Commits

Christian Kellermann committed 1643734

Update to upstream release r119

  • Participants
  • Parent commits cffe22e

Comments (0)

Files changed (5)

+LZ4 Library
+Copyright (c) 2011-2014, Yann Collet
+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.
+
+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 HOLDER 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.
-/*
-   LZ4 - Fast LZ compression algorithm
-   Copyright (C) 2011-2012, 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 parameters
-//**************************************
-// MEMORY_USAGE :
-// Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
-// Increasing memory usage improves compression ratio
-// Reduced memory usage can improve speed, due to cache effect
-// Default value is 14, for 16KB, which nicely fits into Intel x86 L1 cache
-#define MEMORY_USAGE 14
-
-// BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE :
-// This will provide a small boost to performance for big endian cpu, but the resulting compressed stream will be incompatible with little-endian CPU.
-// You can set this option to 1 in situations where data will remain within closed environment
-// This option is useless on Little_Endian CPU (such as x86)
-//#define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1
-
-
-
-//**************************************
-// CPU Feature Detection
-//**************************************
-// 32 or 64 bits ?
-#if (defined(__x86_64__) || defined(__x86_64) || defined(__amd64__) || defined(__amd64) || defined(__ppc64__) || defined(_WIN64) || defined(__LP64__) || defined(_LP64) )   // 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
-#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(__ppc__) || defined(_POWER) || defined(__powerpc__) || defined(_ARCH_PPC) || defined(__PPC__) || defined(__PPC) || defined(PPC) || defined(__powerpc__) || defined(__powerpc) || defined(powerpc) \
-   || 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 __STDC_VERSION__ >= 199901L   // C99
-/* "restrict" is a known keyword */
-#else
-#  define restrict // Disable restrict
-#endif
-
-#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
-
-#ifdef _MSC_VER  // Visual Studio
-#  include <intrin.h>   // For Visual 2005
-#  if LZ4_ARCH64	// 64-bit
-#    pragma intrinsic(_BitScanForward64) // For Visual 2005
-#    pragma intrinsic(_BitScanReverse64) // For Visual 2005
-#  else
-#    pragma intrinsic(_BitScanForward)   // For Visual 2005
-#    pragma intrinsic(_BitScanReverse)   // For Visual 2005
-#  endif
-#endif
-
-#ifdef _MSC_VER
-#  define lz4_bswap16(x) _byteswap_ushort(x)
-#else
-#  define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
-#endif
-
-#if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__)
-#  define expect(expr,value)    (__builtin_expect ((expr),(value)) )
-#else
-#  define expect(expr,value)    (expr)
-#endif
-
-#define likely(expr)     expect((expr) != 0, 1)
-#define unlikely(expr)   expect((expr) != 0, 0)
-
-
-//**************************************
-// Includes
-//**************************************
-#include <stdlib.h>   // for malloc
-#include <string.h>   // for memset
-#include "lz4.h"
-
-
-//**************************************
-// Basic Types
-//**************************************
-#if defined(_MSC_VER)    // Visual Studio does not support 'stdint' natively
-#  define BYTE	unsigned __int8
-#  define U16		unsigned __int16
-#  define U32		unsigned __int32
-#  define S32		__int32
-#  define U64		unsigned __int64
-#else
-#  include <stdint.h>
-#  define BYTE	uint8_t
-#  define U16		uint16_t
-#  define U32		uint32_t
-#  define S32		int32_t
-#  define U64		uint64_t
-#endif
-
-#ifndef LZ4_FORCE_UNALIGNED_ACCESS
-#  pragma pack(push, 1)
-#endif
-
-typedef struct _U16_S { U16 v; } U16_S;
-typedef struct _U32_S { U32 v; } U32_S;
-typedef struct _U64_S { U64 v; } U64_S;
-
-#ifndef LZ4_FORCE_UNALIGNED_ACCESS
-#  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 HASH_LOG (MEMORY_USAGE-2)
-#define HASHTABLESIZE (1 << HASH_LOG)
-#define HASH_MASK (HASHTABLESIZE - 1)
-
-// NOTCOMPRESSIBLE_DETECTIONLEVEL :
-// Decreasing this value will make the algorithm skip faster data segments considered "incompressible"
-// This may decrease compression ratio dramatically, but will be faster on incompressible data
-// Increasing this value will make the algorithm search more before declaring a segment "incompressible"
-// This could improve compression a bit, but will be slower on incompressible data
-// The default value (6) is recommended
-#define NOTCOMPRESSIBLE_DETECTIONLEVEL 6
-#define SKIPSTRENGTH (NOTCOMPRESSIBLE_DETECTIONLEVEL>2?NOTCOMPRESSIBLE_DETECTIONLEVEL:2)
-#define STACKLIMIT 13
-#define HEAPMODE (HASH_LOG>STACKLIMIT)  // Defines if memory is allocated into the stack (local variable), or into the heap (malloc()).
-#define COPYLENGTH 8
-#define LASTLITERALS 5
-#define MFLIMIT (COPYLENGTH+MINMATCH)
-#define MINLENGTH (MFLIMIT+1)
-
-#define MAXD_LOG 16
-#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
-
-#define ML_BITS  4
-#define ML_MASK  ((1U<<ML_BITS)-1)
-#define RUN_BITS (8-ML_BITS)
-#define RUN_MASK ((1U<<RUN_BITS)-1)
-
-
-//**************************************
-// Architecture-specific macros
-//**************************************
-#if LZ4_ARCH64	// 64-bit
-#  define STEPSIZE 8
-#  define UARCH U64
-#  define AARCH A64
-#  define LZ4_COPYSTEP(s,d)       A64(d) = A64(s); d+=8; s+=8;
-#  define LZ4_COPYPACKET(s,d)     LZ4_COPYSTEP(s,d)
-#  define LZ4_SECURECOPY(s,d,e)   if (d<e) LZ4_WILDCOPY(s,d,e)
-#  define HTYPE                   U32
-#  define INITBASE(base)          const BYTE* const base = ip
-#else		// 32-bit
-#  define STEPSIZE 4
-#  define UARCH U32
-#  define AARCH A32
-#  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 LZ4_SECURECOPY          LZ4_WILDCOPY
-#  define HTYPE                   const BYTE*
-#  define INITBASE(base)          const int base = 0
-#endif
-
-#if (defined(LZ4_BIG_ENDIAN) && !defined(BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE))
-#  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 structures
-//**************************************
-struct refTables
-{
-    HTYPE hashTable[HASHTABLESIZE];
-};
-
-
-//**************************************
-// Macros
-//**************************************
-#define LZ4_HASH_FUNCTION(i)	(((i) * 2654435761U) >> ((MINMATCH*8)-HASH_LOG))
-#define LZ4_HASH_VALUE(p)		LZ4_HASH_FUNCTION(A32(p))
-#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; }
-
-
-//****************************
-// Private functions
-//****************************
-#if LZ4_ARCH64
-
-static 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__) && (GCC_VERSION >= 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__) && (GCC_VERSION >= 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
-
-static inline int LZ4_NbCommonBytes (register U32 val)
-{
-#if defined(LZ4_BIG_ENDIAN)
-    #if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
-    unsigned long r = 0;
-    _BitScanReverse( &r, val );
-    return (int)(r>>3);
-    #elif defined(__GNUC__) && (GCC_VERSION >= 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__) && (GCC_VERSION >= 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
-
-
-
-//******************************
-// Compression functions
-//******************************
-
-// LZ4_compressCtx :
-// -----------------
-// Compress 'isize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
-// If it cannot achieve it, compression will stop, and result of the function will be zero.
