1e0c1b49fSNick Terrell /* ****************************************************************** 2e0c1b49fSNick Terrell * huff0 huffman codec, 3e0c1b49fSNick Terrell * part of Finite State Entropy library 4e0c1b49fSNick Terrell * Copyright (c) Yann Collet, Facebook, Inc. 5e0c1b49fSNick Terrell * 6e0c1b49fSNick Terrell * You can contact the author at : 7e0c1b49fSNick Terrell * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy 8e0c1b49fSNick Terrell * 9e0c1b49fSNick Terrell * This source code is licensed under both the BSD-style license (found in the 10e0c1b49fSNick Terrell * LICENSE file in the root directory of this source tree) and the GPLv2 (found 11e0c1b49fSNick Terrell * in the COPYING file in the root directory of this source tree). 12e0c1b49fSNick Terrell * You may select, at your option, one of the above-listed licenses. 13e0c1b49fSNick Terrell ****************************************************************** */ 14e0c1b49fSNick Terrell 15e0c1b49fSNick Terrell 16e0c1b49fSNick Terrell #ifndef HUF_H_298734234 17e0c1b49fSNick Terrell #define HUF_H_298734234 18e0c1b49fSNick Terrell 19e0c1b49fSNick Terrell /* *** Dependencies *** */ 20e0c1b49fSNick Terrell #include "zstd_deps.h" /* size_t */ 21e0c1b49fSNick Terrell 22e0c1b49fSNick Terrell 23e0c1b49fSNick Terrell /* *** library symbols visibility *** */ 24e0c1b49fSNick Terrell /* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual, 25e0c1b49fSNick Terrell * HUF symbols remain "private" (internal symbols for library only). 26e0c1b49fSNick Terrell * Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */ 27e0c1b49fSNick Terrell #if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4) 28e0c1b49fSNick Terrell # define HUF_PUBLIC_API __attribute__ ((visibility ("default"))) 29e0c1b49fSNick Terrell #elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */ 30e0c1b49fSNick Terrell # define HUF_PUBLIC_API __declspec(dllexport) 31e0c1b49fSNick Terrell #elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1) 32e0c1b49fSNick Terrell # define HUF_PUBLIC_API __declspec(dllimport) /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */ 33e0c1b49fSNick Terrell #else 34e0c1b49fSNick Terrell # define HUF_PUBLIC_API 35e0c1b49fSNick Terrell #endif 36e0c1b49fSNick Terrell 37e0c1b49fSNick Terrell 38e0c1b49fSNick Terrell /* ========================== */ 39e0c1b49fSNick Terrell /* *** simple functions *** */ 40e0c1b49fSNick Terrell /* ========================== */ 41e0c1b49fSNick Terrell 42e0c1b49fSNick Terrell /* HUF_compress() : 43e0c1b49fSNick Terrell * Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'. 44e0c1b49fSNick Terrell * 'dst' buffer must be already allocated. 45e0c1b49fSNick Terrell * Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize). 46e0c1b49fSNick Terrell * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB. 47e0c1b49fSNick Terrell * @return : size of compressed data (<= `dstCapacity`). 48e0c1b49fSNick Terrell * Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!! 49e0c1b49fSNick Terrell * if HUF_isError(return), compression failed (more details using HUF_getErrorName()) 50e0c1b49fSNick Terrell */ 51e0c1b49fSNick Terrell HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity, 52e0c1b49fSNick Terrell const void* src, size_t srcSize); 53e0c1b49fSNick Terrell 54e0c1b49fSNick Terrell /* HUF_decompress() : 55e0c1b49fSNick Terrell * Decompress HUF data from buffer 'cSrc', of size 'cSrcSize', 56e0c1b49fSNick Terrell * into already allocated buffer 'dst', of minimum size 'dstSize'. 57e0c1b49fSNick Terrell * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data. 58e0c1b49fSNick Terrell * Note : in contrast with FSE, HUF_decompress can regenerate 59e0c1b49fSNick Terrell * RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data, 60e0c1b49fSNick Terrell * because it knows size to regenerate (originalSize). 61e0c1b49fSNick Terrell * @return : size of regenerated data (== originalSize), 62e0c1b49fSNick Terrell * or an error code, which can be tested using HUF_isError() 63e0c1b49fSNick Terrell */ 64e0c1b49fSNick Terrell HUF_PUBLIC_API size_t HUF_decompress(void* dst, size_t originalSize, 65e0c1b49fSNick Terrell const void* cSrc, size_t cSrcSize); 66e0c1b49fSNick Terrell 67e0c1b49fSNick Terrell 68e0c1b49fSNick Terrell /* *** Tool functions *** */ 69e0c1b49fSNick Terrell #define HUF_BLOCKSIZE_MAX (128 * 1024) /*< maximum input size for a single block compressed with HUF_compress */ 70e0c1b49fSNick Terrell HUF_PUBLIC_API size_t HUF_compressBound(size_t size); /*< maximum compressed size (worst case) */ 71e0c1b49fSNick Terrell 72e0c1b49fSNick Terrell /* Error Management */ 73e0c1b49fSNick Terrell HUF_PUBLIC_API unsigned HUF_isError(size_t code); /*< tells if a return value is an error code */ 74e0c1b49fSNick Terrell HUF_PUBLIC_API const char* HUF_getErrorName(size_t code); /*< provides error code string (useful for debugging) */ 75e0c1b49fSNick Terrell 76e0c1b49fSNick Terrell 77e0c1b49fSNick Terrell /* *** Advanced function *** */ 78e0c1b49fSNick Terrell 79e0c1b49fSNick Terrell /* HUF_compress2() : 80e0c1b49fSNick Terrell * Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`. 