size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity,
+ZSTDLIB_STATIC_API size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity,
const void* src, size_t srcSize, unsigned magicVariant);
Generates a zstd skippable frame containing data given by src, and writes it to dst buffer.
@@ -1230,10 +1230,10 @@
Streaming decompression functions
Memory management
-size_t ZSTD_estimateCCtxSize(int compressionLevel);
-size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams);
-size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params);
-size_t ZSTD_estimateDCtxSize(void);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize(int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateDCtxSize(void);
These functions make it possible to estimate memory usage
of a future {D,C}Ctx, before its creation.
@@ -1258,11 +1258,11 @@
Streaming decompression functions
-size_t ZSTD_estimateCStreamSize(int compressionLevel);
-size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams);
-size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params);
-size_t ZSTD_estimateDStreamSize(size_t windowSize);
-size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize(int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateDStreamSize(size_t windowSize);
+ZSTDLIB_STATIC_API 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_usingCParams() can provide a tighter estimation.
@@ -1277,17 +1277,17 @@
Streaming decompression functions
In this case, get total size by adding ZSTD_estimate?DictSize
-size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel);
-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);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod);
+ZSTDLIB_STATIC_API 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_estimateCDictSize_advanced() makes it possible to control compression parameters precisely, like ZSTD_createCDict_advanced().
Note : dictionaries created by reference (`ZSTD_dlm_byRef`) are logically smaller.
-ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize);
-ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticCCtx() */
+ZSTDLIB_STATIC_API ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize);
+ZSTDLIB_STATIC_API ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticCCtx() */
Initialize an object using a pre-allocated fixed-size buffer.
workspace: The memory area to emplace the object into.
Provided pointer *must be 8-bytes aligned*.
@@ -1310,7 +1310,7 @@
Streaming decompression functions
-ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticDCtx() */
+ZSTDLIB_STATIC_API ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticDCtx() */
typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size);
typedef void (*ZSTD_freeFunction) (void* opaque, void* address);
@@ -1327,9 +1327,9 @@ Streaming decompression functions
typedef struct POOL_ctx_s ZSTD_threadPool;
-ZSTD_threadPool* ZSTD_createThreadPool(size_t numThreads);
-void ZSTD_freeThreadPool (ZSTD_threadPool* pool); /* accept NULL pointer */
-size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool);
+ZSTDLIB_STATIC_API ZSTD_threadPool* ZSTD_createThreadPool(size_t numThreads);
+ZSTDLIB_STATIC_API void ZSTD_freeThreadPool (ZSTD_threadPool* pool); /* accept NULL pointer */
+ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool);
These prototypes make it possible to share a thread pool among multiple compression contexts.
This can limit resources for applications with multiple threads where each one uses
a threaded compression mode (via ZSTD_c_nbWorkers parameter).
@@ -1343,7 +1343,7 @@
Streaming decompression functions
Advanced compression functions
-ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel);
+ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel);
Create a digested dictionary for compression
Dictionary content is just referenced, not duplicated.
As a consequence, `dictBuffer` **must** outlive CDict,
@@ -1351,22 +1351,22 @@
Streaming decompression functions
note: equivalent to ZSTD_createCDict_advanced(), with dictLoadMethod==ZSTD_dlm_byRef
-ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
+ZSTDLIB_STATIC_API ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
@return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize.
`estimatedSrcSize` value is optional, select 0 if not known
-ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
+ZSTDLIB_STATIC_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize);
same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`.
All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0
-size_t ZSTD_checkCParams(ZSTD_compressionParameters params);
+ZSTDLIB_STATIC_API size_t ZSTD_checkCParams(ZSTD_compressionParameters params);
Ensure param values remain within authorized range.
@return 0 on success, or an error code (can be checked with ZSTD_isError())
-ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize);
+ZSTDLIB_STATIC_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize);
optimize params for a given `srcSize` and `dictSize`.
`srcSize` can be unknown, in which case use ZSTD_CONTENTSIZE_UNKNOWN.
`dictSize` must be `0` when there is no dictionary.
@@ -1396,31 +1396,31 @@
Streaming decompression functions
This prototype will generate compilation warnings.
-size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize);
Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx.
It saves some memory, but also requires that `dict` outlives its usage within `cctx`
-size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
Same as ZSTD_CCtx_loadDictionary(), but gives finer control over
how to load the dictionary (by copy ? by reference ?)
and how to interpret it (automatic ? force raw mode ? full mode only ?)
-size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
Same as ZSTD_CCtx_refPrefix(), but gives finer control over
how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?)
-size_t ZSTD_CCtx_getParameter(const ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtx_getParameter(const ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value);
Get the requested compression parameter value, selected by enum ZSTD_cParameter,
and store it into int* value.
@return : 0, or an error code (which can be tested with ZSTD_isError()).
-ZSTD_CCtx_params* ZSTD_createCCtxParams(void);
-size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params); /* accept NULL pointer */
+ZSTDLIB_STATIC_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void);
+ZSTDLIB_STATIC_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params); /* accept NULL pointer */
Quick howto :
- ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure
- ZSTD_CCtxParams_setParameter() : Push parameters one by one into
@@ -1439,24 +1439,24 @@
Streaming decompression functions
-size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params);
Reset params to default values.
-size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init(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_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init_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_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value);
Similar to ZSTD_CCtx_setParameter.
Set one compression parameter, selected by enum ZSTD_cParameter.
Parameters must be applied to a ZSTD_CCtx using
@@ -1466,14 +1466,14 @@
Streaming decompression functions
-size_t ZSTD_CCtxParams_getParameter(const ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value);
+ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_getParameter(const ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value);
Similar to ZSTD_CCtx_getParameter.
Get the requested value of one compression parameter, selected by enum ZSTD_cParameter.
@result : 0, or an error code (which can be tested with ZSTD_isError()).
-size_t ZSTD_CCtx_setParametersUsingCCtxParams(
+ZSTDLIB_STATIC_API 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 can be done even after compression is started,
@@ -1483,7 +1483,7 @@
Streaming decompression functions
-size_t ZSTD_compressStream2_simpleArgs (
+ZSTDLIB_STATIC_API size_t ZSTD_compressStream2_simpleArgs (
ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity, size_t* dstPos,
const void* src, size_t srcSize, size_t* srcPos,
@@ -1497,40 +1497,40 @@ Streaming decompression functions
Advanced decompression functions
-unsigned ZSTD_isFrame(const void* buffer, size_t size);
+ZSTDLIB_STATIC_API unsigned ZSTD_isFrame(const void* buffer, size_t size);
Tells if the content of `buffer` starts with a valid Frame Identifier.
Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0.
Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled.
Note 3 : Skippable Frame Identifiers are considered valid.
-ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize);
+ZSTDLIB_STATIC_API ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize);
Create a digested dictionary, ready to start decompression operation without startup delay.
Dictionary content is referenced, and therefore stays in dictBuffer.
It is important that dictBuffer outlives DDict,
it must remain read accessible throughout the lifetime of DDict
-size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
+ZSTDLIB_STATIC_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize);
Same as ZSTD_DCtx_loadDictionary(),
but references `dict` content instead of copying it into `dctx`.
This saves memory if `dict` remains around.,
However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression.
-size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
+ZSTDLIB_STATIC_API size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType);
Same as ZSTD_DCtx_loadDictionary(),
but gives direct control over
how to load the dictionary (by copy ? by reference ?)
and how to interpret it (automatic ? force raw mode ? full mode only ?).
-size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
+ZSTDLIB_STATIC_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType);
Same as ZSTD_DCtx_refPrefix(), but gives finer control over
how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?)
-size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize);
+ZSTDLIB_STATIC_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize);
Refuses allocating internal buffers for frames requiring a window size larger than provided limit.
This protects a decoder context from reserving too much memory for itself (potential attack scenario).
This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode.
@@ -1539,7 +1539,7 @@
Streaming decompression functions
-size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value);
+ZSTDLIB_STATIC_API size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value);
Get the requested decompression parameter value, selected by enum ZSTD_dParameter,
and store it into int* value.
@return : 0, or an error code (which can be tested with ZSTD_isError()).
@@ -1555,7 +1555,7 @@
Streaming decompression functions
@return : 0, or an error code (which can be tested using ZSTD_isError()).
-size_t ZSTD_decompressStream_simpleArgs (
+ZSTDLIB_STATIC_API size_t ZSTD_decompressStream_simpleArgs (
ZSTD_DCtx* dctx,
void* dst, size_t dstCapacity, size_t* dstPos,
const void* src, size_t srcSize, size_t* srcPos);
@@ -1689,7 +1689,7 @@ Advanced Streaming compression functions
/* MT only : nb of workers actively compressing at probe time */
} ZSTD_frameProgression;
-size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx);
+ZSTDLIB_STATIC_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx);
Tell how many bytes are ready to be flushed immediately.
Useful for multithreading scenarios (nbWorkers >= 1).
Probe the oldest active job, defined as oldest job not yet entirely flushed,
@@ -1705,7 +1705,7 @@
Advanced Streaming compression functions
Advanced Streaming decompression functions
-size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize);
+ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize);
ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
ZSTD_DCtx_loadDictionary(zds, dict, dictSize);
@@ -1715,7 +1715,7 @@
Advanced Streaming decompression functions
-size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict);
+ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict);
ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
ZSTD_DCtx_refDDict(zds, ddict);
@@ -1725,7 +1725,7 @@
Advanced Streaming decompression functions
-size_t ZSTD_resetDStream(ZSTD_DStream* zds);
+ZSTDLIB_STATIC_API size_t ZSTD_resetDStream(ZSTD_DStream* zds);
ZSTD_DCtx_reset(zds, ZSTD_reset_session_only);
@@ -1769,10 +1769,10 @@
Advanced Streaming decompression functions
`ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress again.
-Buffer-less streaming compression functions
size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel);
-size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel);
-size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
/**< note: fails if cdict==NULL */
-size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx, unsigned long long pledgedSrcSize); /**< note: if pledgedSrcSize is not known, use ZSTD_CONTENTSIZE_UNKNOWN */
+Buffer-less streaming compression functions
ZSTDLIB_STATIC_API size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel);
+ZSTDLIB_STATIC_API size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict);
/**< note: fails if cdict==NULL */
+ZSTDLIB_STATIC_API size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx, unsigned long long pledgedSrcSize); /**< note: if pledgedSrcSize is not known, use ZSTD_CONTENTSIZE_UNKNOWN */
size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */
@@ -1858,12 +1858,12 @@ Buffer-less streaming decompression functions
typedef
unsigned checksumFlag;
} ZSTD_frameHeader;
-size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */
+ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */
/*! ZSTD_getFrameHeader_advanced() :
* same as ZSTD_getFrameHeader(),
* with added capability to select a format (like ZSTD_f_zstd1_magicless) */
-size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format);
-size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize); /**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */
+ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format);
+ZSTDLIB_STATIC_API size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize); /**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */
decode Frame Header, or requires larger `srcSize`.
@return : 0, `zfhPtr` is correctly filled,
>0, `srcSize` is too small, value is wanted `srcSize` amount,
@@ -1899,10 +1899,10 @@
Buffer-less streaming decompression functions
typedef
Use ZSTD_insertBlock() for such a case.
-Raw zstd block functions
size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx);
-size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
-size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
-size_t ZSTD_insertBlock (ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize);
/**< insert uncompressed block into `dctx` history. Useful for multi-blocks decompression. */
+Raw zstd block functions
ZSTDLIB_STATIC_API size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx);
+ZSTDLIB_STATIC_API size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
+ZSTDLIB_STATIC_API size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize);
+ZSTDLIB_STATIC_API size_t ZSTD_insertBlock (ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize);
/**< insert uncompressed block into `dctx` history. Useful for multi-blocks decompression. */
diff --git a/examples/streaming_compression_thread_pool.c b/examples/streaming_compression_thread_pool.c
index 471ca863136..21cb3d54999 100644
--- a/examples/streaming_compression_thread_pool.c
+++ b/examples/streaming_compression_thread_pool.c
@@ -60,7 +60,7 @@ static void *compressFile_orDie(void *data)
CHECK_ZSTD( ZSTD_CCtx_setParameter(cctx, ZSTD_c_checksumFlag, 1) );
ZSTD_CCtx_setParameter(cctx, ZSTD_c_nbWorkers, nbThreads);
- /* This loop read from the input file, compresses that entire chunk,
+ /* This loop reads from the input file, compresses that entire chunk,
* and writes all output produced to the output file.
*/
size_t const toRead = buffInSize;
diff --git a/lib/Makefile b/lib/Makefile
index 00fc535b0e9..ef202183d84 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -212,11 +212,11 @@ $(ZSTD_STATLIB_DIR)/%.o : %.c $(ZSTD_STATLIB_DIR)/%.d | $(ZSTD_STATLIB_DIR)
$(ZSTD_DYNLIB_DIR)/%.o : %.S | $(ZSTD_DYNLIB_DIR)
@echo AS $@
- $(COMPILE.c) $(OUTPUT_OPTION) $<
+ $(COMPILE.S) $(OUTPUT_OPTION) $<
$(ZSTD_STATLIB_DIR)/%.o : %.S | $(ZSTD_STATLIB_DIR)
@echo AS $@
- $(COMPILE.c) $(OUTPUT_OPTION) $<
+ $(COMPILE.S) $(OUTPUT_OPTION) $<
MKDIR ?= mkdir
$(BUILD_DIR) $(ZSTD_DYNLIB_DIR) $(ZSTD_STATLIB_DIR):
diff --git a/lib/common/pool.c b/lib/common/pool.c
index ea70b8b65ad..2e37cdd73c8 100644
--- a/lib/common/pool.c
+++ b/lib/common/pool.c
@@ -86,7 +86,7 @@ static void* POOL_thread(void* opaque) {
{ POOL_job const job = ctx->queue[ctx->queueHead];
ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize;
ctx->numThreadsBusy++;
- ctx->queueEmpty = ctx->queueHead == ctx->queueTail;
+ ctx->queueEmpty = (ctx->queueHead == ctx->queueTail);
/* Unlock the mutex, signal a pusher, and run the job */
ZSTD_pthread_cond_signal(&ctx->queuePushCond);
ZSTD_pthread_mutex_unlock(&ctx->queueMutex);
@@ -105,6 +105,7 @@ static void* POOL_thread(void* opaque) {
assert(0); /* Unreachable */
}
+/* ZSTD_createThreadPool() : public access point */
POOL_ctx* ZSTD_createThreadPool(size_t numThreads) {
return POOL_create (numThreads, 0);
}
@@ -114,7 +115,8 @@ POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
}
POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize,
- ZSTD_customMem customMem) {
+ ZSTD_customMem customMem)
+{
POOL_ctx* ctx;
/* Check parameters */
if (!numThreads) { return NULL; }
@@ -192,7 +194,7 @@ void ZSTD_freeThreadPool (ZSTD_threadPool* pool) {
POOL_free (pool);
}
-size_t POOL_sizeof(POOL_ctx *ctx) {
+size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
return sizeof(*ctx)
+ ctx->queueSize * sizeof(POOL_job)
@@ -257,7 +259,8 @@ static int isQueueFull(POOL_ctx const* ctx) {
}
-static void POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
+static void
+POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque)
{
POOL_job const job = {function, opaque};
assert(ctx != NULL);
@@ -313,7 +316,9 @@ POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) {
return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem);
}
-POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem) {
+POOL_ctx*
+POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem)
+{
(void)numThreads;
(void)queueSize;
(void)customMem;
@@ -341,7 +346,7 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) {
return 1;
}
-size_t POOL_sizeof(POOL_ctx* ctx) {
+size_t POOL_sizeof(const POOL_ctx* ctx) {
if (ctx==NULL) return 0; /* supports sizeof NULL */
assert(ctx == &g_poolCtx);
return sizeof(*ctx);
diff --git a/lib/common/pool.h b/lib/common/pool.h
index e18aa0708f7..0ebde1805db 100644
--- a/lib/common/pool.h
+++ b/lib/common/pool.h
@@ -53,7 +53,7 @@ int POOL_resize(POOL_ctx* ctx, size_t numThreads);
* @return threadpool memory usage
* note : compatible with NULL (returns 0 in this case)
*/
-size_t POOL_sizeof(POOL_ctx* ctx);
+size_t POOL_sizeof(const POOL_ctx* ctx);
/*! POOL_function :
* The function type that can be added to a thread pool.
@@ -70,7 +70,7 @@ void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque);
/*! POOL_tryAdd() :
- * Add the job `function(opaque)` to thread pool _if_ a worker is available.
+ * Add the job `function(opaque)` to thread pool _if_ a queue slot is available.
* Returns immediately even if not (does not block).
* @return : 1 if successful, 0 if not.
*/
diff --git a/lib/common/portability_macros.h b/lib/common/portability_macros.h
index 627ef9eed4a..2143817f574 100644
--- a/lib/common/portability_macros.h
+++ b/lib/common/portability_macros.h
@@ -65,6 +65,12 @@
# endif
#endif
+/* Mark the internal assembly functions as hidden */
+#ifdef __ELF__
+# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func
+#else
+# define ZSTD_HIDE_ASM_FUNCTION(func)
+#endif
/* Enable runtime BMI2 dispatch based on the CPU.
* Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default.
diff --git a/lib/common/zstd_internal.h b/lib/common/zstd_internal.h
index 1c053b4120f..e4d36ce0905 100644
--- a/lib/common/zstd_internal.h
+++ b/lib/common/zstd_internal.h
@@ -67,7 +67,6 @@ extern "C" {
#define ZSTD_OPT_NUM (1<<12)
#define ZSTD_REP_NUM 3 /* number of repcodes */
-#define ZSTD_REP_MOVE (ZSTD_REP_NUM-1)
static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 };
#define KB *(1 <<10)
@@ -285,9 +284,9 @@ typedef enum {
* Private declarations
*********************************************/
typedef struct seqDef_s {
- U32 offset; /* offset == rawOffset + ZSTD_REP_NUM, or equivalently, offCode + 1 */
+ U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
U16 litLength;
- U16 matchLength;
+ U16 mlBase; /* mlBase == matchLength - MINMATCH */
} seqDef;
/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */
@@ -329,7 +328,7 @@ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore
{
ZSTD_sequenceLength seqLen;
seqLen.litLength = seq->litLength;
- seqLen.matchLength = seq->matchLength + MINMATCH;
+ seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0xFFFF;
diff --git a/lib/compress/zstd_compress.c b/lib/compress/zstd_compress.c
index 34f8e970f1a..f06456af926 100644
--- a/lib/compress/zstd_compress.c
+++ b/lib/compress/zstd_compress.c
@@ -2396,9 +2396,9 @@ void ZSTD_seqToCodes(const seqStore_t* seqStorePtr)
assert(nbSeq <= seqStorePtr->maxNbSeq);
for (u=0; ulongLengthType==ZSTD_llt_literalLength)
@@ -2910,9 +2910,9 @@ static void ZSTD_copyBlockSequences(ZSTD_CCtx* zc)
assert(zc->seqCollector.maxSequences >= seqStoreSeqSize + 1);
ZSTD_memcpy(updatedRepcodes.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t));
for (i = 0; i < seqStoreSeqSize; ++i) {
- U32 rawOffset = seqStoreSeqs[i].offset - ZSTD_REP_NUM;
+ U32 rawOffset = seqStoreSeqs[i].offBase - ZSTD_REP_NUM;
outSeqs[i].litLength = seqStoreSeqs[i].litLength;
- outSeqs[i].matchLength = seqStoreSeqs[i].matchLength + MINMATCH;
+ outSeqs[i].matchLength = seqStoreSeqs[i].mlBase + MINMATCH;
outSeqs[i].rep = 0;
if (i == seqStore->longLengthPos) {
@@ -2923,9 +2923,9 @@ static void ZSTD_copyBlockSequences(ZSTD_CCtx* zc)
}
}
- if (seqStoreSeqs[i].offset <= ZSTD_REP_NUM) {
+ if (seqStoreSeqs[i].offBase <= ZSTD_REP_NUM) {
/* Derive the correct offset corresponding to a repcode */
- outSeqs[i].rep = seqStoreSeqs[i].offset;
+ outSeqs[i].rep = seqStoreSeqs[i].offBase;
if (outSeqs[i].litLength != 0) {
rawOffset = updatedRepcodes.rep[outSeqs[i].rep - 1];
} else {
@@ -2939,9 +2939,9 @@ static void ZSTD_copyBlockSequences(ZSTD_CCtx* zc)
outSeqs[i].offset = rawOffset;
/* seqStoreSeqs[i].offset == offCode+1, and ZSTD_updateRep() expects offCode
so we provide seqStoreSeqs[i].offset - 1 */
- updatedRepcodes = ZSTD_updateRep(updatedRepcodes.rep,
- seqStoreSeqs[i].offset - 1,
- seqStoreSeqs[i].litLength == 0);
+ ZSTD_updateRep(updatedRepcodes.rep,
+ seqStoreSeqs[i].offBase - 1,
+ seqStoreSeqs[i].litLength == 0);
literalsRead += outSeqs[i].litLength;
}
/* Insert last literals (if any exist) in the block as a sequence with ml == off == 0.
