krak/zstd
krak/zstd
1
0
Fork 0
mirror of https://github.com/facebook/zstd.git synced 2025-03-06 08:49:28 +02:00
Code Issues Releases 1 Activity

Merge pull request #4217 from facebook/ZSTD_compressSequencesAndLiterals

Browse Source 
ZSTD_compressSequencesAndLiterals
This commit is contained in:
Yann Collet 2024-12-26 11:41:44 -08:00 committed by GitHub
commit 788926fe48
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
50 changed files with 1796 additions and 1206 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.github/workflows/dev-long-tests.yml vendored
View File

@ -188,7 +188,7 @@ jobs:
- name: ASan + UBSan + Regression Test
run: make -j uasanregressiontest
clang-ubsan-regression:
clang-asan-ubsan-regression:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # tag=v4.1.1

189
build/VS2010/fullbench-dll/fullbench-dll.vcxproj
View File

@ -1,189 +0,0 @@
<?xml version="1.0" encoding="utf-8"?>
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<ItemGroup Label="ProjectConfigurations">
<ProjectConfiguration Include="Debug|Win32">
<Configuration>Debug</Configuration>
<Platform>Win32</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Debug|x64">
<Configuration>Debug</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Release|Win32">
<Configuration>Release</Configuration>
<Platform>Win32</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Release|x64">
<Configuration>Release</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
</ItemGroup>
<PropertyGroup Label="Globals">
<ProjectGuid>{00000000-1CC8-4FD7-9281-6B8DBB9D3DF8}</ProjectGuid>
<Keyword>Win32Proj</Keyword>
<RootNamespace>fullbench-dll</RootNamespace>
<OutDir>$(SolutionDir)bin\$(Platform)_$(Configuration)\</OutDir>
<IntDir>$(SolutionDir)bin\obj\$(RootNamespace)_$(Platform)_$(Configuration)\</IntDir>
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">
<ConfigurationType>Application</ConfigurationType>
<UseDebugLibraries>true</UseDebugLibraries>
<CharacterSet>MultiByte</CharacterSet>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration">
<ConfigurationType>Application</ConfigurationType>
<UseDebugLibraries>true</UseDebugLibraries>
<CharacterSet>MultiByte</CharacterSet>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
<ConfigurationType>Application</ConfigurationType>
<UseDebugLibraries>false</UseDebugLibraries>
<WholeProgramOptimization>true</WholeProgramOptimization>
<CharacterSet>MultiByte</CharacterSet>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration">
<ConfigurationType>Application</ConfigurationType>
<UseDebugLibraries>false</UseDebugLibraries>
<WholeProgramOptimization>true</WholeProgramOptimization>
<CharacterSet>MultiByte</CharacterSet>
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
</ImportGroup>
<ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
<Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
</ImportGroup>
<ImportGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="PropertySheets">
<Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
</ImportGroup>
<ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
<Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
</ImportGroup>
<ImportGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="PropertySheets">
<Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
</ImportGroup>
<PropertyGroup Label="UserMacros" />
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
<LinkIncremental>true</LinkIncremental>
<IncludePath>$(IncludePath);$(SolutionDir)..\..\lib;$(SolutionDir)..\..\programs;$(SolutionDir)..\..\lib\legacy;$(SolutionDir)..\..\lib\common;$(UniversalCRT_IncludePath);</IncludePath>
<RunCodeAnalysis>false</RunCodeAnalysis>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
<LinkIncremental>true</LinkIncremental>
<IncludePath>$(IncludePath);$(SolutionDir)..\..\lib;$(SolutionDir)..\..\programs;$(SolutionDir)..\..\lib\legacy;$(SolutionDir)..\..\lib\common;$(UniversalCRT_IncludePath);</IncludePath>
<RunCodeAnalysis>false</RunCodeAnalysis>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
<LinkIncremental>false</LinkIncremental>
<IncludePath>$(IncludePath);$(SolutionDir)..\..\lib;$(SolutionDir)..\..\programs;$(SolutionDir)..\..\lib\legacy;$(SolutionDir)..\..\lib\common;$(UniversalCRT_IncludePath);</IncludePath>
<RunCodeAnalysis>false</RunCodeAnalysis>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
<LinkIncremental>false</LinkIncremental>
