From 4314257d8cf9c367cab2333cfdc7b10b0109de90 Mon Sep 17 00:00:00 2001 From: Stefan Schmidt Date: Thu, 7 May 2015 10:52:54 +0200 Subject: [PATCH] lz4: Update our internal copy to release r128 Looking through the git log it is unclear which release we used before as nobody stated it there. :/ We updated after the security issues last year so my best guess is that we have something like r119. To see what changed I now included the NEWS file and also the LICENSE file from upstream. Upstream in now hosted here: https://github.com/Cyan4973/lz4 and http://www.lz4.info I recommend STRONGLY that you check if your distro ships liblz4 as an up to date library package and use the --enable-liblz4 configure option to use the system version. I consider making the system version default for upcoming releases and only carry the internal one as fallback for systems that do not provide it. Fix T2374 --- src/Makefile_Emile.am | 4 +- src/static_libs/lz4/LICENSE | 24 + src/static_libs/lz4/NEWS | 153 ++++++ src/static_libs/lz4/README | 7 - src/static_libs/lz4/README.md | 19 + src/static_libs/lz4/lz4.c | 932 +++++++++++++++++++--------------- src/static_libs/lz4/lz4.h | 208 ++++---- src/static_libs/lz4/lz4hc.c | 888 ++++++++++++++------------------ src/static_libs/lz4/lz4hc.h | 133 ++--- 9 files changed, 1280 insertions(+), 1088 deletions(-) create mode 100644 src/static_libs/lz4/LICENSE create mode 100644 src/static_libs/lz4/NEWS delete mode 100644 src/static_libs/lz4/README create mode 100644 src/static_libs/lz4/README.md diff --git a/src/Makefile_Emile.am b/src/Makefile_Emile.am index 4d6e43f920..fe2515a785 100644 --- a/src/Makefile_Emile.am +++ b/src/Makefile_Emile.am @@ -58,7 +58,9 @@ if ! ENABLE_LIBLZ4 lib_emile_libemile_la_LIBADD += @LIBLZ4_LIBS@ endif -EXTRA_DIST += static_libs/lz4/README \ +EXTRA_DIST += static_libs/lz4/README.md \ +static_libs/lz4/LICENSE \ +static_libs/lz4/NEWS \ static_libs/lz4/lz4.c \ static_libs/lz4/lz4.h \ static_libs/lz4/lz4hc.c \ diff --git a/src/static_libs/lz4/LICENSE b/src/static_libs/lz4/LICENSE new file mode 100644 index 0000000000..b566df3035 --- /dev/null +++ b/src/static_libs/lz4/LICENSE @@ -0,0 +1,24 @@ +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. \ No newline at end of file diff --git a/src/static_libs/lz4/NEWS b/src/static_libs/lz4/NEWS new file mode 100644 index 0000000000..2eeb948ab9 --- /dev/null +++ b/src/static_libs/lz4/NEWS @@ -0,0 +1,153 @@ +r128: +New : lz4cli sparse file support +New : command -m, to compress multiple files in a single command +Fixed : Restored lz4hc compression ratio (was slightly lower since r124) +New : lz4 cli supports long commands +New : lz4frame & lz4cli frame content size support +New : lz4frame supports skippable frames +Changed:Default "make install" directory is /usr/local +New : lz4 cli supports "pass-through" mode +New : datagen can generate sparse files +New : scan-build tests +New : g++ compatibility tests +New : arm cross-compilation test +Fixed : Fuzzer + frametest compatibility with NetBSD (issue #48) +Added : Visual project directory +Updated:Man page & Specification + +r127: +N/A : added a file on SVN + +r126: +New : lz4frame API is now integrated into liblz4 +Fixed : GCC 4.9 bug on highest performance settings, reported by Greg Slazinski +Fixed : bug within LZ4 HC streaming mode, reported by James Boyle +Fixed : older compiler don't like nameless unions, reported by Cheyi Lin +Changed : lz4 is C90 compatible +Changed : added -pedantic option, fixed a few mminor warnings + +r125: +Changed : endian and alignment code +Changed : directory structure : new "lib" directory +Updated : lz4io, now uses lz4frame +Improved: slightly improved decoding speed +Fixed : LZ4_compress_limitedOutput(); Special thanks to Christopher Speller ! +Fixed : some alignment warnings under clang +Fixed : deprecated function LZ4_slideInputBufferHC() + +r124: +New : LZ4 HC streaming mode +Fixed : LZ4F_compressBound() using null preferencesPtr +Updated : xxHash to r38 +Updated library number, to 1.4.0 + +r123: +Added : experimental lz4frame API, thanks to Takayuki Matsuoka and Christopher Jackson for testings +Fix : s390x support, thanks to Nobuhiro Iwamatsu +Fix : test mode (-t) no longer requires confirmation, thanks to Thary Nguyen + +r122: +Fix : AIX & AIX64 support (SamG) +Fix : mips 64-bits support (lew van) +Added : Examples directory, using code examples from Takayuki Matsuoka +Updated : Framing specification, to v1.4.1 +Updated : xxHash, to r36 + +r121: +Added : Makefile : install for kFreeBSD and Hurd (Nobuhiro Iwamatsu) +Fix : Makefile : install for OS-X and BSD, thanks to Takayuki Matsuoka + +r120: +Modified : Streaming API, using strong types +Added : LZ4_versionNumber(), thanks to Takayuki Matsuoka +Fix : OS-X : library install name, thanks to Clemens Lang +Updated : Makefile : synchronize library version number with lz4.h, thanks to Takayuki Matsuoka +Updated : Makefile : stricter compilation flags +Added : pkg-config, thanks to Zbigniew Jędrzejewski-Szmek (issue 135) +Makefile : lz4-test only test native binaries, as suggested by Michał Górny (issue 136) +Updated : xxHash to r35 + +r119: +Fix : Issue 134 : extended malicious address space overflow in 32-bits mode for some specific configurations + +r118: +New : LZ4 Streaming API (Fast version), special thanks to Takayuki Matsuoka +New : datagen : parametrable synthetic data generator for tests +Improved : fuzzer, support more test cases, more parameters, ability to jump to specific test +fix : support ppc64le platform (issue 131) +fix : Issue 52 (malicious address space overflow in 32-bits mode when using large custom format) +fix : Makefile : minor issue 130 : header files permissions + +r117: +Added : man pages for lz4c and lz4cat +Added : automated tests on Travis, thanks to Takayuki Matsuoka ! +fix : block-dependency command line (issue 127) +fix : lz4fullbench (issue 128) + +r116: +hotfix (issue 124 & 125) + +r115: +Added : lz4cat utility, installed on POSX systems (issue 118) +OS-X compatible compilation of dynamic library (issue 115) + +r114: +Makefile : library correctly compiled with -O3 switch (issue 114) +Makefile : library compilation compatible with clang +Makefile : library is versioned and linked (issue 119) +lz4.h : no more static inline prototypes (issue 116) +man : improved header/footer (issue 111) +Makefile : Use system default $(CC) & $(MAKE) variables (issue 112) +xxhash : updated to r34 + +r113: +Large decompression speed improvement for GCC 32-bits. Thanks to Valery Croizier ! +LZ4HC : Compression Level is now a programmable parameter (CLI from 4 to 9) +Separated IO routines from command line (lz4io.c) +Version number into lz4.h (suggested by Francesc Alted) + +r112: +quickfix + +r111 : +Makefile : added capability to install libraries +Modified Directory tree, to better separate libraries from programs. + +r110 : +lz4 & lz4hc : added capability to allocate state & stream state with custom allocator (issue 99) +fuzzer & fullbench : updated to test new functions +man : documented -l command (Legacy format, for Linux kernel compression) (issue 102) +cmake : improved version by Mika Attila, building programs and libraries (issue 100) +xxHash : updated to r33 +Makefile : clean also delete local package .tar.gz + +r109 : +lz4.c : corrected issue 98 (LZ4_compress_limitedOutput()) +Makefile : can specify version number from makefile + +r108 : +lz4.c : corrected compression efficiency issue 97 in 64-bits chained mode (-BD) for streams > 4 GB (thanks Roman Strashkin for reporting) + +r107 : +Makefile : support DESTDIR for staged installs. Thanks Jorge Aparicio. +Makefile : make install installs both lz4 and lz4c (Jorge Aparicio) +Makefile : removed -Wno-implicit-declaration compilation switch +lz4cli.c : include for isatty() (Luca Barbato) +lz4.h : introduced LZ4_MAX_INPUT_SIZE constant (Shay Green) +lz4.h : LZ4_compressBound() : unified macro and inline definitions (Shay Green) +lz4.h : LZ4_decompressSafe_partial() : clarify comments (Shay Green) +lz4.c : LZ4_compress() verify input size condition (Shay Green) +bench.c : corrected a bug in free memory size evaluation +cmake : install into bin/ directory (Richard Yao) +cmake : check for just C compiler (Elan Ruusamae) + +r106 : +Makefile : make dist modify text files in the package to respect Unix EoL convention +lz4cli.c : corrected small display bug in HC mode + +r105 : +Makefile : New install script and man page, contributed by Prasad Pandit +lz4cli.c : Minor modifications, for easier extensibility +COPYING : added license file +LZ4_Streaming_Format.odt : modified file name