@@ -3385,7 +3385,7 @@ static size_t ZSTD_countSeqStoreMatchBytes(const seqStore_t* const seqStore) {
size_t i;
for (i = 0; i < nbSeqs; ++i) {
seqDef seq = seqStore->sequencesStart[i];
- matchBytes += seq.matchLength + MINMATCH;
+ matchBytes += seq.mlBase + MINMATCH;
if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_matchLength) {
matchBytes += 0x10000;
}
@@ -3434,11 +3434,13 @@ static void ZSTD_deriveSeqStoreChunk(seqStore_t* resultSeqStore,
/**
* Returns the raw offset represented by the combination of offCode, ll0, and repcode history.
- * offCode must be an offCode representing a repcode, therefore in the range of [0, 2].
+ * offCode must represent a repcode in the numeric representation of ZSTD_storeSeq().
*/
-static U32 ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32 offCode, const U32 ll0) {
- U32 const adjustedOffCode = offCode + ll0;
- assert(offCode < ZSTD_REP_NUM);
+static U32
+ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32 offCode, const U32 ll0)
+{
+ U32 const adjustedOffCode = STORED_REPCODE(offCode) - 1 + ll0; /* [ 0 - 3 ] */
+ assert(STORED_IS_REPCODE(offCode));
if (adjustedOffCode == ZSTD_REP_NUM) {
/* litlength == 0 and offCode == 2 implies selection of first repcode - 1 */
assert(rep[0] > 0);
@@ -3449,11 +3451,16 @@ static U32 ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32
/**
* ZSTD_seqStore_resolveOffCodes() reconciles any possible divergences in offset history that may arise
- * due to emission of RLE/raw blocks that disturb the offset history, and replaces any repcodes within
- * the seqStore that may be invalid.
+ * due to emission of RLE/raw blocks that disturb the offset history,
+ * and replaces any repcodes within the seqStore that may be invalid.
+ *
+ * dRepcodes are updated as would be on the decompression side.
+ * cRepcodes are updated exactly in accordance with the seqStore.
*
- * dRepcodes are updated as would be on the decompression side. cRepcodes are updated exactly in
- * accordance with the seqStore.
+ * Note : this function assumes seq->offBase respects the following numbering scheme :
+ * 0 : invalid
+ * 1-3 : repcode 1-3
+ * 4+ : real_offset+3
*/
static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_t* const cRepcodes,
seqStore_t* const seqStore, U32 const nbSeq) {
@@ -3461,9 +3468,9 @@ static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_
for (; idx < nbSeq; ++idx) {
seqDef* const seq = seqStore->sequencesStart + idx;
U32 const ll0 = (seq->litLength == 0);
- U32 offCode = seq->offset - 1;
- assert(seq->offset > 0);
- if (offCode <= ZSTD_REP_MOVE) {
+ U32 const offCode = OFFBASE_TO_STORED(seq->offBase);
+ assert(seq->offBase > 0);
+ if (STORED_IS_REPCODE(offCode)) {
U32 const dRawOffset = ZSTD_resolveRepcodeToRawOffset(dRepcodes->rep, offCode, ll0);
U32 const cRawOffset = ZSTD_resolveRepcodeToRawOffset(cRepcodes->rep, offCode, ll0);
/* Adjust simulated decompression repcode history if we come across a mismatch. Replace
@@ -3471,14 +3478,14 @@ static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_
* repcode history.
*/
if (dRawOffset != cRawOffset) {
- seq->offset = cRawOffset + ZSTD_REP_NUM;
+ seq->offBase = cRawOffset + ZSTD_REP_NUM;
}
}
/* Compression repcode history is always updated with values directly from the unmodified seqStore.
* Decompression repcode history may use modified seq->offset value taken from compression repcode history.
*/
- *dRepcodes = ZSTD_updateRep(dRepcodes->rep, seq->offset - 1, ll0);
- *cRepcodes = ZSTD_updateRep(cRepcodes->rep, offCode, ll0);
+ ZSTD_updateRep(dRepcodes->rep, OFFBASE_TO_STORED(seq->offBase), ll0);
+ ZSTD_updateRep(cRepcodes->rep, offCode, ll0);
}
}
@@ -3487,11 +3494,13 @@ static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_
*
* Returns the total size of that block (including header) or a ZSTD error code.
*/
-static size_t ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const seqStore,
- repcodes_t* const dRep, repcodes_t* const cRep,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize,
- U32 lastBlock, U32 isPartition) {
+static size_t
+ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const seqStore,
+ repcodes_t* const dRep, repcodes_t* const cRep,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize,
+ U32 lastBlock, U32 isPartition)
+{
const U32 rleMaxLength = 25;
BYTE* op = (BYTE*)dst;
const BYTE* ip = (const BYTE*)src;
@@ -3500,6 +3509,7 @@ static size_t ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const
/* In case of an RLE or raw block, the simulated decompression repcode history must be reset */
repcodes_t const dRepOriginal = *dRep;
+ DEBUGLOG(5, "ZSTD_compressSeqStore_singleBlock");
if (isPartition)
ZSTD_seqStore_resolveOffCodes(dRep, cRep, seqStore, (U32)(seqStore->sequences - seqStore->sequencesStart));
@@ -3572,8 +3582,10 @@ typedef struct {
* Furthermore, the number of splits is capped by ZSTD_MAX_NB_BLOCK_SPLITS. At ZSTD_MAX_NB_BLOCK_SPLITS == 196 with the current existing blockSize
* maximum of 128 KB, this value is actually impossible to reach.
*/
-static void ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t endIdx,
- ZSTD_CCtx* zc, const seqStore_t* origSeqStore) {
+static void
+ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t endIdx,
+ ZSTD_CCtx* zc, const seqStore_t* origSeqStore)
+{
seqStore_t* fullSeqStoreChunk = &zc->blockSplitCtx.fullSeqStoreChunk;
seqStore_t* firstHalfSeqStore = &zc->blockSplitCtx.firstHalfSeqStore;
seqStore_t* secondHalfSeqStore = &zc->blockSplitCtx.secondHalfSeqStore;
@@ -3628,8 +3640,10 @@ static size_t ZSTD_deriveBlockSplits(ZSTD_CCtx* zc, U32 partitions[], U32 nbSeq)
*
* Returns combined size of all blocks (which includes headers), or a ZSTD error code.
*/
-static size_t ZSTD_compressBlock_splitBlock_internal(ZSTD_CCtx* zc, void* dst, size_t dstCapacity,
- const void* src, size_t blockSize, U32 lastBlock, U32 nbSeq) {
+static size_t
+ZSTD_compressBlock_splitBlock_internal(ZSTD_CCtx* zc, void* dst, size_t dstCapacity,
+ const void* src, size_t blockSize, U32 lastBlock, U32 nbSeq)
+{
size_t cSize = 0;
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
@@ -3715,9 +3729,11 @@ static size_t ZSTD_compressBlock_splitBlock_internal(ZSTD_CCtx* zc, void* dst, s
return cSize;
}
-static size_t ZSTD_compressBlock_splitBlock(ZSTD_CCtx* zc,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize, U32 lastBlock) {
+static size_t
+ZSTD_compressBlock_splitBlock(ZSTD_CCtx* zc,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize, U32 lastBlock)
+{
const BYTE* ip = (const BYTE*)src;
BYTE* op = (BYTE*)dst;
U32 nbSeq;
@@ -3743,9 +3759,10 @@ static size_t ZSTD_compressBlock_splitBlock(ZSTD_CCtx* zc,
return cSize;
}
-static size_t ZSTD_compressBlock_internal(ZSTD_CCtx* zc,
- void* dst, size_t dstCapacity,
- const void* src, size_t srcSize, U32 frame)
+static size_t
+ZSTD_compressBlock_internal(ZSTD_CCtx* zc,
+ void* dst, size_t dstCapacity,
+ const void* src, size_t srcSize, U32 frame)
{
/* This the upper bound for the length of an rle block.
* This isn't the actual upper bound. Finding the real threshold
@@ -5738,39 +5755,39 @@ typedef struct {
size_t posInSrc; /* Number of bytes given by sequences provided so far */
} ZSTD_sequencePosition;
-/* Returns a ZSTD error code if sequence is not valid */
-static size_t ZSTD_validateSequence(U32 offCode, U32 matchLength,
- size_t posInSrc, U32 windowLog, size_t dictSize, U32 minMatch) {
- size_t offsetBound;
- U32 windowSize = 1 << windowLog;
+/* ZSTD_validateSequence() :
+ * @offCode : is presumed to follow format required by ZSTD_storeSeq()
+ * @returns a ZSTD error code if sequence is not valid
+ */
+static size_t
+ZSTD_validateSequence(U32 offCode, U32 matchLength,
+ size_t posInSrc, U32 windowLog, size_t dictSize)
+{
+ U32 const windowSize = 1 << windowLog;
/* posInSrc represents the amount of data the the decoder would decode up to this point.
* As long as the amount of data decoded is less than or equal to window size, offsets may be
* larger than the total length of output decoded in order to reference the dict, even larger than
* window size. After output surpasses windowSize, we're limited to windowSize offsets again.
*/
- offsetBound = posInSrc > windowSize ? (size_t)windowSize : posInSrc + (size_t)dictSize;
- RETURN_ERROR_IF(offCode > offsetBound + ZSTD_REP_MOVE, corruption_detected, "Offset too large!");
- RETURN_ERROR_IF(matchLength < minMatch, corruption_detected, "Matchlength too small");
+ size_t const offsetBound = posInSrc > windowSize ? (size_t)windowSize : posInSrc + (size_t)dictSize;
+ RETURN_ERROR_IF(offCode > STORE_OFFSET(offsetBound), corruption_detected, "Offset too large!");
+ RETURN_ERROR_IF(matchLength < MINMATCH, corruption_detected, "Matchlength too small");
return 0;
}
/* Returns an offset code, given a sequence's raw offset, the ongoing repcode array, and whether litLength == 0 */
-static U32 ZSTD_finalizeOffCode(U32 rawOffset, const U32 rep[ZSTD_REP_NUM], U32 ll0) {
- U32 offCode = rawOffset + ZSTD_REP_MOVE;
- U32 repCode = 0;
+static U32 ZSTD_finalizeOffCode(U32 rawOffset, const U32 rep[ZSTD_REP_NUM], U32 ll0)
+{
+ U32 offCode = STORE_OFFSET(rawOffset);
if (!ll0 && rawOffset == rep[0]) {
- repCode = 1;
+ offCode = STORE_REPCODE_1;
} else if (rawOffset == rep[1]) {
- repCode = 2 - ll0;
+ offCode = STORE_REPCODE(2 - ll0);
} else if (rawOffset == rep[2]) {
- repCode = 3 - ll0;
+ offCode = STORE_REPCODE(3 - ll0);
} else if (ll0 && rawOffset == rep[0] - 1) {
- repCode = 3;
- }
- if (repCode) {
- /* ZSTD_storeSeq expects a number in the range [0, 2] to represent a repcode */
- offCode = repCode - 1;
+ offCode = STORE_REPCODE_3;
}
return offCode;
}
@@ -5778,18 +5795,17 @@ static U32 ZSTD_finalizeOffCode(U32 rawOffset, const U32 rep[ZSTD_REP_NUM], U32
/* Returns 0 on success, and a ZSTD_error otherwise. This function scans through an array of
* ZSTD_Sequence, storing the sequences it finds, until it reaches a block delimiter.
*/
-static size_t ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos,
- const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
- const void* src, size_t blockSize) {
+static size_t
+ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx,
+ ZSTD_sequencePosition* seqPos,
+ const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
+ const void* src, size_t blockSize)
+{
U32 idx = seqPos->idx;
BYTE const* ip = (BYTE const*)(src);
const BYTE* const iend = ip + blockSize;
repcodes_t updatedRepcodes;
U32 dictSize;
- U32 litLength;
- U32 matchLength;
- U32 ll0;
- U32 offCode;
if (cctx->cdict) {
dictSize = (U32)cctx->cdict->dictContentSize;
@@ -5800,23 +5816,22 @@ static size_t ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx, ZS
}
ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t));
for (; (inSeqs[idx].matchLength != 0 || inSeqs[idx].offset != 0) && idx < inSeqsSize; ++idx) {
- litLength = inSeqs[idx].litLength;
- matchLength = inSeqs[idx].matchLength;
- ll0 = litLength == 0;
- offCode = ZSTD_finalizeOffCode(inSeqs[idx].offset, updatedRepcodes.rep, ll0);
- updatedRepcodes = ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0);
+ U32 const litLength = inSeqs[idx].litLength;
+ U32 const ll0 = (litLength == 0);
+ U32 const matchLength = inSeqs[idx].matchLength;
+ U32 const offCode = ZSTD_finalizeOffCode(inSeqs[idx].offset, updatedRepcodes.rep, ll0);
+ ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0);
DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength);
if (cctx->appliedParams.validateSequences) {
seqPos->posInSrc += litLength + matchLength;
FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc,
- cctx->appliedParams.cParams.windowLog, dictSize,
- cctx->appliedParams.cParams.minMatch),
+ cctx->appliedParams.cParams.windowLog, dictSize),
"Sequence validation failed");
}
RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation,
"Not enough memory allocated. Try adjusting ZSTD_c_minMatch.");
- ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength - MINMATCH);
+ ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength);
ip += matchLength + litLength;
}
ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t));
@@ -5843,9 +5858,11 @@ static size_t ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx, ZS
* avoid splitting a match, or to avoid splitting a match such that it would produce a match
* smaller than MINMATCH. In this case, we return the number of bytes that we didn't read from this block.
*/
-static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos,
- const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
- const void* src, size_t blockSize) {
+static size_t
+ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos,
+ const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
+ const void* src, size_t blockSize)
+{
U32 idx = seqPos->idx;
U32 startPosInSequence = seqPos->posInSequence;
U32 endPosInSequence = seqPos->posInSequence + (U32)blockSize;
@@ -5855,10 +5872,6 @@ static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_seq
repcodes_t updatedRepcodes;
U32 bytesAdjustment = 0;
U32 finalMatchSplit = 0;
- U32 litLength;
- U32 matchLength;
- U32 rawOffset;
- U32 offCode;
if (cctx->cdict) {
dictSize = cctx->cdict->dictContentSize;
@@ -5872,9 +5885,10 @@ static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_seq
ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t));
while (endPosInSequence && idx < inSeqsSize && !finalMatchSplit) {
const ZSTD_Sequence currSeq = inSeqs[idx];
- litLength = currSeq.litLength;
- matchLength = currSeq.matchLength;
- rawOffset = currSeq.offset;
+ U32 litLength = currSeq.litLength;
+ U32 matchLength = currSeq.matchLength;
+ U32 const rawOffset = currSeq.offset;
+ U32 offCode;
/* Modify the sequence depending on where endPosInSequence lies */
if (endPosInSequence >= currSeq.litLength + currSeq.matchLength) {
@@ -5927,22 +5941,21 @@ static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_seq
}
}
/* Check if this offset can be represented with a repcode */
- { U32 ll0 = (litLength == 0);
+ { U32 const ll0 = (litLength == 0);
offCode = ZSTD_finalizeOffCode(rawOffset, updatedRepcodes.rep, ll0);
- updatedRepcodes = ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0);
+ ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0);
}
if (cctx->appliedParams.validateSequences) {
seqPos->posInSrc += litLength + matchLength;
FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc,
- cctx->appliedParams.cParams.windowLog, dictSize,
- cctx->appliedParams.cParams.minMatch),
+ cctx->appliedParams.cParams.windowLog, dictSize),
"Sequence validation failed");
}
DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength);
RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation,
"Not enough memory allocated. Try adjusting ZSTD_c_minMatch.");
- ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength - MINMATCH);
+ ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength);
ip += matchLength + litLength;
}
DEBUGLOG(5, "Ending seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength);
@@ -5967,7 +5980,8 @@ static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_seq
typedef size_t (*ZSTD_sequenceCopier) (ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos,
const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
const void* src, size_t blockSize);
-static ZSTD_sequenceCopier ZSTD_selectSequenceCopier(ZSTD_sequenceFormat_e mode) {
+static ZSTD_sequenceCopier ZSTD_selectSequenceCopier(ZSTD_sequenceFormat_e mode)
+{
ZSTD_sequenceCopier sequenceCopier = NULL;
assert(ZSTD_cParam_withinBounds(ZSTD_c_blockDelimiters, mode));
if (mode == ZSTD_sf_explicitBlockDelimiters) {
@@ -5981,12 +5995,15 @@ static ZSTD_sequenceCopier ZSTD_selectSequenceCopier(ZSTD_sequenceFormat_e mode)
/* Compress, block-by-block, all of the sequences given.
*
- * Returns the cumulative size of all compressed blocks (including their headers), otherwise a ZSTD error.
+ * Returns the cumulative size of all compressed blocks (including their headers),
+ * otherwise a ZSTD error.
*/
-static size_t ZSTD_compressSequences_internal(ZSTD_CCtx* cctx,
- void* dst, size_t dstCapacity,
- const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
- const void* src, size_t srcSize) {
+static size_t
+ZSTD_compressSequences_internal(ZSTD_CCtx* cctx,
+ void* dst, size_t dstCapacity,
+ const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
+ const void* src, size_t srcSize)
+{
size_t cSize = 0;
U32 lastBlock;
size_t blockSize;
@@ -5996,7 +6013,7 @@ static size_t ZSTD_compressSequences_internal(ZSTD_CCtx* cctx,
BYTE const* ip = (BYTE const*)src;
BYTE* op = (BYTE*)dst;
- ZSTD_sequenceCopier sequenceCopier = ZSTD_selectSequenceCopier(cctx->appliedParams.blockDelimiters);
+ ZSTD_sequenceCopier const sequenceCopier = ZSTD_selectSequenceCopier(cctx->appliedParams.blockDelimiters);
DEBUGLOG(4, "ZSTD_compressSequences_internal srcSize: %zu, inSeqsSize: %zu", srcSize, inSeqsSize);
/* Special case: empty frame */
@@ -6096,7 +6113,8 @@ static size_t ZSTD_compressSequences_internal(ZSTD_CCtx* cctx,
size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstCapacity,
const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
- const void* src, size_t srcSize) {
+ const void* src, size_t srcSize)
+{
BYTE* op = (BYTE*)dst;
size_t cSize = 0;
size_t compressedBlocksSize = 0;
diff --git a/lib/compress/zstd_compress_internal.h b/lib/compress/zstd_compress_internal.h
index 7360ee8e221..c406e794bdb 100644
--- a/lib/compress/zstd_compress_internal.h
+++ b/lib/compress/zstd_compress_internal.h
@@ -129,7 +129,7 @@ size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr,
*********************************/
typedef struct {
- U32 off; /* Offset code (offset + ZSTD_REP_MOVE) for the match */
+ U32 off; /* Offset sumtype code for the match, using ZSTD_storeSeq() format */
U32 len; /* Raw length of match */
} ZSTD_match_t;
@@ -497,31 +497,6 @@ MEM_STATIC U32 ZSTD_MLcode(U32 mlBase)
return (mlBase > 127) ? ZSTD_highbit32(mlBase) + ML_deltaCode : ML_Code[mlBase];
}
-typedef struct repcodes_s {
- U32 rep[3];
-} repcodes_t;
-
-MEM_STATIC repcodes_t ZSTD_updateRep(U32 const rep[3], U32 const offset, U32 const ll0)
-{
- repcodes_t newReps;
- if (offset >= ZSTD_REP_NUM) { /* full offset */
- newReps.rep[2] = rep[1];
- newReps.rep[1] = rep[0];
- newReps.rep[0] = offset - ZSTD_REP_MOVE;
- } else { /* repcode */
- U32 const repCode = offset + ll0;
- if (repCode > 0) { /* note : if repCode==0, no change */
- U32 const currentOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode];
- newReps.rep[2] = (repCode >= 2) ? rep[1] : rep[2];
- newReps.rep[1] = rep[0];
- newReps.rep[0] = currentOffset;
- } else { /* repCode == 0 */
- ZSTD_memcpy(&newReps, rep, sizeof(newReps));
- }
- }
- return newReps;
-}
-
/* ZSTD_cParam_withinBounds:
* @return 1 if value is within cParam bounds,
* 0 otherwise */
@@ -590,7 +565,9 @@ MEM_STATIC int ZSTD_literalsCompressionIsDisabled(const ZSTD_CCtx_params* cctxPa
* Only called when the sequence ends past ilimit_w, so it only needs to be optimized for single
* large copies.
