Upgrade LZ4.

2 files changed, 529 insertions, 422 deletions

diff --git a/ext/lz4/lz4.c b/ext/lz4/lz4.c
index d58be27a..ed928ced 100644
--- a/ext/lz4/lz4.c
+++ b/ext/lz4/lz4.c
@@ -1,6 +1,6 @@
 /*
    LZ4 - Fast LZ compression algorithm
-   Copyright (C) 2011-2014, Yann Collet.
+   Copyright (C) 2011-2015, 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
@@ -27,84 +27,78 @@
    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 source repository : http://code.google.com/p/lz4
+   - LZ4 source mirror : https://github.com/Cyan4973/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)).
+ * in memory stack (0:default, fastest), or in memory heap (1:requires malloc()).
  */
 #define HEAPMODE 0
 
+/*
+ * CPU_HAS_EFFICIENT_UNALIGNED_MEMORY_ACCESS :
+ * By default, the source code expects the compiler to correctly optimize
+ * 4-bytes and 8-bytes read on architectures able to handle it efficiently.
+ * This is not always the case. In some circumstances (ARM notably),
+ * the compiler will issue cautious code even when target is able to correctly handle unaligned memory accesses.
+ *
+ * You can force the compiler to use unaligned memory access by uncommenting the line below.
+ * One of the below scenarios will happen :
+ * 1 - Your target CPU correctly handle unaligned access, and was not well optimized by compiler (good case).
+ *     You will witness large performance improvements (+50% and up).
+ *     Keep the line uncommented and send a word to upstream (https://groups.google.com/forum/#!forum/lz4c)
+ *     The goal is to automatically detect such situations by adding your target CPU within an exception list.
+ * 2 - Your target CPU correctly handle unaligned access, and was already already optimized by compiler
+ *     No change will be experienced.
+ * 3 - Your target CPU inefficiently handle unaligned access.
+ *     You will experience a performance loss. Comment back the line.
+ * 4 - Your target CPU does not handle unaligned access.
+ *     Program will crash.
+ * If uncommenting results in better performance (case 1)
+ * please report your configuration to upstream (https://groups.google.com/forum/#!forum/lz4c)
+ * An automatic detection macro will be added to match your case within future versions of the library.
+ */
+/* #define CPU_HAS_EFFICIENT_UNALIGNED_MEMORY_ACCESS 1 */
+
 
 /**************************************
    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
-#define LZ4_32BITS (sizeof(void*)==4)
-#define LZ4_64BITS (sizeof(void*)==8)
-
 /*
- * Little Endian or Big Endian ?
- * Overwrite the #define below if you know your architecture endianess
+ * Automated efficient unaligned memory access detection
+ * Based on known hardware architectures
+ * This list will be updated thanks to feedbacks
  */
-#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
+#if defined(CPU_HAS_EFFICIENT_UNALIGNED_MEMORY_ACCESS) \
+    || defined(__ARM_FEATURE_UNALIGNED) \
+    || defined(__i386__) || defined(__x86_64__) \
+    || defined(_M_IX86) || defined(_M_X64) \
+    || defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_8__) \
+    || (defined(_M_ARM) && (_M_ARM >= 7))
+#  define LZ4_UNALIGNED_ACCESS 1
 #else
-/* Little Endian assumed. PDP Endian and other very rare endian format are unsupported. */
+#  define LZ4_UNALIGNED_ACCESS 0
 #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
+ * LZ4_FORCE_SW_BITCOUNT
+ * Define this parameter if your target system or compiler does not support hardware bit count
  */
-#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
+   Compiler Options
 **************************************/
 #if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)   /* C99 */
 /* "restrict" is a known keyword */
@@ -114,28 +108,20 @@
 
 #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
+#  include <intrin.h>
 #  pragma warning(disable : 4127)        /* disable: C4127: conditional expression is constant */
+#  pragma warning(disable : 4293)        /* disable: C4293: too large shift (32-bits) */
 #else
-#  ifdef __GNUC__
-#    define FORCE_INLINE static inline __attribute__((always_inline))
+#  if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)   /* C99 */
+#    ifdef __GNUC__
+#      define FORCE_INLINE static inline __attribute__((always_inline))
+#    else
+#      define FORCE_INLINE static inline
+#    endif
 #  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 FORCE_INLINE static
+#  endif   /* __STDC_VERSION__ */
+#endif  /* _MSC_VER */
 
 #define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
 
@@ -183,46 +169,132 @@
   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
+/**************************************
+   Reading and writing into memory
+**************************************/
+#define STEPSIZE sizeof(size_t)
 
-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;
+static unsigned LZ4_64bits(void) { return sizeof(void*)==8; }
 
-#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
-#  if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
-#    pragma pack(0)
-#  else
-#    pragma pack(pop)
-#  endif
+static unsigned LZ4_isLittleEndian(void)
+{
+    const union { U32 i; BYTE c[4]; } one = { 1 };   /* don't use static : performance detrimental  */
+    return one.c[0];
+}
+
+
+static U16 LZ4_readLE16(const void* memPtr)
+{
+    if ((LZ4_UNALIGNED_ACCESS) && (LZ4_isLittleEndian()))
+        return *(U16*)memPtr;
+    else
+    {
+        const BYTE* p = memPtr;
+        return (U16)((U16)p[0] + (p[1]<<8));
+    }
+}
+
+static void LZ4_writeLE16(void* memPtr, U16 value)
+{
+    if ((LZ4_UNALIGNED_ACCESS) && (LZ4_isLittleEndian()))
+    {
+        *(U16*)memPtr = value;
+        return;
+    }
+    else
+    {
+        BYTE* p = memPtr;
+        p[0] = (BYTE) value;
+        p[1] = (BYTE)(value>>8);
+    }
+}
+
+
+static U16 LZ4_read16(const void* memPtr)
+{
+    if (LZ4_UNALIGNED_ACCESS)
+        return *(U16*)memPtr;
+    else
+    {
+        U16 val16;
+        memcpy(&val16, memPtr, 2);
+        return val16;
+    }
+}
+
+static U32 LZ4_read32(const void* memPtr)
+{
+    if (LZ4_UNALIGNED_ACCESS)
+        return *(U32*)memPtr;
+    else
+    {
+        U32 val32;
+        memcpy(&val32, memPtr, 4);
+        return val32;
+    }
+}
+
+static U64 LZ4_read64(const void* memPtr)
+{
+    if (LZ4_UNALIGNED_ACCESS)
+        return *(U64*)memPtr;
+    else
+    {
+        U64 val64;
+        memcpy(&val64, memPtr, 8);
+        return val64;
+    }
+}
+
+static size_t LZ4_read_ARCH(const void* p)
+{
+    if (LZ4_64bits())
+        return (size_t)LZ4_read64(p);
+    else
+        return (size_t)LZ4_read32(p);
+}
+
+
+static void LZ4_copy4(void* dstPtr, const void* srcPtr)
+{
+    if (LZ4_UNALIGNED_ACCESS)
+    {
+        *(U32*)dstPtr = *(U32*)srcPtr;
+        return;
+    }
+    memcpy(dstPtr, srcPtr, 4);
+}
+
+static void LZ4_copy8(void* dstPtr, const void* srcPtr)
+{
+#if GCC_VERSION!=409  /* disabled on GCC 4.9, as it generates invalid opcode (crash) */
+    if (LZ4_UNALIGNED_ACCESS)
+    {
+        if (LZ4_64bits())
+            *(U64*)dstPtr = *(U64*)srcPtr;
+        else
+            ((U32*)dstPtr)[0] = ((U32*)srcPtr)[0],
+            ((U32*)dstPtr)[1] = ((U32*)srcPtr)[1];
+        return;
+    }
 #endif
+    memcpy(dstPtr, srcPtr, 8);
+}
 
