Source

APKinspector / androguard / classification / libsimilarity / sources / xz-5.0.2 / src / liblzma / lz / lz_encoder_mf.c

The default branch has multiple heads

Full commit
  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
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
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
///////////////////////////////////////////////////////////////////////////////
//
/// \file       lz_encoder_mf.c
/// \brief      Match finders
///
//  Authors:    Igor Pavlov
//              Lasse Collin
//
//  This file has been put into the public domain.
//  You can do whatever you want with this file.
//
///////////////////////////////////////////////////////////////////////////////

#include "lz_encoder.h"
#include "lz_encoder_hash.h"


/// \brief      Find matches starting from the current byte
///
/// \return     The length of the longest match found
extern uint32_t
lzma_mf_find(lzma_mf *mf, uint32_t *count_ptr, lzma_match *matches)
{
	// Call the match finder. It returns the number of length-distance
	// pairs found.
	// FIXME: Minimum count is zero, what _exactly_ is the maximum?
	const uint32_t count = mf->find(mf, matches);

	// Length of the longest match; assume that no matches were found
	// and thus the maximum length is zero.
	uint32_t len_best = 0;

	if (count > 0) {
#ifndef NDEBUG
		// Validate the matches.
		for (uint32_t i = 0; i < count; ++i) {
			assert(matches[i].len <= mf->nice_len);
			assert(matches[i].dist < mf->read_pos);
			assert(memcmp(mf_ptr(mf) - 1,
				mf_ptr(mf) - matches[i].dist - 2,
				matches[i].len) == 0);
		}
#endif

		// The last used element in the array contains
		// the longest match.
		len_best = matches[count - 1].len;

		// If a match of maximum search length was found, try to
		// extend the match to maximum possible length.
		if (len_best == mf->nice_len) {
			// The limit for the match length is either the
			// maximum match length supported by the LZ-based
			// encoder or the number of bytes left in the
			// dictionary, whichever is smaller.
			uint32_t limit = mf_avail(mf) + 1;
			if (limit > mf->match_len_max)
				limit = mf->match_len_max;

			// Pointer to the byte we just ran through
			// the match finder.
			const uint8_t *p1 = mf_ptr(mf) - 1;

			// Pointer to the beginning of the match. We need -1
			// here because the match distances are zero based.
			const uint8_t *p2 = p1 - matches[count - 1].dist - 1;

			while (len_best < limit
					&& p1[len_best] == p2[len_best])
				++len_best;
		}
	}

	*count_ptr = count;

	// Finally update the read position to indicate that match finder was
	// run for this dictionary offset.
	++mf->read_ahead;

	return len_best;
}


/// Hash value to indicate unused element in the hash. Since we start the
/// positions from dict_size + 1, zero is always too far to qualify
/// as usable match position.
#define EMPTY_HASH_VALUE 0


/// Normalization must be done when lzma_mf.offset + lzma_mf.read_pos
/// reaches MUST_NORMALIZE_POS.
#define MUST_NORMALIZE_POS UINT32_MAX


/// \brief      Normalizes hash values
///
/// The hash arrays store positions of match candidates. The positions are
/// relative to an arbitrary offset that is not the same as the absolute
/// offset in the input stream. The relative position of the current byte
/// is lzma_mf.offset + lzma_mf.read_pos. The distances of the matches are
/// the differences of the current read position and the position found from
/// the hash.
///
/// To prevent integer overflows of the offsets stored in the hash arrays,
/// we need to "normalize" the stored values now and then. During the
/// normalization, we drop values that indicate distance greater than the
/// dictionary size, thus making space for new values.
static void
normalize(lzma_mf *mf)
{
	assert(mf->read_pos + mf->offset == MUST_NORMALIZE_POS);

