44cb64ee |
/*
* Copyright (C) 2004 the ffmpeg project
* |
b78e7197 |
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or |
44cb64ee |
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either |
b78e7197 |
* version 2.1 of the License, or (at your option) any later version. |
44cb64ee |
* |
b78e7197 |
* FFmpeg is distributed in the hope that it will be useful, |
44cb64ee |
* 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 |
b78e7197 |
* License along with FFmpeg; if not, write to the Free Software |
5509bffa |
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
44cb64ee |
*/
/** |
ba87f080 |
* @file |
115329f1 |
* Standard C DSP-oriented functions cribbed from the original VP3 |
44cb64ee |
* source code.
*/
#include "avcodec.h" |
e96682e6 |
#include "dsputil.h" |
44cb64ee |
#define IdctAdjustBeforeShift 8
#define xC1S7 64277
#define xC2S6 60547
#define xC3S5 54491
#define xC4S4 46341
#define xC5S3 36410
#define xC6S2 25080
#define xC7S1 12785
|
e003ea06 |
#define M(a,b) (((a) * (b))>>16)
|
849f1035 |
static av_always_inline void idct(uint8_t *dst, int stride, int16_t *input, int type) |
44cb64ee |
{ |
8b6103da |
int16_t *ip = input; |
44cb64ee |
|
d26a9ea0 |
int A, B, C, D, Ad, Bd, Cd, Dd, E, F, G, H;
int Ed, Gd, Add, Bdd, Fd, Hd; |
44cb64ee |
|
bf4e3bd2 |
int i; |
115329f1 |
|
44cb64ee |
/* Inverse DCT on the rows now */
for (i = 0; i < 8; i++) {
/* Check for non-zero values */
if ( ip[0] | ip[1] | ip[2] | ip[3] | ip[4] | ip[5] | ip[6] | ip[7] ) { |
d26a9ea0 |
A = M(xC1S7, ip[1]) + M(xC7S1, ip[7]);
B = M(xC7S1, ip[1]) - M(xC1S7, ip[7]);
C = M(xC3S5, ip[3]) + M(xC5S3, ip[5]);
D = M(xC3S5, ip[5]) - M(xC5S3, ip[3]); |
44cb64ee |
|
d26a9ea0 |
Ad = M(xC4S4, (A - C));
Bd = M(xC4S4, (B - D)); |
44cb64ee |
|
d26a9ea0 |
Cd = A + C;
Dd = B + D; |
44cb64ee |
|
d26a9ea0 |
E = M(xC4S4, (ip[0] + ip[4]));
F = M(xC4S4, (ip[0] - ip[4])); |
44cb64ee |
|
d26a9ea0 |
G = M(xC2S6, ip[2]) + M(xC6S2, ip[6]);
H = M(xC6S2, ip[2]) - M(xC2S6, ip[6]); |
44cb64ee |
|
d26a9ea0 |
Ed = E - G;
Gd = E + G; |
44cb64ee |
|
d26a9ea0 |
Add = F + Ad;
Bdd = Bd - H; |
44cb64ee |
|
d26a9ea0 |
Fd = F - Ad;
Hd = Bd + H; |
44cb64ee |
/* Final sequence of operations over-write original inputs. */ |
d26a9ea0 |
ip[0] = Gd + Cd ;
ip[7] = Gd - Cd ; |
44cb64ee |
|
d26a9ea0 |
ip[1] = Add + Hd;
ip[2] = Add - Hd; |
44cb64ee |
|
d26a9ea0 |
ip[3] = Ed + Dd ;
ip[4] = Ed - Dd ; |
44cb64ee |
|
d26a9ea0 |
ip[5] = Fd + Bdd;
ip[6] = Fd - Bdd; |
44cb64ee |
}
ip += 8; /* next row */
} |
115329f1 |
|
8b6103da |
ip = input; |
44cb64ee |
for ( i = 0; i < 8; i++) {
/* Check for non-zero values (bitwise or faster than ||) */ |
8b6103da |
if ( ip[1 * 8] | ip[2 * 8] | ip[3 * 8] | |
44cb64ee |