-#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)
+/* customized version of memcpy, which may overwrite up to 7 bytes beyond dstEnd */
+static void LZ4_wildCopy(void* dstPtr, const void* srcPtr, void* dstEnd)
+{
+    BYTE* d = dstPtr;
+    const BYTE* s = srcPtr;
+    BYTE* e = dstEnd;
+    do { LZ4_copy8(d,s); d+=8; s+=8; } while (d<e);
+}
 
 
 /**************************************
-   Constants
+   Common 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
@@ -230,13 +302,10 @@ typedef struct {size_t v;} _PACKED size_t_S;
 #define MFLIMIT (COPYLENGTH+MINMATCH)
 static const int LZ4_minLength = (MFLIMIT+1);
 
-#define KB *(1U<<10)
-#define MB *(1U<<20)
+#define KB *(1 <<10)
+#define MB *(1 <<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)
 
@@ -247,130 +316,147 @@ static const int LZ4_minLength = (MFLIMIT+1);
 
 
 /**************************************
-   Structures and local types
+   Common Utils
 **************************************/
-typedef struct {
-    U32  hashTable[HASH_SIZE_U32];
-    U32  currentOffset;
-    U32  initCheck;
-    const BYTE* dictionary;
-    const BYTE* bufferStart;
-    U32  dictSize;
-} LZ4_stream_t_internal;
+#define LZ4_STATIC_ASSERT(c)    { enum { LZ4_static_assert = 1/(int)(!!(c)) }; }   /* use only *after* variable declarations */
 
-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;
+/********************************
+   Common functions
+********************************/
+static unsigned LZ4_NbCommonBytes (register size_t val)
+{
+    if (LZ4_isLittleEndian())
+    {
+        if (LZ4_64bits())
+        {
+#       if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            unsigned long r = 0;
+            _BitScanForward64( &r, (U64)val );
+            return (int)(r>>3);
+#       elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            return (__builtin_ctzll((U64)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
+        }
+        else /* 32 bits */
+        {
+#       if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            unsigned long r;
+            _BitScanForward( &r, (U32)val );
+            return (int)(r>>3);
+#       elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            return (__builtin_ctz((U32)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
+        }
+    }
+    else   /* Big Endian CPU */
+    {
+        if (LZ4_64bits())
+        {
+#       if defined(_MSC_VER) && defined(_WIN64) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            unsigned long r = 0;
+            _BitScanReverse64( &r, val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            return (__builtin_clzll(val) >> 3);
+#       else
+            unsigned 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 /* 32 bits */
+        {
+#       if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            unsigned long r = 0;
+            _BitScanReverse( &r, (unsigned long)val );
+            return (unsigned)(r>>3);
+#       elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+            return (__builtin_clz(val) >> 3);
+#       else
+            unsigned r;
+            if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+            r += (!val);
+            return r;
+#       endif
+        }
+    }
+}
 
-typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
-typedef enum { full = 0, partial = 1 } earlyEnd_directive;
+static unsigned LZ4_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* pInLimit)
+{
+    const BYTE* const pStart = pIn;
 
+    while (likely(pIn<pInLimit-(STEPSIZE-1)))
+    {
+        size_t diff = LZ4_read_ARCH(pMatch) ^ LZ4_read_ARCH(pIn);
+        if (!diff) { pIn+=STEPSIZE; pMatch+=STEPSIZE; continue; }
+        pIn += LZ4_NbCommonBytes(diff);
+        return (unsigned)(pIn - pStart);
+    }
 
+    if (LZ4_64bits()) if ((pIn<(pInLimit-3)) && (LZ4_read32(pMatch) == LZ4_read32(pIn))) { pIn+=4; pMatch+=4; }
+    if ((pIn<(pInLimit-1)) && (LZ4_read16(pMatch) == LZ4_read16(pIn))) { pIn+=2; pMatch+=2; }
+    if ((pIn<pInLimit) && (*pMatch == *pIn)) pIn++;
+    return (unsigned)(pIn - pStart);
+}
+
+
+#ifndef LZ4_COMMONDEFS_ONLY
 /**************************************
-   Architecture-specific macros
+   Local Constants
 **************************************/
-#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
+#define LZ4_HASHLOG   (LZ4_MEMORY_USAGE-2)
+#define HASHTABLESIZE (1 << LZ4_MEMORY_USAGE)
+#define HASH_SIZE_U32 (1 << LZ4_HASHLOG)       /* required as macro for static allocation */
+
+static const int LZ4_64Klimit = ((64 KB) + (MFLIMIT-1));
+static const U32 LZ4_skipTrigger = 6;  /* Increase this value ==> compression run slower on incompressible data */
 
 
 /**************************************
-   Macros
+   Local Utils
 **************************************/
-#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
+int LZ4_versionNumber (void) { return LZ4_VERSION_NUMBER; }
+int LZ4_compressBound(int isize)  { return LZ4_COMPRESSBOUND(isize); }
 
 
-/****************************
-   Private local functions
-****************************/
-#if LZ4_ARCH64
+/**************************************
+   Local Structures and types
+**************************************/
+typedef struct {
+    U32 hashTable[HASH_SIZE_U32];
+    U32 currentOffset;
+    U32 initCheck;
+    const BYTE* dictionary;
+    const BYTE* bufferStart;
+    U32 dictSize;
+} LZ4_stream_t_internal;
 
-static 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
-}
+typedef enum { notLimited = 0, limitedOutput = 1 } limitedOutput_directive;
+typedef enum { byPtr, byU32, byU16 } tableType_t;
 
