Newer
Older
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
* gmc & q-pel & 32/64 bit based MC by Michael Niedermayer <michaelni@gmx.at>
#include "simple_idct.h"
uint8_t cropTbl[256 + 2 * MAX_NEG_CROP];
uint32_t squareTbl[512];
const uint8_t ff_zigzag_direct[64] = {
0, 1, 8, 16, 9, 2, 3, 10,
17, 24, 32, 25, 18, 11, 4, 5,
27, 20, 13, 6, 7, 14, 21, 28,
35, 42, 49, 56, 57, 50, 43, 36,
29, 22, 15, 23, 30, 37, 44, 51,
58, 59, 52, 45, 38, 31, 39, 46,
53, 60, 61, 54, 47, 55, 62, 63
};
/* not permutated inverse zigzag_direct + 1 for MMX quantizer */
uint16_t __align8 inv_zigzag_direct16[64];
const uint8_t ff_alternate_horizontal_scan[64] = {
0, 1, 2, 3, 8, 9, 16, 17,
10, 11, 4, 5, 6, 7, 15, 14,
13, 12, 19, 18, 24, 25, 32, 33,
26, 27, 20, 21, 22, 23, 28, 29,
30, 31, 34, 35, 40, 41, 48, 49,
42, 43, 36, 37, 38, 39, 44, 45,
46, 47, 50, 51, 56, 57, 58, 59,
52, 53, 54, 55, 60, 61, 62, 63,
};
const uint8_t ff_alternate_vertical_scan[64] = {
0, 8, 16, 24, 1, 9, 2, 10,
17, 25, 32, 40, 48, 56, 57, 49,
41, 33, 26, 18, 3, 11, 4, 12,
19, 27, 34, 42, 50, 58, 35, 43,
51, 59, 20, 28, 5, 13, 6, 14,
21, 29, 36, 44, 52, 60, 37, 45,
53, 61, 22, 30, 7, 15, 23, 31,
38, 46, 54, 62, 39, 47, 55, 63,
};
/* a*inverse[b]>>32 == a/b for all 0<=a<=65536 && 2<=b<=255 */
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
0, 4294967295U,2147483648U,1431655766, 1073741824, 858993460, 715827883, 613566757,
536870912, 477218589, 429496730, 390451573, 357913942, 330382100, 306783379, 286331154,
268435456, 252645136, 238609295, 226050911, 214748365, 204522253, 195225787, 186737709,
178956971, 171798692, 165191050, 159072863, 153391690, 148102321, 143165577, 138547333,
134217728, 130150525, 126322568, 122713352, 119304648, 116080198, 113025456, 110127367,
107374183, 104755300, 102261127, 99882961, 97612894, 95443718, 93368855, 91382283,
89478486, 87652394, 85899346, 84215046, 82595525, 81037119, 79536432, 78090315,
76695845, 75350304, 74051161, 72796056, 71582789, 70409300, 69273667, 68174085,
67108864, 66076420, 65075263, 64103990, 63161284, 62245903, 61356676, 60492498,
59652324, 58835169, 58040099, 57266231, 56512728, 55778797, 55063684, 54366675,
53687092, 53024288, 52377650, 51746594, 51130564, 50529028, 49941481, 49367441,
48806447, 48258060, 47721859, 47197443, 46684428, 46182445, 45691142, 45210183,
44739243, 44278014, 43826197, 43383509, 42949673, 42524429, 42107523, 41698712,
41297763, 40904451, 40518560, 40139882, 39768216, 39403370, 39045158, 38693400,
38347923, 38008561, 37675152, 37347542, 37025581, 36709123, 36398028, 36092163,
35791395, 35495598, 35204650, 34918434, 34636834, 34359739, 34087043, 33818641,
33554432, 33294321, 33038210, 32786010, 32537632, 32292988, 32051995, 31814573,
31580642, 31350127, 31122952, 30899046, 30678338, 30460761, 30246249, 30034737,
29826162, 29620465, 29417585, 29217465, 29020050, 28825284, 28633116, 28443493,
28256364, 28071682, 27889399, 27709467, 27531842, 27356480, 27183338, 27012373,
26843546, 26676816, 26512144, 26349493, 26188825, 26030105, 25873297, 25718368,
25565282, 25414008, 25264514, 25116768, 24970741, 24826401, 24683721, 24542671,
24403224, 24265352, 24129030, 23994231, 23860930, 23729102, 23598722, 23469767,
23342214, 23216040, 23091223, 22967740, 22845571, 22724695, 22605092, 22486740,
