generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Removal of Hybrid DWT/DCT experiment.

Browse Source 
Removal of experiment to simplify code base for other
changes.

Change-Id: If0a33952504558511926ad212bc311fc2bffb19a
This commit is contained in:
Paul Wilkins 2013-02-11 12:35:28 +00:00
parent 097f205289
commit 649be94cf0
8 changed files with 4 additions and 2143 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
configure vendored
View File

@ -244,7 +244,6 @@ EXPERIMENT_LIST="
newbintramodes
comp_interintra_pred
tx64x64
dwtdcthybrid
cnvcontext
newcoefcontext
enable_6tap

5
test/dct32x32_test.cc
View File

@ -36,7 +36,6 @@ static int round(double x) {
}
#endif
#if !CONFIG_DWTDCTHYBRID
static const double kPi = 3.141592653589793238462643383279502884;
static void reference2_32x32_idct_2d(double *input, double *output) {
double x;
@ -127,9 +126,7 @@ TEST(VP9Idct32x32Test, AccuracyCheck) {
}
}
}
#else // CONFIG_DWTDCTHYBRID
// TODO(rbultje/debargha): add DWT-specific tests
#endif // CONFIG_DWTDCTHYBRID
TEST(VP9Fdct32x32Test, AccuracyCheck) {
ACMRandom rnd(ACMRandom::DeterministicSeed());
unsigned int max_error = 0;

12
vp9/common/vp9_blockd.h
View File

@ -47,18 +47,6 @@ void vpx_log(const char *format, ...);
#define MAX_MV_REFS 9
#define MAX_MV_REF_CANDIDATES 4
#if CONFIG_DWTDCTHYBRID
#define DWT_MAX_LENGTH 64
#define DWT_TYPE 26 // 26/53/97
#define DWT_PRECISION_BITS 2
#define DWT_PRECISION_RND ((1 << DWT_PRECISION_BITS) / 2)
#define DWTDCT16X16 0
#define DWTDCT16X16_LEAN 1
#define DWTDCT8X8 2
#define DWTDCT_TYPE DWTDCT16X16_LEAN
#endif
typedef struct {
int r, c;
} POS;

