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
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
|
/*
* Copyright (c) 2012
* MIPS Technologies, Inc., California.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the MIPS Technologies, Inc., nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* Author: Nedeljko Babic (nbabic@mips.com)
*
* various filters for ACELP-based codecs optimized for MIPS
*
* 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
*/
/**
* @file
* Reference: libavcodec/acelp_filters.c
*/
#include "config.h"
#include "libavutil/attributes.h"
#include "libavcodec/acelp_filters.h"
#include "libavutil/mips/asmdefs.h"
#if HAVE_INLINE_ASM
#if !HAVE_MIPS32R6 && !HAVE_MIPS64R6
static void ff_acelp_interpolatef_mips(float *out, const float *in,
const float *filter_coeffs, int precision,
int frac_pos, int filter_length, int length)
{
int n, i;
int prec = precision * 4;
int fc_offset = precision - frac_pos;
float in_val_p, in_val_m, fc_val_p, fc_val_m;
for (n = 0; n < length; n++) {
/**
* four pointers are defined in order to minimize number of
* computations done in inner loop
*/
const float *p_in_p = &in[n];
const float *p_in_m = &in[n-1];
const float *p_filter_coeffs_p = &filter_coeffs[frac_pos];
const float *p_filter_coeffs_m = filter_coeffs + fc_offset;
float v = 0;
for (i = 0; i < filter_length;i++) {
__asm__ volatile (
"lwc1 %[in_val_p], 0(%[p_in_p]) \n\t"
"lwc1 %[fc_val_p], 0(%[p_filter_coeffs_p]) \n\t"
"lwc1 %[in_val_m], 0(%[p_in_m]) \n\t"
"lwc1 %[fc_val_m], 0(%[p_filter_coeffs_m]) \n\t"
PTR_ADDIU "%[p_in_p], %[p_in_p], 4 \n\t"
"madd.s %[v],%[v], %[in_val_p],%[fc_val_p] \n\t"
PTR_ADDIU "%[p_in_m], %[p_in_m], -4 \n\t"
PTR_ADDU "%[p_filter_coeffs_p],%[p_filter_coeffs_p], %[prec] \n\t"
PTR_ADDU "%[p_filter_coeffs_m],%[p_filter_coeffs_m], %[prec] \n\t"
"madd.s %[v],%[v],%[in_val_m], %[fc_val_m] \n\t"
: [v] "+&f" (v),[p_in_p] "+r" (p_in_p), [p_in_m] "+r" (p_in_m),
[p_filter_coeffs_p] "+r" (p_filter_coeffs_p),
[in_val_p] "=&f" (in_val_p), [in_val_m] "=&f" (in_val_m),
[fc_val_p] "=&f" (fc_val_p), [fc_val_m] "=&f" (fc_val_m),
[p_filter_coeffs_m] "+r" (p_filter_coeffs_m)
: [prec] "r" (prec)
: "memory"
);
}
out[n] = v;
}
}
static void ff_acelp_apply_order_2_transfer_function_mips(float *out, const float *in,
const float zero_coeffs[2],
const float pole_coeffs[2],
float gain, float mem[2], int n)
{
/**
* loop is unrolled eight times
*/
__asm__ volatile (
"lwc1 $f0, 0(%[mem]) \n\t"
"blez %[n], ff_acelp_apply_order_2_transfer_function_end%= \n\t"
"lwc1 $f1, 4(%[mem]) \n\t"
"lwc1 $f2, 0(%[pole_coeffs]) \n\t"
"lwc1 $f3, 4(%[pole_coeffs]) \n\t"
"lwc1 $f4, 0(%[zero_coeffs]) \n\t"
"lwc1 $f5, 4(%[zero_coeffs]) \n\t"
"ff_acelp_apply_order_2_transfer_function_madd%=: \n\t"
"lwc1 $f6, 0(%[in]) \n\t"
"mul.s $f9, $f3, $f1 \n\t"
"mul.s $f7, $f2, $f0 \n\t"
