danog/Telegram
1
0
Fork 0
mirror of https://github.com/danog/Telegram.git synced 2024-12-04 10:27:46 +01:00
Code Issues Projects Releases Wiki Activity
8da5528cc5
Telegram/thirdparty/lz4/lz4hc.c
Peter 316bfd9598 no message
2015-10-01 19:19:52 +03:00

1010 lines
35 KiB
C
Raw Blame History

/*
LZ4 HC - High Compression Mode of LZ4
Copyright (C) 2011-2014, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
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.
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.
You can contact the author at :
- LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
- LZ4 source repository : http://code.google.com/p/lz4/
*/
/**************************************
Tuning Parameter
**************************************/
#define LZ4HC_DEFAULT_COMPRESSIONLEVEL 8
/**************************************
Memory routines
**************************************/
#include <stdlib.h> /* calloc, free */
#define ALLOCATOR(s) calloc(1,s)
#define FREEMEM free
#include <string.h> /* memset, memcpy */
#define MEM_INIT memset
/**************************************
CPU Feature Detection
**************************************/
/* 32 or 64 bits ? */
#if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
|| defined(__64BIT__) || defined(__mips64) \
|| defined(__powerpc64__) || defined(__powerpc64le__) \
|| defined(__ppc64__) || defined(__ppc64le__) \
|| defined(__PPC64__) || defined(__PPC64LE__) \
|| defined(__ia64) || defined(__itanium__) || defined(_M_IA64) \
|| defined(__s390x__) ) /* Detects 64 bits mode */
# define LZ4_ARCH64 1
#else
# define LZ4_ARCH64 0
#endif
/*
* Little Endian or Big Endian ?
* Overwrite the #define below if you know your architecture endianess
*/
#include <stdlib.h> /* Apparently required to detect endianess */
#if defined (__GLIBC__)
# include <endian.h>
# if (__BYTE_ORDER == __BIG_ENDIAN)
# define LZ4_BIG_ENDIAN 1
# endif
#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))
# define LZ4_BIG_ENDIAN 1
#elif defined(__sparc) || defined(__sparc__) \
|| defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \
|| defined(__hpux) || defined(__hppa) \
|| defined(_MIPSEB) || defined(__s390__)
# define LZ4_BIG_ENDIAN 1
#else
/* Little Endian assumed. PDP Endian and other very rare endian format are unsupported. */
#endif
/*
* Unaligned memory access is automatically enabled for "common" CPU, such as x86.
* For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected
* If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance
*/
#if defined(__ARM_FEATURE_UNALIGNED)
# define LZ4_FORCE_UNALIGNED_ACCESS 1
#endif
/* Define this parameter if your target system or compiler does not support hardware bit count */
#if defined(_MSC_VER) && defined(_WIN32_WCE) /* Visual Studio for Windows CE does not support Hardware bit count */
# define LZ4_FORCE_SW_BITCOUNT
#endif
/**************************************
Compiler Options
**************************************/
#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */
/* "restrict" is a known keyword */
#else
# define restrict /* Disable restrict */
#endif
#ifdef _MSC_VER /* Visual Studio */
# define FORCE_INLINE static __forceinline
# include <intrin.h> /* For Visual 2005 */
# if LZ4_ARCH64 /* 64-bits */
# pragma intrinsic(_BitScanForward64) /* For Visual 2005 */
# pragma intrinsic(_BitScanReverse64) /* For Visual 2005 */
# else /* 32-bits */
# pragma intrinsic(_BitScanForward) /* For Visual 2005 */
# pragma intrinsic(_BitScanReverse) /* For Visual 2005 */
# endif
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# pragma warning(disable : 4701) /* disable: C4701: potentially uninitialized local variable used */