*/
-static void ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const iend, BYTE const* ilimit_w) {
+static void
+ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const iend, BYTE const* ilimit_w)
+{
assert(iend > ilimit_w);
if (ip <= ilimit_w) {
ZSTD_wildcopy(op, ip, ilimit_w - ip, ZSTD_no_overlap);
@@ -600,14 +577,30 @@ static void ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const ie
while (ip < iend) *op++ = *ip++;
}
+#define ZSTD_REP_MOVE (ZSTD_REP_NUM-1)
+#define STORE_REPCODE_1 STORE_REPCODE(1)
+#define STORE_REPCODE_2 STORE_REPCODE(2)
+#define STORE_REPCODE_3 STORE_REPCODE(3)
+#define STORE_REPCODE(r) (assert((r)>=1), assert((r)<=3), (r)-1)
+#define STORE_OFFSET(o) (assert((o)>0), o + ZSTD_REP_MOVE)
+#define STORED_IS_OFFSET(o) ((o) > ZSTD_REP_MOVE)
+#define STORED_IS_REPCODE(o) ((o) <= ZSTD_REP_MOVE)
+#define STORED_OFFSET(o) (assert(STORED_IS_OFFSET(o)), (o)-ZSTD_REP_MOVE)
+#define STORED_REPCODE(o) (assert(STORED_IS_REPCODE(o)), (o)+1) /* returns ID 1,2,3 */
+#define STORED_TO_OFFBASE(o) ((o)+1)
+#define OFFBASE_TO_STORED(o) ((o)-1)
+
/*! ZSTD_storeSeq() :
- * Store a sequence (litlen, litPtr, offCode and mlBase) into seqStore_t.
- * `offCode` : distance to match + ZSTD_REP_MOVE (values <= ZSTD_REP_MOVE are repCodes).
- * `mlBase` : matchLength - MINMATCH
+ * Store a sequence (litlen, litPtr, offCode and matchLength) into seqStore_t.
+ * @offBase_minus1 : Users should use employ macros STORE_REPCODE_X and STORE_OFFSET().
+ * @matchLength : must be >= MINMATCH
* Allowed to overread literals up to litLimit.
*/
-HINT_INLINE UNUSED_ATTR
-void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const BYTE* literals, const BYTE* litLimit, U32 offCode, size_t mlBase)
+HINT_INLINE UNUSED_ATTR void
+ZSTD_storeSeq(seqStore_t* seqStorePtr,
+ size_t litLength, const BYTE* literals, const BYTE* litLimit,
+ U32 offBase_minus1,
+ size_t matchLength)
{
BYTE const* const litLimit_w = litLimit - WILDCOPY_OVERLENGTH;
BYTE const* const litEnd = literals + litLength;
@@ -616,7 +609,7 @@ void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const BYTE* litera
if (g_start==NULL) g_start = (const BYTE*)literals; /* note : index only works for compression within a single segment */
{ U32 const pos = (U32)((const BYTE*)literals - g_start);
DEBUGLOG(6, "Cpos%7u :%3u literals, match%4u bytes at offCode%7u",
- pos, (U32)litLength, (U32)mlBase+MINMATCH, (U32)offCode);
+ pos, (U32)litLength, (U32)matchLength, (U32)offBase_minus1);
}
#endif
assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq);
@@ -647,19 +640,59 @@ void ZSTD_storeSeq(seqStore_t* seqStorePtr, size_t litLength, const BYTE* litera
seqStorePtr->sequences[0].litLength = (U16)litLength;
/* match offset */
- seqStorePtr->sequences[0].offset = offCode + 1;
+ seqStorePtr->sequences[0].offBase = STORED_TO_OFFBASE(offBase_minus1);
/* match Length */
- if (mlBase>0xFFFF) {
- assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */
- seqStorePtr->longLengthType = ZSTD_llt_matchLength;
- seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
+ assert(matchLength >= MINMATCH);
+ { size_t const mlBase = matchLength - MINMATCH;
+ if (mlBase>0xFFFF) {
+ assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */
+ seqStorePtr->longLengthType = ZSTD_llt_matchLength;
+ seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
+ }
+ seqStorePtr->sequences[0].mlBase = (U16)mlBase;
}
- seqStorePtr->sequences[0].matchLength = (U16)mlBase;
seqStorePtr->sequences++;
}
+/* ZSTD_updateRep() :
+ * updates in-place @rep (array of repeat offsets)
+ * @offBase_minus1 : sum-type, with same numeric representation as ZSTD_storeSeq()
+ */
+MEM_STATIC void
+ZSTD_updateRep(U32 rep[ZSTD_REP_NUM], U32 const offBase_minus1, U32 const ll0)
+{
+ if (STORED_IS_OFFSET(offBase_minus1)) { /* full offset */
+ rep[2] = rep[1];
+ rep[1] = rep[0];
+ rep[0] = STORED_OFFSET(offBase_minus1);
+ } else { /* repcode */
+ U32 const repCode = STORED_REPCODE(offBase_minus1) - 1 + ll0;
+ if (repCode > 0) { /* note : if repCode==0, no change */
+ U32 const currentOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode];
+ rep[2] = (repCode >= 2) ? rep[1] : rep[2];
+ rep[1] = rep[0];
+ rep[0] = currentOffset;
+ } else { /* repCode == 0 */
+ /* nothing to do */
+ }
+ }
+}
+
+typedef struct repcodes_s {
+ U32 rep[3];
+} repcodes_t;
+
+MEM_STATIC repcodes_t
+ZSTD_newRep(U32 const rep[ZSTD_REP_NUM], U32 const offBase_minus1, U32 const ll0)
+{
+ repcodes_t newReps;
+ ZSTD_memcpy(&newReps, rep, sizeof(newReps));
+ ZSTD_updateRep(newReps.rep, offBase_minus1, ll0);
+ return newReps;
+}
+
/*-*************************************
* Match length counter
diff --git a/lib/compress/zstd_compress_sequences.c b/lib/compress/zstd_compress_sequences.c
index fa31e6e946f..f1e40af2ea0 100644
--- a/lib/compress/zstd_compress_sequences.c
+++ b/lib/compress/zstd_compress_sequences.c
@@ -313,19 +313,19 @@ ZSTD_encodeSequences_body(
FSE_initCState2(&stateLitLength, CTable_LitLength, llCodeTable[nbSeq-1]);
BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
- BIT_addBits(&blockStream, sequences[nbSeq-1].matchLength, ML_bits[mlCodeTable[nbSeq-1]]);
+ BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
if (longOffsets) {
U32 const ofBits = ofCodeTable[nbSeq-1];
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
- BIT_addBits(&blockStream, sequences[nbSeq-1].offset, extraBits);
+ BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, extraBits);
BIT_flushBits(&blockStream);
}
- BIT_addBits(&blockStream, sequences[nbSeq-1].offset >> extraBits,
+ BIT_addBits(&blockStream, sequences[nbSeq-1].offBase >> extraBits,
ofBits - extraBits);
} else {
- BIT_addBits(&blockStream, sequences[nbSeq-1].offset, ofCodeTable[nbSeq-1]);
+ BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]);
}
BIT_flushBits(&blockStream);
@@ -339,8 +339,8 @@ ZSTD_encodeSequences_body(
U32 const mlBits = ML_bits[mlCode];
DEBUGLOG(6, "encoding: litlen:%2u - matchlen:%2u - offCode:%7u",
(unsigned)sequences[n].litLength,
- (unsigned)sequences[n].matchLength + MINMATCH,
- (unsigned)sequences[n].offset);
+ (unsigned)sequences[n].mlBase + MINMATCH,
+ (unsigned)sequences[n].offBase);
/* 32b*/ /* 64b*/
/* (7)*/ /* (7)*/
FSE_encodeSymbol(&blockStream, &stateOffsetBits, ofCode); /* 15 */ /* 15 */
@@ -351,18 +351,18 @@ ZSTD_encodeSequences_body(
BIT_flushBits(&blockStream); /* (7)*/
BIT_addBits(&blockStream, sequences[n].litLength, llBits);
if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream);
- BIT_addBits(&blockStream, sequences[n].matchLength, mlBits);
+ BIT_addBits(&blockStream, sequences[n].mlBase, mlBits);
if (MEM_32bits() || (ofBits+mlBits+llBits > 56)) BIT_flushBits(&blockStream);
if (longOffsets) {
unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1);
if (extraBits) {
- BIT_addBits(&blockStream, sequences[n].offset, extraBits);
+ BIT_addBits(&blockStream, sequences[n].offBase, extraBits);
BIT_flushBits(&blockStream); /* (7)*/
}
- BIT_addBits(&blockStream, sequences[n].offset >> extraBits,
+ BIT_addBits(&blockStream, sequences[n].offBase >> extraBits,
ofBits - extraBits); /* 31 */
} else {
- BIT_addBits(&blockStream, sequences[n].offset, ofBits); /* 31 */
+ BIT_addBits(&blockStream, sequences[n].offBase, ofBits); /* 31 */
}
BIT_flushBits(&blockStream); /* (7)*/
DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr));
diff --git a/lib/compress/zstd_compress_superblock.c b/lib/compress/zstd_compress_superblock.c
index 67e1abb11e6..10e33785778 100644
--- a/lib/compress/zstd_compress_superblock.c
+++ b/lib/compress/zstd_compress_superblock.c
@@ -539,7 +539,7 @@ static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
repcodes_t rep;
ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep));
for (seq = sstart; seq < sp; ++seq) {
- rep = ZSTD_updateRep(rep.rep, seq->offset - 1, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);
+ ZSTD_updateRep(rep.rep, seq->offBase - 1, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);
}
ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep));
}
diff --git a/lib/compress/zstd_cwksp.h b/lib/compress/zstd_cwksp.h
index 29d027e9cdb..dc3f40c80c3 100644
--- a/lib/compress/zstd_cwksp.h
+++ b/lib/compress/zstd_cwksp.h
@@ -243,12 +243,14 @@ MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignByt
/**
* Internal function. Do not use directly.
- * Reserves the given number of bytes within the aligned/buffer segment of the wksp, which
- * counts from the end of the wksp. (as opposed to the object/table segment)
+ * Reserves the given number of bytes within the aligned/buffer segment of the wksp,
+ * which counts from the end of the wksp (as opposed to the object/table segment).
*
* Returns a pointer to the beginning of that space.
*/
-MEM_STATIC void* ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes) {
+MEM_STATIC void*
+ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes)
+{
void* const alloc = (BYTE*)ws->allocStart - bytes;
void* const bottom = ws->tableEnd;
DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining",
@@ -260,6 +262,8 @@ MEM_STATIC void* ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t
ws->allocFailed = 1;
return NULL;
}
+ /* the area is reserved from the end of wksp.
+ * If it overlaps with tableValidEnd, it voids guarantees on values' range */
if (alloc < ws->tableValidEnd) {
ws->tableValidEnd = alloc;
}
@@ -269,10 +273,12 @@ MEM_STATIC void* ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t
/**
* Moves the cwksp to the next phase, and does any necessary allocations.
+ * cwksp initialization must necessarily go through each phase in order.
* Returns a 0 on success, or zstd error
*/
-MEM_STATIC size_t ZSTD_cwksp_internal_advance_phase(
- ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase) {
+MEM_STATIC size_t
+ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase)
+{
assert(phase >= ws->phase);
if (phase > ws->phase) {
/* Going from allocating objects to allocating buffers */
@@ -295,15 +301,15 @@ MEM_STATIC size_t ZSTD_cwksp_internal_advance_phase(
{ /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */
void* const alloc = ws->objectEnd;
size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES);
- void* const end = (BYTE*)alloc + bytesToAlign;
+ void* const objectEnd = (BYTE*)alloc + bytesToAlign;
DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign);
- RETURN_ERROR_IF(end > ws->workspaceEnd, memory_allocation,
+ RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation,
"table phase - alignment initial allocation failed!");
- ws->objectEnd = end;
- ws->tableEnd = end;
- ws->tableValidEnd = end;
- }
- }
+ ws->objectEnd = objectEnd;
+ ws->tableEnd = objectEnd; /* table area starts being empty */
+ if (ws->tableValidEnd < ws->tableEnd) {
+ ws->tableValidEnd = ws->tableEnd;
+ } } }
ws->phase = phase;
ZSTD_cwksp_assert_internal_consistency(ws);
}
@@ -313,15 +319,17 @@ MEM_STATIC size_t ZSTD_cwksp_internal_advance_phase(
/**
* Returns whether this object/buffer/etc was allocated in this workspace.
*/
-MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr) {
+MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr)
+{
return (ptr != NULL) && (ws->workspace <= ptr) && (ptr <= ws->workspaceEnd);
}
/**
* Internal function. Do not use directly.
*/
-MEM_STATIC void* ZSTD_cwksp_reserve_internal(
- ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase) {
+MEM_STATIC void*
+ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase)
+{
void* alloc;
if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) {
return NULL;
@@ -351,14 +359,16 @@ MEM_STATIC void* ZSTD_cwksp_reserve_internal(
/**
* Reserves and returns unaligned memory.
*/
-MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes) {
+MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes)
+{
return (BYTE*)ZSTD_cwksp_reserve_internal(ws, bytes, ZSTD_cwksp_alloc_buffers);
}
/**
* Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes).
*/
-MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes) {
+MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes)
+{
void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES),
ZSTD_cwksp_alloc_aligned);
assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0);
@@ -370,7 +380,8 @@ MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes) {
* their values remain constrained, allowing us to re-use them without
* memset()-ing them.
*/
-MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) {
+MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes)
+{
const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned;
void* alloc;
void* end;
@@ -408,9 +419,11 @@ MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) {
/**
* Aligned on sizeof(void*).
