ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[libzstd] Refactor ZSTD_compressSequences()

Browse Source
This commit is contained in:
Nick Terrell
2017-07-17 11:36:11 -07:00
parent 830ef4152a
commit 634f012420
1 changed files with 166 additions and 144 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/compress/zstd_compress.c
+147 -125
View File
@@ -1017,143 +1017,78 @@ void ZSTD_seqToCodes(const seqStore_t* seqStorePtr)
mlCodeTable[seqStorePtr->longLengthPos] = MaxML; mlCodeTable[seqStorePtr->longLengthPos] = MaxML;
} }
MEM_STATIC size_t ZSTD_compressSequences (ZSTD_CCtx* zc, MEM_STATIC symbolEncodingType_e ZSTD_selectEncodingType(FSE_repeat* repeatMode,
void* dst, size_t dstCapacity, size_t const mostFrequent, size_t nbSeq, U32 defaultNormLog)
size_t srcSize)
{ {
const int longOffsets = zc->appliedParams.cParams.windowLog > STREAM_ACCUMULATOR_MIN;
const seqStore_t* seqStorePtr = &(zc->seqStore);
U32 count[MaxSeq+1];
S16 norm[MaxSeq+1];
FSE_CTable* CTable_LitLength = zc->entropy->litlengthCTable;
FSE_CTable* CTable_OffsetBits = zc->entropy->offcodeCTable;
FSE_CTable* CTable_MatchLength = zc->entropy->matchlengthCTable;
U32 LLtype, Offtype, MLtype; /* compressed, raw or rle */
const seqDef* const sequences = seqStorePtr->sequencesStart;
const BYTE* const ofCodeTable = seqStorePtr->ofCode;
const BYTE* const llCodeTable = seqStorePtr->llCode;
const BYTE* const mlCodeTable = seqStorePtr->mlCode;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart;
BYTE* seqHead;
BYTE scratchBuffer[1<<MAX(MLFSELog,LLFSELog)];
/* Compress literals */
{ const BYTE* const literals = seqStorePtr->litStart;
size_t const litSize = seqStorePtr->lit - literals;
size_t const cSize = ZSTD_compressLiterals(zc, op, dstCapacity, literals, litSize);
if (ZSTD_isError(cSize)) return cSize;
op += cSize;
}
/* Sequences Header */
if ((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead */) return ERROR(dstSize_tooSmall);
if (nbSeq < 0x7F) *op++ = (BYTE)nbSeq;
else if (nbSeq < LONGNBSEQ) op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
else op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
if (nbSeq==0) goto _check_compressibility;
/* seqHead : flags for FSE encoding type */
seqHead = op++;
#define MIN_SEQ_FOR_DYNAMIC_FSE 64 #define MIN_SEQ_FOR_DYNAMIC_FSE 64
#define MAX_SEQ_FOR_STATIC_FSE 1000 #define MAX_SEQ_FOR_STATIC_FSE 1000
/* convert length/distances into codes */
ZSTD_seqToCodes(seqStorePtr);
/* CTable for Literal Lengths */
{ U32 max = MaxLL;
size_t const mostFrequent = FSE_countFast_wksp(count, &max, llCodeTable, nbSeq, zc->entropy->workspace);
if ((mostFrequent == nbSeq) && (nbSeq > 2)) { if ((mostFrequent == nbSeq) && (nbSeq > 2)) {
*op++ = llCodeTable[0]; *repeatMode = FSE_repeat_check;
FSE_buildCTable_rle(CTable_LitLength, (BYTE)max); return set_rle;
LLtype = set_rle; }
zc->entropy->litlength_repeatMode = FSE_repeat_check; if ((*repeatMode == FSE_repeat_valid) && (nbSeq < MAX_SEQ_FOR_STATIC_FSE)) {
} else if ((zc->entropy->litlength_repeatMode == FSE_repeat_valid) && (nbSeq < MAX_SEQ_FOR_STATIC_FSE)) { return set_repeat;
LLtype = set_repeat; }
} else if ((nbSeq < MIN_SEQ_FOR_DYNAMIC_FSE) || (mostFrequent < (nbSeq >> (LL_defaultNormLog-1)))) { if ((nbSeq < MIN_SEQ_FOR_DYNAMIC_FSE) || (mostFrequent < (nbSeq >> (defaultNormLog-1)))) {