#else
# ifdef __GNUC__
# define FORCE_INLINE static inline __attribute__((always_inline))
# else
# define FORCE_INLINE static inline
# endif
#endif
#ifdef _MSC_VER /* Visual Studio */
# define lz4_bswap16(x) _byteswap_ushort(x)
#else
# define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
#endif
/**************************************
Includes
**************************************/
#include "lz4hc.h"
#include "lz4.h"
/**************************************
Basic Types
**************************************/
#if defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */
# include <stdint.h>
typedef uint8_t BYTE;
typedef uint16_t U16;
typedef uint32_t U32;
typedef int32_t S32;
typedef uint64_t U64;
#else
typedef unsigned char BYTE;
typedef unsigned short U16;
typedef unsigned int U32;
typedef signed int S32;
typedef unsigned long long U64;
#endif
#if defined(__GNUC__) && !defined(LZ4_FORCE_UNALIGNED_ACCESS)
# define _PACKED __attribute__ ((packed))
#else
# define _PACKED
#endif
#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
# ifdef __IBMC__
# pragma pack(1)
# else
# pragma pack(push, 1)
# endif
#endif
typedef struct _U16_S { U16 v; } _PACKED U16_S;
typedef struct _U32_S { U32 v; } _PACKED U32_S;
typedef struct _U64_S { U64 v; } _PACKED U64_S;
#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
# pragma pack(pop)
#endif
#define A64(x) (((U64_S *)(x))->v)
#define A32(x) (((U32_S *)(x))->v)
#define A16(x) (((U16_S *)(x))->v)
/**************************************
Constants
**************************************/
#define MINMATCH 4
#define DICTIONARY_LOGSIZE 16
#define MAXD (1<<DICTIONARY_LOGSIZE)
#define MAXD_MASK ((U32)(MAXD - 1))
#define MAX_DISTANCE (MAXD - 1)
#define HASH_LOG (DICTIONARY_LOGSIZE-1)
#define HASHTABLESIZE (1 << HASH_LOG)
#define HASH_MASK (HASHTABLESIZE - 1)
#define ML_BITS 4
#define ML_MASK (size_t)((1U<<ML_BITS)-1)
#define RUN_BITS (8-ML_BITS)
#define RUN_MASK ((1U<<RUN_BITS)-1)
#define COPYLENGTH 8
#define LASTLITERALS 5
#define MFLIMIT (COPYLENGTH+MINMATCH)
#define MINLENGTH (MFLIMIT+1)
#define OPTIMAL_ML (int)((ML_MASK-1)+MINMATCH)
#define KB *(1<<10)
#define MB *(1<<20)
#define GB *(1U<<30)
/**************************************
Architecture-specific macros
**************************************/
#if LZ4_ARCH64 /* 64-bit */
# define STEPSIZE 8
# define LZ4_COPYSTEP(s,d) A64(d) = A64(s); d+=8; s+=8;
# define LZ4_COPYPACKET(s,d) LZ4_COPYSTEP(s,d)
# define AARCH A64
#else /* 32-bit */
# define STEPSIZE 4
# define LZ4_COPYSTEP(s,d) A32(d) = A32(s); d+=4; s+=4;
# define LZ4_COPYPACKET(s,d) LZ4_COPYSTEP(s,d); LZ4_COPYSTEP(s,d);
# define AARCH A32
#endif
#if defined(LZ4_BIG_ENDIAN)
# define LZ4_READ_LITTLEENDIAN_16(d,s,p) { U16 v = A16(p); v = lz4_bswap16(v); d = (s) - v; }
# define LZ4_WRITE_LITTLEENDIAN_16(p,i) { U16 v = (U16)(i); v = lz4_bswap16(v); A16(p) = v; p+=2; }
#else /* Little Endian */
# define LZ4_READ_LITTLEENDIAN_16(d,s,p) { d = (s) - A16(p); }
# define LZ4_WRITE_LITTLEENDIAN_16(p,v) { A16(p) = v; p+=2; }
#endif
/**************************************
Local Types
**************************************/
typedef struct
{
U32 hashTable[HASHTABLESIZE];
U16 chainTable[MAXD];
const BYTE* end; /* next block here to keep current prefix as prefix */
const BYTE* base; /* All index relative to this position */
const BYTE* dictBase; /* alternate base for extDict */
U32 dictLimit; /* below that point, need extDict */
