Commit | Line | Data |
---|---|---|
2ba45a60 DM |
1 | /* |
2 | * Copyright (c) 2012 | |
3 | * MIPS Technologies, Inc., California. | |
4 | * | |
5 | * Redistribution and use in source and binary forms, with or without | |
6 | * modification, are permitted provided that the following conditions | |
7 | * are met: | |
8 | * 1. Redistributions of source code must retain the above copyright | |
9 | * notice, this list of conditions and the following disclaimer. | |
10 | * 2. Redistributions in binary form must reproduce the above copyright | |
11 | * notice, this list of conditions and the following disclaimer in the | |
12 | * documentation and/or other materials provided with the distribution. | |
13 | * 3. Neither the name of the MIPS Technologies, Inc., nor the names of its | |
14 | * contributors may be used to endorse or promote products derived from | |
15 | * this software without specific prior written permission. | |
16 | * | |
17 | * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND | |
18 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE | |
19 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE | |
20 | * ARE DISCLAIMED. IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE | |
21 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL | |
22 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS | |
23 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) | |
24 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT | |
25 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY | |
26 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF | |
27 | * SUCH DAMAGE. | |
28 | * | |
29 | * Author: Bojan Zivkovic (bojan@mips.com) | |
30 | * | |
31 | * Compute antialias function optimised for MIPS fixed-point architecture | |
32 | * | |
33 | * This file is part of FFmpeg. | |
34 | * | |
35 | * FFmpeg is free software; you can redistribute it and/or | |
36 | * modify it under the terms of the GNU Lesser General Public | |
37 | * License as published by the Free Software Foundation; either | |
38 | * version 2.1 of the License, or (at your option) any later version. | |
39 | * | |
40 | * FFmpeg is distributed in the hope that it will be useful, | |
41 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
42 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
43 | * Lesser General Public License for more details. | |
44 | * | |
45 | * You should have received a copy of the GNU Lesser General Public | |
46 | * License along with FFmpeg; if not, write to the Free Software | |
47 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA | |
48 | */ | |
49 | ||
50 | /** | |
51 | * @file | |
52 | * Reference: libavcodec/mpegaudiodec.c | |
53 | */ | |
54 | ||
55 | #ifndef AVCODEC_MIPS_COMPUTE_ANTIALIAS_FIXED_H | |