81e0c1b49fSNick Terrell * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX . 82e0c1b49fSNick Terrell * `tableLog` must be `<= HUF_TABLELOG_MAX` . */ 83e0c1b49fSNick Terrell HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity, 84e0c1b49fSNick Terrell const void* src, size_t srcSize, 85e0c1b49fSNick Terrell unsigned maxSymbolValue, unsigned tableLog); 86e0c1b49fSNick Terrell 87e0c1b49fSNick Terrell /* HUF_compress4X_wksp() : 88e0c1b49fSNick Terrell * Same as HUF_compress2(), but uses externally allocated `workSpace`. 89*2aa14b1aSNick Terrell * `workspace` must be at least as large as HUF_WORKSPACE_SIZE */ 90*2aa14b1aSNick Terrell #define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */) 91*2aa14b1aSNick Terrell #define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64)) 92e0c1b49fSNick Terrell HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity, 93e0c1b49fSNick Terrell const void* src, size_t srcSize, 94e0c1b49fSNick Terrell unsigned maxSymbolValue, unsigned tableLog, 95e0c1b49fSNick Terrell void* workSpace, size_t wkspSize); 96e0c1b49fSNick Terrell 97e0c1b49fSNick Terrell #endif /* HUF_H_298734234 */ 98e0c1b49fSNick Terrell 99e0c1b49fSNick Terrell /* ****************************************************************** 100e0c1b49fSNick Terrell * WARNING !! 101e0c1b49fSNick Terrell * The following section contains advanced and experimental definitions 102e0c1b49fSNick Terrell * which shall never be used in the context of a dynamic library, 103e0c1b49fSNick Terrell * because they are not guaranteed to remain stable in the future. 104e0c1b49fSNick Terrell * Only consider them in association with static linking. 105e0c1b49fSNick Terrell * *****************************************************************/ 106e0c1b49fSNick Terrell #if !defined(HUF_H_HUF_STATIC_LINKING_ONLY) 107e0c1b49fSNick Terrell #define HUF_H_HUF_STATIC_LINKING_ONLY 108e0c1b49fSNick Terrell 109e0c1b49fSNick Terrell /* *** Dependencies *** */ 110e0c1b49fSNick Terrell #include "mem.h" /* U32 */ 111e0c1b49fSNick Terrell #define FSE_STATIC_LINKING_ONLY 112e0c1b49fSNick Terrell #include "fse.h" 113e0c1b49fSNick Terrell 114e0c1b49fSNick Terrell 115e0c1b49fSNick Terrell /* *** Constants *** */ 116*2aa14b1aSNick Terrell #define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */ 117e0c1b49fSNick Terrell #define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */ 118e0c1b49fSNick Terrell #define HUF_SYMBOLVALUE_MAX 255 119e0c1b49fSNick Terrell 120*2aa14b1aSNick Terrell #define HUF_TABLELOG_ABSOLUTEMAX 12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */ 121e0c1b49fSNick Terrell #if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX) 122e0c1b49fSNick Terrell # error "HUF_TABLELOG_MAX is too large !" 123e0c1b49fSNick Terrell #endif 124e0c1b49fSNick Terrell 125e0c1b49fSNick Terrell 126e0c1b49fSNick Terrell /* **************************************** 127e0c1b49fSNick Terrell * Static allocation 128e0c1b49fSNick Terrell ******************************************/ 129e0c1b49fSNick Terrell /* HUF buffer bounds */ 130e0c1b49fSNick Terrell #define HUF_CTABLEBOUND 129 131e0c1b49fSNick Terrell #define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */ 132e0c1b49fSNick Terrell #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ 133e0c1b49fSNick Terrell 134e0c1b49fSNick Terrell /* static allocation of HUF's Compression Table */ 135e0c1b49fSNick Terrell /* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */ 136*2aa14b1aSNick Terrell typedef size_t HUF_CElt; /* consider it an incomplete type */ 137*2aa14b1aSNick Terrell #define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */ 138*2aa14b1aSNick Terrell #define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t)) 139e0c1b49fSNick Terrell #define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \ 140*2aa14b1aSNick Terrell HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */ 141e0c1b49fSNick Terrell 142e0c1b49fSNick Terrell /* static allocation of HUF's DTable */ 143e0c1b49fSNick Terrell typedef U32 HUF_DTable; 144e0c1b49fSNick Terrell #define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog))) 145e0c1b49fSNick Terrell #define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \ 146e0c1b49fSNick Terrell HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) } 147e0c1b49fSNick Terrell #define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \ 148e0c1b49fSNick Terrell HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) } 149e0c1b49fSNick Terrell 150e0c1b49fSNick Terrell 151e0c1b49fSNick Terrell /* **************************************** 152e0c1b49fSNick Terrell * Advanced decompression functions 153e0c1b49fSNick Terrell ******************************************/ 154e0c1b49fSNick Terrell size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */ 155e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 156e0c1b49fSNick Terrell size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */ 157e0c1b49fSNick Terrell #endif 158e0c1b49fSNick Terrell 159e0c1b49fSNick Terrell size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< decodes RLE and uncompressed */ 160e0c1b49fSNick Terrell size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< considers RLE and uncompressed as errors */ 161e0c1b49fSNick Terrell size_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 */ 162e0c1b49fSNick Terrell size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */ 163e0c1b49fSNick Terrell size_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 */ 164e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 165e0c1b49fSNick Terrell size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */ 166e0c1b49fSNick Terrell size_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 */ 167e0c1b49fSNick Terrell #endif 168e0c1b49fSNick Terrell 169e0c1b49fSNick Terrell 170e0c1b49fSNick Terrell /* **************************************** 171e0c1b49fSNick Terrell * HUF detailed API 172e0c1b49fSNick Terrell * ****************************************/ 173e0c1b49fSNick Terrell 174e0c1b49fSNick Terrell /*! HUF_compress() does the following: 175e0c1b49fSNick Terrell * 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h") 176e0c1b49fSNick Terrell * 2. (optional) refine tableLog using HUF_optimalTableLog() 177e0c1b49fSNick Terrell * 3. build Huffman table from count using HUF_buildCTable() 178e0c1b49fSNick Terrell * 4. save Huffman table to memory buffer using HUF_writeCTable() 179e0c1b49fSNick Terrell * 5. encode the data stream using HUF_compress4X_usingCTable() 180e0c1b49fSNick Terrell * 181e0c1b49fSNick Terrell * The following API allows targeting specific sub-functions for advanced tasks. 182e0c1b49fSNick Terrell * For example, it's possible to compress several blocks using the same 'CTable', 183e0c1b49fSNick Terrell * or to save and regenerate 'CTable' using external methods. 184e0c1b49fSNick Terrell */ 185e0c1b49fSNick Terrell unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue); 186e0c1b49fSNick Terrell size_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 */ 187e0c1b49fSNick Terrell size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog); 188e0c1b49fSNick Terrell size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize); 189e0c1b49fSNick Terrell size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable); 190*2aa14b1aSNick Terrell size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2); 191e0c1b49fSNick Terrell size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue); 192e0c1b49fSNick Terrell int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue); 193e0c1b49fSNick Terrell 194e0c1b49fSNick Terrell typedef enum { 195e0c1b49fSNick Terrell HUF_repeat_none, /*< Cannot use the previous table */ 196e0c1b49fSNick Terrell 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 */ 197e0c1b49fSNick Terrell HUF_repeat_valid /*< Can use the previous table and it is assumed to be valid */ 198e0c1b49fSNick Terrell } HUF_repeat; 199e0c1b49fSNick Terrell /* HUF_compress4X_repeat() : 200e0c1b49fSNick Terrell * Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. 201e0c1b49fSNick Terrell * If it uses hufTable it does not modify hufTable or repeat. 202e0c1b49fSNick Terrell * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. 203*2aa14b1aSNick Terrell * If preferRepeat then the old table will always be used if valid. 204*2aa14b1aSNick Terrell * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */ 205e0c1b49fSNick Terrell size_t HUF_compress4X_repeat(void* dst, size_t dstSize, 206e0c1b49fSNick Terrell const void* src, size_t srcSize, 207e0c1b49fSNick Terrell unsigned maxSymbolValue, unsigned tableLog, 208e0c1b49fSNick Terrell void* workSpace, size_t wkspSize, /*< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */ 209*2aa14b1aSNick Terrell HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible); 210e0c1b49fSNick Terrell 211e0c1b49fSNick Terrell /* HUF_buildCTable_wksp() : 212e0c1b49fSNick Terrell * Same as HUF_buildCTable(), but using externally allocated scratch buffer. 