-// return : the number of bytes written in buffer 'dest', or 0 if the compression fails
-
-static inline int LZ4_compressCtx(void** ctx,
-                 const char* source,
-                 char* dest,
-                 int isize,
-                 int maxOutputSize)
-{
-#if HEAPMODE
-    struct refTables *srt = (struct refTables *) (*ctx);
-    HTYPE* HashTable;
-#else
-    HTYPE HashTable[HASHTABLESIZE] = {0};
-#endif
-
-    const BYTE* ip = (BYTE*) source;
-    INITBASE(base);
-    const BYTE* anchor = ip;
-    const BYTE* const iend = ip + isize;
-    const BYTE* const mflimit = iend - MFLIMIT;
-#define matchlimit (iend - LASTLITERALS)
-
-    BYTE* op = (BYTE*) dest;
-    BYTE* const oend = op + maxOutputSize;
-
-    int length;
-    const int skipStrength = SKIPSTRENGTH;
-    U32 forwardH;
-
-
-    // Init
-    if (isize<MINLENGTH) goto _last_literals;
-#if HEAPMODE
-    if (*ctx == NULL)
-    {
-        srt = (struct refTables *) malloc ( sizeof(struct refTables) );
-        *ctx = (void*) srt;
-    }
-    HashTable = (HTYPE*)(srt->hashTable);
-    memset((void*)HashTable, 0, sizeof(srt->hashTable));
-#else
-    (void) ctx;
-#endif
-
-
-    // First Byte
-    HashTable[LZ4_HASH_VALUE(ip)] = ip - base;
-    ip++; forwardH = LZ4_HASH_VALUE(ip);
-
-    // Main Loop
-    for ( ; ; )
-    {
-        int findMatchAttempts = (1U << skipStrength) + 3;
-        const BYTE* forwardIp = ip;
-        const BYTE* ref;
-        BYTE* token;
-
-        // Find a match
-        do {
-            U32 h = forwardH;
-            int step = findMatchAttempts++ >> skipStrength;
-            ip = forwardIp;
-            forwardIp = ip + step;
-
-            if unlikely(forwardIp > mflimit) { goto _last_literals; }
-
-            forwardH = LZ4_HASH_VALUE(forwardIp);
-            ref = base + HashTable[h];
-            HashTable[h] = ip - base;
-
-        } while ((ref < ip - MAX_DISTANCE) || (A32(ref) != A32(ip)));
-
-        // Catch up
-        while ((ip>anchor) && (ref>(BYTE*)source) && unlikely(ip[-1]==ref[-1])) { ip--; ref--; }
-
-        // Encode Literal length
-        length = (int)(ip - anchor);
-        token = op++;
-        if unlikely(op + length + (2 + 1 + LASTLITERALS) + (length>>8) > oend) return 0; 		// Check output limit
-#ifdef _MSC_VER
-        if (length>=(int)RUN_MASK) 
-        { 
-            int len = length-RUN_MASK; 
-            *token=(RUN_MASK<<ML_BITS); 
-            if (len>254)
-            {
-                do { *op++ = 255; len -= 255; } while (len>254);
-                *op++ = (BYTE)len; 
-                memcpy(op, anchor, length);
-                op += length;
-                goto _next_match;
-            }
-            else
-            *op++ = (BYTE)len; 
-        }
-        else *token = (length<<ML_BITS);
-#else
-        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 = (length<<ML_BITS);
-#endif
-
-        // Copy Literals
-        LZ4_BLINDCOPY(anchor, op, length);
-
-_next_match:
-        // Encode Offset
-        LZ4_WRITE_LITTLEENDIAN_16(op,(U16)(ip-ref));
-
-        // Start Counting
-        ip+=MINMATCH; ref+=MINMATCH;    // MinMatch already verified
-        anchor = ip;
-        while likely(ip<matchlimit-(STEPSIZE-1))
-        {
-            UARCH diff = AARCH(ref) ^ AARCH(ip);
-            if (!diff) { ip+=STEPSIZE; ref+=STEPSIZE; continue; }
-            ip += LZ4_NbCommonBytes(diff);
-            goto _endCount;
-        }
-        if (LZ4_ARCH64) if ((ip<(matchlimit-3)) && (A32(ref) == A32(ip))) { ip+=4; ref+=4; }
-        if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
-        if ((ip<matchlimit) && (*ref == *ip)) ip++;
-_endCount:
-
-        // Encode MatchLength
-        length = (int)(ip - anchor);
-        if unlikely(op + (1 + LASTLITERALS) + (length>>8) > oend) return 0; 		// 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 += length;
-
-        // Test end of chunk
-        if (ip > mflimit) { anchor = ip;  break; }
-
-        // Fill table
-        HashTable[LZ4_HASH_VALUE(ip-2)] = ip - 2 - base;
-
-        // Test next position
-        ref = base + HashTable[LZ4_HASH_VALUE(ip)];
-        HashTable[LZ4_HASH_VALUE(ip)] = ip - base;
-        if ((ref > ip - (MAX_DISTANCE + 1)) && (A32(ref) == A32(ip))) { token = op++; *token=0; goto _next_match; }
-
-        // Prepare next loop
-        anchor = ip++;
-        forwardH = LZ4_HASH_VALUE(ip);
-    }
-
-_last_literals:
-    // Encode Last Literals
-    {
-        int lastRun = (int)(iend - anchor);
-        if (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize) return 0;
-        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++ = (lastRun<<ML_BITS);
-        memcpy(op, anchor, iend - anchor);
-        op += iend-anchor;
-    }
-
-    // End
-    return (int) (((char*)op)-dest);
-}
-
-
-
-// Note : this function is valid only if isize < LZ4_64KLIMIT
-#define LZ4_64KLIMIT ((1<<16) + (MFLIMIT-1))
-#define HASHLOG64K (HASH_LOG+1)
-#define HASH64KTABLESIZE (1U<<HASHLOG64K)
-#define LZ4_HASH64K_FUNCTION(i)	(((i) * 2654435761U) >> ((MINMATCH*8)-HASHLOG64K))
-#define LZ4_HASH64K_VALUE(p)	LZ4_HASH64K_FUNCTION(A32(p))
-static inline int LZ4_compress64kCtx(void** ctx,
-                 const char* source,
-                 char* dest,
-                 int isize,
-                 int maxOutputSize)
-{
-#if HEAPMODE
-    struct refTables *srt = (struct refTables *) (*ctx);
-    U16* HashTable;
-#else
-    U16 HashTable[HASH64KTABLESIZE] = {0};
-#endif
-
-    const BYTE* ip = (BYTE*) source;
-    const BYTE* anchor = ip;
-    const BYTE* const base = ip;
-    const BYTE* const iend = ip + isize;
-    const BYTE* const mflimit = iend - MFLIMIT;
-#define matchlimit (iend - LASTLITERALS)
-
-    BYTE* op = (BYTE*) dest;
-    BYTE* const oend = op + maxOutputSize;
-
-    int len, length;
-    const int skipStrength = SKIPSTRENGTH;
-    U32 forwardH;
-
-
-    // Init
-    if (isize<MINLENGTH) goto _last_literals;
-#if HEAPMODE
-    if (*ctx == NULL)
-    {
-        srt = (struct refTables *) malloc ( sizeof(struct refTables) );
-        *ctx = (void*) srt;
-    }
-    HashTable = (U16*)(srt->hashTable);
-    memset((void*)HashTable, 0, sizeof(srt->hashTable));
-#else
-    (void) ctx;
-#endif
-
-
-    // First Byte
-    ip++; forwardH = LZ4_HASH64K_VALUE(ip);
-
-    // Main Loop
-    for ( ; ; )
-    {
-        int findMatchAttempts = (1U << skipStrength) + 3;
-        const BYTE* forwardIp = ip;
-        const BYTE* ref;
-        BYTE* token;
-
-        // Find a match
-        do {
-            U32 h = forwardH;
-            int step = findMatchAttempts++ >> skipStrength;
-            ip = forwardIp;
-            forwardIp = ip + step;
-
-            if (forwardIp > mflimit) { goto _last_literals; }
-
-            forwardH = LZ4_HASH64K_VALUE(forwardIp);
-            ref = base + HashTable[h];
-            HashTable[h] = (U16)(ip - base);
-
-        } while (A32(ref) != A32(ip));
-
-        // Catch up
-        while ((ip>anchor) && (ref>(BYTE*)source) && (ip[-1]==ref[-1])) { ip--; ref--; }
-
-        // Encode Literal length
-        length = (int)(ip - anchor);
-        token = op++;
-        if unlikely(op + length + (2 + 1 + LASTLITERALS) + (length>>8) > oend) return 0; 		// Check output limit
-#ifdef _MSC_VER
-        if (length>=(int)RUN_MASK) 
-        { 
-            int len = length-RUN_MASK; 
-            *token=(RUN_MASK<<ML_BITS); 
-            if (len>254)
-            {
-                do { *op++ = 255; len -= 255; } while (len>254);
-                *op++ = (BYTE)len; 
-                memcpy(op, anchor, length);
-                op += length;
-                goto _next_match;
-            }
-            else
-            *op++ = (BYTE)len; 
-        }
-        else *token = (length<<ML_BITS);
-#else
-        if (length>=(int)RUN_MASK) { *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *op++ = 255; *op++ = (BYTE)len; }
-        else *token = (length<<ML_BITS);
-#endif
-
-        // Copy Literals
-        LZ4_BLINDCOPY(anchor, op, length);
-
-_next_match:
-        // Encode Offset
-        LZ4_WRITE_LITTLEENDIAN_16(op,(U16)(ip-ref));
-
-        // Start Counting
-        ip+=MINMATCH; ref+=MINMATCH;   // MinMatch verified
-        anchor = ip;
-        while (ip<matchlimit-(STEPSIZE-1))
-        {
-            UARCH diff = AARCH(ref) ^ AARCH(ip);
-            if (!diff) { ip+=STEPSIZE; ref+=STEPSIZE; continue; }
-            ip += LZ4_NbCommonBytes(diff);
-            goto _endCount;
-        }
-        if (LZ4_ARCH64) if ((ip<(matchlimit-3)) && (A32(ref) == A32(ip))) { ip+=4; ref+=4; }
-        if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
-        if ((ip<matchlimit) && (*ref == *ip)) ip++;
-_endCount:
-
-        // Encode MatchLength
-        len = (int)(ip - anchor);
-        if unlikely(op + (1 + LASTLITERALS) + (len>>8) > oend) return 0; 		// Check output limit
-        if (len>=(int)ML_MASK) { *token+=ML_MASK; len-=ML_MASK; for(; len > 509 ; len-=510) { *op++ = 255; *op++ = 255; } if (len > 254) { len-=255; *op++ = 255; } *op++ = (BYTE)len; }
