Only load extra table positions for CDicts

Zstdmt uses prefixes to load the overlap between segments. Loading extra
positions makes compression non-deterministic, depending on the previous
job the context was used for. Since loading extra position takes extra
time as well, only do it when creating a `ZSTD_CDict`.

Fixes #1077.
This commit is contained in:
Nick Terrell
2018-04-02 14:41:30 -07:00
parent daf3ab0aad
commit 295ab0dbfa
8 changed files with 44 additions and 24 deletions
+27 -16
View File
@@ -2190,7 +2190,10 @@ size_t ZSTD_compressBlock(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const
/*! ZSTD_loadDictionaryContent() :
* @return : 0, or an error code
*/
static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, const void* src, size_t srcSize)
static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms,
ZSTD_CCtx_params const* params,
const void* src, size_t srcSize,
ZSTD_dictTableLoadMethod_e dtlm)
{
const BYTE* const ip = (const BYTE*) src;
const BYTE* const iend = ip + srcSize;
@@ -2204,10 +2207,10 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, ZSTD_CCtx_params
switch(params->cParams.strategy)
{
case ZSTD_fast:
ZSTD_fillHashTable(ms, cParams, iend);
ZSTD_fillHashTable(ms, cParams, iend, dtlm);
break;
case ZSTD_dfast:
ZSTD_fillDoubleHashTable(ms, cParams, iend);
ZSTD_fillDoubleHashTable(ms, cParams, iend, dtlm);
break;
case ZSTD_greedy:
@@ -2256,7 +2259,12 @@ static size_t ZSTD_checkDictNCount(short* normalizedCounter, unsigned dictMaxSym
* assumptions : magic number supposed already checked
* dictSize supposed > 8
*/
static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs, ZSTD_matchState_t* ms, ZSTD_CCtx_params const* params, const void* dict, size_t dictSize, void* workspace)
static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs,
ZSTD_matchState_t* ms,
ZSTD_CCtx_params const* params,
const void* dict, size_t dictSize,
ZSTD_dictTableLoadMethod_e dtlm,
void* workspace)
{
const BYTE* dictPtr = (const BYTE*)dict;
const BYTE* const dictEnd = dictPtr + dictSize;
@@ -2336,7 +2344,7 @@ static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs, ZSTD_matc
bs->entropy.offcode_repeatMode = FSE_repeat_valid;
bs->entropy.matchlength_repeatMode = FSE_repeat_valid;
bs->entropy.litlength_repeatMode = FSE_repeat_valid;
CHECK_F(ZSTD_loadDictionaryContent(ms, params, dictPtr, dictContentSize));
CHECK_F(ZSTD_loadDictionaryContent(ms, params, dictPtr, dictContentSize, dtlm));
return dictID;
}
}
@@ -2347,6 +2355,7 @@ static size_t ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, ZS
ZSTD_CCtx_params const* params,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
ZSTD_dictTableLoadMethod_e dtlm,
void* workspace)
{
DEBUGLOG(4, "ZSTD_compress_insertDictionary (dictSize=%u)", (U32)dictSize);
@@ -2356,12 +2365,12 @@ static size_t ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, ZS
/* dict restricted modes */
if (dictContentType == ZSTD_dct_rawContent)
return ZSTD_loadDictionaryContent(ms, params, dict, dictSize);
return ZSTD_loadDictionaryContent(ms, params, dict, dictSize, dtlm);
if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) {
if (dictContentType == ZSTD_dct_auto) {
DEBUGLOG(4, "raw content dictionary detected");
return ZSTD_loadDictionaryContent(ms, params, dict, dictSize);
return ZSTD_loadDictionaryContent(ms, params, dict, dictSize, dtlm);
}
if (dictContentType == ZSTD_dct_fullDict)
return ERROR(dictionary_wrong);
@@ -2369,7 +2378,7 @@ static size_t ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, ZS
}
/* dict as full zstd dictionary */
return ZSTD_loadZstdDictionary(bs, ms, params, dict, dictSize, workspace);
return ZSTD_loadZstdDictionary(bs, ms, params, dict, dictSize, dtlm, workspace);
}
/*! ZSTD_compressBegin_internal() :
@@ -2377,6 +2386,7 @@ static size_t ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, ZS
size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
ZSTD_dictTableLoadMethod_e dtlm,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params, U64 pledgedSrcSize,
