1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
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
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
|
/*
* MMI optimized DSP utils
* Copyright (c) 2000, 2001 Fabrice Bellard.
*
* 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
*
* MMI optimization by Leon van Stuivenberg <leonvs@iae.nl>
*/
#include "../dsputil.h"
#include "mmi.h"
void ff_mmi_idct_put(uint8_t *dest, int line_size, DCTELEM *block);
void ff_mmi_idct_add(uint8_t *dest, int line_size, DCTELEM *block);
static void clear_blocks_mmi(DCTELEM * blocks)
{
int i;
for (i = 0; i < 6; i++) {
asm volatile(
"sq $0, 0(%0) \n\t"
"sq $0, 16(%0) \n\t"
"sq $0, 32(%0) \n\t"
"sq $0, 48(%0) \n\t"
"sq $0, 64(%0) \n\t"
"sq $0, 80(%0) \n\t"
"sq $0, 96(%0) \n\t"
"sq $0, 112(%0) \n\t" :: "r" (blocks) : "memory" );
blocks += 64;
}
}
static void get_pixels_mmi(DCTELEM *block, const uint8_t *pixels, int line_size)
{
int i;
for(i=0;i<8;i++) {
asm volatile(
".set push \n\t"
".set mips3 \n\t"
"ld $8, 0(%1) \n\t"
"add %1, %1, %2 \n\t"
"pextlb $8, $0, $8 \n\t"
"sq $8, 0(%0) \n\t"
".set pop \n\t"
:: "r" (block), "r" (pixels), "r" (line_size) : "$8", "memory" );
block += 8;
}
}
static void put_pixels8_mmi(uint8_t *block, const uint8_t *pixels, int line_size, int h)
{
int i;
for(i=0; i<h; i++) {
asm volatile(
".set push \n\t"
".set mips3 \n\t"
"ldr $8, 0(%1) \n\t"
"ldl $8, 7(%1) \n\t"
"add %1, %1, %2 \n\t"
"sd $8, 0(%0) \n\t"
"add %0, %0, %2 \n\t"
".set pop \n\t"
:: "r" (block), "r" (pixels), "r" (line_size) : "$8", "memory" );
}
}
static void put_pixels16_mmi(uint8_t *block, const uint8_t *pixels, int line_size, int h)
{
int i;
for(i=0; i<(h>>2); i++) {
asm volatile (
".set push \n\t"
".set mips3 \n\t"
#define PUTPIX16 \
"ldr $8, 0(%1) \n\t" \
"ldl $8, 7(%1) \n\t" \
"ldr $9, 8(%1) \n\t" \
"ldl $9, 15(%1) \n\t" \
"add %1, %1, %2 \n\t" \
"pcpyld $8, $9, $8 \n\t" \
"sq $8, 0(%0) \n\t" \
"add %0, %0, %2 \n\t"
PUTPIX16
PUTPIX16
PUTPIX16
PUTPIX16
".set pop \n\t"
:: "r" (block), "r" (pixels), "r" (line_size) : "$8", "$9", "memory" );
}
}
void dsputil_init_mmi(DSPContext* c, AVCodecContext *avctx)
{
const int idct_algo= avctx->idct_algo;
c->clear_blocks = clear_blocks_mmi;
c->put_pixels_tab[1][0] = put_pixels8_mmi;
c->put_no_rnd_pixels_tab[1][0] = put_pixels8_mmi;
c->put_pixels_tab[0][0] = put_pixels16_mmi;
c->put_no_rnd_pixels_tab[0][0] = put_pixels16_mmi;
c->get_pixels = get_pixels_mmi;
if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_PS2){
c->idct_put= ff_mmi_idct_put;
c->idct_add= ff_mmi_idct_add;
c->idct_permutation_type= FF_LIBMPEG2_IDCT_PERM;
}
}
|