U32 lowLimit; /* below that point, no more dict */
U32 nextToUpdate;
U32 compressionLevel;
const BYTE* inputBuffer; /* deprecated */
} LZ4HC_Data_Structure;
/**************************************
Macros
**************************************/
#define LZ4_STATIC_ASSERT(c) { enum { LZ4_static_assert = 1/(!!(c)) }; } /* Visual : use only *after* variable declarations */
#define LZ4_WILDCOPY(s,d,e) do { LZ4_COPYPACKET(s,d) } while (d<e);
#define LZ4_BLINDCOPY(s,d,l) { BYTE* e=d+l; LZ4_WILDCOPY(s,d,e); d=e; }
#define HASH_FUNCTION(i) (((i) * 2654435761U) >> ((MINMATCH*8)-HASH_LOG))
#define DELTANEXT(p) chainTable[(size_t)(p) & MAXD_MASK]
#define GETNEXT(p) ((p) - (size_t)DELTANEXT(p))
static U32 LZ4HC_hashPtr(const void* ptr) { return HASH_FUNCTION(A32(ptr)); }
/**************************************
Private functions
**************************************/
#if LZ4_ARCH64
FORCE_INLINE int LZ4_NbCommonBytes (register U64 val)
{
#if defined(LZ4_BIG_ENDIAN)
# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r = 0;
_BitScanReverse64( &r, val );
return (int)(r>>3);
# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
return (__builtin_clzll(val) >> 3);
# else
int r;
if (!(val>>32)) { r=4; } else { r=0; val>>=32; }
if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
r += (!val);
return r;
# endif
#else
# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r = 0;
_BitScanForward64( &r, val );
return (int)(r>>3);
# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
return (__builtin_ctzll(val) >> 3);
# else
static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 };
return DeBruijnBytePos[((U64)((val & -val) * 0x0218A392CDABBD3F)) >> 58];
# endif
#endif
}
#else
FORCE_INLINE int LZ4_NbCommonBytes (register U32 val)
{
#if defined(LZ4_BIG_ENDIAN)
# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r;
_BitScanReverse( &r, val );
return (int)(r>>3);
# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
return (__builtin_clz(val) >> 3);
# else
int r;
if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
r += (!val);
return r;
# endif
#else
# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
unsigned long r;
_BitScanForward( &r, val );
return (int)(r>>3);
# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
return (__builtin_ctz(val) >> 3);
# else
static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
# endif
#endif
}
#endif
static void LZ4HC_init (LZ4HC_Data_Structure* hc4, const BYTE* base)
{
MEM_INIT((void*)hc4->hashTable, 0, sizeof(hc4->hashTable));
MEM_INIT(hc4->chainTable, 0xFF, sizeof(hc4->chainTable));
hc4->nextToUpdate = 64 KB;
hc4->base = base - 64 KB;
hc4->inputBuffer = base;
hc4->end = base;
hc4->dictBase = base - 64 KB;
hc4->dictLimit = 64 KB;
hc4->lowLimit = 64 KB;
}
/* Update chains up to ip (excluded) */
FORCE_INLINE void LZ4HC_Insert (LZ4HC_Data_Structure* hc4, const BYTE* ip)
{
U16* chainTable = hc4->chainTable;
U32* HashTable = hc4->hashTable;
const BYTE* const base = hc4->base;
const U32 target = (U32)(ip - base);
U32 idx = hc4->nextToUpdate;
while(idx < target)
{
U32 h = LZ4HC_hashPtr(base+idx);
size_t delta = idx - HashTable[h];
if (delta>MAX_DISTANCE) delta = MAX_DISTANCE;
chainTable[idx & 0xFFFF] = (U16)delta;
HashTable[h] = idx;
idx++;
}
hc4->nextToUpdate = target;
}
static void LZ4HC_setExternalDict(LZ4HC_Data_Structure* ctxPtr, const BYTE* newBlock)
{
if (ctxPtr->end >= ctxPtr->base + 4)
LZ4HC_Insert (ctxPtr, ctxPtr->end-3); // finish referencing dictionary content
// Note : need to handle risk of index overflow
