W2 * b[8 * 6] + 4) >> 3; | |
| 77 | 3017024 | a0 = (W0 * b[8 * 0] + W0 * b[8 * 4] ) >> 3; | |
| 78 | 3017024 | a4 = (W0 * b[8 * 0] - W0 * b[8 * 4] ) >> 3; | |
| 79 | |||
| 80 | /* step 2 */ | ||
| 81 | 3017024 | s1 = (int)(181U * (a1 - a5 + a7 - a3) + 128) >> 8; | |
| 82 | 3017024 | s2 = (int)(181U * (a1 - a5 - a7 + a3) + 128) >> 8; | |
| 83 | |||
| 84 | /* step 3 */ | ||
| 85 | 3017024 | b[8 * 0] = (a0 + a2 + a1 + a5 + (1 << 13)) >> 14; | |
| 86 | 3017024 | b[8 * 1] = (a4 + a6 + s1 + (1 << 13)) >> 14; | |
| 87 | 3017024 | b[8 * 2] = (a4 - a6 + s2 + (1 << 13)) >> 14; | |
| 88 | 3017024 | b[8 * 3] = (a0 - a2 + a7 + a3 + (1 << 13)) >> 14; | |
| 89 | |||
| 90 | 3017024 | b[8 * 4] = (a0 - a2 - a7 - a3 + (1 << 13)) >> 14; | |
| 91 | 3017024 | b[8 * 5] = (a4 - a6 - s2 + (1 << 13)) >> 14; | |
| 92 | 3017024 | b[8 * 6] = (a4 + a6 - s1 + (1 << 13)) >> 14; | |
| 93 | 3017024 | b[8 * 7] = (a0 + a2 - a1 - a5 + (1 << 13)) >> 14; | |
| 94 | 3017024 | } | |
| 95 | |||
| 96 | 262288 | static void wmv2_idct_add_c(uint8_t *dest, ptrdiff_t line_size, int16_t *block) | |
| 97 | { | ||
| 98 | int i; | ||
| 99 | |||
| 100 |
2/2✓ Branch 0 taken 2098304 times.
✓ Branch 1 taken 262288 times.
|
2360592 | for (i = 0; i < 64; i += 8) |
| 101 | 2098304 | wmv2_idct_row(block + i); | |
| 102 |
2/2✓ Branch 0 taken 2098304 times.
✓ Branch 1 taken 262288 times.
|
2360592 | for (i = 0; i < 8; i++) |
| 103 | 2098304 | wmv2_idct_col(block + i); | |
| 104 | |||
| 105 |
2/2✓ Branch 0 taken 2098304 times.
✓ Branch 1 taken 262288 times.
|
2360592 | for (i = 0; i < 8; i++) { |
| 106 | 2098304 | dest[0] = av_clip_uint8(dest[0] + block[0]); | |
| 107 | 2098304 | dest[1] = av_clip_uint8(dest[1] + block[1]); | |
| 108 | 2098304 | dest[2] = av_clip_uint8(dest[2] + block[2]); | |
| 109 | 2098304 | dest[3] = av_clip_uint8(dest[3] + block[3]); | |
| 110 | 2098304 | dest[4] = av_clip_uint8(dest[4] + block[4]); | |
| 111 | 2098304 | dest[5] = av_clip_uint8(dest[5] + block[5]); | |
| 112 | 2098304 | dest[6] = av_clip_uint8(dest[6] + block[6]); | |
| 113 | 2098304 | dest[7] = av_clip_uint8(dest[7] + block[7]); | |
| 114 | 2098304 | dest += line_size; | |
| 115 | 2098304 | block += 8; | |
| 116 | } | ||
| 117 | 262288 | } | |
| 118 | |||
| 119 | 114840 | static void wmv2_idct_put_c(uint8_t *dest, ptrdiff_t line_size, int16_t *block) | |
| 120 | { | ||
| 121 | int i; | ||
| 122 | |||
| 123 |
2/2✓ Branch 0 taken 918720 times.
✓ Branch 1 taken 114840 times.
|
1033560 | for (i = 0; i < 64; i += 8) |
| 124 | 918720 | wmv2_idct_row(block + i); | |
| 125 |
2/2✓ Branch 0 taken 918720 times.
✓ Branch 1 taken 114840 times.
|
1033560 | for (i = 0; i < 8; i++) |
| 126 | 918720 | wmv2_idct_col(block + i); | |
| 127 | |||
| 128 |
2/2✓ Branch 0 taken 918720 times.
✓ Branch 1 taken 114840 times.
|
1033560 | for (i = 0; i < 8; i++) { |
| 129 | 918720 | dest[0] = av_clip_uint8(block[0]); | |
| 130 | 918720 | dest[1] = av_clip_uint8(block[1]); | |
| 131 | 918720 | dest[2] = av_clip_uint8(block[2]); | |
| 132 | 918720 | dest[3] = av_clip_uint8(block[3]); | |
| 133 | 918720 | dest[4] = av_clip_uint8(block[4]); | |
| 134 | 918720 | dest[5] = av_clip_uint8(block[5]); | |
| 135 | 918720 | dest[6] = av_clip_uint8(block[6]); | |
| 136 | 918720 | dest[7] = av_clip_uint8(block[7]); | |
| 137 | 918720 | dest += line_size; | |
| 138 | 918720 | block += 8; | |
| 139 | } | ||
| 140 | 114840 | } | |
| 141 | |||
| 142 | 62 | av_cold void ff_wmv2dsp_init(IDCTDSPContext *c) | |
| 143 | { | ||
| 144 | 62 | c->idct_add = wmv2_idct_add_c; | |
| 145 | 62 | c->idct_put = wmv2_idct_put_c; | |
| 146 | 62 | c->idct = NULL; | |
| 147 | 62 | c->perm_type = FF_IDCT_PERM_NONE; | |
| 148 | |||
| 149 | #if ARCH_MIPS | ||
| 150 | ff_wmv2dsp_init_mips(c); | ||
| 151 | #endif | ||
| 152 | 62 | ff_init_scantable_permutation(c->idct_permutation, | |
| 153 | c->perm_type); | ||
| 154 | 62 | } | |
| 155 |