diff --git a/build/VS2008/zstdlib/zstdlib.vcproj b/build/VS2008/zstdlib/zstdlib.vcproj
index ac85f7aeb..ac8f896c3 100644
--- a/build/VS2008/zstdlib/zstdlib.vcproj
+++ b/build/VS2008/zstdlib/zstdlib.vcproj
@@ -388,6 +388,22 @@
RelativePath="..\..\..\lib\compress\zstd_compress.c"
>
+
+
+
+
+
+
+
+
@@ -495,11 +511,27 @@
>
+
+
+
+
+
+
+
+
+
+
+
+
@@ -180,6 +184,10 @@
+
+
+
+
diff --git a/build/VS2010/fuzzer/fuzzer.vcxproj b/build/VS2010/fuzzer/fuzzer.vcxproj
index 12a4b9313..9f00899da 100644
--- a/build/VS2010/fuzzer/fuzzer.vcxproj
+++ b/build/VS2010/fuzzer/fuzzer.vcxproj
@@ -165,6 +165,10 @@
+
+
+
+
@@ -183,6 +187,10 @@
+
+
+
+
diff --git a/build/VS2010/libzstd-dll/libzstd-dll.vcxproj b/build/VS2010/libzstd-dll/libzstd-dll.vcxproj
index 364b3bea5..0a4be69df 100644
--- a/build/VS2010/libzstd-dll/libzstd-dll.vcxproj
+++ b/build/VS2010/libzstd-dll/libzstd-dll.vcxproj
@@ -29,6 +29,10 @@
+
+
+
+
@@ -67,6 +71,10 @@
+
+
+
+
diff --git a/build/VS2010/libzstd/libzstd.vcxproj b/build/VS2010/libzstd/libzstd.vcxproj
index 6087d737c..51b840677 100644
--- a/build/VS2010/libzstd/libzstd.vcxproj
+++ b/build/VS2010/libzstd/libzstd.vcxproj
@@ -29,6 +29,10 @@
+
+
+
+
@@ -67,6 +71,10 @@
+
+
+
+
diff --git a/build/VS2010/zstd/zstd.vcxproj b/build/VS2010/zstd/zstd.vcxproj
index 438dc6173..90470180d 100644
--- a/build/VS2010/zstd/zstd.vcxproj
+++ b/build/VS2010/zstd/zstd.vcxproj
@@ -30,6 +30,10 @@
+
+
+
+
@@ -60,6 +64,10 @@
+
+
+
+
diff --git a/build/cmake/lib/CMakeLists.txt b/build/cmake/lib/CMakeLists.txt
index 8371192c5..f4b7e3753 100644
--- a/build/cmake/lib/CMakeLists.txt
+++ b/build/cmake/lib/CMakeLists.txt
@@ -38,6 +38,10 @@ SET(Sources
${LIBRARY_DIR}/compress/huf_compress.c
${LIBRARY_DIR}/compress/zstd_compress.c
${LIBRARY_DIR}/compress/zstdmt_compress.c
+ ${LIBRARY_DIR}/compress/zstd_fast.c
+ ${LIBRARY_DIR}/compress/zstd_double_fast.c
+ ${LIBRARY_DIR}/compress/zstd_lazy.c
+ ${LIBRARY_DIR}/compress/zstd_opt.c
${LIBRARY_DIR}/decompress/huf_decompress.c
${LIBRARY_DIR}/decompress/zstd_decompress.c
${LIBRARY_DIR}/dictBuilder/cover.c
@@ -58,6 +62,11 @@ SET(Headers
${LIBRARY_DIR}/common/huf.h
${LIBRARY_DIR}/common/mem.h
${LIBRARY_DIR}/common/zstd_internal.h
+ ${LIBRARY_DIR}/compress/zstd_compress.h
+ ${LIBRARY_DIR}/compress/zstd_fast.h
+ ${LIBRARY_DIR}/compress/zstd_double_fast.h
+ ${LIBRARY_DIR}/compress/zstd_lazy.h
+ ${LIBRARY_DIR}/compress/zstd_opt.h
${LIBRARY_DIR}/compress/zstdmt_compress.h
${LIBRARY_DIR}/dictBuilder/zdict.h
${LIBRARY_DIR}/deprecated/zbuff.h)
diff --git a/doc/zstd_manual.html b/doc/zstd_manual.html
index c166e7258..83b75fd86 100644
--- a/doc/zstd_manual.html
+++ b/doc/zstd_manual.html
@@ -27,7 +27,8 @@
Buffer-less and synchronous inner streaming functions
Buffer-less streaming compression (synchronous mode)
Buffer-less streaming decompression (synchronous mode)
-Block functions
+ZSTD_CCtx_params
+Block functions
Introduction
@@ -402,25 +403,29 @@ size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
size_t ZSTD_estimateCCtxSize(int compressionLevel);
-size_t ZSTD_estimateCCtxSize_advanced(ZSTD_compressionParameters cParams);
+size_t ZSTD_estimateCCtxSize_advanced_usingCParams(ZSTD_compressionParameters cParams);
+size_t ZSTD_estimateCCtxSize_advanced_usingCCtxParams(const ZSTD_CCtx_params* params);
size_t ZSTD_estimateDCtxSize(void);
These functions make it possible to estimate memory usage
of a future {D,C}Ctx, before its creation.
ZSTD_estimateCCtxSize() will provide a budget large enough for any compression level up to selected one.
It will also consider src size to be arbitrarily "large", which is worst case.
- If srcSize is known to always be small, ZSTD_estimateCCtxSize_advanced() can provide a tighter estimation.
- ZSTD_estimateCCtxSize_advanced() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
+ If srcSize is known to always be small, ZSTD_estimateCCtxSize_advanced_usingCParams() can provide a tighter estimation.
+ ZSTD_estimateCCtxSize_advanced_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
+ ZSTD_estimateCCtxSize_advanced_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_p_nbThreads is > 1.
Note : CCtx estimation is only correct for single-threaded compression
size_t ZSTD_estimateCStreamSize(int compressionLevel);
-size_t ZSTD_estimateCStreamSize_advanced(ZSTD_compressionParameters cParams);
+size_t ZSTD_estimateCStreamSize_advanced_usingCParams(ZSTD_compressionParameters cParams);
+size_t ZSTD_estimateCStreamSize_advanced_usingCCtxParams(const ZSTD_CCtx_params* params);
size_t ZSTD_estimateDStreamSize(size_t windowSize);
size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize);
ZSTD_estimateCStreamSize() will provide a budget large enough for any compression level up to selected one.
It will also consider src size to be arbitrarily "large", which is worst case.
- If srcSize is known to always be small, ZSTD_estimateCStreamSize_advanced() can provide a tighter estimation.
- ZSTD_estimateCStreamSize_advanced() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
+ If srcSize is known to always be small, ZSTD_estimateCStreamSize_advanced_usingCParams() can provide a tighter estimation.
+ ZSTD_estimateCStreamSize_advanced_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel.
+ ZSTD_estimateCStreamSize_advanced_usingCCtxParams() can be used in tandem with ZSTD_CCtxParam_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_p_nbThreads is set to a value > 1.
Note : CStream estimation is only correct for single-threaded compression.
ZSTD_DStream memory budget depends on window Size.
This information can be passed manually, using ZSTD_estimateDStreamSize,
@@ -430,12 +435,18 @@ size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize);
In this case, get total size by adding ZSTD_estimate?DictSize
+typedef enum {
+ ZSTD_dlm_byCopy = 0, /* Copy dictionary content internally. */
+ ZSTD_dlm_byRef, /* Reference dictionary content -- the dictionary buffer must outlives its users. */
+} ZSTD_dictLoadMethod_e;
+
size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel);
-size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, unsigned byReference);
-size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
+size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod);
+size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod);
ZSTD_estimateCDictSize() will bet that src size is relatively "small", and content is copied, like ZSTD_createCDict().
- ZSTD_estimateCStreamSize_advanced() makes it possible to control precisely compression parameters, like ZSTD_createCDict_advanced().
- Note : dictionary created "byReference" are smaller
+ ZSTD_estimateCStreamSize_advanced_usingCParams() makes it possible to control precisely compression parameters, like ZSTD_createCDict_advanced().
+ Note : dictionary created by reference using ZSTD_dlm_byRef are smaller
+
Advanced compression functions
@@ -461,16 +472,6 @@ size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
-typedef enum {
- ZSTD_p_forceWindow, /* Force back-references to remain < windowSize, even when referencing Dictionary content (default:0) */
- ZSTD_p_forceRawDict /* Force loading dictionary in "content-only" mode (no header analysis) */
-} ZSTD_CCtxParameter;
-
-size_t ZSTD_setCCtxParameter(ZSTD_CCtx* cctx, ZSTD_CCtxParameter param, unsigned value);
- Set advanced parameters, selected through enum ZSTD_CCtxParameter
- @result : 0, or an error code (which can be tested with ZSTD_isError())
-
-
ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel);
Create a digested dictionary for compression
Dictionary content is simply referenced, and therefore stays in dictBuffer.
@@ -483,7 +484,8 @@ size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
} ZSTD_dictMode_e;
ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_t dictSize,
- unsigned byReference, ZSTD_dictMode_e dictMode,
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_dictMode_e dictMode,
ZSTD_compressionParameters cParams,
ZSTD_customMem customMem);
Create a ZSTD_CDict using external alloc and free, and customized compression parameters
@@ -492,7 +494,7 @@ size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
ZSTD_CDict* ZSTD_initStaticCDict(
void* workspace, size_t workspaceSize,
const void* dict, size_t dictSize,
- unsigned byReference, ZSTD_dictMode_e dictMode,
+ ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictMode_e dictMode,
ZSTD_compressionParameters cParams);
Generate a digested dictionary in provided memory area.
workspace: The memory area to emplace the dictionary into.
@@ -580,13 +582,14 @@ size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize,
- unsigned byReference, ZSTD_customMem customMem);
+ ZSTD_dictLoadMethod_e dictLoadMethod,
+ ZSTD_customMem customMem);
Create a ZSTD_DDict using external alloc and free, optionally by reference
ZSTD_DDict* ZSTD_initStaticDDict(void* workspace, size_t workspaceSize,
const void* dict, size_t dictSize,
- unsigned byReference);
+ ZSTD_dictLoadMethod_e dictLoadMethod);
Generate a digested dictionary in provided memory area.
workspace: The memory area to emplace the dictionary into.
Provided pointer must 8-bytes aligned.
@@ -628,9 +631,9 @@ size_t ZSTD_estimateDDictSize(size_t dictSize, unsigned byReference);
Advanced Streaming compression functions
ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem);
ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize);
/**< same as ZSTD_initStaticCCtx() */
size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize); /**< pledgedSrcSize must be correct, a size of 0 means unknown. for a frame size of 0 use initCStream_advanced */
-size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); /**< creates of an internal CDict (incompatible with static CCtx), except if dict == NULL or dictSize < 8, in which case no dict is used. */
+size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); /**< creates of an internal CDict (incompatible with static CCtx), except if dict == NULL or dictSize < 8, in which case no dict is used. Note: dict is loaded with ZSTD_dm_auto (treated as a full zstd dictionary if it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy.*/
size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize,
- ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize is optional and can be 0 (meaning unknown). note: if the contentSizeFlag is set, pledgedSrcSize == 0 means the source size is actually 0 */
+ ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize is optional and can be 0 (meaning unknown). note: if the contentSizeFlag is set, pledgedSrcSize == 0 means the source size is actually 0. dict is loaded with ZSTD_dm_auto and ZSTD_dlm_byCopy. */
size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); /**< note : cdict will just be referenced, and must outlive compression session */
size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize); /**< same as ZSTD_initCStream_usingCDict(), with control over frame parameters */
@@ -819,12 +822,6 @@ void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx);
ZSTD_p_checksumFlag, /* A 32-bits checksum of content is written at end of frame (default:0) */
ZSTD_p_dictIDFlag, /* When applicable, dictID of dictionary is provided in frame header (default:1) */
- /* dictionary parameters (must be set before ZSTD_CCtx_loadDictionary) */
- ZSTD_p_dictMode=300, /* Select how dictionary content must be interpreted. Value must be from type ZSTD_dictMode_e.
- * default : 0==auto : dictionary will be "full" if it respects specification, otherwise it will be "rawContent" */
- ZSTD_p_refDictContent, /* Dictionary content will be referenced, instead of copied (default:0==byCopy).
- * It requires that dictionary buffer outlives its users */
-
/* multi-threading parameters */
ZSTD_p_nbThreads=400, /* Select how many threads a compression job can spawn (default:1)
* More threads improve speed, but also increase memory usage.
@@ -861,18 +858,25 @@ void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx);
size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
+size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
+size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictMode_e dictMode);
Create an internal CDict from dict buffer.
Decompression will have to use same buffer.
@result : 0, or an error code (which can be tested with ZSTD_isError()).
Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary,
meaning "return to no-dictionary mode".
- Note 1 : `dict` content will be copied internally,
- except if ZSTD_p_refDictContent is set before loading.
+ Note 1 : `dict` content will be copied internally. Use
+ ZSTD_CCtx_loadDictionary_byReference() to reference dictionary
+ content instead. The dictionary buffer must then outlive its
+ users.
Note 2 : Loading a dictionary involves building tables, which are dependent on compression parameters.
For this reason, compression parameters cannot be changed anymore after loading a dictionary.
It's also a CPU-heavy operation, with non-negligible impact on latency.
Note 3 : Dictionary will be used for all future compression jobs.
- To return to "no-dictionary" situation, load a NULL dictionary
+ To return to "no-dictionary" situation, load a NULL dictionary
+ Note 5 : Use ZSTD_CCtx_loadDictionary_advanced() to select how dictionary
+ content will be interpreted.
+
size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
@@ -889,6 +893,7 @@ void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx);
size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize);
+size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictMode_e dictMode);
Reference a prefix (single-usage dictionary) for next compression job.
Decompression need same prefix to properly regenerate data.
Prefix is **only used once**. Tables are discarded at end of compression job.
@@ -899,7 +904,9 @@ void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx);
Note 1 : Prefix buffer is referenced. It must outlive compression job.
Note 2 : Referencing a prefix involves building tables, which are dependent on compression parameters.
It's a CPU-heavy operation, with non-negligible impact on latency.
- Note 3 : it's possible to alter ZSTD_p_dictMode using ZSTD_CCtx_setParameter()
+ Note 3 : By default, the prefix is treated as raw content
+ (ZSTD_dm_rawContent). Use ZSTD_CCtx_refPrefix_advanced() to alter
+ dictMode.
typedef enum {
@@ -949,7 +956,57 @@ void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx);
-Block functions
+ZSTD_CCtx_params
+ - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure
+ - ZSTD_CCtxParam_setParameter() : Push parameters one by one into an
+ existing ZSTD_CCtx_params structure. This is similar to
+ ZSTD_CCtx_setParameter().
+ - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to an existing CCtx. These
+ parameters will be applied to all subsequent compression jobs.
+ - ZSTD_compress_generic() : Do compression using the CCtx.
+ - ZSTD_freeCCtxParams() : Free the memory.
+
+ This can be used with ZSTD_estimateCCtxSize_opaque() for static allocation
+ for single-threaded compression.
+
+
+
+size_t ZSTD_resetCCtxParams(ZSTD_CCtx_params* params);
+ Reset params to default, with the default compression level.
+
+
+
+size_t ZSTD_initCCtxParams(ZSTD_CCtx_params* cctxParams, int compressionLevel);
+ Initializes the compression parameters of cctxParams according to
+ compression level. All other parameters are reset to their default values.
+
+
+
+size_t ZSTD_initCCtxParams_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params);
+ Initializes the compression and frame parameters of cctxParams according to
+ params. All other parameters are reset to their default values.
+
+
+
+size_t ZSTD_CCtxParam_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, unsigned value);
+ Similar to ZSTD_CCtx_setParameter.
+ Set one compression parameter, selected by enum ZSTD_cParameter.
+ Parameters must be applied to a ZSTD_CCtx using ZSTD_CCtx_setParametersUsingCCtxParams().
+ Note : when `value` is an enum, cast it to unsigned for proper type checking.
+ @result : 0, or an error code (which can be tested with ZSTD_isError()).
+
+
+
+size_t ZSTD_CCtx_setParametersUsingCCtxParams(
+ ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params);
+ Apply a set of ZSTD_CCtx_params to the compression context.
+ This must be done before the dictionary is loaded.
+ The pledgedSrcSize is treated as unknown.
+ Multithreading parameters are applied only if nbThreads > 1.
+
+
+
+Block functions
Block functions produce and decode raw zstd blocks, without frame metadata.
Frame metadata cost is typically ~18 bytes, which can be non-negligible for very small blocks (< 100 bytes).
User will have to take in charge required information to regenerate data, such as compressed and content sizes.
diff --git a/lib/common/pool.c b/lib/common/pool.c
index 9567d112f..d7080f034 100644
--- a/lib/common/pool.c
+++ b/lib/common/pool.c
@@ -236,7 +236,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customM
}
void POOL_free(POOL_ctx* ctx) {
- assert(ctx == &g_ctx);
+ assert(!ctx || ctx == &g_ctx);
(void)ctx;
}
diff --git a/lib/common/threading.c b/lib/common/threading.c
index 4e47b6b91..a82c975b2 100644
--- a/lib/common/threading.c
+++ b/lib/common/threading.c
@@ -14,12 +14,8 @@
* This file will hold wrapper for systems, which do not support pthreads
*/
-/* When ZSTD_MULTITHREAD is not defined, this file would become an empty translation unit.
- * Include some ISO C header code to prevent this and portably avoid related warnings.
- * (Visual C++: C4206 / GCC: -Wpedantic / Clang: -Wempty-translation-unit)
- */
-#include
-
+/* create fake symbol to avoid empty trnaslation unit warning */
+int g_ZSTD_threading_useles_symbol;
#if defined(ZSTD_MULTITHREAD) && defined(_WIN32)
diff --git a/lib/compress/zstd_compress.c b/lib/compress/zstd_compress.c
index 99d5d7963..3abc1e913 100644
--- a/lib/compress/zstd_compress.c
+++ b/lib/compress/zstd_compress.c
@@ -25,17 +25,13 @@
#include "fse.h"
#define HUF_STATIC_LINKING_ONLY
#include "huf.h"
-#include "zstd_internal.h" /* includes zstd.h */
-#include "zstdmt_compress.h"
+#include "zstd_compress.h"
+#include "zstd_fast.h"
+#include "zstd_double_fast.h"
+#include "zstd_lazy.h"
+#include "zstd_opt.h"
-/*-*************************************
-* Constants
-***************************************/
-static const U32 g_searchStrength = 8; /* control skip over incompressible data */
-#define HASH_READ_SIZE 8
-typedef enum { ZSTDcs_created=0, ZSTDcs_init, ZSTDcs_ongoing, ZSTDcs_ending } ZSTD_compressionStage_e;
-
#define LDM_BUCKET_SIZE_LOG 3
#define LDM_MIN_MATCH_LENGTH 64
#define LDM_WINDOW_LOG 27
@@ -68,8 +64,6 @@ static void ZSTD_resetSeqStore(seqStore_t* ssPtr)
/*-*************************************
* Context memory management
***************************************/
-typedef enum { zcss_init=0, zcss_load, zcss_flush } ZSTD_cStreamStage;
-
struct ZSTD_CDict_s {
void* dictBuffer;
const void* dictContent;
@@ -77,66 +71,6 @@ struct ZSTD_CDict_s {
ZSTD_CCtx* refContext;
}; /* typedef'd to ZSTD_CDict within "zstd.h" */
-typedef struct ZSTD_prefixDict_s {
- const void* dict;
- size_t dictSize;
- ZSTD_dictMode_e dictMode;
-} ZSTD_prefixDict;
-
-struct ZSTD_CCtx_s {
- const BYTE* nextSrc; /* next block here to continue on current prefix */
- const BYTE* base; /* All regular indexes relative to this position */
- const BYTE* dictBase; /* extDict indexes relative to this position */
- U32 dictLimit; /* below that point, need extDict */
- U32 lowLimit; /* below that point, no more data */
- U32 nextToUpdate; /* index from which to continue dictionary update */
- U32 nextToUpdate3; /* index from which to continue dictionary update */
- U32 hashLog3; /* dispatch table : larger == faster, more memory */
- U32 loadedDictEnd; /* index of end of dictionary */
- ZSTD_compressionStage_e stage;
- U32 dictID;
- ZSTD_CCtx_params requestedParams;
- ZSTD_CCtx_params appliedParams;
- void* workSpace;
- size_t workSpaceSize;
- size_t blockSize;
- U64 pledgedSrcSizePlusOne; /* this way, 0 (default) == unknown */
- U64 consumedSrcSize;
- XXH64_state_t xxhState;
- ZSTD_customMem customMem;
- size_t staticSize;
-
- seqStore_t seqStore; /* sequences storage ptrs */
- optState_t optState;
- ldmState_t ldmState; /* long distance matching state */
- U32* hashTable;
- U32* hashTable3;
- U32* chainTable;
- ZSTD_entropyCTables_t* entropy;
-
- /* streaming */
- char* inBuff;
- size_t inBuffSize;
- size_t inToCompress;
- size_t inBuffPos;
- size_t inBuffTarget;
- char* outBuff;
- size_t outBuffSize;
- size_t outBuffContentSize;
- size_t outBuffFlushedSize;
- ZSTD_cStreamStage streamStage;
- U32 frameEnded;
-
- /* Dictionary */
- ZSTD_CDict* cdictLocal;
- const ZSTD_CDict* cdict;
- ZSTD_prefixDict prefixDict; /* single-usage dictionary */
-
- /* Multi-threading */
- ZSTDMT_CCtx* mtctx;
-};
-
-
ZSTD_CCtx* ZSTD_createCCtx(void)
{
return ZSTD_createCCtx_advanced(ZSTD_defaultCMem);
@@ -1088,6 +1022,8 @@ void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx) {
/*! ZSTD_copyCCtx_internal() :
* Duplicate an existing context `srcCCtx` into another one `dstCCtx`.
+ * The "context", in this case, refers to the hash and chain tables, entropy
+ * tables, and dictionary offsets.
* Only works during stage ZSTDcs_init (i.e. after creation, but before first call to ZSTD_compressContinue()).
* pledgedSrcSize=0 means "empty" if fParams.contentSizeFlag=1
* @return : 0, or an error code */
@@ -1102,6 +1038,7 @@ static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx,
memcpy(&dstCCtx->customMem, &srcCCtx->customMem, sizeof(ZSTD_customMem));
{ ZSTD_CCtx_params params = dstCCtx->requestedParams;
+ /* Copy only compression parameters related to tables. */
params.cParams = srcCCtx->appliedParams.cParams;
params.fParams = fParams;
ZSTD_resetCCtx_internal(dstCCtx, params, pledgedSrcSize,
@@ -1330,24 +1267,6 @@ static size_t ZSTD_compressLiterals (ZSTD_entropyCTables_t * entropy,
return lhSize+cLitSize;
}
-static const BYTE LL_Code[64] = { 0, 1, 2, 3, 4, 5, 6, 7,
- 8, 9, 10, 11, 12, 13, 14, 15,
- 16, 16, 17, 17, 18, 18, 19, 19,
- 20, 20, 20, 20, 21, 21, 21, 21,
- 22, 22, 22, 22, 22, 22, 22, 22,
- 23, 23, 23, 23, 23, 23, 23, 23,
- 24, 24, 24, 24, 24, 24, 24, 24,
- 24, 24, 24, 24, 24, 24, 24, 24 };
-
-static const BYTE ML_Code[128] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
- 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31,
- 32, 32, 33, 33, 34, 34, 35, 35, 36, 36, 36, 36, 37, 37, 37, 37,
- 38, 38, 38, 38, 38, 38, 38, 38, 39, 39, 39, 39, 39, 39, 39, 39,
- 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40,
- 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41,
- 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42,
- 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42 };
-
void ZSTD_seqToCodes(const seqStore_t* seqStorePtr)
{
@@ -1633,1510 +1552,6 @@ MEM_STATIC size_t ZSTD_compressSequences(seqStore_t* seqStorePtr,
return cSize;
}
-
-/*! ZSTD_storeSeq() :
- Store a sequence (literal length, literals, offset code and match length code) into seqStore_t.
- `offsetCode` : distance to match, or 0 == repCode.
