Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [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 | #ifndef ZSTD_CCOMMON_H_MODULE |
| 12 | #define ZSTD_CCOMMON_H_MODULE |
| 13 | |
| 14 | /* this module contains definitions which must be identical |
| 15 | * across compression, decompression and dictBuilder. |
| 16 | * It also contains a few functions useful to at least 2 of them |
| 17 | * and which benefit from being inlined */ |
| 18 | |
| 19 | /*-************************************* |
| 20 | * Dependencies |
| 21 | ***************************************/ |
| 22 | #include "compiler.h" |
| 23 | #include "mem.h" |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 24 | #include "debug.h" /* assert, DEBUGLOG, RAWLOG, g_debuglevel */ |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 25 | #include "error_private.h" |
| 26 | #define ZSTD_STATIC_LINKING_ONLY |
| 27 | #include "zstd.h" |
| 28 | #define FSE_STATIC_LINKING_ONLY |
| 29 | #include "fse.h" |
| 30 | #define HUF_STATIC_LINKING_ONLY |
| 31 | #include "huf.h" |
| 32 | #ifndef XXH_STATIC_LINKING_ONLY |
| 33 | # define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */ |
| 34 | #endif |
| 35 | #include "xxhash.h" /* XXH_reset, update, digest */ |
| 36 | |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 37 | #if defined (__cplusplus) |
| 38 | extern "C" { |
| 39 | #endif |
| 40 | |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 41 | /* ---- static assert (debug) --- */ |
| 42 | #define ZSTD_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) |
| 43 | #define ZSTD_isError ERR_isError /* for inlining */ |
| 44 | #define FSE_isError ERR_isError |
| 45 | #define HUF_isError ERR_isError |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 46 | |
| 47 | |
| 48 | /*-************************************* |
| 49 | * shared macros |
| 50 | ***************************************/ |
| 51 | #undef MIN |
| 52 | #undef MAX |
| 53 | #define MIN(a,b) ((a)<(b) ? (a) : (b)) |
| 54 | #define MAX(a,b) ((a)>(b) ? (a) : (b)) |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 55 | |
| 56 | /** |
| 57 | * Return the specified error if the condition evaluates to true. |
| 58 | * |
| 59 | * In debug modes, prints additional information. In order to do that |
| 60 | * (particularly, printing the conditional that failed), this can't just wrap |
| 61 | * RETURN_ERROR(). |
| 62 | */ |
| 63 | #define RETURN_ERROR_IF(cond, err, ...) \ |
| 64 | if (cond) { \ |
| 65 | RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(cond), ZSTD_QUOTE(ERROR(err))); \ |
| 66 | RAWLOG(3, ": " __VA_ARGS__); \ |
| 67 | RAWLOG(3, "\n"); \ |
| 68 | return ERROR(err); \ |
| 69 | } |
| 70 | |
| 71 | /** |
| 72 | * Unconditionally return the specified error. |
| 73 | * |
| 74 | * In debug modes, prints additional information. |
| 75 | */ |
| 76 | #define RETURN_ERROR(err, ...) \ |
| 77 | do { \ |
| 78 | RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", __FILE__, __LINE__, ZSTD_QUOTE(ERROR(err))); \ |
| 79 | RAWLOG(3, ": " __VA_ARGS__); \ |
| 80 | RAWLOG(3, "\n"); \ |
| 81 | return ERROR(err); \ |
| 82 | } while(0); |
| 83 | |
| 84 | /** |
| 85 | * If the provided expression evaluates to an error code, returns that error code. |
| 86 | * |
| 87 | * In debug modes, prints additional information. |
| 88 | */ |
| 89 | #define FORWARD_IF_ERROR(err, ...) \ |
| 90 | do { \ |
| 91 | size_t const err_code = (err); \ |
| 92 | if (ERR_isError(err_code)) { \ |
| 93 | RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", __FILE__, __LINE__, ZSTD_QUOTE(err), ERR_getErrorName(err_code)); \ |
| 94 | RAWLOG(3, ": " __VA_ARGS__); \ |
| 95 | RAWLOG(3, "\n"); \ |
| 96 | return err_code; \ |
| 97 | } \ |
| 98 | } while(0); |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 99 | |
| 100 | |
| 101 | /*-************************************* |