619
vp9/common/vp9_entropy.c
View File

@ -143,624 +143,6 @@ DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_16x16[256]) = {
237, 252, 253, 238, 223, 239, 254, 255,
};
#if CONFIG_DWTDCTHYBRID
#if DWTDCT_TYPE == DWTDCT16X16_LEAN
DECLARE_ALIGNED(16, const int, vp9_coef_bands_32x32[1024]) = {
0, 1, 2, 3, 5, 4, 4, 5, 5, 3, 6, 3, 5, 4, 6, 6,
6, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
};
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_32x32[1024]) = {
0, 1, 32, 64, 33, 2, 3, 34,
65, 96, 128, 97, 66, 35, 4, 5,
36, 67, 98, 129, 160, 192, 161, 130,
99, 68, 37, 6, 7, 38, 69, 100,
131, 162, 193, 224, 256, 225, 194, 163,
132, 101, 70, 39, 8, 9, 40, 71,
102, 133, 164, 195, 226, 257, 288, 320,
289, 258, 227, 196, 165, 134, 103, 72,
41, 10, 11, 42, 73, 104, 135, 166,
197, 228, 259, 290, 321, 352, 384, 353,
322, 291, 260, 229, 198, 167, 136, 105,
74, 43, 12, 13, 44, 75, 106, 137,
168, 199, 230, 261, 292, 323, 354, 385,
416, 448, 417, 386, 355, 324, 293, 262,
231, 200, 169, 138, 107, 76, 45, 14,
15, 46, 77, 108, 139, 170, 201, 232,
263, 294, 325, 356, 387, 418, 449, 480,
481, 450, 419, 388, 357, 326, 295, 264,
233, 202, 171, 140, 109, 78, 47, 79,
110, 141, 172, 203, 234, 265, 296, 327,
358, 389, 420, 451, 482, 483, 452, 421,
390, 359, 328, 297, 266, 235, 204, 173,
142, 111, 143, 174, 205, 236, 267, 298,
329, 360, 391, 422, 453, 484, 485, 454,
423, 392, 361, 330, 299, 268, 237, 206,
175, 207, 238, 269, 300, 331, 362, 393,
424, 455, 486, 487, 456, 425, 394, 363,
332, 301, 270, 239, 271, 302, 333, 364,
395, 426, 457, 488, 489, 458, 427, 396,
365, 334, 303, 335, 366, 397, 428, 459,
490, 491, 460, 429, 398, 367, 399, 430,
461, 492, 493, 462, 431, 463, 494, 495,
16, 512, 528, 17, 513, 529, 48, 544,
560, 80, 576, 592, 49, 545, 561, 18,
514, 530, 19, 515, 531, 50, 546, 562,
81, 577, 593, 112, 608, 624, 144, 640,
656, 113, 609, 625, 82, 578, 594, 51,
547, 563, 20, 516, 532, 21, 517, 533,
52, 548, 564, 83, 579, 595, 114, 610,
626, 145, 641, 657, 176, 672, 688, 208,
704, 720, 177, 673, 689, 146, 642, 658,
115, 611, 627, 84, 580, 596, 53, 549,
565, 22, 518, 534, 23, 519, 535, 54,
550, 566, 85, 581, 597, 116, 612, 628,
147, 643, 659, 178, 674, 690, 209, 705,
721, 240, 736, 752, 272, 768, 784, 241,
737, 753, 210, 706, 722, 179, 675, 691,
148, 644, 660, 117, 613, 629, 86, 582,
598, 55, 551, 567, 24, 520, 536, 25,
521, 537, 56, 552, 568, 87, 583, 599,
118, 614, 630, 149, 645, 661, 180, 676,
692, 211, 707, 723, 242, 738, 754, 273,
769, 785, 304, 800, 816, 336, 832, 848,
305, 801, 817, 274, 770, 786, 243, 739,
755, 212, 708, 724, 181, 677, 693, 150,
646, 662, 119, 615, 631, 88, 584, 600,
57, 553, 569, 26, 522, 538, 27, 523,
539, 58, 554, 570, 89, 585, 601, 120,
616, 632, 151, 647, 663, 182, 678, 694,
213, 709, 725, 244, 740, 756, 275, 771,
787, 306, 802, 818, 337, 833, 849, 368,
864, 880, 400, 896, 912, 369, 865, 881,
338, 834, 850, 307, 803, 819, 276, 772,
788, 245, 741, 757, 214, 710, 726, 183,
679, 695, 152, 648, 664, 121, 617, 633,
90, 586, 602, 59, 555, 571, 28, 524,
540, 29, 525, 541, 60, 556, 572, 91,
587, 603, 122, 618, 634, 153, 649, 665,
184, 680, 696, 215, 711, 727, 246, 742,
758, 277, 773, 789, 308, 804, 820, 339,
835, 851, 370, 866, 882, 401, 897, 913,
432, 928, 944, 464, 960, 976, 433, 929,
945, 402, 898, 914, 371, 867, 883, 340,
836, 852, 309, 805, 821, 278, 774, 790,
247, 743, 759, 216, 712, 728, 185, 681,
697, 154, 650, 666, 123, 619, 635, 92,
588, 604, 61, 557, 573, 30, 526, 542,
31, 527, 543, 62, 558, 574, 93, 589,
605, 124, 620, 636, 155, 651, 667, 186,
682, 698, 217, 713, 729, 248, 744, 760,
279, 775, 791, 310, 806, 822, 341, 837,
853, 372, 868, 884, 403, 899, 915, 434,
930, 946, 465, 961, 977, 496, 992, 1008,
497, 993, 1009, 466, 962, 978, 435, 931,
947, 404, 900, 916, 373, 869, 885, 342,
838, 854, 311, 807, 823, 280, 776, 792,
249, 745, 761, 218, 714, 730, 187, 683,
699, 156, 652, 668, 125, 621, 637, 94,
590, 606, 63, 559, 575, 95, 591, 607,
126, 622, 638, 157, 653, 669, 188, 684,
700, 219, 715, 731, 250, 746, 762, 281,
777, 793, 312, 808, 824, 343, 839, 855,
374, 870, 886, 405, 901, 917, 436, 932,
948, 467, 963, 979, 498, 994, 1010, 499,
995, 1011, 468, 964, 980, 437, 933, 949,
406, 902, 918, 375, 871, 887, 344, 840,
856, 313, 809, 825, 282, 778, 794, 251,
747, 763, 220, 716, 732, 189, 685, 701,
158, 654, 670, 127, 623, 639, 159, 655,
671, 190, 686, 702, 221, 717, 733, 252,
748, 764, 283, 779, 795, 314, 810, 826,
345, 841, 857, 376, 872, 888, 407, 903,
919, 438, 934, 950, 469, 965, 981, 500,
996, 1012, 501, 997, 1013, 470, 966, 982,
439, 935, 951, 408, 904, 920, 377, 873,
889, 346, 842, 858, 315, 811, 827, 284,
780, 796, 253, 749, 765, 222, 718, 734,
191, 687, 703, 223, 719, 735, 254, 750,
766, 285, 781, 797, 316, 812, 828, 347,
843, 859, 378, 874, 890, 409, 905, 921,
440, 936, 952, 471, 967, 983, 502, 998,
1014, 503, 999, 1015, 472, 968, 984, 441,
937, 953, 410, 906, 922, 379, 875, 891,
348, 844, 860, 317, 813, 829, 286, 782,
798, 255, 751, 767, 287, 783, 799, 318,
814, 830, 349, 845, 861, 380, 876, 892,
411, 907, 923, 442, 938, 954, 473, 969,
985, 504, 1000, 1016, 505, 1001, 1017, 474,
970, 986, 443, 939, 955, 412, 908, 924,
381, 877, 893, 350, 846, 862, 319, 815,
831, 351, 847, 863, 382, 878, 894, 413,
909, 925, 444, 940, 956, 475, 971, 987,
506, 1002, 1018, 507, 1003, 1019, 476, 972,
988, 445, 941, 957, 414, 910, 926, 383,
879, 895, 415, 911, 927, 446, 942, 958,
477, 973, 989, 508, 1004, 1020, 509, 1005,
1021, 478, 974, 990, 447, 943, 959, 479,
975, 991, 510, 1006, 1022, 511, 1007, 1023,
};
#elif DWTDCT_TYPE == DWTDCT16X16
DECLARE_ALIGNED(16, const int, vp9_coef_bands_32x32[1024]) = {
0, 1, 2, 3, 5, 4, 4, 5, 5, 3, 6, 3, 5, 4, 6,
6, 6, 6,
6,
6, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
6, 6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
};
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_32x32[1024]) = {
0, 1, 32, 64, 33, 2, 3, 34,
65, 96, 128, 97, 66, 35, 4,
16, 512, 528,
5,
36, 67, 98, 129, 160, 192, 161, 130,
99, 68, 37, 6, 7, 38, 69, 100,
131, 162, 193, 224, 256, 225, 194, 163,
132, 101, 70, 39, 8, 9, 40, 71,
102, 133, 164, 195, 226, 257, 288, 320,
289, 258, 227, 196, 165, 134, 103, 72,
41, 10, 11, 42, 73, 104, 135, 166,
197, 228, 259, 290, 321, 352, 384, 353,
322, 291, 260, 229, 198, 167, 136, 105,
74, 43, 12, 13, 44, 75, 106, 137,
168, 199, 230, 261, 292, 323, 354, 385,
416, 448, 417, 386, 355, 324, 293, 262,
231, 200, 169, 138, 107, 76, 45, 14,
15, 46, 77, 108, 139, 170, 201, 232,
263, 294, 325, 356, 387, 418, 449, 480,
481, 450, 419, 388, 357, 326, 295, 264,
233, 202, 171, 140, 109, 78, 47, 79,
110, 141, 172, 203, 234, 265, 296, 327,
358, 389, 420, 451, 482, 483, 452, 421,
390, 359, 328, 297, 266, 235, 204, 173,
142, 111, 143, 174, 205, 236, 267, 298,
329, 360, 391, 422, 453, 484, 485, 454,
423, 392, 361, 330, 299, 268, 237, 206,
175, 207, 238, 269, 300, 331, 362, 393,
424, 455, 486, 487, 456, 425, 394, 363,
332, 301, 270, 239, 271, 302, 333, 364,
395, 426, 457, 488, 489, 458, 427, 396,
365, 334, 303, 335, 366, 397, 428, 459,
490, 491, 460, 429, 398, 367, 399, 430,
461, 492, 493, 462, 431, 463, 494, 495,
17, 513, 529, 48, 544,
560, 80, 576, 592, 49, 545, 561, 18,
514, 530, 19, 515, 531, 50, 546, 562,
81, 577, 593, 112, 608, 624, 144, 640,
656, 113, 609, 625, 82, 578, 594, 51,
547, 563, 20, 516, 532, 21, 517, 533,
52, 548, 564, 83, 579, 595, 114, 610,
626, 145, 641, 657, 176, 672, 688, 208,
704, 720, 177, 673, 689, 146, 642, 658,
115, 611, 627, 84, 580, 596, 53, 549,
565, 22, 518, 534, 23, 519, 535, 54,
550, 566, 85, 581, 597, 116, 612, 628,
147, 643, 659, 178, 674, 690, 209, 705,
721, 240, 736, 752, 272, 768, 784, 241,
737, 753, 210, 706, 722, 179, 675, 691,
148, 644, 660, 117, 613, 629, 86, 582,
598, 55, 551, 567, 24, 520, 536, 25,
521, 537, 56, 552, 568, 87, 583, 599,
118, 614, 630, 149, 645, 661, 180, 676,
692, 211, 707, 723, 242, 738, 754, 273,
769, 785, 304, 800, 816, 336, 832, 848,
305, 801, 817, 274, 770, 786, 243, 739,
755, 212, 708, 724, 181, 677, 693, 150,
646, 662, 119, 615, 631, 88, 584, 600,
57, 553, 569, 26, 522, 538, 27, 523,
539, 58, 554, 570, 89, 585, 601, 120,
616, 632, 151, 647, 663, 182, 678, 694,
213, 709, 725, 244, 740, 756, 275, 771,
787, 306, 802, 818, 337, 833, 849, 368,
864, 880, 400, 896, 912, 369, 865, 881,
338, 834, 850, 307, 803, 819, 276, 772,
788, 245, 741, 757, 214, 710, 726, 183,
679, 695, 152, 648, 664, 121, 617, 633,
90, 586, 602, 59, 555, 571, 28, 524,
540, 29, 525, 541, 60, 556, 572, 91,
587, 603, 122, 618, 634, 153, 649, 665,
184, 680, 696, 215, 711, 727, 246, 742,
758, 277, 773, 789, 308, 804, 820, 339,
