2 * LZ4 HC - High Compression Mode of LZ4
3 * Copyright (C) 2011-2015, Yann Collet.
5 * BSD 2 - Clause License (http://www.opensource.org/licenses/bsd - license.php)
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions are
9 * * Redistributions of source code must retain the above copyright
10 * notice, this list of conditions and the following disclaimer.
11 * * Redistributions in binary form must reproduce the above
12 * copyright notice, this list of conditions and the following disclaimer
13 * in the documentation and/or other materials provided with the
15 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
16 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
17 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
18 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
19 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
20 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
21 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
22 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
23 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
25 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26 * You can contact the author at :
27 * - LZ4 homepage : http://www.lz4.org
28 * - LZ4 source repository : https://github.com/lz4/lz4
30 * Changed for kernel usage by:
31 * Sven Schmidt <4sschmid@informatik.uni-hamburg.de>
34 /*-************************************
36 **************************************/
37 #include <linux/lz4.h>
39 #include <linux/module.h>
40 #include <linux/kernel.h>
41 #include <linux/string.h> /* memset */
43 /* *************************************
44 * Local Constants and types
45 ***************************************/
47 #define OPTIMAL_ML (int)((ML_MASK - 1) + MINMATCH)
49 #define HASH_FUNCTION(i) (((i) * 2654435761U) \
50 >> ((MINMATCH*8) - LZ4HC_HASH_LOG))
51 #define DELTANEXTU16(p) chainTable[(U16)(p)] /* faster */
53 static U32 LZ4HC_hashPtr(const void *ptr)
55 return HASH_FUNCTION(LZ4_read32(ptr));
58 /**************************************
60 **************************************/
61 static void LZ4HC_init(LZ4HC_CCtx_internal *hc4, const BYTE *start)
63 memset((void *)hc4->hashTable, 0, sizeof(hc4->hashTable));
64 memset(hc4->chainTable, 0xFF, sizeof(hc4->chainTable));
65 hc4->nextToUpdate = 64 * KB;
66 hc4->base = start - 64 * KB;
68 hc4->dictBase = start - 64 * KB;
69 hc4->dictLimit = 64 * KB;
70 hc4->lowLimit = 64 * KB;
73 /* Update chains up to ip (excluded) */
74 static FORCE_INLINE void LZ4HC_Insert(LZ4HC_CCtx_internal *hc4,
77 U16 * const chainTable = hc4->chainTable;
78 U32 * const hashTable = hc4->hashTable;
79 const BYTE * const base = hc4->base;
80 U32 const target = (U32)(ip - base);
81 U32 idx = hc4->nextToUpdate;
83 while (idx < target) {
84 U32 const h = LZ4HC_hashPtr(base + idx);
85 size_t delta = idx - hashTable[h];
87 if (delta > MAX_DISTANCE)
90 DELTANEXTU16(idx) = (U16)delta;
96 hc4->nextToUpdate = target;
99 static FORCE_INLINE int LZ4HC_InsertAndFindBestMatch(
100 LZ4HC_CCtx_internal *hc4, /* Index table will be updated */
102 const BYTE * const iLimit,
103 const BYTE **matchpos,
104 const int maxNbAttempts)