22369622, 22253717, 22139007, 22025474, 21913099, 21801865, 21691755, 21582751,
21474837, 21367997, 21262215, 21157475, 21053762, 20951060, 20849356, 20748635,
20648882, 20550083, 20452226, 20355296, 20259280, 20164166, 20069941, 19976593,
19884108, 19792477, 19701685, 19611723, 19522579, 19434242, 19346700, 19259944,
19173962, 19088744, 19004281, 18920561, 18837576, 18755316, 18673771, 18592933,
18512791, 18433337, 18354562, 18276457, 18199014, 18122225, 18046082, 17970575,
17895698, 17821442, 17747799, 17674763, 17602325, 17530479, 17459217, 17388532,
17318417, 17248865, 17179870, 17111424, 17043522, 16976156, 16909321, 16843010,
};
/* Input permutation for the simple_idct_mmx */
static const uint8_t simple_mmx_permutation[64]={
0x00, 0x08, 0x04, 0x09, 0x01, 0x0C, 0x05, 0x0D,
0x10, 0x18, 0x14, 0x19, 0x11, 0x1C, 0x15, 0x1D,
0x20, 0x28, 0x24, 0x29, 0x21, 0x2C, 0x25, 0x2D,
0x12, 0x1A, 0x16, 0x1B, 0x13, 0x1E, 0x17, 0x1F,
0x02, 0x0A, 0x06, 0x0B, 0x03, 0x0E, 0x07, 0x0F,
0x30, 0x38, 0x34, 0x39, 0x31, 0x3C, 0x35, 0x3D,
0x22, 0x2A, 0x26, 0x2B, 0x23, 0x2E, 0x27, 0x2F,
0x32, 0x3A, 0x36, 0x3B, 0x33, 0x3E, 0x37, 0x3F,
};
static int pix_sum_c(uint8_t * pix, int line_size)
{
int s, i, j;
s = 0;
for (i = 0; i < 16; i++) {
for (j = 0; j < 16; j += 8) {
s += pix[0];
s += pix[1];
s += pix[2];
s += pix[3];
s += pix[4];
s += pix[5];
s += pix[6];
s += pix[7];
pix += 8;
}
pix += line_size - 16;
}
return s;
}
static int pix_norm1_c(uint8_t * pix, int line_size)
s = 0;
for (i = 0; i < 16; i++) {
for (j = 0; j < 16; j += 8) {
Felix von Leitner
committed
#if 0
s += sq[pix[0]];
s += sq[pix[1]];
s += sq[pix[2]];
s += sq[pix[3]];
s += sq[pix[4]];
s += sq[pix[5]];
s += sq[pix[6]];
s += sq[pix[7]];
Felix von Leitner
committed
#else
#if LONG_MAX > 2147483647
register uint64_t x=*(uint64_t*)pix;
s += sq[x&0xff];
s += sq[(x>>8)&0xff];
s += sq[(x>>16)&0xff];
s += sq[(x>>24)&0xff];
s += sq[(x>>32)&0xff];
s += sq[(x>>40)&0xff];
s += sq[(x>>48)&0xff];
s += sq[(x>>56)&0xff];
#else
register uint32_t x=*(uint32_t*)pix;
s += sq[x&0xff];
s += sq[(x>>8)&0xff];
s += sq[(x>>16)&0xff];
s += sq[(x>>24)&0xff];
x=*(uint32_t*)(pix+4);
s += sq[x&0xff];
s += sq[(x>>8)&0xff];
s += sq[(x>>16)&0xff];
s += sq[(x>>24)&0xff];
#endif
#endif
pix += 8;
}
pix += line_size - 16;
}
return s;
}
static int sse8_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size)
s = 0;
for (i = 0; i < 8; i++) {
s += sq[pix1[0] - pix2[0]];
s += sq[pix1[1] - pix2[1]];
s += sq[pix1[2] - pix2[2]];
s += sq[pix1[3] - pix2[3]];
s += sq[pix1[4] - pix2[4]];
s += sq[pix1[5] - pix2[5]];
s += sq[pix1[6] - pix2[6]];
s += sq[pix1[7] - pix2[7]];
pix1 += line_size;
pix2 += line_size;
}
return s;
}
static int sse16_c(void *v, uint8_t *pix1, uint8_t *pix2, int line_size)
int s, i;
uint32_t *sq = squareTbl + 256;
s = 0;
for (i = 0; i < 16; i++) {
s += sq[pix1[ 0] - pix2[ 0]];
s += sq[pix1[ 1] - pix2[ 1]];
s += sq[pix1[ 2] - pix2[ 2]];
s += sq[pix1[ 3] - pix2[ 3]];
s += sq[pix1[ 4] - pix2[ 4]];
s += sq[pix1[ 5] - pix2[ 5]];
s += sq[pix1[ 6] - pix2[ 6]];
s += sq[pix1[ 7] - pix2[ 7]];
s += sq[pix1[ 8] - pix2[ 8]];
s += sq[pix1[ 9] - pix2[ 9]];
s += sq[pix1[10] - pix2[10]];
s += sq[pix1[11] - pix2[11]];
s += sq[pix1[12] - pix2[12]];
s += sq[pix1[13] - pix2[13]];
s += sq[pix1[14] - pix2[14]];