835, 851, 370, 866, 882, 401, 897, 913,
432, 928, 944, 464, 960, 976, 433, 929,
945, 402, 898, 914, 371, 867, 883, 340,
836, 852, 309, 805, 821, 278, 774, 790,
247, 743, 759, 216, 712, 728, 185, 681,
697, 154, 650, 666, 123, 619, 635, 92,
588, 604, 61, 557, 573, 30, 526, 542,
31, 527, 543, 62, 558, 574, 93, 589,
605, 124, 620, 636, 155, 651, 667, 186,
682, 698, 217, 713, 729, 248, 744, 760,
279, 775, 791, 310, 806, 822, 341, 837,
853, 372, 868, 884, 403, 899, 915, 434,
930, 946, 465, 961, 977, 496, 992, 1008,
497, 993, 1009, 466, 962, 978, 435, 931,
947, 404, 900, 916, 373, 869, 885, 342,
838, 854, 311, 807, 823, 280, 776, 792,
249, 745, 761, 218, 714, 730, 187, 683,
699, 156, 652, 668, 125, 621, 637, 94,
590, 606, 63, 559, 575, 95, 591, 607,
126, 622, 638, 157, 653, 669, 188, 684,
700, 219, 715, 731, 250, 746, 762, 281,
777, 793, 312, 808, 824, 343, 839, 855,
374, 870, 886, 405, 901, 917, 436, 932,
948, 467, 963, 979, 498, 994, 1010, 499,
995, 1011, 468, 964, 980, 437, 933, 949,
406, 902, 918, 375, 871, 887, 344, 840,
856, 313, 809, 825, 282, 778, 794, 251,
747, 763, 220, 716, 732, 189, 685, 701,
158, 654, 670, 127, 623, 639, 159, 655,
671, 190, 686, 702, 221, 717, 733, 252,
748, 764, 283, 779, 795, 314, 810, 826,
345, 841, 857, 376, 872, 888, 407, 903,
919, 438, 934, 950, 469, 965, 981, 500,
996, 1012, 501, 997, 1013, 470, 966, 982,
439, 935, 951, 408, 904, 920, 377, 873,
889, 346, 842, 858, 315, 811, 827, 284,
780, 796, 253, 749, 765, 222, 718, 734,
191, 687, 703, 223, 719, 735, 254, 750,
766, 285, 781, 797, 316, 812, 828, 347,
843, 859, 378, 874, 890, 409, 905, 921,
440, 936, 952, 471, 967, 983, 502, 998,
1014, 503, 999, 1015, 472, 968, 984, 441,
937, 953, 410, 906, 922, 379, 875, 891,
348, 844, 860, 317, 813, 829, 286, 782,
798, 255, 751, 767, 287, 783, 799, 318,
814, 830, 349, 845, 861, 380, 876, 892,
411, 907, 923, 442, 938, 954, 473, 969,
985, 504, 1000, 1016, 505, 1001, 1017, 474,
970, 986, 443, 939, 955, 412, 908, 924,
381, 877, 893, 350, 846, 862, 319, 815,
831, 351, 847, 863, 382, 878, 894, 413,
909, 925, 444, 940, 956, 475, 971, 987,
506, 1002, 1018, 507, 1003, 1019, 476, 972,
988, 445, 941, 957, 414, 910, 926, 383,
879, 895, 415, 911, 927, 446, 942, 958,
477, 973, 989, 508, 1004, 1020, 509, 1005,
1021, 478, 974, 990, 447, 943, 959, 479,
975, 991, 510, 1006, 1022, 511, 1007, 1023,
};
#elif DWTDCT_TYPE == DWTDCT8X8
DECLARE_ALIGNED(16, const int, vp9_coef_bands_32x32[1024]) = {
0, 1, 2, 3, 5, 4, 4, 5,
5, 3, 6, 3, 5, 4, 6, 6,
6, 5, 5, 6, 6, 6, 6, 6,
6, 6, 6, 6, 6, 6, 6, 6,
6, 6, 6, 6, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7,
6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
6, 6, 6, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
};
DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_32x32[1024]) = {
0, 1, 32, 64, 33, 2, 3, 34,
65, 96, 128, 97, 66, 35, 4, 5,
36, 67, 98, 129, 160, 192, 161, 130,
99, 68, 37, 6, 7, 38, 69, 100,
131, 162, 193, 224, 225, 194, 163, 132,
101, 70, 39, 71, 102, 133, 164, 195,
226, 227, 196, 165, 134, 103, 135, 166,
197, 228, 229, 198, 167, 199, 230, 231,
8, 256, 264, 9, 257, 265, 40, 288, 296, 72, 320, 328,
41, 289, 297, 10, 258, 266, 11, 259, 267, 42, 290, 298,
73, 321, 329, 104, 352, 360, 136, 384, 392, 105, 353, 361,
74, 322, 330, 43, 291, 299, 12, 260, 268, 13, 261, 269,
44, 292, 300, 75, 323, 331, 106, 354, 362, 137, 385, 393,
168, 416, 424, 200, 448, 456, 169, 417, 425, 138, 386, 394,
107, 355, 363, 76, 324, 332, 45, 293, 301, 14, 262, 270,
15, 263, 271, 46, 294, 302, 77, 325, 333, 108, 356, 364,
139, 387, 395, 170, 418, 426, 201, 449, 457, 232, 480, 488,
233, 481, 489, 202, 450, 458, 171, 419, 427, 140, 388, 396,
109, 357, 365, 78, 326, 334, 47, 295, 303, 79, 327, 335,
110, 358, 366, 141, 389, 397, 172, 420, 428, 203, 451, 459,
234, 482, 490, 235, 483, 491, 204, 452, 460, 173, 421, 429,
142, 390, 398, 111, 359, 367, 143, 391, 399, 174, 422, 430,
205, 453, 461, 236, 484, 492, 237, 485, 493, 206, 454, 462,
175, 423, 431, 207, 455, 463, 238, 486, 494, 239, 487, 495,
16, 512, 528, 17, 513, 529, 18, 514,
530, 19, 515, 531, 20, 516, 532, 21,
517, 533, 22, 518, 534, 23, 519, 535,
24, 520, 536, 25, 521, 537, 26, 522,
538, 27, 523, 539, 28, 524, 540, 29,
525, 541, 30, 526, 542, 31, 527, 543,
48, 544, 560, 49, 545, 561, 50, 546,
562, 51, 547, 563, 52, 548, 564, 53,
549, 565, 54, 550, 566, 55, 551, 567,
56, 552, 568, 57, 553, 569, 58, 554,
570, 59, 555, 571, 60, 556, 572, 61,
557, 573, 62, 558, 574, 63, 559, 575,
80, 576, 592, 81, 577, 593, 82, 578,
594, 83, 579, 595, 84, 580, 596, 85,
581, 597, 86, 582, 598, 87, 583, 599,
88, 584, 600, 89, 585, 601, 90, 586,
602, 91, 587, 603, 92, 588, 604, 93,
589, 605, 94, 590, 606, 95, 591, 607,
112, 608, 624, 113, 609, 625, 114, 610,
626, 115, 611, 627, 116, 612, 628, 117,
613, 629, 118, 614, 630, 119, 615, 631,
120, 616, 632, 121, 617, 633, 122, 618,
634, 123, 619, 635, 124, 620, 636, 125,
621, 637, 126, 622, 638, 127, 623, 639,
144, 640, 656, 145, 641, 657, 146, 642,
658, 147, 643, 659, 148, 644, 660, 149,
645, 661, 150, 646, 662, 151, 647, 663,
152, 648, 664, 153, 649, 665, 154, 650,
666, 155, 651, 667, 156, 652, 668, 157,
653, 669, 158, 654, 670, 159, 655, 671,
176, 672, 688, 177, 673, 689, 178, 674,
690, 179, 675, 691, 180, 676, 692, 181,
677, 693, 182, 678, 694, 183, 679, 695,
184, 680, 696, 185, 681, 697, 186, 682,
698, 187, 683, 699, 188, 684, 700, 189,
685, 701, 190, 686, 702, 191, 687, 703,
208, 704, 720, 209, 705, 721, 210, 706,
722, 211, 707, 723, 212, 708, 724, 213,
709, 725, 214, 710, 726, 215, 711, 727,
216, 712, 728, 217, 713, 729, 218, 714,
730, 219, 715, 731, 220, 716, 732, 221,
717, 733, 222, 718, 734, 223, 719, 735,
240, 736, 752, 241, 737, 753, 242, 738,
754, 243, 739, 755, 244, 740, 756, 245,
741, 757, 246, 742, 758, 247, 743, 759,
248, 744, 760, 249, 745, 761, 250, 746,
762, 251, 747, 763, 252, 748, 764, 253,
749, 765, 254, 750, 766, 255, 751, 767,
272, 768, 784, 273, 769, 785, 274, 770,
786, 275, 771, 787, 276, 772, 788, 277,
773, 789, 278, 774, 790, 279, 775, 791,
280, 776, 792, 281, 777, 793, 282, 778,
794, 283, 779, 795, 284, 780, 796, 285,
781, 797, 286, 782, 798, 287, 783, 799,
304, 800, 816, 305, 801, 817, 306, 802,
818, 307, 803, 819, 308, 804, 820, 309,
805, 821, 310, 806, 822, 311, 807, 823,
312, 808, 824, 313, 809, 825, 314, 810,
826, 315, 811, 827, 316, 812, 828, 317,
813, 829, 318, 814, 830, 319, 815, 831,
336, 832, 848, 337, 833, 849, 338, 834,
850, 339, 835, 851, 340, 836, 852, 341,
837, 853, 342, 838, 854, 343, 839, 855,
344, 840, 856, 345, 841, 857, 346, 842,
858, 347, 843, 859, 348, 844, 860, 349,
845, 861, 350, 846, 862, 351, 847, 863,
368, 864, 880, 369, 865, 881, 370, 866,
882, 371, 867, 883, 372, 868, 884, 373,
869, 885, 374, 870, 886, 375, 871, 887,
376, 872, 888, 377, 873, 889, 378, 874,
890, 379, 875, 891, 380, 876, 892, 381,
877, 893, 382, 878, 894, 383, 879, 895,
400, 896, 912, 401, 897, 913, 402, 898,
914, 403, 899, 915, 404, 900, 916, 405,
901, 917, 406, 902, 918, 407, 903, 919,
408, 904, 920, 409, 905, 921, 410, 906,
922, 411, 907, 923, 412, 908, 924, 413,
909, 925, 414, 910, 926, 415, 911, 927,
432, 928, 944, 433, 929, 945, 434, 930,
946, 435, 931, 947, 436, 932, 948, 437,
933, 949, 438, 934, 950, 439, 935, 951,
440, 936, 952, 441, 937, 953, 442, 938,
954, 443, 939, 955, 444, 940, 956, 445,
941, 957, 446, 942, 958, 447, 943, 959,
464, 960, 976, 465, 961, 977, 466, 962,
978, 467, 963, 979, 468, 964, 980, 469,
965, 981, 470, 966, 982, 471, 967, 983,
472, 968, 984, 473, 969, 985, 474, 970,
986, 475, 971, 987, 476, 972, 988, 477,
973, 989, 478, 974, 990, 479, 975, 991,
496, 992, 1008, 497, 993, 1009, 498, 994,
1010, 499, 995, 1011, 500, 996, 1012, 501,
997, 1013, 502, 998, 1014, 503, 999, 1015,
504, 1000, 1016, 505, 1001, 1017, 506, 1002,
1018, 507, 1003, 1019, 508, 1004, 1020, 509,
1005, 1021, 510, 1006, 1022, 511, 1007, 1023,
};
#endif
#else
DECLARE_ALIGNED(16, const int, vp9_coef_bands_32x32[1024]) = {
0, 1, 2, 3, 5, 4, 4, 5, 5, 3, 6, 3, 5, 4, 6, 6,
6, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
@ -865,7 +247,6 @@ DECLARE_ALIGNED(16, const int, vp9_default_zig_zag1d_32x32[1024]) = {
951, 920, 889, 858, 827, 796, 765, 734, 703, 735, 766, 797, 828, 859, 890, 921, 952, 983, 1014, 1015, 984, 953, 922, 891, 860, 829, 798, 767, 799, 830, 861, 892,
923, 954, 985, 1016, 1017, 986, 955, 924, 893, 862, 831, 863, 894, 925, 956, 987, 1018, 1019, 988, 957, 926, 895, 927, 958, 989, 1020, 1021, 990, 959, 991, 1022, 1023,
};
#endif // CONFIG_DWTDCTHYBRID
/* Array indices are identical to previously-existing CONTEXT_NODE indices */