-#else
+typedef enum { noDict = 0, withPrefix64k, usingExtDict } dict_directive;
+typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive;
 
-static 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
-}
+typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
+typedef enum { full = 0, partial = 1 } earlyEnd_directive;
 
-#endif
 
 
 /********************************
    Compression functions
 ********************************/
-int LZ4_versionNumber (void) { return LZ4_VERSION_NUMBER; }
-int LZ4_compressBound(int isize)  { return LZ4_COMPRESSBOUND(isize); }
 
-static int LZ4_hashSequence(U32 sequence, tableType_t tableType)
+static U32 LZ4_hashSequence(U32 sequence, tableType_t tableType)
 {
     if (tableType == byU16)
         return (((sequence) * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
@@ -378,15 +464,15 @@ static int LZ4_hashSequence(U32 sequence, tableType_t tableType)
         return (((sequence) * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
 }
 
-static int LZ4_hashPosition(const BYTE* p, tableType_t tableType) { return LZ4_hashSequence(A32(p), tableType); }
+static U32 LZ4_hashPosition(const BYTE* p, tableType_t tableType) { return LZ4_hashSequence(LZ4_read32(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; }
+    case byPtr: { const BYTE** hashTable = (const BYTE**)tableBase; hashTable[h] = p; return; }
+    case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = (U32)(p-srcBase); return; }
+    case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = (U16)(p-srcBase); return; }
     }
 }
 
@@ -409,32 +495,12 @@ static const BYTE* LZ4_getPosition(const BYTE* p, void* tableBase, tableType_t t
     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 (LZ4_64BITS) 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,
@@ -457,7 +523,6 @@ static int LZ4_compress_generic(
     BYTE* op = (BYTE*) dest;
     BYTE* const olimit = op + maxOutputSize;
 
-    const int skipStrength = SKIPSTRENGTH;
     U32 forwardH;
     size_t refDelta=0;
 
@@ -479,8 +544,8 @@ static int LZ4_compress_generic(
         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) */
+    if ((tableType == byU16) && (inputSize>=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);
@@ -489,26 +554,26 @@ static int LZ4_compress_generic(
     /* Main Loop */
     for ( ; ; )
     {
-        const BYTE* ref;
+        const BYTE* match;
         BYTE* token;
         {
             const BYTE* forwardIp = ip;
             unsigned step=1;
-            unsigned searchMatchNb = (1U << skipStrength);
+            unsigned searchMatchNb = (1U << LZ4_skipTrigger);
 
             /* Find a match */
             do {
                 U32 h = forwardH;
                 ip = forwardIp;
                 forwardIp += step;
-                step = searchMatchNb++ >> skipStrength;
+                step = searchMatchNb++ >> LZ4_skipTrigger;
 
                 if (unlikely(forwardIp > mflimit)) goto _last_literals;
 
-                ref = LZ4_getPositionOnHash(h, ctx, tableType, base);
+                match = LZ4_getPositionOnHash(h, ctx, tableType, base);
                 if (dict==usingExtDict)
                 {
-                    if (ref<(const BYTE*)source)
+                    if (match<(const BYTE*)source)
                     {
                         refDelta = dictDelta;
                         lowLimit = dictionary;
@@ -522,13 +587,13 @@ static int LZ4_compress_generic(
                 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)) );
+            } while ( ((dictIssue==dictSmall) ? (match < lowRefLimit) : 0)
+                || ((tableType==byU16) ? 0 : (match + MAX_DISTANCE < ip))
+                || (LZ4_read32(match+refDelta) != LZ4_read32(ip)) );
         }
 
         /* Catch up */
-        while ((ip>anchor) && (ref+refDelta > lowLimit) && (unlikely(ip[-1]==ref[refDelta-1]))) { ip--; ref--; }
+        while ((ip>anchor) && (match+refDelta > lowLimit) && (unlikely(ip[-1]==match[refDelta-1]))) { ip--; match--; }
 
         {
             /* Encode Literal length */
@@ -546,12 +611,13 @@ static int LZ4_compress_generic(
             else *token = (BYTE)(litLength<<ML_BITS);
 
             /* Copy Literals */
-            { BYTE* end = op+litLength; LZ4_WILDCOPY(op,anchor,end); op=end; }
+            LZ4_wildCopy(op, anchor, op+litLength);
+            op+=litLength;
         }
 
 _next_match:
         /* Encode Offset */
-        LZ4_WRITE_LITTLEENDIAN_16(op, (U16)(ip-ref));
+        LZ4_writeLE16(op, (U16)(ip-match)); op+=2;
 
         /* Encode MatchLength */
         {
@@ -560,10 +626,10 @@ _next_match:
             if ((dict==usingExtDict) && (lowLimit==dictionary))
             {
                 const BYTE* limit;
-                ref += refDelta;
-                limit = ip + (dictEnd-ref);
+                match += refDelta;
+                limit = ip + (dictEnd-match);
                 if (limit > matchlimit) limit = matchlimit;
-                matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, limit);
+                matchLength = LZ4_count(ip+MINMATCH, match+MINMATCH, limit);
                 ip += MINMATCH + matchLength;
                 if (ip==limit)
                 {
@@ -574,14 +640,14 @@ _next_match:
             }
             else
             {
-                matchLength = LZ4_count(ip+MINMATCH, ref+MINMATCH, matchlimit);
+                matchLength = LZ4_count(ip+MINMATCH, match+MINMATCH, matchlimit);
                 ip += MINMATCH + matchLength;
             }
 
+            if ((outputLimited) && (unlikely(op + (1 + LASTLITERALS) + (matchLength>>8) > olimit)))
+                return 0;    /* Check output limit */
             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; }
@@ -600,10 +666,10 @@ _next_match:
         LZ4_putPosition(ip-2, ctx, tableType, base);
 
         /* Test next position */
-        ref = LZ4_getPosition(ip, ctx, tableType, base);
+        match = LZ4_getPosition(ip, ctx, tableType, base);
         if (dict==usingExtDict)
         {
-            if (ref<(const BYTE*)source)
+            if (match<(const BYTE*)source)
             {
                 refDelta = dictDelta;
                 lowLimit = dictionary;
@@ -615,9 +681,9 @@ _next_match:
             }
         }
         LZ4_putPosition(ip, ctx, tableType, base);
-        if ( ((dictIssue==dictSmall) ? (ref>=lowRefLimit) : 1)
-            && (ref+MAX_DISTANCE>=ip)
-            && (A32(ref+refDelta)==A32(ip)) )
+        if ( ((dictIssue==dictSmall) ? (match>=lowRefLimit) : 1)
+            && (match+MAX_DISTANCE>=ip)
+            && (LZ4_read32(match+refDelta)==LZ4_read32(ip)) )
         { token=op++; *token=0; goto _next_match; }
 
