/*
Copyright (c) 2007-2009, The Musepack Development Team
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* 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.
* Neither the name of the The Musepack Development Team 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 COPYRIGHT HOLDERS AND CONTRIBUTORS
"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 COPYRIGHT
OWNER OR CONTRIBUTORS 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.
*/
#define MAX_ENUM 32
MPC_API int mpc_bits_get_block(mpc_bits_reader * r, mpc_block * p_block);
mpc_int32_t mpc_bits_golomb_dec(mpc_bits_reader * r, const mpc_uint_t k);
MPC_API unsigned int mpc_bits_get_size(mpc_bits_reader * r, mpc_uint64_t * p_size);
mpc_uint32_t mpc_bits_log_dec(mpc_bits_reader * r, mpc_uint_t max);
extern const mpc_uint32_t Cnk [MAX_ENUM / 2][MAX_ENUM];
extern const mpc_uint8_t Cnk_len [MAX_ENUM / 2][MAX_ENUM];
extern const mpc_uint32_t Cnk_lost[MAX_ENUM / 2][MAX_ENUM];
// can read up to 31 bits
static mpc_inline mpc_uint32_t mpc_bits_read(mpc_bits_reader * r, const unsigned int nb_bits)
{
mpc_uint32_t ret;
r->buff -= (int)(r->count - nb_bits) >> 3;
r->count = (r->count - nb_bits) & 0x07;
ret = (r->buff[0] | (r->buff[-1] << 8)) >> r->count;
if (nb_bits > (16 - r->count)) {
ret |= (mpc_uint32_t)((r->buff[-2] << 16) | (r->buff[-3] << 24)) >> r->count;
if (nb_bits > 24 && r->count != 0)
ret |= r->buff[-4] << (32 - r->count);
}
return ret & ((1 << nb_bits) - 1);
}
#if defined(CPU_COLDFIRE)
/* rockbox: This is specific code to optimize demux performance on Coldfire
* CPUs. Coldfire CPUs are very sensible to RAM accesses. As the bitstream
* buffer does not fit into IRAM the read accesses to the uint8 buffer are very
* expensive in terms of CPU cycles.
* The following code uses two variables in IRAM. The variable last_code keeps
* the 4-byte value of buf[0]<<16 | buf[1]<<8 | buf[2]. As long as buf[0] will
* read from the same address the following code will avoid re-reading of the
* buffers. If buf[0] did advance to the next uint8-entry since the last call
* the following will only need to load 1 uint8-entry instead of 3.
*/
static mpc_inline mpc_uint16_t get_code_from_buffer(mpc_bits_reader *r)
{
/* Buffer advanced by 1 entry since last call */
if (r->buff == r->buffered_addr + 1) {
r->buffered_code = (r->buffered_code<<8) | r->buff[2];
r->buffered_addr = r->buff;
}
/* Buffer must be fully re-read */
else if (r->buff != r->buffered_addr) {
r->buffered_code = (r->buff[0] << 16) | (r->buff[1] << 8) | r->buff[2];
r->buffered_addr = r->buff;
}
return (mpc_uint16_t)((r->buffered_code >> r->count) & 0xFFFF);
}
#else
/* Use the decoder's default implementation. This is faster on non-Coldfire targets */
#define get_code_from_buffer(r) (mpc_uint16_t)((((r->buff[0] << 16) | (r->buff[1] << 8) | r->buff[2]) >> r->count) & 0xFFFF);
#endif
// basic huffman decoding routine
// works with maximum lengths up to 16