792
vp9/common/vp9_idctllm.c
View File

@ -1152,8 +1152,6 @@ void vp9_short_idct1_16x16_c(int16_t *input, int16_t *output) {
*output = (out + 32) >> 6;
}
#if !CONFIG_DWTDCTHYBRID
void idct32_1d(int16_t *input, int16_t *output) {
int16_t step1[32], step2[32];
int temp1, temp2;
@ -1521,7 +1519,6 @@ void idct32_1d(int16_t *input, int16_t *output) {
output[31] = step1[0] - step1[31];
}
void vp9_short_idct32x32_c(int16_t *input, int16_t *output, int pitch) {
int16_t out[32 * 32];
int16_t *outptr = &out[0];
@ -1554,792 +1551,3 @@ void vp9_short_idct1_32x32_c(int16_t *input, int16_t *output) {
out = dct_const_round_shift(tmp);
*output = (out + 32) >> 6;
}
#else // !CONFIG_DWTDCTHYBRID
#if DWT_TYPE == 53
// Note: block length must be even for this implementation
static void synthesis_53_row(int length, int16_t *lowpass, int16_t *highpass,
int16_t *x) {
int16_t r, *a, *b;
int n;
n = length >> 1;
b = highpass;
a = lowpass;
r = *highpass;
while (n--) {
*a++ -= (r + (*b) + 1) >> 1;
r = *b++;
}
n = length >> 1;
b = highpass;
a = lowpass;
while (--n) {
*x++ = ((r = *a++) + 1) >> 1;
*x++ = *b++ + ((r + (*a) + 2) >> 2);
}
*x++ = ((r = *a) + 1) >> 1;
*x++ = *b + ((r + 1) >> 1);
}
static void synthesis_53_col(int length, int16_t *lowpass, int16_t *highpass,
int16_t *x) {
int16_t r, *a, *b;
int n;
n = length >> 1;
b = highpass;
a = lowpass;
r = *highpass;
while (n--) {
*a++ -= (r + (*b) + 1) >> 1;
r = *b++;
}
n = length >> 1;
b = highpass;
a = lowpass;
while (--n) {
r = *a++;
*x++ = r;
*x++ = ((*b++) << 1) + ((r + (*a) + 1) >> 1);
}
*x++ = *a;
*x++ = ((*b) << 1) + *a;
}
static void dyadic_synthesize_53(int levels, int width, int height, int16_t *c,
int pitch_c, int16_t *x, int pitch_x) {
int th[16], tw[16], lv, i, j, nh, nw, hh = height, hw = width;
short buffer[2 * DWT_MAX_LENGTH];
th[0] = hh;
tw[0] = hw;
for (i = 1; i <= levels; i++) {
th[i] = (th[i - 1] + 1) >> 1;
tw[i] = (tw[i - 1] + 1) >> 1;
}
for (lv = levels - 1; lv >= 0; lv--) {
nh = th[lv];
nw = tw[lv];
hh = th[lv + 1];
hw = tw[lv + 1];
if ((nh < 2) || (nw < 2)) continue;
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i] = c[i * pitch_c + j];
synthesis_53_col(nh, buffer, buffer + hh, buffer + nh);
for (i = 0; i < nh; i++)
c[i * pitch_c + j] = buffer[i + nh];
}
for (i = 0; i < nh; i++) {
memcpy(buffer, &c[i * pitch_c], nw * sizeof(*buffer));
synthesis_53_row(nw, buffer, buffer + hw, &c[i * pitch_c]);
}
}
for (i = 0; i < height; i++) {
for (j = 0; j < width; j++) {
x[i * pitch_x + j] = c[i * pitch_c + j] >= 0 ?
((c[i * pitch_c + j] + DWT_PRECISION_RND) >> DWT_PRECISION_BITS) :
-((-c[i * pitch_c + j] + DWT_PRECISION_RND) >> DWT_PRECISION_BITS);
}
}
}
#elif DWT_TYPE == 26
// Note: block length must be even for this implementation
static void synthesis_26_row(int length, int16_t *lowpass, int16_t *highpass,
int16_t *x) {
int16_t r, s, *a, *b;
int i, n = length >> 1;
if (n >= 4) {
a = lowpass;
b = highpass;
r = *lowpass;
while (--n) {
*b++ += (r - a[1] + 4) >> 3;
r = *a++;
}
*b += (r - *a + 4) >> 3;
}
a = lowpass;
b = highpass;
for (i = length >> 1; i; i--) {
s = *b++;
r = *a++;
*x++ = (r + s + 1) >> 1;
*x++ = (r - s + 1) >> 1;
}
}
static void synthesis_26_col(int length, int16_t *lowpass, int16_t *highpass,
int16_t *x) {
int16_t r, s, *a, *b;
int i, n = length >> 1;
if (n >= 4) {
a = lowpass;
b = highpass;
r = *lowpass;
while (--n) {
*b++ += (r - a[1] + 4) >> 3;
r = *a++;
}
*b += (r - *a + 4) >> 3;
}
a = lowpass;
b = highpass;
for (i = length >> 1; i; i--) {
s = *b++;
r = *a++;
*x++ = r + s;
*x++ = r - s;
}
}
static void dyadic_synthesize_26(int levels, int width, int height, int16_t *c,
int pitch_c, int16_t *x, int pitch_x) {
int th[16], tw[16], lv, i, j, nh, nw, hh = height, hw = width;
int16_t buffer[2 * DWT_MAX_LENGTH];
th[0] = hh;
tw[0] = hw;
for (i = 1; i <= levels; i++) {
th[i] = (th[i - 1] + 1) >> 1;
tw[i] = (tw[i - 1] + 1) >> 1;
}
for (lv = levels - 1; lv >= 0; lv--) {
nh = th[lv];
nw = tw[lv];
hh = th[lv + 1];
hw = tw[lv + 1];
if ((nh < 2) || (nw < 2)) continue;
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i] = c[i * pitch_c + j];
synthesis_26_col(nh, buffer, buffer + hh, buffer + nh);
for (i = 0; i < nh; i++)
c[i * pitch_c + j] = buffer[i + nh];
}
for (i = 0; i < nh; i++) {
memcpy(buffer, &c[i * pitch_c], nw * sizeof(*buffer));
synthesis_26_row(nw, buffer, buffer + hw, &c[i * pitch_c]);
}
}
for (i = 0; i < height; i++) {
for (j = 0; j < width; j++) {
x[i * pitch_x + j] = c[i * pitch_c + j] >= 0 ?
((c[i * pitch_c + j] + DWT_PRECISION_RND) >> DWT_PRECISION_BITS) :
-((-c[i * pitch_c + j] + DWT_PRECISION_RND) >> DWT_PRECISION_BITS);
}
}
}
#elif DWT_TYPE == 97
static void synthesis_97(int length, double *lowpass, double *highpass,
double *x) {
static const double a_predict1 = -1.586134342;
static const double a_update1 = -0.05298011854;
static const double a_predict2 = 0.8829110762;
static const double a_update2 = 0.4435068522;
static const double s_low = 1.149604398;
static const double s_high = 1/1.149604398;
static const double inv_s_low = 1 / s_low;
static const double inv_s_high = 1 / s_high;
int i;
double y[DWT_MAX_LENGTH];
// Undo pack and scale
for (i = 0; i < length / 2; i++) {
y[i * 2] = lowpass[i] * inv_s_low;
y[i * 2 + 1] = highpass[i] * inv_s_high;
}
memcpy(x, y, sizeof(*y) * length);
// Undo update 2
for (i = 2; i < length; i += 2) {
x[i] -= a_update2 * (x[i-1] + x[i+1]);
}
x[0] -= 2 * a_update2 * x[1];
// Undo predict 2
for (i = 1; i < length - 2; i += 2) {
x[i] -= a_predict2 * (x[i - 1] + x[i + 1]);
}
x[length - 1] -= 2 * a_predict2 * x[length - 2];
// Undo update 1
for (i = 2; i < length; i += 2) {
x[i] -= a_update1 * (x[i - 1] + x[i + 1]);
}
x[0] -= 2 * a_update1 * x[1];
// Undo predict 1
for (i = 1; i < length - 2; i += 2) {
x[i] -= a_predict1 * (x[i - 1] + x[i + 1]);
}
x[length - 1] -= 2 * a_predict1 * x[length - 2];
}
static void dyadic_synthesize_97(int levels, int width, int height, int16_t *c,
int pitch_c, int16_t *x, int pitch_x) {
int th[16], tw[16], lv, i, j, nh, nw, hh = height, hw = width;
double buffer[2 * DWT_MAX_LENGTH];
double y[DWT_MAX_LENGTH * DWT_MAX_LENGTH];
th[0] = hh;
tw[0] = hw;
for (i = 1; i <= levels; i++) {
th[i] = (th[i - 1] + 1) >> 1;
tw[i] = (tw[i - 1] + 1) >> 1;
}
for (lv = levels - 1; lv >= 0; lv--) {
nh = th[lv];
nw = tw[lv];
hh = th[lv + 1];
hw = tw[lv + 1];
if ((nh < 2) || (nw < 2)) continue;
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i] = c[i * pitch_c + j];
synthesis_97(nh, buffer, buffer + hh, buffer + nh);
for (i = 0; i < nh; i++)
y[i * DWT_MAX_LENGTH + j] = buffer[i + nh];
}
for (i = 0; i < nh; i++) {
memcpy(buffer, &y[i * DWT_MAX_LENGTH], nw * sizeof(*buffer));
synthesis_97(nw, buffer, buffer + hw, &y[i * DWT_MAX_LENGTH]);
}
}
for (i = 0; i < height; i++)
for (j = 0; j < width; j++)
x[i * pitch_x + j] = round(y[i * DWT_MAX_LENGTH + j] /
(1 << DWT_PRECISION_BITS));
}
#endif // DWT_TYPE
// TODO(debargha): Implement scaling differently so as not to have to use the
// floating point 16x16 dct
static void butterfly_16x16_idct_1d_f(double input[16], double output[16]) {