         /* Prepare next loop */
@@ -644,16 +710,16 @@ _last_literals:
 int LZ4_compress(const char* source, char* dest, int inputSize)
 {
 #if (HEAPMODE)
-    void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U32, 4);   /* Aligned on 4-bytes boundaries */
+    void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U64, 8);   /* Aligned on 8-bytes boundaries */
 #else
-    U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
+    U64 ctx[LZ4_STREAMSIZE_U64] = {0};      /* Ensure data is aligned on 8-bytes boundaries */
 #endif
     int result;
 
-    if (inputSize < (int)LZ4_64KLIMIT)
+    if (inputSize < 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, LZ4_64BITS ? byU32 : byPtr, noDict, noDictIssue);
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, 0, notLimited, LZ4_64bits() ? byU32 : byPtr, noDict, noDictIssue);
 
 #if (HEAPMODE)
     FREEMEM(ctx);
@@ -664,16 +730,16 @@ int LZ4_compress(const char* source, char* dest, int inputSize)
 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 */
+    void* ctx = ALLOCATOR(LZ4_STREAMSIZE_U64, 8);   /* Aligned on 8-bytes boundaries */
 #else
-    U32 ctx[LZ4_STREAMSIZE_U32] = {0};      /* Ensure data is aligned on 4-bytes boundaries */
+    U64 ctx[LZ4_STREAMSIZE_U64] = {0};      /* Ensure data is aligned on 8-bytes boundaries */
 #endif
     int result;
 
-    if (inputSize < (int)LZ4_64KLIMIT)
+    if (inputSize < 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, LZ4_64BITS ? byU32 : byPtr, noDict, noDictIssue);
+        result = LZ4_compress_generic((void*)ctx, source, dest, inputSize, maxOutputSize, limitedOutput, LZ4_64bits() ? byU32 : byPtr, noDict, noDictIssue);
 
 #if (HEAPMODE)
     FREEMEM(ctx);
@@ -698,7 +764,8 @@ void LZ4_resetStream (LZ4_stream_t* LZ4_stream)
 
 LZ4_stream_t* LZ4_createStream(void)
 {
-    LZ4_stream_t* lz4s = (LZ4_stream_t*)ALLOCATOR(4, LZ4_STREAMSIZE_U32);
+    LZ4_stream_t* lz4s = (LZ4_stream_t*)ALLOCATOR(8, LZ4_STREAMSIZE_U64);
+    LZ4_STATIC_ASSERT(LZ4_STREAMSIZE >= sizeof(LZ4_stream_t_internal));    /* A compilation error here means LZ4_STREAMSIZE is not large enough */
     LZ4_resetStream(lz4s);
     return lz4s;
 }
@@ -717,14 +784,13 @@ int LZ4_loadDict (LZ4_stream_t* LZ4_dict, const char* dictionary, int dictSize)
     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) LZ4_resetStream(LZ4_dict);                         /* Uninitialized structure detected */
 
     if (dictSize < MINMATCH)
     {
         dict->dictionary = NULL;
         dict->dictSize = 0;
-        return 1;
+        return 0;
     }
 
     if (p <= dictEnd - 64 KB) p = dictEnd - 64 KB;
@@ -739,7 +805,7 @@ int LZ4_loadDict (LZ4_stream_t* LZ4_dict, const char* dictionary, int dictSize)
         p+=3;
     }
 
-    return 1;
+    return dict->dictSize;
 }
 
 
@@ -854,12 +920,12 @@ int LZ4_saveDict (LZ4_stream_t* LZ4_dict, char* safeBuffer, int 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);
+    memmove(safeBuffer, previousDictEnd - dictSize, dictSize);
 
     dict->dictionary = (const BYTE*)safeBuffer;
     dict->dictSize = (U32)dictSize;
 
-    return 1;
+    return dictSize;
 }
 
 
@@ -869,13 +935,13 @@ int LZ4_saveDict (LZ4_stream_t* LZ4_dict, char* safeBuffer, int dictSize)
 ****************************/
 /*
  * This generic decompression function cover all use cases.
- * It shall be instanciated several times, using different sets of directives
+ * It shall be instantiated 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.
+ * in order to remove useless branches during compilation optimization.
  */
 FORCE_INLINE int LZ4_decompress_generic(
-                 const char* source,
-                 char* dest,
+                 const char* const source,
+                 char* const dest,
                  int inputSize,
                  int outputSize,         /* If endOnInput==endOnInputSize, this value is the max size of Output Buffer. */
 
@@ -883,24 +949,24 @@ FORCE_INLINE int LZ4_decompress_generic(
                  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 */
+                 const BYTE* const lowPrefix,  /* == dest if dict == noDict */
+                 const BYTE* const dictStart,  /* only if dict==usingExtDict */
+                 const size_t 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 lowLimit = lowPrefix - dictSize;
 
     const BYTE* const dictEnd = (const BYTE*)dictStart + dictSize;
-    const size_t dec32table[] = {4-0, 4-3, 4-2, 4-3, 4-0, 4-0, 4-0, 4-0};   /* note : static reduces speed for LZ4_decompress_safe() on GCC64 */
-    static const size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
+    const size_t dec32table[] = {4, 1, 2, 1, 4, 4, 4, 4};
+    const size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
 
     const int safeDecode = (endOnInput==endOnInputSize);
     const int checkOffset = ((safeDecode) && (dictSize < (int)(64 KB)));
@@ -917,8 +983,9 @@ FORCE_INLINE int LZ4_decompress_generic(
     {
         unsigned token;
         size_t length;
+        const BYTE* match;
 
-        /* get runlength */
+        /* get literal length */
         token = *ip++;
         if ((length=(token>>ML_BITS)) == RUN_MASK)
         {
@@ -929,8 +996,8 @@ FORCE_INLINE int LZ4_decompress_generic(
                 length += s;
             }
             while (likely((endOnInput)?ip<iend-RUN_MASK:1) && (s==255));
-            if ((safeDecode) && LZ4_32BITS && unlikely((size_t)(op+length)<(size_t)(op))) goto _output_error;   /* overflow detection */
-            if ((safeDecode) && LZ4_32BITS && unlikely((size_t)(ip+length)<(size_t)(ip))) goto _output_error;   /* overflow detection */
+            if ((safeDecode) && unlikely((size_t)(op+length)<(size_t)(op))) goto _output_error;   /* overflow detection */
+            if ((safeDecode) && unlikely((size_t)(ip+length)<(size_t)(ip))) goto _output_error;   /* overflow detection */
         }
 