56 | #define AVCODEC_MIPS_COMPUTE_ANTIALIAS_FIXED_H | |
57 | ||
58 | #if HAVE_INLINE_ASM | |
59 | static void compute_antialias_mips_fixed(MPADecodeContext *s, | |
60 | GranuleDef *g) | |
61 | { | |
62 | int32_t *ptr, *csa; | |
63 | int n, i; | |
64 | int MAX_lo = 0xffffffff; | |
65 | ||
66 | /* we antialias only "long" bands */ | |
67 | if (g->block_type == 2) { | |
68 | if (!g->switch_point) | |
69 | return; | |
70 | /* XXX: check this for 8000Hz case */ | |
71 | n = 1; | |
72 | } else { | |
73 | n = SBLIMIT - 1; | |
74 | } | |
75 | ||
76 | ||
77 | ptr = g->sb_hybrid + 18; | |
78 | ||
79 | for(i = n;i > 0;i--) { | |
80 | int tmp0, tmp1, tmp2, tmp00, tmp11; | |
81 | int temp_reg1, temp_reg2, temp_reg3, temp_reg4, temp_reg5, temp_reg6; | |
82 | csa = &csa_table[0][0]; | |
83 | ||
84 | /** | |
85 | * instructions are scheduled to minimize pipeline stall. | |
86 | */ | |
87 | __asm__ volatile ( | |
88 | "lw %[tmp0], -1*4(%[ptr]) \n\t" | |
89 | "lw %[tmp1], 0*4(%[ptr]) \n\t" | |
90 | "lw %[temp_reg1], 0*4(%[csa]) \n\t" | |
91 | "lw %[temp_reg2], 2*4(%[csa]) \n\t" | |
92 | "add %[tmp2], %[tmp0], %[tmp1] \n\t" | |
93 | "lw %[temp_reg3], 3*4(%[csa]) \n\t" | |
94 | "mult $ac0, %[tmp2], %[temp_reg1] \n\t" | |
95 | "mult $ac1, %[tmp2], %[temp_reg1] \n\t" | |
96 | "lw %[tmp00], -2*4(%[ptr]) \n\t" | |
97 | "lw %[tmp11], 1*4(%[ptr]) \n\t" | |
98 | "lw %[temp_reg4], 4*4(%[csa]) \n\t" | |
99 | "mtlo %[MAX_lo], $ac0 \n\t" | |
100 | "mtlo $zero, $ac1 \n\t" | |
101 | "msub $ac0, %[tmp1], %[temp_reg2] \n\t" | |
102 | "madd $ac1, %[tmp0], %[temp_reg3] \n\t" | |
103 | "add %[tmp2], %[tmp00], %[tmp11] \n\t" | |
104 | "lw %[temp_reg5], 6*4(%[csa]) \n\t" | |
105 | "mult $ac2, %[tmp2], %[temp_reg4] \n\t" | |
106 | "mult $ac3, %[tmp2], %[temp_reg4] \n\t" | |
107 | "mfhi %[temp_reg1], $ac0 \n\t" | |
108 | "mfhi %[temp_reg2], $ac1 \n\t" | |
109 | "lw %[temp_reg6], 7*4(%[csa]) \n\t" | |
110 | "mtlo %[MAX_lo], $ac2 \n\t" | |
111 | "msub $ac2, %[tmp11], %[temp_reg5] \n\t" | |
112 | "mtlo $zero, $ac3 \n\t" | |
113 | "madd $ac3, %[tmp00], %[temp_reg6] \n\t" | |
114 | "sll %[temp_reg1], %[temp_reg1], 2 \n\t" | |
115 | "sw %[temp_reg1], -1*4(%[ptr]) \n\t" | |
116 | "mfhi %[temp_reg4], $ac2 \n\t" | |
117 | "sll %[temp_reg2], %[temp_reg2], 2 \n\t" | |
118 | "mfhi %[temp_reg5], $ac3 \n\t" | |
119 | "sw %[temp_reg2], 0*4(%[ptr]) \n\t" | |
120 | "lw %[tmp0], -3*4(%[ptr]) \n\t" | |
121 | "lw %[tmp1], 2*4(%[ptr]) \n\t" | |
122 | "lw %[temp_reg1], 8*4(%[csa]) \n\t" | |
123 | "sll %[temp_reg4], %[temp_reg4], 2 \n\t" | |
124 | "add %[tmp2], %[tmp0], %[tmp1] \n\t" | |
125 | "sll %[temp_reg5], %[temp_reg5], 2 \n\t" | |
126 | "mult $ac0, %[tmp2], %[temp_reg1] \n\t" | |
127 | "mult $ac1, %[tmp2], %[temp_reg1] \n\t" | |