- `matchCode` : matchLength - MINMATCH
-*/
-MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const void* literals, U32 offsetCode, size_t matchCode)
-{
-#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG >= 6)
- static const BYTE* g_start = NULL;
- U32 const pos = (U32)((const BYTE*)literals - g_start);
- if (g_start==NULL) g_start = (const BYTE*)literals;
- if ((pos > 0) && (pos < 1000000000))
- DEBUGLOG(6, "Cpos %6u :%5u literals & match %3u bytes at distance %6u",
- pos, (U32)litLength, (U32)matchCode+MINMATCH, (U32)offsetCode);
-#endif
- /* copy Literals */
- assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + 128 KB);
- ZSTD_wildcopy(seqStorePtr->lit, literals, litLength);
- seqStorePtr->lit += litLength;
-
- /* literal Length */
- if (litLength>0xFFFF) {
- seqStorePtr->longLengthID = 1;
- seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
- }
- seqStorePtr->sequences[0].litLength = (U16)litLength;
-
- /* match offset */
- seqStorePtr->sequences[0].offset = offsetCode + 1;
-
- /* match Length */
- if (matchCode>0xFFFF) {
- seqStorePtr->longLengthID = 2;
- seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
- }
- seqStorePtr->sequences[0].matchLength = (U16)matchCode;
-
- seqStorePtr->sequences++;
-}
-
-
-/*-*************************************
-* Match length counter
-***************************************/
-static unsigned ZSTD_NbCommonBytes (register size_t val)
-{
- if (MEM_isLittleEndian()) {
- if (MEM_64bits()) {
-# if defined(_MSC_VER) && defined(_WIN64)
- unsigned long r = 0;
- _BitScanForward64( &r, (U64)val );
- return (unsigned)(r>>3);
-# elif defined(__GNUC__) && (__GNUC__ >= 3)
- return (__builtin_ctzll((U64)val) >> 3);
-# else
- static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2,
- 0, 3, 1, 3, 1, 4, 2, 7,
- 0, 2, 3, 6, 1, 5, 3, 5,
- 1, 3, 4, 4, 2, 5, 6, 7,
- 7, 0, 1, 2, 3, 3, 4, 6,
- 2, 6, 5, 5, 3, 4, 5, 6,
- 7, 1, 2, 4, 6, 4, 4, 5,
- 7, 2, 6, 5, 7, 6, 7, 7 };
- return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
-# endif
- } else { /* 32 bits */
-# if defined(_MSC_VER)
- unsigned long r=0;
- _BitScanForward( &r, (U32)val );
- return (unsigned)(r>>3);
-# elif defined(__GNUC__) && (__GNUC__ >= 3)
- return (__builtin_ctz((U32)val) >> 3);
-# else
- static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0,
- 3, 2, 2, 1, 3, 2, 0, 1,
- 3, 3, 1, 2, 2, 2, 2, 0,
- 3, 1, 2, 0, 1, 0, 1, 1 };
- return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
-# endif
- }
- } else { /* Big Endian CPU */
- if (MEM_64bits()) {
-# if defined(_MSC_VER) && defined(_WIN64)
- unsigned long r = 0;
- _BitScanReverse64( &r, val );
- return (unsigned)(r>>3);
-# elif defined(__GNUC__) && (__GNUC__ >= 3)
- return (__builtin_clzll(val) >> 3);
-# else
- unsigned r;
- const unsigned n32 = sizeof(size_t)*4; /* calculate this way due to compiler complaining in 32-bits mode */
- if (!(val>>n32)) { r=4; } else { r=0; val>>=n32; }
- if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
- r += (!val);
- return r;
-# endif
- } else { /* 32 bits */
-# if defined(_MSC_VER)
- unsigned long r = 0;
- _BitScanReverse( &r, (unsigned long)val );
- return (unsigned)(r>>3);
-# elif defined(__GNUC__) && (__GNUC__ >= 3)
- return (__builtin_clz((U32)val) >> 3);
-# else
- unsigned r;
- if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
- r += (!val);
- return r;
-# endif
- } }
-}
-
-
-static size_t ZSTD_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* const pInLimit)
-{
- const BYTE* const pStart = pIn;
- const BYTE* const pInLoopLimit = pInLimit - (sizeof(size_t)-1);
-
- while (pIn < pInLoopLimit) {
- size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn);
- if (!diff) { pIn+=sizeof(size_t); pMatch+=sizeof(size_t); continue; }
- pIn += ZSTD_NbCommonBytes(diff);
- return (size_t)(pIn - pStart);
- }
- if (MEM_64bits()) if ((pIn<(pInLimit-3)) && (MEM_read32(pMatch) == MEM_read32(pIn))) { pIn+=4; pMatch+=4; }
- if ((pIn<(pInLimit-1)) && (MEM_read16(pMatch) == MEM_read16(pIn))) { pIn+=2; pMatch+=2; }
- if ((pIn> (32-h) ; }
-MEM_STATIC size_t ZSTD_hash3Ptr(const void* ptr, U32 h) { return ZSTD_hash3(MEM_readLE32(ptr), h); } /* only in zstd_opt.h */
-
-static const U32 prime4bytes = 2654435761U;
-static U32 ZSTD_hash4(U32 u, U32 h) { return (u * prime4bytes) >> (32-h) ; }
-static size_t ZSTD_hash4Ptr(const void* ptr, U32 h) { return ZSTD_hash4(MEM_read32(ptr), h); }
-
-static const U64 prime5bytes = 889523592379ULL;
-static size_t ZSTD_hash5(U64 u, U32 h) { return (size_t)(((u << (64-40)) * prime5bytes) >> (64-h)) ; }
-static size_t ZSTD_hash5Ptr(const void* p, U32 h) { return ZSTD_hash5(MEM_readLE64(p), h); }
-
-static const U64 prime6bytes = 227718039650203ULL;
-static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u << (64-48)) * prime6bytes) >> (64-h)) ; }
-static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); }
-
-static const U64 prime7bytes = 58295818150454627ULL;
-static size_t ZSTD_hash7(U64 u, U32 h) { return (size_t)(((u << (64-56)) * prime7bytes) >> (64-h)) ; }
-static size_t ZSTD_hash7Ptr(const void* p, U32 h) { return ZSTD_hash7(MEM_readLE64(p), h); }
-
-static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL;
-static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; }
-static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); }
-
-static size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls)
-{
- switch(mls)
- {
- default:
- case 4: return ZSTD_hash4Ptr(p, hBits);
- case 5: return ZSTD_hash5Ptr(p, hBits);
- case 6: return ZSTD_hash6Ptr(p, hBits);
- case 7: return ZSTD_hash7Ptr(p, hBits);
- case 8: return ZSTD_hash8Ptr(p, hBits);
- }
-}
-
-
-/*-*************************************
-* Fast Scan
-***************************************/
-static void ZSTD_fillHashTable (ZSTD_CCtx* zc, const void* end, const U32 mls)
-{
- U32* const hashTable = zc->hashTable;
- U32 const hBits = zc->appliedParams.cParams.hashLog;
- const BYTE* const base = zc->base;
- const BYTE* ip = base + zc->nextToUpdate;
- const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
- const size_t fastHashFillStep = 3;
-
- while(ip <= iend) {
- hashTable[ZSTD_hashPtr(ip, hBits, mls)] = (U32)(ip - base);
- ip += fastHashFillStep;
- }
-}
-
-
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_fast_generic(ZSTD_CCtx* cctx,
- const void* src, size_t srcSize,
- const U32 mls)
-{
- U32* const hashTable = cctx->hashTable;
- U32 const hBits = cctx->appliedParams.cParams.hashLog;
- seqStore_t* seqStorePtr = &(cctx->seqStore);
- const BYTE* const base = cctx->base;
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const U32 lowestIndex = cctx->dictLimit;
- const BYTE* const lowest = base + lowestIndex;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - HASH_READ_SIZE;
- U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
- U32 offsetSaved = 0;
-
- /* init */
- ip += (ip==lowest);
- { U32 const maxRep = (U32)(ip-lowest);
- if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
- if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
- }
-
- /* Main Search Loop */
- while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
- size_t mLength;
- size_t const h = ZSTD_hashPtr(ip, hBits, mls);
- U32 const current = (U32)(ip-base);
- U32 const matchIndex = hashTable[h];
- const BYTE* match = base + matchIndex;
- hashTable[h] = current; /* update hash table */
-
- if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
- mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
- ip++;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- U32 offset;
- if ( (matchIndex <= lowestIndex) || (MEM_read32(match) != MEM_read32(ip)) ) {
- ip += ((ip-anchor) >> g_searchStrength) + 1;
- continue;
- }
- mLength = ZSTD_count(ip+4, match+4, iend) + 4;
- offset = (U32)(ip-match);
- while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
- offset_2 = offset_1;
- offset_1 = offset;
-
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
- }
-
- /* match found */
- ip += mLength;
- anchor = ip;
-
- if (ip <= ilimit) {
- /* Fill Table */
- hashTable[ZSTD_hashPtr(base+current+2, hBits, mls)] = current+2; /* here because current+2 could be > iend-8 */
- hashTable[ZSTD_hashPtr(ip-2, hBits, mls)] = (U32)(ip-2-base);
- /* check immediate repcode */
- while ( (ip <= ilimit)
- && ( (offset_2>0)
- & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
- /* store sequence */
- size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
- hashTable[ZSTD_hashPtr(ip, hBits, mls)] = (U32)(ip-base);
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, rLength-MINMATCH);
- ip += rLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } } }
-
- /* save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : offsetSaved;
- seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : offsetSaved;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-static size_t ZSTD_compressBlock_fast(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize)
-{
- const U32 mls = ctx->appliedParams.cParams.searchLength;
- switch(mls)
- {
- default: /* includes case 3 */
- case 4 :
- return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 4);
- case 5 :
- return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 5);
- case 6 :
- return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 6);
- case 7 :
- return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 7);
- }
-}
-
-
-static size_t ZSTD_compressBlock_fast_extDict_generic(
- ZSTD_CCtx* ctx,
- const void* src, size_t srcSize,
- const U32 mls)
-{
- U32* hashTable = ctx->hashTable;
- const U32 hBits = ctx->appliedParams.cParams.hashLog;
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- const BYTE* const base = ctx->base;
- const BYTE* const dictBase = ctx->dictBase;
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const U32 lowestIndex = ctx->lowLimit;
- const BYTE* const dictStart = dictBase + lowestIndex;
- const U32 dictLimit = ctx->dictLimit;
- const BYTE* const lowPrefixPtr = base + dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
-
- /* Search Loop */
- while (ip < ilimit) { /* < instead of <=, because (ip+1) */
- const size_t h = ZSTD_hashPtr(ip, hBits, mls);
- const U32 matchIndex = hashTable[h];
- const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
- const BYTE* match = matchBase + matchIndex;
- const U32 current = (U32)(ip-base);
- const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
- const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* repMatch = repBase + repIndex;
- size_t mLength;
- hashTable[h] = current; /* update hash table */
-
- if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
- && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
- const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
- mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
- ip++;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- if ( (matchIndex < lowestIndex) ||
- (MEM_read32(match) != MEM_read32(ip)) ) {
- ip += ((ip-anchor) >> g_searchStrength) + 1;
- continue;
- }
- { const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
- U32 offset;
- mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
- while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
- offset = current - matchIndex;
- offset_2 = offset_1;
- offset_1 = offset;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
- } }
-
- /* found a match : store it */
- ip += mLength;
- anchor = ip;
-
- if (ip <= ilimit) {
- /* Fill Table */
- hashTable[ZSTD_hashPtr(base+current+2, hBits, mls)] = current+2;
- hashTable[ZSTD_hashPtr(ip-2, hBits, mls)] = (U32)(ip-2-base);
- /* check immediate repcode */
- while (ip <= ilimit) {
- U32 const current2 = (U32)(ip-base);
- U32 const repIndex2 = current2 - offset_2;
- const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
- if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
- && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
- const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
- size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
- U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, repLength2-MINMATCH);
- hashTable[ZSTD_hashPtr(ip, hBits, mls)] = current2;
- ip += repLength2;
- anchor = ip;
- continue;
- }
- break;
- } } }
-
- /* save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-static size_t ZSTD_compressBlock_fast_extDict(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize)
-{
- U32 const mls = ctx->appliedParams.cParams.searchLength;
- switch(mls)
- {
- default: /* includes case 3 */
- case 4 :
- return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 4);
- case 5 :
- return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 5);
- case 6 :
- return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 6);
- case 7 :
- return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 7);
- }
-}
-
-
-/*-*************************************
-* Double Fast
-***************************************/
-static void ZSTD_fillDoubleHashTable (ZSTD_CCtx* cctx, const void* end, const U32 mls)
-{
- U32* const hashLarge = cctx->hashTable;
- U32 const hBitsL = cctx->appliedParams.cParams.hashLog;
- U32* const hashSmall = cctx->chainTable;
- U32 const hBitsS = cctx->appliedParams.cParams.chainLog;
- const BYTE* const base = cctx->base;
- const BYTE* ip = base + cctx->nextToUpdate;
- const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
- const size_t fastHashFillStep = 3;
-
- while(ip <= iend) {
- hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip - base);
- hashLarge[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip - base);
- ip += fastHashFillStep;
- }
-}
-
-
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_doubleFast_generic(ZSTD_CCtx* cctx,
- const void* src, size_t srcSize,
- const U32 mls)
-{
- U32* const hashLong = cctx->hashTable;
- const U32 hBitsL = cctx->appliedParams.cParams.hashLog;
- U32* const hashSmall = cctx->chainTable;
- const U32 hBitsS = cctx->appliedParams.cParams.chainLog;
- seqStore_t* seqStorePtr = &(cctx->seqStore);
- const BYTE* const base = cctx->base;
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const U32 lowestIndex = cctx->dictLimit;
- const BYTE* const lowest = base + lowestIndex;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - HASH_READ_SIZE;
- U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
- U32 offsetSaved = 0;
-
- /* init */
- ip += (ip==lowest);
- { U32 const maxRep = (U32)(ip-lowest);
- if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
- if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
- }
-
- /* Main Search Loop */
- while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
- size_t mLength;
- size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
- size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
- U32 const current = (U32)(ip-base);
- U32 const matchIndexL = hashLong[h2];
- U32 const matchIndexS = hashSmall[h];
- const BYTE* matchLong = base + matchIndexL;
- const BYTE* match = base + matchIndexS;
- hashLong[h2] = hashSmall[h] = current; /* update hash tables */
-
- assert(offset_1 <= current); /* supposed guaranteed by construction */
- if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
- /* favor repcode */
- mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
- ip++;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- U32 offset;
- if ( (matchIndexL > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip)) ) {
- mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8;
- offset = (U32)(ip-matchLong);
- while (((ip>anchor) & (matchLong>lowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
- } else if ( (matchIndexS > lowestIndex) && (MEM_read32(match) == MEM_read32(ip)) ) {
- size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
- U32 const matchIndexL3 = hashLong[hl3];
- const BYTE* matchL3 = base + matchIndexL3;
- hashLong[hl3] = current + 1;
- if ( (matchIndexL3 > lowestIndex) && (MEM_read64(matchL3) == MEM_read64(ip+1)) ) {
- mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8;
- ip++;
- offset = (U32)(ip-matchL3);
- while (((ip>anchor) & (matchL3>lowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
- } else {
- mLength = ZSTD_count(ip+4, match+4, iend) + 4;
- offset = (U32)(ip-match);
- while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
- }
- } else {
- ip += ((ip-anchor) >> g_searchStrength) + 1;
- continue;
- }
-
- offset_2 = offset_1;
- offset_1 = offset;
-
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
- }
-
- /* match found */
- ip += mLength;
- anchor = ip;
-
- if (ip <= ilimit) {
- /* Fill Table */
- hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] =
- hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2; /* here because current+2 could be > iend-8 */
- hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] =
- hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
-
- /* check immediate repcode */
- while ( (ip <= ilimit)
- && ( (offset_2>0)
- & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
- /* store sequence */
- size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
- hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
- hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, rLength-MINMATCH);
- ip += rLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } } }
-
- /* save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : offsetSaved;
- seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : offsetSaved;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-
-static size_t ZSTD_compressBlock_doubleFast(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize)
-{
- const U32 mls = ctx->appliedParams.cParams.searchLength;
- switch(mls)
- {
- default: /* includes case 3 */
- case 4 :
- return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 4);
- case 5 :
- return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 5);
- case 6 :
- return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 6);
- case 7 :
- return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 7);
- }
-}
-
-
-static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
- ZSTD_CCtx* ctx,
- const void* src, size_t srcSize,
- const U32 mls)
-{
- U32* const hashLong = ctx->hashTable;
- U32 const hBitsL = ctx->appliedParams.cParams.hashLog;
- U32* const hashSmall = ctx->chainTable;
- U32 const hBitsS = ctx->appliedParams.cParams.chainLog;
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- const BYTE* const base = ctx->base;
- const BYTE* const dictBase = ctx->dictBase;
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const U32 lowestIndex = ctx->lowLimit;
- const BYTE* const dictStart = dictBase + lowestIndex;
- const U32 dictLimit = ctx->dictLimit;
- const BYTE* const lowPrefixPtr = base + dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
-
- /* Search Loop */
- while (ip < ilimit) { /* < instead of <=, because (ip+1) */
- const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
- const U32 matchIndex = hashSmall[hSmall];
- const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
- const BYTE* match = matchBase + matchIndex;
-
- const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8);
- const U32 matchLongIndex = hashLong[hLong];
- const BYTE* matchLongBase = matchLongIndex < dictLimit ? dictBase : base;
- const BYTE* matchLong = matchLongBase + matchLongIndex;
-
- const U32 current = (U32)(ip-base);
- const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
- const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* repMatch = repBase + repIndex;
- size_t mLength;
- hashSmall[hSmall] = hashLong[hLong] = current; /* update hash table */
-
- if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
- && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
- const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
- mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
- ip++;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
- } else {
- if ((matchLongIndex > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
- const BYTE* matchEnd = matchLongIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchLongIndex < dictLimit ? dictStart : lowPrefixPtr;
- U32 offset;
- mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, lowPrefixPtr) + 8;
- offset = current - matchLongIndex;
- while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
- offset_2 = offset_1;
- offset_1 = offset;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
-
- } else if ((matchIndex > lowestIndex) && (MEM_read32(match) == MEM_read32(ip))) {
- size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
- U32 const matchIndex3 = hashLong[h3];
- const BYTE* const match3Base = matchIndex3 < dictLimit ? dictBase : base;
- const BYTE* match3 = match3Base + matchIndex3;
- U32 offset;
- hashLong[h3] = current + 1;
- if ( (matchIndex3 > lowestIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
- const BYTE* matchEnd = matchIndex3 < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex3 < dictLimit ? dictStart : lowPrefixPtr;
- mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, lowPrefixPtr) + 8;
- ip++;
- offset = current+1 - matchIndex3;
- while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
- } else {
- const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
- const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
- mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
- offset = current - matchIndex;
- while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
- }
- offset_2 = offset_1;
- offset_1 = offset;
- ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
-
- } else {
- ip += ((ip-anchor) >> g_searchStrength) + 1;
- continue;
- } }
-
- /* found a match : store it */
- ip += mLength;
- anchor = ip;
-
- if (ip <= ilimit) {
- /* Fill Table */
- hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2;
- hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] = current+2;
- hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
- hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
- /* check immediate repcode */
- while (ip <= ilimit) {
- U32 const current2 = (U32)(ip-base);
- U32 const repIndex2 = current2 - offset_2;
- const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
- if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
- && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
- const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
- size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
- U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, repLength2-MINMATCH);
- hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
- hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
- ip += repLength2;
- anchor = ip;
- continue;
- }
- break;
- } } }
-
- /* save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-
-static size_t ZSTD_compressBlock_doubleFast_extDict(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize)
-{
- U32 const mls = ctx->appliedParams.cParams.searchLength;
- switch(mls)
- {
- default: /* includes case 3 */
- case 4 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 4);
- case 5 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 5);
- case 6 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 6);
- case 7 :
- return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 7);
- }
-}
-
-
-/*-*************************************
-* Binary Tree search
-***************************************/
-/** ZSTD_insertBt1() : add one or multiple positions to tree.
-* ip : assumed <= iend-8 .