"msub.s $f7, $f7, %[gain], $f6 \n\t"
"sub.s $f7, $f7, $f9 \n\t"
"madd.s $f8, $f7, $f4, $f0 \n\t"
"madd.s $f8, $f8, $f5, $f1 \n\t"
"lwc1 $f11, 4(%[in]) \n\t"
"mul.s $f12, $f3, $f0 \n\t"
"mul.s $f13, $f2, $f7 \n\t"
"msub.s $f13, $f13, %[gain], $f11 \n\t"
"sub.s $f13, $f13, $f12 \n\t"
"madd.s $f14, $f13, $f4, $f7 \n\t"
"madd.s $f14, $f14, $f5, $f0 \n\t"
"swc1 $f8, 0(%[out]) \n\t"
"lwc1 $f6, 8(%[in]) \n\t"
"mul.s $f9, $f3, $f7 \n\t"
"mul.s $f15, $f2, $f13 \n\t"
"msub.s $f15, $f15, %[gain], $f6 \n\t"
"sub.s $f15, $f15, $f9 \n\t"
"madd.s $f8, $f15, $f4, $f13 \n\t"
"madd.s $f8, $f8, $f5, $f7 \n\t"
"swc1 $f14, 4(%[out]) \n\t"
"lwc1 $f11, 12(%[in]) \n\t"
"mul.s $f12, $f3, $f13 \n\t"
"mul.s $f16, $f2, $f15 \n\t"
"msub.s $f16, $f16, %[gain], $f11 \n\t"
"sub.s $f16, $f16, $f12 \n\t"
"madd.s $f14, $f16, $f4, $f15 \n\t"
"madd.s $f14, $f14, $f5, $f13 \n\t"
"swc1 $f8, 8(%[out]) \n\t"
"lwc1 $f6, 16(%[in]) \n\t"
"mul.s $f9, $f3, $f15 \n\t"
"mul.s $f7, $f2, $f16 \n\t"
"msub.s $f7, $f7, %[gain], $f6 \n\t"
"sub.s $f7, $f7, $f9 \n\t"
"madd.s $f8, $f7, $f4, $f16 \n\t"
"madd.s $f8, $f8, $f5, $f15 \n\t"
"swc1 $f14, 12(%[out]) \n\t"
"lwc1 $f11, 20(%[in]) \n\t"
"mul.s $f12, $f3, $f16 \n\t"
"mul.s $f13, $f2, $f7 \n\t"
"msub.s $f13, $f13, %[gain], $f11 \n\t"
"sub.s $f13, $f13, $f12 \n\t"
"madd.s $f14, $f13, $f4, $f7 \n\t"
"madd.s $f14, $f14, $f5, $f16 \n\t"
"swc1 $f8, 16(%[out]) \n\t"
"lwc1 $f6, 24(%[in]) \n\t"
"mul.s $f9, $f3, $f7 \n\t"
"mul.s $f15, $f2, $f13 \n\t"
"msub.s $f15, $f15, %[gain], $f6 \n\t"
"sub.s $f1, $f15, $f9 \n\t"
"madd.s $f8, $f1, $f4, $f13 \n\t"
"madd.s $f8, $f8, $f5, $f7 \n\t"
"swc1 $f14, 20(%[out]) \n\t"
"lwc1 $f11, 28(%[in]) \n\t"
"mul.s $f12, $f3, $f13 \n\t"
"mul.s $f16, $f2, $f1 \n\t"
"msub.s $f16, $f16, %[gain], $f11 \n\t"
"sub.s $f0, $f16, $f12 \n\t"
"madd.s $f14, $f0, $f4, $f1 \n\t"
"madd.s $f14, $f14, $f5, $f13 \n\t"
"swc1 $f8, 24(%[out]) \n\t"
PTR_ADDIU "%[out], 32 \n\t"
PTR_ADDIU "%[in], 32 \n\t"
"addiu %[n], -8 \n\t"
"swc1 $f14, -4(%[out]) \n\t"
"bnez %[n], ff_acelp_apply_order_2_transfer_function_madd%= \n\t"
"swc1 $f1, 4(%[mem]) \n\t"
"swc1 $f0, 0(%[mem]) \n\t"
"ff_acelp_apply_order_2_transfer_function_end%=: \n\t"
: [out] "+r" (out),
[in] "+r" (in), [gain] "+f" (gain),
[n] "+r" (n), [mem] "+r" (mem)
: [zero_coeffs] "r" (zero_coeffs),
[pole_coeffs] "r" (pole_coeffs)
: "$f0", "$f1", "$f2", "$f3", "$f4", "$f5",
"$f6", "$f7", "$f8", "$f9", "$f10", "$f11",
"$f12", "$f13", "$f14", "$f15", "$f16", "memory"
);
}
#endif /* !HAVE_MIPS32R6 && !HAVE_MIPS64R6 */
#endif /* HAVE_INLINE_ASM */
void ff_acelp_filter_init_mips(ACELPFContext *c)
{
#if HAVE_INLINE_ASM
#if !HAVE_MIPS32R6 && !HAVE_MIPS64R6
c->acelp_interpolatef = ff_acelp_interpolatef_mips;
c->acelp_apply_order_2_transfer_function = ff_acelp_apply_order_2_transfer_function_mips;
#endif
#endif
}
|
