Takahiro Suzuki | d7bf820 | 2020-12-17 20:21:59 +0900 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (c) 2016-present, Yann Collet, Facebook, Inc. |
| 3 | * All rights reserved. |
| 4 | * |
| 5 | * This source code is licensed under both the BSD-style license (found in the |
| 6 | * LICENSE file in the root directory of this source tree) and the GPLv2 (found |
| 7 | * in the COPYING file in the root directory of this source tree). |
| 8 | * You may select, at your option, one of the above-listed licenses. |
| 9 | */ |
| 10 | |
| 11 | /* This header contains definitions |
| 12 | * that shall **only** be used by modules within lib/compress. |
| 13 | */ |
| 14 | |
| 15 | #ifndef ZSTD_COMPRESS_H |
| 16 | #define ZSTD_COMPRESS_H |
| 17 | |
| 18 | /*-************************************* |
| 19 | * Dependencies |
| 20 | ***************************************/ |
| 21 | #include "zstd_internal.h" |
| 22 | #ifdef ZSTD_MULTITHREAD |
| 23 | # include "zstdmt_compress.h" |
| 24 | #endif |
| 25 | |
| 26 | #if defined (__cplusplus) |
| 27 | extern "C" { |
| 28 | #endif |
| 29 | |
| 30 | |
| 31 | /*-************************************* |
| 32 | * Constants |
| 33 | ***************************************/ |
| 34 | #define kSearchStrength 8 |
| 35 | #define HASH_READ_SIZE 8 |
| 36 | #define ZSTD_DUBT_UNSORTED_MARK 1 /* For btlazy2 strategy, index ZSTD_DUBT_UNSORTED_MARK==1 means "unsorted". |
| 37 | It could be confused for a real successor at index "1", if sorted as larger than its predecessor. |
| 38 | It's not a big deal though : candidate will just be sorted again. |
| 39 | Additionally, candidate position 1 will be lost. |
| 40 | But candidate 1 cannot hide a large tree of candidates, so it's a minimal loss. |
| 41 | The benefit is that ZSTD_DUBT_UNSORTED_MARK cannot be mishandled after table re-use with a different strategy. |
| 42 | This constant is required by ZSTD_compressBlock_btlazy2() and ZSTD_reduceTable_internal() */ |
| 43 | |
| 44 | |
| 45 | /*-************************************* |
| 46 | * Context memory management |
| 47 | ***************************************/ |
| 48 | typedef enum { ZSTDcs_created=0, ZSTDcs_init, ZSTDcs_ongoing, ZSTDcs_ending } ZSTD_compressionStage_e; |
| 49 | typedef enum { zcss_init=0, zcss_load, zcss_flush } ZSTD_cStreamStage; |
| 50 | |
| 51 | typedef struct ZSTD_prefixDict_s { |
| 52 | const void* dict; |
| 53 | size_t dictSize; |
| 54 | ZSTD_dictContentType_e dictContentType; |
| 55 | } ZSTD_prefixDict; |
| 56 | |
| 57 | typedef struct { |
| 58 | void* dictBuffer; |
| 59 | void const* dict; |
| 60 | size_t dictSize; |
| 61 | ZSTD_dictContentType_e dictContentType; |
| 62 | ZSTD_CDict* cdict; |
| 63 | } ZSTD_localDict; |
| 64 | |
| 65 | typedef struct { |
| 66 | U32 CTable[HUF_CTABLE_SIZE_U32(255)]; |
| 67 | HUF_repeat repeatMode; |
| 68 | } ZSTD_hufCTables_t; |
| 69 | |
| 70 | typedef struct { |
| 71 | FSE_CTable offcodeCTable[FSE_CTABLE_SIZE_U32(OffFSELog, MaxOff)]; |
| 72 | FSE_CTable matchlengthCTable[FSE_CTABLE_SIZE_U32(MLFSELog, MaxML)]; |
| 73 | FSE_CTable litlengthCTable[FSE_CTABLE_SIZE_U32(LLFSELog, MaxLL)]; |
| 74 | FSE_repeat offcode_repeatMode; |
| 75 | FSE_repeat matchlength_repeatMode; |
| 76 | FSE_repeat litlength_repeatMode; |
| 77 | } ZSTD_fseCTables_t; |
| 78 | |
| 79 | typedef struct { |
| 80 | ZSTD_hufCTables_t huf; |
| 81 | ZSTD_fseCTables_t fse; |
| 82 | } ZSTD_entropyCTables_t; |
| 83 | |
| 84 | typedef struct { |
| 85 | U32 off; |
| 86 | U32 len; |
| 87 | } ZSTD_match_t; |
| 88 | |
| 89 | typedef struct { |
| 90 | int price; |
| 91 | U32 off; |
| 92 | U32 mlen; |
| 93 | U32 litlen; |
| 94 | U32 rep[ZSTD_REP_NUM]; |
| 95 | } ZSTD_optimal_t; |
| 96 | |
| 97 | typedef enum { zop_dynamic=0, zop_predef } ZSTD_OptPrice_e; |
| 98 | |
| 99 | typedef struct { |
| 100 | /* All tables are allocated inside cctx->workspace by ZSTD_resetCCtx_internal() */ |
| 101 | unsigned* litFreq; /* table of literals statistics, of size 256 */ |
| 102 | unsigned* litLengthFreq; /* table of litLength statistics, of size (MaxLL+1) */ |
| 103 | unsigned* matchLengthFreq; /* table of matchLength statistics, of size (MaxML+1) */ |
| 104 | unsigned* offCodeFreq; /* table of offCode statistics, of size (MaxOff+1) */ |
| 105 | ZSTD_match_t* matchTable; /* list of found matches, of size ZSTD_OPT_NUM+1 */ |
| 106 | ZSTD_optimal_t* priceTable; /* All positions tracked by optimal parser, of size ZSTD_OPT_NUM+1 */ |
| 107 | |
| 108 | U32 litSum; /* nb of literals */ |
| 109 | U32 litLengthSum; /* nb of litLength codes */ |
| 110 | U32 matchLengthSum; /* nb of matchLength codes */ |
| 111 | U32 offCodeSum; /* nb of offset codes */ |
| 112 | U32 litSumBasePrice; /* to compare to log2(litfreq) */ |
| 113 | U32 litLengthSumBasePrice; /* to compare to log2(llfreq) */ |
| 114 | U32 matchLengthSumBasePrice;/* to compare to log2(mlfreq) */ |
| 115 | U32 offCodeSumBasePrice; /* to compare to log2(offreq) */ |
| 116 | ZSTD_OptPrice_e priceType; /* prices can be determined dynamically, or follow a pre-defined cost structure */ |