static const double C1 = 0.995184726672197;
static const double C2 = 0.98078528040323;
static const double C3 = 0.956940335732209;
static const double C4 = 0.923879532511287;
static const double C5 = 0.881921264348355;
static const double C6 = 0.831469612302545;
static const double C7 = 0.773010453362737;
static const double C8 = 0.707106781186548;
static const double C9 = 0.634393284163646;
static const double C10 = 0.555570233019602;
static const double C11 = 0.471396736825998;
static const double C12 = 0.38268343236509;
static const double C13 = 0.290284677254462;
static const double C14 = 0.195090322016128;
static const double C15 = 0.098017140329561;
vp9_clear_system_state(); // Make it simd safe : __asm emms;
{
double step[16];
double intermediate[16];
double temp1, temp2;
// step 1 and 2
step[ 0] = input[0] + input[8];
step[ 1] = input[0] - input[8];
temp1 = input[4]*C12;
temp2 = input[12]*C4;
temp1 -= temp2;
temp1 *= C8;
step[ 2] = 2*(temp1);
temp1 = input[4]*C4;
temp2 = input[12]*C12;
temp1 += temp2;
temp1 = (temp1);
temp1 *= C8;
step[ 3] = 2*(temp1);
temp1 = input[2]*C8;
temp1 = 2*(temp1);
temp2 = input[6] + input[10];
step[ 4] = temp1 + temp2;
step[ 5] = temp1 - temp2;
temp1 = input[14]*C8;
temp1 = 2*(temp1);
temp2 = input[6] - input[10];
step[ 6] = temp2 - temp1;
step[ 7] = temp2 + temp1;
// for odd input
temp1 = input[3]*C12;
temp2 = input[13]*C4;
temp1 += temp2;
temp1 = (temp1);
temp1 *= C8;
intermediate[ 8] = 2*(temp1);
temp1 = input[3]*C4;
temp2 = input[13]*C12;
temp2 -= temp1;
temp2 = (temp2);
temp2 *= C8;
intermediate[ 9] = 2*(temp2);
intermediate[10] = 2*(input[9]*C8);
intermediate[11] = input[15] - input[1];
intermediate[12] = input[15] + input[1];
intermediate[13] = 2*((input[7]*C8));
temp1 = input[11]*C12;
temp2 = input[5]*C4;
temp2 -= temp1;
temp2 = (temp2);
temp2 *= C8;
intermediate[14] = 2*(temp2);
temp1 = input[11]*C4;
temp2 = input[5]*C12;
temp1 += temp2;
temp1 = (temp1);
temp1 *= C8;
intermediate[15] = 2*(temp1);
step[ 8] = intermediate[ 8] + intermediate[14];
step[ 9] = intermediate[ 9] + intermediate[15];
step[10] = intermediate[10] + intermediate[11];
step[11] = intermediate[10] - intermediate[11];
step[12] = intermediate[12] + intermediate[13];
step[13] = intermediate[12] - intermediate[13];
step[14] = intermediate[ 8] - intermediate[14];
step[15] = intermediate[ 9] - intermediate[15];
// step 3
output[0] = step[ 0] + step[ 3];
output[1] = step[ 1] + step[ 2];
output[2] = step[ 1] - step[ 2];
output[3] = step[ 0] - step[ 3];
temp1 = step[ 4]*C14;
temp2 = step[ 7]*C2;
temp1 -= temp2;
output[4] = (temp1);
temp1 = step[ 4]*C2;
temp2 = step[ 7]*C14;
temp1 += temp2;
output[7] = (temp1);
temp1 = step[ 5]*C10;
temp2 = step[ 6]*C6;
temp1 -= temp2;
output[5] = (temp1);
temp1 = step[ 5]*C6;
temp2 = step[ 6]*C10;
temp1 += temp2;
output[6] = (temp1);
output[8] = step[ 8] + step[11];
output[9] = step[ 9] + step[10];
output[10] = step[ 9] - step[10];
output[11] = step[ 8] - step[11];
output[12] = step[12] + step[15];
output[13] = step[13] + step[14];
output[14] = step[13] - step[14];
output[15] = step[12] - step[15];
// output 4
step[ 0] = output[0] + output[7];
step[ 1] = output[1] + output[6];
step[ 2] = output[2] + output[5];
step[ 3] = output[3] + output[4];
step[ 4] = output[3] - output[4];
step[ 5] = output[2] - output[5];
step[ 6] = output[1] - output[6];
step[ 7] = output[0] - output[7];
temp1 = output[8]*C7;
temp2 = output[15]*C9;
temp1 -= temp2;
step[ 8] = (temp1);
temp1 = output[9]*C11;
temp2 = output[14]*C5;
temp1 += temp2;
step[ 9] = (temp1);
temp1 = output[10]*C3;
temp2 = output[13]*C13;
temp1 -= temp2;
step[10] = (temp1);
temp1 = output[11]*C15;
temp2 = output[12]*C1;
temp1 += temp2;
step[11] = (temp1);
temp1 = output[11]*C1;
temp2 = output[12]*C15;
temp2 -= temp1;
step[12] = (temp2);
temp1 = output[10]*C13;
temp2 = output[13]*C3;
temp1 += temp2;
step[13] = (temp1);
temp1 = output[9]*C5;
temp2 = output[14]*C11;
temp2 -= temp1;
step[14] = (temp2);
temp1 = output[8]*C9;
temp2 = output[15]*C7;
temp1 += temp2;
step[15] = (temp1);
// step 5
output[0] = (step[0] + step[15]);
output[1] = (step[1] + step[14]);
output[2] = (step[2] + step[13]);
output[3] = (step[3] + step[12]);
output[4] = (step[4] + step[11]);
output[5] = (step[5] + step[10]);
output[6] = (step[6] + step[ 9]);
output[7] = (step[7] + step[ 8]);
output[15] = (step[0] - step[15]);
output[14] = (step[1] - step[14]);
output[13] = (step[2] - step[13]);
output[12] = (step[3] - step[12]);
output[11] = (step[4] - step[11]);
output[10] = (step[5] - step[10]);
output[9] = (step[6] - step[ 9]);
output[8] = (step[7] - step[ 8]);
}
vp9_clear_system_state(); // Make it simd safe : __asm emms;
}
static void vp9_short_idct16x16_c_f(int16_t *input, int16_t *output, int pitch,
int scale) {
vp9_clear_system_state(); // Make it simd safe : __asm emms;
{
double out[16*16], out2[16*16];
const int short_pitch = pitch >> 1;
int i, j;
// First transform rows
for (i = 0; i < 16; ++i) {
double temp_in[16], temp_out[16];
for (j = 0; j < 16; ++j)
temp_in[j] = input[j + i*short_pitch];
butterfly_16x16_idct_1d_f(temp_in, temp_out);
for (j = 0; j < 16; ++j)
out[j + i*16] = temp_out[j];
}
// Then transform columns
for (i = 0; i < 16; ++i) {
double temp_in[16], temp_out[16];
for (j = 0; j < 16; ++j)
temp_in[j] = out[j*16 + i];
butterfly_16x16_idct_1d_f(temp_in, temp_out);
for (j = 0; j < 16; ++j)
out2[j*16 + i] = temp_out[j];
}
for (i = 0; i < 16*16; ++i)
output[i] = round(out2[i] / (128 >> scale));
}
vp9_clear_system_state(); // Make it simd safe : __asm emms;
}
static void idct8_1d_f(double *x) {
int i, j;
double t[8];
static const double idctmat[64] = {
0.35355339059327, 0.49039264020162, 0.46193976625564, 0.41573480615127,
0.35355339059327, 0.2777851165098, 0.19134171618254, 0.097545161008064,
0.35355339059327, 0.41573480615127, 0.19134171618254, -0.097545161008064,
-0.35355339059327, -0.49039264020161, -0.46193976625564, -0.2777851165098,
0.35355339059327, 0.2777851165098, -0.19134171618254, -0.49039264020162,
-0.35355339059327, 0.097545161008064, 0.46193976625564, 0.41573480615127,
0.35355339059327, 0.097545161008063, -0.46193976625564, -0.2777851165098,
0.35355339059327, 0.41573480615127, -0.19134171618254, -0.49039264020162,
0.35355339059327, -0.097545161008063, -0.46193976625564, 0.2777851165098,
0.35355339059327, -0.41573480615127, -0.19134171618255, 0.49039264020162,
0.35355339059327, -0.2777851165098, -0.19134171618254, 0.49039264020161,
-0.35355339059327, -0.097545161008064, 0.46193976625564, -0.41573480615127,
0.35355339059327, -0.41573480615127, 0.19134171618254, 0.097545161008065,
-0.35355339059327, 0.49039264020162, -0.46193976625564, 0.2777851165098,
0.35355339059327, -0.49039264020162, 0.46193976625564, -0.41573480615127,
0.35355339059327, -0.2777851165098, 0.19134171618255, -0.097545161008064
};
for (i = 0; i < 8; ++i) {
t[i] = 0;
for (j = 0; j < 8; ++j)
t[i] += idctmat[i * 8 + j] * x[j];
}
for (i = 0; i < 8; ++i) {
x[i] = t[i];
}
}
static void vp9_short_idct8x8_c_f(int16_t *coefs, int16_t *block, int pitch,
int scale) {
