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
|
;*****************************************************************************
;* x86-optimized functions for afir filter
;* Copyright (c) 2017 Paul B Mahol
;*
;* 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 .text
;------------------------------------------------------------------------------
; void ff_fcmul_add(float *sum, const float *t, const float *c, int len)
;------------------------------------------------------------------------------
INIT_XMM sse3
cglobal fcmul_add, 4,4,6, sum, t, c, len
shl lend, 3
add tq, lenq
add cq, lenq
add sumq, lenq
neg lenq
ALIGN 16
.loop:
movsldup m0, [tq + lenq]
movsldup m3, [tq + lenq+mmsize]
movaps m1, [cq + lenq]
movaps m4, [cq + lenq+mmsize]
mulps m0, m1
mulps m3, m4
shufps m1, m1, 0xb1
shufps m4, m4, 0xb1
movshdup m2, [tq + lenq]
movshdup m5, [tq + lenq+mmsize]
mulps m2, m1
mulps m5, m4
addsubps m0, m2
addsubps m3, m5
addps m0, [sumq + lenq]
addps m3, [sumq + lenq+mmsize]
movaps [sumq + lenq], m0
movaps [sumq + lenq+mmsize], m3
add lenq, mmsize*2
jl .loop
movss xm0, [tq + lenq]
mulss xm0, [cq + lenq]
addss xm0, [sumq + lenq]
movss [sumq + lenq], xm0
RET
|