| 117 | const ZSTD_entropyCTables_t* symbolCosts; /* pre-calculated dictionary statistics */ |
| 118 | ZSTD_literalCompressionMode_e literalCompressionMode; |
| 119 | } optState_t; |
| 120 | |
| 121 | typedef struct { |
| 122 | ZSTD_entropyCTables_t entropy; |
| 123 | U32 rep[ZSTD_REP_NUM]; |
| 124 | } ZSTD_compressedBlockState_t; |
| 125 | |
| 126 | typedef struct { |
| 127 | BYTE const* nextSrc; /* next block here to continue on current prefix */ |
| 128 | BYTE const* base; /* All regular indexes relative to this position */ |
| 129 | BYTE const* dictBase; /* extDict indexes relative to this position */ |
| 130 | U32 dictLimit; /* below that point, need extDict */ |
| 131 | U32 lowLimit; /* below that point, no more valid data */ |
| 132 | } ZSTD_window_t; |
| 133 | |
| 134 | typedef struct ZSTD_matchState_t ZSTD_matchState_t; |
| 135 | struct ZSTD_matchState_t { |
| 136 | ZSTD_window_t window; /* State for window round buffer management */ |
| 137 | U32 loadedDictEnd; /* index of end of dictionary, within context's referential. When dict referential is copied into active context (i.e. not attached), effectively same value as dictSize, since referential starts from zero */ |
| 138 | U32 nextToUpdate; /* index from which to continue table update */ |
| 139 | U32 hashLog3; /* dispatch table : larger == faster, more memory */ |
| 140 | U32* hashTable; |
| 141 | U32* hashTable3; |
| 142 | U32* chainTable; |
| 143 | optState_t opt; /* optimal parser state */ |
| 144 | const ZSTD_matchState_t* dictMatchState; |
| 145 | ZSTD_compressionParameters cParams; |
| 146 | }; |
| 147 | |
| 148 | typedef struct { |
| 149 | ZSTD_compressedBlockState_t* prevCBlock; |
| 150 | ZSTD_compressedBlockState_t* nextCBlock; |
| 151 | ZSTD_matchState_t matchState; |
| 152 | } ZSTD_blockState_t; |
| 153 | |
| 154 | typedef struct { |
| 155 | U32 offset; |
| 156 | U32 checksum; |
| 157 | } ldmEntry_t; |
| 158 | |
| 159 | typedef struct { |
| 160 | ZSTD_window_t window; /* State for the window round buffer management */ |
| 161 | ldmEntry_t* hashTable; |
| 162 | BYTE* bucketOffsets; /* Next position in bucket to insert entry */ |
| 163 | U64 hashPower; /* Used to compute the rolling hash. |
| 164 | * Depends on ldmParams.minMatchLength */ |
| 165 | } ldmState_t; |
| 166 | |
| 167 | typedef struct { |
| 168 | U32 enableLdm; /* 1 if enable long distance matching */ |
| 169 | U32 hashLog; /* Log size of hashTable */ |
| 170 | U32 bucketSizeLog; /* Log bucket size for collision resolution, at most 8 */ |
| 171 | U32 minMatchLength; /* Minimum match length */ |
| 172 | U32 hashRateLog; /* Log number of entries to skip */ |
| 173 | U32 windowLog; /* Window log for the LDM */ |
| 174 | } ldmParams_t; |
| 175 | |
| 176 | typedef struct { |
| 177 | U32 offset; |
| 178 | U32 litLength; |
| 179 | U32 matchLength; |
| 180 | } rawSeq; |
| 181 | |
| 182 | typedef struct { |
| 183 | rawSeq* seq; /* The start of the sequences */ |
| 184 | size_t pos; /* The position where reading stopped. <= size. */ |
| 185 | size_t size; /* The number of sequences. <= capacity. */ |
| 186 | size_t capacity; /* The capacity starting from `seq` pointer */ |
| 187 | } rawSeqStore_t; |
| 188 | |
| 189 | struct ZSTD_CCtx_params_s { |
| 190 | ZSTD_format_e format; |
| 191 | ZSTD_compressionParameters cParams; |
| 192 | ZSTD_frameParameters fParams; |
| 193 | |
| 194 | int compressionLevel; |
| 195 | int forceWindow; /* force back-references to respect limit of |
| 196 | * 1<<wLog, even for dictionary */ |
| 197 | size_t targetCBlockSize; /* Tries to fit compressed block size to be around targetCBlockSize. |
| 198 | * No target when targetCBlockSize == 0. |
| 199 | * There is no guarantee on compressed block size */ |
| 200 | |
| 201 | ZSTD_dictAttachPref_e attachDictPref; |
| 202 | ZSTD_literalCompressionMode_e literalCompressionMode; |
| 203 | |
| 204 | /* Multithreading: used to pass parameters to mtctx */ |
| 205 | int nbWorkers; |
| 206 | size_t jobSize; |
| 207 | int overlapLog; |
| 208 | int rsyncable; |
| 209 | |
| 210 | /* Long distance matching parameters */ |
| 211 | ldmParams_t ldmParams; |
| 212 | |
| 213 | /* Internal use, for createCCtxParams() and freeCCtxParams() only */ |
| 214 | ZSTD_customMem customMem; |
| 215 | }; /* typedef'd to ZSTD_CCtx_params within "zstd.h" */ |
| 216 | |
| 217 | struct ZSTD_CCtx_s { |
| 218 | ZSTD_compressionStage_e stage; |
| 219 | int cParamsChanged; /* == 1 if cParams(except wlog) or compression level are changed in requestedParams. Triggers transmission of new params to ZSTDMT (if available) then reset to 0. */ |
| 220 | int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */ |
| 221 | ZSTD_CCtx_params requestedParams; |
| 222 | ZSTD_CCtx_params appliedParams; |
| 223 | U32 dictID; |
| 224 | |
| 225 | int workSpaceOversizedDuration; |
| 226 | void* workSpace; |
| 227 | size_t workSpaceSize; |
| 228 | size_t blockSize; |
| 229 | unsigned long long pledgedSrcSizePlusOne; /* this way, 0 (default) == unknown */ |