-* @return : nb of positions added */
-static U32 ZSTD_insertBt1(ZSTD_CCtx* zc, const BYTE* const ip, const U32 mls, const BYTE* const iend, U32 nbCompares,
- U32 extDict)
-{
- U32* const hashTable = zc->hashTable;
- U32 const hashLog = zc->appliedParams.cParams.hashLog;
- size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
- U32* const bt = zc->chainTable;
- U32 const btLog = zc->appliedParams.cParams.chainLog - 1;
- U32 const btMask = (1 << btLog) - 1;
- U32 matchIndex = hashTable[h];
- size_t commonLengthSmaller=0, commonLengthLarger=0;
- const BYTE* const base = zc->base;
- const BYTE* const dictBase = zc->dictBase;
- const U32 dictLimit = zc->dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const BYTE* match;
- const U32 current = (U32)(ip-base);
- const U32 btLow = btMask >= current ? 0 : current - btMask;
- U32* smallerPtr = bt + 2*(current&btMask);
- U32* largerPtr = smallerPtr + 1;
- U32 dummy32; /* to be nullified at the end */
- U32 const windowLow = zc->lowLimit;
- U32 matchEndIdx = current+8;
- size_t bestLength = 8;
-#ifdef ZSTD_C_PREDICT
- U32 predictedSmall = *(bt + 2*((current-1)&btMask) + 0);
- U32 predictedLarge = *(bt + 2*((current-1)&btMask) + 1);
- predictedSmall += (predictedSmall>0);
- predictedLarge += (predictedLarge>0);
-#endif /* ZSTD_C_PREDICT */
-
- hashTable[h] = current; /* Update Hash Table */
-
- while (nbCompares-- && (matchIndex > windowLow)) {
- U32* const nextPtr = bt + 2*(matchIndex & btMask);
- size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
-
-#ifdef ZSTD_C_PREDICT /* note : can create issues when hlog small <= 11 */
- const U32* predictPtr = bt + 2*((matchIndex-1) & btMask); /* written this way, as bt is a roll buffer */
- if (matchIndex == predictedSmall) {
- /* no need to check length, result known */
- *smallerPtr = matchIndex;
- if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
- matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
- predictedSmall = predictPtr[1] + (predictPtr[1]>0);
- continue;
- }
- if (matchIndex == predictedLarge) {
- *largerPtr = matchIndex;
- if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- largerPtr = nextPtr;
- matchIndex = nextPtr[0];
- predictedLarge = predictPtr[0] + (predictPtr[0]>0);
- continue;
- }
-#endif
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
- match = base + matchIndex;
- if (match[matchLength] == ip[matchLength])
- matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iend) +1;
- } else {
- match = dictBase + matchIndex;
- matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
- if (matchIndex+matchLength >= dictLimit)
- match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
- }
-
- if (matchLength > bestLength) {
- bestLength = matchLength;
- if (matchLength > matchEndIdx - matchIndex)
- matchEndIdx = matchIndex + (U32)matchLength;
- }
-
- if (ip+matchLength == iend) /* equal : no way to know if inf or sup */
- break; /* drop , to guarantee consistency ; miss a bit of compression, but other solutions can corrupt the tree */
-
- if (match[matchLength] < ip[matchLength]) { /* necessarily within correct buffer */
- /* match is smaller than current */
- *smallerPtr = matchIndex; /* update smaller idx */
- commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
- if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
- matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
- } else {
- /* match is larger than current */
- *largerPtr = matchIndex;
- commonLengthLarger = matchLength;
- if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- largerPtr = nextPtr;
- matchIndex = nextPtr[0];
- } }
-
- *smallerPtr = *largerPtr = 0;
- if (bestLength > 384) return MIN(192, (U32)(bestLength - 384)); /* speed optimization */
- if (matchEndIdx > current + 8) return matchEndIdx - current - 8;
- return 1;
-}
-
-
-static size_t ZSTD_insertBtAndFindBestMatch (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iend,
- size_t* offsetPtr,
- U32 nbCompares, const U32 mls,
- U32 extDict)
-{
- U32* const hashTable = zc->hashTable;
- U32 const hashLog = zc->appliedParams.cParams.hashLog;
- size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
- U32* const bt = zc->chainTable;
- U32 const btLog = zc->appliedParams.cParams.chainLog - 1;
- U32 const btMask = (1 << btLog) - 1;
- U32 matchIndex = hashTable[h];
- size_t commonLengthSmaller=0, commonLengthLarger=0;
- const BYTE* const base = zc->base;
- const BYTE* const dictBase = zc->dictBase;
- const U32 dictLimit = zc->dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const U32 current = (U32)(ip-base);
- const U32 btLow = btMask >= current ? 0 : current - btMask;
- const U32 windowLow = zc->lowLimit;
- U32* smallerPtr = bt + 2*(current&btMask);
- U32* largerPtr = bt + 2*(current&btMask) + 1;
- U32 matchEndIdx = current+8;
- U32 dummy32; /* to be nullified at the end */
- size_t bestLength = 0;
-
- hashTable[h] = current; /* Update Hash Table */
-
- while (nbCompares-- && (matchIndex > windowLow)) {
- U32* const nextPtr = bt + 2*(matchIndex & btMask);
- size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
- const BYTE* match;
-
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
- match = base + matchIndex;
- if (match[matchLength] == ip[matchLength])
- matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iend) +1;
- } else {
- match = dictBase + matchIndex;
- matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
- if (matchIndex+matchLength >= dictLimit)
- match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
- }
-
- if (matchLength > bestLength) {
- if (matchLength > matchEndIdx - matchIndex)
- matchEndIdx = matchIndex + (U32)matchLength;
- if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(current-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) )
- bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + current - matchIndex;
- if (ip+matchLength == iend) /* equal : no way to know if inf or sup */
- break; /* drop, to guarantee consistency (miss a little bit of compression) */
- }
-
- if (match[matchLength] < ip[matchLength]) {
- /* match is smaller than current */
- *smallerPtr = matchIndex; /* update smaller idx */
- commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
- if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
- matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
- } else {
- /* match is larger than current */
- *largerPtr = matchIndex;
- commonLengthLarger = matchLength;
- if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- largerPtr = nextPtr;
- matchIndex = nextPtr[0];
- } }
-
- *smallerPtr = *largerPtr = 0;
-
- zc->nextToUpdate = (matchEndIdx > current + 8) ? matchEndIdx - 8 : current+1;
- return bestLength;
-}
-
-
-static void ZSTD_updateTree(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls)
-{
- const BYTE* const base = zc->base;
- const U32 target = (U32)(ip - base);
- U32 idx = zc->nextToUpdate;
-
- while(idx < target)
- idx += ZSTD_insertBt1(zc, base+idx, mls, iend, nbCompares, 0);
-}
-
-/** ZSTD_BtFindBestMatch() : Tree updater, providing best match */
-static size_t ZSTD_BtFindBestMatch (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 mls)
-{
- if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateTree(zc, ip, iLimit, maxNbAttempts, mls);
- return ZSTD_insertBtAndFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, mls, 0);
-}
-
-
-static size_t ZSTD_BtFindBestMatch_selectMLS (
- ZSTD_CCtx* zc, /* Index table will be updated */
- const BYTE* ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 matchLengthSearch)
-{
- switch(matchLengthSearch)
- {
- default : /* includes case 3 */
- case 4 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4);
- case 5 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5);
- case 7 :
- case 6 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6);
- }
-}
-
-
-static void ZSTD_updateTree_extDict(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls)
-{
- const BYTE* const base = zc->base;
- const U32 target = (U32)(ip - base);
- U32 idx = zc->nextToUpdate;
-
- while (idx < target) idx += ZSTD_insertBt1(zc, base+idx, mls, iend, nbCompares, 1);
-}
-
-
-/** Tree updater, providing best match */
-static size_t ZSTD_BtFindBestMatch_extDict (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 mls)
-{
- if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateTree_extDict(zc, ip, iLimit, maxNbAttempts, mls);
- return ZSTD_insertBtAndFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, mls, 1);
-}
-
-
-static size_t ZSTD_BtFindBestMatch_selectMLS_extDict (
- ZSTD_CCtx* zc, /* Index table will be updated */
- const BYTE* ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 matchLengthSearch)
-{
- switch(matchLengthSearch)
- {
- default : /* includes case 3 */
- case 4 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4);
- case 5 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5);
- case 7 :
- case 6 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6);
- }
-}
-
-
-
-/* *********************************
-* Hash Chain
-***********************************/
-#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & mask]
-
-/* Update chains up to ip (excluded)
- Assumption : always within prefix (i.e. not within extDict) */
-FORCE_INLINE_TEMPLATE
-U32 ZSTD_insertAndFindFirstIndex (ZSTD_CCtx* zc, const BYTE* ip, U32 mls)
-{
- U32* const hashTable = zc->hashTable;
- const U32 hashLog = zc->appliedParams.cParams.hashLog;
- U32* const chainTable = zc->chainTable;
- const U32 chainMask = (1 << zc->appliedParams.cParams.chainLog) - 1;
- const BYTE* const base = zc->base;
- const U32 target = (U32)(ip - base);
- U32 idx = zc->nextToUpdate;
-
- while(idx < target) { /* catch up */
- size_t const h = ZSTD_hashPtr(base+idx, hashLog, mls);
- NEXT_IN_CHAIN(idx, chainMask) = hashTable[h];
- hashTable[h] = idx;
- idx++;
- }
-
- zc->nextToUpdate = target;
- return hashTable[ZSTD_hashPtr(ip, hashLog, mls)];
-}
-
-
-/* inlining is important to hardwire a hot branch (template emulation) */
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_HcFindBestMatch_generic (
- ZSTD_CCtx* zc, /* Index table will be updated */
- const BYTE* const ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 mls, const U32 extDict)
-{
- U32* const chainTable = zc->chainTable;
- const U32 chainSize = (1 << zc->appliedParams.cParams.chainLog);
- const U32 chainMask = chainSize-1;
- const BYTE* const base = zc->base;
- const BYTE* const dictBase = zc->dictBase;
- const U32 dictLimit = zc->dictLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const U32 lowLimit = zc->lowLimit;
- const U32 current = (U32)(ip-base);
- const U32 minChain = current > chainSize ? current - chainSize : 0;
- int nbAttempts=maxNbAttempts;
- size_t ml=4-1;
-
- /* HC4 match finder */
- U32 matchIndex = ZSTD_insertAndFindFirstIndex (zc, ip, mls);
-
- for ( ; (matchIndex>lowLimit) & (nbAttempts>0) ; nbAttempts--) {
- const BYTE* match;
- size_t currentMl=0;
- if ((!extDict) || matchIndex >= dictLimit) {
- match = base + matchIndex;
- if (match[ml] == ip[ml]) /* potentially better */
- currentMl = ZSTD_count(ip, match, iLimit);
- } else {
- match = dictBase + matchIndex;
- if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */
- currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4;
- }
-
- /* save best solution */
- if (currentMl > ml) {
- ml = currentMl;
- *offsetPtr = current - matchIndex + ZSTD_REP_MOVE;
- if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
- }
-
- if (matchIndex <= minChain) break;
- matchIndex = NEXT_IN_CHAIN(matchIndex, chainMask);
- }
-
- return ml;
-}
-
-
-FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_selectMLS (
- ZSTD_CCtx* zc,
- const BYTE* ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 matchLengthSearch)
-{
- switch(matchLengthSearch)
- {
- default : /* includes case 3 */
- case 4 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4, 0);
- case 5 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5, 0);
- case 7 :
- case 6 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6, 0);
- }
-}
-
-
-FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_extDict_selectMLS (
- ZSTD_CCtx* zc,
- const BYTE* ip, const BYTE* const iLimit,
- size_t* offsetPtr,
- const U32 maxNbAttempts, const U32 matchLengthSearch)
-{
- switch(matchLengthSearch)
- {
- default : /* includes case 3 */
- case 4 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4, 1);
- case 5 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5, 1);
- case 7 :
- case 6 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6, 1);
- }
-}
-
-
-/* *******************************
-* Common parser - lazy strategy
-*********************************/
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_lazy_generic(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize,
- const U32 searchMethod, const U32 depth)
-{
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- const BYTE* const base = ctx->base + ctx->dictLimit;
-
- U32 const maxSearches = 1 << ctx->appliedParams.cParams.searchLog;
- U32 const mls = ctx->appliedParams.cParams.searchLength;
-
- typedef size_t (*searchMax_f)(ZSTD_CCtx* zc, const BYTE* ip, const BYTE* iLimit,
- size_t* offsetPtr,
- U32 maxNbAttempts, U32 matchLengthSearch);
- searchMax_f const searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS : ZSTD_HcFindBestMatch_selectMLS;
- U32 offset_1 = seqStorePtr->rep[0], offset_2 = seqStorePtr->rep[1], savedOffset=0;
-
- /* init */
- ip += (ip==base);
- ctx->nextToUpdate3 = ctx->nextToUpdate;
- { U32 const maxRep = (U32)(ip-base);
- if (offset_2 > maxRep) savedOffset = offset_2, offset_2 = 0;
- if (offset_1 > maxRep) savedOffset = offset_1, offset_1 = 0;
- }
-
- /* Match Loop */
- while (ip < ilimit) {
- size_t matchLength=0;
- size_t offset=0;
- const BYTE* start=ip+1;
-
- /* check repCode */
- if ((offset_1>0) & (MEM_read32(ip+1) == MEM_read32(ip+1 - offset_1))) {
- /* repcode : we take it */
- matchLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
- if (depth==0) goto _storeSequence;
- }
-
- /* first search (depth 0) */
- { size_t offsetFound = 99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offsetFound, maxSearches, mls);
- if (ml2 > matchLength)
- matchLength = ml2, start = ip, offset=offsetFound;
- }
-
- if (matchLength < 4) {
- ip += ((ip-anchor) >> g_searchStrength) + 1; /* jump faster over incompressible sections */
- continue;
- }
-
- /* let's try to find a better solution */
- if (depth>=1)
- while (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
- size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
- int const gain2 = (int)(mlRep * 3);
- int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
- if ((mlRep >= 4) && (gain2 > gain1))
- matchLength = mlRep, offset = 0, start = ip;
- }
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
- int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
- if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
- continue; /* search a better one */
- } }
-
- /* let's find an even better one */
- if ((depth==2) && (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
- size_t const ml2 = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
- int const gain2 = (int)(ml2 * 4);
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
- if ((ml2 >= 4) && (gain2 > gain1))
- matchLength = ml2, offset = 0, start = ip;
- }
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
- int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
- if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
- continue;
- } } }
- break; /* nothing found : store previous solution */
- }
-
- /* NOTE:
- * start[-offset+ZSTD_REP_MOVE-1] is undefined behavior.
- * (-offset+ZSTD_REP_MOVE-1) is unsigned, and is added to start, which
- * overflows the pointer, which is undefined behavior.
- */
- /* catch up */
- if (offset) {
- while ( (start > anchor)
- && (start > base+offset-ZSTD_REP_MOVE)
- && (start[-1] == (start-offset+ZSTD_REP_MOVE)[-1]) ) /* only search for offset within prefix */
- { start--; matchLength++; }
- offset_2 = offset_1; offset_1 = (U32)(offset - ZSTD_REP_MOVE);
- }
- /* store sequence */
-_storeSequence:
- { size_t const litLength = start - anchor;
- ZSTD_storeSeq(seqStorePtr, litLength, anchor, (U32)offset, matchLength-MINMATCH);
- anchor = ip = start + matchLength;
- }
-
- /* check immediate repcode */
- while ( (ip <= ilimit)
- && ((offset_2>0)
- & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
- /* store sequence */
- matchLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
- offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap repcodes */
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, matchLength-MINMATCH);
- ip += matchLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- } }
-
- /* Save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : savedOffset;
- seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : savedOffset;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-
-static size_t ZSTD_compressBlock_btlazy2(ZSTD_CCtx* ctx, const void* src,
- size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 1, 2);
-}
-
-static size_t ZSTD_compressBlock_lazy2(ZSTD_CCtx* ctx, const void* src,
- size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 2);
-}
-
-static size_t ZSTD_compressBlock_lazy(ZSTD_CCtx* ctx, const void* src,
- size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 1);
-}
-
-static size_t ZSTD_compressBlock_greedy(ZSTD_CCtx* ctx, const void* src,
- size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 0);
-}
-
-
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_lazy_extDict_generic(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize,
- const U32 searchMethod, const U32 depth)
-{
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- const BYTE* const base = ctx->base;
- const U32 dictLimit = ctx->dictLimit;
- const U32 lowestIndex = ctx->lowLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const BYTE* const dictBase = ctx->dictBase;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const dictStart = dictBase + ctx->lowLimit;
-
- const U32 maxSearches = 1 << ctx->appliedParams.cParams.searchLog;
- const U32 mls = ctx->appliedParams.cParams.searchLength;
-
- typedef size_t (*searchMax_f)(ZSTD_CCtx* zc, const BYTE* ip, const BYTE* iLimit,
- size_t* offsetPtr,
- U32 maxNbAttempts, U32 matchLengthSearch);
- searchMax_f searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS_extDict : ZSTD_HcFindBestMatch_extDict_selectMLS;
-
- U32 offset_1 = seqStorePtr->rep[0], offset_2 = seqStorePtr->rep[1];
-
- /* init */
- ctx->nextToUpdate3 = ctx->nextToUpdate;
- ip += (ip == prefixStart);
-
- /* Match Loop */
- while (ip < ilimit) {
- size_t matchLength=0;
- size_t offset=0;
- const BYTE* start=ip+1;
- U32 current = (U32)(ip-base);
-
- /* check repCode */
- { const U32 repIndex = (U32)(current+1 - offset_1);
- const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* const repMatch = repBase + repIndex;
- if (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex > lowestIndex)) /* intentional overflow */
- if (MEM_read32(ip+1) == MEM_read32(repMatch)) {
- /* repcode detected we should take it */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- matchLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repEnd, prefixStart) + 4;
- if (depth==0) goto _storeSequence;
- } }
-
- /* first search (depth 0) */
- { size_t offsetFound = 99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offsetFound, maxSearches, mls);
- if (ml2 > matchLength)
- matchLength = ml2, start = ip, offset=offsetFound;
- }
-
- if (matchLength < 4) {
- ip += ((ip-anchor) >> g_searchStrength) + 1; /* jump faster over incompressible sections */
- continue;
- }
-
- /* let's try to find a better solution */
- if (depth>=1)
- while (ip= 3) & (repIndex > lowestIndex)) /* intentional overflow */
- if (MEM_read32(ip) == MEM_read32(repMatch)) {
- /* repcode detected */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
- int const gain2 = (int)(repLength * 3);
- int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
- if ((repLength >= 4) && (gain2 > gain1))
- matchLength = repLength, offset = 0, start = ip;
- } }
-
- /* search match, depth 1 */
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
- int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
- if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
- continue; /* search a better one */
- } }
-
- /* let's find an even better one */
- if ((depth==2) && (ip= 3) & (repIndex > lowestIndex)) /* intentional overflow */
- if (MEM_read32(ip) == MEM_read32(repMatch)) {
- /* repcode detected */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
- int const gain2 = (int)(repLength * 4);
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
- if ((repLength >= 4) && (gain2 > gain1))
- matchLength = repLength, offset = 0, start = ip;
- } }
-
- /* search match, depth 2 */
- { size_t offset2=99999999;
- size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
- int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
- if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
- continue;
- } } }
- break; /* nothing found : store previous solution */
- }
-
- /* catch up */
- if (offset) {
- U32 const matchIndex = (U32)((start-base) - (offset - ZSTD_REP_MOVE));
- const BYTE* match = (matchIndex < dictLimit) ? dictBase + matchIndex : base + matchIndex;
- const BYTE* const mStart = (matchIndex < dictLimit) ? dictStart : prefixStart;
- while ((start>anchor) && (match>mStart) && (start[-1] == match[-1])) { start--; match--; matchLength++; } /* catch up */
- offset_2 = offset_1; offset_1 = (U32)(offset - ZSTD_REP_MOVE);
- }
-
- /* store sequence */
-_storeSequence:
- { size_t const litLength = start - anchor;
- ZSTD_storeSeq(seqStorePtr, litLength, anchor, (U32)offset, matchLength-MINMATCH);
- anchor = ip = start + matchLength;
- }
-
- /* check immediate repcode */
- while (ip <= ilimit) {
- const U32 repIndex = (U32)((ip-base) - offset_2);
- const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
- const BYTE* const repMatch = repBase + repIndex;
- if (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex > lowestIndex)) /* intentional overflow */
- if (MEM_read32(ip) == MEM_read32(repMatch)) {
- /* repcode detected we should take it */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
- offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap offset history */
- ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, matchLength-MINMATCH);
- ip += matchLength;
- anchor = ip;
- continue; /* faster when present ... (?) */
- }
- break;
- } }
-
- /* Save reps for next block */
- seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-
-size_t ZSTD_compressBlock_greedy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 0);
-}
-
-static size_t ZSTD_compressBlock_lazy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 1);
-}
-
-static size_t ZSTD_compressBlock_lazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 2);
-}
-
-static size_t ZSTD_compressBlock_btlazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
- return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 1, 2);
-}
-
-
-/* The optimal parser */
-#include "zstd_opt.h"
-
-static size_t ZSTD_compressBlock_btopt(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
-#ifdef ZSTD_OPT_H_91842398743
- return ZSTD_compressBlock_opt_generic(ctx, src, srcSize, 0);
-#else
- (void)ctx; (void)src; (void)srcSize;
- return 0;
-#endif
-}
-
-static size_t ZSTD_compressBlock_btultra(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
-#ifdef ZSTD_OPT_H_91842398743
- return ZSTD_compressBlock_opt_generic(ctx, src, srcSize, 1);
-#else
- (void)ctx; (void)src; (void)srcSize;
- return 0;
-#endif
-}
-
-static size_t ZSTD_compressBlock_btopt_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
-#ifdef ZSTD_OPT_H_91842398743
- return ZSTD_compressBlock_opt_extDict_generic(ctx, src, srcSize, 0);
-#else
- (void)ctx; (void)src; (void)srcSize;
- return 0;
-#endif
-}
-
-static size_t ZSTD_compressBlock_btultra_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
-{
-#ifdef ZSTD_OPT_H_91842398743
- return ZSTD_compressBlock_opt_extDict_generic(ctx, src, srcSize, 1);
-#else
- (void)ctx; (void)src; (void)srcSize;
- return 0;
-#endif
-}
-
/* ZSTD_selectBlockCompressor() :
* assumption : strat is a valid strategy */
typedef size_t (*ZSTD_blockCompressor) (ZSTD_CCtx* ctx, const void* src, size_t srcSize);
diff --git a/lib/compress/zstd_compress.h b/lib/compress/zstd_compress.h
new file mode 100644
index 000000000..be208d614
--- /dev/null
+++ b/lib/compress/zstd_compress.h
@@ -0,0 +1,302 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+
+#ifndef ZSTD_COMPRESS_H
+#define ZSTD_COMPRESS_H
+
+/*-*************************************
+* Dependencies
+***************************************/
+#include "zstd_internal.h"
+#include "zstdmt_compress.h"
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+/*-*************************************
+* Constants
+***************************************/
+static const U32 g_searchStrength = 8;
+#define HASH_READ_SIZE 8
+
+
+/*-*************************************
+* Context memory management
+***************************************/
+typedef enum { ZSTDcs_created=0, ZSTDcs_init, ZSTDcs_ongoing, ZSTDcs_ending } ZSTD_compressionStage_e;
+typedef enum { zcss_init=0, zcss_load, zcss_flush } ZSTD_cStreamStage;
+
+typedef struct ZSTD_prefixDict_s {
+ const void* dict;
+ size_t dictSize;
+ ZSTD_dictMode_e dictMode;
+} ZSTD_prefixDict;
+
+struct ZSTD_CCtx_s {
+ const BYTE* nextSrc; /* next block here to continue on current prefix */
+ const BYTE* base; /* All regular indexes relative to this position */
+ const BYTE* dictBase; /* extDict indexes relative to this position */
+ U32 dictLimit; /* below that point, need extDict */
+ U32 lowLimit; /* below that point, no more data */
+ U32 nextToUpdate; /* index from which to continue dictionary update */
+ U32 nextToUpdate3; /* index from which to continue dictionary update */
+ U32 hashLog3; /* dispatch table : larger == faster, more memory */
+ U32 loadedDictEnd; /* index of end of dictionary */
+ ZSTD_compressionStage_e stage;
+ U32 dictID;
+ ZSTD_CCtx_params requestedParams;
+ ZSTD_CCtx_params appliedParams;
+ void* workSpace;
+ size_t workSpaceSize;
+ size_t blockSize;
+ U64 pledgedSrcSizePlusOne; /* this way, 0 (default) == unknown */
+ U64 consumedSrcSize;
+ XXH64_state_t xxhState;
+ ZSTD_customMem customMem;
+ size_t staticSize;
+
+ seqStore_t seqStore; /* sequences storage ptrs */
+ optState_t optState;
+ ldmState_t ldmState; /* long distance matching state */
+ U32* hashTable;
+ U32* hashTable3;
+ U32* chainTable;
+ ZSTD_entropyCTables_t* entropy;
+
+ /* streaming */
+ char* inBuff;
+ size_t inBuffSize;
+ size_t inToCompress;
+ size_t inBuffPos;
+ size_t inBuffTarget;
+ char* outBuff;
+ size_t outBuffSize;
+ size_t outBuffContentSize;
+ size_t outBuffFlushedSize;
+ ZSTD_cStreamStage streamStage;
+ U32 frameEnded;
+
+ /* Dictionary */
+ ZSTD_CDict* cdictLocal;
+ const ZSTD_CDict* cdict;
+ ZSTD_prefixDict prefixDict; /* single-usage dictionary */
+
+ /* Multi-threading */
+ ZSTDMT_CCtx* mtctx;
+};
+
+
+static const BYTE LL_Code[64] = { 0, 1, 2, 3, 4, 5, 6, 7,
+ 8, 9, 10, 11, 12, 13, 14, 15,
+ 16, 16, 17, 17, 18, 18, 19, 19,
+ 20, 20, 20, 20, 21, 21, 21, 21,
+ 22, 22, 22, 22, 22, 22, 22, 22,
+ 23, 23, 23, 23, 23, 23, 23, 23,
+ 24, 24, 24, 24, 24, 24, 24, 24,
+ 24, 24, 24, 24, 24, 24, 24, 24 };
+
+static const BYTE ML_Code[128] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
+ 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31,
+ 32, 32, 33, 33, 34, 34, 35, 35, 36, 36, 36, 36, 37, 37, 37, 37,
+ 38, 38, 38, 38, 38, 38, 38, 38, 39, 39, 39, 39, 39, 39, 39, 39,
+ 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40,
+ 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41,
+ 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42,
+ 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42 };
+
+/*! ZSTD_storeSeq() :
+ Store a sequence (literal length, literals, offset code and match length code) into seqStore_t.
+ `offsetCode` : distance to match, or 0 == repCode.