+ * Note : should happen only once, at workspace first initialization
*/
-MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
- size_t roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
+MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes)
+{
+ size_t const roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*));
void* alloc = ws->objectEnd;
void* end = (BYTE*)alloc + roundedBytes;
@@ -419,7 +432,7 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE;
#endif
- DEBUGLOG(5,
+ DEBUGLOG(4,
"cwksp: reserving %p object %zd bytes (rounded to %zd), %zd bytes remaining",
alloc, bytes, roundedBytes, ZSTD_cwksp_available_space(ws) - roundedBytes);
assert((size_t)alloc % ZSTD_ALIGNOF(void*) == 0);
@@ -427,7 +440,7 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
ZSTD_cwksp_assert_internal_consistency(ws);
/* we must be in the first phase, no advance is possible */
if (ws->phase != ZSTD_cwksp_alloc_objects || end > ws->workspaceEnd) {
- DEBUGLOG(4, "cwksp: object alloc failed!");
+ DEBUGLOG(3, "cwksp: object alloc failed!");
ws->allocFailed = 1;
return NULL;
}
@@ -438,7 +451,7 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE)
/* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on
* either size. */
- alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
+ alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE;
if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) {
__asan_unpoison_memory_region(alloc, bytes);
}
@@ -447,7 +460,8 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) {
return alloc;
}
-MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws) {
+MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws)
+{
DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty");
#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE)
diff --git a/lib/compress/zstd_double_fast.c b/lib/compress/zstd_double_fast.c
index b9393b6a4e0..76933dea262 100644
--- a/lib/compress/zstd_double_fast.c
+++ b/lib/compress/zstd_double_fast.c
@@ -131,7 +131,7 @@ size_t ZSTD_compressBlock_doubleFast_noDict_generic(
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(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength);
goto _match_stored;
}
@@ -217,7 +217,7 @@ size_t ZSTD_compressBlock_doubleFast_noDict_generic(
hashLong[hl1] = (U32)(ip1 - base);
}
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
_match_stored:
/* match found */
@@ -243,7 +243,7 @@ size_t ZSTD_compressBlock_doubleFast_noDict_generic(
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(seqStore, 0, anchor, iend, 0, rLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, rLength);
ip += rLength;
anchor = ip;
continue; /* faster when present ... (?) */
@@ -328,7 +328,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
ip++;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength);
goto _match_stored;
}
@@ -419,7 +419,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
_match_stored:
/* match found */
@@ -448,7 +448,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
@@ -585,7 +585,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength);
} else {
if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) {
const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend;
@@ -596,7 +596,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
} else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) {
size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8);
@@ -621,7 +621,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
}
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
} else {
ip += ((ip-anchor) >> kSearchStrength) + 1;
@@ -653,7 +653,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic(
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2);
hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2;
hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2;
ip += repLength2;
diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
index 2bae2f70e0f..802fc315798 100644
--- a/lib/compress/zstd_fast.c
+++ b/lib/compress/zstd_fast.c
@@ -180,7 +180,7 @@ ZSTD_compressBlock_fast_noDict_generic(
mLength = ip0[-1] == match0[-1];
ip0 -= mLength;
match0 -= mLength;
- offcode = 0;
+ offcode = STORE_REPCODE_1;
mLength += 4;
goto _match;
}
@@ -267,7 +267,7 @@ ZSTD_compressBlock_fast_noDict_generic(
match0 = base + idx;
rep_offset2 = rep_offset1;
rep_offset1 = (U32)(ip0-match0);
- offcode = rep_offset1 + ZSTD_REP_MOVE;
+ offcode = STORE_OFFSET(rep_offset1);
mLength = 4;
/* Count the backwards match length. */
@@ -282,7 +282,7 @@ ZSTD_compressBlock_fast_noDict_generic(
/* Count the forward length. */
mLength += ZSTD_count(ip0 + mLength, match0 + mLength, iend);
- ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength - MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength);
ip0 += mLength;
anchor = ip0;
@@ -306,7 +306,7 @@ ZSTD_compressBlock_fast_noDict_generic(
{ U32 const tmpOff = rep_offset2; rep_offset2 = rep_offset1; rep_offset1 = tmpOff; } /* swap rep_offset2 <=> rep_offset1 */
hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base);
ip0 += rLength;
- ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, 0 /*offCode*/, rLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, STORE_REPCODE_1, rLength);
anchor = ip0;
continue; /* faster when present (confirmed on gcc-8) ... (?) */
} } }
@@ -439,7 +439,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength);
} else if ( (matchIndex <= prefixStartIndex) ) {
size_t const dictHash = ZSTD_hashPtr(ip, dictHLog, mls);
U32 const dictMatchIndex = dictHashTable[dictHash];
@@ -459,7 +459,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
} /* catch up */
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
}
} else if (MEM_read32(match) != MEM_read32(ip)) {
/* it's not a match, and we're not going to check the dictionary */
@@ -474,7 +474,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
&& (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
offset_2 = offset_1;
offset_1 = offset;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
}
/* match found */
@@ -499,7 +499,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, repLength2-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2);
hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
ip += repLength2;
anchor = ip;
@@ -598,7 +598,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend;
size_t const rLength = ZSTD_count_2segments(ip+1 +4, repMatch +4, iend, repMatchEnd, prefixStart) + 4;
ip++;
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, 0, rLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, rLength);
ip += rLength;
anchor = ip;
} else {
@@ -614,7 +614,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
size_t mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4;
while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */
offset_2 = offset_1; offset_1 = offset; /* update offset history */
- ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, offset + ZSTD_REP_MOVE, mLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength);
ip += mLength;
anchor = ip;
} }
@@ -633,7 +633,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic(
const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend;
size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4;
{ U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; } /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, 0 /*offcode*/, repLength2-MINMATCH);
+ ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, STORE_REPCODE_1, repLength2);
hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2;
ip += repLength2;
anchor = ip;
diff --git a/lib/compress/zstd_lazy.c b/lib/compress/zstd_lazy.c
index c40473cad7d..2e38dcb46d2 100644
--- a/lib/compress/zstd_lazy.c
+++ b/lib/compress/zstd_lazy.c
@@ -197,8 +197,8 @@ ZSTD_DUBT_findBetterDictMatch (
U32 matchIndex = dictMatchIndex + dictIndexDelta;
if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) {
DEBUGLOG(9, "ZSTD_DUBT_findBetterDictMatch(%u) : found better match length %u -> %u and offsetCode %u -> %u (dictMatchIndex %u, matchIndex %u)",
- curr, (U32)bestLength, (U32)matchLength, (U32)*offsetPtr, ZSTD_REP_MOVE + curr - matchIndex, dictMatchIndex, matchIndex);
- bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + curr - matchIndex;
+ curr, (U32)bestLength, (U32)matchLength, (U32)*offsetPtr, STORE_OFFSET(curr - matchIndex), dictMatchIndex, matchIndex);
+ bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex);
}
if (ip+matchLength == iend) { /* reached end of input : ip[matchLength] is not valid, no way to know if it's larger or smaller than match */
break; /* drop, to guarantee consistency (miss a little bit of compression) */
@@ -218,7 +218,7 @@ ZSTD_DUBT_findBetterDictMatch (
}
if (bestLength >= MINMATCH) {
- U32 const mIndex = curr - ((U32)*offsetPtr - ZSTD_REP_MOVE); (void)mIndex;
+ U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex;
DEBUGLOG(8, "ZSTD_DUBT_findBetterDictMatch(%u) : found match of length %u and offsetCode %u (pos %u)",
curr, (U32)bestLength, (U32)*offsetPtr, mIndex);
}
@@ -328,7 +328,7 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms,
if (matchLength > matchEndIdx - matchIndex)
matchEndIdx = matchIndex + (U32)matchLength;
if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) )
- bestLength = matchLength, *offsetPtr = ZSTD_REP_MOVE + curr - matchIndex;
+ bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex);
if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */
if (dictMode == ZSTD_dictMatchState) {
nbCompares = 0; /* in addition to avoiding checking any
@@ -368,7 +368,7 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms,
assert(matchEndIdx > curr+8); /* ensure nextToUpdate is increased */
ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */
if (bestLength >= MINMATCH) {
- U32 const mIndex = curr - ((U32)*offsetPtr - ZSTD_REP_MOVE); (void)mIndex;
+ U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex;
DEBUGLOG(8, "ZSTD_DUBT_findBestMatch(%u) : found match of length %u and offsetCode %u (pos %u)",
curr, (U32)bestLength, (U32)*offsetPtr, mIndex);
}
@@ -561,7 +561,7 @@ size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nb
/* save best solution */
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - (matchIndex + ddsIndexDelta) + ZSTD_REP_MOVE;
+ *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta));
if (ip+currentMl == iLimit) {
/* best possible, avoids read overflow on next attempt */
return ml;
@@ -598,7 +598,7 @@ size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nb
/* save best solution */
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - (matchIndex + ddsIndexDelta) + ZSTD_REP_MOVE;
+ *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta));
if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
}
}
@@ -703,7 +703,7 @@ size_t ZSTD_HcFindBestMatch(
/* save best solution */
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - matchIndex + ZSTD_REP_MOVE;
+ *offsetPtr = STORE_OFFSET(curr - matchIndex);
if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
}
@@ -738,7 +738,8 @@ size_t ZSTD_HcFindBestMatch(
/* save best solution */
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - (matchIndex + dmsIndexDelta) + ZSTD_REP_MOVE;
+ assert(curr > matchIndex + dmsIndexDelta);
+ *offsetPtr = STORE_OFFSET(curr - (matchIndex + dmsIndexDelta));
if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
}
@@ -1244,7 +1245,7 @@ size_t ZSTD_RowFindBestMatch(
/* Save best solution */
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - matchIndex + ZSTD_REP_MOVE;
+ *offsetPtr = STORE_OFFSET(curr - matchIndex);
if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */
}
}
@@ -1292,7 +1293,8 @@ size_t ZSTD_RowFindBestMatch(
if (currentMl > ml) {
ml = currentMl;
- *offsetPtr = curr - (matchIndex + dmsIndexDelta) + ZSTD_REP_MOVE;
+ assert(curr > matchIndex + dmsIndexDelta);
+ *offsetPtr = STORE_OFFSET(curr - (matchIndex + dmsIndexDelta));
if (ip+currentMl == iLimit) break;
}
}
@@ -1537,8 +1539,9 @@ ZSTD_compressBlock_lazy_generic(
#endif
while (ip < ilimit) {
size_t matchLength=0;
- size_t offset=0;
+ size_t offcode=STORE_REPCODE_1;
const BYTE* start=ip+1;
+ DEBUGLOG(7, "search baseline (depth 0)");
/* check repCode */
if (isDxS) {
@@ -1564,7 +1567,7 @@ ZSTD_compressBlock_lazy_generic(
{ size_t offsetFound = 999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offsetFound);
if (ml2 > matchLength)
- matchLength = ml2, start = ip, offset=offsetFound;
+ matchLength = ml2, start = ip, offcode=offsetFound;
}
if (matchLength < 4) {
@@ -1575,14 +1578,15 @@ ZSTD_compressBlock_lazy_generic(
/* let's try to find a better solution */
if (depth>=1)
while (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) {
+ && (offcode) && ((offset_1>0) & (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);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((mlRep >= 4) && (gain2 > gain1))
- matchLength = mlRep, offset = 0, start = ip;
+ matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip;
}
if (isDxS) {
const U32 repIndex = (U32)(ip - base) - offset_1;
@@ -1594,30 +1598,31 @@ ZSTD_compressBlock_lazy_generic(
const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
int const gain2 = (int)(mlRep * 3);
- int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)offset+1) + 1);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((mlRep >= 4) && (gain2 > gain1))
- matchLength = mlRep, offset = 0, start = ip;
+ matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip;
}
}
{ size_t offset2=999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offset2);
- 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);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 4);
if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
+ matchLength = ml2, offcode = 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)))) {
+ && (offcode) && ((offset_1>0) & (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 * 4);
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((mlRep >= 4) && (gain2 > gain1))
- matchLength = mlRep, offset = 0, start = ip;
+ matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip;
}
if (isDxS) {
const U32 repIndex = (U32)(ip - base) - offset_1;
@@ -1629,46 +1634,45 @@ ZSTD_compressBlock_lazy_generic(
const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend;
size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4;
int const gain2 = (int)(mlRep * 4);
- int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)offset+1) + 1);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((mlRep >= 4) && (gain2 > gain1))
- matchLength = mlRep, offset = 0, start = ip;
+ matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip;
}
}
{ size_t offset2=999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offset2);
- 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);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 7);
if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
+ matchLength = ml2, offcode = 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.
+ * Pay attention that `start[-value]` can lead to strange undefined behavior
+ * notably if `value` is unsigned, resulting in a large positive `-value`.
*/
/* catch up */
- if (offset) {
+ if (STORED_IS_OFFSET(offcode)) {
if (dictMode == ZSTD_noDict) {
- while ( ((start > anchor) & (start - (offset-ZSTD_REP_MOVE) > prefixLowest))
- && (start[-1] == (start-(offset-ZSTD_REP_MOVE))[-1]) ) /* only search for offset within prefix */
+ while ( ((start > anchor) & (start - STORED_OFFSET(offcode) > prefixLowest))
+ && (start[-1] == (start-STORED_OFFSET(offcode))[-1]) ) /* only search for offset within prefix */
{ start--; matchLength++; }
}
if (isDxS) {
- U32 const matchIndex = (U32)((size_t)(start-base) - (offset - ZSTD_REP_MOVE));
+ U32 const matchIndex = (U32)((size_t)(start-base) - STORED_OFFSET(offcode));
const BYTE* match = (matchIndex < prefixLowestIndex) ? dictBase + matchIndex - dictIndexDelta : base + matchIndex;
const BYTE* const mStart = (matchIndex < prefixLowestIndex) ? dictLowest : prefixLowest;
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);
+ offset_2 = offset_1; offset_1 = (U32)STORED_OFFSET(offcode);
}
/* store sequence */
_storeSequence:
{ size_t const litLength = (size_t)(start - anchor);
- ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offset, matchLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offcode, matchLength);
anchor = ip = start + matchLength;
}
@@ -1684,8 +1688,8 @@ ZSTD_compressBlock_lazy_generic(
&& (MEM_read32(repMatch) == MEM_read32(ip)) ) {
const BYTE* const repEnd2 = repIndex < prefixLowestIndex ? dictEnd : iend;
matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd2, prefixLowest) + 4;
- offset = offset_2; offset_2 = offset_1; offset_1 = (U32)offset; /* swap offset_2 <=> offset_1 */
- ZSTD_storeSeq(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
+ offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap offset_2 <=> offset_1 */
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength);
ip += matchLength;
anchor = ip;
continue;
@@ -1699,8 +1703,8 @@ ZSTD_compressBlock_lazy_generic(
&& (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(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
+ offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap repcodes */
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength);
ip += matchLength;
anchor = ip;
continue; /* faster when present ... (?) */
@@ -1901,7 +1905,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
#endif
while (ip < ilimit) {
size_t matchLength=0;
- size_t offset=0;
+ size_t offcode=STORE_REPCODE_1;
const BYTE* start=ip+1;
U32 curr = (U32)(ip-base);
@@ -1923,7 +1927,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
{ size_t offsetFound = 999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offsetFound);
if (ml2 > matchLength)
- matchLength = ml2, start = ip, offset=offsetFound;
+ matchLength = ml2, start = ip, offcode=offsetFound;
}
if (matchLength < 4) {
@@ -1937,7 +1941,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
ip ++;
curr++;
/* check repCode */
- if (offset) {
+ if (offcode) {
const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog);
const U32 repIndex = (U32)(curr - offset_1);
const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
@@ -1949,18 +1953,18 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
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);
+ int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((repLength >= 4) && (gain2 > gain1))
- matchLength = repLength, offset = 0, start = ip;
+ matchLength = repLength, offcode = STORE_REPCODE_1, start = ip;
} }
/* search match, depth 1 */
{ size_t offset2=999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offset2);
- 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);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 4);
if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
+ matchLength = ml2, offcode = offset2, start = ip;
continue; /* search a better one */
} }
@@ -1969,7 +1973,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
ip ++;
curr++;
/* check repCode */
- if (offset) {
+ if (offcode) {
const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog);
const U32 repIndex = (U32)(curr - offset_1);
const BYTE* const repBase = repIndex < dictLimit ? dictBase : base;
@@ -1981,36 +1985,36 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
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);
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1);
if ((repLength >= 4) && (gain2 > gain1))
- matchLength = repLength, offset = 0, start = ip;
+ matchLength = repLength, offcode = STORE_REPCODE_1, start = ip;
} }
/* search match, depth 2 */
{ size_t offset2=999999999;
size_t const ml2 = searchMax(ms, ip, iend, &offset2);
- 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);
+ int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */
+ int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 7);
if ((ml2 >= 4) && (gain2 > gain1)) {
- matchLength = ml2, offset = offset2, start = ip;
+ matchLength = ml2, offcode = offset2, start = ip;
continue;
} } }
break; /* nothing found : store previous solution */
}
/* catch up */
- if (offset) {
- U32 const matchIndex = (U32)((size_t)(start-base) - (offset - ZSTD_REP_MOVE));
+ if (STORED_IS_OFFSET(offcode)) {
+ U32 const matchIndex = (U32)((size_t)(start-base) - STORED_OFFSET(offcode));
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);
+ offset_2 = offset_1; offset_1 = (U32)STORED_OFFSET(offcode);
}
/* store sequence */
_storeSequence:
{ size_t const litLength = (size_t)(start - anchor);
- ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offset, matchLength-MINMATCH);
+ ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offcode, matchLength);
anchor = ip = start + matchLength;
}
@@ -2027,8 +2031,8 @@ size_t ZSTD_compressBlock_lazy_extDict_generic(
/* 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(seqStore, 0, anchor, iend, 0, matchLength-MINMATCH);
+ offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap offset history */
+ ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength);
ip += matchLength;
anchor = ip;
continue; /* faster when present ... (?) */
diff --git a/lib/compress/zstd_ldm.c b/lib/compress/zstd_ldm.c
index 19b99f27815..f662b2546e0 100644
--- a/lib/compress/zstd_ldm.c
+++ b/lib/compress/zstd_ldm.c
@@ -579,7 +579,9 @@ size_t ZSTD_ldm_generateSequences(
return 0;
}
-void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch) {
+void
+ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch)
+{
while (srcSize > 0 && rawSeqStore->pos < rawSeqStore->size) {
rawSeq* seq = rawSeqStore->seq + rawSeqStore->pos;
if (srcSize <= seq->litLength) {
@@ -709,8 +711,8 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
rep[0] = sequence.offset;
/* Store the sequence */
ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend,
- sequence.offset + ZSTD_REP_MOVE,
- sequence.matchLength - MINMATCH);
+ STORE_OFFSET(sequence.offset),
+ sequence.matchLength);
ip += sequence.matchLength;
}
}
diff --git a/lib/compress/zstd_opt.c b/lib/compress/zstd_opt.c
index 7e5eb8b5764..1b1ddad4289 100644
--- a/lib/compress/zstd_opt.c
+++ b/lib/compress/zstd_opt.c
@@ -204,7 +204,8 @@ ZSTD_rescaleFreqs(optState_t* const optPtr,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1
};
- ZSTD_memcpy(optPtr->litLengthFreq, baseLLfreqs, sizeof(baseLLfreqs)); optPtr->litLengthSum = sum_u32(baseLLfreqs, MaxLL+1);
+ ZSTD_memcpy(optPtr->litLengthFreq, baseLLfreqs, sizeof(baseLLfreqs));
+ optPtr->litLengthSum = sum_u32(baseLLfreqs, MaxLL+1);
}
{ unsigned ml;
@@ -219,7 +220,8 @@ ZSTD_rescaleFreqs(optState_t* const optPtr,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1
};
- ZSTD_memcpy(optPtr->offCodeFreq, baseOFCfreqs, sizeof(baseOFCfreqs)); optPtr->offCodeSum = sum_u32(baseOFCfreqs, MaxOff+1);
+ ZSTD_memcpy(optPtr->offCodeFreq, baseOFCfreqs, sizeof(baseOFCfreqs));
+ optPtr->offCodeSum = sum_u32(baseOFCfreqs, MaxOff+1);
}
@@ -267,7 +269,16 @@ static U32 ZSTD_rawLiteralsCost(const BYTE* const literals, U32 const litLength,
* cost of literalLength symbol */
static U32 ZSTD_litLengthPrice(U32 const litLength, const optState_t* const optPtr, int optLevel)
{
- if (optPtr->priceType == zop_predef) return WEIGHT(litLength, optLevel);
+ assert(litLength <= ZSTD_BLOCKSIZE_MAX);
+ if (optPtr->priceType == zop_predef)
+ return WEIGHT(litLength, optLevel);
+ /* We can't compute the litLength price for sizes >= ZSTD_BLOCKSIZE_MAX
+ * because it isn't representable in the zstd format. So instead just
+ * call it 1 bit more than ZSTD_BLOCKSIZE_MAX - 1. In this case the block
+ * would be all literals.
+ */
+ if (litLength == ZSTD_BLOCKSIZE_MAX)
+ return BITCOST_MULTIPLIER + ZSTD_litLengthPrice(ZSTD_BLOCKSIZE_MAX - 1, optPtr, optLevel);
/* dynamic statistics */
{ U32 const llCode = ZSTD_LLcode(litLength);
@@ -280,15 +291,17 @@ static U32 ZSTD_litLengthPrice(U32 const litLength, const optState_t* const optP
/* ZSTD_getMatchPrice() :
* Provides the cost of the match part (offset + matchLength) of a sequence
* Must be combined with ZSTD_fullLiteralsCost() to get the full cost of a sequence.
- * optLevel: when <2, favors small offset for decompression speed (improved cache efficiency) */
+ * @offcode : expects a scale where 0,1,2 are repcodes 1-3, and 3+ are real_offsets+2
+ * @optLevel: when <2, favors small offset for decompression speed (improved cache efficiency)
+ */
FORCE_INLINE_TEMPLATE U32
-ZSTD_getMatchPrice(U32 const offset,
+ZSTD_getMatchPrice(U32 const offcode,
U32 const matchLength,
const optState_t* const optPtr,
int const optLevel)
{
U32 price;
- U32 const offCode = ZSTD_highbit32(offset+1);
+ U32 const offCode = ZSTD_highbit32(STORED_TO_OFFBASE(offcode));
U32 const mlBase = matchLength - MINMATCH;
assert(matchLength >= MINMATCH);
@@ -331,8 +344,8 @@ static void ZSTD_updateStats(optState_t* const optPtr,
optPtr->litLengthSum++;
}
- /* match offset code (0-2=>repCode; 3+=>offset+2) */
- { U32 const offCode = ZSTD_highbit32(offsetCode+1);
+ /* offset code : expected to follow storeSeq() numeric representation */
+ { U32 const offCode = ZSTD_highbit32(STORED_TO_OFFBASE(offsetCode));
assert(offCode <= MaxOff);
optPtr->offCodeFreq[offCode]++;
optPtr->offCodeSum++;
@@ -631,7 +644,7 @@ U32 ZSTD_insertBtAndGetAllMatches (
DEBUGLOG(8, "found repCode %u (ll0:%u, offset:%u) of length %u",
repCode, ll0, repOffset, repLen);
bestLength = repLen;
- matches[mnum].off = repCode - ll0;
+ matches[mnum].off = STORE_REPCODE(repCode - ll0 + 1); /* expect value between 1 and 3 */
matches[mnum].len = (U32)repLen;
mnum++;
if ( (repLen > sufficient_len)
@@ -660,7 +673,7 @@ U32 ZSTD_insertBtAndGetAllMatches (
bestLength = mlen;
assert(curr > matchIndex3);
assert(mnum==0); /* no prior solution */
- matches[0].off = (curr - matchIndex3) + ZSTD_REP_MOVE;
+ matches[0].off = STORE_OFFSET(curr - matchIndex3);
matches[0].len = (U32)mlen;
mnum = 1;
if ( (mlen > sufficient_len) |
@@ -694,12 +707,12 @@ U32 ZSTD_insertBtAndGetAllMatches (
if (matchLength > bestLength) {
DEBUGLOG(8, "found match of length %u at distance %u (offCode=%u)",
- (U32)matchLength, curr - matchIndex, curr - matchIndex + ZSTD_REP_MOVE);
+ (U32)matchLength, curr - matchIndex, STORE_OFFSET(curr - matchIndex));
assert(matchEndIdx > matchIndex);
if (matchLength > matchEndIdx - matchIndex)
matchEndIdx = matchIndex + (U32)matchLength;
bestLength = matchLength;
- matches[mnum].off = (curr - matchIndex) + ZSTD_REP_MOVE;
+ matches[mnum].off = STORE_OFFSET(curr - matchIndex);
matches[mnum].len = (U32)matchLength;
mnum++;
if ( (matchLength > ZSTD_OPT_NUM)
@@ -742,11 +755,11 @@ U32 ZSTD_insertBtAndGetAllMatches (
if (matchLength > bestLength) {
matchIndex = dictMatchIndex + dmsIndexDelta;
DEBUGLOG(8, "found dms match of length %u at distance %u (offCode=%u)",
- (U32)matchLength, curr - matchIndex, curr - matchIndex + ZSTD_REP_MOVE);
+ (U32)matchLength, curr - matchIndex, STORE_OFFSET(curr - matchIndex));
if (matchLength > matchEndIdx - matchIndex)
matchEndIdx = matchIndex + (U32)matchLength;
bestLength = matchLength;
- matches[mnum].off = (curr - matchIndex) + ZSTD_REP_MOVE;
+ matches[mnum].off = STORE_OFFSET(curr - matchIndex);
matches[mnum].len = (U32)matchLength;
mnum++;
if ( (matchLength > ZSTD_OPT_NUM)
@@ -835,7 +848,8 @@ GEN_ZSTD_BT_GET_ALL_MATCHES(dictMatchState)
ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, 6) \
}
-static ZSTD_getAllMatchesFn ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const dictMode)
+static ZSTD_getAllMatchesFn
+ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const dictMode)
{
ZSTD_getAllMatchesFn const getAllMatchesFns[3][4] = {
ZSTD_BT_GET_ALL_MATCHES_ARRAY(noDict),
@@ -854,16 +868,18 @@ static ZSTD_getAllMatchesFn ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const*
/* Struct containing info needed to make decision about ldm inclusion */
typedef struct {
- rawSeqStore_t seqStore; /* External match candidates store for this block */
- U32 startPosInBlock; /* Start position of the current match candidate */
- U32 endPosInBlock; /* End position of the current match candidate */
- U32 offset; /* Offset of the match candidate */
+ rawSeqStore_t seqStore; /* External match candidates store for this block */
+ U32 startPosInBlock; /* Start position of the current match candidate */
+ U32 endPosInBlock; /* End position of the current match candidate */
+ U32 offset; /* Offset of the match candidate */
} ZSTD_optLdm_t;
/* ZSTD_optLdm_skipRawSeqStoreBytes():
- * Moves forward in rawSeqStore by nbBytes, which will update the fields 'pos' and 'posInSequence'.