| 230 | unsigned long long consumedSrcSize; |
| 231 | unsigned long long producedCSize; |
| 232 | XXH64_state_t xxhState; |
| 233 | ZSTD_customMem customMem; |
| 234 | size_t staticSize; |
| 235 | |
| 236 | seqStore_t seqStore; /* sequences storage ptrs */ |
| 237 | ldmState_t ldmState; /* long distance matching state */ |
| 238 | rawSeq* ldmSequences; /* Storage for the ldm output sequences */ |
| 239 | size_t maxNbLdmSequences; |
| 240 | rawSeqStore_t externSeqStore; /* Mutable reference to external sequences */ |
| 241 | ZSTD_blockState_t blockState; |
| 242 | U32* entropyWorkspace; /* entropy workspace of HUF_WORKSPACE_SIZE bytes */ |
| 243 | |
| 244 | /* streaming */ |
| 245 | char* inBuff; |
| 246 | size_t inBuffSize; |
| 247 | size_t inToCompress; |
| 248 | size_t inBuffPos; |
| 249 | size_t inBuffTarget; |
| 250 | char* outBuff; |
| 251 | size_t outBuffSize; |
| 252 | size_t outBuffContentSize; |
| 253 | size_t outBuffFlushedSize; |
| 254 | ZSTD_cStreamStage streamStage; |
| 255 | U32 frameEnded; |
| 256 | |
| 257 | /* Dictionary */ |
| 258 | ZSTD_localDict localDict; |
| 259 | const ZSTD_CDict* cdict; |
| 260 | ZSTD_prefixDict prefixDict; /* single-usage dictionary */ |
| 261 | |
| 262 | /* Multi-threading */ |
| 263 | #ifdef ZSTD_MULTITHREAD |
| 264 | ZSTDMT_CCtx* mtctx; |
| 265 | #endif |
| 266 | }; |
| 267 | |
| 268 | typedef enum { ZSTD_dtlm_fast, ZSTD_dtlm_full } ZSTD_dictTableLoadMethod_e; |
| 269 | |
| 270 | typedef enum { ZSTD_noDict = 0, ZSTD_extDict = 1, ZSTD_dictMatchState = 2 } ZSTD_dictMode_e; |
| 271 | |
| 272 | |
| 273 | typedef size_t (*ZSTD_blockCompressor) ( |
| 274 | ZSTD_matchState_t* bs, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], |
| 275 | void const* src, size_t srcSize); |
| 276 | ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_dictMode_e dictMode); |
| 277 | |
| 278 | |
| 279 | MEM_STATIC U32 ZSTD_LLcode(U32 litLength) |
| 280 | { |
| 281 | static const BYTE LL_Code[64] = { 0, 1, 2, 3, 4, 5, 6, 7, |
| 282 | 8, 9, 10, 11, 12, 13, 14, 15, |
| 283 | 16, 16, 17, 17, 18, 18, 19, 19, |
| 284 | 20, 20, 20, 20, 21, 21, 21, 21, |
| 285 | 22, 22, 22, 22, 22, 22, 22, 22, |
| 286 | 23, 23, 23, 23, 23, 23, 23, 23, |
| 287 | 24, 24, 24, 24, 24, 24, 24, 24, |
| 288 | 24, 24, 24, 24, 24, 24, 24, 24 }; |
| 289 | static const U32 LL_deltaCode = 19; |
| 290 | return (litLength > 63) ? ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength]; |
| 291 | } |
| 292 | |
| 293 | /* ZSTD_MLcode() : |
| 294 | * note : mlBase = matchLength - MINMATCH; |
| 295 | * because it's the format it's stored in seqStore->sequences */ |
| 296 | MEM_STATIC U32 ZSTD_MLcode(U32 mlBase) |
| 297 | { |
| 298 | static const BYTE ML_Code[128] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, |
| 299 | 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, |
| 300 | 32, 32, 33, 33, 34, 34, 35, 35, 36, 36, 36, 36, 37, 37, 37, 37, |
| 301 | 38, 38, 38, 38, 38, 38, 38, 38, 39, 39, 39, 39, 39, 39, 39, 39, |
| 302 | 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, |
| 303 | 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, |
| 304 | 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, |
| 305 | 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42 }; |
| 306 | static const U32 ML_deltaCode = 36; |
| 307 | return (mlBase > 127) ? ZSTD_highbit32(mlBase) + ML_deltaCode : ML_Code[mlBase]; |
| 308 | } |
| 309 | |
| 310 | /*! ZSTD_storeSeq() : |
| 311 | * Store a sequence (literal length, literals, offset code and match length code) into seqStore_t. |
| 312 | * `offsetCode` : distance to match + 3 (values 1-3 are repCodes). |
| 313 | * `mlBase` : matchLength - MINMATCH |
| 314 | */ |
| 315 | MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const void* literals, U32 offsetCode, size_t mlBase) |
| 316 | { |
| 317 | #if defined(DEBUGLEVEL) && (DEBUGLEVEL >= 6) |
| 318 | static const BYTE* g_start = NULL; |
| 319 | if (g_start==NULL) g_start = (const BYTE*)literals; /* note : index only works for compression within a single segment */ |
| 320 | { U32 const pos = (U32)((const BYTE*)literals - g_start); |
| 321 | DEBUGLOG(6, "Cpos%7u :%3u literals, match%4u bytes at offCode%7u", |
| 322 | pos, (U32)litLength, (U32)mlBase+MINMATCH, (U32)offsetCode); |
| 323 | } |
| 324 | #endif |
| 325 | assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq); |
| 326 | /* copy Literals */ |
| 327 | assert(seqStorePtr->maxNbLit <= 128 KB); |
| 328 | assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + seqStorePtr->maxNbLit); |
| 329 | ZSTD_wildcopy(seqStorePtr->lit, literals, litLength, ZSTD_no_overlap); |
| 330 | seqStorePtr->lit += litLength; |
| 331 | |
| 332 | /* literal Length */ |
| 333 | if (litLength>0xFFFF) { |
| 334 | assert(seqStorePtr->longLengthID == 0); /* there can only be a single long length */ |
| 335 | seqStorePtr->longLengthID = 1; |
| 336 | seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); |
| 337 | } |
| 338 | seqStorePtr->sequences[0].litLength = (U16)litLength; |
| 339 | |
| 340 | /* match offset */ |