static mpc_inline mpc_int32_t mpc_bits_huff_dec(mpc_bits_reader * r, const mpc_huffman *Table)
{
const mpc_uint16_t code = get_code_from_buffer(r);
while (code < Table->Code) Table++;
r->buff -= (int)(r->count - Table->Length) >> 3;
r->count = (r->count - Table->Length) & 0x07;
return Table->Value;
}
static mpc_inline mpc_int32_t mpc_bits_can_dec(mpc_bits_reader * r, const mpc_can_data *can)
{
const mpc_uint16_t code = get_code_from_buffer(r);
const mpc_huff_lut tmp = can->lut[code >> (16 - LUT_DEPTH)];
const mpc_huffman * Table;
if (tmp.Length != 0) {
r->buff -= (int)(r->count - tmp.Length) >> 3;
r->count = (r->count - tmp.Length) & 0x07;
return tmp.Value;
}
Table = can->table + (unsigned char)tmp.Value;
while (code < Table->Code) Table++;
r->buff -= (int)(r->count - Table->Length) >> 3;
r->count = (r->count - Table->Length) & 0x07;
return can->sym[(Table->Value - (code >> (16 - Table->Length))) & 0xFF] ;
}
// LUT-based huffman decoding routine
// works with maximum lengths up to 16
static mpc_inline mpc_int32_t mpc_bits_huff_lut(mpc_bits_reader * r, const mpc_lut_data *lut)
{
const mpc_uint16_t code = get_code_from_buffer(r);
const mpc_huff_lut tmp = lut->lut[code >> (16 - LUT_DEPTH)];
const mpc_huffman * Table;
if (tmp.Length != 0) {
r->buff -= (int)(r->count - tmp.Length) >> 3;
r->count = (r->count - tmp.Length) & 0x07;
return tmp.Value;
}
Table = lut->table + (unsigned char)tmp.Value;
while (code < Table->Code) Table++;
r->buff -= (int)(r->count - Table->Length) >> 3;
r->count = (r->count - Table->Length) & 0x07;
return Table->Value;
}
static mpc_inline mpc_uint32_t mpc_bits_enum_dec(mpc_bits_reader * r, mpc_uint_t k, mpc_uint_t n)
{
mpc_uint32_t bits = 0;
mpc_uint32_t code;
const mpc_uint32_t * C = Cnk[k-1];
code = mpc_bits_read(r, Cnk_len[k-1][n-1] - 1);
if (code >= Cnk_lost[k-1][n-1])
code = ((code << 1) | mpc_bits_read(r, 1)) - Cnk_lost[k-1][n-1];
do {
n--;
if (code >= C[n]) {
bits |= 1 << n;
code -= C[n];
C -= MAX_ENUM;
k--;
}
} while(k > 0);
return bits;
}
04'>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
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2007 by Michael Sevakis
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
.global mpeg2_idct_copy
.type mpeg2_idct_copy, %function
.global mpeg2_idct_add
.type mpeg2_idct_add, %function
/* Custom calling convention:
* r0 contains block pointer and is non-volatile
* all non-volatile c context saved and restored on its behalf
*/
.idct:
add r12, r0, #128
1:
ldrsh r1, [r0, #0] /* d0 */
ldrsh r2, [r0, #2] /* d1 */
ldrsh r3, [r0, #4] /* d2 */
ldrsh r4, [r0, #6] /* d3 */
ldrsh r5, [r0, #8] /* d0 */
ldrsh r6, [r0, #10] /* d1 */
ldrsh r7, [r0, #12] /* d2 */
ldrsh r8, [r0, #14] /* d3 */
orrs r9, r2, r3
orreqs r9, r4, r5
orreqs r9, r6, r7
cmpeq r8, #0
bne 2f
mov r1, r1, asl #15
bic r1, r1, #0x8000
orr r1, r1, r1, lsr #16
str r1, [r0], #4
str r1, [r0], #4
str r1, [r0], #4
str r1, [r0], #4
cmp r0, r12
blo 1b
b 3f
2:
mov r1, r1, asl #11 /* r1 = d0 = (block[0] << 11) + 2048 */
add r1, r1, #2048