213e0c1b49fSNick Terrell * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE. 214e0c1b49fSNick Terrell */ 215e0c1b49fSNick Terrell #define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1) 216e0c1b49fSNick Terrell #define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned)) 217e0c1b49fSNick Terrell size_t HUF_buildCTable_wksp (HUF_CElt* tree, 218e0c1b49fSNick Terrell const unsigned* count, U32 maxSymbolValue, U32 maxNbBits, 219e0c1b49fSNick Terrell void* workSpace, size_t wkspSize); 220e0c1b49fSNick Terrell 221e0c1b49fSNick Terrell /*! HUF_readStats() : 222e0c1b49fSNick Terrell * Read compact Huffman tree, saved by HUF_writeCTable(). 223e0c1b49fSNick Terrell * `huffWeight` is destination buffer. 224e0c1b49fSNick Terrell * @return : size read from `src` , or an error Code . 225e0c1b49fSNick Terrell * Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */ 226e0c1b49fSNick Terrell size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, 227e0c1b49fSNick Terrell U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, 228e0c1b49fSNick Terrell const void* src, size_t srcSize); 229e0c1b49fSNick Terrell 230e0c1b49fSNick Terrell /*! HUF_readStats_wksp() : 231e0c1b49fSNick Terrell * Same as HUF_readStats() but takes an external workspace which must be 232e0c1b49fSNick Terrell * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE. 233e0c1b49fSNick Terrell * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. 234e0c1b49fSNick Terrell */ 235e0c1b49fSNick Terrell #define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1) 236e0c1b49fSNick Terrell #define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned)) 237e0c1b49fSNick Terrell size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, 238e0c1b49fSNick Terrell U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, 239e0c1b49fSNick Terrell const void* src, size_t srcSize, 240e0c1b49fSNick Terrell void* workspace, size_t wkspSize, 241e0c1b49fSNick Terrell int bmi2); 242e0c1b49fSNick Terrell 243e0c1b49fSNick Terrell /* HUF_readCTable() : 244e0c1b49fSNick Terrell * Loading a CTable saved with HUF_writeCTable() */ 245e0c1b49fSNick Terrell size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights); 246e0c1b49fSNick Terrell 247*2aa14b1aSNick Terrell /* HUF_getNbBitsFromCTable() : 248e0c1b49fSNick Terrell * Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX 249*2aa14b1aSNick Terrell * Note 1 : is not inlined, as HUF_CElt definition is private */ 250*2aa14b1aSNick Terrell U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue); 251e0c1b49fSNick Terrell 252e0c1b49fSNick Terrell /* 253e0c1b49fSNick Terrell * HUF_decompress() does the following: 254e0c1b49fSNick Terrell * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics 255e0c1b49fSNick Terrell * 2. build Huffman table from save, using HUF_readDTableX?() 256e0c1b49fSNick Terrell * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable() 257e0c1b49fSNick Terrell */ 258e0c1b49fSNick Terrell 259e0c1b49fSNick Terrell /* HUF_selectDecoder() : 260e0c1b49fSNick Terrell * Tells which decoder is likely to decode faster, 261e0c1b49fSNick Terrell * based on a set of pre-computed metrics. 262e0c1b49fSNick Terrell * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 . 263e0c1b49fSNick Terrell * Assumption : 0 < dstSize <= 128 KB */ 264e0c1b49fSNick Terrell U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize); 265e0c1b49fSNick Terrell 266e0c1b49fSNick Terrell /* 267e0c1b49fSNick Terrell * The minimum workspace size for the `workSpace` used in 268e0c1b49fSNick Terrell * HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp(). 269e0c1b49fSNick Terrell * 270e0c1b49fSNick Terrell * The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when 271e0c1b49fSNick Terrell * HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15. 272e0c1b49fSNick Terrell * Buffer overflow errors may potentially occur if code modifications result in 273e0c1b49fSNick Terrell * a required workspace size greater than that specified in the following 274e0c1b49fSNick Terrell * macro. 