| 341 | seqStorePtr->sequences[0].offset = offsetCode + 1; |
| 342 | |
| 343 | /* match Length */ |
| 344 | if (mlBase>0xFFFF) { |
| 345 | assert(seqStorePtr->longLengthID == 0); /* there can only be a single long length */ |
| 346 | seqStorePtr->longLengthID = 2; |
| 347 | seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); |
| 348 | } |
| 349 | seqStorePtr->sequences[0].matchLength = (U16)mlBase; |
| 350 | |
| 351 | seqStorePtr->sequences++; |
| 352 | } |
| 353 | |
| 354 | |
| 355 | /*-************************************* |
| 356 | * Match length counter |
| 357 | ***************************************/ |
| 358 | static unsigned ZSTD_NbCommonBytes (size_t val) |
| 359 | { |
| 360 | if (MEM_isLittleEndian()) { |
| 361 | if (MEM_64bits()) { |
| 362 | # if defined(_MSC_VER) && defined(_WIN64) |
| 363 | unsigned long r = 0; |
| 364 | _BitScanForward64( &r, (U64)val ); |
| 365 | return (unsigned)(r>>3); |
| 366 | # elif defined(__GNUC__) && (__GNUC__ >= 4) |
| 367 | return (__builtin_ctzll((U64)val) >> 3); |
| 368 | # else |
| 369 | static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, |
| 370 | 0, 3, 1, 3, 1, 4, 2, 7, |
| 371 | 0, 2, 3, 6, 1, 5, 3, 5, |
| 372 | 1, 3, 4, 4, 2, 5, 6, 7, |
| 373 | 7, 0, 1, 2, 3, 3, 4, 6, |
| 374 | 2, 6, 5, 5, 3, 4, 5, 6, |
| 375 | 7, 1, 2, 4, 6, 4, 4, 5, |
| 376 | 7, 2, 6, 5, 7, 6, 7, 7 }; |
| 377 | return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58]; |
| 378 | # endif |
| 379 | } else { /* 32 bits */ |
| 380 | # if defined(_MSC_VER) |
| 381 | unsigned long r=0; |
| 382 | _BitScanForward( &r, (U32)val ); |
| 383 | return (unsigned)(r>>3); |
| 384 | # elif defined(__GNUC__) && (__GNUC__ >= 3) |
| 385 | return (__builtin_ctz((U32)val) >> 3); |
| 386 | # else |
| 387 | static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, |
| 388 | 3, 2, 2, 1, 3, 2, 0, 1, |
| 389 | 3, 3, 1, 2, 2, 2, 2, 0, |
| 390 | 3, 1, 2, 0, 1, 0, 1, 1 }; |
| 391 | return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27]; |
| 392 | # endif |
| 393 | } |
| 394 | } else { /* Big Endian CPU */ |
| 395 | if (MEM_64bits()) { |
| 396 | # if defined(_MSC_VER) && defined(_WIN64) |
| 397 | unsigned long r = 0; |
| 398 | _BitScanReverse64( &r, val ); |
| 399 | return (unsigned)(r>>3); |
| 400 | # elif defined(__GNUC__) && (__GNUC__ >= 4) |
| 401 | return (__builtin_clzll(val) >> 3); |
| 402 | # else |
| 403 | unsigned r; |
| 404 | const unsigned n32 = sizeof(size_t)*4; /* calculate this way due to compiler complaining in 32-bits mode */ |
| 405 | if (!(val>>n32)) { r=4; } else { r=0; val>>=n32; } |
| 406 | if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; } |
| 407 | r += (!val); |
| 408 | return r; |
| 409 | # endif |
| 410 | } else { /* 32 bits */ |
| 411 | # if defined(_MSC_VER) |
| 412 | unsigned long r = 0; |
| 413 | _BitScanReverse( &r, (unsigned long)val ); |
| 414 | return (unsigned)(r>>3); |
| 415 | # elif defined(__GNUC__) && (__GNUC__ >= 3) |
| 416 | return (__builtin_clz((U32)val) >> 3); |
| 417 | # else |
| 418 | unsigned r; |
| 419 | if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; } |
| 420 | r += (!val); |
| 421 | return r; |
| 422 | # endif |
| 423 | } } |
| 424 | } |
| 425 | |
| 426 | |
| 427 | MEM_STATIC size_t ZSTD_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* const pInLimit) |
| 428 | { |
| 429 | const BYTE* const pStart = pIn; |
| 430 | const BYTE* const pInLoopLimit = pInLimit - (sizeof(size_t)-1); |
| 431 | |
| 432 | if (pIn < pInLoopLimit) { |
| 433 | { size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn); |
| 434 | if (diff) return ZSTD_NbCommonBytes(diff); } |
| 435 | pIn+=sizeof(size_t); pMatch+=sizeof(size_t); |
| 436 | while (pIn < pInLoopLimit) { |
| 437 | size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn); |
| 438 | if (!diff) { pIn+=sizeof(size_t); pMatch+=sizeof(size_t); continue; } |
| 439 | pIn += ZSTD_NbCommonBytes(diff); |
| 440 | return (size_t)(pIn - pStart); |
| 441 | } } |
| 442 | if (MEM_64bits() && (pIn<(pInLimit-3)) && (MEM_read32(pMatch) == MEM_read32(pIn))) { pIn+=4; pMatch+=4; } |
| 443 | if ((pIn<(pInLimit-1)) && (MEM_read16(pMatch) == MEM_read16(pIn))) { pIn+=2; pMatch+=2; } |
| 444 | if ((pIn<pInLimit) && (*pMatch == *pIn)) pIn++; |
| 445 | return (size_t)(pIn - pStart); |
| 446 | } |
| 447 | |
| 448 | /** ZSTD_count_2segments() : |
| 449 | * can count match length with `ip` & `match` in 2 different segments. |
| 450 | * convention : on reaching mEnd, match count continue starting from iStart |
| 451 | */ |
| 452 | MEM_STATIC size_t |
| 453 | ZSTD_count_2segments(const BYTE* ip, const BYTE* match, |
| 454 | const BYTE* iEnd, const BYTE* mEnd, const BYTE* iStart) |
| 455 | { |
| 456 | const BYTE* const vEnd = MIN( ip + (mEnd - match), iEnd); |
| 457 | size_t const matchLength = ZSTD_count(ip, match, vEnd); |
| 458 | if (match + matchLength != mEnd) return matchLength; |
| 459 | DEBUGLOG(7, "ZSTD_count_2segments: found a 2-parts match (current length==%zu)", matchLength); |