s += sq[pix1[15] - pix2[15]];
Felix von Leitner
committed
pix1 += line_size;
pix2 += line_size;
static void get_pixels_c(DCTELEM *restrict block, const uint8_t *pixels, int line_size)
{
int i;
/* read the pixels */
for(i=0;i<8;i++) {
block[0] = pixels[0];
block[1] = pixels[1];
block[2] = pixels[2];
block[3] = pixels[3];
block[4] = pixels[4];
block[5] = pixels[5];
block[6] = pixels[6];
block[7] = pixels[7];
pixels += line_size;
block += 8;
static void diff_pixels_c(DCTELEM *restrict block, const uint8_t *s1,
const uint8_t *s2, int stride){
int i;
/* read the pixels */
for(i=0;i<8;i++) {
block[0] = s1[0] - s2[0];
block[1] = s1[1] - s2[1];
block[2] = s1[2] - s2[2];
block[3] = s1[3] - s2[3];
block[4] = s1[4] - s2[4];
block[5] = s1[5] - s2[5];
block[6] = s1[6] - s2[6];
block[7] = s1[7] - s2[7];
s1 += stride;
s2 += stride;
static void put_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
Zdenek Kabelac
committed
int line_size)
uint8_t *cm = cropTbl + MAX_NEG_CROP;
pixels[0] = cm[block[0]];
pixels[1] = cm[block[1]];
pixels[2] = cm[block[2]];
pixels[3] = cm[block[3]];
pixels[4] = cm[block[4]];
pixels[5] = cm[block[5]];
pixels[6] = cm[block[6]];
pixels[7] = cm[block[7]];
pixels += line_size;
block += 8;
static void add_pixels_clamped_c(const DCTELEM *block, uint8_t *restrict pixels,
uint8_t *cm = cropTbl + MAX_NEG_CROP;
pixels[0] = cm[pixels[0] + block[0]];
pixels[1] = cm[pixels[1] + block[1]];
pixels[2] = cm[pixels[2] + block[2]];
pixels[3] = cm[pixels[3] + block[3]];
pixels[4] = cm[pixels[4] + block[4]];
pixels[5] = cm[pixels[5] + block[5]];
pixels[6] = cm[pixels[6] + block[6]];
pixels[7] = cm[pixels[7] + block[7]];
pixels += line_size;
block += 8;
#if 0
#define PIXOP2(OPNAME, OP) \
static void OPNAME ## _pixels(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
{\
int i;\
for(i=0; i<h; i++){\
OP(*((uint64_t*)block), LD64(pixels));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _no_rnd_pixels_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
{\
int i;\
for(i=0; i<h; i++){\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+1);\
OP(*((uint64_t*)block), (a&b) + (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _pixels_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
{\
int i;\
for(i=0; i<h; i++){\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+1);\
OP(*((uint64_t*)block), (a|b) - (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _no_rnd_pixels_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
{\
int i;\
for(i=0; i<h; i++){\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+line_size);\
OP(*((uint64_t*)block), (a&b) + (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _pixels_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
{\
int i;\
for(i=0; i<h; i++){\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+line_size);\
OP(*((uint64_t*)block), (a|b) - (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _pixels_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
{\
int i;\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+1);\
uint64_t l0= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL)\
+ 0x0202020202020202ULL;\
uint64_t h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
uint64_t l1,h1;\
\
pixels+=line_size;\
for(i=0; i<h; i+=2){\