+ `matchCode` : matchLength - MINMATCH
+*/
+MEM_STATIC void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const void* literals, U32 offsetCode, size_t matchCode)
+{
+#if defined(ZSTD_DEBUG) && (ZSTD_DEBUG >= 6)
+ static const BYTE* g_start = NULL;
+ U32 const pos = (U32)((const BYTE*)literals - g_start);
+ if (g_start==NULL) g_start = (const BYTE*)literals;
+ if ((pos > 0) && (pos < 1000000000))
+ DEBUGLOG(6, "Cpos %6u :%5u literals & match %3u bytes at distance %6u",
+ pos, (U32)litLength, (U32)matchCode+MINMATCH, (U32)offsetCode);
+#endif
+ /* copy Literals */
+ assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + 128 KB);
+ ZSTD_wildcopy(seqStorePtr->lit, literals, litLength);
+ seqStorePtr->lit += litLength;
+
+ /* literal Length */
+ if (litLength>0xFFFF) {
+ seqStorePtr->longLengthID = 1;
+ seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
+ }
+ seqStorePtr->sequences[0].litLength = (U16)litLength;
+
+ /* match offset */
+ seqStorePtr->sequences[0].offset = offsetCode + 1;
+
+ /* match Length */
+ if (matchCode>0xFFFF) {
+ seqStorePtr->longLengthID = 2;
+ seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
+ }
+ seqStorePtr->sequences[0].matchLength = (U16)matchCode;
+
+ seqStorePtr->sequences++;
+}
+
+
+/*-*************************************
+* Match length counter
+***************************************/
+static unsigned ZSTD_NbCommonBytes (register size_t val)
+{
+ if (MEM_isLittleEndian()) {
+ if (MEM_64bits()) {
+# if defined(_MSC_VER) && defined(_WIN64)
+ unsigned long r = 0;
+ _BitScanForward64( &r, (U64)val );
+ return (unsigned)(r>>3);
+# elif defined(__GNUC__) && (__GNUC__ >= 3)
+ return (__builtin_ctzll((U64)val) >> 3);
+# else
+ static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2,
+ 0, 3, 1, 3, 1, 4, 2, 7,
+ 0, 2, 3, 6, 1, 5, 3, 5,
+ 1, 3, 4, 4, 2, 5, 6, 7,
+ 7, 0, 1, 2, 3, 3, 4, 6,
+ 2, 6, 5, 5, 3, 4, 5, 6,
+ 7, 1, 2, 4, 6, 4, 4, 5,
+ 7, 2, 6, 5, 7, 6, 7, 7 };
+ return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+# endif
+ } else { /* 32 bits */
+# if defined(_MSC_VER)
+ unsigned long r=0;
+ _BitScanForward( &r, (U32)val );
+ return (unsigned)(r>>3);
+# elif defined(__GNUC__) && (__GNUC__ >= 3)
+ return (__builtin_ctz((U32)val) >> 3);
+# else
+ static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0,
+ 3, 2, 2, 1, 3, 2, 0, 1,
+ 3, 3, 1, 2, 2, 2, 2, 0,
+ 3, 1, 2, 0, 1, 0, 1, 1 };
+ return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+# endif
+ }
+ } else { /* Big Endian CPU */
+ if (MEM_64bits()) {
+# if defined(_MSC_VER) && defined(_WIN64)
+ unsigned long r = 0;
+ _BitScanReverse64( &r, val );
+ return (unsigned)(r>>3);
+# elif defined(__GNUC__) && (__GNUC__ >= 3)
+ return (__builtin_clzll(val) >> 3);
+# else
+ unsigned r;
+ const unsigned n32 = sizeof(size_t)*4; /* calculate this way due to compiler complaining in 32-bits mode */
+ if (!(val>>n32)) { r=4; } else { r=0; val>>=n32; }
+ if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
+ r += (!val);
+ return r;
+# endif
+ } else { /* 32 bits */
+# if defined(_MSC_VER)
+ unsigned long r = 0;
+ _BitScanReverse( &r, (unsigned long)val );
+ return (unsigned)(r>>3);
+# elif defined(__GNUC__) && (__GNUC__ >= 3)
+ return (__builtin_clz((U32)val) >> 3);
+# else
+ unsigned r;
+ if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+ r += (!val);
+ return r;
+# endif
+ } }
+}
+
+
+MEM_STATIC size_t ZSTD_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* const pInLimit)
+{
+ const BYTE* const pStart = pIn;
+ const BYTE* const pInLoopLimit = pInLimit - (sizeof(size_t)-1);
+
+ while (pIn < pInLoopLimit) {
+ size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn);
+ if (!diff) { pIn+=sizeof(size_t); pMatch+=sizeof(size_t); continue; }
+ pIn += ZSTD_NbCommonBytes(diff);
+ return (size_t)(pIn - pStart);
+ }
+ if (MEM_64bits()) if ((pIn<(pInLimit-3)) && (MEM_read32(pMatch) == MEM_read32(pIn))) { pIn+=4; pMatch+=4; }
+ if ((pIn<(pInLimit-1)) && (MEM_read16(pMatch) == MEM_read16(pIn))) { pIn+=2; pMatch+=2; }
+ if ((pIn> (32-h) ; }
+MEM_STATIC size_t ZSTD_hash3Ptr(const void* ptr, U32 h) { return ZSTD_hash3(MEM_readLE32(ptr), h); } /* only in zstd_opt.h */
+
+static const U32 prime4bytes = 2654435761U;
+static U32 ZSTD_hash4(U32 u, U32 h) { return (u * prime4bytes) >> (32-h) ; }
+static size_t ZSTD_hash4Ptr(const void* ptr, U32 h) { return ZSTD_hash4(MEM_read32(ptr), h); }
+
+static const U64 prime5bytes = 889523592379ULL;
+static size_t ZSTD_hash5(U64 u, U32 h) { return (size_t)(((u << (64-40)) * prime5bytes) >> (64-h)) ; }
+static size_t ZSTD_hash5Ptr(const void* p, U32 h) { return ZSTD_hash5(MEM_readLE64(p), h); }
+
+static const U64 prime6bytes = 227718039650203ULL;
+static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u << (64-48)) * prime6bytes) >> (64-h)) ; }
+static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); }
+
+static const U64 prime7bytes = 58295818150454627ULL;
+static size_t ZSTD_hash7(U64 u, U32 h) { return (size_t)(((u << (64-56)) * prime7bytes) >> (64-h)) ; }
+static size_t ZSTD_hash7Ptr(const void* p, U32 h) { return ZSTD_hash7(MEM_readLE64(p), h); }
+
+static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL;
+static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; }
+static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); }
+
+MEM_STATIC size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls)
+{
+ switch(mls)
+ {
+ default:
+ case 4: return ZSTD_hash4Ptr(p, hBits);
+ case 5: return ZSTD_hash5Ptr(p, hBits);
+ case 6: return ZSTD_hash6Ptr(p, hBits);
+ case 7: return ZSTD_hash7Ptr(p, hBits);
+ case 8: return ZSTD_hash8Ptr(p, hBits);
+ }
+}
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* ZSTD_COMPRESS_H */
diff --git a/lib/compress/zstd_double_fast.c b/lib/compress/zstd_double_fast.c
new file mode 100644
index 000000000..323736bf2
--- /dev/null
+++ b/lib/compress/zstd_double_fast.c
@@ -0,0 +1,307 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#include "zstd_double_fast.h"
+
+
+void ZSTD_fillDoubleHashTable(ZSTD_CCtx* cctx, const void* end, const U32 mls)
+{
+ U32* const hashLarge = cctx->hashTable;
+ U32 const hBitsL = cctx->appliedParams.cParams.hashLog;
+ U32* const hashSmall = cctx->chainTable;
+ U32 const hBitsS = cctx->appliedParams.cParams.chainLog;
+ const BYTE* const base = cctx->base;
+ const BYTE* ip = base + cctx->nextToUpdate;
+ const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
+ const size_t fastHashFillStep = 3;
+
+ while(ip <= iend) {
+ hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip - base);
+ hashLarge[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip - base);
+ ip += fastHashFillStep;
+ }
+}
+
+
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_doubleFast_generic(ZSTD_CCtx* cctx,
+ const void* src, size_t srcSize,
+ const U32 mls)
+{
+ U32* const hashLong = cctx->hashTable;
+ const U32 hBitsL = cctx->appliedParams.cParams.hashLog;
+ U32* const hashSmall = cctx->chainTable;
+ const U32 hBitsS = cctx->appliedParams.cParams.chainLog;
+ seqStore_t* seqStorePtr = &(cctx->seqStore);
+ const BYTE* const base = cctx->base;
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const U32 lowestIndex = cctx->dictLimit;
+ const BYTE* const lowest = base + lowestIndex;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - HASH_READ_SIZE;
+ U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
+ U32 offsetSaved = 0;
+
+ /* init */
+ ip += (ip==lowest);
+ { U32 const maxRep = (U32)(ip-lowest);
+ if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
+ if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
+ }
+
+ /* Main Search Loop */
+ while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
+ size_t mLength;
+ size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8);
+ size_t const h = ZSTD_hashPtr(ip, hBitsS, mls);
+ U32 const current = (U32)(ip-base);
+ U32 const matchIndexL = hashLong[h2];
+ U32 const matchIndexS = hashSmall[h];
+ const BYTE* matchLong = base + matchIndexL;
+ const BYTE* match = base + matchIndexS;
+ hashLong[h2] = hashSmall[h] = current; /* update hash tables */
+
+ assert(offset_1 <= current); /* supposed guaranteed by construction */
+ if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
+ /* favor repcode */
+ mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
+ ip++;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
+ } else {
+ U32 offset;
+ if ( (matchIndexL > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip)) ) {
+ mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8;
+ offset = (U32)(ip-matchLong);
+ while (((ip>anchor) & (matchLong>lowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
+ } else if ( (matchIndexS > lowestIndex) && (MEM_read32(match) == MEM_read32(ip)) ) {
+ size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
+ U32 const matchIndexL3 = hashLong[hl3];
+ const BYTE* matchL3 = base + matchIndexL3;
+ hashLong[hl3] = current + 1;
+ if ( (matchIndexL3 > lowestIndex) && (MEM_read64(matchL3) == MEM_read64(ip+1)) ) {
+ mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8;
+ ip++;
+ offset = (U32)(ip-matchL3);
+ while (((ip>anchor) & (matchL3>lowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */
+ } else {
+ mLength = ZSTD_count(ip+4, match+4, iend) + 4;
+ offset = (U32)(ip-match);
+ while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ }
+ } else {
+ ip += ((ip-anchor) >> g_searchStrength) + 1;
+ continue;
+ }
+
+ offset_2 = offset_1;
+ offset_1 = offset;
+
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ }
+
+ /* match found */
+ ip += mLength;
+ anchor = ip;
+
+ if (ip <= ilimit) {
+ /* Fill Table */
+ hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] =
+ hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2; /* here because current+2 could be > iend-8 */
+ hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] =
+ hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
+
+ /* check immediate repcode */
+ while ( (ip <= ilimit)
+ && ( (offset_2>0)
+ & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
+ /* store sequence */
+ size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+ { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
+ hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base);
+ hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base);
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, rLength-MINMATCH);
+ ip += rLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ } } }
+
+ /* save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : offsetSaved;
+ seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : offsetSaved;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_doubleFast(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ const U32 mls = ctx->appliedParams.cParams.searchLength;
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 4);
+ case 5 :
+ return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 5);
+ case 6 :
+ return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 6);
+ case 7 :
+ return ZSTD_compressBlock_doubleFast_generic(ctx, src, srcSize, 7);
+ }
+}
+
+
+static size_t ZSTD_compressBlock_doubleFast_extDict_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize,
+ const U32 mls)
+{
+ U32* const hashLong = ctx->hashTable;
+ U32 const hBitsL = ctx->appliedParams.cParams.hashLog;
+ U32* const hashSmall = ctx->chainTable;
+ U32 const hBitsS = ctx->appliedParams.cParams.chainLog;
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ const BYTE* const base = ctx->base;
+ const BYTE* const dictBase = ctx->dictBase;
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const U32 lowestIndex = ctx->lowLimit;
+ const BYTE* const dictStart = dictBase + lowestIndex;
+ const U32 dictLimit = ctx->dictLimit;
+ const BYTE* const lowPrefixPtr = base + dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
+
+ /* Search Loop */
+ while (ip < ilimit) { /* < instead of <=, because (ip+1) */
+ const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls);
+ const U32 matchIndex = hashSmall[hSmall];
+ const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
+ const BYTE* match = matchBase + matchIndex;
+
+ const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8);
+ const U32 matchLongIndex = hashLong[hLong];
+ const BYTE* matchLongBase = matchLongIndex < dictLimit ? dictBase : base;
+ const BYTE* matchLong = matchLongBase + matchLongIndex;
+
+ const U32 current = (U32)(ip-base);
+ const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
+ const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
+ const BYTE* repMatch = repBase + repIndex;
+ size_t mLength;
+ hashSmall[hSmall] = hashLong[hLong] = current; /* update hash table */
+
+ if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
+ ip++;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
+ } else {
+ if ((matchLongIndex > lowestIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
+ const BYTE* matchEnd = matchLongIndex < dictLimit ? dictEnd : iend;
+ const BYTE* lowMatchPtr = matchLongIndex < dictLimit ? dictStart : lowPrefixPtr;
+ U32 offset;
+ mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, lowPrefixPtr) + 8;
+ offset = current - matchLongIndex;
+ while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
+ offset_2 = offset_1;
+ offset_1 = offset;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+
+ } else if ((matchIndex > lowestIndex) && (MEM_read32(match) == MEM_read32(ip))) {
+ size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
+ U32 const matchIndex3 = hashLong[h3];
+ const BYTE* const match3Base = matchIndex3 < dictLimit ? dictBase : base;
+ const BYTE* match3 = match3Base + matchIndex3;
+ U32 offset;
+ hashLong[h3] = current + 1;
+ if ( (matchIndex3 > lowestIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) {
+ const BYTE* matchEnd = matchIndex3 < dictLimit ? dictEnd : iend;
+ const BYTE* lowMatchPtr = matchIndex3 < dictLimit ? dictStart : lowPrefixPtr;
+ mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, lowPrefixPtr) + 8;
+ ip++;
+ offset = current+1 - matchIndex3;
+ while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */
+ } else {
+ const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
+ const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
+ mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
+ offset = current - matchIndex;
+ while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ }
+ offset_2 = offset_1;
+ offset_1 = offset;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+
+ } else {
+ ip += ((ip-anchor) >> g_searchStrength) + 1;
+ continue;
+ } }
+
+ /* found a match : store it */
+ ip += mLength;
+ anchor = ip;
+
+ if (ip <= ilimit) {
+ /* Fill Table */
+ hashSmall[ZSTD_hashPtr(base+current+2, hBitsS, mls)] = current+2;
+ hashLong[ZSTD_hashPtr(base+current+2, hBitsL, 8)] = current+2;
+ hashSmall[ZSTD_hashPtr(ip-2, hBitsS, mls)] = (U32)(ip-2-base);
+ hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base);
+ /* check immediate repcode */
+ while (ip <= ilimit) {
+ U32 const current2 = (U32)(ip-base);
+ U32 const repIndex2 = current2 - offset_2;
+ const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
+ if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
+ && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
+ U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, repLength2-MINMATCH);
+ hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
+ hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
+ ip += repLength2;
+ anchor = ip;
+ continue;
+ }
+ break;
+ } } }
+
+ /* save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_doubleFast_extDict(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize)
+{
+ U32 const mls = ctx->appliedParams.cParams.searchLength;
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 4);
+ case 5 :
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 5);
+ case 6 :
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 6);
+ case 7 :
+ return ZSTD_compressBlock_doubleFast_extDict_generic(ctx, src, srcSize, 7);
+ }
+}
diff --git a/lib/compress/zstd_double_fast.h b/lib/compress/zstd_double_fast.h
new file mode 100644
index 000000000..872939913
--- /dev/null
+++ b/lib/compress/zstd_double_fast.h
@@ -0,0 +1,27 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#ifndef ZSTD_DOUBLE_FAST_H
+#define ZSTD_DOUBLE_FAST_H
+
+#include "zstd_compress.h"
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+void ZSTD_fillDoubleHashTable(ZSTD_CCtx* cctx, const void* end, const U32 mls);
+size_t ZSTD_compressBlock_doubleFast(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_doubleFast_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* ZSTD_DOUBLE_FAST_H */
diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
new file mode 100644
index 000000000..3849d7b2a
--- /dev/null
+++ b/lib/compress/zstd_fast.c
@@ -0,0 +1,241 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#include "zstd_fast.h"
+
+
+void ZSTD_fillHashTable (ZSTD_CCtx* zc, const void* end, const U32 mls)
+{
+ U32* const hashTable = zc->hashTable;
+ U32 const hBits = zc->appliedParams.cParams.hashLog;
+ const BYTE* const base = zc->base;
+ const BYTE* ip = base + zc->nextToUpdate;
+ const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE;
+ const size_t fastHashFillStep = 3;
+
+ while(ip <= iend) {
+ hashTable[ZSTD_hashPtr(ip, hBits, mls)] = (U32)(ip - base);
+ ip += fastHashFillStep;
+ }
+}
+
+
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_fast_generic(ZSTD_CCtx* cctx,
+ const void* src, size_t srcSize,
+ const U32 mls)
+{
+ U32* const hashTable = cctx->hashTable;
+ U32 const hBits = cctx->appliedParams.cParams.hashLog;
+ seqStore_t* seqStorePtr = &(cctx->seqStore);
+ const BYTE* const base = cctx->base;
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const U32 lowestIndex = cctx->dictLimit;
+ const BYTE* const lowest = base + lowestIndex;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - HASH_READ_SIZE;
+ U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
+ U32 offsetSaved = 0;
+
+ /* init */
+ ip += (ip==lowest);
+ { U32 const maxRep = (U32)(ip-lowest);
+ if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0;
+ if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0;
+ }
+
+ /* Main Search Loop */
+ while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */
+ size_t mLength;
+ size_t const h = ZSTD_hashPtr(ip, hBits, mls);
+ U32 const current = (U32)(ip-base);
+ U32 const matchIndex = hashTable[h];
+ const BYTE* match = base + matchIndex;
+ hashTable[h] = current; /* update hash table */
+
+ if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) {
+ mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
+ ip++;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
+ } else {
+ U32 offset;
+ if ( (matchIndex <= lowestIndex) || (MEM_read32(match) != MEM_read32(ip)) ) {
+ ip += ((ip-anchor) >> g_searchStrength) + 1;
+ continue;
+ }
+ mLength = ZSTD_count(ip+4, match+4, iend) + 4;
+ offset = (U32)(ip-match);
+ while (((ip>anchor) & (match>lowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ offset_2 = offset_1;
+ offset_1 = offset;
+
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ }
+
+ /* match found */
+ ip += mLength;
+ anchor = ip;
+
+ if (ip <= ilimit) {
+ /* Fill Table */
+ hashTable[ZSTD_hashPtr(base+current+2, hBits, mls)] = current+2; /* here because current+2 could be > iend-8 */
+ hashTable[ZSTD_hashPtr(ip-2, hBits, mls)] = (U32)(ip-2-base);
+ /* check immediate repcode */
+ while ( (ip <= ilimit)
+ && ( (offset_2>0)
+ & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
+ /* store sequence */
+ size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+ { U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; } /* swap offset_2 <=> offset_1 */
+ hashTable[ZSTD_hashPtr(ip, hBits, mls)] = (U32)(ip-base);
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, rLength-MINMATCH);
+ ip += rLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ } } }
+
+ /* save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : offsetSaved;
+ seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : offsetSaved;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_fast(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize)
+{
+ const U32 mls = ctx->appliedParams.cParams.searchLength;
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 4);
+ case 5 :
+ return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 5);
+ case 6 :
+ return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 6);
+ case 7 :
+ return ZSTD_compressBlock_fast_generic(ctx, src, srcSize, 7);
+ }
+}
+
+
+static size_t ZSTD_compressBlock_fast_extDict_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize,
+ const U32 mls)
+{
+ U32* hashTable = ctx->hashTable;
+ const U32 hBits = ctx->appliedParams.cParams.hashLog;
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ const BYTE* const base = ctx->base;
+ const BYTE* const dictBase = ctx->dictBase;
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const U32 lowestIndex = ctx->lowLimit;
+ const BYTE* const dictStart = dictBase + lowestIndex;
+ const U32 dictLimit = ctx->dictLimit;
+ const BYTE* const lowPrefixPtr = base + dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ U32 offset_1=seqStorePtr->rep[0], offset_2=seqStorePtr->rep[1];
+
+ /* Search Loop */
+ while (ip < ilimit) { /* < instead of <=, because (ip+1) */
+ const size_t h = ZSTD_hashPtr(ip, hBits, mls);
+ const U32 matchIndex = hashTable[h];
+ const BYTE* matchBase = matchIndex < dictLimit ? dictBase : base;
+ const BYTE* match = matchBase + matchIndex;
+ const U32 current = (U32)(ip-base);
+ const U32 repIndex = current + 1 - offset_1; /* offset_1 expected <= current +1 */
+ const BYTE* repBase = repIndex < dictLimit ? dictBase : base;
+ const BYTE* repMatch = repBase + repIndex;
+ size_t mLength;
+ hashTable[h] = current; /* update hash table */
+
+ if ( (((U32)((dictLimit-1) - repIndex) >= 3) /* intentional underflow */ & (repIndex > lowestIndex))
+ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) {
+ const BYTE* repMatchEnd = repIndex < dictLimit ? dictEnd : iend;
+ mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, lowPrefixPtr) + 4;
+ ip++;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, 0, mLength-MINMATCH);
+ } else {
+ if ( (matchIndex < lowestIndex) ||
+ (MEM_read32(match) != MEM_read32(ip)) ) {
+ ip += ((ip-anchor) >> g_searchStrength) + 1;
+ continue;
+ }
+ { const BYTE* matchEnd = matchIndex < dictLimit ? dictEnd : iend;
+ const BYTE* lowMatchPtr = matchIndex < dictLimit ? dictStart : lowPrefixPtr;
+ U32 offset;
+ mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, lowPrefixPtr) + 4;
+ while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
+ offset = current - matchIndex;
+ offset_2 = offset_1;
+ offset_1 = offset;
+ ZSTD_storeSeq(seqStorePtr, ip-anchor, anchor, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ } }
+
+ /* found a match : store it */
+ ip += mLength;
+ anchor = ip;
+
+ if (ip <= ilimit) {
+ /* Fill Table */
+ hashTable[ZSTD_hashPtr(base+current+2, hBits, mls)] = current+2;
+ hashTable[ZSTD_hashPtr(ip-2, hBits, mls)] = (U32)(ip-2-base);
+ /* check immediate repcode */
+ while (ip <= ilimit) {
+ U32 const current2 = (U32)(ip-base);
+ U32 const repIndex2 = current2 - offset_2;
+ const BYTE* repMatch2 = repIndex2 < dictLimit ? dictBase + repIndex2 : base + repIndex2;
+ if ( (((U32)((dictLimit-1) - repIndex2) >= 3) & (repIndex2 > lowestIndex)) /* intentional overflow */
+ && (MEM_read32(repMatch2) == MEM_read32(ip)) ) {
+ const BYTE* const repEnd2 = repIndex2 < dictLimit ? dictEnd : iend;
+ size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, lowPrefixPtr) + 4;
+ U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, repLength2-MINMATCH);
+ hashTable[ZSTD_hashPtr(ip, hBits, mls)] = current2;
+ ip += repLength2;
+ anchor = ip;
+ continue;
+ }
+ break;
+ } } }
+
+ /* save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_fast_extDict(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize)
+{
+ U32 const mls = ctx->appliedParams.cParams.searchLength;
+ switch(mls)
+ {
+ default: /* includes case 3 */
+ case 4 :
+ return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 4);
+ case 5 :
+ return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 5);
+ case 6 :
+ return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 6);
+ case 7 :
+ return ZSTD_compressBlock_fast_extDict_generic(ctx, src, srcSize, 7);
+ }
+}
diff --git a/lib/compress/zstd_fast.h b/lib/compress/zstd_fast.h
new file mode 100644
index 000000000..ab7a65a91
--- /dev/null
+++ b/lib/compress/zstd_fast.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#ifndef ZSTD_FAST_H
+#define ZSTD_FAST_H
+
+#include "zstd_compress.h"
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+void ZSTD_fillHashTable(ZSTD_CCtx* zc, const void* end, const U32 mls);
+size_t ZSTD_compressBlock_fast(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_fast_extDict(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize);
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* ZSTD_FAST_H */
diff --git a/lib/compress/zstd_lazy.c b/lib/compress/zstd_lazy.c
new file mode 100644
index 000000000..ad8b8aaea
--- /dev/null
+++ b/lib/compress/zstd_lazy.c
@@ -0,0 +1,746 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#include "zstd_lazy.h"
+
+
+/*-*************************************
+* Binary Tree search
+***************************************/
+/** ZSTD_insertBt1() : add one or multiple positions to tree.
+* ip : assumed <= iend-8 .