+ * Moves forward in @rawSeqStore by @nbBytes,
+ * which will update the fields 'pos' and 'posInSequence'.
*/
-static void ZSTD_optLdm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
+static void ZSTD_optLdm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes)
+{
U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes);
while (currPos && rawSeqStore->pos < rawSeqStore->size) {
rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos];
@@ -884,8 +900,10 @@ static void ZSTD_optLdm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t
* Calculates the beginning and end of the next match in the current block.
* Updates 'pos' and 'posInSequence' of the ldmSeqStore.
*/
-static void ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 currPosInBlock,
- U32 blockBytesRemaining) {
+static void
+ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 currPosInBlock,
+ U32 blockBytesRemaining)
+{
rawSeq currSeq;
U32 currBlockEndPos;
U32 literalsBytesRemaining;
@@ -897,8 +915,8 @@ static void ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 cu
optLdm->endPosInBlock = UINT_MAX;
return;
}
- /* Calculate appropriate bytes left in matchLength and litLength after adjusting
- based on ldmSeqStore->posInSequence */
+ /* Calculate appropriate bytes left in matchLength and litLength
+ * after adjusting based on ldmSeqStore->posInSequence */
currSeq = optLdm->seqStore.seq[optLdm->seqStore.pos];
assert(optLdm->seqStore.posInSequence <= currSeq.litLength + currSeq.matchLength);
currBlockEndPos = currPosInBlock + blockBytesRemaining;
@@ -934,15 +952,16 @@ static void ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 cu
}
/* ZSTD_optLdm_maybeAddMatch():
- * Adds a match if it's long enough, based on it's 'matchStartPosInBlock'
- * and 'matchEndPosInBlock', into 'matches'. Maintains the correct ordering of 'matches'
+ * Adds a match if it's long enough,
+ * based on it's 'matchStartPosInBlock' and 'matchEndPosInBlock',
+ * into 'matches'. Maintains the correct ordering of 'matches'.
*/
static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches,
- ZSTD_optLdm_t* optLdm, U32 currPosInBlock) {
- U32 posDiff = currPosInBlock - optLdm->startPosInBlock;
+ const ZSTD_optLdm_t* optLdm, U32 currPosInBlock)
+{
+ U32 const posDiff = currPosInBlock - optLdm->startPosInBlock;
/* Note: ZSTD_match_t actually contains offCode and matchLength (before subtracting MINMATCH) */
- U32 candidateMatchLength = optLdm->endPosInBlock - optLdm->startPosInBlock - posDiff;
- U32 candidateOffCode = optLdm->offset + ZSTD_REP_MOVE;
+ U32 const candidateMatchLength = optLdm->endPosInBlock - optLdm->startPosInBlock - posDiff;
/* Ensure that current block position is not outside of the match */
if (currPosInBlock < optLdm->startPosInBlock
@@ -952,6 +971,7 @@ static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches,
}
if (*nbMatches == 0 || ((candidateMatchLength > matches[*nbMatches-1].len) && *nbMatches < ZSTD_OPT_NUM)) {
+ U32 const candidateOffCode = STORE_OFFSET(optLdm->offset);
DEBUGLOG(6, "ZSTD_optLdm_maybeAddMatch(): Adding ldm candidate match (offCode: %u matchLength %u) at block position=%u",
candidateOffCode, candidateMatchLength, currPosInBlock);
matches[*nbMatches].len = candidateMatchLength;
@@ -963,8 +983,11 @@ static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches,
/* ZSTD_optLdm_processMatchCandidate():
* Wrapper function to update ldm seq store and call ldm functions as necessary.
*/
-static void ZSTD_optLdm_processMatchCandidate(ZSTD_optLdm_t* optLdm, ZSTD_match_t* matches, U32* nbMatches,
- U32 currPosInBlock, U32 remainingBytes) {
+static void
+ZSTD_optLdm_processMatchCandidate(ZSTD_optLdm_t* optLdm,
+ ZSTD_match_t* matches, U32* nbMatches,
+ U32 currPosInBlock, U32 remainingBytes)
+{
if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) {
return;
}
@@ -975,7 +998,7 @@ static void ZSTD_optLdm_processMatchCandidate(ZSTD_optLdm_t* optLdm, ZSTD_match_
* at the end of a match from the ldm seq store, and will often be some bytes
* over beyond matchEndPosInBlock. As such, we need to correct for these "overshoots"
*/
- U32 posOvershoot = currPosInBlock - optLdm->endPosInBlock;
+ U32 const posOvershoot = currPosInBlock - optLdm->endPosInBlock;
ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, posOvershoot);
}
ZSTD_opt_getNextMatchAndUpdateSeqStore(optLdm, currPosInBlock, remainingBytes);
@@ -1075,14 +1098,14 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
/* large match -> immediate encoding */
{ U32 const maxML = matches[nbMatches-1].len;
- U32 const maxOffset = matches[nbMatches-1].off;
+ U32 const maxOffcode = matches[nbMatches-1].off;
DEBUGLOG(6, "found %u matches of maxLength=%u and maxOffCode=%u at cPos=%u => start new series",
- nbMatches, maxML, maxOffset, (U32)(ip-prefixStart));
+ nbMatches, maxML, maxOffcode, (U32)(ip-prefixStart));
if (maxML > sufficient_len) {
lastSequence.litlen = litlen;
lastSequence.mlen = maxML;
- lastSequence.off = maxOffset;
+ lastSequence.off = maxOffcode;
DEBUGLOG(6, "large match (%u>%u), immediate encoding",
maxML, sufficient_len);
cur = 0;
@@ -1099,15 +1122,15 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
opt[pos].price = ZSTD_MAX_PRICE; /* mlen, litlen and price will be fixed during forward scanning */
}
for (matchNb = 0; matchNb < nbMatches; matchNb++) {
- U32 const offset = matches[matchNb].off;
+ U32 const offcode = matches[matchNb].off;
U32 const end = matches[matchNb].len;
for ( ; pos <= end ; pos++ ) {
- U32 const matchPrice = ZSTD_getMatchPrice(offset, pos, optStatePtr, optLevel);
+ U32 const matchPrice = ZSTD_getMatchPrice(offcode, pos, optStatePtr, optLevel);
U32 const sequencePrice = literalsPrice + matchPrice;
DEBUGLOG(7, "rPos:%u => set initial price : %.2f",
pos, ZSTD_fCost(sequencePrice));
opt[pos].mlen = pos;
- opt[pos].off = offset;
+ opt[pos].off = offcode;
opt[pos].litlen = litlen;
opt[pos].price = (int)sequencePrice;
} }
@@ -1152,7 +1175,7 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
assert(cur >= opt[cur].mlen);
if (opt[cur].mlen != 0) {
U32 const prev = cur - opt[cur].mlen;
- repcodes_t newReps = ZSTD_updateRep(opt[prev].rep, opt[cur].off, opt[cur].litlen==0);
+ repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, opt[cur].off, opt[cur].litlen==0);
ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(repcodes_t));
} else {
ZSTD_memcpy(opt[cur].rep, opt[cur - 1].rep, sizeof(repcodes_t));
@@ -1242,7 +1265,7 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
* update them while traversing the sequences.
*/
if (lastSequence.mlen != 0) {
- repcodes_t reps = ZSTD_updateRep(opt[cur].rep, lastSequence.off, lastSequence.litlen==0);
+ repcodes_t const reps = ZSTD_newRep(opt[cur].rep, lastSequence.off, lastSequence.litlen==0);
ZSTD_memcpy(rep, &reps, sizeof(reps));
} else {
ZSTD_memcpy(rep, opt[cur].rep, sizeof(repcodes_t));
@@ -1286,7 +1309,7 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
assert(anchor + llen <= iend);
ZSTD_updateStats(optStatePtr, llen, anchor, offCode, mlen);
- ZSTD_storeSeq(seqStore, llen, anchor, iend, offCode, mlen-MINMATCH);
+ ZSTD_storeSeq(seqStore, llen, anchor, iend, offCode, mlen);
anchor += advance;
ip = anchor;
} }
diff --git a/lib/compress/zstdmt_compress.c b/lib/compress/zstdmt_compress.c
index f564822d4b2..6bc14b035e1 100644
--- a/lib/compress/zstdmt_compress.c
+++ b/lib/compress/zstdmt_compress.c
@@ -102,9 +102,8 @@ typedef struct ZSTDMT_bufferPool_s {
buffer_t bTable[1]; /* variable size */
} ZSTDMT_bufferPool;
-static ZSTDMT_bufferPool* ZSTDMT_createBufferPool(unsigned nbWorkers, ZSTD_customMem cMem)
+static ZSTDMT_bufferPool* ZSTDMT_createBufferPool(unsigned maxNbBuffers, ZSTD_customMem cMem)
{
- unsigned const maxNbBuffers = 2*nbWorkers + 3;
ZSTDMT_bufferPool* const bufPool = (ZSTDMT_bufferPool*)ZSTD_customCalloc(
sizeof(ZSTDMT_bufferPool) + (maxNbBuffers-1) * sizeof(buffer_t), cMem);
if (bufPool==NULL) return NULL;
@@ -160,9 +159,8 @@ static void ZSTDMT_setBufferSize(ZSTDMT_bufferPool* const bufPool, size_t const
}
-static ZSTDMT_bufferPool* ZSTDMT_expandBufferPool(ZSTDMT_bufferPool* srcBufPool, U32 nbWorkers)
+static ZSTDMT_bufferPool* ZSTDMT_expandBufferPool(ZSTDMT_bufferPool* srcBufPool, unsigned maxNbBuffers)
{
- unsigned const maxNbBuffers = 2*nbWorkers + 3;
if (srcBufPool==NULL) return NULL;
if (srcBufPool->totalBuffers >= maxNbBuffers) /* good enough */
return srcBufPool;
@@ -171,7 +169,7 @@ static ZSTDMT_bufferPool* ZSTDMT_expandBufferPool(ZSTDMT_bufferPool* srcBufPool,
size_t const bSize = srcBufPool->bufferSize; /* forward parameters */
ZSTDMT_bufferPool* newBufPool;
ZSTDMT_freeBufferPool(srcBufPool);
- newBufPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ newBufPool = ZSTDMT_createBufferPool(maxNbBuffers, cMem);
if (newBufPool==NULL) return newBufPool;
ZSTDMT_setBufferSize(newBufPool, bSize);
return newBufPool;
@@ -263,6 +261,16 @@ static void ZSTDMT_releaseBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buf)
ZSTD_customFree(buf.start, bufPool->cMem);
}
+/* We need 2 output buffers per worker since each dstBuff must be flushed after it is released.
+ * The 3 additional buffers are as follows:
+ * 1 buffer for input loading
+ * 1 buffer for "next input" when submitting current one
+ * 1 buffer stuck in queue */
+#define BUF_POOL_MAX_NB_BUFFERS(nbWorkers) 2*nbWorkers + 3
+
+/* After a worker releases its rawSeqStore, it is immediately ready for reuse.
+ * So we only need one seq buffer per worker. */
+#define SEQ_POOL_MAX_NB_BUFFERS(nbWorkers) nbWorkers
/* ===== Seq Pool Wrapper ====== */
@@ -316,7 +324,7 @@ static void ZSTDMT_setNbSeq(ZSTDMT_seqPool* const seqPool, size_t const nbSeq)
static ZSTDMT_seqPool* ZSTDMT_createSeqPool(unsigned nbWorkers, ZSTD_customMem cMem)
{
- ZSTDMT_seqPool* const seqPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ ZSTDMT_seqPool* const seqPool = ZSTDMT_createBufferPool(SEQ_POOL_MAX_NB_BUFFERS(nbWorkers), cMem);
if (seqPool == NULL) return NULL;
ZSTDMT_setNbSeq(seqPool, 0);
return seqPool;
@@ -329,7 +337,7 @@ static void ZSTDMT_freeSeqPool(ZSTDMT_seqPool* seqPool)
static ZSTDMT_seqPool* ZSTDMT_expandSeqPool(ZSTDMT_seqPool* pool, U32 nbWorkers)
{
- return ZSTDMT_expandBufferPool(pool, nbWorkers);
+ return ZSTDMT_expandBufferPool(pool, SEQ_POOL_MAX_NB_BUFFERS(nbWorkers));
}
@@ -936,7 +944,7 @@ MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers,
mtctx->jobs = ZSTDMT_createJobsTable(&nbJobs, cMem);
assert(nbJobs > 0); assert((nbJobs & (nbJobs - 1)) == 0); /* ensure nbJobs is a power of 2 */
mtctx->jobIDMask = nbJobs - 1;
- mtctx->bufPool = ZSTDMT_createBufferPool(nbWorkers, cMem);
+ mtctx->bufPool = ZSTDMT_createBufferPool(BUF_POOL_MAX_NB_BUFFERS(nbWorkers), cMem);
mtctx->cctxPool = ZSTDMT_createCCtxPool(nbWorkers, cMem);
mtctx->seqPool = ZSTDMT_createSeqPool(nbWorkers, cMem);
initError = ZSTDMT_serialState_init(&mtctx->serial);
@@ -1039,7 +1047,7 @@ static size_t ZSTDMT_resize(ZSTDMT_CCtx* mtctx, unsigned nbWorkers)
{
if (POOL_resize(mtctx->factory, nbWorkers)) return ERROR(memory_allocation);
FORWARD_IF_ERROR( ZSTDMT_expandJobsTable(mtctx, nbWorkers) , "");
- mtctx->bufPool = ZSTDMT_expandBufferPool(mtctx->bufPool, nbWorkers);
+ mtctx->bufPool = ZSTDMT_expandBufferPool(mtctx->bufPool, BUF_POOL_MAX_NB_BUFFERS(nbWorkers));
if (mtctx->bufPool == NULL) return ERROR(memory_allocation);
mtctx->cctxPool = ZSTDMT_expandCCtxPool(mtctx->cctxPool, nbWorkers);
if (mtctx->cctxPool == NULL) return ERROR(memory_allocation);
diff --git a/lib/compress/zstdmt_compress.h b/lib/compress/zstdmt_compress.h
index 2fee2ec7455..271eb1ac71f 100644
--- a/lib/compress/zstdmt_compress.h
+++ b/lib/compress/zstdmt_compress.h
@@ -65,8 +65,11 @@ size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx);
* Private use only. Init streaming operation.
* expects params to be valid.
* must receive dict, or cdict, or none, but not both.
+ * mtctx can be freshly constructed or reused from a prior compression.
+ * If mtctx is reused, memory allocations from the prior compression may not be freed,
+ * even if they are not needed for the current compression.
* @return : 0, or an error code */
-size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* zcs,
+size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* mtctx,
const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType,
const ZSTD_CDict* cdict,
ZSTD_CCtx_params params, unsigned long long pledgedSrcSize);
diff --git a/lib/decompress/huf_decompress.c b/lib/decompress/huf_decompress.c
index fa61968ba48..2027188255e 100644
--- a/lib/decompress/huf_decompress.c
+++ b/lib/decompress/huf_decompress.c
@@ -664,7 +664,7 @@ size_t HUF_decompress4X1_usingDTable_internal_default(void* dst, size_t dstSize,
#if ZSTD_ENABLE_ASM_X86_64_BMI2
-HUF_ASM_DECL void HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args);
+HUF_ASM_DECL void HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args) ZSTDLIB_HIDDEN;
static HUF_ASM_X86_64_BMI2_ATTRS
size_t
@@ -1380,7 +1380,7 @@ size_t HUF_decompress4X2_usingDTable_internal_default(void* dst, size_t dstSize,
#if ZSTD_ENABLE_ASM_X86_64_BMI2
-HUF_ASM_DECL void HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args);
+HUF_ASM_DECL void HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args) ZSTDLIB_HIDDEN;
static HUF_ASM_X86_64_BMI2_ATTRS size_t
HUF_decompress4X2_usingDTable_internal_bmi2_asm(
diff --git a/lib/decompress/huf_decompress_amd64.S b/lib/decompress/huf_decompress_amd64.S
index 98173cce863..49589cb6114 100644
--- a/lib/decompress/huf_decompress_amd64.S
+++ b/lib/decompress/huf_decompress_amd64.S
@@ -1,14 +1,24 @@
-#include "../common/portability_macros.h"
+/*
+ * Copyright (c) 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).
+ * You may select, at your option, one of the above-listed licenses.
+ */
-#if ZSTD_ENABLE_ASM_X86_64_BMI2
+#include "../common/portability_macros.h"
/* Stack marking
* ref: https://wiki.gentoo.org/wiki/Hardened/GNU_stack_quickstart
*/
-#if defined(__linux__) && defined(__ELF__)
+#if defined(__ELF__) && defined(__GNUC__)
.section .note.GNU-stack,"",%progbits
#endif
+#if ZSTD_ENABLE_ASM_X86_64_BMI2
+
/* Calling convention:
*
* %rdi contains the first argument: HUF_DecompressAsmArgs*.
@@ -20,6 +30,10 @@
* TODO: Support Windows calling convention.