double X[8 * 8], Y[8];
int i, j;
int shortpitch = pitch >> 1;
vp9_clear_system_state(); // Make it simd safe : __asm emms;
{
for (i = 0; i < 8; i++) {
for (j = 0; j < 8; j++) {
X[i * 8 + j] = (double)coefs[i * shortpitch + j];
}
}
for (i = 0; i < 8; i++)
idct8_1d_f(X + 8 * i);
for (i = 0; i < 8; i++) {
for (j = 0; j < 8; ++j)
Y[j] = X[i + 8 * j];
idct8_1d_f(Y);
for (j = 0; j < 8; ++j)
X[i + 8 * j] = Y[j];
}
for (i = 0; i < 8; i++) {
for (j = 0; j < 8; j++) {
block[i * 8 + j] = (int16_t)round(X[i * 8 + j] / (8 >> scale));
}
}
}
vp9_clear_system_state(); // Make it simd safe : __asm emms;
}
#define multiply_bits(d, n) ((n) < 0 ? (d) >> (n) : (d) << (n))
#if DWTDCT_TYPE == DWTDCT16X16_LEAN
void vp9_short_idct32x32_c(int16_t *input, int16_t *output, int pitch) {
// assume output is a 32x32 buffer
// Temporary buffer to hold a 16x16 block for 16x16 inverse dct
int16_t buffer[16 * 16];
// Temporary buffer to hold a 32x32 block for inverse 32x32 dwt
int16_t buffer2[32 * 32];
// Note: pitch is in bytes, short_pitch is in short units
const int short_pitch = pitch >> 1;
int i, j;
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the idct16x16 function
vp9_short_idct16x16_c_f(input, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 32, buffer + i * 16, sizeof(*buffer2) * 16);
}
for (i = 0; i < 16; ++i) {
for (j = 16; j < 32; ++j) {
buffer2[i * 32 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 2);
}
}
for (i = 16; i < 32; ++i) {
for (j = 0; j < 32; ++j) {
buffer2[i * 32 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 2);
}
}
#if DWT_TYPE == 26
dyadic_synthesize_26(1, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 97
dyadic_synthesize_97(1, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 53
dyadic_synthesize_53(1, 32, 32, buffer2, 32, output, 32);
#endif
}
#elif DWTDCT_TYPE == DWTDCT16X16
void vp9_short_idct32x32_c(int16_t *input, int16_t *output, int pitch) {
// assume output is a 32x32 buffer
// Temporary buffer to hold a 16x16 block for 16x16 inverse dct
int16_t buffer[16 * 16];
// Temporary buffer to hold a 32x32 block for inverse 32x32 dwt
int16_t buffer2[32 * 32];
// Note: pitch is in bytes, short_pitch is in short units
const int short_pitch = pitch >> 1;
int i, j;
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the idct16x16 function
vp9_short_idct16x16_c_f(input, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 32, buffer + i * 16, sizeof(*buffer2) * 16);
}
vp9_short_idct16x16_c_f(input + 16, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 32 + 16, buffer + i * 16, sizeof(*buffer2) * 16);
}
vp9_short_idct16x16_c_f(input + 16 * short_pitch, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 32 + 16 * 32, buffer + i * 16,
sizeof(*buffer2) * 16);
}
vp9_short_idct16x16_c_f(input + 16 * short_pitch + 16, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 32 + 16 * 33, buffer + i * 16,
sizeof(*buffer2) * 16);
}
#if DWT_TYPE == 26
dyadic_synthesize_26(1, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 97
dyadic_synthesize_97(1, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 53
dyadic_synthesize_53(1, 32, 32, buffer2, 32, output, 32);
#endif
}
#elif DWTDCT_TYPE == DWTDCT8X8
void vp9_short_idct32x32_c(int16_t *input, int16_t *output, int pitch) {
// assume output is a 32x32 buffer
// Temporary buffer to hold a 16x16 block for 16x16 inverse dct
int16_t buffer[8 * 8];
// Temporary buffer to hold a 32x32 block for inverse 32x32 dwt
int16_t buffer2[32 * 32];
// Note: pitch is in bytes, short_pitch is in short units
const int short_pitch = pitch >> 1;
int i, j;
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the idct16x16 function
vp9_short_idct8x8_c_f(input, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i) {
vpx_memcpy(buffer2 + i * 32, buffer + i * 8, sizeof(*buffer2) * 8);
}
vp9_short_idct8x8_c_f(input + 8, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i) {
vpx_memcpy(buffer2 + i * 32 + 8, buffer + i * 8, sizeof(*buffer2) * 8);
}
vp9_short_idct8x8_c_f(input + 8 * short_pitch, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i) {
vpx_memcpy(buffer2 + i * 32 + 8 * 32, buffer + i * 8,
sizeof(*buffer2) * 8);
}
vp9_short_idct8x8_c_f(input + 8 * short_pitch + 8, buffer, pitch,
1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i) {
vpx_memcpy(buffer2 + i * 32 + 8 * 33, buffer + i * 8,
sizeof(*buffer2) * 8);
}
for (i = 0; i < 16; ++i) {
for (j = 16; j < 32; ++j) {
buffer2[i * 32 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 2);
}
}
for (i = 16; i < 32; ++i) {
for (j = 0; j < 32; ++j) {
buffer2[i * 32 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 2);
}
}
#if DWT_TYPE == 26
dyadic_synthesize_26(2, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 97
dyadic_synthesize_97(2, 32, 32, buffer2, 32, output, 32);
#elif DWT_TYPE == 53
dyadic_synthesize_53(2, 32, 32, buffer2, 32, output, 32);
#endif
}
#endif
#if CONFIG_TX64X64
void vp9_short_idct64x64_c(int16_t *input, int16_t *output, int pitch) {
// assume output is a 64x64 buffer
// Temporary buffer to hold a 16x16 block for 16x16 inverse dct
int16_t buffer[16 * 16];
// Temporary buffer to hold a 32x32 block for inverse 32x32 dwt
int16_t buffer2[64 * 64];
// Note: pitch is in bytes, short_pitch is in short units
const int short_pitch = pitch >> 1;
int i, j;
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the idct16x16 function
vp9_short_idct16x16_c_f(input, buffer, pitch,
2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 64, buffer + i * 16, sizeof(*buffer2) * 16);
}
#if DWTDCT_TYPE == DWTDCT16X16_LEAN
for (i = 0; i < 16; ++i) {
for (j = 16; j < 64; ++j) {
buffer2[i * 64 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 1);
}
}
for (i = 16; i < 64; ++i) {
for (j = 0; j < 64; ++j) {
buffer2[i * 64 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 1);
}
}
#elif DWTDCT_TYPE == DWTDCT16X16
vp9_short_idct16x16_c_f(input + 16, buffer, pitch,
2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 64 + 16, buffer + i * 16, sizeof(*buffer2) * 16);
}
vp9_short_idct16x16_c_f(input + 16 * short_pitch, buffer, pitch,
2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 64 + 16 * 64, buffer + i * 16,
sizeof(*buffer2) * 16);
}
vp9_short_idct16x16_c_f(input + 16 * short_pitch + 16, buffer, pitch,
2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i) {
vpx_memcpy(buffer2 + i * 64 + 16 * 65, buffer + i * 16,
sizeof(*buffer2) * 16);
}
// Copying and scaling highest bands into buffer2
for (i = 0; i < 32; ++i) {
for (j = 32; j < 64; ++j) {
buffer2[i * 64 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 1);
}
}
for (i = 32; i < 64; ++i) {
for (j = 0; j < 64; ++j) {
buffer2[i * 64 + j] =
multiply_bits(input[i * short_pitch + j], DWT_PRECISION_BITS - 1);
}
}
#endif // DWTDCT_TYPE
#if DWT_TYPE == 26
dyadic_synthesize_26(2, 64, 64, buffer2, 64, output, 64);
#elif DWT_TYPE == 97
dyadic_synthesize_97(2, 64, 64, buffer2, 64, output, 64);
#elif DWT_TYPE == 53
dyadic_synthesize_53(2, 64, 64, buffer2, 64, output, 64);
#endif
}
#endif // CONFIG_TX64X64
#endif // !CONFIG_DWTDCTHYBRID