-        else *token += len;
-
-        // Test end of chunk
-        if (ip > mflimit) { anchor = ip;  break; }
-
-        // Fill table
-        HashTable[LZ4_HASH64K_VALUE(ip-2)] = (U16)(ip - 2 - base);
-
-        // Test next position
-        ref = base + HashTable[LZ4_HASH64K_VALUE(ip)];
-        HashTable[LZ4_HASH64K_VALUE(ip)] = (U16)(ip - base);
-        if (A32(ref) == A32(ip)) { token = op++; *token=0; goto _next_match; }
-
-        // Prepare next loop
-        anchor = ip++;
-        forwardH = LZ4_HASH64K_VALUE(ip);
-    }
-
-_last_literals:
-    // Encode Last Literals
-    {
-        int lastRun = (int)(iend - anchor);
-        if (op + lastRun + 1 + (lastRun-RUN_MASK+255)/255 > oend) return 0;
-        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++ = (lastRun<<ML_BITS);
-        memcpy(op, anchor, iend - anchor);
-        op += iend-anchor;
-    }
-
-    // End
-    return (int) (((char*)op)-dest);
-}
-
-
-int LZ4_compress_limitedOutput(const char* source, 
-                               char* dest, 
-                               int isize, 
-                               int maxOutputSize)
-{
-#if HEAPMODE
-    void* ctx = malloc(sizeof(struct refTables));
-    int result;
-    if (isize < LZ4_64KLIMIT)
-        result = LZ4_compress64kCtx(&ctx, source, dest, isize, maxOutputSize);
-    else result = LZ4_compressCtx(&ctx, source, dest, isize, maxOutputSize);
-    free(ctx);
-    return result;
-#else
-    if (isize < (int)LZ4_64KLIMIT) return LZ4_compress64kCtx(NULL, source, dest, isize, maxOutputSize);
-    return LZ4_compressCtx(NULL, source, dest, isize, maxOutputSize);
-#endif
-}
-
-
-int LZ4_compress(const char* source,
-                 char* dest,
-                 int isize)
-{
-    return LZ4_compress_limitedOutput(source, dest, isize, LZ4_compressBound(isize));
-}
-
-
-
-
-//****************************
-// Decompression functions
-//****************************
-
-// Note : The decoding functions LZ4_uncompress() and LZ4_uncompress_unknownOutputSize()
-//		are safe against "buffer overflow" attack type.
-//		They will never write nor read outside of the provided output buffers.
-//      LZ4_uncompress_unknownOutputSize() also insures that it will never read outside of the input buffer.
-//		A corrupted input will produce an error result, a negative int, indicating the position of the error within input stream.
-
-int LZ4_uncompress(const char* source,
-                 char* dest,
-                 int osize)
-{
-    // Local Variables
-    const BYTE* restrict ip = (const BYTE*) source;
-    const BYTE* ref;
-
-    BYTE* op = (BYTE*) dest;
-    BYTE* const oend = op + osize;
-    BYTE* cpy;
-
-    unsigned token;
-
-    size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};
-#if LZ4_ARCH64
-    size_t dec64table[] = {0, 0, 0, -1, 0, 1, 2, 3};
-#endif
-
-
-    // Main Loop
-    while (1)
-    {
-        size_t length;
-
-        // get runlength
-        token = *ip++;
-        if ((length=(token>>ML_BITS)) == RUN_MASK)  { size_t len; for (;(len=*ip++)==255;length+=255){} length += len; }
-
-        // copy literals
-        cpy = op+length;
-        if (cpy>oend-COPYLENGTH)
-        {
-            if (cpy != oend) goto _output_error;         // Error : not enough place for another match (min 4) + 5 literals
-            memcpy(op, ip, length);
-            ip += length;
-            break;                                       // EOF
-        }
-        LZ4_WILDCOPY(ip, op, cpy); ip -= (op-cpy); op = cpy;
-
-        // get offset
-        LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
-        if unlikely(ref < (BYTE* const)dest) goto _output_error;   // Error : offset outside destination buffer
-
-        // get matchlength
-        if ((length=(token&ML_MASK)) == ML_MASK) { for (;*ip==255;length+=255) {ip++;} length += *ip++; }
-
-        // copy repeated sequence
-        if unlikely((op-ref)<STEPSIZE)
-        {
-#if LZ4_ARCH64
-            size_t dec64 = dec64table[op-ref];
-#else
-            const int dec64 = 0;
-#endif
-            op[0] = ref[0];
-            op[1] = ref[1];
-            op[2] = ref[2];
-            op[3] = ref[3];
-            op += 4, ref += 4; ref -= dec32table[op-ref];
-            A32(op) = A32(ref); 
-            op += STEPSIZE-4; ref -= dec64;
-        } else { LZ4_COPYSTEP(ref,op); }
-        cpy = op + length - (STEPSIZE-4);
-
-        if unlikely(cpy>oend-(COPYLENGTH)-(STEPSIZE-4))
-        {
-            if (cpy > oend-LASTLITERALS) goto _output_error;    // Error : last 5 bytes must be literals
-            LZ4_SECURECOPY(ref, op, (oend-COPYLENGTH));
-            while(op<cpy) *op++=*ref++;
-            op=cpy;
-            continue;
-        }
-        
-        LZ4_WILDCOPY(ref, op, cpy);
-        op=cpy;		// correction
-    }
-
-    // end of decoding
-    return (int) (((char*)ip)-source);
-
-    // write overflow error detected
-_output_error:
-    return (int) (-(((char*)ip)-source));
-}
-
-
-int LZ4_uncompress_unknownOutputSize(
-                const char* source,
-                char* dest,
-                int isize,
-                int maxOutputSize)
-{
-    // Local Variables
-    const BYTE* restrict ip = (const BYTE*) source;
-    const BYTE* const iend = ip + isize;
-    const BYTE* ref;
-
-    BYTE* op = (BYTE*) dest;
-    BYTE* const oend = op + maxOutputSize;
-    BYTE* cpy;
-
-    size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};
-#if LZ4_ARCH64
-    size_t dec64table[] = {0, 0, 0, -1, 0, 1, 2, 3};
-#endif
-
-
-    // Special case
-    if unlikely(ip==iend) goto _output_error;    // A correctly formed null-compressed LZ4 must have at least one byte (token=0)
-
-    // Main Loop
-    while (1) 
-    {
-        unsigned token;
-        size_t length;
-
-        // get runlength
-        token = *ip++;
-        if ((length=(token>>ML_BITS)) == RUN_MASK) 
-        { 
-            int s=255; 
-            while (likely(ip<iend) && (s==255)) { s=*ip++; length += s; } 
-        }
-
-        // copy literals
-        cpy = op+length;
-        if ((cpy>oend-MFLIMIT) || (ip+length>iend-(2+1+LASTLITERALS)))
-        {
-            if (cpy > oend) goto _output_error;          // Error : writes beyond output buffer
-            if (ip+length != iend) goto _output_error;   // Error : LZ4 format requires to consume all input at this stage (no match within the last 11 bytes, and at least 8 remaining input bytes for another match+literals)
-            memcpy(op, ip, length);
-            op += length;
-            break;                                       // Necessarily EOF, due to parsing restrictions
-        }
-        LZ4_WILDCOPY(ip, op, cpy); ip -= (op-cpy); op = cpy;
-
-        // get offset
-        LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
-        if unlikely(ref < (BYTE* const)dest) goto _output_error;   // Error : offset outside of destination buffer
-
-        // get matchlength
-        if ((length=(token&ML_MASK)) == ML_MASK)    
-        { 
-            while likely(ip<iend-(LASTLITERALS+1))    // Error : a minimum input bytes must remain for LASTLITERALS + token
-            { 
-                int s = *ip++; 
-                length +=s; 
-                if (s==255) continue; 
-                break; 
-            } 
-        }
-
-        // copy repeated sequence
-        if unlikely(op-ref<STEPSIZE)
-        {
-#if LZ4_ARCH64
-            size_t dec64 = dec64table[op-ref];
-#else
-            const int dec64 = 0;
-#endif
-            op[0] = ref[0];
-            op[1] = ref[1];
-            op[2] = ref[2];
-            op[3] = ref[3];
-            op += 4, ref += 4; ref -= dec32table[op-ref];
-            A32(op) = A32(ref); 
-            op += STEPSIZE-4; ref -= dec64;
-        } else { LZ4_COPYSTEP(ref,op); }
-        cpy = op + length - (STEPSIZE-4);
-
-        if unlikely(cpy>oend-(COPYLENGTH+(STEPSIZE-4)))
-        {
-            if (cpy > oend-LASTLITERALS) goto _output_error;    // Error : last 5 bytes must be literals
-            LZ4_SECURECOPY(ref, op, (oend-COPYLENGTH));
-            while(op<cpy) *op++=*ref++;
-            op=cpy;
-            continue;
-        }
-
-        LZ4_WILDCOPY(ref, op, cpy);
-        op=cpy;		// correction
-    }
-
-    // end of decoding
-    return (int) (((char*)op)-dest);
-
-    // write overflow error detected
-_output_error:
-    return (int) (-(((char*)ip)-source));
-}
-
+/*
+   LZ4 - Fast LZ compression algorithm
+   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 source repository : http://code.google.com/p/lz4/
+   - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
+*/
+
+/**************************************
+   Tuning parameters
+**************************************/
+/*
+ * HEAPMODE :
+ * Select how default compression functions will allocate memory for their hash table,
+ * in memory stack (0:default, fastest), or in memory heap (1:requires memory allocation (malloc)).