	// In future we may not want to touch the lowest bits, because there
	// may be match finders that use larger resolution than one byte.
	const uint32_t subvalue
			= (MUST_NORMALIZE_POS - mf->cyclic_size);
				// & (~(UINT32_C(1) << 10) - 1);

	const uint32_t count = mf->hash_size_sum + mf->sons_count;
	uint32_t *hash = mf->hash;

	for (uint32_t i = 0; i < count; ++i) {
		// If the distance is greater than the dictionary size,
		// we can simply mark the hash element as empty.
		//
		// NOTE: Only the first mf->hash_size_sum elements are
		// initialized for sure. There may be uninitialized elements
		// in mf->son. Since we go through both mf->hash and
		// mf->son here in normalization, Valgrind may complain
		// that the "if" below depends on uninitialized value. In
		// this case it is safe to ignore the warning. See also the
		// comments in lz_encoder_init() in lz_encoder.c.
		if (hash[i] <= subvalue)
			hash[i] = EMPTY_HASH_VALUE;
		else
			hash[i] -= subvalue;
	}

	// Update offset to match the new locations.
	mf->offset -= subvalue;

	return;
}


/// Mark the current byte as processed from point of view of the match finder.
static void
move_pos(lzma_mf *mf)
{
	if (++mf->cyclic_pos == mf->cyclic_size)
		mf->cyclic_pos = 0;

	++mf->read_pos;
	assert(mf->read_pos <= mf->write_pos);

	if (unlikely(mf->read_pos + mf->offset == UINT32_MAX))
		normalize(mf);
}


/// When flushing, we cannot run the match finder unless there is nice_len
/// bytes available in the dictionary. Instead, we skip running the match
/// finder (indicating that no match was found), and count how many bytes we
/// have ignored this way.
///
/// When new data is given after the flushing was completed, the match finder
/// is restarted by rewinding mf->read_pos backwards by mf->pending. Then
/// the missed bytes are added to the hash using the match finder's skip
/// function (with small amount of input, it may start using mf->pending
/// again if flushing).
///
/// Due to this rewinding, we don't touch cyclic_pos or test for
/// normalization. It will be done when the match finder's skip function
/// catches up after a flush.
static void
move_pending(lzma_mf *mf)
{
	++mf->read_pos;
	assert(mf->read_pos <= mf->write_pos);
	++mf->pending;
}


/// Calculate len_limit and determine if there is enough input to run
/// the actual match finder code. Sets up "cur" and "pos". This macro
/// is used by all find functions and binary tree skip functions. Hash
/// chain skip function doesn't need len_limit so a simpler code is used
/// in them.
#define header(is_bt, len_min, ret_op) \
	uint32_t len_limit = mf_avail(mf); \
	if (mf->nice_len <= len_limit) { \
		len_limit = mf->nice_len; \
	} else if (len_limit < (len_min) \
			|| (is_bt && mf->action == LZMA_SYNC_FLUSH)) { \
		assert(mf->action != LZMA_RUN); \
		move_pending(mf); \
		ret_op; \
	} \
	const uint8_t *cur = mf_ptr(mf); \
	const uint32_t pos = mf->read_pos + mf->offset


/// Header for find functions. "return 0" indicates that zero matches
/// were found.
#define header_find(is_bt, len_min) \
	header(is_bt, len_min, return 0); \
	uint32_t matches_count = 0


/// Header for a loop in a skip function. "continue" tells to skip the rest
/// of the code in the loop.
#define header_skip(is_bt, len_min) \
	header(is_bt, len_min, continue)


/// Calls hc_find_func() or bt_find_func() and calculates the total number
/// of matches found. Updates the dictionary position and returns the number
/// of matches found.
#define call_find(func, len_best) \
do { \
	matches_count = func(len_limit, pos, cur, cur_match, mf->depth, \
				mf->son, mf->cyclic_pos, mf->cyclic_size, \
				matches + matches_count, len_best) \
			- matches; \
	move_pos(mf); \
	return matches_count; \
} while (0)


////////////////
// Hash Chain //
////////////////

#if defined(HAVE_MF_HC3) || defined(HAVE_MF_HC4)
///
///
/// \param      len_limit       Don't look for matches longer than len_limit.
/// \param      pos             lzma_mf.read_pos + lzma_mf.offset
/// \param      cur             Pointer to current byte (mf_ptr(mf))
/// \param      cur_match       Start position of the current match candidate
/// \param      depth           Maximum length of the hash chain
/// \param      son             lzma_mf.son (contains the hash chain)
/// \param      cyclic_pos
/// \param      cyclic_size
/// \param      matches         Array to hold the matches.
/// \param      len_best        The length of the longest match found so far.
static lzma_match *
hc_find_func(
		const uint32_t len_limit,
		const uint32_t pos,
		const uint8_t *const cur,
		uint32_t cur_match,
		uint32_t depth,
		uint32_t *const son,
		const uint32_t cyclic_pos,
		const uint32_t cyclic_size,
		lzma_match *matches,
		uint32_t len_best)
{
	son[cyclic_pos] = cur_match;