106 U16 * const chainTable = hc4->chainTable;
107 U32 * const HashTable = hc4->hashTable;
108 const BYTE * const base = hc4->base;
109 const BYTE * const dictBase = hc4->dictBase;
110 const U32 dictLimit = hc4->dictLimit;
111 const U32 lowLimit = (hc4->lowLimit + 64 * KB > (U32)(ip - base))
113 : (U32)(ip - base) - (64 * KB - 1);
115 int nbAttempts = maxNbAttempts;
118 /* HC4 match finder */
119 LZ4HC_Insert(hc4, ip);
120 matchIndex = HashTable[LZ4HC_hashPtr(ip)];
122 while ((matchIndex >= lowLimit)
125 if (matchIndex >= dictLimit) {
126 const BYTE * const match = base + matchIndex;
128 if (*(match + ml) == *(ip + ml)
129 && (LZ4_read32(match) == LZ4_read32(ip))) {
130 size_t const mlt = LZ4_count(ip + MINMATCH,
131 match + MINMATCH, iLimit) + MINMATCH;
139 const BYTE * const match = dictBase + matchIndex;
141 if (LZ4_read32(match) == LZ4_read32(ip)) {
143 const BYTE *vLimit = ip
144 + (dictLimit - matchIndex);
148 mlt = LZ4_count(ip + MINMATCH,
149 match + MINMATCH, vLimit) + MINMATCH;
150 if ((ip + mlt == vLimit)
151 && (vLimit < iLimit))
152 mlt += LZ4_count(ip + mlt,
156 /* virtual matchpos */
158 *matchpos = base + matchIndex;
162 matchIndex -= DELTANEXTU16(matchIndex);
168 static FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch(
169 LZ4HC_CCtx_internal *hc4,
170 const BYTE * const ip,
171 const BYTE * const iLowLimit,
172 const BYTE * const iHighLimit,
174 const BYTE **matchpos,
175 const BYTE **startpos,
176 const int maxNbAttempts)
178 U16 * const chainTable = hc4->chainTable;
179 U32 * const HashTable = hc4->hashTable;
180 const BYTE * const base = hc4->base;
181 const U32 dictLimit = hc4->dictLimit;
182 const BYTE * const lowPrefixPtr = base + dictLimit;
183 const U32 lowLimit = (hc4->lowLimit + 64 * KB > (U32)(ip - base))
185 : (U32)(ip - base) - (64 * KB - 1);
186 const BYTE * const dictBase = hc4->dictBase;
188 int nbAttempts = maxNbAttempts;
189 int delta = (int)(ip - iLowLimit);
192 LZ4HC_Insert(hc4, ip);
193 matchIndex = HashTable[LZ4HC_hashPtr(ip)];
195 while ((matchIndex >= lowLimit)
198 if (matchIndex >= dictLimit) {
199 const BYTE *matchPtr = base + matchIndex;
201 if (*(iLowLimit + longest)
202 == *(matchPtr - delta + longest)) {
203 if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
204 int mlt = MINMATCH + LZ4_count(
210 while ((ip + back > iLowLimit)
211 && (matchPtr + back > lowPrefixPtr)
212 && (ip[back - 1] == matchPtr[back - 1]))
219 *matchpos = matchPtr + back;
220 *startpos = ip + back;
225 const BYTE * const matchPtr = dictBase + matchIndex;
227 if (LZ4_read32(matchPtr) == LZ4_read32(ip)) {
230 const BYTE *vLimit = ip + (dictLimit - matchIndex);
232 if (vLimit > iHighLimit)
235 mlt = LZ4_count(ip + MINMATCH,
236 matchPtr + MINMATCH, vLimit) + MINMATCH;
238 if ((ip + mlt == vLimit) && (vLimit < iHighLimit))
239 mlt += LZ4_count(ip + mlt, base + dictLimit,
241 while ((ip + back > iLowLimit)
242 && (matchIndex + back > lowLimit)
243 && (ip[back - 1] == matchPtr[back - 1]))
248 if ((int)mlt > longest) {
250 *matchpos = base + matchIndex + back;
251 *startpos = ip + back;
256 matchIndex -= DELTANEXTU16(matchIndex);
262 static FORCE_INLINE int LZ4HC_encodeSequence(