         /* copy literals */
@@ -953,14 +1020,16 @@ FORCE_INLINE int LZ4_decompress_generic(
             op += length;
             break;     /* Necessarily EOF, due to parsing restrictions */
         }
-        LZ4_WILDCOPY(op, ip, cpy); ip -= (op-cpy); op = cpy;
+        LZ4_wildCopy(op, ip, cpy);
+        ip += length; 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 */
+        match = cpy - LZ4_readLE16(ip); ip+=2;
+        if ((checkOffset) && (unlikely(match < lowLimit))) goto _output_error;   /* Error : offset outside destination buffer */
 
         /* get matchlength */
-        if ((length=(token&ML_MASK)) == ML_MASK)
+        length = token & ML_MASK;
+        if (length == ML_MASK)
         {
             unsigned s;
             do
@@ -969,35 +1038,38 @@ FORCE_INLINE int LZ4_decompress_generic(
                 s = *ip++;
                 length += s;
             } while (s==255);
-            if ((safeDecode) && LZ4_32BITS && unlikely((size_t)(op+length)<(size_t)op)) goto _output_error;   /* overflow detection */
+            if ((safeDecode) && unlikely((size_t)(op+length)<(size_t)op)) goto _output_error;   /* overflow detection */
         }
+        length += MINMATCH;
 
         /* check external dictionary */
-        if ((dict==usingExtDict) && (ref < (BYTE* const)dest))
+        if ((dict==usingExtDict) && (match < lowPrefix))
         {
-            if (unlikely(op+length+MINMATCH > oend-LASTLITERALS)) goto _output_error;
+            if (unlikely(op+length > oend-LASTLITERALS)) goto _output_error;   /* doesn't respect parsing restriction */
 
-            if (length+MINMATCH <= (size_t)(dest-(char*)ref))
+            if (length <= (size_t)(lowPrefix-match))
             {
-                ref = dictEnd - (dest-(char*)ref);
-                memcpy(op, ref, length+MINMATCH);
-                op += length+MINMATCH;
+                /* match can be copied as a single segment from external dictionary */
+                match = dictEnd - (lowPrefix-match);
+                memcpy(op, match, length);
+                op += length;
             }
             else
             {
-                size_t copySize = (size_t)(dest-(char*)ref);
+                /* match encompass external dictionary and current segment */
+                size_t copySize = (size_t)(lowPrefix-match);
                 memcpy(op, dictEnd - copySize, copySize);
                 op += copySize;
-                copySize = length+MINMATCH - copySize;
-                if (copySize > (size_t)((char*)op-dest))   /* overlap */
+                copySize = length - copySize;
+                if (copySize > (size_t)(op-lowPrefix))   /* overlap within current segment */
                 {
                     BYTE* const endOfMatch = op + copySize;
-                    const BYTE* copyFrom = (BYTE*)dest;
+                    const BYTE* copyFrom = lowPrefix;
                     while (op < endOfMatch) *op++ = *copyFrom++;
                 }
                 else
                 {
-                    memcpy(op, dest, copySize);
+                    memcpy(op, lowPrefix, copySize);
                     op += copySize;
                 }
             }
@@ -1005,28 +1077,32 @@ FORCE_INLINE int LZ4_decompress_generic(
         }
 
         /* copy repeated sequence */
-        if (unlikely((op-ref)<(int)STEPSIZE))
+        cpy = op + length;
+        if (unlikely((op-match)<8))
         {
-            const size_t dec64 = dec64table[LZ4_32BITS ? 0 : op-ref];
-            op[0] = ref[0];
-            op[1] = ref[1];
-            op[2] = ref[2];
-            op[3] = ref[3];
-            ref += dec32table[op-ref];
-            A32(op+4) = A32(ref);
-            op += STEPSIZE; ref -= dec64;
-        } else { LZ4_COPYSTEP(op,ref); }
-        cpy = op + length - (STEPSIZE-4);
-
-        if (unlikely(cpy>oend-COPYLENGTH-(STEPSIZE-4)))
+            const size_t dec64 = dec64table[op-match];
+            op[0] = match[0];
+            op[1] = match[1];
+            op[2] = match[2];
+            op[3] = match[3];
+            match += dec32table[op-match];
+            LZ4_copy4(op+4, match);
+            op += 8; match -= dec64;
+        } else { LZ4_copy8(op, match); op+=8; match+=8; }
+
+        if (unlikely(cpy>oend-12))
         {
-            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;
+            if (cpy > oend-LASTLITERALS) goto _output_error;    /* Error : last LASTLITERALS bytes must be literals */
+            if (op < oend-8)
+            {
+                LZ4_wildCopy(op, match, oend-8);
+                match += (oend-8) - op;
+                op = oend-8;
+            }
+            while (op<cpy) *op++ = *match++;
         }
-        LZ4_WILDCOPY(op, ref, cpy);
+        else
+            LZ4_wildCopy(op, match, cpy);
         op=cpy;   /* correction */
     }
 
@@ -1044,25 +1120,28 @@ _output_error:
 
 int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int maxDecompressedSize)
 {
-    return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize, endOnInputSize, full, 0, noDict, NULL, 0);
+    return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize, endOnInputSize, full, 0, noDict, (BYTE*)dest, NULL, 0);
 }
 
 int LZ4_decompress_safe_partial(const char* source, char* dest, int compressedSize, int targetOutputSize, int maxDecompressedSize)
 {
-    return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize, endOnInputSize, partial, targetOutputSize, noDict, NULL, 0);
+    return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize, endOnInputSize, partial, targetOutputSize, noDict, (BYTE*)dest, 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, 64 KB);
+    return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, withPrefix64k, (BYTE*)(dest - 64 KB), NULL, 64 KB);
 }
 
+
 /* streaming decompression functions */
 
 typedef struct
 {
-    const char* dictionary;
-    int dictSize;
+    BYTE* externalDict;
+    size_t extDictSize;
+    BYTE* prefixEnd;
+    size_t prefixSize;
 } LZ4_streamDecode_t_internal;
 
 /*
@@ -1072,8 +1151,7 @@ typedef struct
  */
 LZ4_streamDecode_t* LZ4_createStreamDecode(void)
 {
-    LZ4_streamDecode_t* lz4s = (LZ4_streamDecode_t*) ALLOCATOR(sizeof(U32), LZ4_STREAMDECODESIZE_U32);
-    MEM_INIT(lz4s, 0, LZ4_STREAMDECODESIZE);
+    LZ4_streamDecode_t* lz4s = (LZ4_streamDecode_t*) ALLOCATOR(sizeof(U64), LZ4_STREAMDECODESIZE_U64);
     return lz4s;
 }
 