| 102 | * Common constants |
| 103 | ***************************************/ |
| 104 | #define ZSTD_OPT_NUM (1<<12) |
| 105 | |
| 106 | #define ZSTD_REP_NUM 3 /* number of repcodes */ |
| 107 | #define ZSTD_REP_MOVE (ZSTD_REP_NUM-1) |
| 108 | static const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 }; |
| 109 | |
| 110 | #define KB *(1 <<10) |
| 111 | #define MB *(1 <<20) |
| 112 | #define GB *(1U<<30) |
| 113 | |
| 114 | #define BIT7 128 |
| 115 | #define BIT6 64 |
| 116 | #define BIT5 32 |
| 117 | #define BIT4 16 |
| 118 | #define BIT1 2 |
| 119 | #define BIT0 1 |
| 120 | |
| 121 | #define ZSTD_WINDOWLOG_ABSOLUTEMIN 10 |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 122 | static const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 }; |
| 123 | static const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 }; |
| 124 | |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 125 | #define ZSTD_FRAMEIDSIZE 4 /* magic number size */ |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 126 | |
| 127 | #define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */ |
| 128 | static const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE; |
| 129 | typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e; |
| 130 | |
| 131 | #define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */ |
| 132 | #define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */ + MIN_SEQUENCES_SIZE /* nbSeq==0 */) /* for a non-null block */ |
| 133 | |
| 134 | #define HufLog 12 |
| 135 | typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e; |
| 136 | |
| 137 | #define LONGNBSEQ 0x7F00 |
| 138 | |
| 139 | #define MINMATCH 3 |
| 140 | |
| 141 | #define Litbits 8 |
| 142 | #define MaxLit ((1<<Litbits) - 1) |
| 143 | #define MaxML 52 |
| 144 | #define MaxLL 35 |
| 145 | #define DefaultMaxOff 28 |
| 146 | #define MaxOff 31 |
| 147 | #define MaxSeq MAX(MaxLL, MaxML) /* Assumption : MaxOff < MaxLL,MaxML */ |
| 148 | #define MLFSELog 9 |
| 149 | #define LLFSELog 9 |
| 150 | #define OffFSELog 8 |
| 151 | #define MaxFSELog MAX(MAX(MLFSELog, LLFSELog), OffFSELog) |
| 152 | |
| 153 | static const U32 LL_bits[MaxLL+1] = { 0, 0, 0, 0, 0, 0, 0, 0, |
| 154 | 0, 0, 0, 0, 0, 0, 0, 0, |
| 155 | 1, 1, 1, 1, 2, 2, 3, 3, |
| 156 | 4, 6, 7, 8, 9,10,11,12, |
| 157 | 13,14,15,16 }; |
| 158 | static const S16 LL_defaultNorm[MaxLL+1] = { 4, 3, 2, 2, 2, 2, 2, 2, |
| 159 | 2, 2, 2, 2, 2, 1, 1, 1, |
| 160 | 2, 2, 2, 2, 2, 2, 2, 2, |
| 161 | 2, 3, 2, 1, 1, 1, 1, 1, |
| 162 | -1,-1,-1,-1 }; |
| 163 | #define LL_DEFAULTNORMLOG 6 /* for static allocation */ |
| 164 | static const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG; |
| 165 | |
| 166 | static const U32 ML_bits[MaxML+1] = { 0, 0, 0, 0, 0, 0, 0, 0, |
| 167 | 0, 0, 0, 0, 0, 0, 0, 0, |
| 168 | 0, 0, 0, 0, 0, 0, 0, 0, |
| 169 | 0, 0, 0, 0, 0, 0, 0, 0, |
| 170 | 1, 1, 1, 1, 2, 2, 3, 3, |
| 171 | 4, 4, 5, 7, 8, 9,10,11, |
| 172 | 12,13,14,15,16 }; |
| 173 | static const S16 ML_defaultNorm[MaxML+1] = { 1, 4, 3, 2, 2, 2, 2, 2, |
| 174 | 2, 1, 1, 1, 1, 1, 1, 1, |
| 175 | 1, 1, 1, 1, 1, 1, 1, 1, |
| 176 | 1, 1, 1, 1, 1, 1, 1, 1, |
| 177 | 1, 1, 1, 1, 1, 1, 1, 1, |
| 178 | 1, 1, 1, 1, 1, 1,-1,-1, |
| 179 | -1,-1,-1,-1,-1 }; |
| 180 | #define ML_DEFAULTNORMLOG 6 /* for static allocation */ |
| 181 | static const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG; |
| 182 | |
| 183 | static const S16 OF_defaultNorm[DefaultMaxOff+1] = { 1, 1, 1, 1, 1, 1, 2, 2, |
| 184 | 2, 1, 1, 1, 1, 1, 1, 1, |
| 185 | 1, 1, 1, 1, 1, 1, 1, 1, |
| 186 | -1,-1,-1,-1,-1 }; |
| 187 | #define OF_DEFAULTNORMLOG 5 /* for static allocation */ |
| 188 | static const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG; |
| 189 | |
| 190 | |
| 191 | /*-******************************************* |
| 192 | * Shared functions to include for inlining |