267 const BYTE * const match,
268 limitedOutput_directive limitedOutputBuffer,
274 /* Encode Literal length */
275 length = (int)(*ip - *anchor);
278 if ((limitedOutputBuffer)
279 && ((*op + (length>>8)
280 + length + (2 + 1 + LASTLITERALS)) > oend)) {
281 /* Check output limit */
284 if (length >= (int)RUN_MASK) {
287 *token = (RUN_MASK<<ML_BITS);
288 len = length - RUN_MASK;
289 for (; len > 254 ; len -= 255)
291 *(*op)++ = (BYTE)len;
293 *token = (BYTE)(length<<ML_BITS);
296 LZ4_wildCopy(*op, *anchor, (*op) + length);
300 LZ4_writeLE16(*op, (U16)(*ip - match));
303 /* Encode MatchLength */
304 length = (int)(matchLength - MINMATCH);
306 if ((limitedOutputBuffer)
307 && (*op + (length>>8)
308 + (1 + LASTLITERALS) > oend)) {
309 /* Check output limit */
313 if (length >= (int)ML_MASK) {
317 for (; length > 509 ; length -= 510) {
327 *(*op)++ = (BYTE)length;
329 *token += (BYTE)(length);
331 /* Prepare next loop */
338 static int LZ4HC_compress_generic(
339 LZ4HC_CCtx_internal *const ctx,
340 const char * const source,
343 int const maxOutputSize,
344 int compressionLevel,
345 limitedOutput_directive limit
348 const BYTE *ip = (const BYTE *) source;
349 const BYTE *anchor = ip;
350 const BYTE * const iend = ip + inputSize;
351 const BYTE * const mflimit = iend - MFLIMIT;
352 const BYTE * const matchlimit = (iend - LASTLITERALS);
354 BYTE *op = (BYTE *) dest;
355 BYTE * const oend = op + maxOutputSize;
357 unsigned int maxNbAttempts;
358 int ml, ml2, ml3, ml0;
359 const BYTE *ref = NULL;
360 const BYTE *start2 = NULL;
361 const BYTE *ref2 = NULL;
362 const BYTE *start3 = NULL;
363 const BYTE *ref3 = NULL;
368 if (compressionLevel > LZ4HC_MAX_CLEVEL)
369 compressionLevel = LZ4HC_MAX_CLEVEL;
370 if (compressionLevel < 1)
371 compressionLevel = LZ4HC_DEFAULT_CLEVEL;
372 maxNbAttempts = 1 << (compressionLevel - 1);
373 ctx->end += inputSize;
378 while (ip < mflimit) {
379 ml = LZ4HC_InsertAndFindBestMatch(ctx, ip,
380 matchlimit, (&ref), maxNbAttempts);
386 /* saved, in case we would skip too much */
392 if (ip + ml < mflimit)
393 ml2 = LZ4HC_InsertAndGetWiderMatch(ctx,
395 matchlimit, ml, &ref2,
396 &start2, maxNbAttempts);
401 /* No better match */
402 if (LZ4HC_encodeSequence(&ip, &op,
403 &anchor, ml, ref, limit, oend))
409 if (start2 < ip + ml0) {
417 /* Here, start0 == ip */
418 if ((start2 - ip) < 3) {
419 /* First Match too small : removed */
428 * Currently we have :
430 * ip1 + 3 <= ip2 (usually < ip1 + ml1)
432 if ((start2 - ip) < OPTIMAL_ML) {
436 if (new_ml > OPTIMAL_ML)
438 if (ip + new_ml > start2 + ml2 - MINMATCH)
439 new_ml = (int)(start2 - ip) + ml2 - MINMATCH;
441 correction = new_ml - (int)(start2 - ip);
443 if (correction > 0) {
444 start2 += correction;
450 * Now, we have start2 = ip + new_ml,
451 * with new_ml = min(ml, OPTIMAL_ML = 18)
454 if (start2 + ml2 < mflimit)
455 ml3 = LZ4HC_InsertAndGetWiderMatch(ctx,
456 start2 + ml2 - 3, start2,
457 matchlimit, ml2, &ref3, &start3,
463 /* No better match : 2 sequences to encode */
464 /* ip & ref are known; Now for ml */
465 if (start2 < ip + ml)
466 ml = (int)(start2 - ip);
467 /* Now, encode 2 sequences */
468 if (LZ4HC_encodeSequence(&ip, &op, &anchor,