<IncludePath>$(IncludePath);$(SolutionDir)..\..\lib;$(SolutionDir)..\..\programs;$(SolutionDir)..\..\lib\legacy;$(SolutionDir)..\..\lib\common;$(UniversalCRT_IncludePath);</IncludePath>
<RunCodeAnalysis>false</RunCodeAnalysis>
</PropertyGroup>
<ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
<ClCompile>
<PrecompiledHeader>
</PrecompiledHeader>
<WarningLevel>Level4</WarningLevel>
<Optimization>Disabled</Optimization>
<PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;ZSTD_DLL_IMPORT=1;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<TreatWarningAsError>true</TreatWarningAsError>
<EnablePREfast>false</EnablePREfast>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<GenerateDebugInformation>true</GenerateDebugInformation>
<AdditionalLibraryDirectories>$(SolutionDir)bin\$(Platform)_$(Configuration);%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
<AdditionalDependencies>libzstd.lib;%(AdditionalDependencies)</AdditionalDependencies>
<ImageHasSafeExceptionHandlers>false</ImageHasSafeExceptionHandlers>
</Link>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
<ClCompile>
<PrecompiledHeader>
</PrecompiledHeader>
<WarningLevel>Level4</WarningLevel>
<Optimization>Disabled</Optimization>
<PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;ZSTD_DLL_IMPORT=1;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<TreatWarningAsError>true</TreatWarningAsError>
<EnablePREfast>false</EnablePREfast>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<GenerateDebugInformation>true</GenerateDebugInformation>
<AdditionalLibraryDirectories>$(SolutionDir)bin\$(Platform)_$(Configuration);%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
<AdditionalDependencies>libzstd.lib;%(AdditionalDependencies)</AdditionalDependencies>
</Link>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
<ClCompile>
<WarningLevel>Level4</WarningLevel>
<PrecompiledHeader>
</PrecompiledHeader>
<Optimization>MaxSpeed</Optimization>
<FunctionLevelLinking>true</FunctionLevelLinking>
<IntrinsicFunctions>true</IntrinsicFunctions>
<PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;ZSTD_DLL_IMPORT=1;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<EnablePREfast>false</EnablePREfast>
<TreatWarningAsError>false</TreatWarningAsError>
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<GenerateDebugInformation>true</GenerateDebugInformation>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<OptimizeReferences>true</OptimizeReferences>
<AdditionalLibraryDirectories>$(SolutionDir)bin\$(Platform)_$(Configuration);%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
<AdditionalDependencies>libzstd.lib;%(AdditionalDependencies)</AdditionalDependencies>
<ImageHasSafeExceptionHandlers>false</ImageHasSafeExceptionHandlers>
</Link>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
<ClCompile>
<WarningLevel>Level4</WarningLevel>
<PrecompiledHeader>
</PrecompiledHeader>
<Optimization>MaxSpeed</Optimization>
<FunctionLevelLinking>true</FunctionLevelLinking>
<IntrinsicFunctions>true</IntrinsicFunctions>
<PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;ZSTD_DLL_IMPORT=1;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<TreatWarningAsError>false</TreatWarningAsError>
<EnablePREfast>false</EnablePREfast>
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<GenerateDebugInformation>true</GenerateDebugInformation>
<EnableCOMDATFolding>true</EnableCOMDATFolding>
<OptimizeReferences>true</OptimizeReferences>
<AdditionalLibraryDirectories>$(SolutionDir)bin\$(Platform)_$(Configuration);%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
<AdditionalDependencies>libzstd.lib;%(AdditionalDependencies)</AdditionalDependencies>
</Link>
</ItemDefinitionGroup>
<ItemGroup>
<ClCompile Include="..\..\..\lib\common\xxhash.c" />
<ClCompile Include="..\..\..\programs\util.c" />
<ClCompile Include="..\..\..\programs\timefn.c" />
<ClCompile Include="..\..\..\programs\datagen.c" />
<ClCompile Include="..\..\..\programs\benchfn.c" />
<ClCompile Include="..\..\..\tests\fullbench.c" />
</ItemGroup>
<ItemGroup>
<ClInclude Include="..\..\..\lib\zstd.h" />
<ClInclude Include="..\..\..\programs\datagen.h" />
<ClInclude Include="..\..\..\programs\benchfn.h" />
<ClInclude Include="..\..\..\programs\util.h" />
</ItemGroup>
<ItemGroup>
<ProjectReference Include="..\libzstd-dll\libzstd-dll.vcxproj">
<Project>{00000000-94d5-4bf9-8a50-7bd9929a0850}</Project>
</ProjectReference>
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
</ImportGroup>
</Project>