to remove white space characters +Makefile : .exe suffix now properly added only for Windows target diff --git a/src/static_libs/lz4/README b/src/static_libs/lz4/README deleted file mode 100644 index 718f773793..0000000000 --- a/src/static_libs/lz4/README +++ /dev/null @@ -1,7 +0,0 @@ -This iz the lz4 tree copied in: - http://lz4.googlecode.com/svn/trunk -by: - yann.collet.73@gmail.com -Copyright/licensing info in source files here. - -this was from revision 84. diff --git a/src/static_libs/lz4/README.md b/src/static_libs/lz4/README.md new file mode 100644 index 0000000000..636c8be0bc --- /dev/null +++ b/src/static_libs/lz4/README.md @@ -0,0 +1,19 @@ +LZ4 - Library Files +================================ + +This directory contains many files, but you don't necessarily need them all. + +If you want to integrate LZ4 compression/decompression into your program, you basically need to include "**lz4.c**" and "**lz4.h**" only. + +If you want more compression, at the cost of compression speed (but preserving decompression speed), you will also have to include "**lz4hc.c**" and "**lz4hc.h**". Note that lz4hc needs lz4 to work properly. + +Next level, if you want to produce files or data streams compatible with lz4 utility, you will have to use and include "**lz4frame.c**" and **lz4frame.h**". This library encapsulate lz4-compressed blocks into the official interoperable frame format. In order to work properly, lz4frame needs lz4 and lz4hc, and also "**xxhash.c**" and "**xxhash.h**", which provide the error detection algorithm. + +A more complex "lz4frame_static.h" is also provided, although its usage is not recommended. It contains definitions which are not guaranteed to remain stable within future versions. Use only if you don't plan to update your lz4 version. + +The other files are not source code. There are : + + - LICENSE : contains the BSD license text + - Makefile : script to compile or install lz4 library (static or dynamic) + - liblz4.pc.in : for pkg-config (make install) + diff --git a/src/static_libs/lz4/lz4.c b/src/static_libs/lz4/lz4.c index 482a8ede33..881d1af020 100644 --- a/src/static_libs/lz4/lz4.c +++ b/src/static_libs/lz4/lz4.c @@ -1,6 +1,7 @@ /* 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 @@ -27,82 +28,77 @@ 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 : 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) + * This way, an automatic detection macro can be added to match your case within later 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 +/* + * Automated efficient unaligned memory access detection + * Based on known hardware architectures + * This list will be updated thanks to feedbacks + */ +#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 -# define LZ4_ARCH64 0 +# define LZ4_UNALIGNED_ACCESS 0 #endif /* - * Little Endian or Big Endian ? - * Overwrite the #define below if you know your architecture endianess + * LZ4_FORCE_SW_BITCOUNT + * Define this parameter if your target system or compiler does not support hardware bit count */ -#include /* Apparently required to detect endianess */ -#if defined (__GLIBC__) -# include -# 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 +* Compiler Options **************************************/ #if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */ /* "restrict" is a known keyword */ @@ -112,28 +108,20 @@ #ifdef _MSC_VER /* Visual Studio */ # define FORCE_INLINE static __forceinline -# include /* 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 # 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__) @@ -181,46 +169,132 @@ typedef unsigned long long U64; #endif -#if defined(__GNUC__) && !defined(LZ4_FORCE_UNALIGNED_ACCESS) -# define _PACKED __attribute__ ((packed)) -#else -# define _PACKED + +/************************************** + Reading and writing into memory +**************************************/ +#define STEPSIZE sizeof(size_t) + +static unsigned LZ4_64bits(void) { return sizeof(void*)==8; } + +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 = (const BYTE*)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 = (BYTE*)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); +} -#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) +/* 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 = (BYTE*)dstPtr; + const BYTE* s = (const BYTE*)srcPtr; + BYTE* e = (BYTE*)dstEnd; + do { LZ4_copy8(d,s); d+=8; s+=8; } while (d>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 + } + } +} + +static unsigned LZ4_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* pInLimit) +{ + const BYTE* const pStart = pIn; + + while (likely(pIn compression run slower on incompressible data */ + + +/************************************** +* Local Utils +**************************************/ +int LZ4_versionNumber (void) { return LZ4_VERSION_NUMBER; } +int LZ4_compressBound(int isize) { return LZ4_COMPRESSBOUND(isize); } + + +/************************************** +* Local Structures and types **************************************/ typedef struct { - U32 hashTable[HASH_SIZE_U32]; - U32 currentOffset; - U32 initCheck; + U32 hashTable[HASH_SIZE_U32]; + U32 currentOffset; + U32 initCheck; const BYTE* dictionary; const BYTE* bufferStart; - U32 dictSize; + U32 dictSize; } LZ4_stream_t_internal; typedef enum { notLimited = 0, limitedOutput = 1 } limitedOutput_directive; @@ -266,108 +451,12 @@ 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; */ -#else -# define LZ4_WILDCOPY(d,s,e) { if (likely(e-d <= 8)) LZ4_COPY8(d,s) else do { LZ4_COPY8(d,s) } while (d>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 +* Compression functions ********************************/ -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 const tableType) { if (tableType == byU16) return (((sequence) * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1))); @@ -375,15 +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) +static void LZ4_putPositionOnHash(const BYTE* p, U32 h, void* tableBase, tableType_t const 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; } } } @@ -406,34 +495,14 @@ 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=(int)LZ4_64KLIMIT)) return 0; /* Size too large (not within 64K limit) */ - if (inputSize=LZ4_64Klimit)) return 0; /* Size too large (not within 64K limit) */ + if (inputSize> skipStrength; - //if (step>8) step=8; // required for valid forwardIp ; slows down uncompressible data a bit + 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; @@ -520,13 +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 */ @@ -544,12 +611,13 @@ static int LZ4_compress_generic( else *token = (BYTE)(litLength< 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) { @@ -572,14 +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; } @@ -598,10 +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; @@ -613,9 +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 */ @@ -642,16 +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, (sizeof(void*)==8) ? 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); @@ -662,16 +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, (sizeof(void*)==8) ? 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); @@ -681,38 +749,48 @@ int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, in /***************************************** - Experimental : Streaming functions +* Experimental : Streaming functions *****************************************/ -void* LZ4_createStream() +/* + * LZ4_initStream + * Use this function once, to init a newly allocated LZ4_stream_t structure + * Return : 1 if OK, 0 if error + */ +void LZ4_resetStream (LZ4_stream_t* LZ4_stream) { - void* lz4s = ALLOCATOR(4, LZ4_STREAMSIZE_U32); - MEM_INIT(lz4s, 0, LZ4_STREAMSIZE); + MEM_INIT(LZ4_stream, 0, sizeof(LZ4_stream_t)); +} + +LZ4_stream_t* LZ4_createStream(void) +{ + 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; } -int LZ4_free (void* LZ4_stream) +int LZ4_freeStream (LZ4_stream_t* LZ4_stream) { FREEMEM(LZ4_stream); return (0); } -int LZ4_loadDict (void* LZ4_dict, const char* dictionary, int dictSize) +int LZ4_loadDict (LZ4_stream_t* 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 (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; @@ -727,11 +805,11 @@ int LZ4_loadDict (void* LZ4_dict, const char* dictionary, int dictSize) p+=3; } - return 1; + return dict->dictSize; } -void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, const BYTE* src) +static 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 */ @@ -802,19 +880,18 @@ FORCE_INLINE int LZ4_compress_continue_generic (void* LZ4_stream, const char* so } } - -int