+* @return : nb of positions added */
+static U32 ZSTD_insertBt1(ZSTD_CCtx* zc, const BYTE* const ip, const U32 mls, const BYTE* const iend, U32 nbCompares,
+ U32 extDict)
+{
+ U32* const hashTable = zc->hashTable;
+ U32 const hashLog = zc->appliedParams.cParams.hashLog;
+ size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
+ U32* const bt = zc->chainTable;
+ U32 const btLog = zc->appliedParams.cParams.chainLog - 1;
+ U32 const btMask = (1 << btLog) - 1;
+ U32 matchIndex = hashTable[h];
+ size_t commonLengthSmaller=0, commonLengthLarger=0;
+ const BYTE* const base = zc->base;
+ const BYTE* const dictBase = zc->dictBase;
+ const U32 dictLimit = zc->dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const BYTE* match;
+ const U32 current = (U32)(ip-base);
+ const U32 btLow = btMask >= current ? 0 : current - btMask;
+ U32* smallerPtr = bt + 2*(current&btMask);
+ U32* largerPtr = smallerPtr + 1;
+ U32 dummy32; /* to be nullified at the end */
+ U32 const windowLow = zc->lowLimit;
+ U32 matchEndIdx = current+8;
+ size_t bestLength = 8;
+#ifdef ZSTD_C_PREDICT
+ U32 predictedSmall = *(bt + 2*((current-1)&btMask) + 0);
+ U32 predictedLarge = *(bt + 2*((current-1)&btMask) + 1);
+ predictedSmall += (predictedSmall>0);
+ predictedLarge += (predictedLarge>0);
+#endif /* ZSTD_C_PREDICT */
+
+ hashTable[h] = current; /* Update Hash Table */
+
+ while (nbCompares-- && (matchIndex > windowLow)) {
+ U32* const nextPtr = bt + 2*(matchIndex & btMask);
+ size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
+
+#ifdef ZSTD_C_PREDICT /* note : can create issues when hlog small <= 11 */
+ const U32* predictPtr = bt + 2*((matchIndex-1) & btMask); /* written this way, as bt is a roll buffer */
+ if (matchIndex == predictedSmall) {
+ /* no need to check length, result known */
+ *smallerPtr = matchIndex;
+ if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
+ matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ predictedSmall = predictPtr[1] + (predictPtr[1]>0);
+ continue;
+ }
+ if (matchIndex == predictedLarge) {
+ *largerPtr = matchIndex;
+ if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ largerPtr = nextPtr;
+ matchIndex = nextPtr[0];
+ predictedLarge = predictPtr[0] + (predictPtr[0]>0);
+ continue;
+ }
+#endif
+ if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
+ match = base + matchIndex;
+ if (match[matchLength] == ip[matchLength])
+ matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iend) +1;
+ } else {
+ match = dictBase + matchIndex;
+ matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
+ if (matchIndex+matchLength >= dictLimit)
+ match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
+ }
+
+ if (matchLength > bestLength) {
+ bestLength = matchLength;
+ if (matchLength > matchEndIdx - matchIndex)
+ matchEndIdx = matchIndex + (U32)matchLength;
+ }
+
+ if (ip+matchLength == iend) /* equal : no way to know if inf or sup */
+ break; /* drop , to guarantee consistency ; miss a bit of compression, but other solutions can corrupt the tree */
+
+ if (match[matchLength] < ip[matchLength]) { /* necessarily within correct buffer */
+ /* match is smaller than current */
+ *smallerPtr = matchIndex; /* update smaller idx */
+ commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
+ if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
+ matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ } else {
+ /* match is larger than current */
+ *largerPtr = matchIndex;
+ commonLengthLarger = matchLength;
+ if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ largerPtr = nextPtr;
+ matchIndex = nextPtr[0];
+ } }
+
+ *smallerPtr = *largerPtr = 0;
+ if (bestLength > 384) return MIN(192, (U32)(bestLength - 384)); /* speed optimization */
+ if (matchEndIdx > current + 8) return matchEndIdx - current - 8;
+ return 1;
+}
+
+
+static size_t ZSTD_insertBtAndFindBestMatch (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iend,
+ size_t* offsetPtr,
+ U32 nbCompares, const U32 mls,
+ U32 extDict)
+{
+ U32* const hashTable = zc->hashTable;
+ U32 const hashLog = zc->appliedParams.cParams.hashLog;
+ size_t const h = ZSTD_hashPtr(ip, hashLog, mls);
+ U32* const bt = zc->chainTable;
+ U32 const btLog = zc->appliedParams.cParams.chainLog - 1;
+ U32 const btMask = (1 << btLog) - 1;
+ U32 matchIndex = hashTable[h];
+ size_t commonLengthSmaller=0, commonLengthLarger=0;
+ const BYTE* const base = zc->base;
+ const BYTE* const dictBase = zc->dictBase;
+ const U32 dictLimit = zc->dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const U32 current = (U32)(ip-base);
+ const U32 btLow = btMask >= current ? 0 : current - btMask;
+ const U32 windowLow = zc->lowLimit;
+ U32* smallerPtr = bt + 2*(current&btMask);
+ U32* largerPtr = bt + 2*(current&btMask) + 1;
+ U32 matchEndIdx = current+8;
+ U32 dummy32; /* to be nullified at the end */
+ size_t bestLength = 0;
+
+ hashTable[h] = current; /* Update Hash Table */
+
+ while (nbCompares-- && (matchIndex > windowLow)) {
+ U32* const nextPtr = bt + 2*(matchIndex & btMask);
+ size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
+ const BYTE* match;
+
+ if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
+ match = base + matchIndex;
+ if (match[matchLength] == ip[matchLength])
+ matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iend) +1;
+ } else {
+ match = dictBase + matchIndex;
+ matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart);
+ if (matchIndex+matchLength >= dictLimit)
+ match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
+ }
+
+ if (matchLength > bestLength) {
+ if (matchLength > matchEndIdx - matchIndex)
+ matchEndIdx = matchIndex + (U32)matchLength;
+ if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(current-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) )
+ bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + current - matchIndex;
+ if (ip+matchLength == iend) /* equal : no way to know if inf or sup */
+ break; /* drop, to guarantee consistency (miss a little bit of compression) */
+ }
+
+ if (match[matchLength] < ip[matchLength]) {
+ /* match is smaller than current */
+ *smallerPtr = matchIndex; /* update smaller idx */
+ commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
+ if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
+ matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ } else {
+ /* match is larger than current */
+ *largerPtr = matchIndex;
+ commonLengthLarger = matchLength;
+ if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ largerPtr = nextPtr;
+ matchIndex = nextPtr[0];
+ } }
+
+ *smallerPtr = *largerPtr = 0;
+
+ zc->nextToUpdate = (matchEndIdx > current + 8) ? matchEndIdx - 8 : current+1;
+ return bestLength;
+}
+
+
+void ZSTD_updateTree(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls)
+{
+ const BYTE* const base = zc->base;
+ const U32 target = (U32)(ip - base);
+ U32 idx = zc->nextToUpdate;
+
+ while(idx < target)
+ idx += ZSTD_insertBt1(zc, base+idx, mls, iend, nbCompares, 0);
+}
+
+/** ZSTD_BtFindBestMatch() : Tree updater, providing best match */
+static size_t ZSTD_BtFindBestMatch (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 mls)
+{
+ if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
+ ZSTD_updateTree(zc, ip, iLimit, maxNbAttempts, mls);
+ return ZSTD_insertBtAndFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, mls, 0);
+}
+
+
+static size_t ZSTD_BtFindBestMatch_selectMLS (
+ ZSTD_CCtx* zc, /* Index table will be updated */
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 matchLengthSearch)
+{
+ switch(matchLengthSearch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4);
+ case 5 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5);
+ case 7 :
+ case 6 : return ZSTD_BtFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6);
+ }
+}
+
+
+void ZSTD_updateTree_extDict(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls)
+{
+ const BYTE* const base = zc->base;
+ const U32 target = (U32)(ip - base);
+ U32 idx = zc->nextToUpdate;
+
+ while (idx < target) idx += ZSTD_insertBt1(zc, base+idx, mls, iend, nbCompares, 1);
+}
+
+
+/** Tree updater, providing best match */
+static size_t ZSTD_BtFindBestMatch_extDict (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 mls)
+{
+ if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
+ ZSTD_updateTree_extDict(zc, ip, iLimit, maxNbAttempts, mls);
+ return ZSTD_insertBtAndFindBestMatch(zc, ip, iLimit, offsetPtr, maxNbAttempts, mls, 1);
+}
+
+
+static size_t ZSTD_BtFindBestMatch_selectMLS_extDict (
+ ZSTD_CCtx* zc, /* Index table will be updated */
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 matchLengthSearch)
+{
+ switch(matchLengthSearch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4);
+ case 5 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5);
+ case 7 :
+ case 6 : return ZSTD_BtFindBestMatch_extDict(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6);
+ }
+}
+
+
+
+/* *********************************
+* Hash Chain
+***********************************/
+#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & mask]
+
+/* Update chains up to ip (excluded)
+ Assumption : always within prefix (i.e. not within extDict) */
+U32 ZSTD_insertAndFindFirstIndex (ZSTD_CCtx* zc, const BYTE* ip, U32 mls)
+{
+ U32* const hashTable = zc->hashTable;
+ const U32 hashLog = zc->appliedParams.cParams.hashLog;
+ U32* const chainTable = zc->chainTable;
+ const U32 chainMask = (1 << zc->appliedParams.cParams.chainLog) - 1;
+ const BYTE* const base = zc->base;
+ const U32 target = (U32)(ip - base);
+ U32 idx = zc->nextToUpdate;
+
+ while(idx < target) { /* catch up */
+ size_t const h = ZSTD_hashPtr(base+idx, hashLog, mls);
+ NEXT_IN_CHAIN(idx, chainMask) = hashTable[h];
+ hashTable[h] = idx;
+ idx++;
+ }
+
+ zc->nextToUpdate = target;
+ return hashTable[ZSTD_hashPtr(ip, hashLog, mls)];
+}
+
+
+/* inlining is important to hardwire a hot branch (template emulation) */
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_HcFindBestMatch_generic (
+ ZSTD_CCtx* zc, /* Index table will be updated */
+ const BYTE* const ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 mls, const U32 extDict)
+{
+ U32* const chainTable = zc->chainTable;
+ const U32 chainSize = (1 << zc->appliedParams.cParams.chainLog);
+ const U32 chainMask = chainSize-1;
+ const BYTE* const base = zc->base;
+ const BYTE* const dictBase = zc->dictBase;
+ const U32 dictLimit = zc->dictLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const U32 lowLimit = zc->lowLimit;
+ const U32 current = (U32)(ip-base);
+ const U32 minChain = current > chainSize ? current - chainSize : 0;
+ int nbAttempts=maxNbAttempts;
+ size_t ml=4-1;
+
+ /* HC4 match finder */
+ U32 matchIndex = ZSTD_insertAndFindFirstIndex (zc, ip, mls);
+
+ for ( ; (matchIndex>lowLimit) & (nbAttempts>0) ; nbAttempts--) {
+ const BYTE* match;
+ size_t currentMl=0;
+ if ((!extDict) || matchIndex >= dictLimit) {
+ match = base + matchIndex;
+ if (match[ml] == ip[ml]) /* potentially better */
+ currentMl = ZSTD_count(ip, match, iLimit);
+ } else {
+ match = dictBase + matchIndex;
+ if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */
+ currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4;
+ }
+
+ /* save best solution */
+ if (currentMl > ml) {
+ ml = currentMl;
+ *offsetPtr = current - matchIndex + ZSTD_REP_MOVE;
+ if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
+ }
+
+ if (matchIndex <= minChain) break;
+ matchIndex = NEXT_IN_CHAIN(matchIndex, chainMask);
+ }
+
+ return ml;
+}
+
+
+FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_selectMLS (
+ ZSTD_CCtx* zc,
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 matchLengthSearch)
+{
+ switch(matchLengthSearch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4, 0);
+ case 5 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5, 0);
+ case 7 :
+ case 6 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6, 0);
+ }
+}
+
+
+FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch_extDict_selectMLS (
+ ZSTD_CCtx* zc,
+ const BYTE* ip, const BYTE* const iLimit,
+ size_t* offsetPtr,
+ const U32 maxNbAttempts, const U32 matchLengthSearch)
+{
+ switch(matchLengthSearch)
+ {
+ default : /* includes case 3 */
+ case 4 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 4, 1);
+ case 5 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 5, 1);
+ case 7 :
+ case 6 : return ZSTD_HcFindBestMatch_generic(zc, ip, iLimit, offsetPtr, maxNbAttempts, 6, 1);
+ }
+}
+
+
+/* *******************************
+* Common parser - lazy strategy
+*********************************/
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_lazy_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize,
+ const U32 searchMethod, const U32 depth)
+{
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ const BYTE* const base = ctx->base + ctx->dictLimit;
+
+ U32 const maxSearches = 1 << ctx->appliedParams.cParams.searchLog;
+ U32 const mls = ctx->appliedParams.cParams.searchLength;
+
+ typedef size_t (*searchMax_f)(ZSTD_CCtx* zc, const BYTE* ip, const BYTE* iLimit,
+ size_t* offsetPtr,
+ U32 maxNbAttempts, U32 matchLengthSearch);
+ searchMax_f const searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS : ZSTD_HcFindBestMatch_selectMLS;
+ U32 offset_1 = seqStorePtr->rep[0], offset_2 = seqStorePtr->rep[1], savedOffset=0;
+
+ /* init */
+ ip += (ip==base);
+ ctx->nextToUpdate3 = ctx->nextToUpdate;
+ { U32 const maxRep = (U32)(ip-base);
+ if (offset_2 > maxRep) savedOffset = offset_2, offset_2 = 0;
+ if (offset_1 > maxRep) savedOffset = offset_1, offset_1 = 0;
+ }
+
+ /* Match Loop */
+ while (ip < ilimit) {
+ size_t matchLength=0;
+ size_t offset=0;
+ const BYTE* start=ip+1;
+
+ /* check repCode */
+ if ((offset_1>0) & (MEM_read32(ip+1) == MEM_read32(ip+1 - offset_1))) {
+ /* repcode : we take it */
+ matchLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4;
+ if (depth==0) goto _storeSequence;
+ }
+
+ /* first search (depth 0) */
+ { size_t offsetFound = 99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offsetFound, maxSearches, mls);
+ if (ml2 > matchLength)
+ matchLength = ml2, start = ip, offset=offsetFound;
+ }
+
+ if (matchLength < 4) {
+ ip += ((ip-anchor) >> g_searchStrength) + 1; /* jump faster over incompressible sections */
+ continue;
+ }
+
+ /* let's try to find a better solution */
+ if (depth>=1)
+ while (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
+ size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
+ int const gain2 = (int)(mlRep * 3);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((mlRep >= 4) && (gain2 > gain1))
+ matchLength = mlRep, offset = 0, start = ip;
+ }
+ { size_t offset2=99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
+ if ((ml2 >= 4) && (gain2 > gain1)) {
+ matchLength = ml2, offset = offset2, start = ip;
+ continue; /* search a better one */
+ } }
+
+ /* let's find an even better one */
+ if ((depth==2) && (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
+ size_t const ml2 = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4;
+ int const gain2 = (int)(ml2 * 4);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((ml2 >= 4) && (gain2 > gain1))
+ matchLength = ml2, offset = 0, start = ip;
+ }
+ { size_t offset2=99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
+ if ((ml2 >= 4) && (gain2 > gain1)) {
+ matchLength = ml2, offset = offset2, start = ip;
+ continue;
+ } } }
+ break; /* nothing found : store previous solution */
+ }
+
+ /* NOTE:
+ * start[-offset+ZSTD_REP_MOVE-1] is undefined behavior.
+ * (-offset+ZSTD_REP_MOVE-1) is unsigned, and is added to start, which
+ * overflows the pointer, which is undefined behavior.
+ */
+ /* catch up */
+ if (offset) {
+ while ( (start > anchor)
+ && (start > base+offset-ZSTD_REP_MOVE)
+ && (start[-1] == (start-offset+ZSTD_REP_MOVE)[-1]) ) /* only search for offset within prefix */
+ { start--; matchLength++; }
+ offset_2 = offset_1; offset_1 = (U32)(offset - ZSTD_REP_MOVE);
+ }
+ /* store sequence */
+_storeSequence:
+ { size_t const litLength = start - anchor;
+ ZSTD_storeSeq(seqStorePtr, litLength, anchor, (U32)offset, matchLength-MINMATCH);
+ anchor = ip = start + matchLength;
+ }
+
+ /* check immediate repcode */
+ while ( (ip <= ilimit)
+ && ((offset_2>0)
+ & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) {
+ /* store sequence */
+ matchLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4;
+ offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap repcodes */
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, matchLength-MINMATCH);
+ ip += matchLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ } }
+
+ /* Save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1 ? offset_1 : savedOffset;
+ seqStorePtr->repToConfirm[1] = offset_2 ? offset_2 : savedOffset;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_btlazy2(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 1, 2);
+}
+
+size_t ZSTD_compressBlock_lazy2(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 2);
+}
+
+size_t ZSTD_compressBlock_lazy(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 1);
+}
+
+size_t ZSTD_compressBlock_greedy(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_generic(ctx, src, srcSize, 0, 0);
+}
+
+
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_lazy_extDict_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize,
+ const U32 searchMethod, const U32 depth)
+{
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ const BYTE* const base = ctx->base;
+ const U32 dictLimit = ctx->dictLimit;
+ const U32 lowestIndex = ctx->lowLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const BYTE* const dictBase = ctx->dictBase;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const dictStart = dictBase + ctx->lowLimit;
+
+ const U32 maxSearches = 1 << ctx->appliedParams.cParams.searchLog;
+ const U32 mls = ctx->appliedParams.cParams.searchLength;
+
+ typedef size_t (*searchMax_f)(ZSTD_CCtx* zc, const BYTE* ip, const BYTE* iLimit,
+ size_t* offsetPtr,
+ U32 maxNbAttempts, U32 matchLengthSearch);
+ searchMax_f searchMax = searchMethod ? ZSTD_BtFindBestMatch_selectMLS_extDict : ZSTD_HcFindBestMatch_extDict_selectMLS;
+
+ U32 offset_1 = seqStorePtr->rep[0], offset_2 = seqStorePtr->rep[1];
+
+ /* init */
+ ctx->nextToUpdate3 = ctx->nextToUpdate;
+ ip += (ip == prefixStart);
+
+ /* Match Loop */
+ while (ip < ilimit) {
+ size_t matchLength=0;
+ size_t offset=0;
+ const BYTE* start=ip+1;
+ U32 current = (U32)(ip-base);
+
+ /* check repCode */
+ { const U32 repIndex = (U32)(current+1 - offset_1);
+ const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
+ const BYTE* const repMatch = repBase + repIndex;
+ if (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex > lowestIndex)) /* intentional overflow */
+ if (MEM_read32(ip+1) == MEM_read32(repMatch)) {
+ /* repcode detected we should take it */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ matchLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repEnd, prefixStart) + 4;
+ if (depth==0) goto _storeSequence;
+ } }
+
+ /* first search (depth 0) */
+ { size_t offsetFound = 99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offsetFound, maxSearches, mls);
+ if (ml2 > matchLength)
+ matchLength = ml2, start = ip, offset=offsetFound;
+ }
+
+ if (matchLength < 4) {
+ ip += ((ip-anchor) >> g_searchStrength) + 1; /* jump faster over incompressible sections */
+ continue;
+ }
+
+ /* let's try to find a better solution */
+ if (depth>=1)
+ while (ip= 3) & (repIndex > lowestIndex)) /* intentional overflow */
+ if (MEM_read32(ip) == MEM_read32(repMatch)) {
+ /* repcode detected */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
+ int const gain2 = (int)(repLength * 3);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((repLength >= 4) && (gain2 > gain1))
+ matchLength = repLength, offset = 0, start = ip;
+ } }
+
+ /* search match, depth 1 */
+ { size_t offset2=99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 4);
+ if ((ml2 >= 4) && (gain2 > gain1)) {
+ matchLength = ml2, offset = offset2, start = ip;
+ continue; /* search a better one */
+ } }
+
+ /* let's find an even better one */
+ if ((depth==2) && (ip= 3) & (repIndex > lowestIndex)) /* intentional overflow */
+ if (MEM_read32(ip) == MEM_read32(repMatch)) {
+ /* repcode detected */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
+ int const gain2 = (int)(repLength * 4);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
+ if ((repLength >= 4) && (gain2 > gain1))
+ matchLength = repLength, offset = 0, start = ip;
+ } }
+
+ /* search match, depth 2 */
+ { size_t offset2=99999999;
+ size_t const ml2 = searchMax(ctx, ip, iend, &offset2, maxSearches, mls);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)offset2+1)); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 7);
+ if ((ml2 >= 4) && (gain2 > gain1)) {
+ matchLength = ml2, offset = offset2, start = ip;
+ continue;
+ } } }
+ break; /* nothing found : store previous solution */
+ }
+
+ /* catch up */
+ if (offset) {
+ U32 const matchIndex = (U32)((start-base) - (offset - ZSTD_REP_MOVE));
+ const BYTE* match = (matchIndex < dictLimit) ? dictBase + matchIndex : base + matchIndex;
+ const BYTE* const mStart = (matchIndex < dictLimit) ? dictStart : prefixStart;
+ while ((start>anchor) && (match>mStart) && (start[-1] == match[-1])) { start--; match--; matchLength++; } /* catch up */
+ offset_2 = offset_1; offset_1 = (U32)(offset - ZSTD_REP_MOVE);
+ }
+
+ /* store sequence */
+_storeSequence:
+ { size_t const litLength = start - anchor;
+ ZSTD_storeSeq(seqStorePtr, litLength, anchor, (U32)offset, matchLength-MINMATCH);
+ anchor = ip = start + matchLength;
+ }
+
+ /* check immediate repcode */
+ while (ip <= ilimit) {
+ const U32 repIndex = (U32)((ip-base) - offset_2);
+ const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
+ const BYTE* const repMatch = repBase + repIndex;
+ if (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex > lowestIndex)) /* intentional overflow */
+ if (MEM_read32(ip) == MEM_read32(repMatch)) {
+ /* repcode detected we should take it */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4;
+ offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap offset history */
+ ZSTD_storeSeq(seqStorePtr, 0, anchor, 0, matchLength-MINMATCH);
+ ip += matchLength;
+ anchor = ip;
+ continue; /* faster when present ... (?) */
+ }
+ break;
+ } }
+
+ /* Save reps for next block */
+ seqStorePtr->repToConfirm[0] = offset_1; seqStorePtr->repToConfirm[1] = offset_2;
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_greedy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 0);
+}
+
+size_t ZSTD_compressBlock_lazy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 1);
+}
+
+size_t ZSTD_compressBlock_lazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 0, 2);
+}
+
+size_t ZSTD_compressBlock_btlazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_lazy_extDict_generic(ctx, src, srcSize, 1, 2);
+}
diff --git a/lib/compress/zstd_lazy.h b/lib/compress/zstd_lazy.h
new file mode 100644
index 000000000..a6c135806
--- /dev/null
+++ b/lib/compress/zstd_lazy.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#ifndef ZSTD_LAZY_H
+#define ZSTD_LAZY_H
+
+#include "zstd_compress.h"
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+U32 ZSTD_insertAndFindFirstIndex (ZSTD_CCtx* zc, const BYTE* ip, U32 mls);
+void ZSTD_updateTree(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls);
+void ZSTD_updateTree_extDict(ZSTD_CCtx* zc, const BYTE* const ip, const BYTE* const iend, const U32 nbCompares, const U32 mls);
+
+size_t ZSTD_compressBlock_btlazy2(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy2(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_greedy(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+
+size_t ZSTD_compressBlock_greedy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_lazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_btlazy2_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+
+#if defined (__cplusplus)
+}
+#endif
+
+#endif /* ZSTD_LAZY_H */
diff --git a/lib/compress/zstd_opt.c b/lib/compress/zstd_opt.c
new file mode 100644
index 000000000..8a71eacdf
--- /dev/null
+++ b/lib/compress/zstd_opt.c
@@ -0,0 +1,948 @@
+/*
+ * Copyright (c) 2016-present, Przemyslaw Skibinski, Yann Collet, Facebook, Inc.
+ * 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).