// Use only one memory segment for dict, so any previous External Dict is lost at this stage
ctxPtr->lowLimit = ctxPtr->dictLimit;
ctxPtr->dictLimit = (U32)(ctxPtr->end - ctxPtr->base);
ctxPtr->dictBase = ctxPtr->base;
ctxPtr->base = newBlock - ctxPtr->dictLimit;
ctxPtr->end = newBlock;
ctxPtr->nextToUpdate = ctxPtr->dictLimit; // reference table must skip to from beginning of block
}
static size_t LZ4HC_CommonLength (const BYTE* p1, const BYTE* p2, const BYTE* const p1Limit)
{
const BYTE* const p1Start = p1;
while (p1 <= p1Limit - STEPSIZE)
{
size_t diff = AARCH(p2) ^ AARCH(p1);
if (!diff) { p1+=STEPSIZE; p2+=STEPSIZE; continue; }
p1 += LZ4_NbCommonBytes(diff);
return (p1 - p1Start);
}
if (LZ4_ARCH64) if ((p1<(p1Limit-3)) && (A32(p2) == A32(p1))) { p1+=4; p2+=4; }
if ((p1<(p1Limit-1)) && (A16(p2) == A16(p1))) { p1+=2; p2+=2; }
if ((p1<p1Limit) && (*p2 == *p1)) p1++;
return (p1 - p1Start);
}
FORCE_INLINE int LZ4HC_InsertAndFindBestMatch (LZ4HC_Data_Structure* hc4, // Index table will be updated
const BYTE* ip, const BYTE* const iLimit,
const BYTE** matchpos,
const int maxNbAttempts)
{
U16* const chainTable = hc4->chainTable;
U32* const HashTable = hc4->hashTable;
const BYTE* const base = hc4->base;
const BYTE* const dictBase = hc4->dictBase;
const U32 dictLimit = hc4->dictLimit;
const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1);
U32 matchIndex;
const BYTE* match;
int nbAttempts=maxNbAttempts;
size_t ml=0;
/* HC4 match finder */
LZ4HC_Insert(hc4, ip);
matchIndex = HashTable[LZ4HC_hashPtr(ip)];
while ((matchIndex>=lowLimit) && (nbAttempts))
{
nbAttempts--;
if (matchIndex >= dictLimit)
{
match = base + matchIndex;
if (*(match+ml) == *(ip+ml)
&& (A32(match) == A32(ip)))
{
size_t mlt = LZ4HC_CommonLength(ip+MINMATCH, match+MINMATCH, iLimit) + MINMATCH;
if (mlt > ml) { ml = mlt; *matchpos = match; }
}
}
else
{
match = dictBase + matchIndex;
if (A32(match) == A32(ip))
{
size_t mlt;
const BYTE* vLimit = ip + (dictLimit - matchIndex);
if (vLimit > iLimit) vLimit = iLimit;
mlt = LZ4HC_CommonLength(ip+MINMATCH, match+MINMATCH, vLimit) + MINMATCH;
if ((ip+mlt == vLimit) && (vLimit < iLimit))
mlt += LZ4HC_CommonLength(ip+mlt, base+dictLimit, iLimit);
if (mlt > ml) { ml = mlt; *matchpos = base + matchIndex; } // virtual matchpos
}
}
matchIndex -= chainTable[matchIndex & 0xFFFF];
}
return (int)ml;
}
FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (
LZ4HC_Data_Structure* hc4,
const BYTE* ip,
const BYTE* iLowLimit,
const BYTE* iHighLimit,
int longest,
const BYTE** matchpos,
const BYTE** startpos,
const int maxNbAttempts)
{
U16* const chainTable = hc4->chainTable;
U32* const HashTable = hc4->hashTable;
const BYTE* const base = hc4->base;
const U32 dictLimit = hc4->dictLimit;
const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1);
const BYTE* const dictBase = hc4->dictBase;
const BYTE* match;
U32 matchIndex;
int nbAttempts = maxNbAttempts;
int delta = (int)(ip-iLowLimit);
/* First Match */
LZ4HC_Insert(hc4, ip);
matchIndex = HashTable[LZ4HC_hashPtr(ip)];
while ((matchIndex>=lowLimit) && (nbAttempts))
{
nbAttempts--;
if (matchIndex >= dictLimit)
{
match = base + matchIndex;
if (*(iLowLimit + longest) == *(match - delta + longest))
if (A32(match) == A32(ip))
{
const BYTE* startt = ip;
const BYTE* tmpMatch = match;
const BYTE* const matchEnd = ip + MINMATCH + LZ4HC_CommonLength(ip+MINMATCH, match+MINMATCH, iHighLimit);
while ((startt>iLowLimit) && (tmpMatch > iLowLimit) && (startt[-1] == tmpMatch[-1])) {startt--; tmpMatch--;}
if ((matchEnd-startt) > longest)
{
longest = (int)(matchEnd-startt);
