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
|
;******************************************************************************
;* SSE-optimized functions for the DCA decoder
;* Copyright (C) 2012-2014 Christophe Gisquet <christophe.gisquet@gmail.com>
;*
;* 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 "libavutil/x86/x86util.asm"
SECTION_RODATA
pf_inv16: times 4 dd 0x3D800000 ; 1/16
SECTION .text
; %1=v0/v1 %2=in1 %3=in2
%macro FIR_LOOP 2-3
.loop%1:
%define va m1
%define vb m2
%if %1
%define OFFSET 0
%else
%define OFFSET NUM_COEF*count
%endif
; for v0, incrementing and for v1, decrementing
mova va, [cf0q + OFFSET]
mova vb, [cf0q + OFFSET + 4*NUM_COEF]
%if %0 == 3
mova m4, [cf0q + OFFSET + mmsize]
mova m0, [cf0q + OFFSET + 4*NUM_COEF + mmsize]
%endif
mulps va, %2
mulps vb, %2
%if %0 == 3
%if cpuflag(fma3)
fmaddps va, m4, %3, va
fmaddps vb, m0, %3, vb
%else
mulps m4, %3
mulps m0, %3
addps va, m4
addps vb, m0
%endif
%endif
; va = va1 va2 va3 va4
; vb = vb1 vb2 vb3 vb4
%if %1
SWAP va, vb
%endif
mova m4, va
unpcklps va, vb ; va3 vb3 va4 vb4
unpckhps m4, vb ; va1 vb1 va2 vb2
addps m4, va ; va1+3 vb1+3 va2+4 vb2+4
movhlps vb, m4 ; va1+3 vb1+3
addps vb, m4 ; va0..4 vb0..4
movlps [outq + count], vb
%if %1
sub cf0q, 8*NUM_COEF
%endif
add count, 8
jl .loop%1
%endmacro
; void dca_lfe_fir(float *out, float *in, float *coefs)
%macro DCA_LFE_FIR 1
cglobal dca_lfe_fir%1, 3,3,6-%1, out, in, cf0
%define IN1 m3
%define IN2 m5
%define count inq
%define NUM_COEF 4*(2-%1)
%define NUM_OUT 32*(%1+1)
movu IN1, [inq + 4 - 1*mmsize]
shufps IN1, IN1, q0123
%if %1 == 0
movu IN2, [inq + 4 - 2*mmsize]
shufps IN2, IN2, q0123
%endif
mov count, -4*NUM_OUT
add cf0q, 4*NUM_COEF*NUM_OUT
add outq, 4*NUM_OUT
; compute v0 first
%if %1 == 0
FIR_LOOP 0, IN1, IN2
%else
FIR_LOOP 0, IN1
%endif
shufps IN1, IN1, q0123
mov count, -4*NUM_OUT
; cf1 already correctly positioned
add outq, 4*NUM_OUT ; outq now at out2
sub cf0q, 8*NUM_COEF
%if %1 == 0
shufps IN2, IN2, q0123
FIR_LOOP 1, IN2, IN1
%else
FIR_LOOP 1, IN1
%endif
RET
%endmacro
INIT_XMM sse
DCA_LFE_FIR 0
DCA_LFE_FIR 1
%if HAVE_FMA3_EXTERNAL
INIT_XMM fma3
DCA_LFE_FIR 0
%endif
|