| 193 | *********************************************/ |
| 194 | static void ZSTD_copy8(void* dst, const void* src) { memcpy(dst, src, 8); } |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 195 | |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 196 | #define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; } |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 197 | static void ZSTD_copy16(void* dst, const void* src) { memcpy(dst, src, 16); } |
| 198 | #define COPY16(d,s) { ZSTD_copy16(d,s); d+=16; s+=16; } |
| 199 | |
| 200 | #define WILDCOPY_OVERLENGTH 8 |
| 201 | #define VECLEN 16 |
| 202 | |
| 203 | typedef enum { |
| 204 | ZSTD_no_overlap, |
| 205 | ZSTD_overlap_src_before_dst, |
| 206 | /* ZSTD_overlap_dst_before_src, */ |
| 207 | } ZSTD_overlap_e; |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 208 | |
| 209 | /*! ZSTD_wildcopy() : |
| 210 | * custom version of memcpy(), can overwrite up to WILDCOPY_OVERLENGTH bytes (if length==0) */ |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 211 | MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE |
| 212 | void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype) |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 213 | { |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 214 | ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src; |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 215 | const BYTE* ip = (const BYTE*)src; |
| 216 | BYTE* op = (BYTE*)dst; |
| 217 | BYTE* const oend = op + length; |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 218 | |
| 219 | assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8)); |
| 220 | if (length < VECLEN || (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN)) { |
| 221 | do |
| 222 | COPY8(op, ip) |
| 223 | while (op < oend); |
| 224 | } |
| 225 | else { |
| 226 | if ((length & 8) == 0) |
| 227 | COPY8(op, ip); |
| 228 | do { |
| 229 | COPY16(op, ip); |
| 230 | } |
| 231 | while (op < oend); |
| 232 | } |
| 233 | } |
| 234 | |
| 235 | /*! ZSTD_wildcopy_16min() : |
| 236 | * same semantics as ZSTD_wilcopy() except guaranteed to be able to copy 16 bytes at the start */ |
| 237 | MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE |
| 238 | void ZSTD_wildcopy_16min(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype) |
| 239 | { |
| 240 | ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src; |
| 241 | const BYTE* ip = (const BYTE*)src; |
| 242 | BYTE* op = (BYTE*)dst; |
| 243 | BYTE* const oend = op + length; |
| 244 | |
| 245 | assert(length >= 8); |
| 246 | assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8)); |
| 247 | |
| 248 | if (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN) { |
| 249 | do |
| 250 | COPY8(op, ip) |
| 251 | while (op < oend); |
| 252 | } |
| 253 | else { |
| 254 | if ((length & 8) == 0) |
| 255 | COPY8(op, ip); |
| 256 | do { |
| 257 | COPY16(op, ip); |
| 258 | } |
| 259 | while (op < oend); |
| 260 | } |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 261 | } |
| 262 | |
| 263 | MEM_STATIC void ZSTD_wildcopy_e(void* dst, const void* src, void* dstEnd) /* should be faster for decoding, but strangely, not verified on all platform */ |
| 264 | { |
| 265 | const BYTE* ip = (const BYTE*)src; |
| 266 | BYTE* op = (BYTE*)dst; |
| 267 | BYTE* const oend = (BYTE*)dstEnd; |
| 268 | do |
| 269 | COPY8(op, ip) |
| 270 | while (op < oend); |
| 271 | } |
| 272 | |
| 273 | |
| 274 | /*-******************************************* |
| 275 | * Private declarations |
| 276 | *********************************************/ |
| 277 | typedef struct seqDef_s { |
| 278 | U32 offset; |
| 279 | U16 litLength; |
| 280 | U16 matchLength; |
| 281 | } seqDef; |
| 282 | |
| 283 | typedef struct { |
| 284 | seqDef* sequencesStart; |
| 285 | seqDef* sequences; |
| 286 | BYTE* litStart; |