469 ml, ref, limit, oend))
472 if (LZ4HC_encodeSequence(&ip, &op, &anchor,
473 ml2, ref2, limit, oend))
478 if (start3 < ip + ml + 3) {
479 /* Not enough space for match 2 : remove it */
480 if (start3 >= (ip + ml)) {
481 /* can write Seq1 immediately
482 * ==> Seq2 is removed,
483 * so Seq3 becomes Seq1
485 if (start2 < ip + ml) {
486 int correction = (int)(ip + ml - start2);
488 start2 += correction;
491 if (ml2 < MINMATCH) {
498 if (LZ4HC_encodeSequence(&ip, &op, &anchor,
499 ml, ref, limit, oend))
518 * OK, now we have 3 ascending matches;
519 * let's write at least the first one
520 * ip & ref are known; Now for ml
522 if (start2 < ip + ml) {
523 if ((start2 - ip) < (int)ML_MASK) {
528 if (ip + ml > start2 + ml2 - MINMATCH)
529 ml = (int)(start2 - ip) + ml2 - MINMATCH;
530 correction = ml - (int)(start2 - ip);
531 if (correction > 0) {
532 start2 += correction;
537 ml = (int)(start2 - ip);
539 if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml,
554 /* Encode Last Literals */
556 int lastRun = (int)(iend - anchor);
559 && (((char *)op - dest) + lastRun + 1
560 + ((lastRun + 255 - RUN_MASK)/255)
561 > (U32)maxOutputSize)) {
562 /* Check output limit */
565 if (lastRun >= (int)RUN_MASK) {
566 *op++ = (RUN_MASK<<ML_BITS);
568 for (; lastRun > 254 ; lastRun -= 255)
570 *op++ = (BYTE) lastRun;
572 *op++ = (BYTE)(lastRun<<ML_BITS);
573 LZ4_memcpy(op, anchor, iend - anchor);
578 return (int) (((char *)op) - dest);
581 static int LZ4_compress_HC_extStateHC(
587 int compressionLevel)
589 LZ4HC_CCtx_internal *ctx = &((LZ4_streamHC_t *)state)->internal_donotuse;
591 if (((size_t)(state)&(sizeof(void *) - 1)) != 0) {
592 /* Error : state is not aligned
593 * for pointers (32 or 64 bits)
598 LZ4HC_init(ctx, (const BYTE *)src);
600 if (maxDstSize < LZ4_compressBound(srcSize))
601 return LZ4HC_compress_generic(ctx, src, dst,
602 srcSize, maxDstSize, compressionLevel, limitedOutput);
604 return LZ4HC_compress_generic(ctx, src, dst,
605 srcSize, maxDstSize, compressionLevel, noLimit);
608 int LZ4_compress_HC(const char *src, char *dst, int srcSize,
609 int maxDstSize, int compressionLevel, void *wrkmem)
611 return LZ4_compress_HC_extStateHC(wrkmem, src, dst,
612 srcSize, maxDstSize, compressionLevel);
614 EXPORT_SYMBOL(LZ4_compress_HC);
616 /**************************************
617 * Streaming Functions
618 **************************************/
619 void LZ4_resetStreamHC(LZ4_streamHC_t *LZ4_streamHCPtr, int compressionLevel)
621 LZ4_streamHCPtr->internal_donotuse.base = NULL;
622 LZ4_streamHCPtr->internal_donotuse.compressionLevel = (unsigned int)compressionLevel;
625 int LZ4_loadDictHC(LZ4_streamHC_t *LZ4_streamHCPtr,
626 const char *dictionary,
629 LZ4HC_CCtx_internal *ctxPtr = &LZ4_streamHCPtr->internal_donotuse;
631 if (dictSize > 64 * KB) {
632 dictionary += dictSize - 64 * KB;
635 LZ4HC_init(ctxPtr, (const BYTE *)dictionary);
637 LZ4HC_Insert(ctxPtr, (const BYTE *)dictionary + (dictSize - 3));
638 ctxPtr->end = (const BYTE *)dictionary + dictSize;
641 EXPORT_SYMBOL(LZ4_loadDictHC);
645 static void LZ4HC_setExternalDict(
646 LZ4HC_CCtx_internal *ctxPtr,
647 const BYTE *newBlock)
