Mercurial > libavcodec.hg
comparison ppc/mpegaudiodec_altivec.c @ 12040:b4888704c11e libavcodec
Altivec-optimized mp{1,2,3} windowing
author | vitor |
---|---|
date | Thu, 01 Jul 2010 21:04:12 +0000 |
parents | |
children |
comparison
equal
deleted
inserted
replaced
12039:d5705b52b76e | 12040:b4888704c11e |
---|---|
1 /* | |
2 * Altivec optimized MP3 decoding functions | |
3 * Copyright (c) 2010 Vitor Sessak | |
4 * | |
5 * This file is part of FFmpeg. | |
6 * | |
7 * FFmpeg is free software; you can redistribute it and/or | |
8 * modify it under the terms of the GNU Lesser General Public | |
9 * License as published by the Free Software Foundation; either | |
10 * version 2.1 of the License, or (at your option) any later version. | |
11 * | |
12 * FFmpeg is distributed in the hope that it will be useful, | |
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
15 * Lesser General Public License for more details. | |
16 * | |
17 * You should have received a copy of the GNU Lesser General Public | |
18 * License along with FFmpeg; if not, write to the Free Software | |
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | |
20 */ | |
21 | |
22 #include "dsputil_altivec.h" | |
23 #include "util_altivec.h" | |
24 | |
25 #define CONFIG_FLOAT 1 | |
26 #include "libavcodec/mpegaudio.h" | |
27 | |
28 #define MACS(rt, ra, rb) rt+=(ra)*(rb) | |
29 #define MLSS(rt, ra, rb) rt-=(ra)*(rb) | |
30 | |
31 #define SUM8(op, sum, w, p) \ | |
32 { \ | |
33 op(sum, (w)[0 * 64], (p)[0 * 64]); \ | |
34 op(sum, (w)[1 * 64], (p)[1 * 64]); \ | |
35 op(sum, (w)[2 * 64], (p)[2 * 64]); \ | |
36 op(sum, (w)[3 * 64], (p)[3 * 64]); \ | |
37 op(sum, (w)[4 * 64], (p)[4 * 64]); \ | |
38 op(sum, (w)[5 * 64], (p)[5 * 64]); \ | |
39 op(sum, (w)[6 * 64], (p)[6 * 64]); \ | |
40 op(sum, (w)[7 * 64], (p)[7 * 64]); \ | |
41 } | |
42 | |
43 static void apply_window(const float *buf, const float *win1, | |
44 const float *win2, float *sum1, float *sum2, int len) | |
45 { | |
46 const vector float *win1a = (const vector float *) win1; | |
47 const vector float *win2a = (const vector float *) win2; | |
48 const vector float *bufa = (const vector float *) buf; | |
49 vector float *sum1a = (vector float *) sum1; | |
50 vector float *sum2a = (vector float *) sum2; | |
51 vector float av_uninit(v0), av_uninit(v4); | |
52 vector float v1, v2, v3; | |
53 | |
54 len = len >> 2; | |
55 | |
56 #define MULT(a, b) \ | |
57 { \ | |
58 v1 = vec_ld(a, win1a); \ | |
59 v2 = vec_ld(b, win2a); \ | |
60 v3 = vec_ld(a, bufa); \ | |
61 v0 = vec_madd(v3, v1, v0); \ | |
62 v4 = vec_madd(v2, v3, v4); \ | |
63 } | |
64 | |
65 while (len--) { | |
66 v0 = vec_xor(v0, v0); | |
67 v4 = vec_xor(v4, v4); | |
68 | |
69 MULT( 0, 0); | |
70 MULT( 256, 64); | |
71 MULT( 512, 128); | |
72 MULT( 768, 192); | |
73 MULT(1024, 256); | |
74 MULT(1280, 320); | |
75 MULT(1536, 384); | |
76 MULT(1792, 448); | |
77 | |
78 vec_st(v0, 0, sum1a); | |
79 vec_st(v4, 0, sum2a); | |
80 sum1a++; | |
81 sum2a++; | |
82 win1a++; | |
83 win2a++; | |
84 bufa++; | |
85 } | |
86 } | |
87 | |
88 static void apply_window_mp3(float *in, float *win, int *unused, float *out, | |
89 int incr) | |
90 { | |
91 LOCAL_ALIGNED_16(float, suma, [17]); | |
92 LOCAL_ALIGNED_16(float, sumb, [17]); | |
93 LOCAL_ALIGNED_16(float, sumc, [17]); | |
94 LOCAL_ALIGNED_16(float, sumd, [17]); | |
95 | |
96 float sum; | |
97 int j; | |
98 float *out2 = out + 32 * incr; | |
99 | |
100 /* copy to avoid wrap */ | |
101 memcpy(in + 512, in, 32 * sizeof(*in)); | |
102 | |
103 apply_window(in + 16, win , win + 512, suma, sumc, 16); | |
104 apply_window(in + 32, win + 48, win + 640, sumb, sumd, 16); | |
105 | |
106 SUM8(MLSS, suma[0], win + 32, in + 48); | |
107 | |
108 sumc[ 0] = 0; | |
109 sumb[16] = 0; | |
110 sumd[16] = 0; | |
111 | |
112 out[0 ] = suma[ 0]; | |
113 out += incr; | |
114 out2 -= incr; | |
115 for(j=1;j<16;j++) { | |
116 *out = suma[ j] - sumd[16-j]; | |
117 *out2 = -sumb[16-j] - sumc[ j]; | |
118 out += incr; | |
119 out2 -= incr; | |
120 } | |
121 | |
122 sum = 0; | |
123 SUM8(MLSS, sum, win + 16 + 32, in + 32); | |
124 *out = sum; | |
125 } | |
126 | |
127 void ff_mpegaudiodec_init_altivec(MPADecodeContext *s) | |
128 { | |
129 s->apply_window_mp3 = apply_window_mp3; | |
130 } |