359 lines · c
1/* ******************************************************************2 * huff0 huffman codec,3 * part of Finite State Entropy library4 * Copyright (c) Yann Collet, Facebook, Inc.5 *6 * You can contact the author at :7 * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy8 *9 * This source code is licensed under both the BSD-style license (found in the10 * LICENSE file in the root directory of this source tree) and the GPLv2 (found11 * in the COPYING file in the root directory of this source tree).12 * You may select, at your option, one of the above-listed licenses.13****************************************************************** */14 15 16#ifndef HUF_H_29873423417#define HUF_H_29873423418 19/* *** Dependencies *** */20#include "zstd_deps.h" /* size_t */21 22 23/* *** library symbols visibility *** */24/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,25 * HUF symbols remain "private" (internal symbols for library only).26 * Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */27#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)28# define HUF_PUBLIC_API __attribute__ ((visibility ("default")))29#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */30# define HUF_PUBLIC_API __declspec(dllexport)31#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)32# define HUF_PUBLIC_API __declspec(dllimport) /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */33#else34# define HUF_PUBLIC_API35#endif36 37 38/* ========================== */39/* *** simple functions *** */40/* ========================== */41 42/* HUF_compress() :43 * Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.44 * 'dst' buffer must be already allocated.45 * Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).46 * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.47 * @return : size of compressed data (<= `dstCapacity`).48 * Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!49 * if HUF_isError(return), compression failed (more details using HUF_getErrorName())50 */51HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,52 const void* src, size_t srcSize);53 54/* HUF_decompress() :55 * Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',56 * into already allocated buffer 'dst', of minimum size 'dstSize'.57 * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.58 * Note : in contrast with FSE, HUF_decompress can regenerate59 * RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,60 * because it knows size to regenerate (originalSize).61 * @return : size of regenerated data (== originalSize),62 * or an error code, which can be tested using HUF_isError()63 */64HUF_PUBLIC_API size_t HUF_decompress(void* dst, size_t originalSize,65 const void* cSrc, size_t cSrcSize);66 67 68/* *** Tool functions *** */69#define HUF_BLOCKSIZE_MAX (128 * 1024) /*< maximum input size for a single block compressed with HUF_compress */70HUF_PUBLIC_API size_t HUF_compressBound(size_t size); /*< maximum compressed size (worst case) */71 72/* Error Management */73HUF_PUBLIC_API unsigned HUF_isError(size_t code); /*< tells if a return value is an error code */74HUF_PUBLIC_API const char* HUF_getErrorName(size_t code); /*< provides error code string (useful for debugging) */75 76 77/* *** Advanced function *** */78 79/* HUF_compress2() :80 * Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.81 * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .82 * `tableLog` must be `<= HUF_TABLELOG_MAX` . */83HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,84 const void* src, size_t srcSize,85 unsigned maxSymbolValue, unsigned tableLog);86 87/* HUF_compress4X_wksp() :88 * Same as HUF_compress2(), but uses externally allocated `workSpace`.89 * `workspace` must be at least as large as HUF_WORKSPACE_SIZE */90#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */)91#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64))92HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,93 const void* src, size_t srcSize,94 unsigned maxSymbolValue, unsigned tableLog,95 void* workSpace, size_t wkspSize);96 97#endif /* HUF_H_298734234 */98 99/* ******************************************************************100 * WARNING !!101 * The following section contains advanced and experimental definitions102 * which shall never be used in the context of a dynamic library,103 * because they are not guaranteed to remain stable in the future.104 * Only consider them in association with static linking.105 * *****************************************************************/106#if !defined(HUF_H_HUF_STATIC_LINKING_ONLY)107#define HUF_H_HUF_STATIC_LINKING_ONLY108 109/* *** Dependencies *** */110#include "mem.h" /* U32 */111#define FSE_STATIC_LINKING_ONLY112#include "fse.h"113 114 115/* *** Constants *** */116#define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */117#define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */118#define HUF_SYMBOLVALUE_MAX 255119 120#define HUF_TABLELOG_ABSOLUTEMAX 12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */121#if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)122# error "HUF_TABLELOG_MAX is too large !"123#endif124 125 126/* ****************************************127* Static allocation128******************************************/129/* HUF buffer bounds */130#define HUF_CTABLEBOUND 129131#define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */132#define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */133 134/* static allocation of HUF's Compression Table */135/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */136typedef size_t HUF_CElt; /* consider it an incomplete type */137#define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */138#define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t))139#define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \140 HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */141 142/* static allocation of HUF's DTable */143typedef U32 HUF_DTable;144#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog)))145#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \146 HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }147#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \148 HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }149 150 151/* ****************************************152* Advanced decompression functions153******************************************/154size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */155#ifndef HUF_FORCE_DECOMPRESS_X1156size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */157#endif158 159size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< decodes RLE and uncompressed */160size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< considers RLE and uncompressed as errors */161size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /*< considers RLE and uncompressed as errors */162size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */163size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /*< single-symbol decoder */164#ifndef HUF_FORCE_DECOMPRESS_X1165size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */166size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /*< double-symbols decoder */167#endif168 169 170/* ****************************************171 * HUF detailed API172 * ****************************************/173 174/*! HUF_compress() does the following:175 * 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")176 * 2. (optional) refine tableLog using HUF_optimalTableLog()177 * 3. build Huffman table from count using HUF_buildCTable()178 * 4. save Huffman table to memory buffer using HUF_writeCTable()179 * 5. encode the data stream using HUF_compress4X_usingCTable()180 *181 * The following API allows targeting specific sub-functions for advanced tasks.182 * For example, it's possible to