@@ -1093,8 +1171,10 @@ int LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream)
 int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize)
 {
     LZ4_streamDecode_t_internal* lz4sd = (LZ4_streamDecode_t_internal*) LZ4_streamDecode;
-    lz4sd->dictionary = dictionary;
-    lz4sd->dictSize = dictSize;
+    lz4sd->prefixSize = (size_t) dictSize;
+    lz4sd->prefixEnd = (BYTE*) dictionary + dictSize;
+    lz4sd->externalDict = NULL;
+    lz4sd->extDictSize  = 0;
     return 1;
 }
 
@@ -1103,23 +1183,32 @@ int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dicti
     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()
+    and indicate where it stands using LZ4_setStreamDecode()
 */
 int LZ4_decompress_safe_continue (LZ4_streamDecode_t* 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)
+    if (lz4sd->prefixEnd == (BYTE*)dest)
     {
-        lz4sd->dictSize += result;
+        result = LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
+                                        endOnInputSize, full, 0,
+                                        usingExtDict, lz4sd->prefixEnd - lz4sd->prefixSize, lz4sd->externalDict, lz4sd->extDictSize);
+        if (result <= 0) return result;
+        lz4sd->prefixSize += result;
+        lz4sd->prefixEnd  += result;
     }
     else
     {
-        lz4sd->dictionary = dest;
-        lz4sd->dictSize = result;
+        lz4sd->extDictSize = lz4sd->prefixSize;
+        lz4sd->externalDict = lz4sd->prefixEnd - lz4sd->extDictSize;
+        result = LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
+                                        endOnInputSize, full, 0,
+                                        usingExtDict, (BYTE*)dest, lz4sd->externalDict, lz4sd->extDictSize);
+        if (result <= 0) return result;
+        lz4sd->prefixSize = result;
+        lz4sd->prefixEnd  = (BYTE*)dest + result;
     }
 
     return result;
@@ -1130,16 +1219,25 @@ int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const ch
     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)
+    if (lz4sd->prefixEnd == (BYTE*)dest)
     {
-        lz4sd->dictSize += result;
+        result = LZ4_decompress_generic(source, dest, 0, originalSize,
+                                        endOnOutputSize, full, 0,
+                                        usingExtDict, lz4sd->prefixEnd - lz4sd->prefixSize, lz4sd->externalDict, lz4sd->extDictSize);
+        if (result <= 0) return result;
+        lz4sd->prefixSize += originalSize;
+        lz4sd->prefixEnd  += originalSize;
     }
     else
     {
-        lz4sd->dictionary = dest;
-        lz4sd->dictSize = result;
+        lz4sd->extDictSize = lz4sd->prefixSize;
+        lz4sd->externalDict = (BYTE*)dest - lz4sd->extDictSize;
+        result = LZ4_decompress_generic(source, dest, 0, originalSize,
+                                        endOnOutputSize, full, 0,
+                                        usingExtDict, (BYTE*)dest, lz4sd->externalDict, lz4sd->extDictSize);
+        if (result <= 0) return result;
+        lz4sd->prefixSize = originalSize;
+        lz4sd->prefixEnd  = (BYTE*)dest + originalSize;
     }
 
     return result;
@@ -1153,14 +1251,33 @@ Advanced decoding functions :
     the dictionary must be explicitly provided within parameters
 */
 
+FORCE_INLINE int LZ4_decompress_usingDict_generic(const char* source, char* dest, int compressedSize, int maxOutputSize, int safe, const char* dictStart, int dictSize)
+{
+    if (dictSize==0)
+        return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, safe, full, 0, noDict, (BYTE*)dest, NULL, 0);
+    if (dictStart+dictSize == dest)
+    {
+        if (dictSize >= (int)(64 KB - 1))
+            return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, safe, full, 0, withPrefix64k, (BYTE*)dest-64 KB, NULL, 0);
+        return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, safe, full, 0, noDict, (BYTE*)dest-dictSize, NULL, 0);
+    }
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, safe, full, 0, usingExtDict, (BYTE*)dest, (BYTE*)dictStart, dictSize);
+}
+
 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);
+    return LZ4_decompress_usingDict_generic(source, dest, compressedSize, maxOutputSize, 1, 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);
+    return LZ4_decompress_usingDict_generic(source, dest, 0, originalSize, 0, dictStart, dictSize);
+}
+
+/* debug function */
+int LZ4_decompress_safe_forceExtDict(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, (BYTE*)dest, (BYTE*)dictStart, dictSize);
 }
 
 
@@ -1196,18 +1313,16 @@ int LZ4_resetStreamState(void* state, const char* inputBuffer)
 
 void* LZ4_create (const char* inputBuffer)
 {
-    void* lz4ds = ALLOCATOR(4, LZ4_STREAMSIZE_U32);
+    void* lz4ds = ALLOCATOR(8, LZ4_STREAMSIZE_U64);
     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);
+    LZ4_stream_t_internal* ctx = (LZ4_stream_t_internal*)LZ4_Data;
+    int dictSize = LZ4_saveDict((LZ4_stream_t*)ctx, (char*)ctx->bufferStart, 64 KB);
+    return (char*)(ctx->bufferStart + dictSize);
 }
 
 /*  Obsolete compresson functions using User-allocated state */
@@ -1219,10 +1334,10 @@ int LZ4_compress_withState (void* state, const char* source, char* dest, int inp
     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)
+    if (inputSize < 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, LZ4_64BITS ? byU32 : byPtr, noDict, noDictIssue);
+        return LZ4_compress_generic(state, source, dest, inputSize, 0, notLimited, LZ4_64bits() ? byU32 : byPtr, noDict, noDictIssue);
 }
 
 int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize)
@@ -1230,20 +1345,23 @@ int LZ4_compress_limitedOutput_withState (void* state, const char* source, char*
     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)
+    if (inputSize < 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, LZ4_64BITS ? byU32 : byPtr, noDict, noDictIssue);
+        return LZ4_compress_generic(state, source, dest, inputSize, maxOutputSize, limitedOutput, LZ4_64bits() ? 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);
+    return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize, endOnInputSize, full, 0, withPrefix64k, (BYTE*)dest - 64 KB, 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);
+    return LZ4_decompress_generic(source, dest, 0, originalSize, endOnOutputSize, full, 0, withPrefix64k, (BYTE*)dest - 64 KB, NULL, 64 KB);
 }
+
+#endif   /* LZ4_COMMONDEFS_ONLY */
+
diff --git a/ext/lz4/lz4.h b/ext/lz4/lz4.h
index f8327f0c..7778caad 100644
--- a/ext/lz4/lz4.h
+++ b/ext/lz4/lz4.h
@@ -37,12 +37,17 @@
 extern "C" {
 #endif
 
