Sereal-Encoder
view release on metacpan or search on metacpan
zstd/compress/zstd_compress.c view on Meta::CPAN
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
* LICENSE file in the root directory of this source tree) and the GPLv2 (found
* in the COPYING file in the root directory of this source tree).
* You may select, at your option, one of the above-listed licenses.
*/
/*-*************************************
* Dependencies
***************************************/
#include "../common/zstd_deps.h" /* INT_MAX, ZSTD_memset, ZSTD_memcpy */
#include "../common/mem.h"
#include "hist.h" /* HIST_countFast_wksp */
#define FSE_STATIC_LINKING_ONLY /* FSE_encodeSymbol */
#include "../common/fse.h"
#define HUF_STATIC_LINKING_ONLY
#include "../common/huf.h"
#include "zstd_compress_internal.h"
#include "zstd_compress_sequences.h"
#include "zstd_compress_literals.h"
#include "zstd_fast.h"
#include "zstd_double_fast.h"
#include "zstd_lazy.h"
#include "zstd_opt.h"
#include "zstd_ldm.h"
#include "zstd_compress_superblock.h"
/* ***************************************************************
* Tuning parameters
*****************************************************************/
/*!
* COMPRESS_HEAPMODE :
* Select how default decompression function ZSTD_compress() allocates its context,
* on stack (0, default), or into heap (1).
* Note that functions with explicit context such as ZSTD_compressCCtx() are unaffected.
*/
#ifndef ZSTD_COMPRESS_HEAPMODE
# define ZSTD_COMPRESS_HEAPMODE 0
#endif
/*!
* ZSTD_HASHLOG3_MAX :
* Maximum size of the hash table dedicated to find 3-bytes matches,
* in log format, aka 17 => 1 << 17 == 128Ki positions.
* This structure is only used in zstd_opt.
* Since allocation is centralized for all strategies, it has to be known here.
* The actual (selected) size of the hash table is then stored in ZSTD_matchState_t.hashLog3,
* so that zstd_opt.c doesn't need to know about this constant.
*/
#ifndef ZSTD_HASHLOG3_MAX
# define ZSTD_HASHLOG3_MAX 17
#endif
/*-*************************************
* Helper functions
***************************************/
/* ZSTD_compressBound()
* Note that the result from this function is only compatible with the "normal"
* full-block strategy.
* When there are a lot of small blocks due to frequent flush in streaming mode
* the overhead of headers can make the compressed data to be larger than the
* return value of ZSTD_compressBound().
*/
size_t ZSTD_compressBound(size_t srcSize) {
return ZSTD_COMPRESSBOUND(srcSize);
}
/*-*************************************
* Context memory management
***************************************/
struct ZSTD_CDict_s {
const void* dictContent;
size_t dictContentSize;
ZSTD_dictContentType_e dictContentType; /* The dictContentType the CDict was created with */
U32* entropyWorkspace; /* entropy workspace of HUF_WORKSPACE_SIZE bytes */
ZSTD_cwksp workspace;
ZSTD_matchState_t matchState;
ZSTD_compressedBlockState_t cBlockState;
ZSTD_customMem customMem;
U32 dictID;
int compressionLevel; /* 0 indicates that advanced API was used to select CDict params */
ZSTD_paramSwitch_e useRowMatchFinder; /* Indicates whether the CDict was created with params that would use
* row-based matchfinder. Unless the cdict is reloaded, we will use
* the same greedy/lazy matchfinder at compression time.
*/
}; /* typedef'd to ZSTD_CDict within "zstd.h" */
ZSTD_CCtx* ZSTD_createCCtx(void)
{
return ZSTD_createCCtx_advanced(ZSTD_defaultCMem);
}
static void ZSTD_initCCtx(ZSTD_CCtx* cctx, ZSTD_customMem memManager)
{
assert(cctx != NULL);
ZSTD_memset(cctx, 0, sizeof(*cctx));
cctx->customMem = memManager;
cctx->bmi2 = ZSTD_cpuSupportsBmi2();
{ size_t const err = ZSTD_CCtx_reset(cctx, ZSTD_reset_parameters);
assert(!ZSTD_isError(err));
(void)err;
}
}
ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem)
{
ZSTD_STATIC_ASSERT(zcss_init==0);
ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN==(0ULL - 1));
if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL;
{ ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_customMalloc(sizeof(ZSTD_CCtx), customMem);
if (!cctx) return NULL;
ZSTD_initCCtx(cctx, customMem);
return cctx;
}
}
ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize)
{
ZSTD_cwksp ws;
zstd/compress/zstd_compress.c view on Meta::CPAN
return ZSTD_estimateCCtxSize_usingCCtxParams_internal(
&cParams, ¶ms->ldmParams, 1, useRowMatchFinder, inBuffSize, outBuffSize,
ZSTD_CONTENTSIZE_UNKNOWN);
}
}
size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams)
{
ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams);
if (ZSTD_rowMatchFinderSupported(cParams.strategy)) {
/* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */
size_t noRowCCtxSize;
size_t rowCCtxSize;
initialParams.useRowMatchFinder = ZSTD_ps_disable;
noRowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams);
initialParams.useRowMatchFinder = ZSTD_ps_enable;
rowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams);
return MAX(noRowCCtxSize, rowCCtxSize);
} else {
return ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams);
}
}
static size_t ZSTD_estimateCStreamSize_internal(int compressionLevel)
{
ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict);
return ZSTD_estimateCStreamSize_usingCParams(cParams);
}
size_t ZSTD_estimateCStreamSize(int compressionLevel)
{
int level;
size_t memBudget = 0;
for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) {
size_t const newMB = ZSTD_estimateCStreamSize_internal(level);
if (newMB > memBudget) memBudget = newMB;
}
return memBudget;
}
/* ZSTD_getFrameProgression():
* tells how much data has been consumed (input) and produced (output) for current frame.