1
build/VS2010/fullbench/fullbench.vcxproj
View File

@ -184,6 +184,7 @@
<ClCompile Include="..\..\..\programs\util.c" />
<ClCompile Include="..\..\..\programs\timefn.c" />
<ClCompile Include="..\..\..\programs\datagen.c" />
<ClCompile Include="..\..\..\programs\lorem.c" />
<ClCompile Include="..\..\..\programs\benchfn.c" />
<ClCompile Include="..\..\..\tests\fullbench.c" />
</ItemGroup>

5
build/VS2010/zstd.sln
View File

@ -7,11 +7,6 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "fuzzer", "fuzzer\fuzzer.vcx
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "fullbench", "fullbench\fullbench.vcxproj", "{61ABD629-1CC8-4FD7-9281-6B8DBB9D3DF8}"
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "fullbench-dll", "fullbench-dll\fullbench-dll.vcxproj", "{00000000-1CC8-4FD7-9281-6B8DBB9D3DF8}"
ProjectSection(ProjectDependencies) = postProject
{00000000-94D5-4BF9-8A50-7BD9929A0850} = {00000000-94D5-4BF9-8A50-7BD9929A0850}
EndProjectSection
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "datagen", "datagen\datagen.vcxproj", "{037E781E-81A6-494B-B1B3-438AB1200523}"
EndProject
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "libzstd", "libzstd\libzstd.vcxproj", "{8BFD8150-94D5-4BF9-8A50-7BD9929A0850}"

1
build/single_file_libs/.gitignore vendored
View File

@ -4,6 +4,7 @@ zstddeclib.c
zstdenclib.c
zstd.c
zstd.h
zstd_errors.h
# test artifacts
temp*

2
contrib/largeNbDicts/largeNbDicts.c
View File

@ -1025,7 +1025,7 @@ int main (int argc, const char** argv)
unsigned nbBlocks = 0; /* determine nbBlocks automatically, from source and blockSize */
ZSTD_dictContentType_e dictContentType = ZSTD_dct_auto;
ZSTD_dictAttachPref_e dictAttachPref = ZSTD_dictDefaultAttach;
ZSTD_paramSwitch_e prefetchCDictTables = ZSTD_ps_auto;
ZSTD_ParamSwitch_e prefetchCDictTables = ZSTD_ps_auto;
metricAggregatePref_e metricAggregatePref = fastest;
for (int argNb = 1; argNb < argc ; argNb++) {