*matchpos = tmpMatch;
*startpos = startt;
}
}
}
else
{
match = dictBase + matchIndex;
if (A32(match) == A32(ip))
{
size_t mlt;
int back=0;
const BYTE* vLimit = ip + (dictLimit - matchIndex);
if (vLimit > iHighLimit) vLimit = iHighLimit;
mlt = LZ4HC_CommonLength(ip+MINMATCH, match+MINMATCH, vLimit) + MINMATCH;
if ((ip+mlt == vLimit) && (vLimit < iHighLimit))
mlt += LZ4HC_CommonLength(ip+mlt, base+dictLimit, iHighLimit);
while ((ip+back > iLowLimit) && (matchIndex+back > lowLimit) && (ip[back-1] == match[back-1])) back--;
mlt -= back;
if ((int)mlt > longest) { longest = (int)mlt; *matchpos = base + matchIndex + back; *startpos = ip+back; }
}
}
matchIndex -= chainTable[matchIndex & 0xFFFF];
}
return longest;
}
typedef enum { noLimit = 0, limitedOutput = 1 } limitedOutput_directive;
//static unsigned debug = 0;
FORCE_INLINE int LZ4HC_encodeSequence (
const BYTE** ip,
BYTE** op,
const BYTE** anchor,
int matchLength,
const BYTE* const match,
limitedOutput_directive limitedOutputBuffer,
BYTE* oend)
{
int length;
BYTE* token;
//if (debug) printf("literal : %u -- match : %u -- offset : %u\n", (U32)(*ip - *anchor), (U32)matchLength, (U32)(*ip-match)); // debug
/* Encode Literal length */
length = (int)(*ip - *anchor);
token = (*op)++;
if ((limitedOutputBuffer) && ((*op + (length>>8) + length + (2 + 1 + LASTLITERALS)) > oend)) return 1; /* Check output limit */
if (length>=(int)RUN_MASK) { int len; *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *(*op)++ = 255; *(*op)++ = (BYTE)len; }
else *token = (BYTE)(length<<ML_BITS);
/* Copy Literals */
LZ4_BLINDCOPY(*anchor, *op, length);
/* Encode Offset */
LZ4_WRITE_LITTLEENDIAN_16(*op,(U16)(*ip-match));
/* Encode MatchLength */
length = (int)(matchLength-MINMATCH);
if ((limitedOutputBuffer) && (*op + (length>>8) + (1 + LASTLITERALS) > oend)) return 1; /* Check output limit */
if (length>=(int)ML_MASK) { *token+=ML_MASK; length-=ML_MASK; for(; length > 509 ; length-=510) { *(*op)++ = 255; *(*op)++ = 255; } if (length > 254) { length-=255; *(*op)++ = 255; } *(*op)++ = (BYTE)length; }
else *token += (BYTE)(length);
/* Prepare next loop */
*ip += matchLength;
*anchor = *ip;
return 0;
}
#define MAX_COMPRESSION_LEVEL 16
static int LZ4HC_compress_generic (
void* ctxvoid,
const char* source,
char* dest,
int inputSize,
int maxOutputSize,
int compressionLevel,
limitedOutput_directive limit
)
{
LZ4HC_Data_Structure* ctx = (LZ4HC_Data_Structure*) ctxvoid;
const BYTE* ip = (const BYTE*) source;
const BYTE* anchor = ip;
const BYTE* const iend = ip + inputSize;
const BYTE* const mflimit = iend - MFLIMIT;
const BYTE* const matchlimit = (iend - LASTLITERALS);
BYTE* op = (BYTE*) dest;
BYTE* const oend = op + maxOutputSize;
unsigned maxNbAttempts;
int ml, ml2, ml3, ml0;
const BYTE* ref=NULL;
const BYTE* start2=NULL;
const BYTE* ref2=NULL;
const BYTE* start3=NULL;
const BYTE* ref3=NULL;
const BYTE* start0;
const BYTE* ref0;
/* init */
if (compressionLevel > MAX_COMPRESSION_LEVEL) compressionLevel = MAX_COMPRESSION_LEVEL;
if (compressionLevel == 0) compressionLevel = LZ4HC_DEFAULT_COMPRESSIONLEVEL;
maxNbAttempts = 1 << compressionLevel;
ctx->end += inputSize;
ip++;
/* Main Loop */
while (ip < mflimit)
{
ml = LZ4HC_InsertAndFindBestMatch (ctx, ip, matchlimit, (&ref), maxNbAttempts);
if (!ml) { ip++; continue; }
/* saved, in case we would skip too much */
start0 = ip;
ref0 = ref;
ml0 = ml;
_Search2:
if (ip+ml < mflimit)
ml2 = LZ4HC_InsertAndGetWiderMatch(ctx, ip + ml - 2, ip + 1, matchlimit, ml, &ref2, &start2, maxNbAttempts);
else ml2 = ml;