| 460 | DEBUGLOG(7, "distance from match beginning to end dictionary = %zi", mEnd - match); |
| 461 | DEBUGLOG(7, "distance from current pos to end buffer = %zi", iEnd - ip); |
| 462 | DEBUGLOG(7, "next byte : ip==%02X, istart==%02X", ip[matchLength], *iStart); |
| 463 | DEBUGLOG(7, "final match length = %zu", matchLength + ZSTD_count(ip+matchLength, iStart, iEnd)); |
| 464 | return matchLength + ZSTD_count(ip+matchLength, iStart, iEnd); |
| 465 | } |
| 466 | |
| 467 | |
| 468 | /*-************************************* |
| 469 | * Hashes |
| 470 | ***************************************/ |
| 471 | static const U32 prime3bytes = 506832829U; |
| 472 | static U32 ZSTD_hash3(U32 u, U32 h) { return ((u << (32-24)) * prime3bytes) >> (32-h) ; } |
| 473 | MEM_STATIC size_t ZSTD_hash3Ptr(const void* ptr, U32 h) { return ZSTD_hash3(MEM_readLE32(ptr), h); } /* only in zstd_opt.h */ |
| 474 | |
| 475 | static const U32 prime4bytes = 2654435761U; |
| 476 | static U32 ZSTD_hash4(U32 u, U32 h) { return (u * prime4bytes) >> (32-h) ; } |
| 477 | static size_t ZSTD_hash4Ptr(const void* ptr, U32 h) { return ZSTD_hash4(MEM_read32(ptr), h); } |
| 478 | |
| 479 | static const U64 prime5bytes = 889523592379ULL; |
| 480 | static size_t ZSTD_hash5(U64 u, U32 h) { return (size_t)(((u << (64-40)) * prime5bytes) >> (64-h)) ; } |
| 481 | static size_t ZSTD_hash5Ptr(const void* p, U32 h) { return ZSTD_hash5(MEM_readLE64(p), h); } |
| 482 | |
| 483 | static const U64 prime6bytes = 227718039650203ULL; |
| 484 | static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u << (64-48)) * prime6bytes) >> (64-h)) ; } |
| 485 | static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); } |
| 486 | |
| 487 | static const U64 prime7bytes = 58295818150454627ULL; |
| 488 | static size_t ZSTD_hash7(U64 u, U32 h) { return (size_t)(((u << (64-56)) * prime7bytes) >> (64-h)) ; } |
| 489 | static size_t ZSTD_hash7Ptr(const void* p, U32 h) { return ZSTD_hash7(MEM_readLE64(p), h); } |
| 490 | |
| 491 | static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL; |
| 492 | static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; } |
| 493 | static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); } |
| 494 | |
| 495 | MEM_STATIC size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls) |
| 496 | { |
| 497 | switch(mls) |
| 498 | { |
| 499 | default: |
| 500 | case 4: return ZSTD_hash4Ptr(p, hBits); |
| 501 | case 5: return ZSTD_hash5Ptr(p, hBits); |
| 502 | case 6: return ZSTD_hash6Ptr(p, hBits); |
| 503 | case 7: return ZSTD_hash7Ptr(p, hBits); |
| 504 | case 8: return ZSTD_hash8Ptr(p, hBits); |
| 505 | } |
| 506 | } |
| 507 | |
| 508 | /** ZSTD_ipow() : |
| 509 | * Return base^exponent. |
| 510 | */ |
| 511 | static U64 ZSTD_ipow(U64 base, U64 exponent) |
| 512 | { |
| 513 | U64 power = 1; |
| 514 | while (exponent) { |
| 515 | if (exponent & 1) power *= base; |
| 516 | exponent >>= 1; |
| 517 | base *= base; |
| 518 | } |
| 519 | return power; |
| 520 | } |
| 521 | |
| 522 | #define ZSTD_ROLL_HASH_CHAR_OFFSET 10 |
| 523 | |
| 524 | /** ZSTD_rollingHash_append() : |
| 525 | * Add the buffer to the hash value. |
| 526 | */ |
| 527 | static U64 ZSTD_rollingHash_append(U64 hash, void const* buf, size_t size) |
| 528 | { |
| 529 | BYTE const* istart = (BYTE const*)buf; |
| 530 | size_t pos; |
| 531 | for (pos = 0; pos < size; ++pos) { |
| 532 | hash *= prime8bytes; |
| 533 | hash += istart[pos] + ZSTD_ROLL_HASH_CHAR_OFFSET; |
| 534 | } |
| 535 | return hash; |
| 536 | } |
| 537 | |
| 538 | /** ZSTD_rollingHash_compute() : |
| 539 | * Compute the rolling hash value of the buffer. |
| 540 | */ |
| 541 | MEM_STATIC U64 ZSTD_rollingHash_compute(void const* buf, size_t size) |
| 542 | { |
| 543 | return ZSTD_rollingHash_append(0, buf, size); |
| 544 | } |
| 545 | |
| 546 | /** ZSTD_rollingHash_primePower() : |
| 547 | * Compute the primePower to be passed to ZSTD_rollingHash_rotate() for a hash |
| 548 | * over a window of length bytes. |
| 549 | */ |
| 550 | MEM_STATIC U64 ZSTD_rollingHash_primePower(U32 length) |
| 551 | { |
| 552 | return ZSTD_ipow(prime8bytes, length - 1); |
| 553 | } |
| 554 | |
| 555 | /** ZSTD_rollingHash_rotate() : |
| 556 | * Rotate the rolling hash by one byte. |
| 557 | */ |
| 558 | MEM_STATIC U64 ZSTD_rollingHash_rotate(U64 hash, BYTE toRemove, BYTE toAdd, U64 primePower) |
| 559 | { |
| 560 | hash -= (toRemove + ZSTD_ROLL_HASH_CHAR_OFFSET) * primePower; |
| 561 | hash *= prime8bytes; |
| 562 | hash += toAdd + ZSTD_ROLL_HASH_CHAR_OFFSET; |
| 563 | return hash; |
| 564 | } |
| 565 | |
| 566 | /*-************************************* |
| 567 | * Round buffer management |
| 568 | ***************************************/ |
| 569 | #if (ZSTD_WINDOWLOG_MAX_64 > 31) |
| 570 | # error "ZSTD_WINDOWLOG_MAX is too large : would overflow ZSTD_CURRENT_MAX" |
| 571 | #endif |
| 572 | /* Max current allowed */ |
| 573 | #define ZSTD_CURRENT_MAX ((3U << 29) + (1U << ZSTD_WINDOWLOG_MAX)) |
| 574 | /* Maximum chunk size before overflow correction needs to be called again */ |