LZ4_compress_continue (void* LZ4_stream, const char* source, char* dest, int inputSize) +int LZ4_compress_continue (LZ4_stream_t* 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) +int LZ4_compress_limitedOutput_continue (LZ4_stream_t* 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 +/* 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; @@ -835,7 +912,7 @@ int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* } -int LZ4_saveDict (void* LZ4_dict, char* safeBuffer, int dictSize) +int LZ4_saveDict (LZ4_stream_t* LZ4_dict, char* safeBuffer, int dictSize) { LZ4_stream_t_internal* dict = (LZ4_stream_t_internal*) LZ4_dict; const BYTE* previousDictEnd = dict->dictionary + dict->dictSize; @@ -843,28 +920,28 @@ int LZ4_saveDict (void* 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; } -/**************************** - Decompression functions -****************************/ +/******************************* +* Decompression functions +*******************************/ /* * 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. */ @@ -872,36 +949,32 @@ 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; -//#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 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 checkOffset = (endOnInput) && (dictSize < (int)(64 KB)); + const int safeDecode = (endOnInput==endOnInputSize); + const int checkOffset = ((safeDecode) && (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 ((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); @@ -910,8 +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) { @@ -922,9 +996,8 @@ FORCE_INLINE int LZ4_decompress_generic( length += s; } while (likely((endOnInput)?ipLZ4_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 */ + 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 */ @@ -945,16 +1018,18 @@ FORCE_INLINE int LZ4_decompress_generic( memcpy(op, ip, length); ip += length; op += length; - break; /* Necessarily EOF, due to parsing restrictions */ + 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 @@ -963,36 +1038,38 @@ FORCE_INLINE int LZ4_decompress_generic( 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 */ + 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 cpy = op + copySize; - const BYTE* ref = (BYTE*)dest; - while (op < cpy) *op++ = *ref++; + BYTE* const endOfMatch = op + copySize; + const BYTE* copyFrom = lowPrefix; + while (op < endOfMatch) *op++ = *copyFrom++; } else { - memcpy(op, dest, copySize); + memcpy(op, lowPrefix, copySize); op += copySize; } } @@ -1000,34 +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[(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); + 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-COPYLENGTH-(STEPSIZE-4))) + if (unlikely(cpy>oend-12)) { - if (cpy > oend-LASTLITERALS) goto _output_error; /* Error : last 5 bytes must be literals */ - if (op 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 (opdictionary = dictionary; - lz4sd->dictSize = dictSize; + lz4sd->prefixSize = (size_t) dictSize; + lz4sd->prefixEnd = (BYTE*) dictionary + dictSize; + lz4sd->externalDict = NULL; + lz4sd->extDictSize = 0; return 1; } @@ -1101,43 +1183,61 @@ int LZ4_setDictDecode (void* LZ4_streamDecode, const char* dictionary, int dictS 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 (void* 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 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; } -int LZ4_decompress_fast_continue (void* LZ4_streamDecode, const char* source, char* dest, int originalSize) +int LZ4_decompress_fast_continue (LZ4_streamDecode_t* 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) + 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; @@ -1151,19 +1251,38 @@ 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); } /*************************************************** - Obsolete Functions +* Obsolete Functions ***************************************************/ /* These function names are deprecated and should no longer be used. @@ -1179,7 +1298,7 @@ int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int LZ4_sizeofStreamState() { return LZ4_STREAMSIZE; } -void LZ4_init(LZ4_stream_t_internal* lz4ds, const BYTE* base) +static void LZ4_init(LZ4_stream_t_internal* lz4ds, const BYTE* base) { MEM_INIT(lz4ds, 0, LZ4_STREAMSIZE); lz4ds->bufferStart = base; @@ -1194,18 +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*)LZ4_Data, (char*)ctx->bufferStart, 64 KB); + return (char*)(ctx->bufferStart + dictSize); } /* Obsolete compresson functions using User-allocated state */ @@ -1217,10 +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, (sizeof(void*)==8) ? 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) @@ -1228,20 +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, (sizeof(void*)==8) ? 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/src/static_libs/lz4/lz4.h b/src/static_libs/lz4/lz4.h index 1064fa115d..de43fc0a20 100644 --- a/src/static_libs/lz4/lz4.h +++ b/src/static_libs/lz4/lz4.h @@ -1,7 +1,8 @@ /* LZ4 - Fast LZ compression algorithm Header File - 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 @@ -28,7 +29,7 @@ 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 : https://github.com/Cyan4973/lz4 - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c */ #pragma once @@ -37,17 +38,23 @@ extern "C" { #endif +/* + * lz4.h provides block compression functions, for optimal performance. + * If you need to generate inter-operable compressed data (respecting LZ4 frame specification), + * please use lz4frame.h instead. +*/ /************************************** - Version +* Version **************************************/ -#define LZ4_VERSION_MAJOR 1 /* for major interface/format changes */ -#define LZ4_VERSION_MINOR 2 /* for minor interface/format changes */ +#define LZ4_VERSION_MAJOR 1 /* for breaking interface changes */ +#define LZ4_VERSION_MINOR 6 /* 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); /************************************** - Tuning parameter +* Tuning parameter **************************************/ /* * LZ4_MEMORY_USAGE : @@ -60,52 +67,45 @@ extern "C" { /************************************** - Simple Functions +* Simple Functions **************************************/ -int LZ4_compress (const char* source, char* dest, int inputSize); -int LZ4_decompress_safe (const char* source, char* dest, int compressedSize, int maxOutputSize); +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() - inputSize : Max supported value is LZ4_MAX_INPUT_VALUE + inputSize : Max supported value is LZ4_MAX_INPUT_SIZE return : the number of bytes written in buffer dest or 0 if the compression fails LZ4_decompress_safe() : compressedSize : is obviously the source size - maxOutputSize : is the size of the destination buffer, which must be already allocated. - return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize) + maxDecompressedSize : is the size of the destination buffer, which must be already allocated. + 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. */ /************************************** - Advanced Functions +* Advanced Functions **************************************/ #define LZ4_MAX_INPUT_SIZE 0x7E000000 /* 2 113 929 216 bytes */ #define LZ4_COMPRESSBOUND(isize) ((unsigned int)(isize) > (unsigned int)LZ4_MAX_INPUT_SIZE ? 0 : (isize) + ((isize)/255) + 16) /* 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 @@ -116,28 +116,40 @@ 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); + + +/* +LZ4_compress_withState() : + Same compression functions, but using an externally allocated memory space to store compression state. + Use LZ4_sizeofState() to know how much memory must be allocated, + and then, provide it as 'void* state' to compression functions. +*/ +int LZ4_sizeofState(void); +int LZ4_compress_withState (void* state, const char* source, char* dest, int inputSize); +int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize); /* LZ4_decompress_fast() : originalSize : is the original and therefore uncompressed size return : the number of bytes read from the source buffer (in other words, the compressed size) - If the source stream is malformed, the function will stop decoding and return a negative result. + If the source stream is detected malformed, the function will stop decoding and return a negative result. Destination buffer must be already allocated. Its size must be a minimum of 'originalSize' bytes. - note : This function is a bit faster than LZ4_decompress_safe() - It provides fast decompression and fully respect memory boundaries for properly formed compressed data. - It does not provide