if (ml2 == ml) /* No better match */
{
if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
continue;
}
if (start0 < ip)
{
if (start2 < ip + ml0) /* empirical */
{
ip = start0;
ref = ref0;
ml = ml0;
}
}
/* Here, start0==ip */
if ((start2 - ip) < 3) /* First Match too small : removed */
{
ml = ml2;
ip = start2;
ref =ref2;
goto _Search2;
}
_Search3:
/*
* Currently we have :
* ml2 > ml1, and
* ip1+3 <= ip2 (usually < ip1+ml1)
*/
if ((start2 - ip) < OPTIMAL_ML)
{
int correction;
int new_ml = ml;
if (new_ml > OPTIMAL_ML) new_ml = OPTIMAL_ML;
if (ip+new_ml > start2 + ml2 - MINMATCH) new_ml = (int)(start2 - ip) + ml2 - MINMATCH;
correction = new_ml - (int)(start2 - ip);
if (correction > 0)
{
start2 += correction;
ref2 += correction;
ml2 -= correction;
}
}
/* Now, we have start2 = ip+new_ml, with new_ml = min(ml, OPTIMAL_ML=18) */
if (start2 + ml2 < mflimit)
ml3 = LZ4HC_InsertAndGetWiderMatch(ctx, start2 + ml2 - 3, start2, matchlimit, ml2, &ref3, &start3, maxNbAttempts);
else ml3 = ml2;
if (ml3 == ml2) /* No better match : 2 sequences to encode */
{
/* ip & ref are known; Now for ml */
if (start2 < ip+ml) ml = (int)(start2 - ip);
/* Now, encode 2 sequences */
if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
ip = start2;
if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml2, ref2, limit, oend)) return 0;
continue;
}
if (start3 < ip+ml+3) /* Not enough space for match 2 : remove it */
{
if (start3 >= (ip+ml)) /* can write Seq1 immediately ==> Seq2 is removed, so Seq3 becomes Seq1 */
{
if (start2 < ip+ml)
{
int correction = (int)(ip+ml - start2);
start2 += correction;
ref2 += correction;
ml2 -= correction;
if (ml2 < MINMATCH)
{
start2 = start3;
ref2 = ref3;
ml2 = ml3;
}
}
if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
ip = start3;
ref = ref3;
ml = ml3;
start0 = start2;
ref0 = ref2;
ml0 = ml2;
goto _Search2;
}
start2 = start3;
ref2 = ref3;
ml2 = ml3;
goto _Search3;
}
/*
* OK, now we have 3 ascending matches; let's write at least the first one
* ip & ref are known; Now for ml
*/
if (start2 < ip+ml)
{
if ((start2 - ip) < (int)ML_MASK)
{
int correction;
if (ml > OPTIMAL_ML) ml = OPTIMAL_ML;
if (ip + ml > start2 + ml2 - MINMATCH) ml = (int)(start2 - ip) + ml2 - MINMATCH;
correction = ml - (int)(start2 - ip);
if (correction > 0)
{
start2 += correction;
ref2 += correction;
ml2 -= correction;
}
}
else
{
ml = (int)(start2 - ip);
}
}
if (LZ4HC_encodeSequence(&ip, &op, &anchor, ml, ref, limit, oend)) return 0;
ip = start2;
ref = ref2;
ml = ml2;
start2 = start3;
ref2 = ref3;
ml2 = ml3;
goto _Search3;
}
/* Encode Last Literals */
{
int lastRun = (int)(iend - anchor);
if ((limit) && (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize)) return 0; /* Check output limit */
if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun > 254 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
else *op++ = (BYTE)(lastRun<<ML_BITS);
memcpy(op, anchor, iend - anchor);
op += iend-anchor;
}
/* End */
return (int) (((char*)op)-dest);
}
int LZ4_compressHC2(const char* source, char* dest, int inputSize, int compressionLevel)
{
LZ4HC_Data_Structure ctx;
LZ4HC_init(&ctx, (const BYTE*)source);
return LZ4HC_compress_generic (&ctx, source, dest, inputSize, 0, compressionLevel, noLimit);
}
int LZ4_compressHC(const char* source, char* dest, int inputSize) { return LZ4_compressHC2(source, dest, inputSize, 0); }
int LZ4_compressHC2_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel)
{
LZ4HC_Data_Structure ctx;
LZ4HC_init(&ctx, (const BYTE*)source);
