| rev |
line source |
|
nengel@2
|
1 /*
|
|
nengel@2
|
2 * Copyright (c) 2010 Mans Rullgard <mans@mansr.com>
|
|
nengel@2
|
3 *
|
|
nengel@2
|
4 * This file is part of FFmpeg.
|
|
nengel@2
|
5 *
|
|
nengel@2
|
6 * FFmpeg is free software; you can redistribute it and/or
|
|
nengel@2
|
7 * modify it under the terms of the GNU Lesser General Public
|
|
nengel@2
|
8 * License as published by the Free Software Foundation; either
|
|
nengel@2
|
9 * version 2.1 of the License, or (at your option) any later version.
|
|
nengel@2
|
10 *
|
|
nengel@2
|
11 * FFmpeg is distributed in the hope that it will be useful,
|
|
nengel@2
|
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
nengel@2
|
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
nengel@2
|
14 * Lesser General Public License for more details.
|
|
nengel@2
|
15 *
|
|
nengel@2
|
16 * You should have received a copy of the GNU Lesser General Public
|
|
nengel@2
|
17 * License along with FFmpeg; if not, write to the Free Software
|
|
nengel@2
|
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
nengel@2
|
19 */
|
|
nengel@2
|
20
|
|
nengel@2
|
21 #include "asm.S"
|
|
nengel@2
|
22
|
|
nengel@2
|
23 function ff_dca_lfe_fir_neon, export=1
|
|
nengel@2
|
24 push {r4-r6,lr}
|
|
nengel@2
|
25
|
|
nengel@2
|
26 add r4, r0, r3, lsl #2 @ out2
|
|
nengel@2
|
27 add r5, r2, #256*4-16 @ cf1
|
|
nengel@2
|
28 sub r1, r1, #12
|
|
nengel@2
|
29 cmp r3, #32
|
|
nengel@2
|
30 moveq r6, #256/32
|
|
nengel@2
|
31 movne r6, #256/64
|
|
nengel@2
|
32 NOVFP vldr d0, [sp, #16] @ scale, bias
|
|
nengel@2
|
33 mov lr, #-16
|
|
nengel@2
|
34 1:
|
|
nengel@2
|
35 vmov.f32 q2, #0.0 @ v0
|
|
nengel@2
|
36 vmov.f32 q3, #0.0 @ v1
|
|
nengel@2
|
37 mov r12, r6
|
|
nengel@2
|
38 2:
|
|
nengel@2
|
39 vld1.32 {q8}, [r2,:128]! @ cf0
|
|
nengel@2
|
40 vld1.32 {q9}, [r5,:128], lr @ cf1
|
|
nengel@2
|
41 vld1.32 {q1}, [r1], lr @ in
|
|
nengel@2
|
42 subs r12, r12, #4
|
|
nengel@2
|
43 vrev64.32 q10, q8
|
|
nengel@2
|
44 vmla.f32 q3, q1, q9
|
|
nengel@2
|
45 vmla.f32 d4, d2, d21
|
|
nengel@2
|
46 vmla.f32 d5, d3, d20
|
|
nengel@2
|
47 bne 2b
|
|
nengel@2
|
48
|
|
nengel@2
|
49 add r1, r1, r6, lsl #2
|
|
nengel@2
|
50 subs r3, r3, #1
|
|
nengel@2
|
51 vadd.f32 d4, d4, d5
|
|
nengel@2
|
52 vadd.f32 d6, d6, d7
|
|
nengel@2
|
53 vpadd.f32 d4, d4, d6
|
|
nengel@2
|
54 vdup.32 d5, d0[1]
|
|
nengel@2
|
55 vmla.f32 d5, d4, d0[0]
|
|
nengel@2
|
56 vst1.32 {d5[0]}, [r0,:32]!
|
|
nengel@2
|
57 vst1.32 {d5[1]}, [r4,:32]!
|
|
nengel@2
|
58 bne 1b
|
|
nengel@2
|
59
|
|
nengel@2
|
60 pop {r4-r6,pc}
|
|
nengel@2
|
61 endfunc
|