+ */
+
+#include "zstd_opt.h"
+#include "zstd_lazy.h"
+
+
+#define ZSTD_LITFREQ_ADD 2
+#define ZSTD_FREQ_DIV 4
+#define ZSTD_MAX_PRICE (1<<30)
+
+/*-*************************************
+* Price functions for optimal parser
+***************************************/
+static void ZSTD_setLog2Prices(optState_t* optPtr)
+{
+ optPtr->log2matchLengthSum = ZSTD_highbit32(optPtr->matchLengthSum+1);
+ optPtr->log2litLengthSum = ZSTD_highbit32(optPtr->litLengthSum+1);
+ optPtr->log2litSum = ZSTD_highbit32(optPtr->litSum+1);
+ optPtr->log2offCodeSum = ZSTD_highbit32(optPtr->offCodeSum+1);
+ optPtr->factor = 1 + ((optPtr->litSum>>5) / optPtr->litLengthSum) + ((optPtr->litSum<<1) / (optPtr->litSum + optPtr->matchSum));
+}
+
+
+static void ZSTD_rescaleFreqs(optState_t* optPtr, const BYTE* src, size_t srcSize)
+{
+ unsigned u;
+
+ optPtr->cachedLiterals = NULL;
+ optPtr->cachedPrice = optPtr->cachedLitLength = 0;
+ optPtr->staticPrices = 0;
+
+ if (optPtr->litLengthSum == 0) {
+ if (srcSize <= 1024) optPtr->staticPrices = 1;
+
+ assert(optPtr->litFreq!=NULL);
+ for (u=0; u<=MaxLit; u++)
+ optPtr->litFreq[u] = 0;
+ for (u=0; ulitFreq[src[u]]++;
+
+ optPtr->litSum = 0;
+ optPtr->litLengthSum = MaxLL+1;
+ optPtr->matchLengthSum = MaxML+1;
+ optPtr->offCodeSum = (MaxOff+1);
+ optPtr->matchSum = (ZSTD_LITFREQ_ADD<litFreq[u] = 1 + (optPtr->litFreq[u]>>ZSTD_FREQ_DIV);
+ optPtr->litSum += optPtr->litFreq[u];
+ }
+ for (u=0; u<=MaxLL; u++)
+ optPtr->litLengthFreq[u] = 1;
+ for (u=0; u<=MaxML; u++)
+ optPtr->matchLengthFreq[u] = 1;
+ for (u=0; u<=MaxOff; u++)
+ optPtr->offCodeFreq[u] = 1;
+ } else {
+ optPtr->matchLengthSum = 0;
+ optPtr->litLengthSum = 0;
+ optPtr->offCodeSum = 0;
+ optPtr->matchSum = 0;
+ optPtr->litSum = 0;
+
+ for (u=0; u<=MaxLit; u++) {
+ optPtr->litFreq[u] = 1 + (optPtr->litFreq[u]>>(ZSTD_FREQ_DIV+1));
+ optPtr->litSum += optPtr->litFreq[u];
+ }
+ for (u=0; u<=MaxLL; u++) {
+ optPtr->litLengthFreq[u] = 1 + (optPtr->litLengthFreq[u]>>(ZSTD_FREQ_DIV+1));
+ optPtr->litLengthSum += optPtr->litLengthFreq[u];
+ }
+ for (u=0; u<=MaxML; u++) {
+ optPtr->matchLengthFreq[u] = 1 + (optPtr->matchLengthFreq[u]>>ZSTD_FREQ_DIV);
+ optPtr->matchLengthSum += optPtr->matchLengthFreq[u];
+ optPtr->matchSum += optPtr->matchLengthFreq[u] * (u + 3);
+ }
+ optPtr->matchSum *= ZSTD_LITFREQ_ADD;
+ for (u=0; u<=MaxOff; u++) {
+ optPtr->offCodeFreq[u] = 1 + (optPtr->offCodeFreq[u]>>ZSTD_FREQ_DIV);
+ optPtr->offCodeSum += optPtr->offCodeFreq[u];
+ }
+ }
+
+ ZSTD_setLog2Prices(optPtr);
+}
+
+
+static U32 ZSTD_getLiteralPrice(optState_t* optPtr, U32 litLength, const BYTE* literals)
+{
+ U32 price, u;
+
+ if (optPtr->staticPrices)
+ return ZSTD_highbit32((U32)litLength+1) + (litLength*6);
+
+ if (litLength == 0)
+ return optPtr->log2litLengthSum - ZSTD_highbit32(optPtr->litLengthFreq[0]+1);
+
+ /* literals */
+ if (optPtr->cachedLiterals == literals) {
+ U32 const additional = litLength - optPtr->cachedLitLength;
+ const BYTE* literals2 = optPtr->cachedLiterals + optPtr->cachedLitLength;
+ price = optPtr->cachedPrice + additional * optPtr->log2litSum;
+ for (u=0; u < additional; u++)
+ price -= ZSTD_highbit32(optPtr->litFreq[literals2[u]]+1);
+ optPtr->cachedPrice = price;
+ optPtr->cachedLitLength = litLength;
+ } else {
+ price = litLength * optPtr->log2litSum;
+ for (u=0; u < litLength; u++)
+ price -= ZSTD_highbit32(optPtr->litFreq[literals[u]]+1);
+
+ if (litLength >= 12) {
+ optPtr->cachedLiterals = literals;
+ optPtr->cachedPrice = price;
+ optPtr->cachedLitLength = litLength;
+ }
+ }
+
+ /* literal Length */
+ { const BYTE LL_deltaCode = 19;
+ const BYTE llCode = (litLength>63) ? (BYTE)ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength];
+ price += LL_bits[llCode] + optPtr->log2litLengthSum - ZSTD_highbit32(optPtr->litLengthFreq[llCode]+1);
+ }
+
+ return price;
+}
+
+
+FORCE_INLINE_TEMPLATE U32 ZSTD_getPrice(optState_t* optPtr, U32 litLength, const BYTE* literals, U32 offset, U32 matchLength, const int ultra)
+{
+ /* offset */
+ U32 price;
+ BYTE const offCode = (BYTE)ZSTD_highbit32(offset+1);
+
+ if (optPtr->staticPrices)
+ return ZSTD_getLiteralPrice(optPtr, litLength, literals) + ZSTD_highbit32((U32)matchLength+1) + 16 + offCode;
+
+ price = offCode + optPtr->log2offCodeSum - ZSTD_highbit32(optPtr->offCodeFreq[offCode]+1);
+ if (!ultra && offCode >= 20) price += (offCode-19)*2;
+
+ /* match Length */
+ { const BYTE ML_deltaCode = 36;
+ const BYTE mlCode = (matchLength>127) ? (BYTE)ZSTD_highbit32(matchLength) + ML_deltaCode : ML_Code[matchLength];
+ price += ML_bits[mlCode] + optPtr->log2matchLengthSum - ZSTD_highbit32(optPtr->matchLengthFreq[mlCode]+1);
+ }
+
+ return price + ZSTD_getLiteralPrice(optPtr, litLength, literals) + optPtr->factor;
+}
+
+
+static void ZSTD_updatePrice(optState_t* optPtr, U32 litLength, const BYTE* literals, U32 offset, U32 matchLength)
+{
+ U32 u;
+
+ /* literals */
+ optPtr->litSum += litLength*ZSTD_LITFREQ_ADD;
+ for (u=0; u < litLength; u++)
+ optPtr->litFreq[literals[u]] += ZSTD_LITFREQ_ADD;
+
+ /* literal Length */
+ { const BYTE LL_deltaCode = 19;
+ const BYTE llCode = (litLength>63) ? (BYTE)ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength];
+ optPtr->litLengthFreq[llCode]++;
+ optPtr->litLengthSum++;
+ }
+
+ /* match offset */
+ { BYTE const offCode = (BYTE)ZSTD_highbit32(offset+1);
+ optPtr->offCodeSum++;
+ optPtr->offCodeFreq[offCode]++;
+ }
+
+ /* match Length */
+ { const BYTE ML_deltaCode = 36;
+ const BYTE mlCode = (matchLength>127) ? (BYTE)ZSTD_highbit32(matchLength) + ML_deltaCode : ML_Code[matchLength];
+ optPtr->matchLengthFreq[mlCode]++;
+ optPtr->matchLengthSum++;
+ }
+
+ ZSTD_setLog2Prices(optPtr);
+}
+
+
+#define SET_PRICE(pos, mlen_, offset_, litlen_, price_) \
+ { \
+ while (last_pos < pos) { opt[last_pos+1].price = ZSTD_MAX_PRICE; last_pos++; } \
+ opt[pos].mlen = mlen_; \
+ opt[pos].off = offset_; \
+ opt[pos].litlen = litlen_; \
+ opt[pos].price = price_; \
+ }
+
+
+/* function safe only for comparisons */
+static U32 ZSTD_readMINMATCH(const void* memPtr, U32 length)
+{
+ switch (length)
+ {
+ default :
+ case 4 : return MEM_read32(memPtr);
+ case 3 : if (MEM_isLittleEndian())
+ return MEM_read32(memPtr)<<8;
+ else
+ return MEM_read32(memPtr)>>8;
+ }
+}
+
+
+/* Update hashTable3 up to ip (excluded)
+ Assumption : always within prefix (i.e. not within extDict) */
+static
+U32 ZSTD_insertAndFindFirstIndexHash3 (ZSTD_CCtx* zc, const BYTE* ip)
+{
+ U32* const hashTable3 = zc->hashTable3;
+ U32 const hashLog3 = zc->hashLog3;
+ const BYTE* const base = zc->base;
+ U32 idx = zc->nextToUpdate3;
+ const U32 target = zc->nextToUpdate3 = (U32)(ip - base);
+ const size_t hash3 = ZSTD_hash3Ptr(ip, hashLog3);
+
+ while(idx < target) {
+ hashTable3[ZSTD_hash3Ptr(base+idx, hashLog3)] = idx;
+ idx++;
+ }
+
+ return hashTable3[hash3];
+}
+
+
+/*-*************************************
+* Binary Tree search
+***************************************/
+static U32 ZSTD_insertBtAndGetAllMatches (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iLimit,
+ U32 nbCompares, const U32 mls,
+ U32 extDict, ZSTD_match_t* matches, const U32 minMatchLen)
+{
+ const BYTE* const base = zc->base;
+ const U32 current = (U32)(ip-base);
+ const U32 hashLog = zc->appliedParams.cParams.hashLog;
+ const size_t h = ZSTD_hashPtr(ip, hashLog, mls);
+ U32* const hashTable = zc->hashTable;
+ U32 matchIndex = hashTable[h];
+ U32* const bt = zc->chainTable;
+ const U32 btLog = zc->appliedParams.cParams.chainLog - 1;
+ const U32 btMask= (1U << btLog) - 1;
+ size_t commonLengthSmaller=0, commonLengthLarger=0;
+ const BYTE* const dictBase = zc->dictBase;
+ const U32 dictLimit = zc->dictLimit;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const U32 btLow = btMask >= current ? 0 : current - btMask;
+ const U32 windowLow = zc->lowLimit;
+ U32* smallerPtr = bt + 2*(current&btMask);
+ U32* largerPtr = bt + 2*(current&btMask) + 1;
+ U32 matchEndIdx = current+8;
+ U32 dummy32; /* to be nullified at the end */
+ U32 mnum = 0;
+
+ const U32 minMatch = (mls == 3) ? 3 : 4;
+ size_t bestLength = minMatchLen-1;
+
+ if (minMatch == 3) { /* HC3 match finder */
+ U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3 (zc, ip);
+ if (matchIndex3>windowLow && (current - matchIndex3 < (1<<18))) {
+ const BYTE* match;
+ size_t currentMl=0;
+ if ((!extDict) || matchIndex3 >= dictLimit) {
+ match = base + matchIndex3;
+ if (match[bestLength] == ip[bestLength]) currentMl = ZSTD_count(ip, match, iLimit);
+ } else {
+ match = dictBase + matchIndex3;
+ if (ZSTD_readMINMATCH(match, MINMATCH) == ZSTD_readMINMATCH(ip, MINMATCH)) /* assumption : matchIndex3 <= dictLimit-4 (by table construction) */
+ currentMl = ZSTD_count_2segments(ip+MINMATCH, match+MINMATCH, iLimit, dictEnd, prefixStart) + MINMATCH;
+ }
+
+ /* save best solution */
+ if (currentMl > bestLength) {
+ bestLength = currentMl;
+ matches[mnum].off = ZSTD_REP_MOVE_OPT + current - matchIndex3;
+ matches[mnum].len = (U32)currentMl;
+ mnum++;
+ if (currentMl > ZSTD_OPT_NUM) goto update;
+ if (ip+currentMl == iLimit) goto update; /* best possible, and avoid read overflow*/
+ }
+ }
+ }
+
+ hashTable[h] = current; /* Update Hash Table */
+
+ while (nbCompares-- && (matchIndex > windowLow)) {
+ U32* nextPtr = bt + 2*(matchIndex & btMask);
+ size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
+ const BYTE* match;
+
+ if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
+ match = base + matchIndex;
+ if (match[matchLength] == ip[matchLength]) {
+ matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iLimit) +1;
+ }
+ } else {
+ match = dictBase + matchIndex;
+ matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iLimit, dictEnd, prefixStart);
+ if (matchIndex+matchLength >= dictLimit)
+ match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
+ }
+
+ if (matchLength > bestLength) {
+ if (matchLength > matchEndIdx - matchIndex) matchEndIdx = matchIndex + (U32)matchLength;
+ bestLength = matchLength;
+ matches[mnum].off = ZSTD_REP_MOVE_OPT + current - matchIndex;
+ matches[mnum].len = (U32)matchLength;
+ mnum++;
+ if (matchLength > ZSTD_OPT_NUM) break;
+ if (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */
+ break; /* drop, to guarantee consistency (miss a little bit of compression) */
+ }
+
+ if (match[matchLength] < ip[matchLength]) {
+ /* match is smaller than current */
+ *smallerPtr = matchIndex; /* update smaller idx */
+ commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
+ if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
+ matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
+ } else {
+ /* match is larger than current */
+ *largerPtr = matchIndex;
+ commonLengthLarger = matchLength;
+ if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
+ largerPtr = nextPtr;
+ matchIndex = nextPtr[0];
+ } }
+
+ *smallerPtr = *largerPtr = 0;
+
+update:
+ zc->nextToUpdate = (matchEndIdx > current + 8) ? matchEndIdx - 8 : current+1;
+ return mnum;
+}
+
+
+/** Tree updater, providing best match */
+static U32 ZSTD_BtGetAllMatches (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iLimit,
+ const U32 maxNbAttempts, const U32 mls, ZSTD_match_t* matches, const U32 minMatchLen)
+{
+ if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
+ ZSTD_updateTree(zc, ip, iLimit, maxNbAttempts, mls);
+ return ZSTD_insertBtAndGetAllMatches(zc, ip, iLimit, maxNbAttempts, mls, 0, matches, minMatchLen);
+}
+
+
+static U32 ZSTD_BtGetAllMatches_selectMLS (
+ ZSTD_CCtx* zc, /* Index table will be updated */
+ const BYTE* ip, const BYTE* const iHighLimit,
+ const U32 maxNbAttempts, const U32 matchLengthSearch, ZSTD_match_t* matches, const U32 minMatchLen)
+{
+ switch(matchLengthSearch)
+ {
+ case 3 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 3, matches, minMatchLen);
+ default :
+ case 4 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 4, matches, minMatchLen);
+ case 5 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 5, matches, minMatchLen);
+ case 7 :
+ case 6 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 6, matches, minMatchLen);
+ }
+}
+
+/** Tree updater, providing best match */
+static U32 ZSTD_BtGetAllMatches_extDict (
+ ZSTD_CCtx* zc,
+ const BYTE* const ip, const BYTE* const iLimit,
+ const U32 maxNbAttempts, const U32 mls, ZSTD_match_t* matches, const U32 minMatchLen)
+{
+ if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
+ ZSTD_updateTree_extDict(zc, ip, iLimit, maxNbAttempts, mls);
+ return ZSTD_insertBtAndGetAllMatches(zc, ip, iLimit, maxNbAttempts, mls, 1, matches, minMatchLen);
+}
+
+
+static U32 ZSTD_BtGetAllMatches_selectMLS_extDict (
+ ZSTD_CCtx* zc, /* Index table will be updated */
+ const BYTE* ip, const BYTE* const iHighLimit,
+ const U32 maxNbAttempts, const U32 matchLengthSearch, ZSTD_match_t* matches, const U32 minMatchLen)
+{
+ switch(matchLengthSearch)
+ {
+ case 3 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 3, matches, minMatchLen);
+ default :
+ case 4 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 4, matches, minMatchLen);
+ case 5 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 5, matches, minMatchLen);
+ case 7 :
+ case 6 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 6, matches, minMatchLen);
+ }
+}
+
+
+/*-*******************************
+* Optimal parser
+*********************************/
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_opt_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize, const int ultra)
+{
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ optState_t* optStatePtr = &(ctx->optState);
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ const BYTE* const base = ctx->base;
+ const BYTE* const prefixStart = base + ctx->dictLimit;
+
+ const U32 maxSearches = 1U << ctx->appliedParams.cParams.searchLog;
+ const U32 sufficient_len = ctx->appliedParams.cParams.targetLength;
+ const U32 mls = ctx->appliedParams.cParams.searchLength;
+ const U32 minMatch = (ctx->appliedParams.cParams.searchLength == 3) ? 3 : 4;
+
+ ZSTD_optimal_t* opt = optStatePtr->priceTable;
+ ZSTD_match_t* matches = optStatePtr->matchTable;
+ const BYTE* inr;
+ U32 offset, rep[ZSTD_REP_NUM];
+
+ /* init */
+ ctx->nextToUpdate3 = ctx->nextToUpdate;
+ ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize);
+ ip += (ip==prefixStart);
+ { U32 i; for (i=0; irep[i]; }
+
+ /* Match Loop */
+ while (ip < ilimit) {
+ U32 cur, match_num, last_pos, litlen, price;
+ U32 u, mlen, best_mlen, best_off, litLength;
+ memset(opt, 0, sizeof(ZSTD_optimal_t));
+ last_pos = 0;
+ litlen = (U32)(ip - anchor);
+
+ /* check repCode */
+ { U32 i, last_i = ZSTD_REP_CHECK + (ip==anchor);
+ for (i=(ip == anchor); i 0) && (repCur < (S32)(ip-prefixStart))
+ && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(ip - repCur, minMatch))) {
+ mlen = (U32)ZSTD_count(ip+minMatch, ip+minMatch-repCur, iend) + minMatch;
+ if (mlen > sufficient_len || mlen >= ZSTD_OPT_NUM) {
+ best_mlen = mlen; best_off = i; cur = 0; last_pos = 1;
+ goto _storeSequence;
+ }
+ best_off = i - (ip == anchor);
+ do {
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
+ if (mlen > last_pos || price < opt[mlen].price)
+ SET_PRICE(mlen, mlen, i, litlen, price); /* note : macro modifies last_pos */
+ mlen--;
+ } while (mlen >= minMatch);
+ } } }
+
+ match_num = ZSTD_BtGetAllMatches_selectMLS(ctx, ip, iend, maxSearches, mls, matches, minMatch);
+
+ if (!last_pos && !match_num) { ip++; continue; }
+
+ if (match_num && (matches[match_num-1].len > sufficient_len || matches[match_num-1].len >= ZSTD_OPT_NUM)) {
+ best_mlen = matches[match_num-1].len;
+ best_off = matches[match_num-1].off;
+ cur = 0;
+ last_pos = 1;
+ goto _storeSequence;
+ }
+
+ /* set prices using matches at position = 0 */
+ best_mlen = (last_pos) ? last_pos : minMatch;
+ for (u = 0; u < match_num; u++) {
+ mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
+ best_mlen = matches[u].len;
+ while (mlen <= best_mlen) {
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
+ if (mlen > last_pos || price < opt[mlen].price)
+ SET_PRICE(mlen, mlen, matches[u].off, litlen, price); /* note : macro modifies last_pos */
+ mlen++;
+ } }
+
+ if (last_pos < minMatch) { ip++; continue; }
+
+ /* initialize opt[0] */
+ { U32 i ; for (i=0; i litlen) {
+ price = opt[cur - litlen].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-litlen);
+ } else
+ price = ZSTD_getLiteralPrice(optStatePtr, litlen, anchor);
+ } else {
+ litlen = 1;
+ price = opt[cur - 1].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-1);
+ }
+
+ if (cur > last_pos || price <= opt[cur].price)
+ SET_PRICE(cur, 1, 0, litlen, price);
+
+ if (cur == last_pos) break;
+
+ if (inr > ilimit) /* last match must start at a minimum distance of 8 from oend */
+ continue;
+
+ mlen = opt[cur].mlen;
+ if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
+ opt[cur].rep[2] = opt[cur-mlen].rep[1];
+ opt[cur].rep[1] = opt[cur-mlen].rep[0];
+ opt[cur].rep[0] = opt[cur].off - ZSTD_REP_MOVE_OPT;
+ } else {
+ opt[cur].rep[2] = (opt[cur].off > 1) ? opt[cur-mlen].rep[1] : opt[cur-mlen].rep[2];
+ opt[cur].rep[1] = (opt[cur].off > 0) ? opt[cur-mlen].rep[0] : opt[cur-mlen].rep[1];
+ opt[cur].rep[0] = ((opt[cur].off==ZSTD_REP_MOVE_OPT) && (mlen != 1)) ? (opt[cur-mlen].rep[0] - 1) : (opt[cur-mlen].rep[opt[cur].off]);
+ }
+
+ best_mlen = minMatch;
+ { U32 i, last_i = ZSTD_REP_CHECK + (mlen != 1);
+ for (i=(opt[cur].mlen != 1); i 0) && (repCur < (S32)(inr-prefixStart))
+ && (ZSTD_readMINMATCH(inr, minMatch) == ZSTD_readMINMATCH(inr - repCur, minMatch))) {
+ mlen = (U32)ZSTD_count(inr+minMatch, inr+minMatch - repCur, iend) + minMatch;
+
+ if (mlen > sufficient_len || cur + mlen >= ZSTD_OPT_NUM) {
+ best_mlen = mlen; best_off = i; last_pos = cur + 1;
+ goto _storeSequence;
+ }
+
+ best_off = i - (opt[cur].mlen != 1);
+ if (mlen > best_mlen) best_mlen = mlen;
+
+ do {
+ if (opt[cur].mlen == 1) {
+ litlen = opt[cur].litlen;
+ if (cur > litlen) {
+ price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, inr-litlen, best_off, mlen - MINMATCH, ultra);
+ } else
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
+ } else {
+ litlen = 0;
+ price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, best_off, mlen - MINMATCH, ultra);
+ }
+
+ if (cur + mlen > last_pos || price <= opt[cur + mlen].price)
+ SET_PRICE(cur + mlen, mlen, i, litlen, price);
+ mlen--;
+ } while (mlen >= minMatch);
+ } } }
+
+ match_num = ZSTD_BtGetAllMatches_selectMLS(ctx, inr, iend, maxSearches, mls, matches, best_mlen);
+
+ if (match_num > 0 && (matches[match_num-1].len > sufficient_len || cur + matches[match_num-1].len >= ZSTD_OPT_NUM)) {
+ best_mlen = matches[match_num-1].len;
+ best_off = matches[match_num-1].off;
+ last_pos = cur + 1;
+ goto _storeSequence;
+ }
+
+ /* set prices using matches at position = cur */
+ for (u = 0; u < match_num; u++) {
+ mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
+ best_mlen = matches[u].len;
+
+ while (mlen <= best_mlen) {
+ if (opt[cur].mlen == 1) {
+ litlen = opt[cur].litlen;
+ if (cur > litlen)
+ price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, ip+cur-litlen, matches[u].off-1, mlen - MINMATCH, ultra);
+ else
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
+ } else {
+ litlen = 0;
+ price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, matches[u].off-1, mlen - MINMATCH, ultra);
+ }
+
+ if (cur + mlen > last_pos || (price < opt[cur + mlen].price))
+ SET_PRICE(cur + mlen, mlen, matches[u].off, litlen, price);
+
+ mlen++;
+ } } }
+
+ best_mlen = opt[last_pos].mlen;
+ best_off = opt[last_pos].off;
+ cur = last_pos - best_mlen;
+
+ /* store sequence */
+_storeSequence: /* cur, last_pos, best_mlen, best_off have to be set */
+ opt[0].mlen = 1;
+
+ while (1) {
+ mlen = opt[cur].mlen;
+ offset = opt[cur].off;
+ opt[cur].mlen = best_mlen;
+ opt[cur].off = best_off;
+ best_mlen = mlen;
+ best_off = offset;
+ if (mlen > cur) break;
+ cur -= mlen;
+ }
+
+ for (u = 0; u <= last_pos;) {
+ u += opt[u].mlen;
+ }
+
+ for (cur=0; cur < last_pos; ) {
+ mlen = opt[cur].mlen;
+ if (mlen == 1) { ip++; cur++; continue; }
+ offset = opt[cur].off;
+ cur += mlen;
+ litLength = (U32)(ip - anchor);
+
+ if (offset > ZSTD_REP_MOVE_OPT) {
+ rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = offset - ZSTD_REP_MOVE_OPT;
+ offset--;
+ } else {
+ if (offset != 0) {
+ best_off = (offset==ZSTD_REP_MOVE_OPT) ? (rep[0] - 1) : (rep[offset]);
+ if (offset != 1) rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = best_off;
+ }
+ if (litLength==0) offset--;
+ }
+
+ ZSTD_updatePrice(optStatePtr, litLength, anchor, offset, mlen-MINMATCH);
+ ZSTD_storeSeq(seqStorePtr, litLength, anchor, offset, mlen-MINMATCH);
+ anchor = ip = ip + mlen;
+ } } /* for (cur=0; cur < last_pos; ) */
+
+ /* Save reps for next block */
+ { int i; for (i=0; irepToConfirm[i] = rep[i]; }
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_btopt(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_generic(ctx, src, srcSize, 0);
+}
+
+size_t ZSTD_compressBlock_btultra(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_generic(ctx, src, srcSize, 1);
+}
+
+
+FORCE_INLINE_TEMPLATE
+size_t ZSTD_compressBlock_opt_extDict_generic(ZSTD_CCtx* ctx,
+ const void* src, size_t srcSize, const int ultra)
+{
+ seqStore_t* seqStorePtr = &(ctx->seqStore);
+ optState_t* optStatePtr = &(ctx->optState);
+ const BYTE* const istart = (const BYTE*)src;
+ const BYTE* ip = istart;
+ const BYTE* anchor = istart;
+ const BYTE* const iend = istart + srcSize;
+ const BYTE* const ilimit = iend - 8;
+ const BYTE* const base = ctx->base;
+ const U32 lowestIndex = ctx->lowLimit;
+ const U32 dictLimit = ctx->dictLimit;
+ const BYTE* const prefixStart = base + dictLimit;
+ const BYTE* const dictBase = ctx->dictBase;
+ const BYTE* const dictEnd = dictBase + dictLimit;
+
+ const U32 maxSearches = 1U << ctx->appliedParams.cParams.searchLog;
+ const U32 sufficient_len = ctx->appliedParams.cParams.targetLength;
+ const U32 mls = ctx->appliedParams.cParams.searchLength;
+ const U32 minMatch = (ctx->appliedParams.cParams.searchLength == 3) ? 3 : 4;
+
+ ZSTD_optimal_t* opt = optStatePtr->priceTable;
+ ZSTD_match_t* matches = optStatePtr->matchTable;
+ const BYTE* inr;
+
+ /* init */
+ U32 offset, rep[ZSTD_REP_NUM];
+ { U32 i; for (i=0; irep[i]; }
+
+ ctx->nextToUpdate3 = ctx->nextToUpdate;
+ ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize);
+ ip += (ip==prefixStart);
+
+ /* Match Loop */
+ while (ip < ilimit) {
+ U32 cur, match_num, last_pos, litlen, price;
+ U32 u, mlen, best_mlen, best_off, litLength;
+ U32 current = (U32)(ip-base);
+ memset(opt, 0, sizeof(ZSTD_optimal_t));
+ last_pos = 0;
+ opt[0].litlen = (U32)(ip - anchor);
+
+ /* check repCode */
+ { U32 i, last_i = ZSTD_REP_CHECK + (ip==anchor);
+ for (i = (ip==anchor); i 0 && repCur <= (S32)current)
+ && (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex>lowestIndex)) /* intentional overflow */
+ && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
+ /* repcode detected we should take it */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ mlen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iend, repEnd, prefixStart) + minMatch;
+
+ if (mlen > sufficient_len || mlen >= ZSTD_OPT_NUM) {
+ best_mlen = mlen; best_off = i; cur = 0; last_pos = 1;
+ goto _storeSequence;
+ }
+
+ best_off = i - (ip==anchor);
+ litlen = opt[0].litlen;
+ do {
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
+ if (mlen > last_pos || price < opt[mlen].price)
+ SET_PRICE(mlen, mlen, i, litlen, price); /* note : macro modifies last_pos */
+ mlen--;
+ } while (mlen >= minMatch);
+ } } }
+
+ match_num = ZSTD_BtGetAllMatches_selectMLS_extDict(ctx, ip, iend, maxSearches, mls, matches, minMatch); /* first search (depth 0) */
+
+ if (!last_pos && !match_num) { ip++; continue; }
+
+ { U32 i; for (i=0; i sufficient_len || matches[match_num-1].len >= ZSTD_OPT_NUM)) {
+ best_mlen = matches[match_num-1].len;
+ best_off = matches[match_num-1].off;
+ cur = 0;
+ last_pos = 1;
+ goto _storeSequence;
+ }
+
+ best_mlen = (last_pos) ? last_pos : minMatch;
+
+ /* set prices using matches at position = 0 */
+ for (u = 0; u < match_num; u++) {
+ mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
+ best_mlen = matches[u].len;
+ litlen = opt[0].litlen;
+ while (mlen <= best_mlen) {
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
+ if (mlen > last_pos || price < opt[mlen].price)
+ SET_PRICE(mlen, mlen, matches[u].off, litlen, price);
+ mlen++;
+ } }
+
+ if (last_pos < minMatch) {
+ ip++; continue;
+ }
+
+ /* check further positions */
+ for (cur = 1; cur <= last_pos; cur++) {
+ inr = ip + cur;
+
+ if (opt[cur-1].mlen == 1) {
+ litlen = opt[cur-1].litlen + 1;
+ if (cur > litlen) {
+ price = opt[cur - litlen].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-litlen);
+ } else
+ price = ZSTD_getLiteralPrice(optStatePtr, litlen, anchor);
+ } else {
+ litlen = 1;
+ price = opt[cur - 1].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-1);
+ }
+
+ if (cur > last_pos || price <= opt[cur].price)
+ SET_PRICE(cur, 1, 0, litlen, price);
+
+ if (cur == last_pos) break;
+
+ if (inr > ilimit) /* last match must start at a minimum distance of 8 from oend */
+ continue;
+
+ mlen = opt[cur].mlen;
+ if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
+ opt[cur].rep[2] = opt[cur-mlen].rep[1];
+ opt[cur].rep[1] = opt[cur-mlen].rep[0];
+ opt[cur].rep[0] = opt[cur].off - ZSTD_REP_MOVE_OPT;
+ } else {
+ opt[cur].rep[2] = (opt[cur].off > 1) ? opt[cur-mlen].rep[1] : opt[cur-mlen].rep[2];
+ opt[cur].rep[1] = (opt[cur].off > 0) ? opt[cur-mlen].rep[0] : opt[cur-mlen].rep[1];
+ opt[cur].rep[0] = ((opt[cur].off==ZSTD_REP_MOVE_OPT) && (mlen != 1)) ? (opt[cur-mlen].rep[0] - 1) : (opt[cur-mlen].rep[opt[cur].off]);
+ }
+
+ best_mlen = minMatch;
+ { U32 i, last_i = ZSTD_REP_CHECK + (mlen != 1);
+ for (i = (mlen != 1); i 0 && repCur <= (S32)(current+cur))
+ && (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex>lowestIndex)) /* intentional overflow */
+ && (ZSTD_readMINMATCH(inr, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
+ /* repcode detected */
+ const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
+ mlen = (U32)ZSTD_count_2segments(inr+minMatch, repMatch+minMatch, iend, repEnd, prefixStart) + minMatch;
+
+ if (mlen > sufficient_len || cur + mlen >= ZSTD_OPT_NUM) {
+ best_mlen = mlen; best_off = i; last_pos = cur + 1;
+ goto _storeSequence;
+ }
+
+ best_off = i - (opt[cur].mlen != 1);
+ if (mlen > best_mlen) best_mlen = mlen;
+
+ do {
+ if (opt[cur].mlen == 1) {
+ litlen = opt[cur].litlen;
+ if (cur > litlen) {
+ price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, inr-litlen, best_off, mlen - MINMATCH, ultra);
+ } else
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
+ } else {
+ litlen = 0;
+ price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, best_off, mlen - MINMATCH, ultra);
+ }
+
+ if (cur + mlen > last_pos || price <= opt[cur + mlen].price)
+ SET_PRICE(cur + mlen, mlen, i, litlen, price);
+ mlen--;
+ } while (mlen >= minMatch);
+ } } }
+
+ match_num = ZSTD_BtGetAllMatches_selectMLS_extDict(ctx, inr, iend, maxSearches, mls, matches, minMatch);
+
+ if (match_num > 0 && (matches[match_num-1].len > sufficient_len || cur + matches[match_num-1].len >= ZSTD_OPT_NUM)) {
+ best_mlen = matches[match_num-1].len;
+ best_off = matches[match_num-1].off;
+ last_pos = cur + 1;
+ goto _storeSequence;
+ }
+
+ /* set prices using matches at position = cur */
+ for (u = 0; u < match_num; u++) {
+ mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
+ best_mlen = matches[u].len;
+
+ while (mlen <= best_mlen) {
+ if (opt[cur].mlen == 1) {
+ litlen = opt[cur].litlen;
+ if (cur > litlen)
+ price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, ip+cur-litlen, matches[u].off-1, mlen - MINMATCH, ultra);
+ else
+ price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
+ } else {
+ litlen = 0;
+ price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, matches[u].off-1, mlen - MINMATCH, ultra);
+ }
+
+ if (cur + mlen > last_pos || (price < opt[cur + mlen].price))
+ SET_PRICE(cur + mlen, mlen, matches[u].off, litlen, price);
+
+ mlen++;
+ } } } /* for (cur = 1; cur <= last_pos; cur++) */
+
+ best_mlen = opt[last_pos].mlen;
+ best_off = opt[last_pos].off;
+ cur = last_pos - best_mlen;
+
+ /* store sequence */
+_storeSequence: /* cur, last_pos, best_mlen, best_off have to be set */
+ opt[0].mlen = 1;
+
+ while (1) {
+ mlen = opt[cur].mlen;
+ offset = opt[cur].off;
+ opt[cur].mlen = best_mlen;
+ opt[cur].off = best_off;
+ best_mlen = mlen;
+ best_off = offset;
+ if (mlen > cur) break;
+ cur -= mlen;
+ }
+
+ for (u = 0; u <= last_pos; ) {
+ u += opt[u].mlen;
+ }
+
+ for (cur=0; cur < last_pos; ) {
+ mlen = opt[cur].mlen;
+ if (mlen == 1) { ip++; cur++; continue; }
+ offset = opt[cur].off;
+ cur += mlen;
+ litLength = (U32)(ip - anchor);
+
+ if (offset > ZSTD_REP_MOVE_OPT) {
+ rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = offset - ZSTD_REP_MOVE_OPT;
+ offset--;
+ } else {
+ if (offset != 0) {
+ best_off = (offset==ZSTD_REP_MOVE_OPT) ? (rep[0] - 1) : (rep[offset]);
+ if (offset != 1) rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = best_off;
+ }
+
+ if (litLength==0) offset--;
+ }
+
+ ZSTD_updatePrice(optStatePtr, litLength, anchor, offset, mlen-MINMATCH);
+ ZSTD_storeSeq(seqStorePtr, litLength, anchor, offset, mlen-MINMATCH);
+ anchor = ip = ip + mlen;
+ } } /* for (cur=0; cur < last_pos; ) */
+
+ /* Save reps for next block */
+ { int i; for (i=0; irepToConfirm[i] = rep[i]; }
+
+ /* Return the last literals size */
+ return iend - anchor;
+}
+
+
+size_t ZSTD_compressBlock_btopt_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_extDict_generic(ctx, src, srcSize, 0);
+}
+
+size_t ZSTD_compressBlock_btultra_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize)
+{
+ return ZSTD_compressBlock_opt_extDict_generic(ctx, src, srcSize, 1);
+}
diff --git a/lib/compress/zstd_opt.h b/lib/compress/zstd_opt.h
index d5f503d0f..c15c15f87 100644
--- a/lib/compress/zstd_opt.h
+++ b/lib/compress/zstd_opt.h
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2016-present, Przemyslaw Skibinski, Yann Collet, Facebook, Inc.