return LZ4HC_compress_generic (&ctx, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput);
}
int LZ4_compressHC_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
{
return LZ4_compressHC2_limitedOutput(source, dest, inputSize, maxOutputSize, 0);
}
/*****************************
Using external allocation
*****************************/
int LZ4_sizeofStateHC(void) { return sizeof(LZ4HC_Data_Structure); }
int LZ4_compressHC2_withStateHC (void* state, const char* source, char* dest, int inputSize, int compressionLevel)
{
if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */
LZ4HC_init ((LZ4HC_Data_Structure*)state, (const BYTE*)source);
return LZ4HC_compress_generic (state, source, dest, inputSize, 0, compressionLevel, noLimit);
}
int LZ4_compressHC_withStateHC (void* state, const char* source, char* dest, int inputSize)
{ return LZ4_compressHC2_withStateHC (state, source, dest, inputSize, 0); }
int LZ4_compressHC2_limitedOutput_withStateHC (void* state, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel)
{
if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */
LZ4HC_init ((LZ4HC_Data_Structure*)state, (const BYTE*)source);
return LZ4HC_compress_generic (state, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput);
}
int LZ4_compressHC_limitedOutput_withStateHC (void* state, const char* source, char* dest, int inputSize, int maxOutputSize)
{ return LZ4_compressHC2_limitedOutput_withStateHC (state, source, dest, inputSize, maxOutputSize, 0); }
/**************************************
Experimental Streaming Functions
**************************************/
/* allocation */
LZ4_streamHC_t* LZ4_createStreamHC(void) { return (LZ4_streamHC_t*)malloc(sizeof(LZ4_streamHC_t)); }
int LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr) { free(LZ4_streamHCPtr); return 0; };
/* initialization */
void LZ4_resetStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel)
{
LZ4_STATIC_ASSERT(sizeof(LZ4HC_Data_Structure) <= LZ4_STREAMHCSIZE); /* if compilation fails here, LZ4_STREAMHCSIZE must be increased */
((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->base = NULL;
((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->compressionLevel = (unsigned)compressionLevel;
}
int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, const char* dictionary, int dictSize)
{
LZ4HC_Data_Structure* streamPtr = (LZ4HC_Data_Structure*) LZ4_streamHCPtr;
if (dictSize > 64 KB)
{
dictionary += dictSize - 64 KB;
dictSize = 64 KB;
}
LZ4HC_init (streamPtr, (const BYTE*)dictionary);
if (dictSize >= 4) LZ4HC_Insert (streamPtr, (const BYTE*)dictionary +(dictSize-3));
streamPtr->end = (const BYTE*)dictionary + dictSize;
return dictSize;
}
/* compression */
static int LZ4_compressHC_continue_generic (LZ4HC_Data_Structure* dsPtr,
const char* source, char* dest,
int inputSize, int maxOutputSize, limitedOutput_directive limit)
{
/* auto-init if forgotten */
if (dsPtr->base == NULL)
LZ4HC_init (dsPtr, (const BYTE*) source);
/* Check overflow */
if ((size_t)(dsPtr->end - dsPtr->base) > 2 GB)
{
size_t dictSize = (size_t)(dsPtr->end - dsPtr->base) - dsPtr->dictLimit;
if (dictSize > 64 KB) dictSize = 64 KB;
LZ4_loadDictHC((LZ4_streamHC_t*)dsPtr, (const char*)(dsPtr->end) - dictSize, (int)dictSize);
}
/* Check if blocks follow each other */
if ((const BYTE*)source != dsPtr->end) LZ4HC_setExternalDict(dsPtr, (const BYTE*)source);
/* Check overlapping input/dictionary space */
{
const BYTE* sourceEnd = (const BYTE*) source + inputSize;
const BYTE* dictBegin = dsPtr->dictBase + dsPtr->lowLimit;
const BYTE* dictEnd = dsPtr->dictBase + dsPtr->dictLimit;