compress several blocks using the same 'CTable',183 * or to save and regenerate 'CTable' using external methods.184 */185unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);186size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits); /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */187size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);188size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize);189size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);190size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);191size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);192int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);193 194typedef enum {195 HUF_repeat_none, /*< Cannot use the previous table */196 HUF_repeat_check, /*< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */197 HUF_repeat_valid /*< Can use the previous table and it is assumed to be valid */198 } HUF_repeat;199/* HUF_compress4X_repeat() :200 * Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.201 * If it uses hufTable it does not modify hufTable or repeat.202 * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.203 * If preferRepeat then the old table will always be used if valid.204 * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */205size_t HUF_compress4X_repeat(void* dst, size_t dstSize,206 const void* src, size_t srcSize,207 unsigned maxSymbolValue, unsigned tableLog,208 void* workSpace, size_t wkspSize, /*< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */209 HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);210 211/* HUF_buildCTable_wksp() :212 * Same as HUF_buildCTable(), but using externally allocated scratch buffer.213 * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.214 */215#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)216#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))217size_t HUF_buildCTable_wksp (HUF_CElt* tree,218 const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,219 void* workSpace, size_t wkspSize);220 221/*! HUF_readStats() :222 * Read compact Huffman tree, saved by HUF_writeCTable().223 * `huffWeight` is destination buffer.224 * @return : size read from `src` , or an error Code .225 * Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */226size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,227 U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,228 const void* src, size_t srcSize);229 230/*! HUF_readStats_wksp() :231 * Same as HUF_readStats() but takes an external workspace which must be232 * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE.233 * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.234 */235#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1)236#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned))237size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize,238 U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,239 const void* src, size_t srcSize,240 void* workspace, size_t wkspSize,241 int bmi2);242 243/* HUF_readCTable() :244 * Loading a CTable saved with HUF_writeCTable() */245size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);246 247/* HUF_getNbBitsFromCTable() :248 * Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX249 * Note 1 : is not inlined, as HUF_CElt definition is private */250U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue);251 252/*253 * HUF_decompress() does the following:254 * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics255 * 2. build Huffman table from save, using HUF_readDTableX?()256 * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()257 */258 259/* HUF_selectDecoder() :260 * Tells which decoder is likely to decode faster,261 * based on a set of pre-computed metrics.262 * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .263 * Assumption : 0 < dstSize <= 128 KB */264U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);265 266/*267 * The minimum workspace size for the `workSpace` used in268 * HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().269 *270 * The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when271 * HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.272 * Buffer overflow errors may potentially occur if code modifications result in273 * a required workspace size greater than that specified in the following274 * macro.275 */276#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9))277#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))278 279#ifndef HUF_FORCE_DECOMPRESS_X2280size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);281size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);282#endif283#ifndef HUF_FORCE_DECOMPRESS_X1284size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);285size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);286#endif287 288size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);289#ifndef HUF_FORCE_DECOMPRESS_X2290size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);291#endif292#ifndef HUF_FORCE_DECOMPRESS_X1293size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);294#endif295 296 297/* ====================== */298/* single stream variants */299/* ====================== */300 301size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);302size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); /*< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U64 U64 */303size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);304size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2);305/* HUF_compress1X_repeat() :306 * Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.307 * If it uses hufTable it does not modify hufTable or repeat.308 * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.309 * If preferRepeat then the old table will always be used if valid.310 * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */311size_t HUF_compress1X_repeat(void* dst, size_t dstSize,312 const void* src, size_t srcSize,313 unsigned maxSymbolValue, unsigned tableLog,314 void* workSpace, size_t wkspSize, /*< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */315 HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible);316 317size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */318#ifndef HUF_FORCE_DECOMPRESS_X1319size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */320#endif321 322size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);323size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize);324#ifndef HUF_FORCE_DECOMPRESS_X2325size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */326size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /*< single-symbol decoder */327#endif328#ifndef HUF_FORCE_DECOMPRESS_X1329size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */330size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /*< double-symbols decoder */331#endif332 333size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); /*< automatic selection of sing or double symbol decoder, based on DTable */334#ifndef HUF_FORCE_DECOMPRESS_X2335size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);336#endif337#ifndef HUF_FORCE_DECOMPRESS_X1338size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);339#endif340 341/* BMI2 variants.342 * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.343 */344size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);345#ifndef HUF_FORCE_DECOMPRESS_X2346size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);347#endif348size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);349size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2);350#ifndef HUF_FORCE_DECOMPRESS_X2351size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);352#endif353#ifndef HUF_FORCE_DECOMPRESS_X1354size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2);355#endif356 357#endif /* HUF_STATIC_LINKING_ONLY */358 359