full protection against intentionnally modified data stream. - Use this function in a trusted environment (data to decode comes from a trusted source). + 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 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); @@ -145,117 +157,120 @@ int LZ4_decompress_fast (const char* source, char* dest, int originalSize); /* LZ4_decompress_safe_partial() : This function decompress a compressed block of size 'compressedSize' at position 'source' - into output buffer 'dest' of size 'maxOutputSize'. + into destination buffer 'dest' of size 'maxDecompressedSize'. The function tries to stop decompressing operation as soon as 'targetOutputSize' has been reached, reducing decompression time. - return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize) + return : the number of bytes decoded in the destination buffer (necessarily <= maxDecompressedSize) Note : this number can be < 'targetOutputSize' should the compressed block to decode be smaller. Always control how many bytes were decoded. If the source stream is detected malformed, the function will stop decoding and return a negative result. This function never writes outside of output buffer, and never reads outside of input buffer. It is therefore protected against malicious data packets */ -int LZ4_decompress_safe_partial (const char* source, char* dest, int compressedSize, int targetOutputSize, int maxOutputSize); +int LZ4_decompress_safe_partial (const char* source, char* dest, int compressedSize, int targetOutputSize, int maxDecompressedSize); /*********************************************** - 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 : set this structure content to zero before first use ! + * 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; /* - * If you prefer dynamic allocation methods, - * LZ4_createStream - * provides a pointer (void*) towards an initialized LZ4_stream_t structure. - * LZ4_free just frees it. + * LZ4_resetStream + * Use this function to init an allocated LZ4_stream_t structure */ -void* LZ4_createStream(); -int LZ4_free (void* LZ4_stream); +void LZ4_resetStream (LZ4_stream_t* LZ4_streamPtr); +/* + * 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_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 (same effect as init). - * Return : 1 if OK, 0 if error + * Loading a size of 0 is allowed. + * Return : dictionary size, in bytes (necessarily <= 64 KB) */ -int LZ4_loadDict (void* 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. + * dest buffer must be already allocated, and sized to at least LZ4_compressBound(inputSize) */ -int LZ4_compress_continue (void* 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 (void* 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 at its previous memory location - * save it into a safe place (char* safeBuffer) + * If previously compressed data block is not guaranteed to remain available at its memory location + * 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 - * Note : any dictSize > 64 KB will be interpreted as 64KB. + * Return : saved dictionary size in bytes (necessarily <= dictSize), or 0 if error */ -int LZ4_saveDict (void* 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 : set this structure content to zero 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); /* - * If you prefer dynamic allocation methods, - * LZ4_createStreamDecode() - * provides a pointer (void*) towards an initialized LZ4_streamDecode_t structure. - * LZ4_free just frees it. + * LZ4_setStreamDecode + * Use this function to instruct where to find the dictionary. + * Setting a size of 0 is allowed (same effect as reset). + * Return : 1 if OK, 0 if error */ -void* LZ4_createStreamDecode(); -int LZ4_free (void* LZ4_stream); /* yes, it's the same one as for compression */ +int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize); /* *_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() + 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 (void* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize); -int LZ4_decompress_fast_continue (void* LZ4_streamDecode, const char* source, char* dest, int originalSize); - -/* - * LZ4_setDictDecode - * 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). - * Return : 1 if OK, 0 if error - */ -int LZ4_setDictDecode (void* LZ4_streamDecode, const char* dictionary, int dictSize); +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); /* @@ -263,17 +278,15 @@ 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); - /************************************** - Obsolete Functions +* Obsolete Functions **************************************/ /* Obsolete decompression functions @@ -281,14 +294,11 @@ 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 the same as LZ4_decompress_fast - LZ4_uncompress_unknownOutputSize is the same as LZ4_decompress_safe -*/ -int LZ4_uncompress (const char* source, char* dest, int outputSize); -int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize); +These function prototypes are now disabled; uncomment them if you really need them. +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); */ -/* Obsolete functions for externally allocated state; use streaming interface instead */ -int LZ4_sizeofState(void); -int LZ4_compress_withState (void* state, const char* source, char* dest, int inputSize); -int LZ4_compress_limitedOutput_withState (void* state, const char* source, char* dest, int inputSize, int maxOutputSize); /* Obsolete streaming functions; use new streaming interface whenever possible */ void* LZ4_create (const char* inputBuffer); diff --git a/src/static_libs/lz4/lz4hc.c b/src/static_libs/lz4/lz4hc.c index 6086749025..c7a94a0678 100644 --- a/src/static_libs/lz4/lz4hc.c +++ b/src/static_libs/lz4/lz4hc.c @@ -1,34 +1,35 @@ /* - LZ4 HC - High Compression Mode of LZ4 - Copyright (C) 2011-2014, Yann Collet. - BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) +LZ4 HC - High Compression Mode of LZ4 +Copyright (C) 2011-2015, Yann Collet. - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions are - met: +BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) - * 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. +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: - 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. +* 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. - You can contact the author at : - - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html - - LZ4 source repository : http://code.google.com/p/lz4/ +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 : https://github.com/Cyan4973/lz4 + - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c */ @@ -36,213 +37,48 @@ /************************************** Tuning Parameter **************************************/ -#define LZ4HC_DEFAULT_COMPRESSIONLEVEL 8 - - -/************************************** - Memory routines -**************************************/ -#include /* calloc, free */ -#define ALLOCATOR(s) calloc(1,s) -#define FREEMEM free -#include /* memset, memcpy */ -#define MEM_INIT memset - - -/************************************** - 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 /* Apparently required to detect endianess */ -#if defined (__GLIBC__) -# include -# if (__BYTE_ORDER == __BIG_ENDIAN) -# define LZ4_BIG_ENDIAN 1 -# endif -#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN)) -# define LZ4_BIG_ENDIAN 1 -#elif defined(__sparc) || defined(__sparc__) \ - || defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \ - || defined(__hpux) || defined(__hppa) \ - || defined(_MIPSEB) || defined(__s390__) -# define LZ4_BIG_ENDIAN 1 -#else -/* Little Endian assumed. PDP Endian and other very rare endian format are unsupported. */ -#endif - -/* - * Unaligned memory access is automatically enabled for "common" CPU, such as x86. - * For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected - * If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance - */ -#if defined(__ARM_FEATURE_UNALIGNED) -# define LZ4_FORCE_UNALIGNED_ACCESS 1 -#endif - -/* Define this parameter if your target system or compiler does not support hardware bit count */ -#if defined(_MSC_VER) && defined(_WIN32_WCE) /* Visual Studio for Windows CE does not support Hardware bit count */ -# define LZ4_FORCE_SW_BITCOUNT -#endif - - -/************************************** - Compiler Options -**************************************/ -#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */ -/* "restrict" is a known keyword */ -#else -# define restrict /* Disable restrict */ -#endif - -#ifdef _MSC_VER /* Visual Studio */ -# define