ZSTD_buffered_policy_e zbuff)
@@ -2397,7 +2407,7 @@ size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
{
size_t const dictID = ZSTD_compress_insertDictionary(
cctx->blockState.prevCBlock, &cctx->blockState.matchState,
&params, dict, dictSize, dictContentType, cctx->entropyWorkspace);
&params, dict, dictSize, dictContentType, dtlm, cctx->entropyWorkspace);
if (ZSTD_isError(dictID)) return dictID;
assert(dictID <= (size_t)(U32)-1);
cctx->dictID = (U32)dictID;
@@ -2408,6 +2418,7 @@ size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx,
size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
ZSTD_dictTableLoadMethod_e dtlm,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params,
unsigned long long pledgedSrcSize)
@@ -2416,7 +2427,7 @@ size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
/* compression parameters verification and optimization */
CHECK_F( ZSTD_checkCParams(params.cParams) );
return ZSTD_compressBegin_internal(cctx,
dict, dictSize, dictContentType,
dict, dictSize, dictContentType, dtlm,
cdict,
params, pledgedSrcSize,
ZSTDb_not_buffered);
@@ -2431,7 +2442,7 @@ size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx,
ZSTD_CCtx_params const cctxParams =
ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
return ZSTD_compressBegin_advanced_internal(cctx,
dict, dictSize, ZSTD_dct_auto,
dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast,
NULL /*cdict*/,
cctxParams, pledgedSrcSize);
}
@@ -2442,7 +2453,7 @@ size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t di
ZSTD_CCtx_params const cctxParams =
ZSTD_assignParamsToCCtxParams(cctx->requestedParams, params);
DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (U32)dictSize);
return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, NULL,
return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL,
cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, ZSTDb_not_buffered);
}
@@ -2553,7 +2564,7 @@ size_t ZSTD_compress_advanced_internal(
{
DEBUGLOG(4, "ZSTD_compress_advanced_internal (srcSize:%u)",
(U32)srcSize);
CHECK_F( ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, NULL,
CHECK_F( ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL,
params, srcSize, ZSTDb_not_buffered) );
return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize);
}
@@ -2654,7 +2665,7 @@ static size_t ZSTD_initCDict_internal(
{ size_t const dictID = ZSTD_compress_insertDictionary(
&cdict->cBlockState, &cdict->matchState, &params,
cdict->dictContent, cdict->dictContentSize,
dictContentType, cdict->workspace);
dictContentType, ZSTD_dtlm_full, cdict->workspace);
if (ZSTD_isError(dictID)) return dictID;
assert(dictID <= (size_t)(U32)-1);
cdict->dictID = (U32)dictID;
@@ -2799,7 +2810,7 @@ size_t ZSTD_compressBegin_usingCDict_advanced(
}
params.fParams = fParams;
return ZSTD_compressBegin_internal(cctx,
NULL, 0, ZSTD_dct_auto,
NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast,
cdict,
params, pledgedSrcSize,
ZSTDb_not_buffered);
@@ -2889,7 +2900,7 @@ static size_t ZSTD_resetCStream_internal(ZSTD_CStream* cctx,
assert(!((dict) && (cdict))); /* either dict or cdict, not both */
CHECK_F( ZSTD_compressBegin_internal(cctx,
dict, dictSize, dictContentType,
dict, dictSize, dictContentType, ZSTD_dtlm_fast,
cdict,
params, pledgedSrcSize,
ZSTDb_buffered) );
+3 -1
View File
@@ -235,6 +235,7 @@ struct ZSTD_CCtx_s {
#endif
};
typedef enum { ZSTD_dtlm_fast, ZSTD_dtlm_full } ZSTD_dictTableLoadMethod_e;
typedef size_t (*ZSTD_blockCompressor) (
ZSTD_matchState_t* bs, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
@@ -640,7 +641,7 @@ MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window,
* ============================================================== */
/* ZSTD_getCParamsFromCCtxParams() :
* cParams are built depending on compressionLevel, src size hints,
* cParams are built depending on compressionLevel, src size hints,
* LDM and manually set compression parameters.