+ * Copyright (c) 2016-present, Yann Collet, Facebook, Inc.
* All rights reserved.
*
* This source code is licensed under both the BSD-style license (found in the
@@ -7,926 +7,23 @@
* in the COPYING file in the root directory of this source tree).
*/
+#ifndef ZSTD_OPT_H
+#define ZSTD_OPT_H
-/* Note : this file is intended to be included within zstd_compress.c */
+#include "zstd_compress.h"
+#if defined (__cplusplus)
+extern "C" {
+#endif
-#ifndef ZSTD_OPT_H_91842398743
-#define ZSTD_OPT_H_91842398743
+size_t ZSTD_compressBlock_btopt(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_btultra(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_btopt_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
+size_t ZSTD_compressBlock_btultra_extDict(ZSTD_CCtx* ctx, const void* src, size_t srcSize);
-#define ZSTD_LITFREQ_ADD 2
-#define ZSTD_FREQ_DIV 4
-#define ZSTD_MAX_PRICE (1<<30)
-
-/*-*************************************
-* Price functions for optimal parser
-***************************************/
-static void ZSTD_setLog2Prices(optState_t* optPtr)
-{
- optPtr->log2matchLengthSum = ZSTD_highbit32(optPtr->matchLengthSum+1);
- optPtr->log2litLengthSum = ZSTD_highbit32(optPtr->litLengthSum+1);
- optPtr->log2litSum = ZSTD_highbit32(optPtr->litSum+1);
- optPtr->log2offCodeSum = ZSTD_highbit32(optPtr->offCodeSum+1);
- optPtr->factor = 1 + ((optPtr->litSum>>5) / optPtr->litLengthSum) + ((optPtr->litSum<<1) / (optPtr->litSum + optPtr->matchSum));
+#if defined (__cplusplus)
}
+#endif
-
-static void ZSTD_rescaleFreqs(optState_t* optPtr, const BYTE* src, size_t srcSize)
-{
- unsigned u;
-
- optPtr->cachedLiterals = NULL;
- optPtr->cachedPrice = optPtr->cachedLitLength = 0;
- optPtr->staticPrices = 0;
-
- if (optPtr->litLengthSum == 0) {
- if (srcSize <= 1024) optPtr->staticPrices = 1;
-
- assert(optPtr->litFreq!=NULL);
- for (u=0; u<=MaxLit; u++)
- optPtr->litFreq[u] = 0;
- for (u=0; ulitFreq[src[u]]++;
-
- optPtr->litSum = 0;
- optPtr->litLengthSum = MaxLL+1;
- optPtr->matchLengthSum = MaxML+1;
- optPtr->offCodeSum = (MaxOff+1);
- optPtr->matchSum = (ZSTD_LITFREQ_ADD<litFreq[u] = 1 + (optPtr->litFreq[u]>>ZSTD_FREQ_DIV);
- optPtr->litSum += optPtr->litFreq[u];
- }
- for (u=0; u<=MaxLL; u++)
- optPtr->litLengthFreq[u] = 1;
- for (u=0; u<=MaxML; u++)
- optPtr->matchLengthFreq[u] = 1;
- for (u=0; u<=MaxOff; u++)
- optPtr->offCodeFreq[u] = 1;
- } else {
- optPtr->matchLengthSum = 0;
- optPtr->litLengthSum = 0;
- optPtr->offCodeSum = 0;
- optPtr->matchSum = 0;
- optPtr->litSum = 0;
-
- for (u=0; u<=MaxLit; u++) {
- optPtr->litFreq[u] = 1 + (optPtr->litFreq[u]>>(ZSTD_FREQ_DIV+1));
- optPtr->litSum += optPtr->litFreq[u];
- }
- for (u=0; u<=MaxLL; u++) {
- optPtr->litLengthFreq[u] = 1 + (optPtr->litLengthFreq[u]>>(ZSTD_FREQ_DIV+1));
- optPtr->litLengthSum += optPtr->litLengthFreq[u];
- }
- for (u=0; u<=MaxML; u++) {
- optPtr->matchLengthFreq[u] = 1 + (optPtr->matchLengthFreq[u]>>ZSTD_FREQ_DIV);
- optPtr->matchLengthSum += optPtr->matchLengthFreq[u];
- optPtr->matchSum += optPtr->matchLengthFreq[u] * (u + 3);
- }
- optPtr->matchSum *= ZSTD_LITFREQ_ADD;
- for (u=0; u<=MaxOff; u++) {
- optPtr->offCodeFreq[u] = 1 + (optPtr->offCodeFreq[u]>>ZSTD_FREQ_DIV);
- optPtr->offCodeSum += optPtr->offCodeFreq[u];
- }
- }
-
- ZSTD_setLog2Prices(optPtr);
-}
-
-
-static U32 ZSTD_getLiteralPrice(optState_t* optPtr, U32 litLength, const BYTE* literals)
-{
- U32 price, u;
-
- if (optPtr->staticPrices)
- return ZSTD_highbit32((U32)litLength+1) + (litLength*6);
-
- if (litLength == 0)
- return optPtr->log2litLengthSum - ZSTD_highbit32(optPtr->litLengthFreq[0]+1);
-
- /* literals */
- if (optPtr->cachedLiterals == literals) {
- U32 const additional = litLength - optPtr->cachedLitLength;
- const BYTE* literals2 = optPtr->cachedLiterals + optPtr->cachedLitLength;
- price = optPtr->cachedPrice + additional * optPtr->log2litSum;
- for (u=0; u < additional; u++)
- price -= ZSTD_highbit32(optPtr->litFreq[literals2[u]]+1);
- optPtr->cachedPrice = price;
- optPtr->cachedLitLength = litLength;
- } else {
- price = litLength * optPtr->log2litSum;
- for (u=0; u < litLength; u++)
- price -= ZSTD_highbit32(optPtr->litFreq[literals[u]]+1);
-
- if (litLength >= 12) {
- optPtr->cachedLiterals = literals;
- optPtr->cachedPrice = price;
- optPtr->cachedLitLength = litLength;
- }
- }
-
- /* literal Length */
- { const BYTE LL_deltaCode = 19;
- const BYTE llCode = (litLength>63) ? (BYTE)ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength];
- price += LL_bits[llCode] + optPtr->log2litLengthSum - ZSTD_highbit32(optPtr->litLengthFreq[llCode]+1);
- }
-
- return price;
-}
-
-
-FORCE_INLINE_TEMPLATE U32 ZSTD_getPrice(optState_t* optPtr, U32 litLength, const BYTE* literals, U32 offset, U32 matchLength, const int ultra)
-{
- /* offset */
- U32 price;
- BYTE const offCode = (BYTE)ZSTD_highbit32(offset+1);
-
- if (optPtr->staticPrices)
- return ZSTD_getLiteralPrice(optPtr, litLength, literals) + ZSTD_highbit32((U32)matchLength+1) + 16 + offCode;
-
- price = offCode + optPtr->log2offCodeSum - ZSTD_highbit32(optPtr->offCodeFreq[offCode]+1);
- if (!ultra && offCode >= 20) price += (offCode-19)*2;
-
- /* match Length */
- { const BYTE ML_deltaCode = 36;
- const BYTE mlCode = (matchLength>127) ? (BYTE)ZSTD_highbit32(matchLength) + ML_deltaCode : ML_Code[matchLength];
- price += ML_bits[mlCode] + optPtr->log2matchLengthSum - ZSTD_highbit32(optPtr->matchLengthFreq[mlCode]+1);
- }
-
- return price + ZSTD_getLiteralPrice(optPtr, litLength, literals) + optPtr->factor;
-}
-
-
-static void ZSTD_updatePrice(optState_t* optPtr, U32 litLength, const BYTE* literals, U32 offset, U32 matchLength)
-{
- U32 u;
-
- /* literals */
- optPtr->litSum += litLength*ZSTD_LITFREQ_ADD;
- for (u=0; u < litLength; u++)
- optPtr->litFreq[literals[u]] += ZSTD_LITFREQ_ADD;
-
- /* literal Length */
- { const BYTE LL_deltaCode = 19;
- const BYTE llCode = (litLength>63) ? (BYTE)ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength];
- optPtr->litLengthFreq[llCode]++;
- optPtr->litLengthSum++;
- }
-
- /* match offset */
- { BYTE const offCode = (BYTE)ZSTD_highbit32(offset+1);
- optPtr->offCodeSum++;
- optPtr->offCodeFreq[offCode]++;
- }
-
- /* match Length */
- { const BYTE ML_deltaCode = 36;
- const BYTE mlCode = (matchLength>127) ? (BYTE)ZSTD_highbit32(matchLength) + ML_deltaCode : ML_Code[matchLength];
- optPtr->matchLengthFreq[mlCode]++;
- optPtr->matchLengthSum++;
- }
-
- ZSTD_setLog2Prices(optPtr);
-}
-
-
-#define SET_PRICE(pos, mlen_, offset_, litlen_, price_) \
- { \
- while (last_pos < pos) { opt[last_pos+1].price = ZSTD_MAX_PRICE; last_pos++; } \
- opt[pos].mlen = mlen_; \
- opt[pos].off = offset_; \
- opt[pos].litlen = litlen_; \
- opt[pos].price = price_; \
- }
-
-
-/* function safe only for comparisons */
-static U32 ZSTD_readMINMATCH(const void* memPtr, U32 length)
-{
- switch (length)
- {
- default :
- case 4 : return MEM_read32(memPtr);
- case 3 : if (MEM_isLittleEndian())
- return MEM_read32(memPtr)<<8;
- else
- return MEM_read32(memPtr)>>8;
- }
-}
-
-
-/* Update hashTable3 up to ip (excluded)
- Assumption : always within prefix (i.e. not within extDict) */
-static
-U32 ZSTD_insertAndFindFirstIndexHash3 (ZSTD_CCtx* zc, const BYTE* ip)
-{
- U32* const hashTable3 = zc->hashTable3;
- U32 const hashLog3 = zc->hashLog3;
- const BYTE* const base = zc->base;
- U32 idx = zc->nextToUpdate3;
- const U32 target = zc->nextToUpdate3 = (U32)(ip - base);
- const size_t hash3 = ZSTD_hash3Ptr(ip, hashLog3);
-
- while(idx < target) {
- hashTable3[ZSTD_hash3Ptr(base+idx, hashLog3)] = idx;
- idx++;
- }
-
- return hashTable3[hash3];
-}
-
-
-/*-*************************************
-* Binary Tree search
-***************************************/
-static U32 ZSTD_insertBtAndGetAllMatches (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iLimit,
- U32 nbCompares, const U32 mls,
- U32 extDict, ZSTD_match_t* matches, const U32 minMatchLen)
-{
- const BYTE* const base = zc->base;
- const U32 current = (U32)(ip-base);
- const U32 hashLog = zc->appliedParams.cParams.hashLog;
- const size_t h = ZSTD_hashPtr(ip, hashLog, mls);
- U32* const hashTable = zc->hashTable;
- U32 matchIndex = hashTable[h];
- U32* const bt = zc->chainTable;
- const U32 btLog = zc->appliedParams.cParams.chainLog - 1;
- const U32 btMask= (1U << btLog) - 1;
- size_t commonLengthSmaller=0, commonLengthLarger=0;
- const BYTE* const dictBase = zc->dictBase;
- const U32 dictLimit = zc->dictLimit;
- const BYTE* const dictEnd = dictBase + dictLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const U32 btLow = btMask >= current ? 0 : current - btMask;
- const U32 windowLow = zc->lowLimit;
- U32* smallerPtr = bt + 2*(current&btMask);
- U32* largerPtr = bt + 2*(current&btMask) + 1;
- U32 matchEndIdx = current+8;
- U32 dummy32; /* to be nullified at the end */
- U32 mnum = 0;
-
- const U32 minMatch = (mls == 3) ? 3 : 4;
- size_t bestLength = minMatchLen-1;
-
- if (minMatch == 3) { /* HC3 match finder */
- U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3 (zc, ip);
- if (matchIndex3>windowLow && (current - matchIndex3 < (1<<18))) {
- const BYTE* match;
- size_t currentMl=0;
- if ((!extDict) || matchIndex3 >= dictLimit) {
- match = base + matchIndex3;
- if (match[bestLength] == ip[bestLength]) currentMl = ZSTD_count(ip, match, iLimit);
- } else {
- match = dictBase + matchIndex3;
- if (ZSTD_readMINMATCH(match, MINMATCH) == ZSTD_readMINMATCH(ip, MINMATCH)) /* assumption : matchIndex3 <= dictLimit-4 (by table construction) */
- currentMl = ZSTD_count_2segments(ip+MINMATCH, match+MINMATCH, iLimit, dictEnd, prefixStart) + MINMATCH;
- }
-
- /* save best solution */
- if (currentMl > bestLength) {
- bestLength = currentMl;
- matches[mnum].off = ZSTD_REP_MOVE_OPT + current - matchIndex3;
- matches[mnum].len = (U32)currentMl;
- mnum++;
- if (currentMl > ZSTD_OPT_NUM) goto update;
- if (ip+currentMl == iLimit) goto update; /* best possible, and avoid read overflow*/
- }
- }
- }
-
- hashTable[h] = current; /* Update Hash Table */
-
- while (nbCompares-- && (matchIndex > windowLow)) {
- U32* nextPtr = bt + 2*(matchIndex & btMask);
- size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */
- const BYTE* match;
-
- if ((!extDict) || (matchIndex+matchLength >= dictLimit)) {
- match = base + matchIndex;
- if (match[matchLength] == ip[matchLength]) {
- matchLength += ZSTD_count(ip+matchLength+1, match+matchLength+1, iLimit) +1;
- }
- } else {
- match = dictBase + matchIndex;
- matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iLimit, dictEnd, prefixStart);
- if (matchIndex+matchLength >= dictLimit)
- match = base + matchIndex; /* to prepare for next usage of match[matchLength] */
- }
-
- if (matchLength > bestLength) {
- if (matchLength > matchEndIdx - matchIndex) matchEndIdx = matchIndex + (U32)matchLength;
- bestLength = matchLength;
- matches[mnum].off = ZSTD_REP_MOVE_OPT + current - matchIndex;
- matches[mnum].len = (U32)matchLength;
- mnum++;
- if (matchLength > ZSTD_OPT_NUM) break;
- if (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */
- break; /* drop, to guarantee consistency (miss a little bit of compression) */
- }
-
- if (match[matchLength] < ip[matchLength]) {
- /* match is smaller than current */
- *smallerPtr = matchIndex; /* update smaller idx */
- commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */
- if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- smallerPtr = nextPtr+1; /* new "smaller" => larger of match */
- matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */
- } else {
- /* match is larger than current */
- *largerPtr = matchIndex;
- commonLengthLarger = matchLength;
- if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */
- largerPtr = nextPtr;
- matchIndex = nextPtr[0];
- } }
-
- *smallerPtr = *largerPtr = 0;
-
-update:
- zc->nextToUpdate = (matchEndIdx > current + 8) ? matchEndIdx - 8 : current+1;
- return mnum;
-}
-
-
-/** Tree updater, providing best match */
-static U32 ZSTD_BtGetAllMatches (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iLimit,
- const U32 maxNbAttempts, const U32 mls, ZSTD_match_t* matches, const U32 minMatchLen)
-{
- if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateTree(zc, ip, iLimit, maxNbAttempts, mls);
- return ZSTD_insertBtAndGetAllMatches(zc, ip, iLimit, maxNbAttempts, mls, 0, matches, minMatchLen);
-}
-
-
-static U32 ZSTD_BtGetAllMatches_selectMLS (
- ZSTD_CCtx* zc, /* Index table will be updated */
- const BYTE* ip, const BYTE* const iHighLimit,
- const U32 maxNbAttempts, const U32 matchLengthSearch, ZSTD_match_t* matches, const U32 minMatchLen)
-{
- switch(matchLengthSearch)
- {
- case 3 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 3, matches, minMatchLen);
- default :
- case 4 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 4, matches, minMatchLen);
- case 5 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 5, matches, minMatchLen);
- case 7 :
- case 6 : return ZSTD_BtGetAllMatches(zc, ip, iHighLimit, maxNbAttempts, 6, matches, minMatchLen);
- }
-}
-
-/** Tree updater, providing best match */
-static U32 ZSTD_BtGetAllMatches_extDict (
- ZSTD_CCtx* zc,
- const BYTE* const ip, const BYTE* const iLimit,
- const U32 maxNbAttempts, const U32 mls, ZSTD_match_t* matches, const U32 minMatchLen)
-{
- if (ip < zc->base + zc->nextToUpdate) return 0; /* skipped area */
- ZSTD_updateTree_extDict(zc, ip, iLimit, maxNbAttempts, mls);
- return ZSTD_insertBtAndGetAllMatches(zc, ip, iLimit, maxNbAttempts, mls, 1, matches, minMatchLen);
-}
-
-
-static U32 ZSTD_BtGetAllMatches_selectMLS_extDict (
- ZSTD_CCtx* zc, /* Index table will be updated */
- const BYTE* ip, const BYTE* const iHighLimit,
- const U32 maxNbAttempts, const U32 matchLengthSearch, ZSTD_match_t* matches, const U32 minMatchLen)
-{
- switch(matchLengthSearch)
- {
- case 3 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 3, matches, minMatchLen);
- default :
- case 4 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 4, matches, minMatchLen);
- case 5 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 5, matches, minMatchLen);
- case 7 :
- case 6 : return ZSTD_BtGetAllMatches_extDict(zc, ip, iHighLimit, maxNbAttempts, 6, matches, minMatchLen);
- }
-}
-
-
-/*-*******************************
-* Optimal parser
-*********************************/
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_opt_generic(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize, const int ultra)
-{
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- optState_t* optStatePtr = &(ctx->optState);
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- const BYTE* const base = ctx->base;
- const BYTE* const prefixStart = base + ctx->dictLimit;
-
- const U32 maxSearches = 1U << ctx->appliedParams.cParams.searchLog;
- const U32 sufficient_len = ctx->appliedParams.cParams.targetLength;
- const U32 mls = ctx->appliedParams.cParams.searchLength;
- const U32 minMatch = (ctx->appliedParams.cParams.searchLength == 3) ? 3 : 4;
-
- ZSTD_optimal_t* opt = optStatePtr->priceTable;
- ZSTD_match_t* matches = optStatePtr->matchTable;
- const BYTE* inr;
- U32 offset, rep[ZSTD_REP_NUM];
-
- /* init */
- ctx->nextToUpdate3 = ctx->nextToUpdate;
- ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize);
- ip += (ip==prefixStart);
- { U32 i; for (i=0; irep[i]; }
-
- /* Match Loop */
- while (ip < ilimit) {
- U32 cur, match_num, last_pos, litlen, price;
- U32 u, mlen, best_mlen, best_off, litLength;
- memset(opt, 0, sizeof(ZSTD_optimal_t));
- last_pos = 0;
- litlen = (U32)(ip - anchor);
-
- /* check repCode */
- { U32 i, last_i = ZSTD_REP_CHECK + (ip==anchor);
- for (i=(ip == anchor); i 0) && (repCur < (S32)(ip-prefixStart))
- && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(ip - repCur, minMatch))) {
- mlen = (U32)ZSTD_count(ip+minMatch, ip+minMatch-repCur, iend) + minMatch;
- if (mlen > sufficient_len || mlen >= ZSTD_OPT_NUM) {
- best_mlen = mlen; best_off = i; cur = 0; last_pos = 1;
- goto _storeSequence;
- }
- best_off = i - (ip == anchor);
- do {
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
- if (mlen > last_pos || price < opt[mlen].price)
- SET_PRICE(mlen, mlen, i, litlen, price); /* note : macro modifies last_pos */
- mlen--;
- } while (mlen >= minMatch);
- } } }
-
- match_num = ZSTD_BtGetAllMatches_selectMLS(ctx, ip, iend, maxSearches, mls, matches, minMatch);
-
- if (!last_pos && !match_num) { ip++; continue; }
-
- if (match_num && (matches[match_num-1].len > sufficient_len || matches[match_num-1].len >= ZSTD_OPT_NUM)) {
- best_mlen = matches[match_num-1].len;
- best_off = matches[match_num-1].off;
- cur = 0;
- last_pos = 1;
- goto _storeSequence;
- }
-
- /* set prices using matches at position = 0 */
- best_mlen = (last_pos) ? last_pos : minMatch;
- for (u = 0; u < match_num; u++) {
- mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
- best_mlen = matches[u].len;
- while (mlen <= best_mlen) {
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
- if (mlen > last_pos || price < opt[mlen].price)
- SET_PRICE(mlen, mlen, matches[u].off, litlen, price); /* note : macro modifies last_pos */
- mlen++;
- } }
-
- if (last_pos < minMatch) { ip++; continue; }
-
- /* initialize opt[0] */
- { U32 i ; for (i=0; i litlen) {
- price = opt[cur - litlen].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-litlen);
- } else
- price = ZSTD_getLiteralPrice(optStatePtr, litlen, anchor);
- } else {
- litlen = 1;
- price = opt[cur - 1].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-1);
- }
-
- if (cur > last_pos || price <= opt[cur].price)
- SET_PRICE(cur, 1, 0, litlen, price);
-
- if (cur == last_pos) break;
-
- if (inr > ilimit) /* last match must start at a minimum distance of 8 from oend */
- continue;
-
- mlen = opt[cur].mlen;
- if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
- opt[cur].rep[2] = opt[cur-mlen].rep[1];
- opt[cur].rep[1] = opt[cur-mlen].rep[0];
- opt[cur].rep[0] = opt[cur].off - ZSTD_REP_MOVE_OPT;
- } else {
- opt[cur].rep[2] = (opt[cur].off > 1) ? opt[cur-mlen].rep[1] : opt[cur-mlen].rep[2];
- opt[cur].rep[1] = (opt[cur].off > 0) ? opt[cur-mlen].rep[0] : opt[cur-mlen].rep[1];
- opt[cur].rep[0] = ((opt[cur].off==ZSTD_REP_MOVE_OPT) && (mlen != 1)) ? (opt[cur-mlen].rep[0] - 1) : (opt[cur-mlen].rep[opt[cur].off]);
- }
-
- best_mlen = minMatch;
- { U32 i, last_i = ZSTD_REP_CHECK + (mlen != 1);
- for (i=(opt[cur].mlen != 1); i 0) && (repCur < (S32)(inr-prefixStart))
- && (ZSTD_readMINMATCH(inr, minMatch) == ZSTD_readMINMATCH(inr - repCur, minMatch))) {
- mlen = (U32)ZSTD_count(inr+minMatch, inr+minMatch - repCur, iend) + minMatch;