275e0c1b49fSNick Terrell */ 276e0c1b49fSNick Terrell #define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9)) 277e0c1b49fSNick Terrell #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32)) 278e0c1b49fSNick Terrell 279e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 280e0c1b49fSNick Terrell size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize); 281e0c1b49fSNick Terrell size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize); 282e0c1b49fSNick Terrell #endif 283e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 284e0c1b49fSNick Terrell size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize); 285e0c1b49fSNick Terrell size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize); 286e0c1b49fSNick Terrell #endif 287e0c1b49fSNick Terrell 288e0c1b49fSNick Terrell size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); 289e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 290e0c1b49fSNick Terrell size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); 291e0c1b49fSNick Terrell #endif 292e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 293e0c1b49fSNick Terrell size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); 294e0c1b49fSNick Terrell #endif 295e0c1b49fSNick Terrell 296e0c1b49fSNick Terrell 297e0c1b49fSNick Terrell /* ====================== */ 298e0c1b49fSNick Terrell /* single stream variants */ 299e0c1b49fSNick Terrell /* ====================== */ 300e0c1b49fSNick Terrell 301e0c1b49fSNick Terrell size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog); 302*2aa14b1aSNick Terrell size_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 */ 303e0c1b49fSNick Terrell size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable); 304*2aa14b1aSNick Terrell size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2); 305e0c1b49fSNick Terrell /* HUF_compress1X_repeat() : 306e0c1b49fSNick Terrell * Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. 307e0c1b49fSNick Terrell * If it uses hufTable it does not modify hufTable or repeat. 308e0c1b49fSNick Terrell * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. 309*2aa14b1aSNick Terrell * If preferRepeat then the old table will always be used if valid. 310*2aa14b1aSNick Terrell * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */ 311e0c1b49fSNick Terrell size_t HUF_compress1X_repeat(void* dst, size_t dstSize, 312e0c1b49fSNick Terrell const void* src, size_t srcSize, 313e0c1b49fSNick Terrell unsigned maxSymbolValue, unsigned tableLog, 314e0c1b49fSNick Terrell void* workSpace, size_t wkspSize, /*< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */ 315*2aa14b1aSNick Terrell HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible); 316e0c1b49fSNick Terrell 317e0c1b49fSNick Terrell size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */ 318e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 319e0c1b49fSNick Terrell size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */ 320e0c1b49fSNick Terrell #endif 321e0c1b49fSNick Terrell 322e0c1b49fSNick Terrell size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); 323e0c1b49fSNick Terrell size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); 324e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 325e0c1b49fSNick Terrell size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< single-symbol decoder */ 326e0c1b49fSNick Terrell size_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 */ 327e0c1b49fSNick Terrell #endif 328e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 329e0c1b49fSNick Terrell size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /*< double-symbols decoder */ 330e0c1b49fSNick Terrell size_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 */ 331e0c1b49fSNick Terrell #endif 332e0c1b49fSNick Terrell 333e0c1b49fSNick Terrell size_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 */ 334e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 335e0c1b49fSNick Terrell size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); 336e0c1b49fSNick Terrell #endif 337e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 338e0c1b49fSNick Terrell size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); 339e0c1b49fSNick Terrell #endif 340e0c1b49fSNick Terrell 341e0c1b49fSNick Terrell /* BMI2 variants. 342e0c1b49fSNick Terrell * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. 343e0c1b49fSNick Terrell */ 344e0c1b49fSNick Terrell size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2); 345e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 346e0c1b49fSNick Terrell size_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); 347e0c1b49fSNick Terrell #endif 348e0c1b49fSNick Terrell size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2); 349e0c1b49fSNick Terrell size_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); 350e0c1b49fSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X2 351e0c1b49fSNick Terrell size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); 352e0c1b49fSNick Terrell #endif 353*2aa14b1aSNick Terrell #ifndef HUF_FORCE_DECOMPRESS_X1 354*2aa14b1aSNick Terrell size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); 355*2aa14b1aSNick Terrell #endif 356e0c1b49fSNick Terrell 357e0c1b49fSNick Terrell #endif /* HUF_STATIC_LINKING_ONLY */ 358e0c1b49fSNick Terrell 359