ip[4 * 8] | ip[5 * 8] | ip[6 * 8] | ip[7 * 8] ) {
|
d26a9ea0 |
A = M(xC1S7, ip[1*8]) + M(xC7S1, ip[7*8]);
B = M(xC7S1, ip[1*8]) - M(xC1S7, ip[7*8]);
C = M(xC3S5, ip[3*8]) + M(xC5S3, ip[5*8]);
D = M(xC3S5, ip[5*8]) - M(xC5S3, ip[3*8]); |
44cb64ee |
|
d26a9ea0 |
Ad = M(xC4S4, (A - C));
Bd = M(xC4S4, (B - D)); |
44cb64ee |
|
d26a9ea0 |
Cd = A + C;
Dd = B + D; |
44cb64ee |
|
d9cf32d9 |
E = M(xC4S4, (ip[0*8] + ip[4*8])) + 8;
F = M(xC4S4, (ip[0*8] - ip[4*8])) + 8;
if(type==1){ //HACK
E += 16*128;
F += 16*128;
} |
44cb64ee |
|
d26a9ea0 |
G = M(xC2S6, ip[2*8]) + M(xC6S2, ip[6*8]);
H = M(xC6S2, ip[2*8]) - M(xC2S6, ip[6*8]); |
44cb64ee |
|
d26a9ea0 |
Ed = E - G;
Gd = E + G; |
44cb64ee |
|
d26a9ea0 |
Add = F + Ad;
Bdd = Bd - H; |
44cb64ee |
|
d26a9ea0 |
Fd = F - Ad;
Hd = Bd + H; |
44cb64ee |
/* Final sequence of operations over-write original inputs. */ |
8b6103da |
if(type==0){ |
d26a9ea0 |
ip[0*8] = (Gd + Cd ) >> 4;
ip[7*8] = (Gd - Cd ) >> 4; |
115329f1 |
|
d26a9ea0 |
ip[1*8] = (Add + Hd ) >> 4;
ip[2*8] = (Add - Hd ) >> 4; |
115329f1 |
|
d26a9ea0 |
ip[3*8] = (Ed + Dd ) >> 4;
ip[4*8] = (Ed - Dd ) >> 4; |
115329f1 |
|
d26a9ea0 |
ip[5*8] = (Fd + Bdd ) >> 4;
ip[6*8] = (Fd - Bdd ) >> 4; |
8b6103da |
}else if(type==1){ |
c23acbae |
dst[0*stride] = av_clip_uint8((Gd + Cd ) >> 4);
dst[7*stride] = av_clip_uint8((Gd - Cd ) >> 4); |
115329f1 |
|
c23acbae |
dst[1*stride] = av_clip_uint8((Add + Hd ) >> 4);
dst[2*stride] = av_clip_uint8((Add - Hd ) >> 4); |
115329f1 |
|
c23acbae |
dst[3*stride] = av_clip_uint8((Ed + Dd ) >> 4);
dst[4*stride] = av_clip_uint8((Ed - Dd ) >> 4); |
115329f1 |
|
c23acbae |
dst[5*stride] = av_clip_uint8((Fd + Bdd ) >> 4);
dst[6*stride] = av_clip_uint8((Fd - Bdd ) >> 4); |
8b6103da |
}else{ |
c23acbae |
dst[0*stride] = av_clip_uint8(dst[0*stride] + ((Gd + Cd ) >> 4));
dst[7*stride] = av_clip_uint8(dst[7*stride] + ((Gd - Cd ) >> 4)); |
115329f1 |
|
c23acbae |
dst[1*stride] = av_clip_uint8(dst[1*stride] + ((Add + Hd ) >> 4));
dst[2*stride] = av_clip_uint8(dst[2*stride] + ((Add - Hd ) >> 4)); |
115329f1 |
|
c23acbae |
dst[3*stride] = av_clip_uint8(dst[3*stride] + ((Ed + Dd ) >> 4));
dst[4*stride] = av_clip_uint8(dst[4*stride] + ((Ed - Dd ) >> 4)); |
115329f1 |
|
c23acbae |
dst[5*stride] = av_clip_uint8(dst[5*stride] + ((Fd + Bdd ) >> 4));
dst[6*stride] = av_clip_uint8(dst[6*stride] + ((Fd - Bdd ) >> 4)); |
8b6103da |
} |
44cb64ee |
} else { |
8b6103da |
if(type==0){ |
115329f1 |
ip[0*8] =
ip[1*8] =
ip[2*8] =
ip[3*8] =
ip[4*8] =
ip[5*8] = |
8b6103da |
ip[6*8] =
ip[7*8] = ((xC4S4 * ip[0*8] + (IdctAdjustBeforeShift<<16))>>20);
}else if(type==1){
dst[0*stride]=
dst[1*stride]=
dst[2*stride]=
dst[3*stride]=
dst[4*stride]=
dst[5*stride]=