* able to count progression inside worker threads (non-blocking mode).
*/
ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx)
{
#ifdef ZSTD_MULTITHREAD
if (cctx->appliedParams.nbWorkers > 0) {
return ZSTDMT_getFrameProgression(cctx->mtctx);
}
#endif
{ ZSTD_frameProgression fp;
size_t const buffered = (cctx->inBuff == NULL) ? 0 :
cctx->inBuffPos - cctx->inToCompress;
if (buffered) assert(cctx->inBuffPos >= cctx->inToCompress);
assert(buffered <= ZSTD_BLOCKSIZE_MAX);
fp.ingested = cctx->consumedSrcSize + buffered;
fp.consumed = cctx->consumedSrcSize;
fp.produced = cctx->producedCSize;
fp.flushed = cctx->producedCSize; /* simplified; some data might still be left within streaming output buffer */
fp.currentJobID = 0;
fp.nbActiveWorkers = 0;
return fp;
} }
/*! ZSTD_toFlushNow()
* Only useful for multithreading scenarios currently (nbWorkers >= 1).
*/
size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx)
{
#ifdef ZSTD_MULTITHREAD
if (cctx->appliedParams.nbWorkers > 0) {
return ZSTDMT_toFlushNow(cctx->mtctx);
}
#endif
(void)cctx;
return 0; /* over-simplification; could also check if context is currently running in streaming mode, and in which case, report how many bytes are left to be flushed within output buffer */
}
static void ZSTD_assertEqualCParams(ZSTD_compressionParameters cParams1,
ZSTD_compressionParameters cParams2)
{
(void)cParams1;
(void)cParams2;
assert(cParams1.windowLog == cParams2.windowLog);
assert(cParams1.chainLog == cParams2.chainLog);
assert(cParams1.hashLog == cParams2.hashLog);
assert(cParams1.searchLog == cParams2.searchLog);
assert(cParams1.minMatch == cParams2.minMatch);
assert(cParams1.targetLength == cParams2.targetLength);
assert(cParams1.strategy == cParams2.strategy);
}
void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs)
{
int i;
for (i = 0; i < ZSTD_REP_NUM; ++i)
bs->rep[i] = repStartValue[i];
bs->entropy.huf.repeatMode = HUF_repeat_none;
bs->entropy.fse.offcode_repeatMode = FSE_repeat_none;
bs->entropy.fse.matchlength_repeatMode = FSE_repeat_none;
bs->entropy.fse.litlength_repeatMode = FSE_repeat_none;
}
/*! ZSTD_invalidateMatchState()
* Invalidate all the matches in the match finder tables.
* Requires nextSrc and base to be set (can be NULL).
*/
static void ZSTD_invalidateMatchState(ZSTD_matchState_t* ms)
{
ZSTD_window_clear(&ms->window);
ms->nextToUpdate = ms->window.dictLimit;
ms->loadedDictEnd = 0;
ms->opt.litLengthSum = 0; /* force reset of btopt stats */
ms->dictMatchState = NULL;
}
/**
* Controls, for this matchState reset, whether the tables need to be cleared /
* prepared for the coming compression (ZSTDcrp_makeClean), or whether the
* tables can be left unclean (ZSTDcrp_leaveDirty), because we know that a
* subsequent operation will overwrite the table space anyways (e.g., copying
* the matchState contents in from a CDict).
*/
typedef enum {
ZSTDcrp_makeClean,
ZSTDcrp_leaveDirty
} ZSTD_compResetPolicy_e;
/**
* Controls, for this matchState reset, whether indexing can continue where it
* left off (ZSTDirp_continue), or whether it needs to be restarted from zero
* (ZSTDirp_reset).
*/
typedef enum {
ZSTDirp_continue,
zstd/compress/zstd_compress.c view on Meta::CPAN
ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_noAttachDict);
ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel);
}
DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (unsigned)dictSize);
return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL,
&cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, ZSTDb_not_buffered);
}
size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel)
{
return ZSTD_compressBegin_usingDict(cctx, NULL, 0, compressionLevel);
}
/*! ZSTD_writeEpilogue() :
* Ends a frame.
* @return : nb of bytes written into dst (or an error code) */
static size_t ZSTD_writeEpilogue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity)
{
BYTE* const ostart = (BYTE*)dst;
BYTE* op = ostart;
size_t fhSize = 0;
DEBUGLOG(4, "ZSTD_writeEpilogue");
RETURN_ERROR_IF(cctx->stage == ZSTDcs_created, stage_wrong, "init missing");
/* special case : empty frame */
if (cctx->stage == ZSTDcs_init) {
fhSize = ZSTD_writeFrameHeader(dst, dstCapacity, &cctx->appliedParams, 0, 0);
FORWARD_IF_ERROR(fhSize, "ZSTD_writeFrameHeader failed");
dstCapacity -= fhSize;
op += fhSize;
cctx->stage = ZSTDcs_ongoing;
}
if (cctx->stage != ZSTDcs_ending) {
/* write one last empty block, make it the "last" block */
U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1) + 0;
RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for epilogue");
MEM_writeLE32(op, cBlockHeader24);
op += ZSTD_blockHeaderSize;
dstCapacity -= ZSTD_blockHeaderSize;
}
if (cctx->appliedParams.fParams.checksumFlag) {
U32 const checksum = (U32) XXH64_digest(&cctx->xxhState);
RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for checksum");
DEBUGLOG(4, "ZSTD_writeEpilogue: write checksum : %08X", (unsigned)checksum);
MEM_writeLE32(op, checksum);
op += 4;
}
cctx->stage = ZSTDcs_created; /* return to "created but no init" status */
return op-ostart;
}
void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize)
{
#if ZSTD_TRACE
if (cctx->traceCtx && ZSTD_trace_compress_end != NULL) {
int const streaming = cctx->inBuffSize > 0 || cctx->outBuffSize > 0 || cctx->appliedParams.nbWorkers > 0;
ZSTD_Trace trace;
ZSTD_memset(&trace, 0, sizeof(trace));
trace.version = ZSTD_VERSION_NUMBER;
trace.streaming = streaming;
trace.dictionaryID = cctx->dictID;
trace.dictionarySize = cctx->dictContentSize;
trace.uncompressedSize = cctx->consumedSrcSize;
trace.compressedSize = cctx->producedCSize + extraCSize;
trace.params = &cctx->appliedParams;
trace.cctx = cctx;
ZSTD_trace_compress_end(cctx->traceCtx, &trace);
}
cctx->traceCtx = 0;
#else
(void)cctx;
(void)extraCSize;
#endif
}
size_t ZSTD_compressEnd (ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize)
{
size_t endResult;
size_t const cSize = ZSTD_compressContinue_internal(cctx,
dst, dstCapacity, src, srcSize,
1 /* frame mode */, 1 /* last chunk */);
FORWARD_IF_ERROR(cSize, "ZSTD_compressContinue_internal failed");
endResult = ZSTD_writeEpilogue(cctx, (char*)dst + cSize, dstCapacity-cSize);
FORWARD_IF_ERROR(endResult, "ZSTD_writeEpilogue failed");
assert(!(cctx->appliedParams.fParams.contentSizeFlag && cctx->pledgedSrcSizePlusOne == 0));
if (cctx->pledgedSrcSizePlusOne != 0) { /* control src size */
ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN == (unsigned long long)-1);
DEBUGLOG(4, "end of frame : controlling src size");
RETURN_ERROR_IF(
cctx->pledgedSrcSizePlusOne != cctx->consumedSrcSize+1,
srcSize_wrong,
"error : pledgedSrcSize = %u, while realSrcSize = %u",
(unsigned)cctx->pledgedSrcSizePlusOne-1,
(unsigned)cctx->consumedSrcSize);
}
ZSTD_CCtx_trace(cctx, endResult);
return cSize + endResult;
}
size_t ZSTD_compress_advanced (ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const void* dict,size_t dictSize,
ZSTD_parameters params)
{
DEBUGLOG(4, "ZSTD_compress_advanced");
FORWARD_IF_ERROR(ZSTD_checkCParams(params.cParams), "");
ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, ZSTD_NO_CLEVEL);
return ZSTD_compress_advanced_internal(cctx,
dst, dstCapacity,
src, srcSize,
dict, dictSize,
&cctx->simpleApiParams);
}
/* Internal */
size_t ZSTD_compress_advanced_internal(
ZSTD_CCtx* cctx,
( run in 0.981 second using v1.01-cache-2.11-cpan-39bf76dae61 )