128 | "sw %[temp_reg4], -2*4(%[ptr]) \n\t" | |
129 | "sw %[temp_reg5], 1*4(%[ptr]) \n\t" | |
130 | "lw %[temp_reg2], 10*4(%[csa]) \n\t" | |
131 | "mtlo %[MAX_lo], $ac0 \n\t" | |
132 | "lw %[temp_reg3], 11*4(%[csa]) \n\t" | |
133 | "msub $ac0, %[tmp1], %[temp_reg2] \n\t" | |
134 | "mtlo $zero, $ac1 \n\t" | |
135 | "madd $ac1, %[tmp0], %[temp_reg3] \n\t" | |
136 | "lw %[tmp00], -4*4(%[ptr]) \n\t" | |
137 | "lw %[tmp11], 3*4(%[ptr]) \n\t" | |
138 | "mfhi %[temp_reg1], $ac0 \n\t" | |
139 | "lw %[temp_reg4], 12*4(%[csa]) \n\t" | |
140 | "mfhi %[temp_reg2], $ac1 \n\t" | |
141 | "add %[tmp2], %[tmp00], %[tmp11] \n\t" | |
142 | "mult $ac2, %[tmp2], %[temp_reg4] \n\t" | |
143 | "mult $ac3, %[tmp2], %[temp_reg4] \n\t" | |
144 | "lw %[temp_reg5], 14*4(%[csa]) \n\t" | |
145 | "lw %[temp_reg6], 15*4(%[csa]) \n\t" | |
146 | "sll %[temp_reg1], %[temp_reg1], 2 \n\t" | |
147 | "mtlo %[MAX_lo], $ac2 \n\t" | |
148 | "msub $ac2, %[tmp11], %[temp_reg5] \n\t" | |
149 | "mtlo $zero, $ac3 \n\t" | |
150 | "madd $ac3, %[tmp00], %[temp_reg6] \n\t" | |
151 | "sll %[temp_reg2], %[temp_reg2], 2 \n\t" | |
152 | "sw %[temp_reg1], -3*4(%[ptr]) \n\t" | |
153 | "mfhi %[temp_reg4], $ac2 \n\t" | |
154 | "sw %[temp_reg2], 2*4(%[ptr]) \n\t" | |
155 | "mfhi %[temp_reg5], $ac3 \n\t" | |
156 | "lw %[tmp0], -5*4(%[ptr]) \n\t" | |
157 | "lw %[tmp1], 4*4(%[ptr]) \n\t" | |
158 | "lw %[temp_reg1], 16*4(%[csa]) \n\t" | |
159 | "lw %[temp_reg2], 18*4(%[csa]) \n\t" | |
160 | "add %[tmp2], %[tmp0], %[tmp1] \n\t" | |
161 | "lw %[temp_reg3], 19*4(%[csa]) \n\t" | |
162 | "mult $ac0, %[tmp2], %[temp_reg1] \n\t" | |
163 | "mult $ac1, %[tmp2], %[temp_reg1] \n\t" | |
164 | "sll %[temp_reg4], %[temp_reg4], 2 \n\t" | |
165 | "sll %[temp_reg5], %[temp_reg5], 2 \n\t" | |
166 | "sw %[temp_reg4], -4*4(%[ptr]) \n\t" | |
167 | "mtlo %[MAX_lo], $ac0 \n\t" | |
168 | "msub $ac0, %[tmp1], %[temp_reg2] \n\t" | |
169 | "mtlo $zero, $ac1 \n\t" | |
170 | "madd $ac1, %[tmp0], %[temp_reg3] \n\t" | |
171 | "sw %[temp_reg5], 3*4(%[ptr]) \n\t" | |
172 | "lw %[tmp00], -6*4(%[ptr]) \n\t" | |
173 | "mfhi %[temp_reg1], $ac0 \n\t" | |
174 | "lw %[tmp11], 5*4(%[ptr]) \n\t" | |
175 | "mfhi %[temp_reg2], $ac1 \n\t" | |
176 | "lw %[temp_reg4], 20*4(%[csa]) \n\t" | |
177 | "add %[tmp2], %[tmp00], %[tmp11] \n\t" | |
178 | "lw %[temp_reg5], 22*4(%[csa]) \n\t" | |
179 | "mult $ac2, %[tmp2], %[temp_reg4] \n\t" | |
180 | "mult $ac3, %[tmp2], %[temp_reg4] \n\t" | |
181 | "lw %[temp_reg6], 23*4(%[csa]) \n\t" | |
182 | "sll %[temp_reg1], %[temp_reg1], 2 \n\t" | |
183 | "sll %[temp_reg2], %[temp_reg2], 2 \n\t" | |
184 | "mtlo %[MAX_lo], $ac2 \n\t" | |
185 | "msub $ac2, %[tmp11], %[temp_reg5] \n\t" | |
186 | "mtlo $zero, $ac3 \n\t" | |
187 | "madd $ac3, %[tmp00], %[temp_reg6] \n\t" | |
188 | "sw %[temp_reg1], -5*4(%[ptr]) \n\t" | |