dst[6*stride]= |
c23acbae |
dst[7*stride]= av_clip_uint8(128 + ((xC4S4 * ip[0*8] + (IdctAdjustBeforeShift<<16))>>20)); |
8b6103da |
}else{
if(ip[0*8]){
int v= ((xC4S4 * ip[0*8] + (IdctAdjustBeforeShift<<16))>>20); |
c23acbae |
dst[0*stride] = av_clip_uint8(dst[0*stride] + v);
dst[1*stride] = av_clip_uint8(dst[1*stride] + v);
dst[2*stride] = av_clip_uint8(dst[2*stride] + v);
dst[3*stride] = av_clip_uint8(dst[3*stride] + v);
dst[4*stride] = av_clip_uint8(dst[4*stride] + v);
dst[5*stride] = av_clip_uint8(dst[5*stride] + v);
dst[6*stride] = av_clip_uint8(dst[6*stride] + v);
dst[7*stride] = av_clip_uint8(dst[7*stride] + v); |
8b6103da |
}
} |
44cb64ee |
}
ip++; /* next column */ |
8b6103da |
dst++; |
44cb64ee |
}
} |
8b6103da |
void ff_vp3_idct_c(DCTELEM *block/* align 16*/){
idct(NULL, 0, block, 0);
} |
115329f1 |
|
8b6103da |
void ff_vp3_idct_put_c(uint8_t *dest/*align 8*/, int line_size, DCTELEM *block/*align 16*/){
idct(dest, line_size, block, 1);
}
void ff_vp3_idct_add_c(uint8_t *dest/*align 8*/, int line_size, DCTELEM *block/*align 16*/){
idct(dest, line_size, block, 2);
} |
9971331d |
|
eb6a6cd7 |
void ff_vp3_idct_dc_add_c(uint8_t *dest/*align 8*/, int line_size, const DCTELEM *block/*align 16*/){ |
c0fda017 |
int i, dc = (block[0] + 15) >> 5; |
eb6a6cd7 |
for(i = 0; i < 8; i++){ |
c23acbae |
dest[0] = av_clip_uint8(dest[0] + dc);
dest[1] = av_clip_uint8(dest[1] + dc);
dest[2] = av_clip_uint8(dest[2] + dc);
dest[3] = av_clip_uint8(dest[3] + dc);
dest[4] = av_clip_uint8(dest[4] + dc);
dest[5] = av_clip_uint8(dest[5] + dc);
dest[6] = av_clip_uint8(dest[6] + dc);
dest[7] = av_clip_uint8(dest[7] + dc); |
eb6a6cd7 |
dest += line_size;
}
}
|
9971331d |
void ff_vp3_v_loop_filter_c(uint8_t *first_pixel, int stride, int *bounding_values)
{
unsigned char *end;
int filter_value;
const int nstride= -stride;
for (end= first_pixel + 8; first_pixel < end; first_pixel++) {
filter_value =
(first_pixel[2 * nstride] - first_pixel[ stride])
+3*(first_pixel[0 ] - first_pixel[nstride]);
filter_value = bounding_values[(filter_value + 4) >> 3];
first_pixel[nstride] = av_clip_uint8(first_pixel[nstride] + filter_value);
first_pixel[0] = av_clip_uint8(first_pixel[0] - filter_value);
}
}
void ff_vp3_h_loop_filter_c(uint8_t *first_pixel, int stride, int *bounding_values)
{
unsigned char *end;
int filter_value;
for (end= first_pixel + 8*stride; first_pixel != end; first_pixel += stride) {
filter_value =
(first_pixel[-2] - first_pixel[ 1])
+3*(first_pixel[ 0] - first_pixel[-1]);
filter_value = bounding_values[(filter_value + 4) >> 3];
first_pixel[-1] = av_clip_uint8(first_pixel[-1] + filter_value);
first_pixel[ 0] = av_clip_uint8(first_pixel[ 0] - filter_value);
}
} |