*/
ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
@@ -672,6 +673,7 @@ ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict);
size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx,
const void* dict, size_t dictSize,
ZSTD_dictContentType_e dictContentType,
ZSTD_dictTableLoadMethod_e dtlm,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params,
unsigned long long pledgedSrcSize);
+4 -1
View File
@@ -14,7 +14,7 @@
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
ZSTD_compressionParameters const* cParams,
void const* end)
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
U32* const hashLarge = ms->hashTable;
U32 const hBitsL = cParams->hashLog;
@@ -40,6 +40,9 @@ void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
hashSmall[smHash] = current + i;
if (i == 0 || hashLarge[lgHash] == 0)
hashLarge[lgHash] = current + i;
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
}
}
}
+1 -1
View File
@@ -20,7 +20,7 @@ extern "C" {
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
ZSTD_compressionParameters const* cParams,
void const* end);
void const* end, ZSTD_dictTableLoadMethod_e dtlm);
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+4 -1
View File
@@ -14,7 +14,7 @@
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
ZSTD_compressionParameters const* cParams,
void const* end)
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
U32* const hashTable = ms->hashTable;
U32 const hBits = cParams->hashLog;
@@ -34,6 +34,9 @@ void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
size_t const hash = ZSTD_hashPtr(ip + i, hBits, mls);
if (i == 0 || hashTable[hash] == 0)
hashTable[hash] = current + i;
/* Only load extra positions for ZSTD_dtlm_full */
if (dtlm == ZSTD_dtlm_fast)
break;
}
}
}
+1 -1
View File
@@ -20,7 +20,7 @@ extern "C" {
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
ZSTD_compressionParameters const* cParams,
void const* end);
void const* end, ZSTD_dictTableLoadMethod_e dtlm);
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_compressionParameters const* cParams, void const* src, size_t srcSize);
+2 -2
View File
@@ -224,12 +224,12 @@ static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
switch(cParams->strategy)
{
case ZSTD_fast:
ZSTD_fillHashTable(ms, cParams, iend);
ZSTD_fillHashTable(ms, cParams, iend, ZSTD_dtlm_fast);
ms->nextToUpdate = (U32)(iend - ms->window.base);
break;
case ZSTD_dfast:
ZSTD_fillDoubleHashTable(ms, cParams, iend);
ZSTD_fillDoubleHashTable(ms, cParams, iend, ZSTD_dtlm_fast);
ms->nextToUpdate = (U32)(iend - ms->window.base);
break;
+2 -1
View File
@@ -625,7 +625,7 @@ void ZSTDMT_compressionJob(void* jobDescription)
/* init */
if (job->cdict) {
size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, job->cdict, jobParams, job->fullFrameSize);
size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, job->cdict, jobParams, job->fullFrameSize);
assert(job->firstJob); /* only allowed for first job */
if (ZSTD_isError(initError)) { job->cSize = initError; goto _endJob; }
} else { /* srcStart points at reloaded section */
@@ -637,6 +637,7 @@ void ZSTDMT_compressionJob(void* jobDescription)
} }
{ size_t const initError = ZSTD_compressBegin_advanced_internal(cctx,
job->prefix.start, job->prefix.size, ZSTD_dct_rawContent, /* load dictionary in "content-only" mode (no header analysis) */
ZSTD_dtlm_fast,
NULL, /*cdict*/
jobParams, pledgedSrcSize);
if (ZSTD_isError(initError)) {