FORCE_INLINE static __forceinline -# include /* For Visual 2005 */ -# if LZ4_ARCH64 /* 64-bits */ -# pragma intrinsic(_BitScanForward64) /* For Visual 2005 */ -# pragma intrinsic(_BitScanReverse64) /* For Visual 2005 */ -# else /* 32-bits */ -# pragma intrinsic(_BitScanForward) /* For Visual 2005 */ -# pragma intrinsic(_BitScanReverse) /* For Visual 2005 */ -# endif -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -# pragma warning(disable : 4701) /* disable: C4701: potentially uninitialized local variable used */ -#else -# ifdef __GNUC__ -# define FORCE_INLINE static inline __attribute__((always_inline)) -# else -# define FORCE_INLINE static inline -# endif -#endif - -#ifdef _MSC_VER /* Visual Studio */ -# define lz4_bswap16(x) _byteswap_ushort(x) -#else -# define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8))) -#endif +static const int LZ4HC_compressionLevel_default = 9; /************************************** Includes **************************************/ #include "lz4hc.h" -#include "lz4.h" /************************************** - Basic Types + Local Compiler Options **************************************/ -#if defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */ -# include - 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; +#if defined(__GNUC__) +# pragma GCC diagnostic ignored "-Wunused-function" #endif -#if defined(__GNUC__) && !defined(LZ4_FORCE_UNALIGNED_ACCESS) -# define _PACKED __attribute__ ((packed)) -#else -# define _PACKED +#if defined (__clang__) +# pragma clang diagnostic ignored "-Wunused-function" #endif -#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__) -# ifdef __IBMC__ -# pragma pack(1) -# else -# pragma pack(push, 1) -# endif -#endif - -typedef struct _U16_S { U16 v; } _PACKED U16_S; -typedef struct _U32_S { U32 v; } _PACKED U32_S; -typedef struct _U64_S { U64 v; } _PACKED U64_S; - -#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__) -# pragma pack(pop) -#endif - -#define A64(x) (((U64_S *)(x))->v) -#define A32(x) (((U32_S *)(x))->v) -#define A16(x) (((U16_S *)(x))->v) - /************************************** - Constants + Common LZ4 definition **************************************/ -#define MINMATCH 4 +#define LZ4_COMMONDEFS_ONLY +#include "lz4.c" + +/************************************** + Local Constants +**************************************/ #define DICTIONARY_LOGSIZE 16 #define MAXD (1<> ((MINMATCH*8)-HASH_LOG)) -#define HASH_VALUE(p) HASH_FUNCTION(A32(p)) -#define HASH_POINTER(p) (HashTable[HASH_VALUE(p)] + base) #define DELTANEXT(p) chainTable[(size_t)(p) & MAXD_MASK] #define GETNEXT(p) ((p) - (size_t)DELTANEXT(p)) +static U32 LZ4HC_hashPtr(const void* ptr) { return HASH_FUNCTION(LZ4_read32(ptr)); } + + /************************************** - Private functions + HC Compression **************************************/ -#if LZ4_ARCH64 - -FORCE_INLINE int LZ4_NbCommonBytes (register U64 val) -{ -#if defined(LZ4_BIG_ENDIAN) -# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT) - unsigned long r = 0; - _BitScanReverse64( &r, val ); - return (int)(r>>3); -# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT) - return (__builtin_clzll(val) >> 3); -# else - int r; - if (!(val>>32)) { r=4; } else { r=0; val>>=32; } - if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; } - r += (!val); - return r; -# endif -#else -# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT) - unsigned long r = 0; - _BitScanForward64( &r, val ); - return (int)(r>>3); -# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT) - return (__builtin_ctzll(val) >> 3); -# else - static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 }; - return DeBruijnBytePos[((U64)((val & -val) * 0x0218A392CDABBD3F)) >> 58]; -# endif -#endif -} - -#else - -FORCE_INLINE int LZ4_NbCommonBytes (register U32 val) -{ -#if defined(LZ4_BIG_ENDIAN) -# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT) - unsigned long r; - _BitScanReverse( &r, val ); - return (int)(r>>3); -# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT) - return (__builtin_clz(val) >> 3); -# else - int r; - if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; } - r += (!val); - return r; -# endif -#else -# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT) - unsigned long r; - _BitScanForward( &r, val ); - return (int)(r>>3); -# elif defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT) - return (__builtin_ctz(val) >> 3); -# else - static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 }; - return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27]; -# endif -#endif -} - -#endif - - -int LZ4_sizeofStreamStateHC() -{ - return sizeof(LZ4HC_Data_Structure); -} - -FORCE_INLINE void LZ4_initHC (LZ4HC_Data_Structure* hc4, const BYTE* base) +static void LZ4HC_init (LZ4HC_Data_Structure* hc4, const BYTE* start) { MEM_INIT((void*)hc4->hashTable, 0, sizeof(hc4->hashTable)); MEM_INIT(hc4->chainTable, 0xFF, sizeof(hc4->chainTable)); - hc4->nextToUpdate = base + 1; - hc4->base = base; - hc4->inputBuffer = base; - hc4->end = base; -} - -int LZ4_resetStreamStateHC(void* state, const char* inputBuffer) -{ - if ((((size_t)state) & (sizeof(void*)-1)) != 0) return 1; /* Error : pointer is not aligned for pointer (32 or 64 bits) */ - LZ4_initHC((LZ4HC_Data_Structure*)state, (const BYTE*)inputBuffer); - return 0; -} - - -void* LZ4_createHC (const char* inputBuffer) -{ - void* hc4 = ALLOCATOR(sizeof(LZ4HC_Data_Structure)); - LZ4_initHC ((LZ4HC_Data_Structure*)hc4, (const BYTE*)inputBuffer); - return hc4; -} - - -int LZ4_freeHC (void* LZ4HC_Data) -{ - FREEMEM(LZ4HC_Data); - return (0); + hc4->nextToUpdate = 64 KB; + hc4->base = start - 64 KB; + hc4->inputBuffer = start; + hc4->end = start; + hc4->dictBase = start - 64 KB; + hc4->dictLimit = 64 KB; + hc4->lowLimit = 64 KB; } /* Update chains up to ip (excluded) */ FORCE_INLINE void LZ4HC_Insert (LZ4HC_Data_Structure* hc4, const BYTE* ip) { - U16* chainTable = hc4->chainTable; - HTYPE* HashTable = hc4->hashTable; - INITBASE(base,hc4->base); + U16* chainTable = hc4->chainTable; + U32* HashTable = hc4->hashTable; + const BYTE* const base = hc4->base; + const U32 target = (U32)(ip - base); + U32 idx = hc4->nextToUpdate; - while(hc4->nextToUpdate < ip) + while(idx < target) { - const BYTE* const p = hc4->nextToUpdate; - size_t delta = (p) - HASH_POINTER(p); + U32 h = LZ4HC_hashPtr(base+idx); + size_t delta = idx - HashTable[h]; if (delta>MAX_DISTANCE) delta = MAX_DISTANCE; - DELTANEXT(p) = (U16)delta; - HashTable[HASH_VALUE(p)] = (HTYPE)((p) - base); - hc4->nextToUpdate++; + chainTable[idx & 0xFFFF] = (U16)delta; + HashTable[h] = idx; + idx++; } + + hc4->nextToUpdate = target; } -char* LZ4_slideInputBufferHC(void* LZ4HC_Data) -{ - LZ4HC_Data_Structure* hc4 = (LZ4HC_Data_Structure*)LZ4HC_Data; - U32 distance = (U32)(hc4->end - hc4->inputBuffer) - 64 KB; - distance = (distance >> 16) << 16; /* Must be a multiple of 64 KB */ - LZ4HC_Insert(hc4, hc4->end - MINMATCH); - memcpy((void*)(hc4->end - 64 KB - distance), (const void*)(hc4->end - 64 KB), 64 KB); - hc4->nextToUpdate -= distance; - hc4->base -= distance; - if ((U32)(hc4->inputBuffer - hc4->base) > 1 GB + 64 KB) /* Avoid overflow */ - { - int i; - hc4->base += 1 GB; - for (i=0; ihashTable[i] -= 1 GB; - } - hc4->end -= distance; - return (char*)(hc4->end); -} - - -FORCE_INLINE size_t LZ4HC_CommonLength (const BYTE* p1, const BYTE* p2, const BYTE* const matchlimit) -{ - const BYTE* p1t = p1; - - while (p1tchainTable; - HTYPE* const HashTable = hc4->hashTable; - const BYTE* ref; - INITBASE(base,hc4->base); + U32* const HashTable = hc4->hashTable; + const BYTE* const base = hc4->base; + const BYTE* const dictBase = hc4->dictBase; + const U32 dictLimit = hc4->dictLimit; + const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1); + U32 matchIndex; + const BYTE* match; int nbAttempts=maxNbAttempts; - size_t repl=0, ml=0; - U16 delta=0; /* useless assignment, to remove an uninitialization warning */ + size_t ml=0; /* HC4 match finder */ LZ4HC_Insert(hc4, ip); - ref = HASH_POINTER(ip); + matchIndex = HashTable[LZ4HC_hashPtr(ip)]; -#define REPEAT_OPTIMIZATION -#ifdef REPEAT_OPTIMIZATION - /* Detect repetitive sequences of length <= 4 */ - if ((U32)(ip-ref) <= 4) /* potential repetition */ - { - if (A32(ref) == A32(ip)) /* confirmed */ - { - delta = (U16)(ip-ref); - repl = ml = LZ4HC_CommonLength(ip+MINMATCH, ref+MINMATCH, matchlimit) + MINMATCH; - *matchpos = ref; - } - ref = GETNEXT(ref); - } -#endif - - while (((U32)(ip-ref) <= MAX_DISTANCE) && (nbAttempts)) + while ((matchIndex>=lowLimit) && (nbAttempts)) { nbAttempts--; - if (*(ref+ml) == *(ip+ml)) - if (A32(ref) == A32(ip)) + if (matchIndex >= dictLimit) { - size_t mlt = LZ4HC_CommonLength(ip+MINMATCH, ref+MINMATCH, matchlimit) + MINMATCH; - if (mlt > ml) { ml = mlt; *matchpos = ref; } + match = base + matchIndex; + if (*(match+ml) == *(ip+ml) + && (LZ4_read32(match) == LZ4_read32(ip))) + { + size_t mlt = LZ4_count(ip+MINMATCH, match+MINMATCH, iLimit) + MINMATCH; + if (mlt > ml) { ml = mlt; *matchpos = match; } + } } - ref = GETNEXT(ref); - } - -#ifdef REPEAT_OPTIMIZATION - /* Complete table */ - if (repl) - { - const BYTE* ptr = ip; - const BYTE* end; - - end = ip + repl - (MINMATCH-1); - while(ptr < end-delta) + else { - DELTANEXT(ptr) = delta; /* Pre-Load */ - ptr++; + match = dictBase + matchIndex; + if (LZ4_read32(match) == LZ4_read32(ip)) + { + size_t mlt; + const BYTE* vLimit = ip + (dictLimit - matchIndex); + if (vLimit > iLimit) vLimit = iLimit; + mlt = LZ4_count(ip+MINMATCH, match+MINMATCH, vLimit) + MINMATCH; + if ((ip+mlt == vLimit) && (vLimit < iLimit)) + mlt += LZ4_count(ip+mlt, base+dictLimit, iLimit); + if (mlt > ml) { ml = mlt; *matchpos = base + matchIndex; } /* virtual matchpos */ + } } - do - { - DELTANEXT(ptr) = delta; - HashTable[HASH_VALUE(ptr)] = (HTYPE)((ptr) - base); /* Head of chain */ - ptr++; - } while(ptr < end); - hc4->nextToUpdate = end; + matchIndex -= chainTable[matchIndex & 0xFFFF]; } -#endif return (int)ml; } -FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch (LZ4HC_Data_Structure* hc4, const BYTE* ip, const BYTE* startLimit, const BYTE* matchlimit, int longest, const BYTE** matchpos, const BYTE** startpos, const int maxNbAttempts) +FORCE_INLINE int LZ4HC_InsertAndGetWiderMatch ( + LZ4HC_Data_Structure* hc4, + const BYTE* const ip, + const BYTE* const iLowLimit, + const BYTE* const iHighLimit, + int longest, + const BYTE** matchpos, + const BYTE** startpos, + const int maxNbAttempts) { - U16* const chainTable = hc4->chainTable; - HTYPE* const HashTable = hc4->hashTable; - INITBASE(base,hc4->base); - const BYTE* ref; + U16* const chainTable = hc4->chainTable; + U32* const HashTable = hc4->hashTable; + const BYTE* const base = hc4->base; + const U32 dictLimit = hc4->dictLimit; + const BYTE* const lowPrefixPtr = base + dictLimit; + const U32 lowLimit = (hc4->lowLimit + 64 KB > (U32)(ip-base)) ? hc4->lowLimit : (U32)(ip - base) - (64 KB - 1); + const BYTE* const dictBase = hc4->dictBase; + U32 matchIndex; int nbAttempts = maxNbAttempts; - int delta = (int)(ip-startLimit); + int delta = (int)(ip-iLowLimit); + /* First Match */ LZ4HC_Insert(hc4, ip); - ref = HASH_POINTER(ip); + matchIndex = HashTable[LZ4HC_hashPtr(ip)]; - while (((U32)(ip-ref) <= MAX_DISTANCE) && (nbAttempts)) + while ((matchIndex>=lowLimit) && (nbAttempts)) { nbAttempts--; - if (*(startLimit + longest) == *(ref - delta + longest)) - if (A32(ref) == A32(ip)) + if (matchIndex >= dictLimit) { -#if 1 - const BYTE* reft = ref+MINMATCH; - const BYTE* ipt = ip+MINMATCH; - const BYTE* startt = ip; + const BYTE* matchPtr = base + matchIndex; + if (*(iLowLimit + longest) == *(matchPtr - delta + longest)) + if (LZ4_read32(matchPtr) == LZ4_read32(ip)) + { + int mlt = MINMATCH + LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, iHighLimit); + int back = 0; - while (iptiLowLimit) + && (matchPtr+back > lowPrefixPtr) + && (ip[back-1] == matchPtr[back-1])) + back--; + + mlt -= back; + + if (mlt > longest) + { + longest = (int)mlt; + *matchpos = matchPtr+back; + *startpos = ip+back; + } + } + } + else + { + const BYTE* matchPtr = dictBase + matchIndex; + if (LZ4_read32(matchPtr) == LZ4_read32(ip)) { - size_t diff = AARCH(reft) ^ AARCH(ipt); - if (!diff) { ipt+=STEPSIZE; reft+=STEPSIZE; continue; } - ipt += LZ4_NbCommonBytes(diff); - goto _endCount; - } - if (LZ4_ARCH64) if ((ipt<(matchlimit-3)) && (A32(reft) == A32(ipt))) { ipt+=4; reft+=4; } - if ((ipt<(matchlimit-1)) && (A16(reft) == A16(ipt))) { ipt+=2; reft+=2; } - if ((iptstartLimit) && (reft > hc4->inputBuffer) && (startt[-1] == reft[-1])) {startt--; reft--;} - - if ((ipt-startt) > longest) - { - longest = (int)(ipt-startt); - *matchpos = reft; - *startpos = startt; + size_t mlt; + int back=0; + const BYTE* vLimit = ip + (dictLimit - matchIndex); + if (vLimit > iHighLimit) vLimit = iHighLimit; + mlt = LZ4_count(ip+MINMATCH, matchPtr+MINMATCH, vLimit) + MINMATCH; + if ((ip+mlt == vLimit) && (vLimit < iHighLimit)) + mlt += LZ4_count(ip+mlt, base+dictLimit, iHighLimit); + while ((ip+back > iLowLimit) && (matchIndex+back > lowLimit) && (ip[back-1] == matchPtr[back-1])) back--; + mlt -= back; + if ((int)mlt > longest) { longest = (int)mlt; *matchpos = base + matchIndex + back; *startpos = ip+back; } } } - ref = GETNEXT(ref); + matchIndex -= chainTable[matchIndex & 0xFFFF]; } return longest; @@ -564,34 +283,44 @@ _endCount: typedef enum { noLimit = 0, limitedOutput = 1 } limitedOutput_directive; +#define LZ4HC_DEBUG 0 +#if LZ4HC_DEBUG +static unsigned debug = 0; +#endif + FORCE_INLINE int LZ4HC_encodeSequence ( - const BYTE** ip, - BYTE** op, - const BYTE** anchor, - int matchLength, - const BYTE* ref, - limitedOutput_directive limitedOutputBuffer, - BYTE* oend) + const BYTE** ip, + BYTE** op, + const BYTE** anchor, + int matchLength, + const BYTE* const match, + limitedOutput_directive limitedOutputBuffer, + BYTE* oend) { int length; BYTE* token; +#if LZ4HC_DEBUG + if (debug) printf("literal : %u -- match : %u -- offset : %u\n", (U32)(*ip - *anchor), (U32)matchLength, (U32)(*ip-match)); +#endif + /* Encode Literal length */ length = (int)(*ip - *anchor); token = (*op)++; - if ((limitedOutputBuffer) && ((*op + length + (2 + 1 + LASTLITERALS) + (length>>8)) > oend)) return 1; /* Check output limit */ + if ((limitedOutputBuffer) && ((*op + (length>>8) + length + (2 + 1 + LASTLITERALS)) > oend)) return 1; /* Check output limit */ if (length>=(int)RUN_MASK) { int len; *token=(RUN_MASK< 254 ; len-=255) *(*op)++ = 255; *(*op)++ = (BYTE)len; } else *token = (BYTE)(length<>8) > oend)) return 1; /* Check output limit */ + if ((limitedOutputBuffer) && (*op + (length>>8) + (1 + LASTLITERALS) > oend)) return 1; /* Check output limit */ if (length>=(int)ML_MASK) { *token+=ML_MASK; length-=ML_MASK; for(; length > 509 ; length-=510) { *(*op)++ = 255; *(*op)++ = 255; } if (length > 254) { length-=255; *(*op)++ = 255; } *(*op)++ = (BYTE)length; } else *token += (BYTE)(length); @@ -603,16 +332,15 @@ FORCE_INLINE int LZ4HC_encodeSequence ( } -#define MAX_COMPRESSION_LEVEL 16 static int LZ4HC_compress_generic ( - void* ctxvoid, - const char* source, - char* dest, - int inputSize, - int maxOutputSize, - int compressionLevel, - limitedOutput_directive limit - ) + void* ctxvoid, + const char* source, + char* dest, + int inputSize, + int maxOutputSize, + int compressionLevel, + limitedOutput_directive limit + ) { LZ4HC_Data_Structure* ctx = (LZ4HC_Data_Structure*) ctxvoid; const BYTE* ip = (const BYTE*) source; @@ -624,7 +352,7 @@ static int LZ4HC_compress_generic ( BYTE* op = (BYTE*) dest; BYTE* const oend = op + maxOutputSize; - const int maxNbAttempts = compressionLevel > MAX_COMPRESSION_LEVEL ? 1 << MAX_COMPRESSION_LEVEL : compressionLevel ? 1<<(compressionLevel-1) : 1<end) return 0; + /* init */ + if (compressionLevel > g_maxCompressionLevel) compressionLevel = g_maxCompressionLevel; + if (compressionLevel < 1) compressionLevel = LZ4HC_compressionLevel_default; + maxNbAttempts = 1 << (compressionLevel-1); ctx->end += inputSize; ip++; @@ -684,10 +414,10 @@ _Search2: _Search3: /* - * Currently we have : - * ml2 > ml1, and - * ip1+3 <= ip2 (usually < ip1+ml1) - */ + * Currently we have : + * ml2 > ml1, and + * ip1+3 <= ip2 (usually < ip1+ml1) + */ if ((start2 - ip) < OPTIMAL_ML) { int correction; @@ -755,9 +485,9 @@ _Search3: } /* - * OK, now we have 3 ascending matches; let's write at least the first one - * ip & ref are known; Now for ml - */ + * OK, now we have 3 ascending matches; let's write at least the first one + * ip & ref are known; Now for ml + */ if (start2 < ip+ml) { if ((start2 - ip) < (int)ML_MASK) @@ -789,7 +519,6 @@ _Search3: ml2 = ml3; goto _Search3; - } /* Encode Last Literals */ @@ -809,28 +538,18 @@ _Search3: int LZ4_compressHC2(const char* source, char* dest, int inputSize, int compressionLevel) { - void* ctx = LZ4_createHC(source); - int result; - if (ctx==NULL) return 0; - - result = LZ4HC_compress_generic (ctx, source, dest, inputSize, 0, compressionLevel, noLimit); - - LZ4_freeHC(ctx); - return result; + LZ4HC_Data_Structure ctx; + LZ4HC_init(&ctx, (const BYTE*)source); + return LZ4HC_compress_generic (&ctx, source, dest, inputSize, 0, compressionLevel, noLimit); } int LZ4_compressHC(const char* source, char* dest, int inputSize) { return LZ4_compressHC2(source, dest, inputSize, 0); } int LZ4_compressHC2_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel) { - void* ctx = LZ4_createHC(source); - int result; - if (ctx==NULL) return 0; - - result = LZ4HC_compress_generic (ctx, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput); - - LZ4_freeHC(ctx); - return result; + LZ4HC_Data_Structure ctx; + LZ4HC_init(&ctx, (const BYTE*)source); + return LZ4HC_compress_generic (&ctx, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput); } int LZ4_compressHC_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize) @@ -840,15 +559,15 @@ int LZ4_compressHC_limitedOutput(const char* source, char* dest, int inputSize, /***************************** - Using external allocation -*****************************/ -int LZ4_sizeofStateHC() { return sizeof(LZ4HC_Data_Structure); } + * Using external allocation + * ***************************/ +int LZ4_sizeofStateHC(void) { return sizeof(LZ4HC_Data_Structure); } int LZ4_compressHC2_withStateHC (void* state, const char* source, char* dest, int inputSize, int compressionLevel) { if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */ - LZ4_initHC ((LZ4HC_Data_Structure*)state, (const BYTE*)source); + LZ4HC_init ((LZ4HC_Data_Structure*)state, (const BYTE*)source); return LZ4HC_compress_generic (state, source, dest, inputSize, 0, compressionLevel, noLimit); } @@ -859,7 +578,7 @@ int LZ4_compressHC_withStateHC (void* state, const char* source, char* dest, int int LZ4_compressHC2_limitedOutput_withStateHC (void* state, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel) { if (((size_t)(state)&(sizeof(void*)-1)) != 0) return 0; /* Error : state is not aligned for pointers (32 or 64 bits) */ - LZ4_initHC ((LZ4HC_Data_Structure*)state, (const BYTE*)source); + LZ4HC_init ((LZ4HC_Data_Structure*)state, (const BYTE*)source); return LZ4HC_compress_generic (state, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput); } @@ -867,26 +586,171 @@ int LZ4_compressHC_limitedOutput_withStateHC (void* state, const char* source, c { return LZ4_compressHC2_limitedOutput_withStateHC (state, source, dest, inputSize, maxOutputSize, 0); } -/**************************** - Stream functions -****************************/ +/************************************** + * Streaming Functions + * ************************************/ +/* allocation */ +LZ4_streamHC_t* LZ4_createStreamHC(void) { return (LZ4_streamHC_t*)malloc(sizeof(LZ4_streamHC_t)); } +int LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr) { free(LZ4_streamHCPtr); return 0; } + + +/* initialization */ +void LZ4_resetStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel) +{ + LZ4_STATIC_ASSERT(sizeof(LZ4HC_Data_Structure) <= sizeof(LZ4_streamHC_t)); /* if compilation fails here, LZ4_STREAMHCSIZE must be increased */ + ((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->base = NULL; + ((LZ4HC_Data_Structure*)LZ4_streamHCPtr)->compressionLevel = (unsigned)compressionLevel; +} + +int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, const char* dictionary, int dictSize) +{ + LZ4HC_Data_Structure* ctxPtr = (LZ4HC_Data_Structure*) LZ4_streamHCPtr; + if (dictSize > 64 KB) + { + dictionary += dictSize - 64 KB; + dictSize = 64 KB; + } + LZ4HC_init (ctxPtr, (const BYTE*)dictionary); + if (dictSize >= 4) LZ4HC_Insert (ctxPtr, (const BYTE*)dictionary +(dictSize-3)); + ctxPtr->end = (const BYTE*)dictionary + dictSize; + return dictSize; +} + + +/* compression */ + +static void LZ4HC_setExternalDict(LZ4HC_Data_Structure* ctxPtr, const BYTE* newBlock) +{ + if (ctxPtr->end >= ctxPtr->base + 4) + LZ4HC_Insert (ctxPtr, ctxPtr->end-3); /* Referencing remaining dictionary content */ + /* Only one memory segment for extDict, so any previous extDict is lost at this stage */ + ctxPtr->lowLimit = ctxPtr->dictLimit; + ctxPtr->dictLimit = (U32)(ctxPtr->end - ctxPtr->base); + ctxPtr->dictBase = ctxPtr->base; + ctxPtr->base = newBlock - ctxPtr->dictLimit; + ctxPtr->end = newBlock; + ctxPtr->nextToUpdate = ctxPtr->dictLimit; /* match referencing will resume from there */ +} + +static int LZ4_compressHC_continue_generic (LZ4HC_Data_Structure* ctxPtr, + const char* source, char* dest, + int inputSize, int maxOutputSize, limitedOutput_directive limit) +{ + /* auto-init if forgotten */ + if (ctxPtr->base == NULL) + LZ4HC_init (ctxPtr, (const BYTE*) source); + + /* Check overflow */ + if ((size_t)(ctxPtr->end - ctxPtr->base) > 2 GB) + { + size_t dictSize = (size_t)(ctxPtr->end - ctxPtr->base) - ctxPtr->dictLimit; + if (dictSize > 64 KB) dictSize = 64 KB; + + LZ4_loadDictHC((LZ4_streamHC_t*)ctxPtr, (const char*)(ctxPtr->end) - dictSize, (int)dictSize); + } + + /* Check if blocks follow each other */ + if ((const BYTE*)source != ctxPtr->end) LZ4HC_setExternalDict(ctxPtr, (const BYTE*)source); + + /* Check overlapping input/dictionary space */ + { + const BYTE* sourceEnd = (const BYTE*) source + inputSize; + const BYTE* dictBegin = ctxPtr->dictBase + ctxPtr->lowLimit; + const BYTE* dictEnd = ctxPtr->dictBase + ctxPtr->dictLimit; + if ((sourceEnd > dictBegin) && ((BYTE*)source < dictEnd)) + { + if (sourceEnd > dictEnd) sourceEnd = dictEnd; + ctxPtr->lowLimit = (U32)(sourceEnd - ctxPtr->dictBase); + if (ctxPtr->dictLimit - ctxPtr->lowLimit < 4) ctxPtr->lowLimit = ctxPtr->dictLimit; + } + } + + return LZ4HC_compress_generic (ctxPtr, source, dest, inputSize, maxOutputSize, ctxPtr->compressionLevel, limit); +} + +int LZ4_compressHC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize) +{ + return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, 0, noLimit); +} + +int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize, int maxOutputSize) +{ + return LZ4_compressHC_continue_generic ((LZ4HC_Data_Structure*)LZ4_streamHCPtr, source, dest, inputSize, maxOutputSize, limitedOutput); +} + + +/* dictionary saving */ + +int LZ4_saveDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, char* safeBuffer, int dictSize) +{ + LZ4HC_Data_Structure* streamPtr = (LZ4HC_Data_Structure*)LZ4_streamHCPtr; + int prefixSize = (int)(streamPtr->end - (streamPtr->base + streamPtr->dictLimit)); + if (dictSize > 64 KB) dictSize = 64 KB; + if (dictSize < 4) dictSize = 0; + if (dictSize > prefixSize) dictSize = prefixSize; + memcpy(safeBuffer, streamPtr->end - dictSize, dictSize); + { + U32 endIndex = (U32)(streamPtr->end - streamPtr->base); + streamPtr->end = (const BYTE*)safeBuffer + dictSize; + streamPtr->base = streamPtr->end - endIndex; + streamPtr->dictLimit = endIndex - dictSize; + streamPtr->lowLimit = endIndex - dictSize; + if (streamPtr->nextToUpdate < streamPtr->dictLimit) streamPtr->nextToUpdate = streamPtr->dictLimit; + } + return dictSize; +} + + +/*********************************** + * Deprecated Functions + ***********************************/ +int LZ4_sizeofStreamStateHC(void) { return LZ4_STREAMHCSIZE; } + +int LZ4_resetStreamStateHC(void* state, const char* inputBuffer) +{ + if ((((size_t)state) & (sizeof(void*)-1)) != 0) return 1; /* Error : pointer is not aligned for pointer (32 or 64 bits) */ + LZ4HC_init((LZ4HC_Data_Structure*)state, (const BYTE*)inputBuffer); + return 0; +} + +void* LZ4_createHC (const char* inputBuffer) +{ + void* hc4 = ALLOCATOR(1, sizeof(LZ4HC_Data_Structure)); + LZ4HC_init ((LZ4HC_Data_Structure*)hc4, (const BYTE*)inputBuffer); + return hc4; +} + +int LZ4_freeHC (void* LZ4HC_Data) +{ + FREEMEM(LZ4HC_Data); + return (0); +} + +/* int LZ4_compressHC_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize) { - return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, 0, noLimit); +return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, 0, noLimit); } +int LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize) +{ +return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, 0, limitedOutput); +} +*/ int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel) { return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, 0, compressionLevel, noLimit); } -int LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize) -{ - return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, 0, limitedOutput); -} - int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel) { return LZ4HC_compress_generic (LZ4HC_Data, source, dest, inputSize, maxOutputSize, compressionLevel, limitedOutput); } + +char* LZ4_slideInputBufferHC(void* LZ4HC_Data) +{ + LZ4HC_Data_Structure* hc4 = (LZ4HC_Data_Structure*)LZ4HC_Data; + int dictSize = LZ4_saveDictHC((LZ4_streamHC_t*)LZ4HC_Data, (char*)(hc4->inputBuffer), 64 KB); + return (char*)(hc4->inputBuffer + dictSize); +} diff --git a/src/static_libs/lz4/lz4hc.h