4
vp9/decoder/vp9_dequantize.c
View File

@ -357,21 +357,17 @@ void vp9_dequant_idct_add_32x32_c(int16_t *input, const int16_t *dq,
if (eob) {
input[0] = input[0] * dq[0] / 2;
#if !CONFIG_DWTDCTHYBRID
if (eob == 1) {
vp9_short_idct1_32x32_c(input, output);
add_constant_residual(output[0], pred, pitch, dest, stride, 32, 32);
input[0] = 0;
} else {
#endif
for (i = 1; i < 1024; i++)
input[i] = input[i] * dq[1] / 2;
vp9_short_idct32x32_c(input, output, 64);
vpx_memset(input, 0, 2048);
add_residual(output, pred, pitch, dest, stride, 32, 32);
#if !CONFIG_DWTDCTHYBRID
}
#endif
}
}

705
vp9/encoder/vp9_dct.c
View File

@ -1471,8 +1471,6 @@ void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {
#undef ROUNDING
#endif
#if !CONFIG_DWTDCTHYBRID
#define TEST_INT_32x32_DCT 1
#if !TEST_INT_32x32_DCT
@ -2210,706 +2208,3 @@ void vp9_short_fdct32x32_c(int16_t *input, int16_t *out, int pitch) {
}
#endif
#else // CONFIG_DWTDCTHYBRID
#if DWT_TYPE == 53
// Note: block length must be even for this implementation
static void analysis_53_row(int length, short *x,
short *lowpass, short *highpass) {
int n;
short r, *a, *b;
n = length >> 1;
b = highpass;
a = lowpass;
while (--n) {
*a++ = (r = *x++) << 1;
*b++ = *x - ((r + x[1] + 1) >> 1);
x++;
}
*a = (r = *x++) << 1;
*b = *x - r;
n = length >> 1;
b = highpass;
a = lowpass;
r = *highpass;
while (n--) {
*a++ += (r + (*b) + 1) >> 1;
r = *b++;
}
}
static void analysis_53_col(int length, short *x,
short *lowpass, short *highpass) {
int n;
short r, *a, *b;
n = length >> 1;
b = highpass;
a = lowpass;
while (--n) {
*a++ = (r = *x++);
*b++ = (((*x) << 1) - (r + x[1]) + 2) >> 2;
x++;
}
*a = (r = *x++);
*b = (*x - r + 1) >> 1;
n = length >> 1;
b = highpass;
a = lowpass;
r = *highpass;
while (n--) {
*a++ += (r + (*b) + 1) >> 1;
r = *b++;
}
}
static void dyadic_analyze_53(int levels, int width, int height,
short *x, int pitch_x, short *c, int pitch_c) {
int lv, i, j, nh, nw, hh = height, hw = width;
short buffer[2 * DWT_MAX_LENGTH];
for (i = 0; i < height; i++) {
for (j = 0; j < width; j++) {
c[i * pitch_c + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
}
}
for (lv = 0; lv < levels; lv++) {
nh = hh;
hh = (hh + 1) >> 1;
nw = hw;
hw = (hw + 1) >> 1;
if ((nh < 2) || (nw < 2)) return;
for (i = 0; i < nh; i++) {
memcpy(buffer, &c[i * pitch_c], nw * sizeof(short));
analysis_53_row(nw, buffer, &c[i * pitch_c], &c[i * pitch_c] + hw);
}
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i + nh] = c[i * pitch_c + j];
analysis_53_col(nh, buffer + nh, buffer, buffer + hh);
for (i = 0; i < nh; i++)
c[i * pitch_c + j] = buffer[i];
}
}
}
#elif DWT_TYPE == 26
static void analysis_26_row(int length, short *x,
short *lowpass, short *highpass) {
int i, n;
short r, s, *a, *b;
a = lowpass;
b = highpass;
for (i = length >> 1; i; i--) {
r = *x++;
s = *x++;
*a++ = r + s;
*b++ = r - s;
}
n = length >> 1;
if (n >= 4) {
a = lowpass;
b = highpass;
r = *lowpass;
while (--n) {
*b++ -= (r - a[1] + 4) >> 3;
r = *a++;
}
*b -= (r - *a + 4) >> 3;
}
}
static void analysis_26_col(int length, short *x,
short *lowpass, short *highpass) {
int i, n;
short r, s, *a, *b;
a = lowpass;
b = highpass;
for (i = length >> 1; i; i--) {
r = *x++;
s = *x++;
*a++ = (r + s + 1) >> 1;
*b++ = (r - s + 1) >> 1;
}
n = length >> 1;
if (n >= 4) {
a = lowpass;
b = highpass;
r = *lowpass;
while (--n) {
*b++ -= (r - a[1] + 4) >> 3;
r = *a++;
}
*b -= (r - *a + 4) >> 3;
}
}
static void dyadic_analyze_26(int levels, int width, int height,
short *x, int pitch_x, short *c, int pitch_c) {
int lv, i, j, nh, nw, hh = height, hw = width;
short buffer[2 * DWT_MAX_LENGTH];
for (i = 0; i < height; i++) {
for (j = 0; j < width; j++) {
c[i * pitch_c + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
}
}
for (lv = 0; lv < levels; lv++) {
nh = hh;
hh = (hh + 1) >> 1;
nw = hw;
hw = (hw + 1) >> 1;
if ((nh < 2) || (nw < 2)) return;
for (i = 0; i < nh; i++) {
memcpy(buffer, &c[i * pitch_c], nw * sizeof(short));
analysis_26_row(nw, buffer, &c[i * pitch_c], &c[i * pitch_c] + hw);
}
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i + nh] = c[i * pitch_c + j];
analysis_26_col(nh, buffer + nh, buffer, buffer + hh);
for (i = 0; i < nh; i++)
c[i * pitch_c + j] = buffer[i];
}
}
}
#elif DWT_TYPE == 97
static void analysis_97(int length, double *x,
double *lowpass, double *highpass) {
static const double a_predict1 = -1.586134342;
static const double a_update1 = -0.05298011854;
static const double a_predict2 = 0.8829110762;
static const double a_update2 = 0.4435068522;
static const double s_low = 1.149604398;
static const double s_high = 1/1.149604398;
int i;
double y[DWT_MAX_LENGTH];
// Predict 1
for (i = 1; i < length - 2; i += 2) {
x[i] += a_predict1 * (x[i - 1] + x[i + 1]);
}
x[length - 1] += 2 * a_predict1 * x[length - 2];
// Update 1
for (i = 2; i < length; i += 2) {
x[i] += a_update1 * (x[i - 1] + x[i + 1]);
}
x[0] += 2 * a_update1 * x[1];
// Predict 2
for (i = 1; i < length - 2; i += 2) {
x[i] += a_predict2 * (x[i - 1] + x[i + 1]);
}
x[length - 1] += 2 * a_predict2 * x[length - 2];
// Update 2
for (i = 2; i < length; i += 2) {
x[i] += a_update2 * (x[i - 1] + x[i + 1]);
}
x[0] += 2 * a_update2 * x[1];
memcpy(y, x, sizeof(*y) * length);
// Scale and pack
for (i = 0; i < length / 2; i++) {
lowpass[i] = y[2 * i] * s_low;
highpass[i] = y[2 * i + 1] * s_high;
}
}
static void dyadic_analyze_97(int levels, int width, int height,
short *x, int pitch_x, short *c, int pitch_c) {
int lv, i, j, nh, nw, hh = height, hw = width;
double buffer[2 * DWT_MAX_LENGTH];
double y[DWT_MAX_LENGTH * DWT_MAX_LENGTH];
for (i = 0; i < height; i++) {
for (j = 0; j < width; j++) {
y[i * DWT_MAX_LENGTH + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
}
}
for (lv = 0; lv < levels; lv++) {
nh = hh;
hh = (hh + 1) >> 1;
nw = hw;
hw = (hw + 1) >> 1;
if ((nh < 2) || (nw < 2)) return;
for (i = 0; i < nh; i++) {
memcpy(buffer, &y[i * DWT_MAX_LENGTH], nw * sizeof(*buffer));
analysis_97(nw, buffer, &y[i * DWT_MAX_LENGTH],
&y[i * DWT_MAX_LENGTH] + hw);
}
for (j = 0; j < nw; j++) {
for (i = 0; i < nh; i++)
buffer[i + nh] = y[i * DWT_MAX_LENGTH + j];
analysis_97(nh, buffer + nh, buffer, buffer + hh);
for (i = 0; i < nh; i++)
c[i * pitch_c + j] = round(buffer[i]);
}
}
}
#endif // DWT_TYPE
// TODO(debargha): Implement the scaling differently so as not to have to
// use the floating point dct
static void dct16x16_1d_f(double input[16], double output[16]) {
static const double C1 = 0.995184726672197;
static const double C2 = 0.98078528040323;
static const double C3 = 0.956940335732209;
static const double C4 = 0.923879532511287;
static const double C5 = 0.881921264348355;
static const double C6 = 0.831469612302545;
static const double C7 = 0.773010453362737;
static const double C8 = 0.707106781186548;
static const double C9 = 0.634393284163646;
static const double C10 = 0.555570233019602;
static const double C11 = 0.471396736825998;
static const double C12 = 0.38268343236509;
static const double C13 = 0.290284677254462;
static const double C14 = 0.195090322016128;
static const double C15 = 0.098017140329561;
vp9_clear_system_state(); // Make it simd safe : __asm emms;
{
double step[16];
double intermediate[16];
double temp1, temp2;
// step 1
step[ 0] = input[0] + input[15];
step[ 1] = input[1] + input[14];
step[ 2] = input[2] + input[13];
step[ 3] = input[3] + input[12];
step[ 4] = input[4] + input[11];
step[ 5] = input[5] + input[10];
step[ 6] = input[6] + input[ 9];
step[ 7] = input[7] + input[ 8];
step[ 8] = input[7] - input[ 8];
step[ 9] = input[6] - input[ 9];
step[10] = input[5] - input[10];
step[11] = input[4] - input[11];
step[12] = input[3] - input[12];
step[13] = input[2] - input[13];
step[14] = input[1] - input[14];
step[15] = input[0] - input[15];
// step 2
output[0] = step[0] + step[7];
output[1] = step[1] + step[6];
output[2] = step[2] + step[5];
output[3] = step[3] + step[4];
output[4] = step[3] - step[4];
output[5] = step[2] - step[5];
output[6] = step[1] - step[6];
output[7] = step[0] - step[7];
temp1 = step[ 8]*C7;
temp2 = step[15]*C9;
output[ 8] = temp1 + temp2;
temp1 = step[ 9]*C11;
temp2 = step[14]*C5;
output[ 9] = temp1 - temp2;
temp1 = step[10]*C3;
temp2 = step[13]*C13;
output[10] = temp1 + temp2;
temp1 = step[11]*C15;
temp2 = step[12]*C1;
output[11] = temp1 - temp2;
temp1 = step[11]*C1;
temp2 = step[12]*C15;
output[12] = temp2 + temp1;
temp1 = step[10]*C13;
temp2 = step[13]*C3;
output[13] = temp2 - temp1;
temp1 = step[ 9]*C5;
temp2 = step[14]*C11;
output[14] = temp2 + temp1;
temp1 = step[ 8]*C9;
temp2 = step[15]*C7;
output[15] = temp2 - temp1;
// step 3
step[ 0] = output[0] + output[3];
step[ 1] = output[1] + output[2];
step[ 2] = output[1] - output[2];
step[ 3] = output[0] - output[3];
temp1 = output[4]*C14;
temp2 = output[7]*C2;
step[ 4] = temp1 + temp2;
temp1 = output[5]*C10;
temp2 = output[6]*C6;
step[ 5] = temp1 + temp2;
temp1 = output[5]*C6;
temp2 = output[6]*C10;
step[ 6] = temp2 - temp1;
temp1 = output[4]*C2;
temp2 = output[7]*C14;
step[ 7] = temp2 - temp1;
step[ 8] = output[ 8] + output[11];
step[ 9] = output[ 9] + output[10];
step[10] = output[ 9] - output[10];
step[11] = output[ 8] - output[11];
step[12] = output[12] + output[15];
step[13] = output[13] + output[14];
step[14] = output[13] - output[14];
step[15] = output[12] - output[15];
// step 4
output[ 0] = (step[ 0] + step[ 1]);
output[ 8] = (step[ 0] - step[ 1]);
temp1 = step[2]*C12;
temp2 = step[3]*C4;
temp1 = temp1 + temp2;
output[ 4] = 2*(temp1*C8);
temp1 = step[2]*C4;
temp2 = step[3]*C12;
temp1 = temp2 - temp1;
output[12] = 2*(temp1*C8);
output[ 2] = 2*((step[4] + step[ 5])*C8);
output[14] = 2*((step[7] - step[ 6])*C8);
temp1 = step[4] - step[5];
temp2 = step[6] + step[7];
output[ 6] = (temp1 + temp2);
output[10] = (temp1 - temp2);
intermediate[8] = step[8] + step[14];
