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
130
131
132
133
134
135
136
137
138
|
;*****************************************************************************
;* x86-optimized functions for volume filter
;* Copyright (c) 2012 Justin Ruggles <justin.ruggles@gmail.com>
;*
;* This file is part of Libav.
;*
;* Libav 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.
;*
;* Libav 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 Libav; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************
%include "libavutil/x86/x86util.asm"
SECTION_RODATA 32
pd_1_256: times 4 dq 0x3F70000000000000
pd_int32_max: times 4 dq 0x41DFFFFFFFC00000
pw_1: times 8 dw 1
pw_128: times 8 dw 128
pq_128: times 2 dq 128
SECTION_TEXT
;------------------------------------------------------------------------------
; void ff_scale_samples_s16(uint8_t *dst, const uint8_t *src, int len,
; int volume)
;------------------------------------------------------------------------------
INIT_XMM sse2
cglobal scale_samples_s16, 4,4,4, dst, src, len, volume
movd m0, volumem
pshuflw m0, m0, 0
punpcklwd m0, [pw_1]
mova m1, [pw_128]
lea lenq, [lend*2-mmsize]
.loop:
; dst[i] = av_clip_int16((src[i] * volume + 128) >> 8);
mova m2, [srcq+lenq]
punpcklwd m3, m2, m1
punpckhwd m2, m1
pmaddwd m3, m0
pmaddwd m2, m0
psrad m3, 8
psrad m2, 8
packssdw m3, m2
mova [dstq+lenq], m3
sub lenq, mmsize
jge .loop
REP_RET
;------------------------------------------------------------------------------
; void ff_scale_samples_s32(uint8_t *dst, const uint8_t *src, int len,
; int volume)
;------------------------------------------------------------------------------
%macro SCALE_SAMPLES_S32 0
cglobal scale_samples_s32, 4,4,4, dst, src, len, volume
%if ARCH_X86_32 && cpuflag(avx)
vbroadcastss xmm2, volumem
%else
movd xmm2, volumed
pshufd xmm2, xmm2, 0
%endif
CVTDQ2PD m2, xmm2
mulpd m2, m2, [pd_1_256]
mova m3, [pd_int32_max]
lea lenq, [lend*4-mmsize]
.loop:
CVTDQ2PD m0, [srcq+lenq ]
CVTDQ2PD m1, [srcq+lenq+mmsize/2]
mulpd m0, m0, m2
mulpd m1, m1, m2
minpd m0, m0, m3
minpd m1, m1, m3
cvtpd2dq xmm0, m0
cvtpd2dq xmm1, m1
%if cpuflag(avx)
vmovdqa [dstq+lenq ], xmm0
vmovdqa [dstq+lenq+mmsize/2], xmm1
%else
movq [dstq+lenq ], xmm0
movq [dstq+lenq+mmsize/2], xmm1
%endif
sub lenq, mmsize
jge .loop
REP_RET
%endmacro
INIT_XMM sse2
%define CVTDQ2PD cvtdq2pd
SCALE_SAMPLES_S32
%define CVTDQ2PD vcvtdq2pd
INIT_YMM avx
SCALE_SAMPLES_S32
%undef CVTDQ2PD
; NOTE: This is not bit-identical with the C version because it clips to
; [-INT_MAX, INT_MAX] instead of [INT_MIN, INT_MAX]
INIT_XMM ssse3, atom
cglobal scale_samples_s32, 4,4,8, dst, src, len, volume
movd m4, volumem
pshufd m4, m4, 0
mova m5, [pq_128]
pxor m6, m6
lea lenq, [lend*4-mmsize]
.loop:
; src[i] = av_clipl_int32((src[i] * volume + 128) >> 8);
mova m7, [srcq+lenq]
pabsd m3, m7
pshufd m0, m3, q0100
pshufd m1, m3, q0302
pmuludq m0, m4
pmuludq m1, m4
paddq m0, m5
paddq m1, m5
psrlq m0, 7
psrlq m1, 7
shufps m2, m0, m1, q3131
shufps m0, m0, m1, q2020
pcmpgtd m2, m6
por m0, m2
psrld m0, 1
psignd m0, m7
mova [dstq+lenq], m0
sub lenq, mmsize
jge .loop
REP_RET
|