uint64_t a= LD64(pixels );\
uint64_t b= LD64(pixels+1);\
l1= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL);\
h1= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
pixels+=line_size;\
block +=line_size;\
a= LD64(pixels );\
b= LD64(pixels+1);\
l0= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL)\
+ 0x0202020202020202ULL;\
h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static void OPNAME ## _no_rnd_pixels_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
{\
int i;\
const uint64_t a= LD64(pixels );\
const uint64_t b= LD64(pixels+1);\
uint64_t l0= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL)\
+ 0x0101010101010101ULL;\
uint64_t h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
uint64_t l1,h1;\
\
pixels+=line_size;\
for(i=0; i<h; i+=2){\
uint64_t a= LD64(pixels );\
uint64_t b= LD64(pixels+1);\
l1= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL);\
h1= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
pixels+=line_size;\
block +=line_size;\
a= LD64(pixels );\
b= LD64(pixels+1);\
l0= (a&0x0303030303030303ULL)\
+ (b&0x0303030303030303ULL)\
+ 0x0101010101010101ULL;\
h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
+ ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
CALL_2X_PIXELS(OPNAME ## _pixels16_c , OPNAME ## _pixels_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_x2_c , OPNAME ## _pixels_x2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_y2_c , OPNAME ## _pixels_y2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_xy2_c, OPNAME ## _pixels_xy2_c, 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_x2_c , OPNAME ## _no_rnd_pixels_x2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_y2_c , OPNAME ## _no_rnd_pixels_y2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_xy2_c, OPNAME ## _no_rnd_pixels_xy2_c, 8)
#define op_avg(a, b) a = ( ((a)|(b)) - ((((a)^(b))&0xFEFEFEFEFEFEFEFEULL)>>1) )
#else // 64 bit variant
#define PIXOP2(OPNAME, OP) \
static void OPNAME ## _pixels4_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
int i;\
for(i=0; i<h; i++){\
OP(*((uint32_t*)(block )), LD32(pixels ));\
pixels+=line_size;\
block +=line_size;\
}\
}\
static void OPNAME ## _pixels8_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
int i;\
for(i=0; i<h; i++){\
OP(*((uint32_t*)(block )), LD32(pixels ));\
OP(*((uint32_t*)(block+4)), LD32(pixels+4));\
pixels+=line_size;\
block +=line_size;\
}\
}\
static inline void OPNAME ## _no_rnd_pixels8_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
OPNAME ## _pixels8_c(block, pixels, line_size, h);\
static inline void OPNAME ## _no_rnd_pixels8_l2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
uint32_t a,b;\
a= LD32(&src1[i*src_stride1 ]);\
b= LD32(&src2[i*src_stride2 ]);\
OP(*((uint32_t*)&dst[i*dst_stride ]), (a&b) + (((a^b)&0xFEFEFEFEUL)>>1));\
a= LD32(&src1[i*src_stride1+4]);\
b= LD32(&src2[i*src_stride2+4]);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), (a&b) + (((a^b)&0xFEFEFEFEUL)>>1));\
static inline void OPNAME ## _pixels8_l2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
uint32_t a,b;\
a= LD32(&src1[i*src_stride1 ]);\
b= LD32(&src2[i*src_stride2 ]);\
OP(*((uint32_t*)&dst[i*dst_stride ]), (a|b) - (((a^b)&0xFEFEFEFEUL)>>1));\
a= LD32(&src1[i*src_stride1+4]);\