	while (true) {
		const uint32_t delta = pos - cur_match;
		if (depth-- == 0 || delta >= cyclic_size)
			return matches;

		const uint8_t *const pb = cur - delta;
		cur_match = son[cyclic_pos - delta
				+ (delta > cyclic_pos ? cyclic_size : 0)];

		if (pb[len_best] == cur[len_best] && pb[0] == cur[0]) {
			uint32_t len = 0;
			while (++len != len_limit)
				if (pb[len] != cur[len])
					break;

			if (len_best < len) {
				len_best = len;
				matches->len = len;
				matches->dist = delta - 1;
				++matches;

				if (len == len_limit)
					return matches;
			}
		}
	}
}


#define hc_find(len_best) \
	call_find(hc_find_func, len_best)


#define hc_skip() \
do { \
	mf->son[mf->cyclic_pos] = cur_match; \
	move_pos(mf); \
} while (0)

#endif


#ifdef HAVE_MF_HC3
extern uint32_t
lzma_mf_hc3_find(lzma_mf *mf, lzma_match *matches)
{
	header_find(false, 3);

	hash_3_calc();

	const uint32_t delta2 = pos - mf->hash[hash_2_value];
	const uint32_t cur_match = mf->hash[FIX_3_HASH_SIZE + hash_value];

	mf->hash[hash_2_value] = pos;
	mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;

	uint32_t len_best = 2;

	if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
		for ( ; len_best != len_limit; ++len_best)
			if (*(cur + len_best - delta2) != cur[len_best])
				break;

		matches[0].len = len_best;
		matches[0].dist = delta2 - 1;
		matches_count = 1;

		if (len_best == len_limit) {
			hc_skip();
			return 1; // matches_count
		}
	}

	hc_find(len_best);
}


extern void
lzma_mf_hc3_skip(lzma_mf *mf, uint32_t amount)
{
	do {
		if (mf_avail(mf) < 3) {
			move_pending(mf);
			continue;
		}

		const uint8_t *cur = mf_ptr(mf);
		const uint32_t pos = mf->read_pos + mf->offset;

		hash_3_calc();

		const uint32_t cur_match
				= mf->hash[FIX_3_HASH_SIZE + hash_value];

		mf->hash[hash_2_value] = pos;
		mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;

		hc_skip();

	} while (--amount != 0);
}
#endif


#ifdef HAVE_MF_HC4
extern uint32_t
lzma_mf_hc4_find(lzma_mf *mf, lzma_match *matches)
{
	header_find(false, 4);

	hash_4_calc();

	uint32_t delta2 = pos - mf->hash[hash_2_value];
	const uint32_t delta3
			= pos - mf->hash[FIX_3_HASH_SIZE + hash_3_value];
	const uint32_t cur_match = mf->hash[FIX_4_HASH_SIZE + hash_value];

	mf->hash[hash_2_value ] = pos;
	mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
	mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;

	uint32_t len_best = 1;

	if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
		len_best = 2;
		matches[0].len = 2;
		matches[0].dist = delta2 - 1;
		matches_count = 1;
	}

	if (delta2 != delta3 && delta3 < mf->cyclic_size
			&& *(cur - delta3) == *cur) {
		len_best = 3;
		matches[matches_count++].dist = delta3 - 1;
		delta2 = delta3;
	}

	if (matches_count != 0) {
		for ( ; len_best != len_limit; ++len_best)
			if (*(cur + len_best - delta2) != cur[len_best])
				break;

		matches[matches_count - 1].len = len_best;

		if (len_best == len_limit) {
			hc_skip();
			return matches_count;
		}
	}

	if (len_best < 3)
		len_best = 3;

	hc_find(len_best);
}


extern void
lzma_mf_hc4_skip(lzma_mf *mf, uint32_t amount)
{
	do {
		if (mf_avail(mf) < 4) {
			move_pending(mf);
			continue;
		}

		const uint8_t *cur = mf_ptr(mf);
		const uint32_t pos = mf->read_pos + mf->offset;

		hash_4_calc();

		const uint32_t cur_match
				= mf->hash[FIX_4_HASH_SIZE + hash_value];

		mf->hash[hash_2_value] = pos;
		mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
		mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;

		hc_skip();