| 575 | #define ZSTD_CHUNKSIZE_MAX \ |
| 576 | ( ((U32)-1) /* Maximum ending current index */ \ |
| 577 | - ZSTD_CURRENT_MAX) /* Maximum beginning lowLimit */ |
| 578 | |
| 579 | /** |
| 580 | * ZSTD_window_clear(): |
| 581 | * Clears the window containing the history by simply setting it to empty. |
| 582 | */ |
| 583 | MEM_STATIC void ZSTD_window_clear(ZSTD_window_t* window) |
| 584 | { |
| 585 | size_t const endT = (size_t)(window->nextSrc - window->base); |
| 586 | U32 const end = (U32)endT; |
| 587 | |
| 588 | window->lowLimit = end; |
| 589 | window->dictLimit = end; |
| 590 | } |
| 591 | |
| 592 | /** |
| 593 | * ZSTD_window_hasExtDict(): |
| 594 | * Returns non-zero if the window has a non-empty extDict. |
| 595 | */ |
| 596 | MEM_STATIC U32 ZSTD_window_hasExtDict(ZSTD_window_t const window) |
| 597 | { |
| 598 | return window.lowLimit < window.dictLimit; |
| 599 | } |
| 600 | |
| 601 | /** |
| 602 | * ZSTD_matchState_dictMode(): |
| 603 | * Inspects the provided matchState and figures out what dictMode should be |
| 604 | * passed to the compressor. |
| 605 | */ |
| 606 | MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_matchState_t *ms) |
| 607 | { |
| 608 | return ZSTD_window_hasExtDict(ms->window) ? |
| 609 | ZSTD_extDict : |
| 610 | ms->dictMatchState != NULL ? |
| 611 | ZSTD_dictMatchState : |
| 612 | ZSTD_noDict; |
| 613 | } |
| 614 | |
| 615 | /** |
| 616 | * ZSTD_window_needOverflowCorrection(): |
| 617 | * Returns non-zero if the indices are getting too large and need overflow |
| 618 | * protection. |
| 619 | */ |
| 620 | MEM_STATIC U32 ZSTD_window_needOverflowCorrection(ZSTD_window_t const window, |
| 621 | void const* srcEnd) |
| 622 | { |
| 623 | U32 const current = (U32)((BYTE const*)srcEnd - window.base); |
| 624 | return current > ZSTD_CURRENT_MAX; |
| 625 | } |
| 626 | |
| 627 | /** |
| 628 | * ZSTD_window_correctOverflow(): |
| 629 | * Reduces the indices to protect from index overflow. |
| 630 | * Returns the correction made to the indices, which must be applied to every |
| 631 | * stored index. |
| 632 | * |
| 633 | * The least significant cycleLog bits of the indices must remain the same, |
| 634 | * which may be 0. Every index up to maxDist in the past must be valid. |
| 635 | * NOTE: (maxDist & cycleMask) must be zero. |
| 636 | */ |
| 637 | MEM_STATIC U32 ZSTD_window_correctOverflow(ZSTD_window_t* window, U32 cycleLog, |
| 638 | U32 maxDist, void const* src) |
| 639 | { |
| 640 | /* preemptive overflow correction: |
| 641 | * 1. correction is large enough: |
| 642 | * lowLimit > (3<<29) ==> current > 3<<29 + 1<<windowLog |
| 643 | * 1<<windowLog <= newCurrent < 1<<chainLog + 1<<windowLog |
| 644 | * |
| 645 | * current - newCurrent |
| 646 | * > (3<<29 + 1<<windowLog) - (1<<windowLog + 1<<chainLog) |
| 647 | * > (3<<29) - (1<<chainLog) |
| 648 | * > (3<<29) - (1<<30) (NOTE: chainLog <= 30) |
| 649 | * > 1<<29 |
| 650 | * |
| 651 | * 2. (ip+ZSTD_CHUNKSIZE_MAX - cctx->base) doesn't overflow: |
| 652 | * After correction, current is less than (1<<chainLog + 1<<windowLog). |
| 653 | * In 64-bit mode we are safe, because we have 64-bit ptrdiff_t. |
| 654 | * In 32-bit mode we are safe, because (chainLog <= 29), so |
| 655 | * ip+ZSTD_CHUNKSIZE_MAX - cctx->base < 1<<32. |
| 656 | * 3. (cctx->lowLimit + 1<<windowLog) < 1<<32: |
| 657 | * windowLog <= 31 ==> 3<<29 + 1<<windowLog < 7<<29 < 1<<32. |
| 658 | */ |
| 659 | U32 const cycleMask = (1U << cycleLog) - 1; |
| 660 | U32 const current = (U32)((BYTE const*)src - window->base); |
| 661 | U32 const newCurrent = (current & cycleMask) + maxDist; |
| 662 | U32 const correction = current - newCurrent; |
| 663 | assert((maxDist & cycleMask) == 0); |
| 664 | assert(current > newCurrent); |
| 665 | /* Loose bound, should be around 1<<29 (see above) */ |
| 666 | assert(correction > 1<<28); |
| 667 | |
| 668 | window->base += correction; |
| 669 | window->dictBase += correction; |
| 670 | window->lowLimit -= correction; |
| 671 | window->dictLimit -= correction; |
| 672 | |
| 673 | DEBUGLOG(4, "Correction of 0x%x bytes to lowLimit=0x%x", correction, |
| 674 | window->lowLimit); |
| 675 | return correction; |
| 676 | } |
| 677 | |
| 678 | /** |
| 679 | * ZSTD_window_enforceMaxDist(): |
| 680 | * Updates lowLimit so that: |
| 681 | * (srcEnd - base) - lowLimit == maxDist + loadedDictEnd |
| 682 | * |
| 683 | * It ensures index is valid as long as index >= lowLimit. |
| 684 | * This must be called before a block compression call. |
| 685 | * |
| 686 | * loadedDictEnd is only defined if a dictionary is in use for current compression. |
| 687 | * As the name implies, loadedDictEnd represents the index at end of dictionary. |
| 688 | * The value lies within context's referential, it can be directly compared to blockEndIdx. |
| 689 | * |
| 690 | * If loadedDictEndPtr is NULL, no dictionary is in use, and we use loadedDictEnd == 0. |
| 691 | * If loadedDictEndPtr is not NULL, we set it to zero after updating lowLimit. |