add r1, r1, r3, asl #11 /* r1 = t0 = d0 + (block[2] << 11) */
sub r3, r1, r3, asl #12 /* r3 = t1 = d0 - (block[2] << 11) */
add r9, r2, r4 /* r9 = tmp = (d1+d3)*(1108/4) */
add r10, r9, r9, asl #2
add r10, r10, r9, asl #4
add r9, r10, r9, asl #8
add r10, r2, r2, asl #4 /* r2 = t2 = tmp + (d1*(1568/32)*8) */
add r2, r10, r2, asl #5
add r2, r9, r2, asl #3
add r10, r4, r4, asl #2 /* r4 = t3 = tmp - (d3*(3784/8)*2) */
rsb r10, r10, r4, asl #6
add r4, r4, r10, asl #3
sub r4, r9, r4, asl #1
/* t2 & t3 are 1/4 final value here */
add r1, r1, r2, asl #2 /* r1 = a0 = t0 + t2 */
sub r2, r1, r2, asl #3 /* r2 = a3 = t0 - t2 */
add r3, r3, r4, asl #2 /* r3 = a1 = t1 + t3 */
sub r4, r3, r4, asl #3 /* r4 = a2 = t1 - t3 */
add r9, r8, r5 /* r9 = tmp = 565*(d3 + d0) */
add r10, r9, r9, asl #4
add r10, r10, r10, asl #5
add r9, r10, r9, asl #2
add r10, r5, r5, asl #4 /* r5 = t0 = tmp + (((2276/4)*d0)*4) */
add r10, r10, r10, asl #5
add r5, r10, r5, asl #3
add r5, r9, r5, asl #2
add r10, r8, r8, asl #2 /* r8 = t1 = tmp - (((3406/2)*d3)*2) */
add r10, r10, r10, asl #4
add r10, r10, r8, asl #7
rsb r8, r8, r10, asl #3
sub r8, r9, r8, asl #1
add r9, r6, r7 /* r9 = tmp = (2408/8)*(d1 + d2) */
add r10, r9, r9, asl #3
add r10, r10, r10, asl #5
add r9, r10, r9, asl #2
add r10, r7, r7, asl #3 /* r7 = t2 = (tmp*8) - 799*d2 */
add r10, r10, r7, asl #4
rsb r7, r7, r10, asl #5
rsb r7, r7, r9, asl #3
sub r10, r6, r6, asl #4 /* r6 = t3 = (tmp*8) - 4017*d1 */
sub r10, r10, r6, asl #6
add r10, r10, r6, asl #12
add r6, r10, r6
rsb r6, r6, r9, asl #3
/* t0 = r5, t1 = r8, t2 = r7, t3 = r6*/
add r9, r5, r7 /* r9 = b0 = t0 + t2 */
add r10, r8, r6 /* r10 = b3 = t1 + t3 */
sub r5, r5, r7 /* t0 -= t2 */
sub r8, r8, r6 /* t1 -= t3 */
add r6, r5, r8 /* r6 = t0 + t1 */
sub r7, r5, r8 /* r7 = t0 - t1 */
add r11, r6, r6, asr #2 /* r6 = b1 = r6*(181/128) */
add r11, r11, r11, asr #5
add r6, r11, r6, asr #3
add r11, r7, r7, asr #2 /* r7 = b2 = r7*(181/128) */
add r11, r11, r11, asr #5
add r7, r11, r7, asr #3
/* r1 = a0, r3 = a1, r4 = a2, r2 = a3 */
/* r9 = b0, r6 = b1*2, r7 = b2*2, r10 = b3 */
add r5, r1, r9 /* block[0] = (a0 + b0) >> 12 */
mov r5, r5, asr #12
strh r5, [r0], #2
add r8, r3, r6, asr #1 /* block[1] = (a1 + b1) >> 12 */
mov r8, r8, asr #12
strh r8, [r0], #2
add r5, r4, r7, asr #1 /* block[2] = (a2 + b2) >> 12 */
mov r5, r5, asr #12
strh r5, [r0], #2
add r8, r2, r10 /* block[3] = (a3 + b3) >> 12 */
mov r8, r8, asr #12
strh r8, [r0], #2
sub r5, r2, r10 /* block[4] = (a3 - b3) >> 12 */
mov r5, r5, asr #12
strh r5, [r0], #2
sub r8, r4, r7, asr #1 /* block[5] = (a2 - b2) >> 12 */
mov r8, r8, asr #12
strh r8, [r0], #2
sub r5, r3, r6, asr #1 /* block[6] = (a1 - b1) >> 12 */
mov r5, r5, asr #12
strh r5, [r0], #2
sub r8, r1, r9 /* block[7] = (a0 - b0) >> 12 */
mov r8, r8, asr #12
strh r8, [r0], #2
cmp r0, r12
blo 1b
3:
sub r0, r0, #128
add r12, r0, #16
4:
ldrsh r1, [r0, #0*8] /* d0 */
ldrsh r2, [r0, #2*8] /* d1 */
ldrsh r3, [r0, #4*8] /* d2 */