75
doc/zstd_manual.html
View File

@ -154,7 +154,7 @@ size_t ZSTD_compressBound(size_t srcSize); </b>/*!< maximum compressed size in w
it's recommended to provide @dstCapacity >= ZSTD_compressBound(srcSize)
as it eliminates one potential failure scenario,
aka not enough room in dst buffer to write the compressed frame.
Note : ZSTD_compressBound() itself can fail, if @srcSize > ZSTD_MAX_INPUT_SIZE .
Note : ZSTD_compressBound() itself can fail, if @srcSize >= ZSTD_MAX_INPUT_SIZE .
In which case, ZSTD_compressBound() will return an error code
which can be tested using ZSTD_isError().
@ -1084,7 +1084,7 @@ size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
*
* Note: This field is optional. ZSTD_generateSequences() will calculate the value of
* 'rep', but repeat offsets do not necessarily need to be calculated from an external
* sequence provider's perspective. For example, ZSTD_compressSequences() does not
* sequence provider perspective. For example, ZSTD_compressSequences() does not
* use this 'rep' field at all (as of now).
*/
} ZSTD_Sequence;
@ -1189,14 +1189,14 @@ size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict);
} ZSTD_literalCompressionMode_e;
</b></pre><BR>
<pre><b>typedef enum {
</b>/* Note: This enum controls features which are conditionally beneficial. Zstd typically will make a final<b>
* decision on whether or not to enable the feature (ZSTD_ps_auto), but setting the switch to ZSTD_ps_enable
* or ZSTD_ps_disable allow for a force enable/disable the feature.
</b>/* Note: This enum controls features which are conditionally beneficial.<b>
* Zstd can take a decision on whether or not to enable the feature (ZSTD_ps_auto),
* but setting the switch to ZSTD_ps_enable or ZSTD_ps_disable force enable/disable the feature.
*/
ZSTD_ps_auto = 0, </b>/* Let the library automatically determine whether the feature shall be enabled */<b>
ZSTD_ps_enable = 1, </b>/* Force-enable the feature */<b>
ZSTD_ps_disable = 2 </b>/* Do not use the feature */<b>
} ZSTD_paramSwitch_e;
} ZSTD_ParamSwitch_e;
</b></pre><BR>
<a name="Chapter15"></a><h2>Frame header and size functions</h2><pre></pre>
@ -1315,9 +1315,9 @@ ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr,
</p></pre><BR>
<pre><b>typedef enum {
ZSTD_sf_noBlockDelimiters = 0, </b>/* Representation of ZSTD_Sequence has no block delimiters, sequences only */<b>
ZSTD_sf_explicitBlockDelimiters = 1 </b>/* Representation of ZSTD_Sequence contains explicit block delimiters */<b>
} ZSTD_sequenceFormat_e;
ZSTD_sf_noBlockDelimiters = 0, </b>/* ZSTD_Sequence[] has no block delimiters, just sequences */<b>
ZSTD_sf_explicitBlockDelimiters = 1 </b>/* ZSTD_Sequence[] contains explicit block delimiters */<b>
} ZSTD_SequenceFormat_e;
</b></pre><BR>
<pre><b>ZSTDLIB_STATIC_API size_t ZSTD_sequenceBound(size_t srcSize);
</b><p> `srcSize` : size of the input buffer
@ -1331,7 +1331,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr,
<pre><b>ZSTD_DEPRECATED("For debugging only, will be replaced by ZSTD_extractSequences()")
ZSTDLIB_STATIC_API size_t
ZSTD_generateSequences(ZSTD_CCtx* zc,
ZSTD_Sequence* outSeqs, size_t outSeqsSize,
ZSTD_Sequence* outSeqs, size_t outSeqsCapacity,
const void* src, size_t srcSize);
</b><p> WARNING: This function is meant for debugging and informational purposes ONLY!
Its implementation is flawed, and it will be deleted in a future version.
@ -1345,7 +1345,7 @@ ZSTD_generateSequences(ZSTD_CCtx* zc,
@param zc The compression context to be used for ZSTD_compress2(). Set any
compression parameters you need on this context.
@param outSeqs The output sequences buffer of size @p outSeqsSize
@param outSeqsSize The size of the output sequences buffer.
@param outSeqsCapacity The size of the output sequences buffer.
ZSTD_sequenceBound(srcSize) is an upper bound on the number
of sequences that can be generated.
@param src The source buffer to generate sequences from of size @p srcSize.
@ -1376,13 +1376,14 @@ ZSTD_generateSequences(ZSTD_CCtx* zc,
</p></pre><BR>
<pre><b>ZSTDLIB_STATIC_API size_t
ZSTD_compressSequences( ZSTD_CCtx* cctx, void* dst, size_t dstSize,
const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
const void* src, size_t srcSize);
ZSTD_compressSequences(ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
const void* src, size_t srcSize);
</b><p> Compress an array of ZSTD_Sequence, associated with @src buffer, into dst.
@src contains the entire input (not just the literals).
If @srcSize > sum(sequence.length), the remaining bytes are considered all literals
If a dictionary is included, then the cctx should reference the dict. (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.)
If a dictionary is included, then the cctx should reference the dict (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.).
The entire source is compressed into a single frame.
The compression behavior changes based on cctx params. In particular:
@ -1391,11 +1392,17 @@ ZSTD_compressSequences( ZSTD_CCtx* cctx, void* dst, size_t dstSize,
the block size derived from the cctx, and sequences may be split. This is the default setting.
If ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, the array of ZSTD_Sequence is expected to contain
block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided.
valid block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided.
If ZSTD_c_validateSequences == 0, this function will blindly accept the sequences provided. Invalid sequences cause undefined
behavior. If ZSTD_c_validateSequences == 1, then if sequence is invalid (see doc/zstd_compression_format.md for
specifics regarding offset/matchlength requirements) then the function will bail out and return an error.
When ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, it's possible to decide generating repcodes
using the advanced parameter ZSTD_c_repcodeResolution. Repcodes will improve compression ratio, though the benefit
can vary greatly depending on Sequences. On the other hand, repcode resolution is an expensive operation.
By default, it's disabled at low (<10) compression levels, and enabled above the threshold (>=10).
ZSTD_c_repcodeResolution makes it possible to directly manage this processing in either direction.
If ZSTD_c_validateSequences == 0, this function blindly accepts the Sequences provided. Invalid Sequences cause undefined
behavior. If ZSTD_c_validateSequences == 1, then the function will detect invalid Sequences (see doc/zstd_compression_format.md for
specifics regarding offset/matchlength requirements) and then bail out and return an error.
In addition to the two adjustable experimental params, there are other important cctx params.
- ZSTD_c_minMatch MUST be set as less than or equal to the smallest match generated by the match finder. It has a minimum value of ZSTD_MINMATCH_MIN.
@ -1403,9 +1410,33 @@ ZSTD_compressSequences( ZSTD_CCtx* cctx, void* dst, size_t dstSize,
- ZSTD_c_windowLog affects offset validation: this function will return an error at higher debug levels if a provided offset
is larger than what the spec allows for a given window log and dictionary (if present). See: doc/zstd_compression_format.md
Note: Repcodes are, as of now, always re-calculated within this function, so ZSTD_Sequence::rep is unused.
Note 2: Once we integrate ability to ingest repcodes, the explicit block delims mode must respect those repcodes exactly,
and cannot emit an RLE block that disagrees with the repcode history
Note: Repcodes are, as of now, always re-calculated within this function, ZSTD_Sequence.rep is effectively unused.
Dev Note: Once ability to ingest repcodes become available, the explicit block delims mode must respect those repcodes exactly,
and cannot emit an RLE block that disagrees with the repcode history.
@return : final compressed size, or a ZSTD error code.
</p></pre><BR>
<pre><b>ZSTDLIB_STATIC_API size_t
ZSTD_compressSequencesAndLiterals(ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const ZSTD_Sequence* inSeqs, size_t nbSequences,
const void* literals, size_t litSize, size_t litCapacity,
size_t decompressedSize);
</b><p> This is a variant of ZSTD_compressSequences() which,
instead of receiving (src,srcSize) as input parameter, receives (literals,litSize),
aka all the literals, already extracted and laid out into a single continuous buffer.
This can be useful if the process generating the sequences also happens to generate the buffer of literals,
thus skipping an extraction + caching stage.
It's a speed optimization, useful when the right conditions are met,
but it also features the following limitations:
- Only supports explicit delimiter mode
- Currently does not support Sequences validation (so input Sequences are trusted)
- Not compatible with frame checksum, which must be disabled
- If any block is incompressible, will fail and return an error
- @litSize must be == sum of all @.litLength fields in @inSeqs. Any discrepancy will generate an error.
- the buffer @literals must have a size @litCapacity which is larger than @litSize by at least 8 bytes.
- @decompressedSize must be correct, and correspond to the sum of all Sequences. Any discrepancy will generate an error.
@return : final compressed size, or a ZSTD error code.
</p></pre><BR>