| 692 | * This is because dictionaries are allowed to be referenced fully |
| 693 | * as long as the last byte of the dictionary is in the window. |
| 694 | * Once input has progressed beyond window size, dictionary cannot be referenced anymore. |
| 695 | * |
| 696 | * In normal dict mode, the dictionary lies between lowLimit and dictLimit. |
| 697 | * In dictMatchState mode, lowLimit and dictLimit are the same, |
| 698 | * and the dictionary is below them. |
| 699 | * forceWindow and dictMatchState are therefore incompatible. |
| 700 | */ |
| 701 | MEM_STATIC void |
| 702 | ZSTD_window_enforceMaxDist(ZSTD_window_t* window, |
| 703 | const void* blockEnd, |
| 704 | U32 maxDist, |
| 705 | U32* loadedDictEndPtr, |
| 706 | const ZSTD_matchState_t** dictMatchStatePtr) |
| 707 | { |
| 708 | U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base); |
| 709 | U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0; |
| 710 | DEBUGLOG(5, "ZSTD_window_enforceMaxDist: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u", |
| 711 | (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd); |
| 712 | |
| 713 | /* - When there is no dictionary : loadedDictEnd == 0. |
| 714 | In which case, the test (blockEndIdx > maxDist) is merely to avoid |
| 715 | overflowing next operation `newLowLimit = blockEndIdx - maxDist`. |
| 716 | - When there is a standard dictionary : |
| 717 | Index referential is copied from the dictionary, |
| 718 | which means it starts from 0. |
| 719 | In which case, loadedDictEnd == dictSize, |
| 720 | and it makes sense to compare `blockEndIdx > maxDist + dictSize` |
| 721 | since `blockEndIdx` also starts from zero. |
| 722 | - When there is an attached dictionary : |
| 723 | loadedDictEnd is expressed within the referential of the context, |
| 724 | so it can be directly compared against blockEndIdx. |
| 725 | */ |
| 726 | if (blockEndIdx > maxDist + loadedDictEnd) { |
| 727 | U32 const newLowLimit = blockEndIdx - maxDist; |
| 728 | if (window->lowLimit < newLowLimit) window->lowLimit = newLowLimit; |
| 729 | if (window->dictLimit < window->lowLimit) { |
| 730 | DEBUGLOG(5, "Update dictLimit to match lowLimit, from %u to %u", |
| 731 | (unsigned)window->dictLimit, (unsigned)window->lowLimit); |
| 732 | window->dictLimit = window->lowLimit; |
| 733 | } |
| 734 | /* On reaching window size, dictionaries are invalidated */ |
| 735 | if (loadedDictEndPtr) *loadedDictEndPtr = 0; |
| 736 | if (dictMatchStatePtr) *dictMatchStatePtr = NULL; |
| 737 | } |
| 738 | } |
| 739 | |
| 740 | /* Similar to ZSTD_window_enforceMaxDist(), |
| 741 | * but only invalidates dictionary |
| 742 | * when input progresses beyond window size. */ |
| 743 | MEM_STATIC void |
| 744 | ZSTD_checkDictValidity(ZSTD_window_t* window, |
| 745 | const void* blockEnd, |
| 746 | U32 maxDist, |
| 747 | U32* loadedDictEndPtr, |
| 748 | const ZSTD_matchState_t** dictMatchStatePtr) |
| 749 | { |
| 750 | U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base); |
| 751 | U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0; |
| 752 | DEBUGLOG(5, "ZSTD_checkDictValidity: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u", |
| 753 | (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd); |
| 754 | |
| 755 | if (loadedDictEnd && (blockEndIdx > maxDist + loadedDictEnd)) { |
| 756 | /* On reaching window size, dictionaries are invalidated */ |
| 757 | if (loadedDictEndPtr) *loadedDictEndPtr = 0; |
| 758 | if (dictMatchStatePtr) *dictMatchStatePtr = NULL; |
| 759 | } |
| 760 | } |
| 761 | |
| 762 | /** |
| 763 | * ZSTD_window_update(): |
| 764 | * Updates the window by appending [src, src + srcSize) to the window. |
| 765 | * If it is not contiguous, the current prefix becomes the extDict, and we |
| 766 | * forget about the extDict. Handles overlap of the prefix and extDict. |
| 767 | * Returns non-zero if the segment is contiguous. |
| 768 | */ |
| 769 | MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window, |
| 770 | void const* src, size_t srcSize) |
| 771 | { |
| 772 | BYTE const* const ip = (BYTE const*)src; |
| 773 | U32 contiguous = 1; |
| 774 | DEBUGLOG(5, "ZSTD_window_update"); |
| 775 | /* Check if blocks follow each other */ |
| 776 | if (src != window->nextSrc) { |
| 777 | /* not contiguous */ |
| 778 | size_t const distanceFromBase = (size_t)(window->nextSrc - window->base); |
| 779 | DEBUGLOG(5, "Non contiguous blocks, new segment starts at %u", window->dictLimit); |
| 780 | window->lowLimit = window->dictLimit; |
| 781 | assert(distanceFromBase == (size_t)(U32)distanceFromBase); /* should never overflow */ |
| 782 | window->dictLimit = (U32)distanceFromBase; |
| 783 | window->dictBase = window->base; |
| 784 | window->base = ip - distanceFromBase; |
| 785 | // ms->nextToUpdate = window->dictLimit; |
| 786 | if (window->dictLimit - window->lowLimit < HASH_READ_SIZE) window->lowLimit = window->dictLimit; /* too small extDict */ |
| 787 | contiguous = 0; |
| 788 | } |
| 789 | window->nextSrc = ip + srcSize; |
| 790 | /* if input and dictionary overlap : reduce dictionary (area presumed modified by input) */ |