+ */
+#define HEAPMODE 0
+
+
+/**************************************
+   CPU Feature Detection
+**************************************/
+/* 32 or 64 bits ? */
+#if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
+  || defined(__powerpc64__) || defined(__powerpc64le__) \
+  || defined(__ppc64__) || defined(__ppc64le__) \
+  || defined(__PPC64__) || defined(__PPC64LE__) \
+  || defined(__ia64) || defined(__itanium__) || defined(_M_IA64) )   /* 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, such as ARM, the compiler may be more cautious, inserting unnecessary extra code to ensure aligned access property
+ * 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
+
+/*
+ * BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE :
+ * This option may provide a small boost to performance for some big endian cpu, although probably modest.
+ * You may set this option to 1 if data will remain within closed environment.
+ * This option is useless on Little_Endian CPU (such as x86)
+ */
+
+/* #define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1 */
+
+
+/**************************************
+ 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 */
+#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
+
+#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
+
+#if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__)
+#  define expect(expr,value)    (__builtin_expect ((expr),(value)) )
+#else
+#  define expect(expr,value)    (expr)
+#endif
+
+#define likely(expr)     expect((expr) != 0, 1)
+#define unlikely(expr)   expect((expr) != 0, 0)
+
+
+/**************************************
+   Memory routines
+**************************************/
+#include <stdlib.h>   /* malloc, calloc, free */
+#define ALLOCATOR(n,s) calloc(n,s)
+#define FREEMEM        free
+#include <string.h>   /* memset, memcpy */
+#define MEM_INIT       memset
+
+
+/**************************************
+   Includes
+**************************************/
+#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__)
+#  if defined(__IBMC__) || defined(__SUNPRO_C) || defined(__SUNPRO_CC)
+#    pragma pack(1)
+#  else
+#    pragma pack(push, 1)
+#  endif
+#endif
+
+typedef struct { U16 v; }  _PACKED U16_S;
+typedef struct { U32 v; }  _PACKED U32_S;
+typedef struct { U64 v; }  _PACKED U64_S;
+typedef struct {size_t v;} _PACKED size_t_S;
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+#  if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
+#    pragma pack(0)
+#  else
+#    pragma pack(pop)
+#  endif
+#endif
+
+#define A16(x)   (((U16_S *)(x))->v)
+#define A32(x)   (((U32_S *)(x))->v)
+#define A64(x)   (((U64_S *)(x))->v)
+#define AARCH(x) (((size_t_S *)(x))->v)
+
+
+/**************************************
+   Constants
+**************************************/
+#define LZ4_HASHLOG   (LZ4_MEMORY_USAGE-2)
+#define HASHTABLESIZE (1 << LZ4_MEMORY_USAGE)
+#define HASH_SIZE_U32 (1 << LZ4_HASHLOG)
+
+#define MINMATCH 4
+
+#define COPYLENGTH 8
+#define LASTLITERALS 5
+#define MFLIMIT (COPYLENGTH+MINMATCH)
+static const int LZ4_minLength = (MFLIMIT+1);
+
+#define KB *(1U<<10)
+#define MB *(1U<<20)
+#define GB *(1U<<30)
+
+#define LZ4_64KLIMIT ((64 KB) + (MFLIMIT-1))
+#define SKIPSTRENGTH 6   /* Increasing this value will make the compression run slower on incompressible data */
+
+#define MAXD_LOG 16
+#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
+
+#define ML_BITS  4
+#define ML_MASK  ((1U<<ML_BITS)-1)
+#define RUN_BITS (8-ML_BITS)
+#define RUN_MASK ((1U<<RUN_BITS)-1)
+
+
+/**************************************
+   Structures and local types
+**************************************/
+typedef struct {
+    U32  hashTable[HASH_SIZE_U32];
+    U32  currentOffset;
+    U32  initCheck;
+    const BYTE* dictionary;
+    const BYTE* bufferStart;
+    U32  dictSize;
+} LZ4_stream_t_internal;
+
+typedef enum { notLimited = 0, limitedOutput = 1 } limitedOutput_directive;
+typedef enum { byPtr, byU32, byU16 } tableType_t;
+
+typedef enum { noDict = 0, withPrefix64k, usingExtDict } dict_directive;
+typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive;
+
+typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
+typedef enum { full = 0, partial = 1 } earlyEnd_directive;
+
+
+/**************************************
+   Architecture-specific macros
+**************************************/
+#define STEPSIZE                  sizeof(size_t)
+#define LZ4_COPYSTEP(d,s)         { AARCH(d) = AARCH(s); d+=STEPSIZE; s+=STEPSIZE; }
+#define LZ4_COPY8(d,s)            { LZ4_COPYSTEP(d,s); if (STEPSIZE<8) LZ4_COPYSTEP(d,s); }
+
+#if (defined(LZ4_BIG_ENDIAN) && !defined(BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE))
+#  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
+
+
+/**************************************
+   Macros
+**************************************/
+#define LZ4_STATIC_ASSERT(c)    { enum { LZ4_static_assert = 1/(!!(c)) }; }   /* use only *after* variable declarations */
+#if LZ4_ARCH64 || !defined(__GNUC__)
+#  define LZ4_WILDCOPY(d,s,e)   { do { LZ4_COPY8(d,s) } while (d<e); }        /* at the end, d>=e; */
+#else
+#  define LZ4_WILDCOPY(d,s,e)   { if (likely(e-d <= 8)) LZ4_COPY8(d,s) else do { LZ4_COPY8(d,s) } while (d<e); }
+#endif
+
+
+/****************************
+   Private local functions
+****************************/
+#if LZ4_ARCH64
+
+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__) && (GCC_VERSION >= 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__) && (GCC_VERSION >= 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 & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+#   endif
+# endif
+}
+
+#else
+
+int LZ4_NbCommonBytes (register U32 val)
+{
+# if defined(LZ4_BIG_ENDIAN)
+#   if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+    unsigned long r = 0;
+    _BitScanReverse( &r, val );
+    return (int)(r>>3);
+#   elif defined(__GNUC__) && (GCC_VERSION >= 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__) && (GCC_VERSION >= 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
+
+
+/********************************
+   Compression functions
+********************************/
+int LZ4_compressBound(int isize)  { return LZ4_COMPRESSBOUND(isize); }
+
+static int LZ4_hashSequence(U32 sequence, tableType_t tableType)
+{
+    if (tableType == byU16)
+        return (((sequence) * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
+    else
+        return (((sequence) * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
+}
+
+static int LZ4_hashPosition(const BYTE* p, tableType_t tableType) { return LZ4_hashSequence(A32(p), tableType); }
+
+static void LZ4_putPositionOnHash(const BYTE* p, U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    switch (tableType)
+    {
+    case byPtr: { const BYTE** hashTable = (const BYTE**) tableBase; hashTable[h] = p; break; }
+    case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = (U32)(p-srcBase); break; }
+    case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = (U16)(p-srcBase); break; }
+    }
+}
+
+static void LZ4_putPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    U32 h = LZ4_hashPosition(p, tableType);
+    LZ4_putPositionOnHash(p, h, tableBase, tableType, srcBase);
+}
+
+static const BYTE* LZ4_getPositionOnHash(U32 h, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    if (tableType == byPtr) { const BYTE** hashTable = (const BYTE**) tableBase; return hashTable[h]; }
+    if (tableType == byU32) { U32* hashTable = (U32*) tableBase; return hashTable[h] + srcBase; }
+    { U16* hashTable = (U16*) tableBase; return hashTable[h] + srcBase; }   /* default, to ensure a return */
+}
+
+static const BYTE* LZ4_getPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
+{