FSE_buildCTable_wksp(CTable_LitLength, LL_defaultNorm, MaxLL, LL_defaultNormLog, scratchBuffer, sizeof(scratchBuffer)); *repeatMode = FSE_repeat_valid;
LLtype = set_basic; return set_basic;
zc->entropy->litlength_repeatMode = FSE_repeat_valid; }
} else { *repeatMode = FSE_repeat_check;
return set_compressed;
}
MEM_STATIC size_t ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* CTable, U32 FSELog, symbolEncodingType_e type,
U32* count, U32 max,
BYTE const* codeTable, size_t nbSeq,
S16 const* defaultNorm, U32 defaultNormLog, U32 defaultMax,
void* workspace, size_t workspaceSize)
{
BYTE* op = (BYTE*)dst;
BYTE const* const oend = op + dstCapacity;
switch (type) {
case set_rle:
*op = codeTable[0];
CHECK_F(FSE_buildCTable_rle(CTable, (BYTE)max));
return 1;
case set_repeat:
return 0;
case set_basic:
CHECK_F(FSE_buildCTable_wksp(CTable, defaultNorm, defaultMax, defaultNormLog, workspace, workspaceSize));
return 0;
case set_compressed: {
S16 norm[MaxSeq + 1];
size_t nbSeq_1 = nbSeq; size_t nbSeq_1 = nbSeq;
const U32 tableLog = FSE_optimalTableLog(LLFSELog, nbSeq, max); const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max);
if (count[llCodeTable[nbSeq-1]]>1) { count[llCodeTable[nbSeq-1]]--; nbSeq_1--; } if (count[codeTable[nbSeq-1]] > 1) {
FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max); count[codeTable[nbSeq-1]]--;
nbSeq_1--;
}
CHECK_F(FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max));
{ size_t const NCountSize = FSE_writeNCount(op, oend - op, norm, max, tableLog); /* overflow protected */ { size_t const NCountSize = FSE_writeNCount(op, oend - op, norm, max, tableLog); /* overflow protected */
if (FSE_isError(NCountSize)) return NCountSize; if (FSE_isError(NCountSize)) return NCountSize;
op += NCountSize; } CHECK_F(FSE_buildCTable_wksp(CTable, norm, max, tableLog, workspace, workspaceSize));
FSE_buildCTable_wksp(CTable_LitLength, norm, max, tableLog, scratchBuffer, sizeof(scratchBuffer)); return NCountSize;
LLtype = set_compressed; }
zc->entropy->litlength_repeatMode = FSE_repeat_check; }
} } default: return assert(0), ERROR(GENERIC);
}
}
/* CTable for Offsets */ MEM_STATIC size_t ZSTD_encodeSequences(void* dst, size_t dstCapacity,
{ U32 max = MaxOff; FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
size_t const mostFrequent = FSE_countFast_wksp(count, &max, ofCodeTable, nbSeq, zc->entropy->workspace); FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
if ((mostFrequent == nbSeq) && (nbSeq > 2)) { FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
*op++ = ofCodeTable[0]; seqDef const* sequences, size_t nbSeq, int longOffsets)
FSE_buildCTable_rle(CTable_OffsetBits, (BYTE)max); {
Offtype = set_rle; BIT_CStream_t blockStream;
zc->entropy->offcode_repeatMode = FSE_repeat_check;
} else if ((zc->entropy->offcode_repeatMode == FSE_repeat_valid) && (nbSeq < MAX_SEQ_FOR_STATIC_FSE)) {
Offtype = set_repeat;
} else if ((nbSeq < MIN_SEQ_FOR_DYNAMIC_FSE) || (mostFrequent < (nbSeq >> (OF_defaultNormLog-1)))) {
FSE_buildCTable_wksp(CTable_OffsetBits, OF_defaultNorm, MaxOff, OF_defaultNormLog, scratchBuffer, sizeof(scratchBuffer));
Offtype = set_basic;
zc->entropy->offcode_repeatMode = FSE_repeat_valid;
} else {
size_t nbSeq_1 = nbSeq;
const U32 tableLog = FSE_optimalTableLog(OffFSELog, nbSeq, max);
if (count[ofCodeTable[nbSeq-1]]>1) { count[ofCodeTable[nbSeq-1]]--; nbSeq_1--; }
FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max);
{ size_t const NCountSize = FSE_writeNCount(op, oend-op, norm, max, tableLog); /* overflow protected */
if (FSE_isError(NCountSize)) return NCountSize;
op += NCountSize; }
FSE_buildCTable_wksp(CTable_OffsetBits, norm, max, tableLog, scratchBuffer, sizeof(scratchBuffer));
Offtype = set_compressed;
zc->entropy->offcode_repeatMode = FSE_repeat_check;
} }
/* CTable for MatchLengths */
{ U32 max = MaxML;
size_t const mostFrequent = FSE_countFast_wksp(count, &max, mlCodeTable, nbSeq, zc->entropy->workspace);
if ((mostFrequent == nbSeq) && (nbSeq > 2)) {
*op++ = *mlCodeTable;
FSE_buildCTable_rle(CTable_MatchLength, (BYTE)max);
MLtype = set_rle;
zc->entropy->matchlength_repeatMode = FSE_repeat_check;
} else if ((zc->entropy->matchlength_repeatMode == FSE_repeat_valid) && (nbSeq < MAX_SEQ_FOR_STATIC_FSE)) {
MLtype = set_repeat;
} else if ((nbSeq < MIN_SEQ_FOR_DYNAMIC_FSE) || (mostFrequent < (nbSeq >> (ML_defaultNormLog-1)))) {
FSE_buildCTable_wksp(CTable_MatchLength, ML_defaultNorm, MaxML, ML_defaultNormLog, scratchBuffer, sizeof(scratchBuffer));
MLtype = set_basic;
zc->entropy->matchlength_repeatMode = FSE_repeat_valid;
} else {
size_t nbSeq_1 = nbSeq;
const U32 tableLog = FSE_optimalTableLog(MLFSELog, nbSeq, max);
if (count[mlCodeTable[nbSeq-1]]>1) { count[mlCodeTable[nbSeq-1]]--; nbSeq_1--; }
FSE_normalizeCount(norm, tableLog, count, nbSeq_1, max);
{ size_t const NCountSize = FSE_writeNCount(op, oend-op, norm, max, tableLog); /* overflow protected */
if (FSE_isError(NCountSize)) return NCountSize;
op += NCountSize; }
FSE_buildCTable_wksp(CTable_MatchLength, norm, max, tableLog, scratchBuffer, sizeof(scratchBuffer));
MLtype = set_compressed;
zc->entropy->matchlength_repeatMode = FSE_repeat_check;
} }
*seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
/* Encoding Sequences */
{ BIT_CStream_t blockStream;
FSE_CState_t stateMatchLength; FSE_CState_t stateMatchLength;
FSE_CState_t stateOffsetBits; FSE_CState_t stateOffsetBits;
FSE_CState_t stateLitLength; FSE_CState_t stateLitLength;
CHECK_E(BIT_initCStream(&blockStream, op, oend-op), dstSize_tooSmall); /* not enough space remaining */ CHECK_E(BIT_initCStream(&blockStream, dst, dstCapacity), dstSize_tooSmall); /* not enough space remaining */
/* first symbols */ /* first symbols */
FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]); FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]);
@@ -1216,8 +1151,95 @@ MEM_STATIC size_t ZSTD_compressSequences (ZSTD_CCtx* zc,
{ size_t const streamSize = BIT_closeCStream(&blockStream); { size_t const streamSize = BIT_closeCStream(&blockStream);
if (streamSize==0) return ERROR(dstSize_tooSmall); /* not enough space */ if (streamSize==0) return ERROR(dstSize_tooSmall); /* not enough space */
op += streamSize; return streamSize;
}
}
MEM_STATIC size_t ZSTD_compressSequences (ZSTD_CCtx* zc,
void* dst, size_t dstCapacity,
size_t srcSize)
{
const int longOffsets = zc->appliedParams.cParams.windowLog > STREAM_ACCUMULATOR_MIN;
const seqStore_t* seqStorePtr = &(zc->seqStore);
U32 count[MaxSeq+1];
FSE_CTable* CTable_LitLength = zc->entropy->litlengthCTable;
FSE_CTable* CTable_OffsetBits = zc->entropy->offcodeCTable;
FSE_CTable* CTable_MatchLength = zc->entropy->matchlengthCTable;
U32 LLtype, Offtype, MLtype; /* compressed, raw or rle */
const seqDef* const sequences = seqStorePtr->sequencesStart;