1
lib/common/error_private.c
View File

@ -40,6 +40,7 @@ const char* ERR_getErrorString(ERR_enum code)
case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported";
case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large";
case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small";
case PREFIX(cannotProduce_uncompressedBlock): return "This mode cannot generate an uncompressed block";
case PREFIX(stabilityCondition_notRespected): return "pledged buffer stability condition is not respected";
case PREFIX(dictionary_corrupted): return "Dictionary is corrupted";
case PREFIX(dictionary_wrong): return "Dictionary mismatch";

62
lib/common/zstd_internal.h
View File

@ -95,7 +95,7 @@ typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e;
#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */) /* for a non-null block */
#define MIN_LITERALS_FOR_4_STREAMS 6
typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e;
typedef enum { set_basic, set_rle, set_compressed, set_repeat } SymbolEncodingType_e;
#define LONGNBSEQ 0x7F00
@ -278,62 +278,6 @@ typedef enum {
/*-*******************************************
* Private declarations
*********************************************/
typedef struct seqDef_s {
U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
U16 litLength;
U16 mlBase; /* mlBase == matchLength - MINMATCH */
} seqDef;
/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */
typedef enum {
ZSTD_llt_none = 0, /* no longLengthType */
ZSTD_llt_literalLength = 1, /* represents a long literal */
ZSTD_llt_matchLength = 2 /* represents a long match */
} ZSTD_longLengthType_e;
typedef struct {
seqDef* sequencesStart;
seqDef* sequences; /* ptr to end of sequences */
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
size_t maxNbSeq;
size_t maxNbLit;
/* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength
* in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
* the existing value of the litLength or matchLength by 0x10000.
*/
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
} seqStore_t;
typedef struct {
U32 litLength;
U32 matchLength;
} ZSTD_sequenceLength;
/**
* Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences
* indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength.
*/
MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq)
{
ZSTD_sequenceLength seqLen;
seqLen.litLength = seq->litLength;
seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0x10000;
}
if (seqStore->longLengthType == ZSTD_llt_matchLength) {
seqLen.matchLength += 0x10000;
}
}
return seqLen;
}
/**
* Contains the compressed frame size and an upper-bound for the decompressed frame size.
@ -347,10 +291,6 @@ typedef struct {
unsigned long long decompressedBound;
} ZSTD_frameSizeInfo; /* decompress & legacy */
const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */
int ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
/* ZSTD_invalidateRepCodes() :
* ensures next compression will not use repcodes from previous block.
* Note : only works with regular variant;

2
lib/common/zstd_trace.h
View File

@ -65,7 +65,7 @@ typedef struct {
/**
* Non-zero if streaming (de)compression is used.
*/
unsigned streaming;
int streaming;
/**
* The dictionary ID.
*/