b= LD32(&src2[i*src_stride2+4]);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), (a|b) - (((a^b)&0xFEFEFEFEUL)>>1));\
static inline void OPNAME ## _pixels4_l2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
int i;\
for(i=0; i<h; i++){\
uint32_t a,b;\
a= LD32(&src1[i*src_stride1 ]);\
b= LD32(&src2[i*src_stride2 ]);\
OP(*((uint32_t*)&dst[i*dst_stride ]), (a|b) - (((a^b)&0xFEFEFEFEUL)>>1));\
}\
}\
\
static inline void OPNAME ## _pixels16_l2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
OPNAME ## _pixels8_l2(dst , src1 , src2 , dst_stride, src_stride1, src_stride2, h);\
OPNAME ## _pixels8_l2(dst+8, src1+8, src2+8, dst_stride, src_stride1, src_stride2, h);\
}\
\
static inline void OPNAME ## _no_rnd_pixels16_l2(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
OPNAME ## _no_rnd_pixels8_l2(dst , src1 , src2 , dst_stride, src_stride1, src_stride2, h);\
OPNAME ## _no_rnd_pixels8_l2(dst+8, src1+8, src2+8, dst_stride, src_stride1, src_stride2, h);\
}\
\
static inline void OPNAME ## _no_rnd_pixels8_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
OPNAME ## _no_rnd_pixels8_l2(block, pixels, pixels+1, line_size, line_size, line_size, h);\
}\
\
static inline void OPNAME ## _pixels8_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
OPNAME ## _pixels8_l2(block, pixels, pixels+1, line_size, line_size, line_size, h);\
}\
\
static inline void OPNAME ## _no_rnd_pixels8_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
OPNAME ## _no_rnd_pixels8_l2(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
}\
\
static inline void OPNAME ## _pixels8_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
OPNAME ## _pixels8_l2(block, pixels, pixels+line_size, line_size, line_size, line_size, h);\
}\
\
static inline void OPNAME ## _pixels8_l4(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,\
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
uint32_t a, b, c, d, l0, l1, h0, h1;\
a= LD32(&src1[i*src_stride1]);\
b= LD32(&src2[i*src_stride2]);\
c= LD32(&src3[i*src_stride3]);\
d= LD32(&src4[i*src_stride4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
a= LD32(&src1[i*src_stride1+4]);\
b= LD32(&src2[i*src_stride2+4]);\
c= LD32(&src3[i*src_stride3+4]);\
d= LD32(&src4[i*src_stride4+4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
static inline void OPNAME ## _no_rnd_pixels8_l4(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,\
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
uint32_t a, b, c, d, l0, l1, h0, h1;\
a= LD32(&src1[i*src_stride1]);\
b= LD32(&src2[i*src_stride2]);\
c= LD32(&src3[i*src_stride3]);\
d= LD32(&src4[i*src_stride4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
a= LD32(&src1[i*src_stride1+4]);\
b= LD32(&src2[i*src_stride2+4]);\
c= LD32(&src3[i*src_stride3+4]);\
d= LD32(&src4[i*src_stride4+4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
static inline void OPNAME ## _pixels16_l4(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,\
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
OPNAME ## _pixels8_l4(dst , src1 , src2 , src3 , src4 , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
OPNAME ## _pixels8_l4(dst+8, src1+8, src2+8, src3+8, src4+8, dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
}\