189 | "sw %[temp_reg2], 4*4(%[ptr]) \n\t" | |
190 | "mfhi %[temp_reg4], $ac2 \n\t" | |
191 | "lw %[tmp0], -7*4(%[ptr]) \n\t" | |
192 | "mfhi %[temp_reg5], $ac3 \n\t" | |
193 | "lw %[tmp1], 6*4(%[ptr]) \n\t" | |
194 | "lw %[temp_reg1], 24*4(%[csa]) \n\t" | |
195 | "lw %[temp_reg2], 26*4(%[csa]) \n\t" | |
196 | "add %[tmp2], %[tmp0], %[tmp1] \n\t" | |
197 | "lw %[temp_reg3], 27*4(%[csa]) \n\t" | |
198 | "mult $ac0, %[tmp2], %[temp_reg1] \n\t" | |
199 | "mult $ac1, %[tmp2], %[temp_reg1] \n\t" | |
200 | "sll %[temp_reg4], %[temp_reg4], 2 \n\t" | |
201 | "sll %[temp_reg5], %[temp_reg5], 2 \n\t" | |
202 | "sw %[temp_reg4], -6*4(%[ptr]) \n\t" | |
203 | "mtlo %[MAX_lo], $ac0 \n\t" | |
204 | "msub $ac0, %[tmp1], %[temp_reg2] \n\t" | |
205 | "mtlo $zero, $ac1 \n\t" | |
206 | "madd $ac1, %[tmp0], %[temp_reg3] \n\t" | |
207 | "sw %[temp_reg5], 5*4(%[ptr]) \n\t" | |
208 | "lw %[tmp00], -8*4(%[ptr]) \n\t" | |
209 | "mfhi %[temp_reg1], $ac0 \n\t" | |
210 | "lw %[tmp11], 7*4(%[ptr]) \n\t" | |
211 | "mfhi %[temp_reg2], $ac1 \n\t" | |
212 | "lw %[temp_reg4], 28*4(%[csa]) \n\t" | |
213 | "add %[tmp2], %[tmp00], %[tmp11] \n\t" | |
214 | "lw %[temp_reg5], 30*4(%[csa]) \n\t" | |
215 | "mult $ac2, %[tmp2], %[temp_reg4] \n\t" | |
216 | "mult $ac3, %[tmp2], %[temp_reg4] \n\t" | |
217 | "lw %[temp_reg6], 31*4(%[csa]) \n\t" | |
218 | "sll %[temp_reg1], %[temp_reg1], 2 \n\t" | |
219 | "sll %[temp_reg2], %[temp_reg2], 2 \n\t" | |
220 | "mtlo %[MAX_lo], $ac2 \n\t" | |
221 | "msub $ac2, %[tmp11], %[temp_reg5] \n\t" | |
222 | "mtlo $zero, $ac3 \n\t" | |
223 | "madd $ac3, %[tmp00], %[temp_reg6] \n\t" | |
224 | "sw %[temp_reg1], -7*4(%[ptr]) \n\t" | |
225 | "sw %[temp_reg2], 6*4(%[ptr]) \n\t" | |
226 | "mfhi %[temp_reg4], $ac2 \n\t" | |
227 | "mfhi %[temp_reg5], $ac3 \n\t" | |
228 | "sll %[temp_reg4], %[temp_reg4], 2 \n\t" | |
229 | "sll %[temp_reg5], %[temp_reg5], 2 \n\t" | |
230 | "sw %[temp_reg4], -8*4(%[ptr]) \n\t" | |
231 | "sw %[temp_reg5], 7*4(%[ptr]) \n\t" | |
232 | ||
233 | : [tmp0] "=&r" (tmp0), [tmp1] "=&r" (tmp1), [tmp2] "=&r" (tmp2), | |
234 | [tmp00] "=&r" (tmp00), [tmp11] "=&r" (tmp11), | |
235 | [temp_reg1] "=&r" (temp_reg1), [temp_reg2] "=&r" (temp_reg2), | |
236 | [temp_reg3] "=&r" (temp_reg3), [temp_reg4] "=&r" (temp_reg4), | |
237 | [temp_reg5] "=&r" (temp_reg5), [temp_reg6] "=&r" (temp_reg6) | |
238 | : [csa] "r" (csa), [ptr] "r" (ptr), | |
239 | [MAX_lo] "r" (MAX_lo) | |
240 | : "memory", "hi", "lo", "$ac1hi", "$ac1lo", "$ac2hi", "$ac2lo", | |
241 | "$ac3hi", "$ac3lo" | |
242 | ); | |
243 | ||
244 | ptr += 18; | |
245 | } | |
246 | } | |
247 | #define compute_antialias compute_antialias_mips_fixed | |
248 | #endif /* HAVE_INLINE_ASM */ | |
249 | ||
250 | #endif /* AVCODEC_MIPS_COMPUTE_ANTIALIAS_FIXED_H */ |