const BYTE* const ofCodeTable = seqStorePtr->ofCode;
const BYTE* const llCodeTable = seqStorePtr->llCode;
const BYTE* const mlCodeTable = seqStorePtr->mlCode;
BYTE* const ostart = (BYTE*)dst;
BYTE* const oend = ostart + dstCapacity;
BYTE* op = ostart;
size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart;
BYTE* seqHead;
ZSTD_STATIC_ASSERT(sizeof(zc->entropy->workspace) >= (1<<MAX(MLFSELog,LLFSELog)));
/* Compress literals */
{ const BYTE* const literals = seqStorePtr->litStart;
size_t const litSize = seqStorePtr->lit - literals;
size_t const cSize = ZSTD_compressLiterals(zc, op, dstCapacity, literals, litSize);
if (ZSTD_isError(cSize)) return cSize;
op += cSize;
}
/* Sequences Header */
if ((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead */) return ERROR(dstSize_tooSmall);
if (nbSeq < 0x7F) *op++ = (BYTE)nbSeq;
else if (nbSeq < LONGNBSEQ) op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2;
else op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3;
if (nbSeq==0) goto _check_compressibility;
/* seqHead : flags for FSE encoding type */
seqHead = op++;
/* convert length/distances into codes */
ZSTD_seqToCodes(seqStorePtr);
/* CTable for Literal Lengths */
{ U32 max = MaxLL;
size_t const mostFrequent = FSE_countFast_wksp(count, &max, llCodeTable, nbSeq, zc->entropy->workspace);
LLtype = ZSTD_selectEncodingType(&zc->entropy->litlength_repeatMode, mostFrequent, nbSeq, LL_defaultNormLog);
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_LitLength, LLFSELog, (symbolEncodingType_e)LLtype,
count, max, llCodeTable, nbSeq, LL_defaultNorm, LL_defaultNormLog, MaxLL,
zc->entropy->workspace, sizeof(zc->entropy->workspace));
if (ZSTD_isError(countSize)) return countSize;
op += countSize;
} } } }
/* CTable for Offsets */
{ U32 max = MaxOff;
size_t const mostFrequent = FSE_countFast_wksp(count, &max, ofCodeTable, nbSeq, zc->entropy->workspace);
Offtype = ZSTD_selectEncodingType(&zc->entropy->offcode_repeatMode, mostFrequent, nbSeq, OF_defaultNormLog);
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_OffsetBits, OffFSELog, (symbolEncodingType_e)Offtype,
count, max, ofCodeTable, nbSeq, OF_defaultNorm, OF_defaultNormLog, MaxOff,
zc->entropy->workspace, sizeof(zc->entropy->workspace));
if (ZSTD_isError(countSize)) return countSize;
op += countSize;
} }
/* CTable for MatchLengths */
{ U32 max = MaxML;
size_t const mostFrequent = FSE_countFast_wksp(count, &max, mlCodeTable, nbSeq, zc->entropy->workspace);
MLtype = ZSTD_selectEncodingType(&zc->entropy->matchlength_repeatMode, mostFrequent, nbSeq, ML_defaultNormLog);
{ size_t const countSize = ZSTD_buildCTable(op, oend - op, CTable_MatchLength, MLFSELog, (symbolEncodingType_e)MLtype,
count, max, mlCodeTable, nbSeq, ML_defaultNorm, ML_defaultNormLog, MaxML,
zc->entropy->workspace, sizeof(zc->entropy->workspace));
if (ZSTD_isError(countSize)) return countSize;
op += countSize;
} }
*seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2));
{ size_t const streamSize = ZSTD_encodeSequences(op, oend - op,
CTable_MatchLength, mlCodeTable,
CTable_OffsetBits, ofCodeTable,
CTable_LitLength, llCodeTable,
sequences, nbSeq, longOffsets);
if (ZSTD_isError(streamSize)) return streamSize;
op += streamSize;
}
/* check compressibility */ /* check compressibility */
_check_compressibility: _check_compressibility: