module/zstd/lib/common/huf.h

   1 /* ******************************************************************
   2  * huff0 huffman codec,
   3  * part of Finite State Entropy library
   4  * Copyright (c) 2013-2020, Yann Collet, Facebook, Inc.
   5  *
   6  * You can contact the author at :
   7  * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy
   8  *
   9  * This source code is licensed under both the BSD-style license (found in the
  10  * LICENSE file in the root directory of this source tree) and the GPLv2 (found
  11  * 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 #if defined (__cplusplus)
  16 extern "C" {
  17 #endif
  18
  19 #ifndef HUF_H_298734234
  20 #define HUF_H_298734234
  21
  22 /* *** Dependencies *** */
  23 #include <stddef.h>    /* size_t */
  24
  25
  26 /* *** library symbols visibility *** */
  27 /* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual,
  28  *        HUF symbols remain "private" (internal symbols for library only).
  29  *        Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */
  30 #if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4)
  31 #  define HUF_PUBLIC_API __attribute__ ((visibility ("default")))
  32 #elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1)   /* Visual expected */
  33 #  define HUF_PUBLIC_API __declspec(dllexport)
  34 #elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1)
  35 #  define HUF_PUBLIC_API __declspec(dllimport)  /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */
  36 #else
  37 #  define HUF_PUBLIC_API
  38 #endif
  39
  40
  41 /* ========================== */
  42 /* ***  simple functions  *** */
  43 /* ========================== */
  44
  45 /** HUF_compress() :
  46  *  Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'.
  47  * 'dst' buffer must be already allocated.
  48  *  Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize).
  49  * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB.
  50  * @return : size of compressed data (<= `dstCapacity`).
  51  *  Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!!
  52  *                   if HUF_isError(return), compression failed (more details using HUF_getErrorName())
  53  */
  54 HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity,
  55                              const void* src, size_t srcSize);
  56
  57 /** HUF_decompress() :
  58  *  Decompress HUF data from buffer 'cSrc', of size 'cSrcSize',
  59  *  into already allocated buffer 'dst', of minimum size 'dstSize'.
  60  * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data.
  61  *  Note : in contrast with FSE, HUF_decompress can regenerate
  62  *         RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data,
  63  *         because it knows size to regenerate (originalSize).
  64  * @return : size of regenerated data (== originalSize),
  65  *           or an error code, which can be tested using HUF_isError()
  66  */
  67 HUF_PUBLIC_API size_t HUF_decompress(void* dst,  size_t originalSize,
  68                                const void* cSrc, size_t cSrcSize);
  69
  70
  71 /* ***   Tool functions *** */
  72 #define HUF_BLOCKSIZE_MAX (128 * 1024)                  /**< maximum input size for a single block compressed with HUF_compress */
  73 HUF_PUBLIC_API size_t HUF_compressBound(size_t size);   /**< maximum compressed size (worst case) */
  74
  75 /* Error Management */
  76 HUF_PUBLIC_API unsigned    HUF_isError(size_t code);       /**< tells if a return value is an error code */
  77 HUF_PUBLIC_API const char* HUF_getErrorName(size_t code);  /**< provides error code string (useful for debugging) */
  78
  79
  80 /* ***   Advanced function   *** */
  81
  82 /** HUF_compress2() :
  83  *  Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`.
  84  * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX .
  85  * `tableLog` must be `<= HUF_TABLELOG_MAX` . */
  86 HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity,
  87                                const void* src, size_t srcSize,
  88                                unsigned maxSymbolValue, unsigned tableLog);
  89
  90 /** HUF_compress4X_wksp() :
  91  *  Same as HUF_compress2(), but uses externally allocated `workSpace`.
  92  * `workspace` must have minimum alignment of 4, and be at least as large as HUF_WORKSPACE_SIZE */
  93 #define HUF_WORKSPACE_SIZE ((6 << 10) + 256)
  94 #define HUF_WORKSPACE_SIZE_U32 (HUF_WORKSPACE_SIZE / sizeof(U32))
  95 HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity,
  96                                      const void* src, size_t srcSize,
  97                                      unsigned maxSymbolValue, unsigned tableLog,
  98                                      void* workSpace, size_t wkspSize);
  99
 100 #endif   /* HUF_H_298734234 */
 101
 102 /* ******************************************************************
 103  *  WARNING !!
 104  *  The following section contains advanced and experimental definitions
 105  *  which shall never be used in the context of a dynamic library,
 106  *  because they are not guaranteed to remain stable in the future.
 107  *  Only consider them in association with static linking.
 108  * *****************************************************************/
 109 #if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY)
 110 #define HUF_H_HUF_STATIC_LINKING_ONLY
 111
 112 /* *** Dependencies *** */
 113 #include "mem.h"   /* U32 */
 114
 115
 116 /* *** Constants *** */
 117 #define HUF_TABLELOG_MAX      12      /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_ABSOLUTEMAX_TABLELOG */
 118 #define HUF_TABLELOG_DEFAULT  11      /* default tableLog value when none specified */
 119 #define HUF_SYMBOLVALUE_MAX  255
 120
 121 #define HUF_TABLELOG_ABSOLUTEMAX  15  /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */
 122 #if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX)
 123 #  error "HUF_TABLELOG_MAX is too large !"
 124 #endif
 125
 126
 127 /* ****************************************
 128 *  Static allocation
 129 ******************************************/
 130 /* HUF buffer bounds */
 131 #define HUF_CTABLEBOUND 129
 132 #define HUF_BLOCKBOUND(size) (size + (size>>8) + 8)   /* only true when incompressible is pre-filtered with fast heuristic */
 133 #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size))   /* Macro version, useful for static allocation */
 134
 135 /* static allocation of HUF's Compression Table */
 136 #define HUF_CTABLE_SIZE_U32(maxSymbolValue)   ((maxSymbolValue)+1)   /* Use tables of U32, for proper alignment */
 137 #define HUF_CTABLE_SIZE(maxSymbolValue)       (HUF_CTABLE_SIZE_U32(maxSymbolValue) * sizeof(U32))
 138 #define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \
 139     U32 name##hb[HUF_CTABLE_SIZE_U32(maxSymbolValue)]; \
 140     void* name##hv = &(name##hb); \
 141     HUF_CElt* name = (HUF_CElt*)(name##hv)   /* no final ; */
 142
 143 /* static allocation of HUF's DTable */
 144 typedef U32 HUF_DTable;
 145 #define HUF_DTABLE_SIZE(maxTableLog)   (1 + (1<<(maxTableLog)))
 146 #define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \
 147         HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) }
 148 #define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \
 149         HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) }
 150
 151
 152 /* ****************************************
 153 *  Advanced decompression functions
 154 ******************************************/
 155 size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
 156 #ifndef HUF_FORCE_DECOMPRESS_X1
 157 size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
 158 #endif
 159
 160 size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< decodes RLE and uncompressed */
 161 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 */
 162 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 */
 163 size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
 164 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 */
 165 #ifndef HUF_FORCE_DECOMPRESS_X1
 166 size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
 167 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 */
 168 #endif
 169
 170
 171 /* ****************************************
 172  *  HUF detailed API
 173  * ****************************************/
 174
 175 /*! HUF_compress() does the following:
 176  *  1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h")
 177  *  2. (optional) refine tableLog using HUF_optimalTableLog()
 178  *  3. build Huffman table from count using HUF_buildCTable()
 179  *  4. save Huffman table to memory buffer using HUF_writeCTable()
 180  *  5. encode the data stream using HUF_compress4X_usingCTable()
 181  *
 182  *  The following API allows targeting specific sub-functions for advanced tasks.
 183  *  For example, it's possible to compress several blocks using the same 'CTable',
 184  *  or to save and regenerate 'CTable' using external methods.
 185  */
 186 unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue);
 187 typedef struct HUF_CElt_s HUF_CElt;   /* incomplete type */
 188 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 */
 189 size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog);
 190 size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
 191 size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
 192 int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue);
 193
 194 typedef 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 size_t HUF_compress4X_repeat(void* dst, size_t dstSize,
 205                        const void* src, size_t srcSize,
 206                        unsigned maxSymbolValue, unsigned tableLog,
 207                        void* workSpace, size_t wkspSize,    /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
 208                        HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
 209
 210 /** HUF_buildCTable_wksp() :
 211  *  Same as HUF_buildCTable(), but using externally allocated scratch buffer.
 212  * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE.