| 791 | if ( (ip+srcSize > window->dictBase + window->lowLimit) |
| 792 | & (ip < window->dictBase + window->dictLimit)) { |
| 793 | ptrdiff_t const highInputIdx = (ip + srcSize) - window->dictBase; |
| 794 | U32 const lowLimitMax = (highInputIdx > (ptrdiff_t)window->dictLimit) ? window->dictLimit : (U32)highInputIdx; |
| 795 | window->lowLimit = lowLimitMax; |
| 796 | DEBUGLOG(5, "Overlapping extDict and input : new lowLimit = %u", window->lowLimit); |
| 797 | } |
| 798 | return contiguous; |
| 799 | } |
| 800 | |
| 801 | |
| 802 | /* debug functions */ |
| 803 | #if (DEBUGLEVEL>=2) |
| 804 | |
| 805 | MEM_STATIC double ZSTD_fWeight(U32 rawStat) |
| 806 | { |
| 807 | U32 const fp_accuracy = 8; |
| 808 | U32 const fp_multiplier = (1 << fp_accuracy); |
| 809 | U32 const newStat = rawStat + 1; |
| 810 | U32 const hb = ZSTD_highbit32(newStat); |
| 811 | U32 const BWeight = hb * fp_multiplier; |
| 812 | U32 const FWeight = (newStat << fp_accuracy) >> hb; |
| 813 | U32 const weight = BWeight + FWeight; |
| 814 | assert(hb + fp_accuracy < 31); |
| 815 | return (double)weight / fp_multiplier; |
| 816 | } |
| 817 | |
| 818 | /* display a table content, |
| 819 | * listing each element, its frequency, and its predicted bit cost */ |
| 820 | MEM_STATIC void ZSTD_debugTable(const U32* table, U32 max) |
| 821 | { |
| 822 | unsigned u, sum; |
| 823 | for (u=0, sum=0; u<=max; u++) sum += table[u]; |
| 824 | DEBUGLOG(2, "total nb elts: %u", sum); |
| 825 | for (u=0; u<=max; u++) { |
| 826 | DEBUGLOG(2, "%2u: %5u (%.2f)", |
| 827 | u, table[u], ZSTD_fWeight(sum) - ZSTD_fWeight(table[u]) ); |
| 828 | } |
| 829 | } |
| 830 | |
| 831 | #endif |
| 832 | |
| 833 | |
| 834 | #if defined (__cplusplus) |
| 835 | } |
| 836 | #endif |
| 837 | |
| 838 | |
| 839 | /* ============================================================== |
| 840 | * Private declarations |
| 841 | * These prototypes shall only be called from within lib/compress |
| 842 | * ============================================================== */ |
| 843 | |
| 844 | /* ZSTD_getCParamsFromCCtxParams() : |
| 845 | * cParams are built depending on compressionLevel, src size hints, |
| 846 | * LDM and manually set compression parameters. |
| 847 | */ |
| 848 | ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams( |
| 849 | const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize); |
| 850 | |
| 851 | /*! ZSTD_initCStream_internal() : |
| 852 | * Private use only. Init streaming operation. |
| 853 | * expects params to be valid. |
| 854 | * must receive dict, or cdict, or none, but not both. |
| 855 | * @return : 0, or an error code */ |
| 856 | size_t ZSTD_initCStream_internal(ZSTD_CStream* zcs, |
| 857 | const void* dict, size_t dictSize, |
| 858 | const ZSTD_CDict* cdict, |
| 859 | ZSTD_CCtx_params params, unsigned long long pledgedSrcSize); |
| 860 | |
| 861 | void ZSTD_resetSeqStore(seqStore_t* ssPtr); |
| 862 | |
| 863 | /*! ZSTD_getCParamsFromCDict() : |
| 864 | * as the name implies */ |
| 865 | ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict); |
| 866 | |
| 867 | /* ZSTD_compressBegin_advanced_internal() : |
| 868 | * Private use only. To be called from zstdmt_compress.c. */ |
| 869 | size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx, |
| 870 | const void* dict, size_t dictSize, |
| 871 | ZSTD_dictContentType_e dictContentType, |
| 872 | ZSTD_dictTableLoadMethod_e dtlm, |
| 873 | const ZSTD_CDict* cdict, |
| 874 | ZSTD_CCtx_params params, |
| 875 | unsigned long long pledgedSrcSize); |
| 876 | |
| 877 | /* ZSTD_compress_advanced_internal() : |
| 878 | * Private use only. To be called from zstdmt_compress.c. */ |
| 879 | size_t ZSTD_compress_advanced_internal(ZSTD_CCtx* cctx, |
| 880 | void* dst, size_t dstCapacity, |
| 881 | const void* src, size_t srcSize, |
| 882 | const void* dict,size_t dictSize, |
| 883 | ZSTD_CCtx_params params); |
| 884 | |
| 885 | |
| 886 | /* ZSTD_writeLastEmptyBlock() : |
| 887 | * output an empty Block with end-of-frame mark to complete a frame |
| 888 | * @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h)) |
| 889 | * or an error code if `dstCapacity` is too small (<ZSTD_blockHeaderSize) |
| 890 | */ |
| 891 | size_t ZSTD_writeLastEmptyBlock(void* dst, size_t dstCapacity); |
| 892 | |
| 893 | |
| 894 | /* ZSTD_referenceExternalSequences() : |
| 895 | * Must be called before starting a compression operation. |
| 896 | * seqs must parse a prefix of the source. |
| 897 | * This cannot be used when long range matching is enabled. |
| 898 | * Zstd will use these sequences, and pass the literals to a secondary block |
| 899 | * compressor. |
| 900 | * @return : An error code on failure. |
| 901 | * NOTE: seqs are not verified! Invalid sequences can cause out-of-bounds memory |
| 902 | * access and data corruption. |
| 903 | */ |
| 904 | size_t ZSTD_referenceExternalSequences(ZSTD_CCtx* cctx, rawSeq* seq, size_t nbSeq); |
| 905 | |
| 906 | |
| 907 | #endif /* ZSTD_COMPRESS_H */ |