+    U32 h = LZ4_hashPosition(p, tableType);
+    return LZ4_getPositionOnHash(h, tableBase, tableType, srcBase);
+}
+
+static unsigned LZ4_count(const BYTE* pIn, const BYTE* pRef, const BYTE* pInLimit)
+{
+    const BYTE* const pStart = pIn;
+
+    while (likely(pIn<pInLimit-(STEPSIZE-1)))
+    {
+        size_t diff = AARCH(pRef) ^ AARCH(pIn);
+        if (!diff) { pIn+=STEPSIZE; pRef+=STEPSIZE; continue; }
+        pIn += LZ4_NbCommonBytes(diff);
+        return (unsigned)(pIn - pStart);
+    }
+    if (sizeof(void*)==8) if ((pIn<(pInLimit-3)) && (A32(pRef) == A32(pIn))) { pIn+=4; pRef+=4; }
+    if ((pIn<(pInLimit-1)) && (A16(pRef) == A16(pIn))) { pIn+=2; pRef+=2; }
+    if ((pIn<pInLimit) && (*pRef == *pIn)) pIn++;
+
+    return (unsigned)(pIn - pStart);
+}
+
+
+static int LZ4_compress_generic(
+                 void* ctx,
+                 const char* source,
+                 char* dest,
+                 int inputSize,
+                 int maxOutputSize,
+
+                 limitedOutput_directive outputLimited,
+                 tableType_t tableType,
+                 dict_directive dict,
+                 dictIssue_directive dictIssue)
+{
+    LZ4_stream_t_internal* const dictPtr = (LZ4_stream_t_internal*)ctx;
+
+    const BYTE* ip = (const BYTE*) source;
+    const BYTE* base;
+    const BYTE* lowLimit;
+    const BYTE* const lowRefLimit = ip - dictPtr->dictSize;
+    const BYTE* const dictionary = dictPtr->dictionary;
+    const BYTE* const dictEnd = dictionary + dictPtr->dictSize;
+    const size_t dictDelta = dictEnd - (const BYTE*)source;
+    const BYTE* anchor = (const BYTE*) source;
+    const BYTE* const iend = ip + inputSize;
+    const BYTE* const mflimit = iend - MFLIMIT;
+    const BYTE* const matchlimit = iend - LASTLITERALS;
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const olimit = op + maxOutputSize;
+
+    const int skipStrength = SKIPSTRENGTH;
+    U32 forwardH;
+    size_t refDelta=0;
+
+    /* Init conditions */
+    if ((U32)inputSize > (U32)LZ4_MAX_INPUT_SIZE) return 0;          /* Unsupported input size, too large (or negative) */
+    switch(dict)
+    {
+    case noDict:
+    default:
+        base = (const BYTE*)source;
+        lowLimit = (const BYTE*)source;
+        break;
+    case withPrefix64k:
+        base = (const BYTE*)source - dictPtr->currentOffset;
+        lowLimit = (const BYTE*)source - dictPtr->dictSize;
+        break;
+    case usingExtDict:
+        base = (const BYTE*)source - dictPtr->currentOffset;
+        lowLimit = (const BYTE*)source;
+        break;
+    }
+    if ((tableType == byU16) && (inputSize>=(int)LZ4_64KLIMIT)) return 0;   /* Size too large (not within 64K limit) */
+    if (inputSize<LZ4_minLength) goto _last_literals;                       /* Input too small, no compression (all literals) */
+
+    /* First Byte */
+    LZ4_putPosition(ip, ctx, tableType, base);
+    ip++; forwardH = LZ4_hashPosition(ip, tableType);
+
+    /* Main Loop */
+    for ( ; ; )
+    {
+        const BYTE* ref;
+        BYTE* token;
+        {
+            const BYTE* forwardIp = ip;
+            unsigned step=1;
+            unsigned searchMatchNb = (1U << skipStrength);
+
+            /* Find a match */
+            do {
+                U32 h = forwardH;
+                ip = forwardIp;
+                forwardIp += step;
+                step = searchMatchNb++ >> skipStrength;
+                //if (step>8) step=8;   // required for valid forwardIp ; slows down uncompressible data a bit
+
+                if (unlikely(forwardIp > mflimit)) goto _last_literals;
+
+                ref = LZ4_getPositionOnHash(h, ctx, tableType, base);
+                if (dict==usingExtDict)
+                {
+                    if (ref<(const BYTE*)source)
+                    {
+                        refDelta = dictDelta;
+                        lowLimit = dictionary;
+                    }
+                    else
+                    {
+                        refDelta = 0;
+                        lowLimit = (const BYTE*)source;
+                    }
+                }
+                forwardH = LZ4_hashPosition(forwardIp, tableType);
+                LZ4_putPositionOnHash(ip, h, ctx, tableType, base);
+
+            } while ( ((dictIssue==dictSmall) ? (ref < lowRefLimit) : 0)
+                || ((tableType==byU16) ? 0 : (ref + MAX_DISTANCE < ip))
+                || (A32(ref+refDelta) != A32(ip)) );
+        }
+
+        /* Catch up */
+        while ((ip>anchor) && (ref+refDelta > lowLimit) && (unlikely(ip[-1]==ref[refDelta-1]))) { ip--; ref--; }
+
+        {
+            /* Encode Literal length */
+            unsigned litLength = (unsigned)(ip - anchor);
+            token = op++;
+            if ((outputLimited) && (unlikely(op + litLength + (2 + 1 + LASTLITERALS) + (litLength/255) > olimit)))
+                return 0;   /* Check output limit */
+            if (litLength>=RUN_MASK)
+            {
+                int len = (int)litLength-RUN_MASK;
+                *token=(RUN_MASK<<ML_BITS);
+                for(; len >= 255 ; len-=255) *op++ = 255;
+                *op++ = (BYTE)len;
+            }
+            else *token = (BYTE)(litLength<<ML_BITS);
+
+            /* Copy Literals */
+            { BYTE* end = op+litLength; LZ4_WILDCOPY(op,anchor,end); op=end; }
+        }
+
+_next_match:
+        /* Encode Offset */
+        LZ4_WRITE_LITTLEENDIAN_16(op, (U16)(ip-ref));
+
+        /* Encode MatchLength */
+        {
+            unsigned matchLength;
+
+            if ((dict==usingExtDict) && (lowLimit==dictionary))
+            {
+                const BYTE* limit;
+                ref += refDelta;
+                limit = ip + (dictEnd-ref);
+                if (limit > matchlimit) limit = matchlimit;
+                matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, limit);
+                ip += MINMATCH + matchLength;
+                if (ip==limit)
+                {
+                    unsigned more = LZ4_count(ip, (const BYTE*)source, matchlimit);
+                    matchLength += more;
+                    ip += more;
+                }
+            }
+            else
+            {
+                matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, matchlimit);
+                ip += MINMATCH + matchLength;
+            }
+
+            if (matchLength>=ML_MASK)
+            {
+                if ((outputLimited) && (unlikely(op + (1 + LASTLITERALS) + (matchLength>>8) > olimit)))
+                    return 0;    /* Check output limit */
+                *token += ML_MASK;
+                matchLength -= ML_MASK;
+                for (; matchLength >= 510 ; matchLength-=510) { *op++ = 255; *op++ = 255; }
+                if (matchLength >= 255) { matchLength-=255; *op++ = 255; }
+                *op++ = (BYTE)matchLength;
+            }
+            else *token += (BYTE)(matchLength);
+        }
+
+        anchor = ip;
+
+        /* Test end of chunk */
+        if (ip > mflimit) break;
+
+        /* Fill table */
+        LZ4_putPosition(ip-2, ctx, tableType, base);
+
+        /* Test next position */
+        ref = LZ4_getPosition(ip, ctx, tableType, base);
+        if (dict==usingExtDict)
+        {
+            if (ref<(const BYTE*)source)
+            {
+                refDelta = dictDelta;
+                lowLimit = dictionary;
+            }
+            else
+            {
+                refDelta = 0;
+                lowLimit = (const BYTE*)source;
+            }
+        }
+        LZ4_putPosition(ip, ctx, tableType, base);
+        if ( ((dictIssue==dictSmall) ? (ref>=lowRefLimit) : 1)
+            && (ref+MAX_DISTANCE>=ip)
+            && (A32(ref+refDelta)==A32(ip)) )