 213  */
 214 #define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1)
 215 #define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned))
 216 size_t HUF_buildCTable_wksp (HUF_CElt* tree,
 217                        const unsigned* count, U32 maxSymbolValue, U32 maxNbBits,
 218                              void* workSpace, size_t wkspSize);
 219
 220 /*! HUF_readStats() :
 221  *  Read compact Huffman tree, saved by HUF_writeCTable().
 222  * `huffWeight` is destination buffer.
 223  * @return : size read from `src` , or an error Code .
 224  *  Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */
 225 size_t HUF_readStats(BYTE* huffWeight, size_t hwSize,
 226                      U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr,
 227                      const void* src, size_t srcSize);
 228
 229 /** HUF_readCTable() :
 230  *  Loading a CTable saved with HUF_writeCTable() */
 231 size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights);
 232
 233 /** HUF_getNbBits() :
 234  *  Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX
 235  *  Note 1 : is not inlined, as HUF_CElt definition is private
 236  *  Note 2 : const void* used, so that it can provide a statically allocated table as argument (which uses type U32) */
 237 U32 HUF_getNbBits(const void* symbolTable, U32 symbolValue);
 238
 239 /*
 240  * HUF_decompress() does the following:
 241  * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics
 242  * 2. build Huffman table from save, using HUF_readDTableX?()
 243  * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable()
 244  */
 245
 246 /** HUF_selectDecoder() :
 247  *  Tells which decoder is likely to decode faster,
 248  *  based on a set of pre-computed metrics.
 249  * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 .
 250  *  Assumption : 0 < dstSize <= 128 KB */
 251 U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize);
 252
 253 /**
 254  *  The minimum workspace size for the `workSpace` used in
 255  *  HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp().
 256  *
 257  *  The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when
 258  *  HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15.
 259  *  Buffer overflow errors may potentially occur if code modifications result in
 260  *  a required workspace size greater than that specified in the following
 261  *  macro.
 262  */
 263 #define HUF_DECOMPRESS_WORKSPACE_SIZE (2 << 10)
 264 #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32))
 265
 266 #ifndef HUF_FORCE_DECOMPRESS_X2
 267 size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize);
 268 size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
 269 #endif
 270 #ifndef HUF_FORCE_DECOMPRESS_X1
 271 size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize);
 272 size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize);
 273 #endif
 274
 275 size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 276 #ifndef HUF_FORCE_DECOMPRESS_X2
 277 size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 278 #endif
 279 #ifndef HUF_FORCE_DECOMPRESS_X1
 280 size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 281 #endif
 282
 283
 284 /* ====================== */
 285 /* single stream variants */
 286 /* ====================== */
 287
 288 size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog);
 289 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_U32 unsigned */
 290 size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable);
 291 /** HUF_compress1X_repeat() :
 292  *  Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none.
 293  *  If it uses hufTable it does not modify hufTable or repeat.
 294  *  If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used.
 295  *  If preferRepeat then the old table will always be used if valid. */
 296 size_t HUF_compress1X_repeat(void* dst, size_t dstSize,
 297                        const void* src, size_t srcSize,
 298                        unsigned maxSymbolValue, unsigned tableLog,
 299                        void* workSpace, size_t wkspSize,   /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */
 300                        HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2);
 301
 302 size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* single-symbol decoder */
 303 #ifndef HUF_FORCE_DECOMPRESS_X1
 304 size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /* double-symbol decoder */
 305 #endif
 306
 307 size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);
 308 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);
 309 #ifndef HUF_FORCE_DECOMPRESS_X2
 310 size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< single-symbol decoder */
 311 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 */
 312 #endif
 313 #ifndef HUF_FORCE_DECOMPRESS_X1
 314 size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize);   /**< double-symbols decoder */
 315 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 */
 316 #endif
 317
 318 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 */
 319 #ifndef HUF_FORCE_DECOMPRESS_X2
 320 size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 321 #endif
 322 #ifndef HUF_FORCE_DECOMPRESS_X1
 323 size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable);
 324 #endif
 325
 326 /* BMI2 variants.
 327  * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0.
 328  */
 329 size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
 330 #ifndef HUF_FORCE_DECOMPRESS_X2
 331 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);
 332 #endif
 333 size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2);
 334 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);
 335
 336 #endif /* HUF_STATIC_LINKING_ONLY */
 337
 338 #if defined (__cplusplus)
 339 }
 340 #endif