static inline void OPNAME ## _no_rnd_pixels16_l4(uint8_t *dst, const uint8_t *src1, uint8_t *src2, uint8_t *src3, uint8_t *src4,\
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\
OPNAME ## _no_rnd_pixels8_l4(dst , src1 , src2 , src3 , src4 , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
OPNAME ## _no_rnd_pixels8_l4(dst+8, src1+8, src2+8, src3+8, src4+8, dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
}\
static inline void OPNAME ## _pixels8_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
{\
int j;\
for(j=0; j<2; j++){\
int i;\
const uint32_t a= LD32(pixels );\
const uint32_t b= LD32(pixels+1);\
uint32_t l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
uint32_t l1,h1;\
\
pixels+=line_size;\
for(i=0; i<h; i+=2){\
uint32_t a= LD32(pixels );\
uint32_t b= LD32(pixels+1);\
l1= (a&0x03030303UL)\
+ (b&0x03030303UL);\
h1= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
a= LD32(pixels );\
b= LD32(pixels+1);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
}\
pixels+=4-line_size*(h+1);\
block +=4-line_size*h;\
}\
}\
\
static inline void OPNAME ## _no_rnd_pixels8_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
{\
int j;\
for(j=0; j<2; j++){\
int i;\
const uint32_t a= LD32(pixels );\
const uint32_t b= LD32(pixels+1);\
uint32_t l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
uint32_t l1,h1;\
\
pixels+=line_size;\
for(i=0; i<h; i+=2){\
uint32_t a= LD32(pixels );\
uint32_t b= LD32(pixels+1);\
l1= (a&0x03030303UL)\
+ (b&0x03030303UL);\
h1= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
a= LD32(pixels );\
b= LD32(pixels+1);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
}\
pixels+=4-line_size*(h+1);\
block +=4-line_size*h;\
}\
}\
\
CALL_2X_PIXELS(OPNAME ## _pixels16_c , OPNAME ## _pixels8_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_x2_c , OPNAME ## _pixels8_x2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_y2_c , OPNAME ## _pixels8_y2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _pixels16_xy2_c, OPNAME ## _pixels8_xy2_c, 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_c , OPNAME ## _pixels8_c , 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_x2_c , OPNAME ## _no_rnd_pixels8_x2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_y2_c , OPNAME ## _no_rnd_pixels8_y2_c , 8)\
CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_xy2_c, OPNAME ## _no_rnd_pixels8_xy2_c, 8)\
#define op_avg(a, b) a = ( ((a)|(b)) - ((((a)^(b))&0xFEFEFEFEUL)>>1) )
#endif
#define op_put(a, b) a = b
PIXOP2(avg, op_avg)
PIXOP2(put, op_put)
#undef op_avg
#undef op_put
#define avg2(a,b) ((a+b+1)>>1)
#define avg4(a,b,c,d) ((a+b+c+d+2)>>2)
static void gmc1_c(uint8_t *dst, uint8_t *src, int stride, int h, int x16, int y16, int rounder)
{
const int A=(16-x16)*(16-y16);
const int B=( x16)*(16-y16);
const int C=(16-x16)*( y16);
const int D=( x16)*( y16);
int i;
for(i=0; i<h; i++)
{
dst[0]= (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1] + rounder)>>8;
dst[1]= (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2] + rounder)>>8;
dst[2]= (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3] + rounder)>>8;
dst[3]= (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4] + rounder)>>8;
dst[4]= (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5] + rounder)>>8;