649 if (ctxPtr->end >= ctxPtr->base + 4) {
650 /* Referencing remaining dictionary content */
651 LZ4HC_Insert(ctxPtr, ctxPtr->end - 3);
655 * Only one memory segment for extDict,
656 * so any previous extDict is lost at this stage
658 ctxPtr->lowLimit = ctxPtr->dictLimit;
659 ctxPtr->dictLimit = (U32)(ctxPtr->end - ctxPtr->base);
660 ctxPtr->dictBase = ctxPtr->base;
661 ctxPtr->base = newBlock - ctxPtr->dictLimit;
662 ctxPtr->end = newBlock;
663 /* match referencing will resume from there */
664 ctxPtr->nextToUpdate = ctxPtr->dictLimit;
667 static int LZ4_compressHC_continue_generic(
668 LZ4_streamHC_t *LZ4_streamHCPtr,
673 limitedOutput_directive limit)
675 LZ4HC_CCtx_internal *ctxPtr = &LZ4_streamHCPtr->internal_donotuse;
677 /* auto - init if forgotten */
678 if (ctxPtr->base == NULL)
679 LZ4HC_init(ctxPtr, (const BYTE *) source);
682 if ((size_t)(ctxPtr->end - ctxPtr->base) > 2 * GB) {
683 size_t dictSize = (size_t)(ctxPtr->end - ctxPtr->base)
685 if (dictSize > 64 * KB)
687 LZ4_loadDictHC(LZ4_streamHCPtr,
688 (const char *)(ctxPtr->end) - dictSize, (int)dictSize);
691 /* Check if blocks follow each other */
692 if ((const BYTE *)source != ctxPtr->end)
693 LZ4HC_setExternalDict(ctxPtr, (const BYTE *)source);
695 /* Check overlapping input/dictionary space */
697 const BYTE *sourceEnd = (const BYTE *) source + inputSize;
698 const BYTE * const dictBegin = ctxPtr->dictBase + ctxPtr->lowLimit;
699 const BYTE * const dictEnd = ctxPtr->dictBase + ctxPtr->dictLimit;
701 if ((sourceEnd > dictBegin)
702 && ((const BYTE *)source < dictEnd)) {
703 if (sourceEnd > dictEnd)
705 ctxPtr->lowLimit = (U32)(sourceEnd - ctxPtr->dictBase);
707 if (ctxPtr->dictLimit - ctxPtr->lowLimit < 4)
708 ctxPtr->lowLimit = ctxPtr->dictLimit;
712 return LZ4HC_compress_generic(ctxPtr, source, dest,
713 inputSize, maxOutputSize, ctxPtr->compressionLevel, limit);
716 int LZ4_compress_HC_continue(
717 LZ4_streamHC_t *LZ4_streamHCPtr,
723 if (maxOutputSize < LZ4_compressBound(inputSize))
724 return LZ4_compressHC_continue_generic(LZ4_streamHCPtr,
725 source, dest, inputSize, maxOutputSize, limitedOutput);
727 return LZ4_compressHC_continue_generic(LZ4_streamHCPtr,
728 source, dest, inputSize, maxOutputSize, noLimit);
730 EXPORT_SYMBOL(LZ4_compress_HC_continue);
732 /* dictionary saving */
735 LZ4_streamHC_t *LZ4_streamHCPtr,
739 LZ4HC_CCtx_internal *const streamPtr = &LZ4_streamHCPtr->internal_donotuse;
740 int const prefixSize = (int)(streamPtr->end
741 - (streamPtr->base + streamPtr->dictLimit));
743 if (dictSize > 64 * KB)
747 if (dictSize > prefixSize)
748 dictSize = prefixSize;
750 memmove(safeBuffer, streamPtr->end - dictSize, dictSize);
753 U32 const endIndex = (U32)(streamPtr->end - streamPtr->base);
755 streamPtr->end = (const BYTE *)safeBuffer + dictSize;
756 streamPtr->base = streamPtr->end - endIndex;
757 streamPtr->dictLimit = endIndex - dictSize;
758 streamPtr->lowLimit = endIndex - dictSize;
760 if (streamPtr->nextToUpdate < streamPtr->dictLimit)
761 streamPtr->nextToUpdate = streamPtr->dictLimit;
765 EXPORT_SYMBOL(LZ4_saveDictHC);
767 MODULE_LICENSE("Dual BSD/GPL");
768 MODULE_DESCRIPTION("LZ4 HC compressor");