| 287 | BYTE* lit; |
| 288 | BYTE* llCode; |
| 289 | BYTE* mlCode; |
| 290 | BYTE* ofCode; |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 291 | size_t maxNbSeq; |
| 292 | size_t maxNbLit; |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 293 | U32 longLengthID; /* 0 == no longLength; 1 == Lit.longLength; 2 == Match.longLength; */ |
| 294 | U32 longLengthPos; |
| 295 | } seqStore_t; |
| 296 | |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 297 | /** |
| 298 | * Contains the compressed frame size and an upper-bound for the decompressed frame size. |
| 299 | * Note: before using `compressedSize`, check for errors using ZSTD_isError(). |
| 300 | * similarly, before using `decompressedBound`, check for errors using: |
| 301 | * `decompressedBound != ZSTD_CONTENTSIZE_ERROR` |
| 302 | */ |
| 303 | typedef struct { |
| 304 | size_t compressedSize; |
| 305 | unsigned long long decompressedBound; |
| 306 | } ZSTD_frameSizeInfo; /* decompress & legacy */ |
| 307 | |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 308 | const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */ |
| 309 | void ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */ |
| 310 | |
| 311 | /* custom memory allocation functions */ |
| 312 | void* ZSTD_malloc(size_t size, ZSTD_customMem customMem); |
| 313 | void* ZSTD_calloc(size_t size, ZSTD_customMem customMem); |
| 314 | void ZSTD_free(void* ptr, ZSTD_customMem customMem); |
| 315 | |
| 316 | |
| 317 | MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */ |
| 318 | { |
| 319 | assert(val != 0); |
| 320 | { |
| 321 | # if defined(_MSC_VER) /* Visual */ |
| 322 | unsigned long r=0; |
| 323 | _BitScanReverse(&r, val); |
| 324 | return (unsigned)r; |
| 325 | # elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ |
| 326 | return 31 - __builtin_clz(val); |
| 327 | # else /* Software version */ |
| 328 | static const U32 DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30, 8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31 }; |
| 329 | U32 v = val; |
| 330 | v |= v >> 1; |
| 331 | v |= v >> 2; |
| 332 | v |= v >> 4; |
| 333 | v |= v >> 8; |
| 334 | v |= v >> 16; |
| 335 | return DeBruijnClz[(v * 0x07C4ACDDU) >> 27]; |
| 336 | # endif |
| 337 | } |
| 338 | } |
| 339 | |
| 340 | |
| 341 | /* ZSTD_invalidateRepCodes() : |
| 342 | * ensures next compression will not use repcodes from previous block. |
| 343 | * Note : only works with regular variant; |
| 344 | * do not use with extDict variant ! */ |
| 345 | void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx); /* zstdmt, adaptive_compression (shouldn't get this definition from here) */ |
| 346 | |
| 347 | |
| 348 | typedef struct { |
| 349 | blockType_e blockType; |
| 350 | U32 lastBlock; |
| 351 | U32 origSize; |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 352 | } blockProperties_t; /* declared here for decompress and fullbench */ |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 353 | |
| 354 | /*! ZSTD_getcBlockSize() : |
| 355 | * Provides the size of compressed block from block header `src` */ |
| 356 | /* Used by: decompress, fullbench (does not get its definition from here) */ |
| 357 | size_t ZSTD_getcBlockSize(const void* src, size_t srcSize, |
| 358 | blockProperties_t* bpPtr); |
| 359 | |
Scott Baker | 8487c5d | 2019-10-18 12:49:46 -0700 | [diff] [blame] | 360 | /*! ZSTD_decodeSeqHeaders() : |
| 361 | * decode sequence header from src */ |
| 362 | /* Used by: decompress, fullbench (does not get its definition from here) */ |
| 363 | size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, |
| 364 | const void* src, size_t srcSize); |
| 365 | |
| 366 | |
Scott Baker | 2d89798 | 2019-09-24 11:50:08 -0700 | [diff] [blame] | 367 | #if defined (__cplusplus) |
| 368 | } |
| 369 | #endif |
| 370 | |
| 371 | #endif /* ZSTD_CCOMMON_H_MODULE */ |