dst[5]= (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6] + rounder)>>8;
dst[6]= (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7] + rounder)>>8;
dst[7]= (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8] + rounder)>>8;
dst+= stride;
src+= stride;
static void gmc_c(uint8_t *dst, uint8_t *src, int stride, int h, int ox, int oy,
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
int dxx, int dxy, int dyx, int dyy, int shift, int r, int width, int height)
{
int y, vx, vy;
const int s= 1<<shift;
width--;
height--;
for(y=0; y<h; y++){
int x;
vx= ox;
vy= oy;
for(x=0; x<8; x++){ //XXX FIXME optimize
int src_x, src_y, frac_x, frac_y, index;
src_x= vx>>16;
src_y= vy>>16;
frac_x= src_x&(s-1);
frac_y= src_y&(s-1);
src_x>>=shift;
src_y>>=shift;
if((unsigned)src_x < width){
if((unsigned)src_y < height){
index= src_x + src_y*stride;
dst[y*stride + x]= ( ( src[index ]*(s-frac_x)
+ src[index +1]* frac_x )*(s-frac_y)
+ ( src[index+stride ]*(s-frac_x)
+ src[index+stride+1]* frac_x )* frac_y
+ r)>>(shift*2);
}else{
index= src_x + clip(src_y, 0, height)*stride;
dst[y*stride + x]= ( ( src[index ]*(s-frac_x)
+ src[index +1]* frac_x )*s
+ r)>>(shift*2);
}
}else{
if((unsigned)src_y < height){
index= clip(src_x, 0, width) + src_y*stride;
dst[y*stride + x]= ( ( src[index ]*(s-frac_y)
+ src[index+stride ]* frac_y )*s
+ r)>>(shift*2);
}else{
index= clip(src_x, 0, width) + clip(src_y, 0, height)*stride;
dst[y*stride + x]= src[index ];
}
}
vx+= dxx;
vy+= dyx;
}
ox += dxy;
oy += dyy;
}
}
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
#define H264_CHROMA_MC(OPNAME, OP)\
static void OPNAME ## h264_chroma_mc2_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){\
const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\
const int C=(8-x)*( y);\
const int D=( x)*( y);\
int i;\
\
assert(x<8 && y<8 && x>=0 && y>=0);\
\
for(i=0; i<h; i++)\
{\
OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
dst+= stride;\
src+= stride;\
}\
}\
\
static void OPNAME ## h264_chroma_mc4_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){\
const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\
const int C=(8-x)*( y);\
const int D=( x)*( y);\
int i;\
\
assert(x<8 && y<8 && x>=0 && y>=0);\
\
for(i=0; i<h; i++)\
{\
OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
dst+= stride;\
src+= stride;\
}\
}\
\
static void OPNAME ## h264_chroma_mc8_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){\
const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\
const int C=(8-x)*( y);\
const int D=( x)*( y);\
int i;\
\
assert(x<8 && y<8 && x>=0 && y>=0);\
\
for(i=0; i<h; i++)\
{\
OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
OP(dst[1], (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2]));\
OP(dst[2], (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3]));\
OP(dst[3], (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4]));\
OP(dst[4], (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5]));\
OP(dst[5], (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6]));\
OP(dst[6], (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7]));\
OP(dst[7], (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8]));\
dst+= stride;\
src+= stride;\
}\
}
#define op_avg(a, b) a = (((a)+(((b) + 32)>>6)+1)>>1)
#define op_put(a, b) a = (((b) + 32)>>6)