+/*
+ * lz4.h provides raw compression format functions, for optimal performance and integration into programs.
+ * If you need to generate data using an inter-operable format (respecting the framing specification),
+ * please use lz4frame.h instead.
+*/
 
 /**************************************
    Version
 **************************************/
-#define LZ4_VERSION_MAJOR    1    /* for major interface/format changes  */
-#define LZ4_VERSION_MINOR    3    /* for minor interface/format changes  */
+#define LZ4_VERSION_MAJOR    1    /* for breaking interface changes  */
+#define LZ4_VERSION_MINOR    5    /* for new (non-breaking) interface capabilities */
 #define LZ4_VERSION_RELEASE  0    /* for tweaks, bug-fixes, or development */
 #define LZ4_VERSION_NUMBER (LZ4_VERSION_MAJOR *100*100 + LZ4_VERSION_MINOR *100 + LZ4_VERSION_RELEASE)
 int LZ4_versionNumber (void);
@@ -64,12 +69,12 @@ int LZ4_versionNumber (void);
    Simple Functions
 **************************************/
 
-int LZ4_compress        (const char* source, char* dest, int inputSize);
+int LZ4_compress        (const char* source, char* dest, int sourceSize);
 int LZ4_decompress_safe (const char* source, char* dest, int compressedSize, int maxDecompressedSize);
 
 /*
 LZ4_compress() :
-    Compresses 'inputSize' bytes from 'source' into 'dest'.
+    Compresses 'sourceSize' 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()
@@ -83,16 +88,9 @@ LZ4_decompress_safe() :
     return : the number of bytes decompressed into the destination buffer (necessarily <= maxDecompressedSize)
              If the destination buffer is not large enough, decoding will stop and output an error code (<0).
              If the source stream is detected malformed, the function will stop decoding and return a negative result.
-             This function is protected against buffer overflow exploits :
-             it never writes outside of output buffer, and never reads outside of input buffer.
-             Therefore, it is protected against malicious data packets.
-*/
-
-
-/*
-Note :
-    Should you prefer to explicitly allocate compression-table memory using your own allocation method,
-    use the streaming functions provided below, simply reset the memory area between each call to LZ4_compress_continue()
+             This function is protected against buffer overflow exploits,
+             and never writes outside of output buffer, nor reads outside of input buffer.
+             It is also protected against malicious data packets.
 */
 
 
@@ -104,9 +102,9 @@ Note :
 
 /*
 LZ4_compressBound() :
-    Provides the maximum size that LZ4 may output in a "worst case" scenario (input data not compressible)
-    primarily useful for memory allocation of output buffer.
-    macro is also provided when result needs to be evaluated at compilation (such as stack memory allocation).
+    Provides the maximum size that LZ4 compression may output in a "worst case" scenario (input data not compressible)
+    This function is primarily useful for memory allocation purposes (output buffer size).
+    Macro LZ4_COMPRESSBOUND() is also provided for compilation-time evaluation (stack memory allocation for example).
 
     isize  : is the input size. Max supported value is LZ4_MAX_INPUT_SIZE
     return : maximum output size in a "worst case" scenario
@@ -117,16 +115,17 @@ int LZ4_compressBound(int isize);
 
 /*
 LZ4_compress_limitedOutput() :
-    Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+    Compress 'sourceSize' 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.
+    This saves time and memory on detecting non-compressible (or barely compressible) data.
     This function never writes outside of provided output buffer.
 
-    inputSize  : Max supported value is LZ4_MAX_INPUT_VALUE
+    sourceSize  : Max supported value is LZ4_MAX_INPUT_VALUE
     maxOutputSize : is the size of the destination buffer (which must be already allocated)
     return : the number of bytes written in buffer 'dest'
-             or 0 if the compression fails
+             or 0 if compression fails
 */
-int LZ4_compress_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize);
+int LZ4_compress_limitedOutput (const char* source, char* dest, int sourceSize, int maxOutputSize);
 
 
 /*
@@ -148,7 +147,7 @@ LZ4_decompress_fast() :
              Destination buffer must be already allocated. Its size must be a minimum of 'originalSize' bytes.
     note : This function fully respect memory boundaries for properly formed compressed data.
            It is a bit faster than LZ4_decompress_safe().
-           However, it does not provide any protection against intentionnally modified data stream (malicious input).
+           However, it does not provide any protection against intentionally modified data stream (malicious input).
            Use this function in trusted environment only (data to decode comes from a trusted source).
 */
 int LZ4_decompress_fast (const char* source, char* dest, int originalSize);
@@ -170,107 +169,106 @@ int LZ4_decompress_safe_partial (const char* source, char* dest, int compressedS
 
 
 /***********************************************
-   Experimental Streaming Compression Functions
+   Streaming Compression Functions
 ***********************************************/
 
-#define LZ4_STREAMSIZE_U32 ((1 << (LZ4_MEMORY_USAGE-2)) + 8)
-#define LZ4_STREAMSIZE     (LZ4_STREAMSIZE_U32 * sizeof(unsigned int))
+#define LZ4_STREAMSIZE_U64 ((1 << (LZ4_MEMORY_USAGE-3)) + 4)
+#define LZ4_STREAMSIZE     (LZ4_STREAMSIZE_U64 * sizeof(long long))
 /*
  * LZ4_stream_t
  * information structure to track an LZ4 stream.
  * important : init this structure content before first use !
+ * note : only allocated directly the structure if you are statically linking LZ4
+ *        If you are using liblz4 as a DLL, please use below construction methods instead.
  */
-typedef struct { unsigned int table[LZ4_STREAMSIZE_U32]; } LZ4_stream_t;
+typedef struct { long long table[LZ4_STREAMSIZE_U64]; } LZ4_stream_t;
 
 /*
  * LZ4_resetStream
- * Use this function to init a newly allocated LZ4_stream_t structure
- * You can also reset an existing LZ4_stream_t structure
+ * Use this function to init an allocated LZ4_stream_t structure
  */
-void LZ4_resetStream (LZ4_stream_t* LZ4_stream);
+void LZ4_resetStream (LZ4_stream_t* LZ4_streamPtr);
 
 /*
- * If you prefer dynamic allocation methods,
  * LZ4_createStream will allocate and initialize an LZ4_stream_t structure
  * LZ4_freeStream releases its memory.
+ * In the context of a DLL (liblz4), please use these methods rather than the static struct.
+ * They are more future proof, in case of a change of LZ4_stream_t size.
  */
 LZ4_stream_t* LZ4_createStream(void);
-int           LZ4_freeStream (LZ4_stream_t* LZ4_stream);
+int           LZ4_freeStream (LZ4_stream_t* LZ4_streamPtr);
 