*/
+ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop)
+ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop)
+ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop)
+ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop)
.global HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop
.global HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop
.global _HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop
diff --git a/lib/dictBuilder/zdict.c b/lib/dictBuilder/zdict.c
index 8b8b381edd9..006aba7c9c4 100644
--- a/lib/dictBuilder/zdict.c
+++ b/lib/dictBuilder/zdict.c
@@ -682,8 +682,8 @@ static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
if (nbSeq >= 2) { /* rep offsets */
const seqDef* const seq = seqStorePtr->sequencesStart;
- U32 offset1 = seq[0].offset - 3;
- U32 offset2 = seq[1].offset - 3;
+ U32 offset1 = seq[0].offBase - ZSTD_REP_NUM;
+ U32 offset2 = seq[1].offBase - ZSTD_REP_NUM;
if (offset1 >= MAXREPOFFSET) offset1 = 0;
if (offset2 >= MAXREPOFFSET) offset2 = 0;
repOffsets[offset1] += 3;
diff --git a/lib/libzstd.mk b/lib/libzstd.mk
index af12daffe12..6e9a643954b 100644
--- a/lib/libzstd.mk
+++ b/lib/libzstd.mk
@@ -34,6 +34,8 @@ ZSTD_NO_ASM ?= 0
# libzstd helpers
##################################################################
+VOID ?= /dev/null
+
# Make 4.3 doesn't support '\#' anymore (https://lwn.net/Articles/810071/)
NUM_SYMBOL := \#
@@ -79,7 +81,7 @@ endif
CFLAGS += -fno-stack-protector -fomit-frame-pointer -fno-ident \
-DDYNAMIC_BMI2=0 -DNDEBUG
else
- CFLAGS += -O3
+ CFLAGS ?= -O3
endif
DEBUGLEVEL ?= 0
@@ -93,8 +95,24 @@ DEBUGFLAGS= -Wall -Wextra -Wcast-qual -Wcast-align -Wshadow \
-Wvla -Wformat=2 -Winit-self -Wfloat-equal -Wwrite-strings \
-Wredundant-decls -Wmissing-prototypes -Wc++-compat
CFLAGS += $(DEBUGFLAGS) $(MOREFLAGS)
+ASFLAGS += $(DEBUGFLAGS) $(MOREFLAGS) $(CFLAGS)
LDFLAGS += $(MOREFLAGS)
-FLAGS = $(CPPFLAGS) $(CFLAGS) $(LDFLAGS)
+FLAGS = $(CPPFLAGS) $(CFLAGS) $(ASFLAGS) $(LDFLAGS)
+
+ifndef ALREADY_APPENDED_NOEXECSTACK
+export ALREADY_APPENDED_NOEXECSTACK := 1
+ifeq ($(shell echo "int main(int argc, char* argv[]) { (void)argc; (void)argv; return 0; }" | $(CC) $(FLAGS) -z noexecstack -x c -Werror - -o $(VOID) 2>$(VOID) && echo 1 || echo 0),1)
+LDFLAGS += -z noexecstack
+endif
+ifeq ($(shell echo | $(CC) $(FLAGS) -Wa,--noexecstack -x assembler -Werror -c - -o $(VOID) 2>$(VOID) && echo 1 || echo 0),1)
+CFLAGS += -Wa,--noexecstack
+# CFLAGS are also added to ASFLAGS
+else ifeq ($(shell echo | $(CC) $(FLAGS) -Qunused-arguments -Wa,--noexecstack -x assembler -Werror -c - -o $(VOID) 2>$(VOID) && echo 1 || echo 0),1)
+# See e.g.: https://github.com/android/ndk/issues/171
+CFLAGS += -Qunused-arguments -Wa,--noexecstack
+# CFLAGS are also added to ASFLAGS
+endif
+endif
HAVE_COLORNEVER = $(shell echo a | grep --color=never a > /dev/null 2> /dev/null && echo 1 || echo 0)
GREP_OPTIONS ?=
diff --git a/lib/module.modulemap b/lib/module.modulemap
new file mode 100644
index 00000000000..bbb939782e1
--- /dev/null
+++ b/lib/module.modulemap
@@ -0,0 +1,25 @@
+module libzstd [extern_c] {
+ header "zstd.h"
+ export *
+ config_macros [exhaustive] /* zstd.h */ \
+ ZSTD_STATIC_LINKING_ONLY, \
+ ZSTDLIB_VISIBLE, \
+ ZSTD_DLL_EXPORT, \
+ ZSTDLIB_STATIC_API, \
+ ZSTD_DISABLE_DEPRECATE_WARNINGS, \
+ ZSTD_CLEVEL_DEFAULT, \
+ /* zdict.h */ ZDICT_STATIC_LINKING_ONLY, \
+ ZDICTLIB_VISIBILITY, \
+ ZDICT_DISABLE_DEPRECATE_WARNINGS, \
+ /* zstd_errors.h */ ZSTDERRORLIB_VISIBILITY
+
+ module dictbuilder [extern_c] {
+ header "zdict.h"
+ export *
+ }
+
+ module errors [extern_c] {
+ header "zstd_errors.h"
+ export *
+ }
+}
diff --git a/lib/modulemap/module.modulemap b/lib/modulemap/module.modulemap
deleted file mode 100644
index eeda6988543..00000000000
--- a/lib/modulemap/module.modulemap
+++ /dev/null
@@ -1,4 +0,0 @@
-module libzstd [extern_c] {
- header "../zstd.h"
- export *
-}
diff --git a/lib/zstd.h b/lib/zstd.h
index 66ec125d00e..a88ae7bf8ed 100644
--- a/lib/zstd.h
+++ b/lib/zstd.h
@@ -74,7 +74,7 @@ extern "C" {
/*------ Version ------*/
#define ZSTD_VERSION_MAJOR 1
#define ZSTD_VERSION_MINOR 5
-#define ZSTD_VERSION_RELEASE 1
+#define ZSTD_VERSION_RELEASE 2
#define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE)
/*! ZSTD_versionNumber() :
diff --git a/programs/Makefile b/programs/Makefile
index a54900cc1e9..f77e1b7f10f 100644
--- a/programs/Makefile
+++ b/programs/Makefile
@@ -62,8 +62,6 @@ else
EXT =
endif
-VOID = /dev/null
-
# thread detection
NO_THREAD_MSG := ==> no threads, building without multithreading support
HAVE_PTHREAD := $(shell printf '$(NUM_SYMBOL)include \nint main(void) { return 0; }' > have_pthread.c && $(CC) $(FLAGS) -o have_pthread$(EXT) have_pthread.c -pthread 2> $(VOID) && rm have_pthread$(EXT) && echo 1 || echo 0; rm have_pthread.c)
@@ -322,7 +320,7 @@ $(BUILD_DIR)/%.o : %.c $(BUILD_DIR)/%.d | $(BUILD_DIR)
$(BUILD_DIR)/%.o : %.S | $(BUILD_DIR)
@echo AS $@
- $(COMPILE.c) $(OUTPUT_OPTION) $<
+ $(COMPILE.S) $(OUTPUT_OPTION) $<
MKDIR ?= mkdir
$(BUILD_DIR): ; $(MKDIR) -p $@
diff --git a/programs/benchzstd.c b/programs/benchzstd.c
index 1e4d717d153..fa2659efbbb 100644
--- a/programs/benchzstd.c
+++ b/programs/benchzstd.c
@@ -31,9 +31,14 @@
#include "timefn.h" /* UTIL_time_t */
#include "benchfn.h"
#include "../lib/common/mem.h"
+#ifndef ZSTD_STATIC_LINKING_ONLY
#define ZSTD_STATIC_LINKING_ONLY
+#endif
#include "../lib/zstd.h"
#include "datagen.h" /* RDG_genBuffer */
+#ifndef XXH_INLINE_ALL
+#define XXH_INLINE_ALL
+#endif
#include "../lib/common/xxhash.h"
#include "benchzstd.h"
#include "../lib/zstd_errors.h"
diff --git a/programs/dibio.c b/programs/dibio.c
index 04860dbbfa6..d19f954486f 100644
--- a/programs/dibio.c
+++ b/programs/dibio.c
@@ -31,7 +31,6 @@
#include "timefn.h" /* UTIL_time_t, UTIL_clockSpanMicro, UTIL_getTime */
#include "../lib/common/mem.h" /* read */
-#include "../lib/common/error_private.h"
#include "dibio.h"
diff --git a/programs/fileio.c b/programs/fileio.c
index 379d334eb02..5338fa62955 100644
--- a/programs/fileio.c
+++ b/programs/fileio.c
@@ -1679,6 +1679,7 @@ static int FIO_compressFilename_dstFile(FIO_ctx_t* const fCtx,
if (ress.dstFile == NULL) {
int dstFilePermissions = DEFAULT_FILE_PERMISSIONS;
if ( strcmp (srcFileName, stdinmark)
+ && strcmp (dstFileName, stdoutmark)
&& UTIL_stat(srcFileName, &statbuf)
&& UTIL_isRegularFileStat(&statbuf) ) {
dstFilePermissions = statbuf.st_mode;
@@ -2245,7 +2246,7 @@ FIO_decompressZstdFrame(FIO_ctx_t* const fCtx, dRess_t* ress, FILE* finput,
ZSTD_inBuffer inBuff = { ress->srcBuffer, ress->srcBufferLoaded, 0 };
ZSTD_outBuffer outBuff= { ress->dstBuffer, ress->dstBufferSize, 0 };
size_t const readSizeHint = ZSTD_decompressStream(ress->dctx, &outBuff, &inBuff);
- const int displayLevel = (!fCtx->hasStdoutOutput || g_display_prefs.progressSetting == FIO_ps_always) ? 1 : 2;
+ const int displayLevel = (g_display_prefs.progressSetting == FIO_ps_always) ? 1 : 2;
UTIL_HumanReadableSize_t const hrs = UTIL_makeHumanReadableSize(alreadyDecoded+frameSize);
if (ZSTD_isError(readSizeHint)) {
DISPLAYLEVEL(1, "%s : Decoding error (36) : %s \n",
@@ -2634,6 +2635,7 @@ static int FIO_decompressDstFile(FIO_ctx_t* const fCtx,
if ((ress.dstFile == NULL) && (prefs->testMode==0)) {
int dstFilePermissions = DEFAULT_FILE_PERMISSIONS;
if ( strcmp(srcFileName, stdinmark) /* special case : don't transfer permissions from stdin */
+ && strcmp(dstFileName, stdoutmark)
&& UTIL_stat(srcFileName, &statbuf)
&& UTIL_isRegularFileStat(&statbuf) ) {
dstFilePermissions = statbuf.st_mode;
diff --git a/programs/zstd.1 b/programs/zstd.1
index da5814963dd..c7a19dbacac 100644
--- a/programs/zstd.1
+++ b/programs/zstd.1
@@ -1,345 +1,501 @@
-.TH "ZSTD" "1" "December 2021" "zstd 1.5.1" "User Commands"
+.
+.TH "ZSTD" "1" "January 2022" "zstd 1.5.2" "User Commands"
+.
.SH "NAME"
\fBzstd\fR \- zstd, zstdmt, unzstd, zstdcat \- Compress or decompress \.zst files
+.
.SH "SYNOPSIS"
-.TS
-allbox;
-\fBzstd\fR [\fIOPTIONS\fR] [\- \fIINPUT\-FILE\fR] [\-o \fIOUTPUT\-FILE\fR]
-.TE
+\fBzstd\fR [\fIOPTIONS\fR] [\-|\fIINPUT\-FILE\fR] [\-o \fIOUTPUT\-FILE\fR]
+.
.P
\fBzstdmt\fR is equivalent to \fBzstd \-T0\fR
+.
.P
\fBunzstd\fR is equivalent to \fBzstd \-d\fR
+.
.P
\fBzstdcat\fR is equivalent to \fBzstd \-dcf\fR
+.
.SH "DESCRIPTION"
\fBzstd\fR is a fast lossless compression algorithm and data compression tool, with command line syntax similar to \fBgzip (1)\fR and \fBxz (1)\fR\. It is based on the \fBLZ77\fR family, with further FSE & huff0 entropy stages\. \fBzstd\fR offers highly configurable compression speed, with fast modes at > 200 MB/s per core, and strong modes nearing lzma compression ratios\. It also features a very fast decoder, with speeds > 500 MB/s per core\.
+.
.P
\fBzstd\fR command line syntax is generally similar to gzip, but features the following differences :
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
Source files are preserved by default\. It\'s possible to remove them automatically by using the \fB\-\-rm\fR command\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
When compressing a single file, \fBzstd\fR displays progress notifications and result summary by default\. Use \fB\-q\fR to turn them off\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fBzstd\fR does not accept input from console, but it properly accepts \fBstdin\fR when it\'s not the console\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fBzstd\fR displays a short help page when command line is an error\. Use \fB\-q\fR to turn it off\.
+.
.IP "" 0
+.
.P
\fBzstd\fR compresses or decompresses each \fIfile\fR according to the selected operation mode\. If no \fIfiles\fR are given or \fIfile\fR is \fB\-\fR, \fBzstd\fR reads from standard input and writes the processed data to standard output\. \fBzstd\fR will refuse to write compressed data to standard output if it is a terminal : it will display an error message and skip the \fIfile\fR\. Similarly, \fBzstd\fR will refuse to read compressed data from standard input if it is a terminal\.
+.
.P
Unless \fB\-\-stdout\fR or \fB\-o\fR is specified, \fIfiles\fR are written to a new file whose name is derived from the source \fIfile\fR name:
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
When compressing, the suffix \fB\.zst\fR is appended to the source filename to get the target filename\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
When decompressing, the \fB\.zst\fR suffix is removed from the source filename to get the target filename
+.
.IP "" 0
+.
.SS "Concatenation with \.zst files"
It is possible to concatenate \fB\.zst\fR files as is\. \fBzstd\fR will decompress such files as if they were a single \fB\.zst\fR file\.
+.
.SH "OPTIONS"
+.
.SS "Integer suffixes and special values"
In most places where an integer argument is expected, an optional suffix is supported to easily indicate large integers\. There must be no space between the integer and the suffix\.
+.
.TP
\fBKiB\fR
-Multiply the integer by 1,024 (2\e^10)\. \fBKi\fR, \fBK\fR, and \fBKB\fR are accepted as synonyms for \fBKiB\fR\.
+Multiply the integer by 1,024 (2^10)\. \fBKi\fR, \fBK\fR, and \fBKB\fR are accepted as synonyms for \fBKiB\fR\.
+.
.TP
\fBMiB\fR
-Multiply the integer by 1,048,576 (2\e^20)\. \fBMi\fR, \fBM\fR, and \fBMB\fR are accepted as synonyms for \fBMiB\fR\.
+Multiply the integer by 1,048,576 (2^20)\. \fBMi\fR, \fBM\fR, and \fBMB\fR are accepted as synonyms for \fBMiB\fR\.
+.
.SS "Operation mode"
If multiple operation mode options are given, the last one takes effect\.
+.
.TP
\fB\-z\fR, \fB\-\-compress\fR
Compress\. This is the default operation mode when no operation mode option is specified and no other operation mode is implied from the command name (for example, \fBunzstd\fR implies \fB\-\-decompress\fR)\.
+.
.TP
\fB\-d\fR, \fB\-\-decompress\fR, \fB\-\-uncompress\fR
Decompress\.
+.
.TP
\fB\-t\fR, \fB\-\-test\fR
Test the integrity of compressed \fIfiles\fR\. This option is equivalent to \fB\-\-decompress \-\-stdout\fR except that the decompressed data is discarded instead of being written to standard output\. No files are created or removed\.
+.
.TP
\fB\-b#\fR
Benchmark file(s) using compression level #
+.
.TP
\fB\-\-train FILEs\fR
Use FILEs as a training set to create a dictionary\. The training set should contain a lot of small files (> 100)\.
+.
.TP
\fB\-l\fR, \fB\-\-list\fR
Display information related to a zstd compressed file, such as size, ratio, and checksum\. Some of these fields may not be available\. This command can be augmented with the \fB\-v\fR modifier\.
+.
.SS "Operation modifiers"
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-#\fR: \fB#\fR compression level [1\-19] (default: 3)
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-ultra\fR: unlocks high compression levels 20+ (maximum 22), using a lot more memory\. Note that decompression will also require more memory when using these levels\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-fast[=#]\fR: switch to ultra\-fast compression levels\. If \fB=#\fR is not present, it defaults to \fB1\fR\. The higher the value, the faster the compression speed, at the cost of some compression ratio\. This setting overwrites compression level if one was set previously\. Similarly, if a compression level is set after \fB\-\-fast\fR, it overrides it\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-T#\fR, \fB\-\-threads=#\fR: Compress using \fB#\fR working threads (default: 1)\. If \fB#\fR is 0, attempt to detect and use the number of physical CPU cores\. In all cases, the nb of threads is capped to \fBZSTDMT_NBWORKERS_MAX\fR, which is either 64 in 32\-bit mode, or 256 for 64\-bit environments\. This modifier does nothing if \fBzstd\fR is compiled without multithread support\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-single\-thread\fR: Does not spawn a thread for compression, use a single thread for both I/O and compression\. In this mode, compression is serialized with I/O, which is slightly slower\. (This is different from \fB\-T1\fR, which spawns 1 compression thread in parallel of I/O)\. This mode is the only one available when multithread support is disabled\. Single\-thread mode features lower memory usage\. Final compressed result is slightly different from \fB\-T1\fR\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-auto\-threads={physical,logical} (default: physical)\fR: When using a default amount of threads via \fB\-T0\fR, choose the default based on the number of detected physical or logical cores\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-adapt[=min=#,max=#]\fR : \fBzstd\fR will dynamically adapt compression level to perceived I/O conditions\. Compression level adaptation can be observed live by using command \fB\-v\fR\. Adaptation can be constrained between supplied \fBmin\fR and \fBmax\fR levels\. The feature works when combined with multi\-threading and \fB\-\-long\fR mode\. It does not work with \fB\-\-single\-thread\fR\. It sets window size to 8 MB by default (can be changed manually, see \fBwlog\fR)\. Due to the chaotic nature of dynamic adaptation, compressed result is not reproducible\. \fInote\fR : at the time of this writing, \fB\-\-adapt\fR can remain stuck at low speed when combined with multiple worker threads (>=2)\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-long[=#]\fR: enables long distance matching with \fB#\fR \fBwindowLog\fR, if not \fB#\fR is not present it defaults to \fB27\fR\. This increases the window size (\fBwindowLog\fR) and memory usage for both the compressor and decompressor\. This setting is designed to improve the compression ratio for files with long matches at a large distance\.
+.
.IP
Note: If \fBwindowLog\fR is set to larger than 27, \fB\-\-long=windowLog\fR or \fB\-\-memory=windowSize\fR needs to be passed to the decompressor\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-D DICT\fR: use \fBDICT\fR as Dictionary to compress or decompress FILE(s)
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-patch\-from FILE\fR: Specify the file to be used as a reference point for zstd\'s diff engine\. This is effectively dictionary compression with some convenient parameter selection, namely that windowSize > srcSize\.
+.
.IP
Note: cannot use both this and \-D together Note: \fB\-\-long\fR mode will be automatically activated if chainLog < fileLog (fileLog being the windowLog required to cover the whole file)\. You can also manually force it\. Node: for all levels, you can use \-\-patch\-from in \-\-single\-thread mode to improve compression ratio at the cost of speed Note: for level 19, you can get increased compression ratio at the cost of speed by specifying \fB\-\-zstd=targetLength=\fR to be something large (i\.e 4096), and by setting a large \fB\-\-zstd=chainLog=\fR
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-rsyncable\fR : \fBzstd\fR will periodically synchronize the compression state to make the compressed file more rsync\-friendly\. There is a negligible impact to compression ratio, and the faster compression levels will see a small compression speed hit\. This feature does not work with \fB\-\-single\-thread\fR\. You probably don\'t want to use it with long range mode, since it will decrease the effectiveness of the synchronization points, but your mileage may vary\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-C\fR, \fB\-\-[no\-]check\fR: add integrity check computed from uncompressed data (default: enabled)
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-[no\-]content\-size\fR: enable / disable whether or not the original size of the file is placed in the header of the compressed file\. The default option is \-\-content\-size (meaning that the original size will be placed in the header)\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-no\-dictID\fR: do not store dictionary ID within frame header (dictionary compression)\. The decoder will have to rely on implicit knowledge about which dictionary to use, it won\'t be able to check if it\'s correct\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-M#\fR, \fB\-\-memory=#\fR: Set a memory usage limit\. By default, Zstandard uses 128 MB for decompression as the maximum amount of memory the decompressor is allowed to use, but you can override this manually if need be in either direction (ie\. you can increase or decrease it)\.
+.
.IP
This is also used during compression when using with \-\-patch\-from=\. In this case, this parameter overrides that maximum size allowed for a dictionary\. (128 MB)\.
+.