+        { token=op++; *token=0; goto _next_match; }
+
+        /* Prepare next loop */
+        forwardH = LZ4_hashPosition(++ip, tableType);
+    }
+
+_last_literals:
+    /* Encode Last Literals */
+    {
+        int lastRun = (int)(iend - anchor);
+        if ((outputLimited) && (((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 >= 255 ; 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_compress(const char* source, char* dest, int inputSize)
+{
+#if (HEAPMODE)
+    void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U32, 4);   /* Aligned on 4-bytes boundaries */
+#else
+    U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
+#endif
+    int result;
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, byU16, noDict, noDictIssue);
+    else
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
+
+#if (HEAPMODE)
+    FREEMEM(ctx);
+#endif
+    return result;
+}
+
+int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+#if (HEAPMODE)
+    void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U32, 4);   /* Aligned on 4-bytes boundaries */
+#else
+    U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
+#endif
+    int result;
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue);
+    else
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limitedOutput, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
+
+#if (HEAPMODE)
+    FREEMEM(ctx);
+#endif
+    return result;
+}
+
+
+/*****************************************
+   Experimental : Streaming functions
+*****************************************/
+
+void* LZ4_createStream()
+{
+    void* lz4s = ALLOCATOR(4, LZ4_STREAMSIZE_U32);
+    MEM_INIT(lz4s, 0, LZ4_STREAMSIZE);
+    return lz4s;
+}
+
+int LZ4_free (void* LZ4_stream)
+{
+    FREEMEM(LZ4_stream);
+    return (0);
+}
+
+
+int LZ4_loadDict (void* LZ4_dict, const char* dictionary, int dictSize)
+{
+    LZ4_stream_t_internal* dict = (LZ4_stream_t_internal*) LZ4_dict;
+    const BYTE* p = (const BYTE*)dictionary;
+    const BYTE* const dictEnd = p + dictSize;
+    const BYTE* base;
+
+    LZ4_STATIC_ASSERT(LZ4_STREAMSIZE >= sizeof(LZ4_stream_t_internal));      /* A compilation error here means LZ4_STREAMSIZE is not large enough */
+    if (dict->initCheck) MEM_INIT(dict, 0, sizeof(LZ4_stream_t_internal));   /* Uninitialized structure detected */
+
+    if (dictSize < MINMATCH)
+    {
+        dict->dictionary = NULL;
+        dict->dictSize = 0;
+        return 1;
+    }
+
+    if (p <= dictEnd - 64 KB) p = dictEnd - 64 KB;
+    base = p - dict->currentOffset;
+    dict->dictionary = p;
+    dict->dictSize = (U32)(dictEnd - p);
+    dict->currentOffset += dict->dictSize;
+
+    while (p <= dictEnd-MINMATCH)
+    {
+        LZ4_putPosition(p, dict, byU32, base);
+        p+=3;
+    }
+
+    return 1;
+}
+
+
+void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, const BYTE* src)
+{
+    if ((LZ4_dict->currentOffset > 0x80000000) ||
+        ((size_t)LZ4_dict->currentOffset > (size_t)src))   /* address space overflow */
+    {
+        /* rescale hash table */
+        U32 delta = LZ4_dict->currentOffset - 64 KB;
+        const BYTE* dictEnd = LZ4_dict->dictionary + LZ4_dict->dictSize;
+        int i;
+        for (i=0; i<HASH_SIZE_U32; i++)
+        {
+            if (LZ4_dict->hashTable[i] < delta) LZ4_dict->hashTable[i]=0;
+            else LZ4_dict->hashTable[i] -= delta;
+        }
+        LZ4_dict->currentOffset = 64 KB;
+        if (LZ4_dict->dictSize > 64 KB) LZ4_dict->dictSize = 64 KB;
+        LZ4_dict->dictionary = dictEnd - LZ4_dict->dictSize;
+    }
+}
+
+
+FORCE_INLINE int LZ4_compress_continue_generic (void* LZ4_stream, const char* source, char* dest, int inputSize,
+                                                int maxOutputSize, limitedOutput_directive limit)
+{
+    LZ4_stream_t_internal* streamPtr = (LZ4_stream_t_internal*)LZ4_stream;
+    const BYTE* const dictEnd = streamPtr->dictionary + streamPtr->dictSize;
+
+    const BYTE* smallest = (const BYTE*) source;
+    if (streamPtr->initCheck) return 0;   /* Uninitialized structure detected */
+    if ((streamPtr->dictSize>0) && (smallest>dictEnd)) smallest = dictEnd;
+    LZ4_renormDictT(streamPtr, smallest);
+
+    /* Check overlapping input/dictionary space */
+    {
+        const BYTE* sourceEnd = (const BYTE*) source + inputSize;
+        if ((sourceEnd > streamPtr->dictionary) && (sourceEnd < dictEnd))
+        {
+            streamPtr->dictSize = (U32)(dictEnd - sourceEnd);
+            if (streamPtr->dictSize > 64 KB) streamPtr->dictSize = 64 KB;
+            if (streamPtr->dictSize < 4) streamPtr->dictSize = 0;
+            streamPtr->dictionary = dictEnd - streamPtr->dictSize;
+        }
+    }
+
+    /* prefix mode : source data follows dictionary */
+    if (dictEnd == (const BYTE*)source)
+    {
+        int result;
+        if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset))
+            result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, withPrefix64k, dictSmall);
+        else
+            result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, withPrefix64k, noDictIssue);
+        streamPtr->dictSize += (U32)inputSize;
+        streamPtr->currentOffset += (U32)inputSize;
+        return result;
+    }
+
+    /* external dictionary mode */
+    {
+        int result;
+        if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset))
+            result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, usingExtDict, dictSmall);
+        else
+            result = LZ4_compress_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limit, byU32, usingExtDict, noDictIssue);
+        streamPtr->dictionary = (const BYTE*)source;
+        streamPtr->dictSize = (U32)inputSize;
+        streamPtr->currentOffset += (U32)inputSize;
+        return result;
+    }
+}
+
+
+int LZ4_compress_continue (void* LZ4_stream, const char* source, char* dest, int inputSize)
+{
+    return LZ4_compress_continue_generic(LZ4_stream, source, dest, inputSize, 0, notLimited);
+}
+
+int LZ4_compress_limitedOutput_continue (void* LZ4_stream, const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    return LZ4_compress_continue_generic(LZ4_stream, source, dest, inputSize, maxOutputSize, limitedOutput);
+}
+
+
+// Hidden debug function, to force separate dictionary mode
+int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* dest, int inputSize)
+{
+    LZ4_stream_t_internal* streamPtr = (LZ4_stream_t_internal*)LZ4_dict;
+    int result;
+    const BYTE* const dictEnd = streamPtr->dictionary + streamPtr->dictSize;
+
+    const BYTE* smallest = dictEnd;
+    if (smallest > (const BYTE*) source) smallest = (const BYTE*) source;
+    LZ4_renormDictT((LZ4_stream_t_internal*)LZ4_dict, smallest);
+
+    result = LZ4_compress_generic(LZ4_dict, source, dest, inputSize, 0, notLimited, byU32, usingExtDict, noDictIssue);
+
+    streamPtr->dictionary = (const BYTE*)source;
+    streamPtr->dictSize = (U32)inputSize;
+    streamPtr->currentOffset += (U32)inputSize;
+
+    return result;
+}
+
+
+int LZ4_saveDict (void* LZ4_dict, char* safeBuffer, int dictSize)
+{
+    LZ4_stream_t_internal* dict = (LZ4_stream_t_internal*) LZ4_dict;
+    const BYTE* previousDictEnd = dict->dictionary + dict->dictSize;
+
+    if ((U32)dictSize > 64 KB) dictSize = 64 KB;   /* useless to define a dictionary > 64 KB */
+    if ((U32)dictSize > dict->dictSize) dictSize = dict->dictSize;
+
+    memcpy(safeBuffer, previousDictEnd - dictSize, dictSize);
+
+    dict->dictionary = (const BYTE*)safeBuffer;
+    dict->dictSize = (U32)dictSize;
+
+    return 1;
+}
+
+
+
+/****************************
+   Decompression functions
+****************************/
+/*
+ * This generic decompression function cover all use cases.