ldrsh r4, [r0, #6*8] /* d3 */
ldrsh r5, [r0, #8*8] /* d0 */
ldrsh r6, [r0, #10*8] /* d1 */
ldrsh r7, [r0, #12*8] /* d2 */
ldrsh r8, [r0, #14*8] /* d3 */
mov r1, r1, asl #11 /* r1 = d0 = (block[0] << 11) + 2048 */
add r1, r1, #65536
add r1, r1, r3, asl #11 /* r1 = t0 = d0 + d2:(block[2] << 11) */
sub r3, r1, r3, asl #12 /* r3 = t1 = d0 - d2:(block[2] << 11) */
add r9, r2, r4 /* r9 = tmp = (d1+d3)*(1108/4) */
add r10, r9, r9, asl #2
add r10, r10, r9, asl #4
add r9, r10, r9, asl #8
add r11, r2, r2, asl #4 /* r2 = t2 = tmp + (d1*(1568/32)*8) */
add r2, r11, r2, asl #5
add r2, r9, r2, asl #3
add r10, r4, r4, asl #2 /* r4 = t3 = tmp - (d3*(3784/8)*2) */
rsb r10, r10, r4, asl #6
add r4, r4, r10, asl #3
sub r4, r9, r4, asl #1
/* t2 & t3 are 1/4 final value here */
add r1, r1, r2, asl #2 /* r1 = a0 = t0 + t2 */
sub r2, r1, r2, asl #3 /* r2 = a3 = t0 - t2 */
add r3, r3, r4, asl #2 /* r3 = a1 = t1 + t3 */
sub r4, r3, r4, asl #3 /* r4 = a2 = t1 - t3 */
add r9, r8, r5 /* r9 = tmp = 565*(d3 + d0) */
add r10, r9, r9, asl #4
add r10, r10, r10, asl #5
add r9, r10, r9, asl #2
add r10, r5, r5, asl #4 /* r5 = t0 = tmp + (((2276/4)*d0)*4) */
add r10, r10, r10, asl #5
add r5, r10, r5, asl #3
add r5, r9, r5, asl #2
add r10, r8, r8, asl #2 /* r8 = t1 = tmp - (((3406/2)*d3)*2) */
add r10, r10, r10, asl #4
add r10, r10, r8, asl #7
rsb r8, r8, r10, asl #3
sub r8, r9, r8, asl #1
add r9, r6, r7 /* r9 = tmp = (2408/8)*(d1 + d2) */
add r10, r9, r9, asl #3
add r10, r10, r10, asl #5
add r9, r10, r9, asl #2
add r10, r7, r7, asl #3 /* r7 = t2 = (tmp*8) - 799*d2 */
add r10, r10, r7, asl #4
rsb r7, r7, r10, asl #5
rsb r7, r7, r9, asl #3
sub r10, r6, r6, asl #4 /* r6 = t3 = (tmp*8) - 4017*d1 */
sub r10, r10, r6, asl #6
add r10, r10, r6, asl #12
add r6, r10, r6
rsb r6, r6, r9, asl #3
/* t0=r5, t1=r8, t2=r7, t3=r6*/
add r9, r5, r7 /* r9 = b0 = t0 + t2 */
add r10, r8, r6 /* r10 = b3 = t1 + t3 */
sub r5, r5, r7 /* t0 -= t2 */
sub r8, r8, r6 /* t1 -= t3 */
add r6, r5, r8 /* r6 = t0 + t1 */
sub r7, r5, r8 /* r7 = t0 - t1 */
add r11, r6, r6, asr #2 /* r6 = b1 = r5*(181/128) */
add r11, r11, r11, asr #5
add r6, r11, r6, asr #3
add r11, r7, r7, asr #2 /* r7 = b2 = r6*(181/128) */
add r11, r11, r11, asr #5
add r7, r11, r7, asr #3
/* r1 = a0, r3 = a1, r4 = a2, r2 = a3 */
/* r9 = b0, r6 = b1*2, r7 = b2*2, r10 = b3 */
add r5, r1, r9 /* block[0] = (a0 + b0) >> 17 */
mov r5, r5, asr #17
strh r5, [r0, #0*8]
add r8, r3, r6, asr #1 /* block[1] = (a1 + b1) >> 17 */
mov r8, r8, asr #17
strh r8, [r0, #2*8]
add r5, r4, r7, asr #1 /* block[2] = (a2 + b2) >> 17 */
mov r5, r5, asr #17
strh r5, [r0, #4*8]
add r8, r2, r10 /* block[3] = (a3 + b3) >> 17 */
mov r8, r8, asr #17
strh r8, [r0, #6*8]
sub r5, r2, r10 /* block[4] = (a3 - b3) >> 17 */
mov r5, r5, asr #17
strh r5, [r0, #8*8]
sub r8, r4, r7, asr #1 /* block[5] = (a2 - b2) >> 17 */
mov r8, r8, asr #17
strh r8, [r0, #10*8]
sub r5, r3, r6, asr #1 /* block[6] = (a1 - b1) >> 17 */
mov r5, r5, asr #17
strh r5, [r0, #12*8]
|