-
- if (mlen > sufficient_len || cur + mlen >= ZSTD_OPT_NUM) {
- best_mlen = mlen; best_off = i; last_pos = cur + 1;
- goto _storeSequence;
- }
-
- best_off = i - (opt[cur].mlen != 1);
- if (mlen > best_mlen) best_mlen = mlen;
-
- do {
- if (opt[cur].mlen == 1) {
- litlen = opt[cur].litlen;
- if (cur > litlen) {
- price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, inr-litlen, best_off, mlen - MINMATCH, ultra);
- } else
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
- } else {
- litlen = 0;
- price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, best_off, mlen - MINMATCH, ultra);
- }
-
- if (cur + mlen > last_pos || price <= opt[cur + mlen].price)
- SET_PRICE(cur + mlen, mlen, i, litlen, price);
- mlen--;
- } while (mlen >= minMatch);
- } } }
-
- match_num = ZSTD_BtGetAllMatches_selectMLS(ctx, inr, iend, maxSearches, mls, matches, best_mlen);
-
- if (match_num > 0 && (matches[match_num-1].len > sufficient_len || cur + matches[match_num-1].len >= ZSTD_OPT_NUM)) {
- best_mlen = matches[match_num-1].len;
- best_off = matches[match_num-1].off;
- last_pos = cur + 1;
- goto _storeSequence;
- }
-
- /* set prices using matches at position = cur */
- for (u = 0; u < match_num; u++) {
- mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
- best_mlen = matches[u].len;
-
- while (mlen <= best_mlen) {
- if (opt[cur].mlen == 1) {
- litlen = opt[cur].litlen;
- if (cur > litlen)
- price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, ip+cur-litlen, matches[u].off-1, mlen - MINMATCH, ultra);
- else
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
- } else {
- litlen = 0;
- price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, matches[u].off-1, mlen - MINMATCH, ultra);
- }
-
- if (cur + mlen > last_pos || (price < opt[cur + mlen].price))
- SET_PRICE(cur + mlen, mlen, matches[u].off, litlen, price);
-
- mlen++;
- } } }
-
- best_mlen = opt[last_pos].mlen;
- best_off = opt[last_pos].off;
- cur = last_pos - best_mlen;
-
- /* store sequence */
-_storeSequence: /* cur, last_pos, best_mlen, best_off have to be set */
- opt[0].mlen = 1;
-
- while (1) {
- mlen = opt[cur].mlen;
- offset = opt[cur].off;
- opt[cur].mlen = best_mlen;
- opt[cur].off = best_off;
- best_mlen = mlen;
- best_off = offset;
- if (mlen > cur) break;
- cur -= mlen;
- }
-
- for (u = 0; u <= last_pos;) {
- u += opt[u].mlen;
- }
-
- for (cur=0; cur < last_pos; ) {
- mlen = opt[cur].mlen;
- if (mlen == 1) { ip++; cur++; continue; }
- offset = opt[cur].off;
- cur += mlen;
- litLength = (U32)(ip - anchor);
-
- if (offset > ZSTD_REP_MOVE_OPT) {
- rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = offset - ZSTD_REP_MOVE_OPT;
- offset--;
- } else {
- if (offset != 0) {
- best_off = (offset==ZSTD_REP_MOVE_OPT) ? (rep[0] - 1) : (rep[offset]);
- if (offset != 1) rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = best_off;
- }
- if (litLength==0) offset--;
- }
-
- ZSTD_updatePrice(optStatePtr, litLength, anchor, offset, mlen-MINMATCH);
- ZSTD_storeSeq(seqStorePtr, litLength, anchor, offset, mlen-MINMATCH);
- anchor = ip = ip + mlen;
- } } /* for (cur=0; cur < last_pos; ) */
-
- /* Save reps for next block */
- { int i; for (i=0; irepToConfirm[i] = rep[i]; }
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-
-FORCE_INLINE_TEMPLATE
-size_t ZSTD_compressBlock_opt_extDict_generic(ZSTD_CCtx* ctx,
- const void* src, size_t srcSize, const int ultra)
-{
- seqStore_t* seqStorePtr = &(ctx->seqStore);
- optState_t* optStatePtr = &(ctx->optState);
- const BYTE* const istart = (const BYTE*)src;
- const BYTE* ip = istart;
- const BYTE* anchor = istart;
- const BYTE* const iend = istart + srcSize;
- const BYTE* const ilimit = iend - 8;
- const BYTE* const base = ctx->base;
- const U32 lowestIndex = ctx->lowLimit;
- const U32 dictLimit = ctx->dictLimit;
- const BYTE* const prefixStart = base + dictLimit;
- const BYTE* const dictBase = ctx->dictBase;
- const BYTE* const dictEnd = dictBase + dictLimit;
-
- const U32 maxSearches = 1U << ctx->appliedParams.cParams.searchLog;
- const U32 sufficient_len = ctx->appliedParams.cParams.targetLength;
- const U32 mls = ctx->appliedParams.cParams.searchLength;
- const U32 minMatch = (ctx->appliedParams.cParams.searchLength == 3) ? 3 : 4;
-
- ZSTD_optimal_t* opt = optStatePtr->priceTable;
- ZSTD_match_t* matches = optStatePtr->matchTable;
- const BYTE* inr;
-
- /* init */
- U32 offset, rep[ZSTD_REP_NUM];
- { U32 i; for (i=0; irep[i]; }
-
- ctx->nextToUpdate3 = ctx->nextToUpdate;
- ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize);
- ip += (ip==prefixStart);
-
- /* Match Loop */
- while (ip < ilimit) {
- U32 cur, match_num, last_pos, litlen, price;
- U32 u, mlen, best_mlen, best_off, litLength;
- U32 current = (U32)(ip-base);
- memset(opt, 0, sizeof(ZSTD_optimal_t));
- last_pos = 0;
- opt[0].litlen = (U32)(ip - anchor);
-
- /* check repCode */
- { U32 i, last_i = ZSTD_REP_CHECK + (ip==anchor);
- for (i = (ip==anchor); i 0 && repCur <= (S32)current)
- && (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex>lowestIndex)) /* intentional overflow */
- && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
- /* repcode detected we should take it */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- mlen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iend, repEnd, prefixStart) + minMatch;
-
- if (mlen > sufficient_len || mlen >= ZSTD_OPT_NUM) {
- best_mlen = mlen; best_off = i; cur = 0; last_pos = 1;
- goto _storeSequence;
- }
-
- best_off = i - (ip==anchor);
- litlen = opt[0].litlen;
- do {
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
- if (mlen > last_pos || price < opt[mlen].price)
- SET_PRICE(mlen, mlen, i, litlen, price); /* note : macro modifies last_pos */
- mlen--;
- } while (mlen >= minMatch);
- } } }
-
- match_num = ZSTD_BtGetAllMatches_selectMLS_extDict(ctx, ip, iend, maxSearches, mls, matches, minMatch); /* first search (depth 0) */
-
- if (!last_pos && !match_num) { ip++; continue; }
-
- { U32 i; for (i=0; i sufficient_len || matches[match_num-1].len >= ZSTD_OPT_NUM)) {
- best_mlen = matches[match_num-1].len;
- best_off = matches[match_num-1].off;
- cur = 0;
- last_pos = 1;
- goto _storeSequence;
- }
-
- best_mlen = (last_pos) ? last_pos : minMatch;
-
- /* set prices using matches at position = 0 */
- for (u = 0; u < match_num; u++) {
- mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
- best_mlen = matches[u].len;
- litlen = opt[0].litlen;
- while (mlen <= best_mlen) {
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
- if (mlen > last_pos || price < opt[mlen].price)
- SET_PRICE(mlen, mlen, matches[u].off, litlen, price);
- mlen++;
- } }
-
- if (last_pos < minMatch) {
- ip++; continue;
- }
-
- /* check further positions */
- for (cur = 1; cur <= last_pos; cur++) {
- inr = ip + cur;
-
- if (opt[cur-1].mlen == 1) {
- litlen = opt[cur-1].litlen + 1;
- if (cur > litlen) {
- price = opt[cur - litlen].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-litlen);
- } else
- price = ZSTD_getLiteralPrice(optStatePtr, litlen, anchor);
- } else {
- litlen = 1;
- price = opt[cur - 1].price + ZSTD_getLiteralPrice(optStatePtr, litlen, inr-1);
- }
-
- if (cur > last_pos || price <= opt[cur].price)
- SET_PRICE(cur, 1, 0, litlen, price);
-
- if (cur == last_pos) break;
-
- if (inr > ilimit) /* last match must start at a minimum distance of 8 from oend */
- continue;
-
- mlen = opt[cur].mlen;
- if (opt[cur].off > ZSTD_REP_MOVE_OPT) {
- opt[cur].rep[2] = opt[cur-mlen].rep[1];
- opt[cur].rep[1] = opt[cur-mlen].rep[0];
- opt[cur].rep[0] = opt[cur].off - ZSTD_REP_MOVE_OPT;
- } else {
- opt[cur].rep[2] = (opt[cur].off > 1) ? opt[cur-mlen].rep[1] : opt[cur-mlen].rep[2];
- opt[cur].rep[1] = (opt[cur].off > 0) ? opt[cur-mlen].rep[0] : opt[cur-mlen].rep[1];
- opt[cur].rep[0] = ((opt[cur].off==ZSTD_REP_MOVE_OPT) && (mlen != 1)) ? (opt[cur-mlen].rep[0] - 1) : (opt[cur-mlen].rep[opt[cur].off]);
- }
-
- best_mlen = minMatch;
- { U32 i, last_i = ZSTD_REP_CHECK + (mlen != 1);
- for (i = (mlen != 1); i 0 && repCur <= (S32)(current+cur))
- && (((U32)((dictLimit-1) - repIndex) >= 3) & (repIndex>lowestIndex)) /* intentional overflow */
- && (ZSTD_readMINMATCH(inr, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) {
- /* repcode detected */
- const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend;
- mlen = (U32)ZSTD_count_2segments(inr+minMatch, repMatch+minMatch, iend, repEnd, prefixStart) + minMatch;
-
- if (mlen > sufficient_len || cur + mlen >= ZSTD_OPT_NUM) {
- best_mlen = mlen; best_off = i; last_pos = cur + 1;
- goto _storeSequence;
- }
-
- best_off = i - (opt[cur].mlen != 1);
- if (mlen > best_mlen) best_mlen = mlen;
-
- do {
- if (opt[cur].mlen == 1) {
- litlen = opt[cur].litlen;
- if (cur > litlen) {
- price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, inr-litlen, best_off, mlen - MINMATCH, ultra);
- } else
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, best_off, mlen - MINMATCH, ultra);
- } else {
- litlen = 0;
- price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, best_off, mlen - MINMATCH, ultra);
- }
-
- if (cur + mlen > last_pos || price <= opt[cur + mlen].price)
- SET_PRICE(cur + mlen, mlen, i, litlen, price);
- mlen--;
- } while (mlen >= minMatch);
- } } }
-
- match_num = ZSTD_BtGetAllMatches_selectMLS_extDict(ctx, inr, iend, maxSearches, mls, matches, minMatch);
-
- if (match_num > 0 && (matches[match_num-1].len > sufficient_len || cur + matches[match_num-1].len >= ZSTD_OPT_NUM)) {
- best_mlen = matches[match_num-1].len;
- best_off = matches[match_num-1].off;
- last_pos = cur + 1;
- goto _storeSequence;
- }
-
- /* set prices using matches at position = cur */
- for (u = 0; u < match_num; u++) {
- mlen = (u>0) ? matches[u-1].len+1 : best_mlen;
- best_mlen = matches[u].len;
-
- while (mlen <= best_mlen) {
- if (opt[cur].mlen == 1) {
- litlen = opt[cur].litlen;
- if (cur > litlen)
- price = opt[cur - litlen].price + ZSTD_getPrice(optStatePtr, litlen, ip+cur-litlen, matches[u].off-1, mlen - MINMATCH, ultra);
- else
- price = ZSTD_getPrice(optStatePtr, litlen, anchor, matches[u].off-1, mlen - MINMATCH, ultra);
- } else {
- litlen = 0;
- price = opt[cur].price + ZSTD_getPrice(optStatePtr, 0, NULL, matches[u].off-1, mlen - MINMATCH, ultra);
- }
-
- if (cur + mlen > last_pos || (price < opt[cur + mlen].price))
- SET_PRICE(cur + mlen, mlen, matches[u].off, litlen, price);
-
- mlen++;
- } } } /* for (cur = 1; cur <= last_pos; cur++) */
-
- best_mlen = opt[last_pos].mlen;
- best_off = opt[last_pos].off;
- cur = last_pos - best_mlen;
-
- /* store sequence */
-_storeSequence: /* cur, last_pos, best_mlen, best_off have to be set */
- opt[0].mlen = 1;
-
- while (1) {
- mlen = opt[cur].mlen;
- offset = opt[cur].off;
- opt[cur].mlen = best_mlen;
- opt[cur].off = best_off;
- best_mlen = mlen;
- best_off = offset;
- if (mlen > cur) break;
- cur -= mlen;
- }
-
- for (u = 0; u <= last_pos; ) {
- u += opt[u].mlen;
- }
-
- for (cur=0; cur < last_pos; ) {
- mlen = opt[cur].mlen;
- if (mlen == 1) { ip++; cur++; continue; }
- offset = opt[cur].off;
- cur += mlen;
- litLength = (U32)(ip - anchor);
-
- if (offset > ZSTD_REP_MOVE_OPT) {
- rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = offset - ZSTD_REP_MOVE_OPT;
- offset--;
- } else {
- if (offset != 0) {
- best_off = (offset==ZSTD_REP_MOVE_OPT) ? (rep[0] - 1) : (rep[offset]);
- if (offset != 1) rep[2] = rep[1];
- rep[1] = rep[0];
- rep[0] = best_off;
- }
-
- if (litLength==0) offset--;
- }
-
- ZSTD_updatePrice(optStatePtr, litLength, anchor, offset, mlen-MINMATCH);
- ZSTD_storeSeq(seqStorePtr, litLength, anchor, offset, mlen-MINMATCH);
- anchor = ip = ip + mlen;
- } } /* for (cur=0; cur < last_pos; ) */
-
- /* Save reps for next block */
- { int i; for (i=0; irepToConfirm[i] = rep[i]; }
-
- /* Return the last literals size */
- return iend - anchor;
-}
-
-#endif /* ZSTD_OPT_H_91842398743 */
+#endif /* ZSTD_OPT_H */
diff --git a/lib/compress/zstdmt_compress.c b/lib/compress/zstdmt_compress.c
index ac451c35e..a725e66b5 100644
--- a/lib/compress/zstdmt_compress.c
+++ b/lib/compress/zstdmt_compress.c
@@ -528,7 +528,7 @@ size_t ZSTDMT_CCtxParam_setMTCtxParameter(
params->jobSize = value;
return 0;
case ZSTDMT_p_overlapSectionLog :
- DEBUGLOG(5, "ZSTDMT_p_overlapSectionLog : %u", value);
+ DEBUGLOG(4, "ZSTDMT_p_overlapSectionLog : %u", value);
params->overlapSizeLog = (value >= 9) ? 9 : value;
return 0;
default :
diff --git a/lib/decompress/zstd_decompress.c b/lib/decompress/zstd_decompress.c
index 8e82d85a2..ada773b9a 100644
--- a/lib/decompress/zstd_decompress.c
+++ b/lib/decompress/zstd_decompress.c
@@ -1088,7 +1088,10 @@ static size_t ZSTD_decompressSequences(
}
-FORCE_INLINE_TEMPLATE seq_t ZSTD_decodeSequenceLong_generic(seqState_t* seqState, int const longOffsets)
+typedef enum { ZSTD_lo_isRegularOffset, ZSTD_lo_isLongOffset=1 } ZSTD_longOffset_e;
+
+HINT_INLINE
+seq_t ZSTD_decodeSequenceLong(seqState_t* seqState, ZSTD_longOffset_e const longOffsets)
{
seq_t seq;
@@ -1180,19 +1183,12 @@ FORCE_INLINE_TEMPLATE seq_t ZSTD_decodeSequenceLong_generic(seqState_t* seqState
return seq;
}
-static seq_t ZSTD_decodeSequenceLong(seqState_t* seqState, unsigned const windowSize) {
- if (ZSTD_highbit32(windowSize) > STREAM_ACCUMULATOR_MIN) {
- return ZSTD_decodeSequenceLong_generic(seqState, 1);
- } else {
- return ZSTD_decodeSequenceLong_generic(seqState, 0);
- }
-}
HINT_INLINE
size_t ZSTD_execSequenceLong(BYTE* op,
- BYTE* const oend, seq_t sequence,
- const BYTE** litPtr, const BYTE* const litLimit,
- const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
+ BYTE* const oend, seq_t sequence,
+ const BYTE** litPtr, const BYTE* const litLimit,
+ const BYTE* const base, const BYTE* const vBase, const BYTE* const dictEnd)
{
BYTE* const oLitEnd = op + sequence.litLength;
size_t const sequenceLength = sequence.litLength + sequence.matchLength;
@@ -1202,11 +1198,9 @@ size_t ZSTD_execSequenceLong(BYTE* op,
const BYTE* match = sequence.match;
/* check */
-#if 1
if (oMatchEnd>oend) return ERROR(dstSize_tooSmall); /* last match must start at a minimum distance of WILDCOPY_OVERLENGTH from oend */
if (iLitEnd > litLimit) return ERROR(corruption_detected); /* over-read beyond lit buffer */
if (oLitEnd>oend_w) return ZSTD_execSequenceLast7(op, oend, sequence, litPtr, litLimit, base, vBase, dictEnd);
-#endif
/* copy Literals */
ZSTD_copy8(op, *litPtr);
@@ -1216,7 +1210,6 @@ size_t ZSTD_execSequenceLong(BYTE* op,
*litPtr = iLitEnd; /* update for next sequence */
/* copy Match */
-#if 1
if (sequence.offset > (size_t)(oLitEnd - base)) {
/* offset beyond prefix */
if (sequence.offset > (size_t)(oLitEnd - vBase)) return ERROR(corruption_detected);
@@ -1236,8 +1229,8 @@ size_t ZSTD_execSequenceLong(BYTE* op,
return sequenceLength;
}
} }
- /* Requirement: op <= oend_w && sequence.matchLength >= MINMATCH */
-#endif
+ assert(op <= oend_w);
+ assert(sequence.matchLength >= MINMATCH);
/* match within prefix */
if (sequence.offset < 8) {
@@ -1285,9 +1278,12 @@ static size_t ZSTD_decompressSequencesLong(
const BYTE* const base = (const BYTE*) (dctx->base);
const BYTE* const vBase = (const BYTE*) (dctx->vBase);
const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd);
- unsigned const windowSize32 = (unsigned)dctx->fParams.windowSize;
int nbSeq;
+ unsigned long long const regularWindowSizeMax = 1ULL << STREAM_ACCUMULATOR_MIN;
+ ZSTD_longOffset_e const isLongOffset = (ZSTD_longOffset_e)(MEM_32bits() && (dctx->fParams.windowSize >= regularWindowSizeMax));
+ ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1);
+
/* Build Decoding Tables */
{ size_t const seqHSize = ZSTD_decodeSeqHeaders(dctx, &nbSeq, ip, seqSize);
if (ZSTD_isError(seqHSize)) return seqHSize;
@@ -1315,13 +1311,13 @@ static size_t ZSTD_decompressSequencesLong(
/* prepare in advance */
for (seqNb=0; (BIT_reloadDStream(&seqState.DStream) <= BIT_DStream_completed) && seqNb 4) /* do not enable prefetching on 32-bits x86, as it's performance detrimental */
- /* likely because of register pressure */
- /* if that's the correct cause, then 32-bits ARM should be affected differently */
- /* it would be good to test this on ARM real hardware, to see if prefetch version improves speed */
- if (dctx->fParams.windowSize > (1<<23))
- return ZSTD_decompressSequencesLong(dctx, dst, dstCapacity, ip, srcSize);
+ if (dctx->fParams.windowSize > (1<<23))
+ return ZSTD_decompressSequencesLong(dctx, dst, dstCapacity, ip, srcSize);
return ZSTD_decompressSequences(dctx, dst, dstCapacity, ip, srcSize);
}
diff --git a/programs/bench.c b/programs/bench.c
index 68ac8c886..ac76a0023 100644
--- a/programs/bench.c
+++ b/programs/bench.c
@@ -407,10 +407,8 @@ static int BMK_benchMem(const void* srcBuffer, size_t srcSize,
blockTable[blockNb].cPtr, blockTable[blockNb].cSize,
ddict);
if (ZSTD_isError(regenSize)) {
- DISPLAY("ZSTD_decompress_usingDDict() failed on block %u of size %u : %s \n",
+ EXM_THROW(2, "ZSTD_decompress_usingDDict() failed on block %u of size %u : %s \n",
blockNb, (U32)blockTable[blockNb].cSize, ZSTD_getErrorName(regenSize));
- clockLoop = 0; /* force immediate test end */
- break;
}
blockTable[blockNb].resSize = regenSize;
}
diff --git a/tests/fuzz/Makefile b/tests/fuzz/Makefile
index c6327068f..dfb8f1913 100644
--- a/tests/fuzz/Makefile
+++ b/tests/fuzz/Makefile
@@ -48,7 +48,11 @@ LIBFUZZER ?= -lFuzzer
default: all
-all: round_trip simple_decompress
+all: \
+ simple_round_trip \
+ stream_round_trip \
+ simple_decompress \
+ stream_decompress
%.o: %.c
$(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $^ -c -o $@
diff --git a/tests/fuzz/regression_driver.c b/tests/fuzz/regression_driver.c
index 41fdcc940..36ae3884f 100644
--- a/tests/fuzz/regression_driver.c
+++ b/tests/fuzz/regression_driver.c
@@ -39,7 +39,7 @@ int main(int argc, char const **argv) {
FILE *file;
/* Check that it is a regular file, and that the fileSize is valid */
- FUZZ_ASSERT_MSG(UTIL_isRegFile(fileName), fileName);
+ FUZZ_ASSERT_MSG(UTIL_isRegularFile(fileName), fileName);
FUZZ_ASSERT_MSG(fileSize <= kMaxFileSize, fileName);
/* Ensure we have a large enough buffer allocated */
if (fileSize > bufferSize) {
diff --git a/tests/fuzz/stream_decompress.c b/tests/fuzz/stream_decompress.c
index ae853dc8c..dfd746972 100644
--- a/tests/fuzz/stream_decompress.c
+++ b/tests/fuzz/stream_decompress.c
@@ -20,7 +20,7 @@
#include "fuzz_helpers.h"
#include "zstd.h"
-static size_t const kBufSize = ZSTD_BLOCKSIZE_ABSOLUTEMAX;
+static size_t const kBufSize = ZSTD_BLOCKSIZE_MAX;
static ZSTD_DStream *dstream = NULL;
static void* buf = NULL;