.IP
Additionally, this can be used to limit memory for dictionary training\. This parameter overrides the default limit of 2 GB\. zstd will load training samples up to the memory limit and ignore the rest\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-stream\-size=#\fR : Sets the pledged source size of input coming from a stream\. This value must be exact, as it will be included in the produced frame header\. Incorrect stream sizes will cause an error\. This information will be used to better optimize compression parameters, resulting in better and potentially faster compression, especially for smaller source sizes\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-size\-hint=#\fR: When handling input from a stream, \fBzstd\fR must guess how large the source size will be when optimizing compression parameters\. If the stream size is relatively small, this guess may be a poor one, resulting in a higher compression ratio than expected\. This feature allows for controlling the guess when needed\. Exact guesses result in better compression ratios\. Overestimates result in slightly degraded compression ratios, while underestimates may result in significant degradation\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-o FILE\fR: save result into \fBFILE\fR
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-f\fR, \fB\-\-force\fR: disable input and output checks\. Allows overwriting existing files, input from console, output to stdout, operating on links, block devices, etc\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-c\fR, \fB\-\-stdout\fR: write to standard output (even if it is the console)
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-[no\-]sparse\fR: enable / disable sparse FS support, to make files with many zeroes smaller on disk\. Creating sparse files may save disk space and speed up decompression by reducing the amount of disk I/O\. default: enabled when output is into a file, and disabled when output is stdout\. This setting overrides default and can force sparse mode over stdout\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-rm\fR: remove source file(s) after successful compression or decompression\. If used in combination with \-o, will trigger a confirmation prompt (which can be silenced with \-f), as this is a destructive operation\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-k\fR, \fB\-\-keep\fR: keep source file(s) after successful compression or decompression\. This is the default behavior\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-r\fR: operate recursively on directories\. It selects all files in the named directory and all its subdirectories\. This can be useful both to reduce command line typing, and to circumvent shell expansion limitations, when there are a lot of files and naming breaks the maximum size of a command line\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-filelist FILE\fR read a list of files to process as content from \fBFILE\fR\. Format is compatible with \fBls\fR output, with one file per line\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-output\-dir\-flat DIR\fR: resulting files are stored into target \fBDIR\fR directory, instead of same directory as origin file\. Be aware that this command can introduce name collision issues, if multiple files, from different directories, end up having the same name\. Collision resolution ensures first file with a given name will be present in \fBDIR\fR, while in combination with \fB\-f\fR, the last file will be present instead\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-output\-dir\-mirror DIR\fR: similar to \fB\-\-output\-dir\-flat\fR, the output files are stored underneath target \fBDIR\fR directory, but this option will replicate input directory hierarchy into output \fBDIR\fR\.
+.
.IP
If input directory contains "\.\.", the files in this directory will be ignored\. If input directory is an absolute directory (i\.e\. "/var/tmp/abc"), it will be stored into the "output\-dir/var/tmp/abc"\. If there are multiple input files or directories, name collision resolution will follow the same rules as \fB\-\-output\-dir\-flat\fR\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-format=FORMAT\fR: compress and decompress in other formats\. If compiled with support, zstd can compress to or decompress from other compression algorithm formats\. Possibly available options are \fBzstd\fR, \fBgzip\fR, \fBxz\fR, \fBlzma\fR, and \fBlz4\fR\. If no such format is provided, \fBzstd\fR is the default\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-h\fR/\fB\-H\fR, \fB\-\-help\fR: display help/long help and exit
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-V\fR, \fB\-\-version\fR: display version number and exit\. Advanced : \fB\-vV\fR also displays supported formats\. \fB\-vvV\fR also displays POSIX support\. \fB\-q\fR will only display the version number, suitable for machine reading\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-v\fR, \fB\-\-verbose\fR: verbose mode, display more information
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-q\fR, \fB\-\-quiet\fR: suppress warnings, interactivity, and notifications\. specify twice to suppress errors too\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-no\-progress\fR: do not display the progress bar, but keep all other messages\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-show\-default\-cparams\fR: Shows the default compression parameters that will be used for a particular src file\. If the provided src file is not a regular file (eg\. named pipe), the cli will just output the default parameters\. That is, the parameters that are used when the src size is unknown\.
-.IP "\[ci]" 4
+.
+.IP "\(bu" 4
\fB\-\-\fR: All arguments after \fB\-\-\fR are treated as files
+.
.IP "" 0
+.
.SS "Restricted usage of Environment Variables"
Using environment variables to set parameters has security implications\. Therefore, this avenue is intentionally restricted\. Only \fBZSTD_CLEVEL\fR and \fBZSTD_NBTHREADS\fR are currently supported\. They set the compression level and number of threads to use during compression, respectively\.
+.
.P
\fBZSTD_CLEVEL\fR can be used to set the level between 1 and 19 (the "normal" range)\. If the value of \fBZSTD_CLEVEL\fR is not a valid integer, it will be ignored with a warning message\. \fBZSTD_CLEVEL\fR just replaces the default compression level (\fB3\fR)\.
+.
.P
\fBZSTD_NBTHREADS\fR can be used to set the number of threads \fBzstd\fR will attempt to use during compression\. If the value of \fBZSTD_NBTHREADS\fR is not a valid unsigned integer, it will be ignored with a warning message\. \fBZSTD_NBTHREADS\fR has a default value of (\fB1\fR), and is capped at ZSTDMT_NBWORKERS_MAX==200\. \fBzstd\fR must be compiled with multithread support for this to have any effect\.
+.
.P
They can both be overridden by corresponding command line arguments: \fB\-#\fR for compression level and \fB\-T#\fR for number of compression threads\.
+.
.SH "DICTIONARY BUILDER"
\fBzstd\fR offers \fIdictionary\fR compression, which greatly improves efficiency on small files and messages\. It\'s possible to train \fBzstd\fR with a set of samples, the result of which is saved into a file called a \fBdictionary\fR\. Then during compression and decompression, reference the same dictionary, using command \fB\-D dictionaryFileName\fR\. Compression of small files similar to the sample set will be greatly improved\.
+.
.TP
\fB\-\-train FILEs\fR
Use FILEs as training set to create a dictionary\. The training set should contain a lot of small files (> 100), and weight typically 100x the target dictionary size (for example, 10 MB for a 100 KB dictionary)\. \fB\-\-train\fR can be combined with \fB\-r\fR to indicate a directory rather than listing all the files, which can be useful to circumvent shell expansion limits\.
+.
.IP
\fB\-\-train\fR supports multithreading if \fBzstd\fR is compiled with threading support (default)\. Additional parameters can be specified with \fB\-\-train\-fastcover\fR\. The legacy dictionary builder can be accessed with \fB\-\-train\-legacy\fR\. The slower cover dictionary builder can be accessed with \fB\-\-train\-cover\fR\. Default is equivalent to \fB\-\-train\-fastcover=d=8,steps=4\fR\.
+.
.TP
\fB\-o file\fR
Dictionary saved into \fBfile\fR (default name: dictionary)\.
+.
.TP
\fB\-\-maxdict=#\fR
Limit dictionary to specified size (default: 112640)\.
+.
.TP
\fB\-#\fR
Use \fB#\fR compression level during training (optional)\. Will generate statistics more tuned for selected compression level, resulting in a \fIsmall\fR compression ratio improvement for this level\.
+.
.TP
\fB\-B#\fR
Split input files into blocks of size # (default: no split)
+.
.TP
\fB\-M#\fR, \fB\-\-memory=#\fR
Limit the amount of sample data loaded for training (default: 2 GB)\. See above for details\.
+.
.TP
\fB\-\-dictID=#\fR
A dictionary ID is a locally unique ID that a decoder can use to verify it is using the right dictionary\. By default, zstd will create a 4\-bytes random number ID\. It\'s possible to give a precise number instead\. Short numbers have an advantage : an ID < 256 will only need 1 byte in the compressed frame header, and an ID < 65536 will only need 2 bytes\. This compares favorably to 4 bytes default\. However, it\'s up to the dictionary manager to not assign twice the same ID to 2 different dictionaries\.
+.
.TP
\fB\-\-train\-cover[=k#,d=#,steps=#,split=#,shrink[=#]]\fR
Select parameters for the default dictionary builder algorithm named cover\. If \fId\fR is not specified, then it tries \fId\fR = 6 and \fId\fR = 8\. If \fIk\fR is not specified, then it tries \fIsteps\fR values in the range [50, 2000]\. If \fIsteps\fR is not specified, then the default value of 40 is used\. If \fIsplit\fR is not specified or split <= 0, then the default value of 100 is used\. Requires that \fId\fR <= \fIk\fR\. If \fIshrink\fR flag is not used, then the default value for \fIshrinkDict\fR of 0 is used\. If \fIshrink\fR is not specified, then the default value for \fIshrinkDictMaxRegression\fR of 1 is used\.
+.
.IP
Selects segments of size \fIk\fR with highest score to put in the dictionary\. The score of a segment is computed by the sum of the frequencies of all the subsegments of size \fId\fR\. Generally \fId\fR should be in the range [6, 8], occasionally up to 16, but the algorithm will run faster with d <= \fI8\fR\. Good values for \fIk\fR vary widely based on the input data, but a safe range is [2 * \fId\fR, 2000]\. If \fIsplit\fR is 100, all input samples are used for both training and testing to find optimal \fId\fR and \fIk\fR to build dictionary\. Supports multithreading if \fBzstd\fR is compiled with threading support\. Having \fIshrink\fR enabled takes a truncated dictionary of minimum size and doubles in size until compression ratio of the truncated dictionary is at most \fIshrinkDictMaxRegression%\fR worse than the compression ratio of the largest dictionary\.
+.
.IP
Examples:
+.
.IP
\fBzstd \-\-train\-cover FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=k=50,d=8 FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=d=8,steps=500 FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=k=50 FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=k=50,split=60 FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=shrink FILEs\fR
+.
.IP
\fBzstd \-\-train\-cover=shrink=2 FILEs\fR
+.
.TP
\fB\-\-train\-fastcover[=k#,d=#,f=#,steps=#,split=#,accel=#]\fR
Same as cover but with extra parameters \fIf\fR and \fIaccel\fR and different default value of split If \fIsplit\fR is not specified, then it tries \fIsplit\fR = 75\. If \fIf\fR is not specified, then it tries \fIf\fR = 20\. Requires that 0 < \fIf\fR < 32\. If \fIaccel\fR is not specified, then it tries \fIaccel\fR = 1\. Requires that 0 < \fIaccel\fR <= 10\. Requires that \fId\fR = 6 or \fId\fR = 8\.
+.
.IP
\fIf\fR is log of size of array that keeps track of frequency of subsegments of size \fId\fR\. The subsegment is hashed to an index in the range [0,2^\fIf\fR \- 1]\. It is possible that 2 different subsegments are hashed to the same index, and they are considered as the same subsegment when computing frequency\. Using a higher \fIf\fR reduces collision but takes longer\.
+.
.IP
Examples:
+.
.IP
\fBzstd \-\-train\-fastcover FILEs\fR
+.
.IP
\fBzstd \-\-train\-fastcover=d=8,f=15,accel=2 FILEs\fR
+.
.TP
\fB\-\-train\-legacy[=selectivity=#]\fR
Use legacy dictionary builder algorithm with the given dictionary \fIselectivity\fR (default: 9)\. The smaller the \fIselectivity\fR value, the denser the dictionary, improving its efficiency but reducing its possible maximum size\. \fB\-\-train\-legacy=s=#\fR is also accepted\.
+.
.IP
Examples:
+.
.IP
\fBzstd \-\-train\-legacy FILEs\fR
+.
.IP
\fBzstd \-\-train\-legacy=selectivity=8 FILEs\fR
+.
.SH "BENCHMARK"
+.
.TP
\fB\-b#\fR
benchmark file(s) using compression level #
+.
.TP
\fB\-e#\fR
benchmark file(s) using multiple compression levels, from \fB\-b#\fR to \fB\-e#\fR (inclusive)
+.
.TP
\fB\-i#\fR
minimum evaluation time, in seconds (default: 3s), benchmark mode only
+.
.TP
\fB\-B#\fR, \fB\-\-block\-size=#\fR
cut file(s) into independent blocks of size # (default: no block)
+.
.TP
\fB\-\-priority=rt\fR
set process priority to real\-time
+.
.P
\fBOutput Format:\fR CompressionLevel#Filename : IntputSize \-> OutputSize (CompressionRatio), CompressionSpeed, DecompressionSpeed
+.
.P
\fBMethodology:\fR For both compression and decompression speed, the entire input is compressed/decompressed in\-memory to measure speed\. A run lasts at least 1 sec, so when files are small, they are compressed/decompressed several times per run, in order to improve measurement accuracy\.
+.
.SH "ADVANCED COMPRESSION OPTIONS"
-### \-B#: Select the size of each compression job\. This parameter is only available when multi\-threading is enabled\. Each compression job is run in parallel, so this value indirectly impacts the nb of active threads\. Default job size varies depending on compression level (generally \fB4 * windowSize\fR)\. \fB\-B#\fR makes it possible to manually select a custom size\. Note that job size must respect a minimum value which is enforced transparently\. This minimum is either 512 KB, or \fBoverlapSize\fR, whichever is largest\. Different job sizes will lead to (slightly) different compressed frames\.
+.
+.SS "\-B#:"
+Select the size of each compression job\. This parameter is only available when multi\-threading is enabled\. Each compression job is run in parallel, so this value indirectly impacts the nb of active threads\. Default job size varies depending on compression level (generally \fB4 * windowSize\fR)\. \fB\-B#\fR makes it possible to manually select a custom size\. Note that job size must respect a minimum value which is enforced transparently\. This minimum is either 512 KB, or \fBoverlapSize\fR, whichever is largest\. Different job sizes will lead to (slightly) different compressed frames\.
+.
.SS "\-\-zstd[=options]:"
\fBzstd\fR provides 22 predefined compression levels\. The selected or default predefined compression level can be changed with advanced compression options\. The \fIoptions\fR are provided as a comma\-separated list\. You may specify only the options you want to change and the rest will be taken from the selected or default compression level\. The list of available \fIoptions\fR:
+.
.TP
\fBstrategy\fR=\fIstrat\fR, \fBstrat\fR=\fIstrat\fR
Specify a strategy used by a match finder\.
+.
.IP
There are 9 strategies numbered from 1 to 9, from faster to stronger: 1=ZSTD_fast, 2=ZSTD_dfast, 3=ZSTD_greedy, 4=ZSTD_lazy, 5=ZSTD_lazy2, 6=ZSTD_btlazy2, 7=ZSTD_btopt, 8=ZSTD_btultra, 9=ZSTD_btultra2\.
+.
.TP
\fBwindowLog\fR=\fIwlog\fR, \fBwlog\fR=\fIwlog\fR
Specify the maximum number of bits for a match distance\.
+.
.IP
The higher number of increases the chance to find a match which usually improves compression ratio\. It also increases memory requirements for the compressor and decompressor\. The minimum \fIwlog\fR is 10 (1 KiB) and the maximum is 30 (1 GiB) on 32\-bit platforms and 31 (2 GiB) on 64\-bit platforms\.
+.
.IP
Note: If \fBwindowLog\fR is set to larger than 27, \fB\-\-long=windowLog\fR or \fB\-\-memory=windowSize\fR needs to be passed to the decompressor\.
+.
.TP
\fBhashLog\fR=\fIhlog\fR, \fBhlog\fR=\fIhlog\fR
Specify the maximum number of bits for a hash table\.
+.
.IP
Bigger hash tables cause less collisions which usually makes compression faster, but requires more memory during compression\.
+.
.IP
The minimum \fIhlog\fR is 6 (64 B) and the maximum is 30 (1 GiB)\.
+.
.TP
\fBchainLog\fR=\fIclog\fR, \fBclog\fR=\fIclog\fR
Specify the maximum number of bits for a hash chain or a binary tree\.
+.
.IP
Higher numbers of bits increases the chance to find a match which usually improves compression ratio\. It also slows down compression speed and increases memory requirements for compression\. This option is ignored for the ZSTD_fast strategy\.
+.
.IP
The minimum \fIclog\fR is 6 (64 B) and the maximum is 29 (524 Mib) on 32\-bit platforms and 30 (1 Gib) on 64\-bit platforms\.
+.
.TP
\fBsearchLog\fR=\fIslog\fR, \fBslog\fR=\fIslog\fR
Specify the maximum number of searches in a hash chain or a binary tree using logarithmic scale\.
+.
.IP
More searches increases the chance to find a match which usually increases compression ratio but decreases compression speed\.
+.
.IP
The minimum \fIslog\fR is 1 and the maximum is \'windowLog\' \- 1\.
+.
.TP
\fBminMatch\fR=\fImml\fR, \fBmml\fR=\fImml\fR
Specify the minimum searched length of a match in a hash table\.
+.
.IP
Larger search lengths usually decrease compression ratio but improve decompression speed\.
+.
.IP
The minimum \fImml\fR is 3 and the maximum is 7\.
+.
.TP
\fBtargetLength\fR=\fItlen\fR, \fBtlen\fR=\fItlen\fR
The impact of this field vary depending on selected strategy\.
+.
.IP
For ZSTD_btopt, ZSTD_btultra and ZSTD_btultra2, it specifies the minimum match length that causes match finder to stop searching\. A larger \fBtargetLength\fR usually improves compression ratio but decreases compression speed\. t For ZSTD_fast, it triggers ultra\-fast mode when > 0\. The value represents the amount of data skipped between match sampling\. Impact is reversed : a larger \fBtargetLength\fR increases compression speed but decreases compression ratio\.
+.
.IP
For all other strategies, this field has no impact\.
+.
.IP
The minimum \fItlen\fR is 0 and the maximum is 128 Kib\.
+.
.TP
\fBoverlapLog\fR=\fIovlog\fR, \fBovlog\fR=\fIovlog\fR
Determine \fBoverlapSize\fR, amount of data reloaded from previous job\. This parameter is only available when multithreading is enabled\. Reloading more data improves compression ratio, but decreases speed\.
+.
.IP
The minimum \fIovlog\fR is 0, and the maximum is 9\. 1 means "no overlap", hence completely independent jobs\. 9 means "full overlap", meaning up to \fBwindowSize\fR is reloaded from previous job\. Reducing \fIovlog\fR by 1 reduces the reloaded amount by a factor 2\. For example, 8 means "windowSize/2", and 6 means "windowSize/8"\. Value 0 is special and means "default" : \fIovlog\fR is automatically determined by \fBzstd\fR\. In which case, \fIovlog\fR will range from 6 to 9, depending on selected \fIstrat\fR\.
+.
.TP
\fBldmHashLog\fR=\fIlhlog\fR, \fBlhlog\fR=\fIlhlog\fR
Specify the maximum size for a hash table used for long distance matching\.
+.
.IP
This option is ignored unless long distance matching is enabled\.
+.
.IP
Bigger hash tables usually improve compression ratio at the expense of more memory during compression and a decrease in compression speed\.
+.
.IP
The minimum \fIlhlog\fR is 6 and the maximum is 30 (default: 20)\.
+.
.TP
\fBldmMinMatch\fR=\fIlmml\fR, \fBlmml\fR=\fIlmml\fR
Specify the minimum searched length of a match for long distance matching\.
+.
.IP
This option is ignored unless long distance matching is enabled\.
+.
.IP
Larger/very small values usually decrease compression ratio\.
+.
.IP
The minimum \fIlmml\fR is 4 and the maximum is 4096 (default: 64)\.
+.
.TP
\fBldmBucketSizeLog\fR=\fIlblog\fR, \fBlblog\fR=\fIlblog\fR
Specify the size of each bucket for the hash table used for long distance matching\.
+.
.IP
This option is ignored unless long distance matching is enabled\.
+.
.IP
Larger bucket sizes improve collision resolution but decrease compression speed\.
+.
.IP
The minimum \fIlblog\fR is 1 and the maximum is 8 (default: 3)\.
+.
.TP
\fBldmHashRateLog\fR=\fIlhrlog\fR, \fBlhrlog\fR=\fIlhrlog\fR
Specify the frequency of inserting entries into the long distance matching hash table\.
+.
.IP
This option is ignored unless long distance matching is enabled\.
+.
.IP
Larger values will improve compression speed\. Deviating far from the default value will likely result in a decrease in compression ratio\.
+.
.IP
The default value is \fBwlog \- lhlog\fR\.
+.
.SS "Example"
The following parameters sets advanced compression options to something similar to predefined level 19 for files bigger than 256 KB:
+.
.P
\fB\-\-zstd\fR=wlog=23,clog=23,hlog=22,slog=6,mml=3,tlen=48,strat=6
+.
.SH "BUGS"
Report bugs at: https://github\.com/facebook/zstd/issues
+.
.SH "AUTHOR"
Yann Collet
diff --git a/programs/zstdgrep.1 b/programs/zstdgrep.1
index 06465cebc35..563696d339e 100644
--- a/programs/zstdgrep.1
+++ b/programs/zstdgrep.1
@@ -1,17 +1,26 @@
-.TH "ZSTDGREP" "1" "December 2021" "zstd 1.5.1" "User Commands"
+.