+ * It shall be instanciated several times, using different sets of directives
+ * Note that it is essential this generic function is really inlined,
+ * in order to remove useless branches during compilation optimisation.
+ */
+FORCE_INLINE int LZ4_decompress_generic(
+                 const char* source,
+                 char* dest,
+                 int inputSize,
+                 int outputSize,         /* If endOnInput==endOnInputSize, this value is the max size of Output Buffer. */
+
+                 int endOnInput,         /* endOnOutputSize, endOnInputSize */
+                 int partialDecoding,    /* full, partial */
+                 int targetOutputSize,   /* only used if partialDecoding==partial */
+                 int dict,               /* noDict, withPrefix64k, usingExtDict */
+                 const char* dictStart,  /* only if dict==usingExtDict */
+                 int dictSize            /* note : = 0 if noDict */
+                 )
+{
+    /* Local Variables */
+    const BYTE* restrict ip = (const BYTE*) source;
+    const BYTE* ref;
+    const BYTE* const iend = ip + inputSize;
+
+    BYTE* op = (BYTE*) dest;
+    BYTE* const oend = op + outputSize;
+    BYTE* cpy;
+    BYTE* oexit = op + targetOutputSize;
+    const BYTE* const lowLimit = (const BYTE*)dest - dictSize;
+
+    const BYTE* const dictEnd = (const BYTE*)dictStart + dictSize;
+//#define OLD
+#ifdef OLD
+    const size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};   /* static reduces speed for LZ4_decompress_safe() on GCC64 */
+#else
+    const size_t dec32table[] = {4-0, 4-3, 4-2, 4-3, 4-0, 4-0, 4-0, 4-0};   /* static reduces speed for LZ4_decompress_safe() on GCC64 */
+#endif
+    static const size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
+
+    const int checkOffset = (endOnInput) && (dictSize < (int)(64 KB));
+
+
+    /* Special cases */
+    if ((partialDecoding) && (oexit> oend-MFLIMIT)) oexit = oend-MFLIMIT;                        /* targetOutputSize too high => decode everything */
+    if ((endOnInput) && (unlikely(outputSize==0))) return ((inputSize==1) && (*ip==0)) ? 0 : -1;   /* Empty output buffer */
+    if ((!endOnInput) && (unlikely(outputSize==0))) return (*ip==0?1:-1);
+
+
+    /* Main Loop */
+    while (1)
+    {
+        unsigned token;
+        size_t length;
+
+        /* get runlength */
+        token = *ip++;
+        if ((length=(token>>ML_BITS)) == RUN_MASK)
+        {
+            unsigned s;
+            do
+            {
+                s = *ip++;
+                length += s;
+            }
+            while (likely((endOnInput)?ip<iend-RUN_MASK:1) && (s==255));
+            //if ((sizeof(void*)==4) && unlikely(length>LZ4_MAX_INPUT_SIZE)) goto _output_error;   /* overflow detection */
+            if ((sizeof(void*)==4) && unlikely((size_t)(op+length)<(size_t)(op))) goto _output_error;   /* quickfix issue 134 */
+            if ((endOnInput) && (sizeof(void*)==4) && unlikely((size_t)(ip+length)<(size_t)(ip))) goto _output_error;   /* quickfix issue 134 */
+        }
+
+        /* copy literals */
+        cpy = op+length;
+        if (((endOnInput) && ((cpy>(partialDecoding?oexit:oend-MFLIMIT)) || (ip+length>iend-(2+1+LASTLITERALS))) )
+            || ((!endOnInput) && (cpy>oend-COPYLENGTH)))
+        {
+            if (partialDecoding)
+            {
+                if (cpy > oend) goto _output_error;                           /* Error : write attempt beyond end of output buffer */
+                if ((endOnInput) && (ip+length > iend)) goto _output_error;   /* Error : read attempt beyond end of input buffer */
+            }
+            else
+            {
+                if ((!endOnInput) && (cpy != oend)) goto _output_error;       /* Error : block decoding must stop exactly there */
+                if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error;   /* Error : input must be consumed */
+            }
+            memcpy(op, ip, length);
+            ip += length;
+            op += length;
+            break;                                       /* Necessarily EOF, due to parsing restrictions */
+        }
+        LZ4_WILDCOPY(op, ip, cpy); ip -= (op-cpy); op = cpy;
+
+        /* get offset */
+        LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
+        if ((checkOffset) && (unlikely(ref < lowLimit))) goto _output_error;   /* Error : offset outside destination buffer */
+
+        /* get matchlength */
+        if ((length=(token&ML_MASK)) == ML_MASK)
+        {
+            unsigned s;
+            do
+            {
+                if ((endOnInput) && (ip > iend-LASTLITERALS)) goto _output_error;
+                s = *ip++;
+                length += s;
+            } while (s==255);
+            //if ((sizeof(void*)==4) && unlikely(length>LZ4_MAX_INPUT_SIZE)) goto _output_error;   /* overflow detection */
+            if ((sizeof(void*)==4) && unlikely((size_t)(op+length)<(size_t)op)) goto _output_error;   /* quickfix issue 134 */
+        }
+
+        /* check external dictionary */
+        if ((dict==usingExtDict) && (ref < (BYTE* const)dest))
+        {
+            if (unlikely(op+length+MINMATCH > oend-LASTLITERALS)) goto _output_error;
+
+            if (length+MINMATCH <= (size_t)(dest-(char*)ref))
+            {
+                ref = dictEnd - (dest-(char*)ref);
+                memcpy(op, ref, length+MINMATCH);
+                op += length+MINMATCH;
+            }
+            else
+            {
+                size_t copySize = (size_t)(dest-(char*)ref);
+                memcpy(op, dictEnd - copySize, copySize);
+                op += copySize;
+                copySize = length+MINMATCH - copySize;
+                if (copySize > (size_t)((char*)op-dest))   /* overlap */
+                {
+                    BYTE* const cpy = op + copySize;
+                    const BYTE* ref = (BYTE*)dest;
+                    while (op < cpy) *op++ = *ref++;
+                }
+                else
+                {
+                    memcpy(op, dest, copySize);
+                    op += copySize;
+                }
+            }
+            continue;
+        }
+
+        /* copy repeated sequence */
+        if (unlikely((op-ref)<(int)STEPSIZE))
+        {
+            const size_t dec64 = dec64table[(sizeof(void*)==4) ? 0 : op-ref];
+            op[0] = ref[0];
+            op[1] = ref[1];
+            op[2] = ref[2];
+            op[3] = ref[3];
+#ifdef OLD
+            op += 4, ref += 4; ref -= dec32table[op-ref];
+            A32(op) = A32(ref);
+            op += STEPSIZE-4; ref -= dec64;
+#else
+            ref += dec32table[op-ref];
+            A32(op+4) = A32(ref);
+            op += STEPSIZE; ref -= dec64;
+#endif
+        } else { LZ4_COPYSTEP(op,ref); }
+        cpy = op + length - (STEPSIZE-4);
+
+        if (unlikely(cpy>oend-COPYLENGTH-(STEPSIZE-4)))
+        {
+            if (cpy > oend-LASTLITERALS) goto _output_error;    /* Error : last 5 bytes must be literals */
+            if (op<oend-COPYLENGTH) LZ4_WILDCOPY(op, ref, (oend-COPYLENGTH));
+            while(op<cpy) *op++=*ref++;
+            op=cpy;
+            continue;
+        }
+        LZ4_WILDCOPY(op, ref, cpy);
+        op=cpy;   /* correction */
+    }
+
+    /* end of decoding */
+    if (endOnInput)
+       return (int) (((char*)op)-dest);     /* Nb of output bytes decoded */
+    else
+       return (int) (((char*)ip)-source);   /* Nb of input bytes read */
+
+    /* Overflow error detected */
+_output_error:
+    return (int) (-(((char*)ip)-source))-1;
+}
+
+
+int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, noDict, NULL, 0);
+}
+
+int LZ4_decompress_safe_partial(const char* source, char* dest, int compressedSize, int targetOutputSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, partial, targetOutputSize, noDict, NULL, 0);
+}
+
+int LZ4_decompress_fast(const char* source, char* dest, int originalSize)
+{
+    return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, withPrefix64k, NULL, 0);
+}
+
+/* streaming decompression functions */
+
+//#define LZ4_STREAMDECODESIZE_U32 4
+//#define LZ4_STREAMDECODESIZE     (LZ4_STREAMDECODESIZE_U32 * sizeof(unsigned int))
+//typedef struct { unsigned int table[LZ4_STREAMDECODESIZE_U32]; } LZ4_streamDecode_t;
+typedef struct
+{
+    const char* dictionary;
+    int dictSize;
+} LZ4_streamDecode_t_internal;
+
+/*
+ * If you prefer dynamic allocation methods,
+ * LZ4_createStreamDecode()
+ * provides a pointer (void*) towards an initialized LZ4_streamDecode_t structure.