intermediate[9] = step[9] + step[15];
temp1 = intermediate[8]*C12;
temp2 = intermediate[9]*C4;
temp1 = temp1 - temp2;
output[3] = 2*(temp1*C8);
temp1 = intermediate[8]*C4;
temp2 = intermediate[9]*C12;
temp1 = temp2 + temp1;
output[13] = 2*(temp1*C8);
output[ 9] = 2*((step[10] + step[11])*C8);
intermediate[11] = step[10] - step[11];
intermediate[12] = step[12] + step[13];
intermediate[13] = step[12] - step[13];
intermediate[14] = step[ 8] - step[14];
intermediate[15] = step[ 9] - step[15];
output[15] = (intermediate[11] + intermediate[12]);
output[ 1] = -(intermediate[11] - intermediate[12]);
output[ 7] = 2*(intermediate[13]*C8);
temp1 = intermediate[14]*C12;
temp2 = intermediate[15]*C4;
temp1 = temp1 - temp2;
output[11] = -2*(temp1*C8);
temp1 = intermediate[14]*C4;
temp2 = intermediate[15]*C12;
temp1 = temp2 + temp1;
output[ 5] = 2*(temp1*C8);
}
vp9_clear_system_state(); // Make it simd safe : __asm emms;
}
static void vp9_short_fdct16x16_c_f(short *input, short *out, int pitch,
int scale) {
vp9_clear_system_state(); // Make it simd safe : __asm emms;
{
int shortpitch = pitch >> 1;
int i, j;
double output[256];
// First transform columns
for (i = 0; i < 16; i++) {
double temp_in[16], temp_out[16];
for (j = 0; j < 16; j++)
temp_in[j] = input[j*shortpitch + i];
dct16x16_1d_f(temp_in, temp_out);
for (j = 0; j < 16; j++)
output[j*16 + i] = temp_out[j];
}
// Then transform rows
for (i = 0; i < 16; ++i) {
double temp_in[16], temp_out[16];
for (j = 0; j < 16; ++j)
temp_in[j] = output[j + i*16];
dct16x16_1d_f(temp_in, temp_out);
for (j = 0; j < 16; ++j)
output[j + i*16] = temp_out[j];
}
// Scale by some magic number
for (i = 0; i < 256; i++)
out[i] = (short)round(output[i] / (2 << scale));
}
vp9_clear_system_state(); // Make it simd safe : __asm emms;
}
void vp9_short_fdct8x8_c_f(short *block, short *coefs, int pitch, int scale) {
int j1, i, j, k;
float b[8];
float b1[8];
float d[8][8];
float f0 = (float) .7071068;
float f1 = (float) .4903926;
float f2 = (float) .4619398;
float f3 = (float) .4157348;
float f4 = (float) .3535534;
float f5 = (float) .2777851;
float f6 = (float) .1913417;
float f7 = (float) .0975452;
pitch = pitch / 2;
for (i = 0, k = 0; i < 8; i++, k += pitch) {
for (j = 0; j < 8; j++) {
b[j] = (float)(block[k + j] << (3 - scale));
}
/* Horizontal transform */
for (j = 0; j < 4; j++) {
j1 = 7 - j;
b1[j] = b[j] + b[j1];
b1[j1] = b[j] - b[j1];
}
b[0] = b1[0] + b1[3];
b[1] = b1[1] + b1[2];
b[2] = b1[1] - b1[2];
b[3] = b1[0] - b1[3];
b[4] = b1[4];
b[5] = (b1[6] - b1[5]) * f0;
b[6] = (b1[6] + b1[5]) * f0;
b[7] = b1[7];
d[i][0] = (b[0] + b[1]) * f4;
d[i][4] = (b[0] - b[1]) * f4;
d[i][2] = b[2] * f6 + b[3] * f2;
d[i][6] = b[3] * f6 - b[2] * f2;
b1[4] = b[4] + b[5];
b1[7] = b[7] + b[6];
b1[5] = b[4] - b[5];
b1[6] = b[7] - b[6];
d[i][1] = b1[4] * f7 + b1[7] * f1;
d[i][5] = b1[5] * f3 + b1[6] * f5;
d[i][7] = b1[7] * f7 - b1[4] * f1;
d[i][3] = b1[6] * f3 - b1[5] * f5;
}
/* Vertical transform */
for (i = 0; i < 8; i++) {
for (j = 0; j < 4; j++) {
j1 = 7 - j;
b1[j] = d[j][i] + d[j1][i];
b1[j1] = d[j][i] - d[j1][i];
}
b[0] = b1[0] + b1[3];
b[1] = b1[1] + b1[2];
b[2] = b1[1] - b1[2];
b[3] = b1[0] - b1[3];
b[4] = b1[4];
b[5] = (b1[6] - b1[5]) * f0;
b[6] = (b1[6] + b1[5]) * f0;
b[7] = b1[7];
d[0][i] = (b[0] + b[1]) * f4;
d[4][i] = (b[0] - b[1]) * f4;
d[2][i] = b[2] * f6 + b[3] * f2;
d[6][i] = b[3] * f6 - b[2] * f2;
b1[4] = b[4] + b[5];
b1[7] = b[7] + b[6];
b1[5] = b[4] - b[5];
b1[6] = b[7] - b[6];
d[1][i] = b1[4] * f7 + b1[7] * f1;
d[5][i] = b1[5] * f3 + b1[6] * f5;
d[7][i] = b1[7] * f7 - b1[4] * f1;
d[3][i] = b1[6] * f3 - b1[5] * f5;
}
for (i = 0; i < 8; i++) {
for (j = 0; j < 8; j++) {
*(coefs + j + i * 8) = (short) floor(d[i][j] + 0.5);
}
}
return;
}
#define divide_bits(d, n) ((n) < 0 ? (d) << (n) : (d) >> (n))
#if DWTDCT_TYPE == DWTDCT16X16_LEAN
void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
// assume out is a 32x32 buffer
short buffer[16 * 16];
int i, j;
const int short_pitch = pitch >> 1;
#if DWT_TYPE == 26
dyadic_analyze_26(1, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 97
dyadic_analyze_97(1, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 53
dyadic_analyze_53(1, 32, 32, input, short_pitch, out, 32);
#endif
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the dct16x16 function
vp9_short_fdct16x16_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 32, buffer + i * 16, sizeof(short) * 16);
for (i = 0; i < 16; ++i) {
for (j = 16; j < 32; ++j) {
out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
}
}
for (i = 16; i < 32; ++i) {
for (j = 0; j < 32; ++j) {
out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
}
}
}
#elif DWTDCT_TYPE == DWTDCT16X16
void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
// assume out is a 32x32 buffer
short buffer[16 * 16];
int i, j;
const int short_pitch = pitch >> 1;
#if DWT_TYPE == 26
dyadic_analyze_26(1, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 97
dyadic_analyze_97(1, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 53
dyadic_analyze_53(1, 32, 32, input, short_pitch, out, 32);
#endif
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the dct16x16 function
vp9_short_fdct16x16_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 32, buffer + i * 16, sizeof(short) * 16);
vp9_short_fdct16x16_c_f(out + 16, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 32 + 16, buffer + i * 16, sizeof(short) * 16);
vp9_short_fdct16x16_c_f(out + 32 * 16, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 32 + 32 * 16, buffer + i * 16, sizeof(short) * 16);
vp9_short_fdct16x16_c_f(out + 33 * 16, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 32 + 33 * 16, buffer + i * 16, sizeof(short) * 16);
}
#elif DWTDCT_TYPE == DWTDCT8X8
void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
// assume out is a 32x32 buffer
short buffer[8 * 8];
int i, j;
const int short_pitch = pitch >> 1;
#if DWT_TYPE == 26
dyadic_analyze_26(2, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 97
dyadic_analyze_97(2, 32, 32, input, short_pitch, out, 32);
#elif DWT_TYPE == 53
dyadic_analyze_53(2, 32, 32, input, short_pitch, out, 32);
#endif
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the dct16x16 function
vp9_short_fdct8x8_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i)
vpx_memcpy(out + i * 32, buffer + i * 8, sizeof(short) * 8);
vp9_short_fdct8x8_c_f(out + 8, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i)
vpx_memcpy(out + i * 32 + 8, buffer + i * 8, sizeof(short) * 8);
vp9_short_fdct8x8_c_f(out + 32 * 8, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i)
vpx_memcpy(out + i * 32 + 32 * 8, buffer + i * 8, sizeof(short) * 8);
vp9_short_fdct8x8_c_f(out + 33 * 8, buffer, 64, 1 + DWT_PRECISION_BITS);
for (i = 0; i < 8; ++i)
vpx_memcpy(out + i * 32 + 33 * 8, buffer + i * 8, sizeof(short) * 8);
for (i = 0; i < 16; ++i) {
for (j = 16; j < 32; ++j) {
out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
}
}
for (i = 16; i < 32; ++i) {
for (j = 0; j < 32; ++j) {
out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
}
}
}
#endif
#if CONFIG_TX64X64
void vp9_short_fdct64x64_c(short *input, short *out, int pitch) {
// assume out is a 64x64 buffer
short buffer[16 * 16];
int i, j;
const int short_pitch = pitch >> 1;
#if DWT_TYPE == 26
dyadic_analyze_26(2, 64, 64, input, short_pitch, out, 64);
#elif DWT_TYPE == 97
dyadic_analyze_97(2, 64, 64, input, short_pitch, out, 64);
#elif DWT_TYPE == 53
dyadic_analyze_53(2, 64, 64, input, short_pitch, out, 64);
#endif
// TODO(debargha): Implement more efficiently by adding output pitch
// argument to the dct16x16 function
vp9_short_fdct16x16_c_f(out, buffer, 128, 2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 64, buffer + i * 16, sizeof(short) * 16);
#if DWTDCT_TYPE == DWTDCT16X16_LEAN
for (i = 0; i < 16; ++i) {
for (j = 16; j < 48; ++j) {
out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
}
}
for (i = 16; i < 64; ++i) {
for (j = 0; j < 64; ++j) {
out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
}
}
#elif DWTDCT_TYPE == DWTDCT16X16
vp9_short_fdct16x16_c_f(out + 16, buffer, 128, 2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 64 + 16, buffer + i * 16, sizeof(short) * 16);
vp9_short_fdct16x16_c_f(out + 64 * 16, buffer, 128, 2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 64 + 64 * 16, buffer + i * 16, sizeof(short) * 16);
vp9_short_fdct16x16_c_f(out + 65 * 16, buffer, 128, 2 + DWT_PRECISION_BITS);
for (i = 0; i < 16; ++i)
vpx_memcpy(out + i * 64 + 65 * 16, buffer + i * 16, sizeof(short) * 16);
// There is no dct used on the highest bands for now.
// Need to scale these coeffs by a factor of 2/2^DWT_PRECISION_BITS
// TODO(debargha): experiment with turning these coeffs to 0
for (i = 0; i < 32; ++i) {
for (j = 32; j < 64; ++j) {
out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
}
}
for (i = 32; i < 64; ++i) {
for (j = 0; j < 64; ++j) {
out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
}
}
#endif // DWTDCT_TYPE
}
#endif // CONFIG_TX64X64
#endif // CONFIG_DWTDCTHYBRID

9
vp9/encoder/vp9_rdopt.c
View File

@ -859,21 +859,18 @@ static void super_block_yrd_32x32(MACROBLOCK *x,
SUPERBLOCK * const x_sb = &x->sb_coeff_data;
MACROBLOCKD * const xd = &x->e_mbd;
SUPERBLOCKD * const xd_sb = &xd->sb_coeff_data;
#if DEBUG_ERROR || CONFIG_DWTDCTHYBRID
#if DEBUG_ERROR
int16_t out[1024];
#endif
vp9_transform_sby_32x32(x);
vp9_quantize_sby_32x32(x);
#if DEBUG_ERROR || CONFIG_DWTDCTHYBRID
#if DEBUG_ERROR
vp9_short_idct32x32(xd_sb->dqcoeff, out, 64);
#endif
#if !CONFIG_DWTDCTHYBRID
*distortion = vp9_sb_block_error_c(x_sb->coeff, xd_sb->dqcoeff, 1024);
#else
*distortion = vp9_block_error_c(x_sb->src_diff, out, 1024) << 4;
#endif
#if DEBUG_ERROR
printf("IDCT/FDCT error 32x32: %d (d: %d)\n",
vp9_block_error_c(x_sb->src_diff, out, 1024), *distortion);