+.TH "ZSTDGREP" "1" "January 2022" "zstd 1.5.2" "User Commands"
+.
.SH "NAME"
\fBzstdgrep\fR \- print lines matching a pattern in zstandard\-compressed files
+.
.SH "SYNOPSIS"
-\fBzstdgrep\fR [\fIgrep\-flags\fR] [\-\-] \fIpattern\fR [\fIfiles\fR \|\.\|\.\|\.]
+\fBzstdgrep\fR [\fIgrep\-flags\fR] [\-\-] \fIpattern\fR [\fIfiles\fR \.\.\.]
+.
.SH "DESCRIPTION"
\fBzstdgrep\fR runs \fBgrep (1)\fR on files, or \fBstdin\fR if no files argument is given, after decompressing them with \fBzstdcat (1)\fR\.
+.
.P
The grep\-flags and pattern arguments are passed on to \fBgrep (1)\fR\. If an \fB\-e\fR flag is found in the \fBgrep\-flags\fR, \fBzstdgrep\fR will not look for a pattern argument\.
+.
.P
Note that modern \fBgrep\fR alternatives such as \fBripgrep\fR (\fBrg\fR) support \fBzstd\fR\-compressed files out of the box, and can prove better alternatives than \fBzstdgrep\fR notably for unsupported complex pattern searches\. Note though that such alternatives may also feature some minor command line differences\.
+.
.SH "EXIT STATUS"
In case of missing arguments or missing pattern, 1 will be returned, otherwise 0\.
+.
.SH "SEE ALSO"
\fBzstd (1)\fR
+.
.SH "AUTHORS"
Thomas Klausner \fIwiz@NetBSD\.org\fR
diff --git a/programs/zstdless.1 b/programs/zstdless.1
index d90ea0f5465..ab38e7a7f45 100644
--- a/programs/zstdless.1
+++ b/programs/zstdless.1
@@ -1,9 +1,14 @@
-.TH "ZSTDLESS" "1" "December 2021" "zstd 1.5.1" "User Commands"
+.
+.TH "ZSTDLESS" "1" "January 2022" "zstd 1.5.2" "User Commands"
+.
.SH "NAME"
\fBzstdless\fR \- view zstandard\-compressed files
+.
.SH "SYNOPSIS"
-\fBzstdless\fR [\fIflags\fR] [\fIfile\fR \|\.\|\.\|\.]
+\fBzstdless\fR [\fIflags\fR] [\fIfile\fR \.\.\.]
+.
.SH "DESCRIPTION"
\fBzstdless\fR runs \fBless (1)\fR on files or stdin, if no files argument is given, after decompressing them with \fBzstdcat (1)\fR\.
+.
.SH "SEE ALSO"
\fBzstd (1)\fR
diff --git a/tests/Makefile b/tests/Makefile
index efcbe9761e9..132fa7a0818 100644
--- a/tests/Makefile
+++ b/tests/Makefile
@@ -118,7 +118,7 @@ zstdd_%.o : $(ZSTDDIR)/decompress/%.c
$(CC) -c $(CPPFLAGS) $(CFLAGS) $< -o $@
zstdd_%.o : $(ZSTDDIR)/decompress/%.S
- $(CC) -c $(CPPFLAGS) $(CFLAGS) $< -o $@
+ $(CC) -c $(CPPFLAGS) $(ASFLAGS) $< -o $@
zstdmt%.o : CPPFLAGS += $(MULTITHREAD_CPP)
@@ -132,7 +132,7 @@ zstdmt_d_%.o : $(ZSTDDIR)/decompress/%.c
$(CC) -c $(CPPFLAGS) $(CFLAGS) $< -o $@
zstdmt_d_%.o : $(ZSTDDIR)/decompress/%.S
- $(CC) -c $(CPPFLAGS) $(CFLAGS) $< -o $@
+ $(CC) -c $(CPPFLAGS) $(ASFLAGS) $< -o $@
fullbench32: CPPFLAGS += -m32
fullbench fullbench32 : CPPFLAGS += $(MULTITHREAD_CPP) -Wno-deprecated-declarations
diff --git a/tests/decodecorpus.c b/tests/decodecorpus.c
index 4f4db85753b..1037a36596c 100644
--- a/tests/decodecorpus.c
+++ b/tests/decodecorpus.c
@@ -633,17 +633,16 @@ static inline void initSeqStore(seqStore_t *seqStore) {
}
/* Randomly generate sequence commands */
-static U32 generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
- size_t contentSize, size_t literalsSize, dictInfo info)
+static U32
+generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
+ size_t contentSize, size_t literalsSize, dictInfo info)
{
/* The total length of all the matches */
size_t const remainingMatch = contentSize - literalsSize;
size_t excessMatch = 0;
U32 numSequences = 0;
-
U32 i;
-
const BYTE* literals = LITERAL_BUFFER;
BYTE* srcPtr = frame->src;
@@ -703,32 +702,31 @@ static U32 generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
lenPastStart = MIN(lenPastStart+MIN_SEQ_LEN, (U32)info.dictContentSize);
offset = (U32)((BYTE*)srcPtr - (BYTE*)frame->srcStart) + lenPastStart;
}
- {
- U32 const matchLenBound = MIN(frame->header.windowSize, lenPastStart);
+ { U32 const matchLenBound = MIN(frame->header.windowSize, lenPastStart);
matchLen = MIN(matchLen, matchLenBound);
}
}
}
- offsetCode = offset + ZSTD_REP_MOVE;
+ offsetCode = STORE_OFFSET(offset);
repIndex = 2;
} else {
/* do a repeat offset */
- offsetCode = RAND(seed) % 3;
+ U32 const randomRepIndex = RAND(seed) % 3;
+ offsetCode = STORE_REPCODE(randomRepIndex + 1); /* expects values between 1 & 3 */
if (literalLen > 0) {
- offset = frame->stats.rep[offsetCode];
- repIndex = offsetCode;
+ offset = frame->stats.rep[randomRepIndex];
+ repIndex = randomRepIndex;
} else {
- /* special case */
- offset = offsetCode == 2 ? frame->stats.rep[0] - 1
- : frame->stats.rep[offsetCode + 1];
- repIndex = MIN(2, offsetCode + 1);
+ /* special case : literalLen == 0 */
+ offset = randomRepIndex == 2 ? frame->stats.rep[0] - 1
+ : frame->stats.rep[randomRepIndex + 1];
+ repIndex = MIN(2, randomRepIndex + 1);
}
}
} while (((!info.useDict) && (offset > (size_t)((BYTE*)srcPtr - (BYTE*)frame->srcStart))) || offset == 0);
- {
+ { BYTE* const dictEnd = info.dictContent + info.dictContentSize;
size_t j;
- BYTE* const dictEnd = info.dictContent + info.dictContentSize;
for (j = 0; j < matchLen; j++) {
if ((U32)((BYTE*)srcPtr - (BYTE*)frame->srcStart) < offset) {
/* copy from dictionary instead of literals */
@@ -739,8 +737,7 @@ static U32 generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
*srcPtr = *(srcPtr-offset);
}
srcPtr++;
- }
- }
+ } }
{ int r;
for (r = repIndex; r > 0; r--) {
@@ -754,12 +751,12 @@ static U32 generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
DISPLAYLEVEL(7, " srcPos: %8u seqNb: %3u",
(unsigned)((BYTE*)srcPtr - (BYTE*)frame->srcStart), (unsigned)i);
DISPLAYLEVEL(6, "\n");
- if (offsetCode < 3) {
+ if (STORED_IS_REPCODE(offsetCode)) { /* expects sumtype numeric representation of ZSTD_storeSeq() */
DISPLAYLEVEL(7, " repeat offset: %d\n", (int)repIndex);
}
/* use libzstd sequence handling */
ZSTD_storeSeq(seqStore, literalLen, literals, literals + literalLen,
- offsetCode, matchLen - MINMATCH);
+ offsetCode, matchLen);
literalsSize -= literalLen;
excessMatch -= (matchLen - MIN_SEQ_LEN);
@@ -949,9 +946,9 @@ static size_t writeSequences(U32* seed, frame_t* frame, seqStore_t* seqStorePtr,
FSE_initCState2(&stateLitLength, CTable_LitLength, llCodeTable[nbSeq-1]);
BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
- BIT_addBits(&blockStream, sequences[nbSeq-1].matchLength, ML_bits[mlCodeTable[nbSeq-1]]);
+ BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]);
if (MEM_32bits()) BIT_flushBits(&blockStream);
- BIT_addBits(&blockStream, sequences[nbSeq-1].offset, ofCodeTable[nbSeq-1]);
+ BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]);
BIT_flushBits(&blockStream);
{ size_t n;
@@ -971,9 +968,9 @@ static size_t writeSequences(U32* seed, frame_t* frame, seqStore_t* seqStorePtr,
BIT_flushBits(&blockStream); /* (7)*/
BIT_addBits(&blockStream, sequences[n].litLength, llBits);
if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream);
- BIT_addBits(&blockStream, sequences[n].matchLength, mlBits);
+ BIT_addBits(&blockStream, sequences[n].mlBase, mlBits);
if (MEM_32bits()) BIT_flushBits(&blockStream); /* (7)*/
- BIT_addBits(&blockStream, sequences[n].offset, ofBits); /* 31 */
+ BIT_addBits(&blockStream, sequences[n].offBase, ofBits); /* 31 */
BIT_flushBits(&blockStream); /* (7)*/
} }
diff --git a/tests/fullbench.c b/tests/fullbench.c
index f610fef4c75..b55ff767f2f 100644
--- a/tests/fullbench.c
+++ b/tests/fullbench.c
@@ -108,7 +108,25 @@ local_ZSTD_compress(const void* src, size_t srcSize,
p.fParams = f;
p.cParams = *(ZSTD_compressionParameters*)payload;
return ZSTD_compress_advanced (g_zcc, dst, dstSize, src, srcSize, NULL ,0, p);
- //return ZSTD_compress(dst, dstSize, src, srcSize, cLevel);
+}
+
+static size_t
+local_ZSTD_compress_freshCCtx(const void* src, size_t srcSize,
+ void* dst, size_t dstSize,
+ void* payload)
+{
+ ZSTD_parameters p;
+ ZSTD_frameParameters f = { 1 /* contentSizeHeader*/, 0, 0 };
+ p.fParams = f;
+ p.cParams = *(ZSTD_compressionParameters*)payload;
+ if (g_zcc != NULL) ZSTD_freeCCtx(g_zcc);
+ g_zcc = ZSTD_createCCtx();
+ assert(g_zcc != NULL);
+ { size_t const r = ZSTD_compress_advanced (g_zcc, dst, dstSize, src, srcSize, NULL ,0, p);
+ ZSTD_freeCCtx(g_zcc);
+ g_zcc = NULL;
+ return r;
+ }
}
static size_t g_cSize = 0;
@@ -230,14 +248,15 @@ local_ZSTD_compressStream_freshCCtx(const void* src, size_t srcSize,
void* dst, size_t dstCapacity,
void* payload)
{
- ZSTD_CCtx* const cctx = ZSTD_createCCtx();
- size_t r;
- assert(cctx != NULL);
-
- r = local_ZSTD_compressStream(src, srcSize, dst, dstCapacity, payload);
-
- ZSTD_freeCCtx(cctx);
- return r;
+ if (g_cstream != NULL) ZSTD_freeCCtx(g_cstream);
+ g_cstream = ZSTD_createCCtx();
+ assert(g_cstream != NULL);
+
+ { size_t const r = local_ZSTD_compressStream(src, srcSize, dst, dstCapacity, payload);
+ ZSTD_freeCCtx(g_cstream);
+ g_cstream = NULL;
+ return r;
+ }
}
static size_t
@@ -430,6 +449,9 @@ static int benchMem(unsigned benchNb,
case 2:
benchFunction = local_ZSTD_decompress; benchName = "decompress";
break;
+ case 3:
+ benchFunction = local_ZSTD_compress_freshCCtx; benchName = "compress_freshCCtx";
+ break;
#ifndef ZSTD_DLL_IMPORT
case 11:
benchFunction = local_ZSTD_compressContinue; benchName = "compressContinue";
@@ -508,7 +530,6 @@ static int benchMem(unsigned benchNb,
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_targetLength, (int)cparams.targetLength);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_strategy, cparams.strategy);
-
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_compressionLevel, cLevel);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_windowLog, (int)cparams.windowLog);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_hashLog, (int)cparams.hashLog);
@@ -527,6 +548,9 @@ static int benchMem(unsigned benchNb,
case 2:
g_cSize = ZSTD_compress(dstBuff2, dstBuffSize, src, srcSize, cLevel);
break;
+ case 3:
+ payload = &cparams;
+ break;
#ifndef ZSTD_DLL_IMPORT
case 11:
payload = &cparams;
diff --git a/tests/fuzz/.gitignore b/tests/fuzz/.gitignore
index 02c2f10beb1..28f2b8da4db 100644
--- a/tests/fuzz/.gitignore
+++ b/tests/fuzz/.gitignore
@@ -17,6 +17,7 @@ decompress_dstSize_tooSmall
fse_read_ncount
sequence_compression_api
seekable_roundtrip
+huf_decompress
huf_round_trip
fuzz-*.log
rt_lib_*
diff --git a/tests/fuzz/Makefile b/tests/fuzz/Makefile
index b707bd95176..912348c3843 100644
--- a/tests/fuzz/Makefile
+++ b/tests/fuzz/Makefile
@@ -12,6 +12,7 @@
CFLAGS ?= -O3
CXXFLAGS ?= -O3
CPPFLAGS ?=
+ASFLAGS ?=
LDFLAGS ?=
ARFLAGS ?=
LIB_FUZZING_ENGINE ?= libregression.a
@@ -43,6 +44,7 @@ FUZZ_EXTRA_FLAGS := -Wall -Wextra -Wcast-qual -Wcast-align -Wshadow \
-Wredundant-decls -Wno-deprecated-declarations \
-g -fno-omit-frame-pointer
FUZZ_CFLAGS := $(FUZZ_EXTRA_FLAGS) $(CFLAGS)
+FUZZ_ASFLAGS := $(FUZZ_EXTRA_FLAGS) $(ASFLAGS)
FUZZ_CXXFLAGS := $(FUZZ_EXTRA_FLAGS) -std=c++11 $(CXXFLAGS)
FUZZ_LDFLAGS := -pthread $(LDFLAGS)
FUZZ_ARFLAGS := $(ARFLAGS)
@@ -127,7 +129,7 @@ rt_lib_decompress_%.o: $(ZSTDDIR)/decompress/%.c
$(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $(FUZZ_ROUND_TRIP_FLAGS) $< -c -o $@
rt_lib_decompress_%.o: $(ZSTDDIR)/decompress/%.S
- $(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $(FUZZ_ROUND_TRIP_FLAGS) $< -c -o $@
+ $(CC) $(FUZZ_CPPFLAGS) $(FUZZ_ASFLAGS) $(FUZZ_ROUND_TRIP_FLAGS) $< -c -o $@
rt_lib_dictBuilder_%.o: $(ZSTDDIR)/dictBuilder/%.c
$(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $(FUZZ_ROUND_TRIP_FLAGS) $< -c -o $@
@@ -151,7 +153,7 @@ d_lib_decompress_%.o: $(ZSTDDIR)/decompress/%.c
$(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $< -c -o $@
d_lib_decompress_%.o: $(ZSTDDIR)/decompress/%.S
- $(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $< -c -o $@
+ $(CC) $(FUZZ_CPPFLAGS) $(FUZZ_ASFLAGS) $< -c -o $@
d_lib_dictBuilder_%.o: $(ZSTDDIR)/dictBuilder/%.c
$(CC) $(FUZZ_CPPFLAGS) $(FUZZ_CFLAGS) $< -c -o $@
diff --git a/tests/fuzz/sequence_compression_api.c b/tests/fuzz/sequence_compression_api.c
index cc840bf801a..a2959e1aca6 100644
--- a/tests/fuzz/sequence_compression_api.c
+++ b/tests/fuzz/sequence_compression_api.c
@@ -72,10 +72,10 @@ static size_t decodeSequences(void* dst, size_t nbSequences,
size_t literalsSize, const void* dict, size_t dictSize) {
const uint8_t* litPtr = literalsBuffer;
const uint8_t* const litBegin = literalsBuffer;
- const uint8_t* const litEnd = literalsBuffer + literalsSize;
+ const uint8_t* const litEnd = litBegin + literalsSize;
const uint8_t* dictPtr = dict;
uint8_t* op = dst;
- const uint8_t* const oend = dst + ZSTD_FUZZ_GENERATED_SRC_MAXSIZE;
+ const uint8_t* const oend = (uint8_t*)dst + ZSTD_FUZZ_GENERATED_SRC_MAXSIZE;
size_t generatedSrcBufferSize = 0;
size_t bytesWritten = 0;
uint32_t lastLLSize;
diff --git a/tests/fuzz/zstd_helpers.c b/tests/fuzz/zstd_helpers.c
index 0fbf3bed7ab..f66579754ff 100644
--- a/tests/fuzz/zstd_helpers.c
+++ b/tests/fuzz/zstd_helpers.c
@@ -123,7 +123,7 @@ FUZZ_dict_t FUZZ_train(void const* src, size_t srcSize, FUZZ_dataProducer_t *pro
size_t const offset = FUZZ_dataProducer_uint32Range(producer, 0, MAX(srcSize, 1) - 1);
size_t const limit = MIN(srcSize - offset, remaining);
size_t const toCopy = MIN(limit, remaining / (nbSamples - sample));
- memcpy(samples + pos, src + offset, toCopy);
+ memcpy(samples + pos, (const char*)src + offset, toCopy);
pos += toCopy;
samplesSizes[sample] = toCopy;
}
diff --git a/tests/playTests.sh b/tests/playTests.sh
index a772b61ff7c..b7a3d88a817 100755
--- a/tests/playTests.sh
+++ b/tests/playTests.sh
@@ -304,6 +304,13 @@ zstd -d -f tmp_corrupt.zst --no-check
zstd -d -f tmp_corrupt.zst --check --no-check # final flag overrides
zstd -d -f tmp.zst --no-check
+if [ "$isWindows" = false ]; then
+ if [ -n "$(which readelf)" ]; then
+ println "test: check if binary has executable stack (#2963)"
+ readelf -lW "$ZSTD_BIN" | grep 'GNU_STACK .* RW ' || die "zstd binary has executable stack!"
+ fi
+fi
+
println "\n===> zstdgrep tests"
ln -sf "$ZSTD_BIN" zstdcat
rm -f tmp_grep
@@ -1359,21 +1366,21 @@ zstd -d tmp.tzst
rm -f tmp.tar tmp.tzst
if [ $GZIPMODE -eq 1 ]; then
- tar -c tmp | gzip > tmp.tgz
+ tar -f - -c tmp | gzip > tmp.tgz
zstd -d tmp.tgz
[ -e tmp.tar ] || die ".tgz failed to decompress to .tar!"
rm -f tmp.tar tmp.tgz
fi
if [ $LZMAMODE -eq 1 ]; then
- tar -c tmp | zstd --format=xz > tmp.txz
+ tar -f - -c tmp | zstd --format=xz > tmp.txz
zstd -d tmp.txz
[ -e tmp.tar ] || die ".txz failed to decompress to .tar!"
rm -f tmp.tar tmp.txz
fi
if [ $LZ4MODE -eq 1 ]; then
- tar -c tmp | zstd --format=lz4 > tmp.tlz4
+ tar -f - -c tmp | zstd --format=lz4 > tmp.tlz4
zstd -d tmp.tlz4
[ -e tmp.tar ] || die ".tlz4 failed to decompress to .tar!"
rm -f tmp.tar tmp.tlz4
diff --git a/tests/test-license.py b/tests/test-license.py
index 224776536dc..f27cb677431 100755
--- a/tests/test-license.py
+++ b/tests/test-license.py
@@ -43,6 +43,7 @@ def to_abs(d):
"Makefile",
".mk",
".py",
+ ".S",
]
# License should certainly be in the first 10 KB.