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
|
/*
* SIMD-optimized IDCT functions for HEVC decoding
* Copyright (c) Alexandra Hajkova
*
* This file is part of FFmpeg.
*
* FFmpeg 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 2.1 of the License, or (at your option) any later version.
*
* FFmpeg 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 FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "config.h"
#include "libavutil/attributes.h"
#include "libavutil/cpu.h"
#include "libavutil/ppc/cpu.h"
#include "libavutil/ppc/util_altivec.h"
#include "libavcodec/hevcdsp.h"
#if HAVE_ALTIVEC
static const vec_s16 trans4[4] = {
{ 64, 64, 64, 64, 64, 64, 64, 64 },
{ 83, 36, 83, 36, 83, 36, 83, 36 },
{ 64, -64, 64, -64, 64, -64, 64, -64 },
{ 36, -83, 36, -83, 36, -83, 36, -83 },
};
static const vec_u8 mask[2] = {
{ 0x00, 0x01, 0x08, 0x09, 0x10, 0x11, 0x18, 0x19, 0x02, 0x03, 0x0A, 0x0B, 0x12, 0x13, 0x1A, 0x1B },
{ 0x04, 0x05, 0x0C, 0x0D, 0x14, 0x15, 0x1C, 0x1D, 0x06, 0x07, 0x0E, 0x0F, 0x16, 0x17, 0x1E, 0x1F },
};
static av_always_inline void transform4x4(vec_s16 src_01, vec_s16 src_23,
vec_s32 res[4], const int shift,
int16_t *coeffs)
{
vec_s16 src_02, src_13;
vec_s32 zero = vec_splat_s32(0);
vec_s32 e0, o0, e1, o1;
vec_s32 add;
src_13 = vec_mergel(src_01, src_23);
src_02 = vec_mergeh(src_01, src_23);
e0 = vec_msums(src_02, trans4[0], zero);
o0 = vec_msums(src_13, trans4[1], zero);
e1 = vec_msums(src_02, trans4[2], zero);
o1 = vec_msums(src_13, trans4[3], zero);
add = vec_sl(vec_splat_s32(1), vec_splat_u32(shift - 1));
e0 = vec_add(e0, add);
e1 = vec_add(e1, add);
res[0] = vec_add(e0, o0);
res[1] = vec_add(e1, o1);
res[2] = vec_sub(e1, o1);
res[3] = vec_sub(e0, o0);
}
static av_always_inline void scale(vec_s32 res[4], vec_s16 res_packed[2],
const int shift)
{
int i;
vec_u32 v_shift = vec_splat_u32(shift);
for (i = 0; i < 4; i++)
res[i] = vec_sra(res[i], v_shift);
// clip16
res_packed[0] = vec_packs(res[0], res[1]);
res_packed[1] = vec_packs(res[2], res[3]);
}
#define FUNCDECL(a, depth) a ## _ ## depth ## _altivec
#define FUNC(a, b) FUNCDECL(a, b)
#define BIT_DEPTH 8
#include "hevcdsp_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 10
#include "hevcdsp_template.c"
#undef BIT_DEPTH
#endif /* HAVE_ALTIVEC */
av_cold void ff_hevc_dsp_init_ppc(HEVCDSPContext *c, const int bit_depth)
{
#if HAVE_ALTIVEC
if (!PPC_ALTIVEC(av_get_cpu_flags()))
return;
if (bit_depth == 8)
c->idct[0] = ff_hevc_idct_4x4_8_altivec;
if (bit_depth == 10)
c->idct[0] = ff_hevc_idct_4x4_10_altivec;
#endif /* HAVE_ALTIVEC */
}
|