H264_CHROMA_MC(put_ , op_put)
H264_CHROMA_MC(avg_ , op_avg)
#undef op_avg
#undef op_put
static inline void copy_block4(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
{
int i;
for(i=0; i<h; i++)
{
ST32(dst , LD32(src ));
dst+=dstStride;
src+=srcStride;
}
}
static inline void copy_block8(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
{
int i;
for(i=0; i<h; i++)
{
ST32(dst , LD32(src ));
ST32(dst+4 , LD32(src+4 ));
dst+=dstStride;
src+=srcStride;
}
}
static inline void copy_block16(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
{
int i;
for(i=0; i<h; i++)
{
ST32(dst , LD32(src ));
ST32(dst+4 , LD32(src+4 ));
ST32(dst+8 , LD32(src+8 ));
ST32(dst+12, LD32(src+12));
dst+=dstStride;
src+=srcStride;
}
}
static inline void copy_block17(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
ST32(dst , LD32(src ));
ST32(dst+4 , LD32(src+4 ));
ST32(dst+8 , LD32(src+8 ));
ST32(dst+12, LD32(src+12));
dst[16]= src[16];
dst+=dstStride;
src+=srcStride;
}
}
static inline void copy_block9(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
ST32(dst , LD32(src ));
ST32(dst+4 , LD32(src+4 ));
dst[8]= src[8];
dst+=dstStride;
src+=srcStride;
}
}
static void OPNAME ## mpeg4_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){\
uint8_t *cm = cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<h; i++)\
{\
OP(dst[0], (src[0]+src[1])*20 - (src[0]+src[2])*6 + (src[1]+src[3])*3 - (src[2]+src[4]));\
OP(dst[1], (src[1]+src[2])*20 - (src[0]+src[3])*6 + (src[0]+src[4])*3 - (src[1]+src[5]));\
OP(dst[2], (src[2]+src[3])*20 - (src[1]+src[4])*6 + (src[0]+src[5])*3 - (src[0]+src[6]));\
OP(dst[3], (src[3]+src[4])*20 - (src[2]+src[5])*6 + (src[1]+src[6])*3 - (src[0]+src[7]));\
OP(dst[4], (src[4]+src[5])*20 - (src[3]+src[6])*6 + (src[2]+src[7])*3 - (src[1]+src[8]));\
OP(dst[5], (src[5]+src[6])*20 - (src[4]+src[7])*6 + (src[3]+src[8])*3 - (src[2]+src[8]));\
OP(dst[6], (src[6]+src[7])*20 - (src[5]+src[8])*6 + (src[4]+src[8])*3 - (src[3]+src[7]));\
OP(dst[7], (src[7]+src[8])*20 - (src[6]+src[8])*6 + (src[5]+src[7])*3 - (src[4]+src[6]));\
dst+=dstStride;\
src+=srcStride;\
}\
static void OPNAME ## mpeg4_qpel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
uint8_t *cm = cropTbl + MAX_NEG_CROP;\
int i;\
for(i=0; i<w; i++)\
{\
const int src0= src[0*srcStride];\
const int src1= src[1*srcStride];\
const int src2= src[2*srcStride];\
const int src3= src[3*srcStride];\
const int src4= src[4*srcStride];\
const int src5= src[5*srcStride];\
const int src6= src[6*srcStride];\
const int src7= src[7*srcStride];\
const int src8= src[8*srcStride];\
OP(dst[0*dstStride], (src0+src1)*20 - (src0+src2)*6 + (src1+src3)*3 - (src2+src4));\
OP(dst[1*dstStride], (src1+src2)*20 - (src0+src3)*6 + (src0+src4)*3 - (src1+src5));\
OP(dst[2*dstStride], (src2+src3)*20 - (src1+src4)*6 + (src0+src5)*3 - (src0+src6));\
OP(dst[3*dstStride], (src3+src4)*20 - (src2+src5)*6 + (src1+src6)*3 - (src0+src7));\
OP(dst[4*dstStride], (src4+src5)*20 - (src3+src6)*6 + (src2+src7)*3 - (src1+src8));\
OP(dst[5*dstStride], (src5+src6)*20 - (src4+src7)*6 + (src3+src8)*3 - (src2+src8));\
OP(dst[6*dstStride], (src6+src7)*20 - (src5+src8)*6 + (src4+src8)*3 - (src3+src7));\
OP(dst[7*dstStride], (src7+src8)*20 - (src6+src8)*6 + (src5+src7)*3 - (src4+src6));\