+ */
+void* LZ4_createStreamDecode()
+{
+    void* lz4s = ALLOCATOR(sizeof(U32), LZ4_STREAMDECODESIZE_U32);
+    MEM_INIT(lz4s, 0, LZ4_STREAMDECODESIZE);
+    return lz4s;
+}
+
+/*
+ * LZ4_setDictDecode
+ * Use this function to instruct where to find the dictionary
+ * This function is not necessary if previous data is still available where it was decoded.
+ * Loading a size of 0 is allowed (same effect as no dictionary).
+ * Return : 1 if OK, 0 if error
+ */
+int LZ4_setDictDecode (void* LZ4_streamDecode, const char* dictionary, int dictSize)
+{
+    LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
+    lz4sd->dictionary = dictionary;
+    lz4sd->dictSize = dictSize;
+    return 1;
+}
+
+/*
+*_continue() :
+    These decoding functions allow decompression of multiple blocks in "streaming" mode.
+    Previously decoded blocks must still be available at the memory position where they were decoded.
+    If it's not possible, save the relevant part of decoded data into a safe buffer,
+    and indicate where it stands using LZ4_setDictDecode()
+*/
+int LZ4_decompress_safe_continue (void* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize)
+{
+    LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
+    int result;
+
+    result = LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, usingExtDict, lz4sd->dictionary, lz4sd->dictSize);
+    if (result <= 0) return result;
+    if (lz4sd->dictionary + lz4sd->dictSize == dest)
+    {
+        lz4sd->dictSize += result;
+    }
+    else
+    {
+        lz4sd->dictionary = dest;
+        lz4sd->dictSize = result;
+    }
+
+    return result;
+}
+
+int LZ4_decompress_fast_continue (void* LZ4_streamDecode, const char* source, char* dest, int originalSize)
+{
+    LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
+    int result;
+
+    result = LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, usingExtDict, lz4sd->dictionary, lz4sd->dictSize);
+    if (result <= 0) return result;
+    if (lz4sd->dictionary + lz4sd->dictSize == dest)
+    {
+        lz4sd->dictSize += result;
+    }
+    else
+    {
+        lz4sd->dictionary = dest;
+        lz4sd->dictSize = result;
+    }
+
+    return result;
+}
+
+
+/*
+Advanced decoding functions :
+*_usingDict() :
+    These decoding functions work the same as "_continue" ones,
+    the dictionary must be explicitly provided within parameters
+*/
+
+int LZ4_decompress_safe_usingDict(const char* source, char* dest, int compressedSize, int maxOutputSize, const char* dictStart, int dictSize)
+{
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, usingExtDict, dictStart, dictSize);
+}
+
+int LZ4_decompress_fast_usingDict(const char* source, char* dest, int originalSize, const char* dictStart, int dictSize)
+{
+    return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, usingExtDict, dictStart, dictSize);
+}
+
+
+/***************************************************
+    Obsolete Functions
+***************************************************/
+/*
+These function names are deprecated and should no longer be used.
+They are only provided here for compatibility with older user programs.
+- LZ4_uncompress is totally equivalent to LZ4_decompress_fast
+- LZ4_uncompress_unknownOutputSize is totally equivalent to LZ4_decompress_safe
+*/
+int LZ4_uncompress (const char* source, char* dest, int outputSize) { return LZ4_decompress_fast(source, dest, outputSize); }
+int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize) { return LZ4_decompress_safe(source, dest, isize, maxOutputSize); }
+
+
+/* Obsolete Streaming functions */
+
+int LZ4_sizeofStreamState() { return LZ4_STREAMSIZE; }
+
+void LZ4_init(LZ4_stream_t_internal* lz4ds, const BYTE* base)
+{
+    MEM_INIT(lz4ds, 0, LZ4_STREAMSIZE);
+    lz4ds->bufferStart = base;
+}
+
+int LZ4_resetStreamState(void* state, const char* inputBuffer)
+{
+    if ((((size_t)state) & 3) != 0) return 1;   /* Error : pointer is not aligned on 4-bytes boundary */
+    LZ4_init((LZ4_stream_t_internal*)state, (const BYTE*)inputBuffer);
+    return 0;
+}
+
+void* LZ4_create (const char* inputBuffer)
+{
+    void* lz4ds = ALLOCATOR(4, LZ4_STREAMSIZE_U32);
+    LZ4_init ((LZ4_stream_t_internal*)lz4ds, (const BYTE*)inputBuffer);
+    return lz4ds;
+}
+
+char* LZ4_slideInputBuffer (void* LZ4_Data)
+{
+    LZ4_stream_t_internal* lz4ds = (LZ4_stream_t_internal*)LZ4_Data;
+
+    LZ4_saveDict((LZ4_stream_t*)LZ4_Data, (char*)lz4ds->bufferStart, 64 KB);
+
+    return (char*)(lz4ds->bufferStart + 64 KB);
+}
+
+/*  Obsolete compresson functions using User-allocated state */
+
+int LZ4_sizeofState() { return LZ4_STREAMSIZE; }
+
+int LZ4_compress_withState (void* state, const char* source, char* dest, int inputSize)
+{
+    if (((size_t)(state)&3) != 0) return 0;   /* Error : state is not aligned on 4-bytes boundary */
+    MEM_INIT(state, 0, LZ4_STREAMSIZE);
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        return LZ4_compress_generic(state, source, dest, inputSize, 0, notLimited, byU16, noDict, noDictIssue);
+    else
+        return LZ4_compress_generic(state, source, dest, inputSize, 0, notLimited, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
+}
+
+int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+    if (((size_t)(state)&3) != 0) return 0;   /* Error : state is not aligned on 4-bytes boundary */
+    MEM_INIT(state, 0, LZ4_STREAMSIZE);
+
+    if (inputSize < (int)LZ4_64KLIMIT)
+        return LZ4_compress_generic(state, source, dest, inputSize, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue);
+    else
+        return LZ4_compress_generic(state, source, dest, inputSize, maxOutputSize, limitedOutput, (sizeof(void*)==8) ? byU32 : byPtr, noDict, noDictIssue);
+}
+
+/* Obsolete streaming decompression functions */
+
+int LZ4_decompress_safe_withPrefix64k(const char* source, char* dest, int compressedSize, int maxOutputSize)
+{
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, withPrefix64k, NULL, 64 KB);
+}
+
+int LZ4_decompress_fast_withPrefix64k(const char* source, char* dest, int originalSize)
+{
+    return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, withPrefix64k, NULL, 64 KB);
+}
-/*
-   LZ4 - Fast LZ compression algorithm
-   Header File
-   Copyright (C) 2011-2012, 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/
-*/
-#pragma once
-
-#if defined (__cplusplus)
-extern "C" {
-#endif
-
-
-//**************************************
-// Compiler Options
-//**************************************
-#if defined(_MSC_VER) && !defined(__cplusplus)   // Visual Studio
-#  define inline __inline           // Visual is not C99, but supports some kind of inline
-#endif
-
-
-//****************************
-// Simple Functions
-//****************************
-
-int LZ4_compress   (const char* source, char* dest, int isize);
-int LZ4_uncompress (const char* source, char* dest, int osize);
-
-/*
-LZ4_compress() :
-    Compresses 'isize' bytes from 'source' into 'dest'.
-    Destination buffer must be already allocated,
-    and must be sized to handle worst cases situations (input data not compressible)
-    Worst case size evaluation is provided by function LZ4_compressBound()
-
-    isize  : is the input size. Max supported value is ~1.9GB
-    return : the number of bytes written in buffer dest
-
-
-LZ4_uncompress() :
-    osize  : is the output size, therefore the original size
-    return : the number of bytes read in the source buffer
-             If the source stream is malformed, the function will stop decoding and return a negative result, indicating the byte position of the faulty instruction
-             This function never writes outside of provided buffers, and never modifies input buffer.
-    note : destination buffer must be already allocated.
-           its size must be a minimum of 'osize' bytes.
-*/
-
-