b/src/static_libs/lz4/lz4hc.h index deb2394062..4a05845423 100644 --- a/src/static_libs/lz4/lz4hc.h +++ b/src/static_libs/lz4/lz4hc.h @@ -1,7 +1,7 @@ /* LZ4 HC - High Compression Mode of LZ4 Header File - 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 @@ -28,8 +28,8 @@ 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/ + - LZ4 source repository : https://github.com/Cyan4973/lz4 + - LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c */ #pragma once @@ -74,12 +74,12 @@ int LZ4_compressHC2_limitedOutput (const char* source, char* dest, int inputSize */ /* Note : -Decompression functions are provided within LZ4 source code (see "lz4.h") (BSD license) + Decompression functions are provided within LZ4 source code (see "lz4.h") (BSD license) */ /************************************** - Using an external allocation +* Using an external allocation **************************************/ int LZ4_sizeofStateHC(void); int LZ4_compressHC_withStateHC (void* state, const char* source, char* dest, int inputSize); @@ -95,77 +95,84 @@ int LZ4_sizeofStateHC(); Note that tables must be aligned for pointer (32 or 64 bits), otherwise compression will fail (return code 0). -The allocated memory can be provided to the compressions functions using 'void* state' parameter. +The allocated memory can be provided to the compression functions using 'void* state' parameter. LZ4_compress_withStateHC() and LZ4_compress_limitedOutput_withStateHC() are equivalent to previously described functions. -They just use the externally allocated memory area instead of allocating their own (on stack, or on heap). +They just use the externally allocated memory for state instead of allocating their own (on stack, or on heap). */ + +/***************************** +* Includes +*****************************/ +#include /* size_t */ + + /************************************** - Streaming Functions +* Experimental Streaming Functions **************************************/ -/* Note : these streaming functions still follows the older model */ +#define LZ4_STREAMHCSIZE 262192 +#define LZ4_STREAMHCSIZE_SIZET (LZ4_STREAMHCSIZE / sizeof(size_t)) +typedef struct { size_t table[LZ4_STREAMHCSIZE_SIZET]; } LZ4_streamHC_t; +/* +LZ4_streamHC_t +This structure allows static allocation of LZ4 HC streaming state. +State must then be initialized using LZ4_resetStreamHC() before first use. + +Static allocation should only be used with statically linked library. +If you want to use LZ4 as a DLL, please use construction functions below, which are more future-proof. +*/ + + +LZ4_streamHC_t* LZ4_createStreamHC(void); +int LZ4_freeStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr); +/* +These functions create and release memory for LZ4 HC streaming state. +Newly created states are already initialized. +Existing state space can be re-used anytime using LZ4_resetStreamHC(). +If you use LZ4 as a DLL, please use these functions instead of direct struct allocation, +to avoid size mismatch between different versions. +*/ + +void LZ4_resetStreamHC (LZ4_streamHC_t* LZ4_streamHCPtr, int compressionLevel); +int LZ4_loadDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, const char* dictionary, int dictSize); + +int LZ4_compressHC_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize); +int LZ4_compressHC_limitedOutput_continue (LZ4_streamHC_t* LZ4_streamHCPtr, const char* source, char* dest, int inputSize, int maxOutputSize); + +int LZ4_saveDictHC (LZ4_streamHC_t* LZ4_streamHCPtr, char* safeBuffer, int maxDictSize); + +/* +These functions compress data in successive blocks of any size, using previous blocks as dictionary. +One key assumption is that each previous block will remain read-accessible while compressing next block. + +Before starting compression, state must be properly initialized, using LZ4_resetStreamHC(). +A first "fictional block" can then be designated as initial dictionary, using LZ4_loadDictHC() (Optional). + +Then, use LZ4_compressHC_continue() or LZ4_compressHC_limitedOutput_continue() to compress each successive block. +They work like usual LZ4_compressHC() or LZ4_compressHC_limitedOutput(), but use previous memory blocks to improve compression. +Previous memory blocks (including initial dictionary when present) must remain accessible and unmodified during compression. + +If, for any reason, previous data block can't be preserved in memory during next compression block, +you must save it to a safer memory space, +using LZ4_saveDictHC(). +*/ + + + +/************************************** + * Deprecated Streaming Functions + * ************************************/ +/* Note : these streaming functions follows the older model, and should no longer be used */ void* LZ4_createHC (const char* inputBuffer); -int LZ4_compressHC_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize); -int LZ4_compressHC_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize); char* LZ4_slideInputBufferHC (void* LZ4HC_Data); int LZ4_freeHC (void* LZ4HC_Data); int LZ4_compressHC2_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int compressionLevel); int LZ4_compressHC2_limitedOutput_continue (void* LZ4HC_Data, const char* source, char* dest, int inputSize, int maxOutputSize, int compressionLevel); -/* -These functions allow the compression of dependent blocks, where each block benefits from prior 64 KB within preceding blocks. -In order to achieve this, it is necessary to start creating the LZ4HC Data Structure, thanks to the function : - -void* LZ4_createHC (const char* inputBuffer); -The result of the function is the (void*) pointer on the LZ4HC Data Structure. -This pointer will be needed in all other functions. -If the pointer returned is NULL, then the allocation has failed, and compression must be aborted. -The only parameter 'const char* inputBuffer' must, obviously, point at the beginning of input buffer. -The input buffer must be already allocated, and size at least 192KB. -'inputBuffer' will also be the 'const char* source' of the first block. - -All blocks are expected to lay next to each other within the input buffer, starting from 'inputBuffer'. -To compress each block, use either LZ4_compressHC_continue() or LZ4_compressHC_limitedOutput_continue(). -Their behavior are identical to LZ4_compressHC() or LZ4_compressHC_limitedOutput(), -but require the LZ4HC Data Structure as their first argument, and check that each block starts right after the previous one. -If next block does not begin immediately after the previous one, the compression will fail (return 0). - -When it's no longer possible to lay the next block after the previous one (not enough space left into input buffer), a call to : -char* LZ4_slideInputBufferHC(void* LZ4HC_Data); -must be performed. It will typically copy the latest 64KB of input at the beginning of input buffer. -Note that, for this function to work properly, minimum size of an input buffer must be 192KB. -==> The memory position where the next input data block must start is provided as the result of the function. - -Compression can then resume, using LZ4_compressHC_continue() or LZ4_compressHC_limitedOutput_continue(), as usual. - -When compression is completed, a call to LZ4_freeHC() will release the memory used by the LZ4HC Data Structure. -*/ - -int LZ4_sizeofStreamStateHC(void); -int LZ4_resetStreamStateHC(void* state, const char* inputBuffer); - -/* -These functions achieve the same result as : -void* LZ4_createHC (const char* inputBuffer); - -They are provided here to allow the user program to allocate memory using its own routines. - -To know how much space must be allocated, use LZ4_sizeofStreamStateHC(); -Note also that space must be aligned for pointers (32 or 64 bits). - -Once space is allocated, you must initialize it using : LZ4_resetStreamStateHC(void* state, const char* inputBuffer); -void* state is a pointer to the space allocated. -It must be aligned for pointers (32 or 64 bits), and be large enough. -The parameter 'const char* inputBuffer' must, obviously, point at the beginning of input buffer. -The input buffer must be already allocated, and size at least 192KB. -'inputBuffer' will also be the 'const char* source' of the first block. - -The same space can be re-used multiple times, just by initializing it each time with LZ4_resetStreamState(). -return value of LZ4_resetStreamStateHC() must be 0 is OK. -Any other value means there was an error (typically, state is not aligned for pointers (32 or 64 bits)). -*/ +int LZ4_sizeofStreamStateHC(void); +int LZ4_resetStreamStateHC(void* state, const char* inputBuffer); #if defined (__cplusplus)