if ((sourceEnd > dictBegin) && ((BYTE*)source < dictEnd))
{
if (sourceEnd > dictEnd) sourceEnd = dictEnd;
dsPtr->lowLimit = (U32)(sourceEnd - dsPtr->dictBase);
if (dsPtr->dictLimit - dsPtr->lowLimit < 4) dsPtr->lowLimit = dsPtr->dictLimit;
}
}
return LZ4HC_compress_generic (dsPtr, source, dest, inputSize, maxOutputSize, dsPtr->compressionLevel, limit);
}
int LZ4_compressHC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize)
{
return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, 0, noLimit);
}
int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize, int maxOutputSize)
{
return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, maxOutputSize, limitedOutput);
}
/* dictionary saving */
int LZ4_saveDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, char* safeBuffer, int dictSize)
{
LZ4HC_Data_Structure* streamPtr = (LZ4HC_Data_Structure*)LZ4_streamHCPtr;
int prefixSize = (int)(streamPtr->end - (streamPtr->base + streamPtr->dictLimit));
if (dictSize > 64 KB) dictSize = 64 KB;
if (dictSize < 4) dictSize = 0;
if (dictSize > prefixSize) dictSize = prefixSize;
memcpy(safeBuffer, streamPtr->end - dictSize, dictSize);
//LZ4_loadDictHC(LZ4_streamHCPtr, safeBuffer, dictSize);
{
U32 endIndex = (U32)(streamPtr->end - streamPtr->base);
streamPtr->end = (const BYTE*)safeBuffer + dictSize;
streamPtr->base = streamPtr->end - endIndex;
streamPtr->dictLimit = endIndex - dictSize;
streamPtr->lowLimit = endIndex - dictSize;
}
return dictSize;
}
/***********************************
* Deprecated Functions
***********************************/
int LZ4_sizeofStreamStateHC(void) { return LZ4_STREAMHCSIZE; }
int LZ4_resetStreamStateHC(void* state, const char* inputBuffer)
{
if ((((size_t)state) & (sizeof(void*)-1)) != 0) return 1; /* Error : pointer is not aligned for pointer (32 or 64 bits) */
LZ4HC_init((LZ4HC_Data_Structure*)state, (const BYTE*)inputBuffer);
return 0;
}
void* LZ4_createHC (const char* inputBuffer)
{
void* hc4 = ALLOCATOR(sizeof(LZ4HC_Data_Structure));
LZ4HC_init ((LZ4HC_Data_Structure*)hc4, (const BYTE*)inputBuffer);
return hc4;
}
int LZ4_freeHC (void* LZ4HC_Data)
{
FREEMEM(LZ4HC_Data);
return (0);
}
/*
int LZ4_compressHC_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize)
{
return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, 0, noLimit);
}
int LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize)
{
return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, 0, limitedOutput);
}
*/
int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel)
{
return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, compressionLevel, noLimit);
}
int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel)
{
return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput);
}
char* LZ4_slideInputBufferHC(void* LZ4HC_Data)
{
LZ4HC_Data_Structure* hc4 = (LZ4HC_Data_Structure*)LZ4HC_Data;
size_t distance = (hc4->end - 64 KB) - hc4->inputBuffer;
if (hc4->end <= hc4->inputBuffer + 64 KB) return (char*)(hc4->end); /* no update : less than 64KB within buffer */
distance = (distance >> 16) << 16; /* Must be a multiple of 64 KB */
LZ4HC_Insert(hc4, hc4->end - MINMATCH);
memcpy((void*)(hc4->end - 64 KB - distance), (const void*)(hc4->end - 64 KB), 64 KB);
hc4->base -= distance;
if ((U32)(hc4->inputBuffer - hc4->base) > 1 GB + 64 KB) /* Avoid overflow */
{
int i;
hc4->base += 1 GB;
for (i=0; i<HASHTABLESIZE; i++) hc4->hashTable[i] -= 1 GB;
}
hc4->end -= distance;
return (char*)(hc4->end);
}
Reference in New Issue View Git Blame Copy Permalink