 /*
  * LZ4_loadDict
  * Use this function to load a static dictionary into LZ4_stream.
  * Any previous data will be forgotten, only 'dictionary' will remain in memory.
  * Loading a size of 0 is allowed.
- * Return : 1 if OK, 0 if error
+ * Return : dictionary size, in bytes (necessarily <= 64 KB)
  */
-int LZ4_loadDict (LZ4_stream_t* LZ4_stream, const char* dictionary, int dictSize);
+int LZ4_loadDict (LZ4_stream_t* LZ4_streamPtr, const char* dictionary, int dictSize);
 
 /*
  * LZ4_compress_continue
  * Compress data block 'source', using blocks compressed before as dictionary to improve compression ratio
  * Previous data blocks are assumed to still be present at their previous location.
  */
-int LZ4_compress_continue (LZ4_stream_t* LZ4_stream, const char* source, char* dest, int inputSize);
+int LZ4_compress_continue (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize);
 
 /*
  * LZ4_compress_limitedOutput_continue
  * Same as before, but also specify a maximum target compressed size (maxOutputSize)
  * If objective cannot be met, compression exits, and returns a zero.
  */
-int LZ4_compress_limitedOutput_continue (LZ4_stream_t* LZ4_stream, const char* source, char* dest, int inputSize, int maxOutputSize);
+int LZ4_compress_limitedOutput_continue (LZ4_stream_t* LZ4_streamPtr, const char* source, char* dest, int inputSize, int maxOutputSize);
 
 /*
  * LZ4_saveDict
  * If previously compressed data block is not guaranteed to remain available at its memory location
- * save it into a safe place (char* safeBuffer)
+ * save it into a safer place (char* safeBuffer)
  * Note : you don't need to call LZ4_loadDict() afterwards,
  *        dictionary is immediately usable, you can therefore call again LZ4_compress_continue()
- * Return : 1 if OK, 0 if error
+ * Return : dictionary size in bytes, or 0 if error
  * Note : any dictSize > 64 KB will be interpreted as 64KB.
  */
-int LZ4_saveDict (LZ4_stream_t* LZ4_stream, char* safeBuffer, int dictSize);
+int LZ4_saveDict (LZ4_stream_t* LZ4_streamPtr, char* safeBuffer, int dictSize);
 
 
 /************************************************
-  Experimental Streaming Decompression Functions
+   Streaming Decompression Functions
 ************************************************/
 
-#define LZ4_STREAMDECODESIZE_U32 4
-#define LZ4_STREAMDECODESIZE     (LZ4_STREAMDECODESIZE_U32 * sizeof(unsigned int))
+#define LZ4_STREAMDECODESIZE_U64  4
+#define LZ4_STREAMDECODESIZE     (LZ4_STREAMDECODESIZE_U64 * sizeof(unsigned long long))
+typedef struct { unsigned long long table[LZ4_STREAMDECODESIZE_U64]; } LZ4_streamDecode_t;
 /*
  * LZ4_streamDecode_t
  * information structure to track an LZ4 stream.
- * important : init this structure content using LZ4_setStreamDecode or memset() before first use !
+ * init this structure content using LZ4_setStreamDecode or memset() before first use !
+ *
+ * In the context of a DLL (liblz4) please prefer usage of construction methods below.
+ * They are more future proof, in case of a change of LZ4_streamDecode_t size in the future.
+ * LZ4_createStreamDecode will allocate and initialize an LZ4_streamDecode_t structure
+ * LZ4_freeStreamDecode releases its memory.
  */
-typedef struct { unsigned int table[LZ4_STREAMDECODESIZE_U32]; } LZ4_streamDecode_t;
+LZ4_streamDecode_t* LZ4_createStreamDecode(void);
+int                 LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream);
 
 /*
  * LZ4_setStreamDecode
  * Use this function to instruct where to find the dictionary.
- * This function can be used to specify a static dictionary,
- * or to instruct where to find some previously decoded data saved into a different memory space.
- * Setting a size of 0 is allowed (same effect as no dictionary).
+ * Setting a size of 0 is allowed (same effect as reset).
  * Return : 1 if OK, 0 if error
  */
 int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize);
 
 /*
- * If you prefer dynamic allocation methods,
- * LZ4_createStreamDecode will allocate and initialize an LZ4_streamDecode_t structure
- * LZ4_freeStreamDecode releases its memory.
- */
-LZ4_streamDecode_t* LZ4_createStreamDecode(void);
-int                 LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream);
-
-/*
 *_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 its new address using LZ4_setDictDecode()
+    Previously decoded blocks *must* remain available at the memory position where they were decoded (up to 64 KB)
+    If this condition is not possible, save the relevant part of decoded data into a safe buffer,
+    and indicate where is its new address using LZ4_setStreamDecode()
 */
-int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize);
+int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxDecompressedSize);
 int LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int originalSize);
 
 
@@ -279,10 +277,9 @@ Advanced decoding functions :
 *_usingDict() :
     These decoding functions work the same as
     a combination of LZ4_setDictDecode() followed by LZ4_decompress_x_continue()
-    all together into a single function call.
-    It doesn't use nor update an LZ4_streamDecode_t structure.
+    They are stand-alone and don't use nor update an LZ4_streamDecode_t structure.
 */
-int LZ4_decompress_safe_usingDict (const char* source, char* dest, int compressedSize, int maxOutputSize, const char* dictStart, int dictSize);
+int LZ4_decompress_safe_usingDict (const char* source, char* dest, int compressedSize, int maxDecompressedSize, const char* dictStart, int dictSize);
 int LZ4_decompress_fast_usingDict (const char* source, char* dest, int originalSize, const char* dictStart, int dictSize);
 
 
@@ -297,18 +294,10 @@ They are only provided here for compatibility with older user programs.
 - LZ4_uncompress is the same as LZ4_decompress_fast
 - LZ4_uncompress_unknownOutputSize is the same as LZ4_decompress_safe
 These function prototypes are now disabled; uncomment them if you really need them.
-It is highly recommended to stop using these functions and migrated to newer ones */
+It is highly recommended to stop using these functions and migrate to newer ones */
 /* int LZ4_uncompress (const char* source, char* dest, int outputSize); */
 /* int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize); */
 
-/*
- * If you prefer dynamic allocation methods,
- * LZ4_createStreamDecode()
- * provides a pointer (void*) towards an initialized LZ4_streamDecode_t structure.
- * LZ4_free just frees it.
- */
-/* void* LZ4_createStreamDecode(void); */
-/*int   LZ4_free (void* LZ4_stream);    yes, it's the same one as for compression */
 
 /* Obsolete streaming functions; use new streaming interface whenever possible */
 void* LZ4_create (const char* inputBuffer);