	} while (--amount != 0);
}
#endif


/////////////////
// Binary Tree //
/////////////////

#if defined(HAVE_MF_BT2) || defined(HAVE_MF_BT3) || defined(HAVE_MF_BT4)
static lzma_match *
bt_find_func(
		const uint32_t len_limit,
		const uint32_t pos,
		const uint8_t *const cur,
		uint32_t cur_match,
		uint32_t depth,
		uint32_t *const son,
		const uint32_t cyclic_pos,
		const uint32_t cyclic_size,
		lzma_match *matches,
		uint32_t len_best)
{
	uint32_t *ptr0 = son + (cyclic_pos << 1) + 1;
	uint32_t *ptr1 = son + (cyclic_pos << 1);

	uint32_t len0 = 0;
	uint32_t len1 = 0;

	while (true) {
		const uint32_t delta = pos - cur_match;
		if (depth-- == 0 || delta >= cyclic_size) {
			*ptr0 = EMPTY_HASH_VALUE;
			*ptr1 = EMPTY_HASH_VALUE;
			return matches;
		}

		uint32_t *const pair = son + ((cyclic_pos - delta
				+ (delta > cyclic_pos ? cyclic_size : 0))
				<< 1);

		const uint8_t *const pb = cur - delta;
		uint32_t len = my_min(len0, len1);

		if (pb[len] == cur[len]) {
			while (++len != len_limit)
				if (pb[len] != cur[len])
					break;

			if (len_best < len) {
				len_best = len;
				matches->len = len;
				matches->dist = delta - 1;
				++matches;

				if (len == len_limit) {
					*ptr1 = pair[0];
					*ptr0 = pair[1];
					return matches;
				}
			}
		}

		if (pb[len] < cur[len]) {
			*ptr1 = cur_match;
			ptr1 = pair + 1;
			cur_match = *ptr1;
			len1 = len;
		} else {
			*ptr0 = cur_match;
			ptr0 = pair;
			cur_match = *ptr0;
			len0 = len;
		}
	}
}


static void
bt_skip_func(
		const uint32_t len_limit,
		const uint32_t pos,
		const uint8_t *const cur,
		uint32_t cur_match,
		uint32_t depth,
		uint32_t *const son,
		const uint32_t cyclic_pos,
		const uint32_t cyclic_size)
{
	uint32_t *ptr0 = son + (cyclic_pos << 1) + 1;
	uint32_t *ptr1 = son + (cyclic_pos << 1);

	uint32_t len0 = 0;
	uint32_t len1 = 0;

	while (true) {
		const uint32_t delta = pos - cur_match;
		if (depth-- == 0 || delta >= cyclic_size) {
			*ptr0 = EMPTY_HASH_VALUE;
			*ptr1 = EMPTY_HASH_VALUE;
			return;
		}

		uint32_t *pair = son + ((cyclic_pos - delta
				+ (delta > cyclic_pos ? cyclic_size : 0))
				<< 1);
		const uint8_t *pb = cur - delta;
		uint32_t len = my_min(len0, len1);

		if (pb[len] == cur[len]) {
			while (++len != len_limit)
				if (pb[len] != cur[len])
					break;

			if (len == len_limit) {
				*ptr1 = pair[0];
				*ptr0 = pair[1];
				return;
			}
		}

		if (pb[len] < cur[len]) {
			*ptr1 = cur_match;
			ptr1 = pair + 1;
			cur_match = *ptr1;
			len1 = len;
		} else {
			*ptr0 = cur_match;
			ptr0 = pair;
			cur_match = *ptr0;
			len0 = len;
		}
	}
}


#define bt_find(len_best) \
	call_find(bt_find_func, len_best)

#define bt_skip() \
do { \
	bt_skip_func(len_limit, pos, cur, cur_match, mf->depth, \
			mf->son, mf->cyclic_pos, \
			mf->cyclic_size); \
	move_pos(mf); \
} while (0)

#endif


#ifdef HAVE_MF_BT2
extern uint32_t
lzma_mf_bt2_find(lzma_mf *mf, lzma_match *matches)
{
	header_find(true, 2);

	hash_2_calc();

	const uint32_t cur_match = mf->hash[hash_value];
	mf->hash[hash_value] = pos;

	bt_find(1);
}


extern void
lzma_mf_bt2_skip(lzma_mf *mf, uint32_t amount)
{
	do {
		header_skip(true, 2);

		hash_2_calc();

		const uint32_t cur_match = mf->hash[hash_value];
		mf->hash[hash_value] = pos;

		bt_skip();

	} while (--amount != 0);
}
#endif


#ifdef HAVE_MF_BT3
extern uint32_t
lzma_mf_bt3_find(lzma_mf *mf, lzma_match *matches)
{
	header_find(true, 3);

	hash_3_calc();

	const uint32_t delta2 = pos - mf->hash[hash_2_value];
	const uint32_t cur_match = mf->hash[FIX_3_HASH_SIZE + hash_value];

	mf->hash[hash_2_value] = pos;
	mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;

	uint32_t len_best = 2;

	if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
		for ( ; len_best != len_limit; ++len_best)
			if (*(cur + len_best - delta2) != cur[len_best])
				break;

		matches[0].len = len_best;
		matches[0].dist = delta2 - 1;
		matches_count = 1;

		if (len_best == len_limit) {
			bt_skip();
			return 1; // matches_count
		}
	}

	bt_find(len_best);
}


extern void
lzma_mf_bt3_skip(lzma_mf *mf, uint32_t amount)
{
	do {
		header_skip(true, 3);

		hash_3_calc();

		const uint32_t cur_match
				= mf->hash[FIX_3_HASH_SIZE + hash_value];

		mf->hash[hash_2_value] = pos;
		mf->hash[FIX_3_HASH_SIZE + hash_value] = pos;

		bt_skip();

	} while (--amount != 0);
}
#endif


#ifdef HAVE_MF_BT4
extern uint32_t
lzma_mf_bt4_find(lzma_mf *mf, lzma_match *matches)
{
	header_find(true, 4);

	hash_4_calc();

	uint32_t delta2 = pos - mf->hash[hash_2_value];
	const uint32_t delta3
			= pos - mf->hash[FIX_3_HASH_SIZE + hash_3_value];
	const uint32_t cur_match = mf->hash[FIX_4_HASH_SIZE + hash_value];

	mf->hash[hash_2_value] = pos;
	mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
	mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;

	uint32_t len_best = 1;

	if (delta2 < mf->cyclic_size && *(cur - delta2) == *cur) {
		len_best = 2;
		matches[0].len = 2;
		matches[0].dist = delta2 - 1;
		matches_count = 1;
	}

	if (delta2 != delta3 && delta3 < mf->cyclic_size
			&& *(cur - delta3) == *cur) {
		len_best = 3;
		matches[matches_count++].dist = delta3 - 1;
		delta2 = delta3;
	}

	if (matches_count != 0) {
		for ( ; len_best != len_limit; ++len_best)
			if (*(cur + len_best - delta2) != cur[len_best])
				break;

		matches[matches_count - 1].len = len_best;

		if (len_best == len_limit) {
			bt_skip();
			return matches_count;
		}
	}

	if (len_best < 3)
		len_best = 3;

	bt_find(len_best);
}


extern void
lzma_mf_bt4_skip(lzma_mf *mf, uint32_t amount)
{
	do {
		header_skip(true, 4);

		hash_4_calc();

		const uint32_t cur_match
				= mf->hash[FIX_4_HASH_SIZE + hash_value];

		mf->hash[hash_2_value] = pos;
		mf->hash[FIX_3_HASH_SIZE + hash_3_value] = pos;
		mf->hash[FIX_4_HASH_SIZE + hash_value] = pos;

		bt_skip();

	} while (--amount != 0);
}
#endif