860
lib/compress/zstd_compress.c
View File

File diff suppressed because it is too large Load Diff

264
lib/compress/zstd_compress_internal.h
View File

@ -83,6 +83,70 @@ typedef struct {
ZSTD_fseCTables_t fse;
} ZSTD_entropyCTables_t;
/***********************************************
* Sequences *
***********************************************/
typedef struct SeqDef_s {
U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */
U16 litLength;
U16 mlBase; /* mlBase == matchLength - MINMATCH */
} SeqDef;
/* Controls whether seqStore has a single "long" litLength or matchLength. See SeqStore_t. */
typedef enum {
ZSTD_llt_none = 0, /* no longLengthType */
ZSTD_llt_literalLength = 1, /* represents a long literal */
ZSTD_llt_matchLength = 2 /* represents a long match */
} ZSTD_longLengthType_e;
typedef struct {
SeqDef* sequencesStart;
SeqDef* sequences; /* ptr to end of sequences */
BYTE* litStart;
BYTE* lit; /* ptr to end of literals */
BYTE* llCode;
BYTE* mlCode;
BYTE* ofCode;
size_t maxNbSeq;
size_t maxNbLit;
/* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength
* in the seqStore that has a value larger than U16 (if it exists). To do so, we increment
* the existing value of the litLength or matchLength by 0x10000.
*/
ZSTD_longLengthType_e longLengthType;
U32 longLengthPos; /* Index of the sequence to apply long length modification to */
} SeqStore_t;
typedef struct {
U32 litLength;
U32 matchLength;
} ZSTD_SequenceLength;
/**
* Returns the ZSTD_SequenceLength for the given sequences. It handles the decoding of long sequences
* indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength.
*/
MEM_STATIC ZSTD_SequenceLength ZSTD_getSequenceLength(SeqStore_t const* seqStore, SeqDef const* seq)
{
ZSTD_SequenceLength seqLen;
seqLen.litLength = seq->litLength;
seqLen.matchLength = seq->mlBase + MINMATCH;
if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) {
if (seqStore->longLengthType == ZSTD_llt_literalLength) {
seqLen.litLength += 0x10000;
}
if (seqStore->longLengthType == ZSTD_llt_matchLength) {
seqLen.matchLength += 0x10000;
}
}
return seqLen;
}
const SeqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */
int ZSTD_seqToCodes(const SeqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */
/***********************************************
* Entropy buffer statistics structs and funcs *
***********************************************/
@ -92,7 +156,7 @@ typedef struct {
* hufDesSize refers to the size of huffman tree description in bytes.
* This metadata is populated in ZSTD_buildBlockEntropyStats_literals() */
typedef struct {
symbolEncodingType_e hType;
SymbolEncodingType_e hType;
BYTE hufDesBuffer[ZSTD_MAX_HUF_HEADER_SIZE];
size_t hufDesSize;
} ZSTD_hufCTablesMetadata_t;
@ -103,9 +167,9 @@ typedef struct {
* fseTablesSize refers to the size of fse tables in bytes.
* This metadata is populated in ZSTD_buildBlockEntropyStats_sequences() */
typedef struct {
symbolEncodingType_e llType;
symbolEncodingType_e ofType;
symbolEncodingType_e mlType;
SymbolEncodingType_e llType;
SymbolEncodingType_e ofType;
SymbolEncodingType_e mlType;
BYTE fseTablesBuffer[ZSTD_MAX_FSE_HEADERS_SIZE];
size_t fseTablesSize;
size_t lastCountSize; /* This is to account for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */
@ -120,7 +184,7 @@ typedef struct {
* Builds entropy for the block.
* @return : 0 on success or error code */
size_t ZSTD_buildBlockEntropyStats(
const seqStore_t* seqStorePtr,
const SeqStore_t* seqStorePtr,
const ZSTD_entropyCTables_t* prevEntropy,
ZSTD_entropyCTables_t* nextEntropy,
const ZSTD_CCtx_params* cctxParams,
@ -149,15 +213,9 @@ typedef struct {
stopped. posInSequence <= seq[pos].litLength + seq[pos].matchLength */
size_t size; /* The number of sequences. <= capacity. */
size_t capacity; /* The capacity starting from `seq` pointer */
} rawSeqStore_t;
} RawSeqStore_t;
typedef struct {
U32 idx; /* Index in array of ZSTD_Sequence */
U32 posInSequence; /* Position within sequence at idx */
size_t posInSrc; /* Number of bytes given by sequences provided so far */
} ZSTD_sequencePosition;
UNUSED_ATTR static const rawSeqStore_t kNullRawSeqStore = {NULL, 0, 0, 0, 0};
UNUSED_ATTR static const RawSeqStore_t kNullRawSeqStore = {NULL, 0, 0, 0, 0};
typedef struct {
int price; /* price from beginning of segment to this position */
@ -189,7 +247,7 @@ typedef struct {
U32 offCodeSumBasePrice; /* to compare to log2(offreq) */
ZSTD_OptPrice_e priceType; /* prices can be determined dynamically, or follow a pre-defined cost structure */
const ZSTD_entropyCTables_t* symbolCosts; /* pre-calculated dictionary statistics */
ZSTD_paramSwitch_e literalCompressionMode;
ZSTD_ParamSwitch_e literalCompressionMode;
} optState_t;
typedef struct {
@ -211,11 +269,11 @@ typedef struct {
#define ZSTD_WINDOW_START_INDEX 2
typedef struct ZSTD_matchState_t ZSTD_matchState_t;
typedef struct ZSTD_MatchState_t ZSTD_MatchState_t;
#define ZSTD_ROW_HASH_CACHE_SIZE 8 /* Size of prefetching hash cache for row-based matchfinder */
struct ZSTD_matchState_t {
struct ZSTD_MatchState_t {
ZSTD_window_t window; /* State for window round buffer management */
U32 loadedDictEnd; /* index of end of dictionary, within context's referential.
* When loadedDictEnd != 0, a dictionary is in use, and still valid.
@ -237,15 +295,15 @@ struct ZSTD_matchState_t {
U32* hashTable3;
U32* chainTable;
U32 forceNonContiguous; /* Non-zero if we should force non-contiguous load for the next window update. */
int forceNonContiguous; /* Non-zero if we should force non-contiguous load for the next window update. */
int dedicatedDictSearch; /* Indicates whether this matchState is using the
* dedicated dictionary search structure.
*/
optState_t opt; /* optimal parser state */
const ZSTD_matchState_t* dictMatchState;
const ZSTD_MatchState_t* dictMatchState;
ZSTD_compressionParameters cParams;
const rawSeqStore_t* ldmSeqStore;
const RawSeqStore_t* ldmSeqStore;
/* Controls prefetching in some dictMatchState matchfinders.
* This behavior is controlled from the cctx ms.
@ -263,7 +321,7 @@ struct ZSTD_matchState_t {
typedef struct {
ZSTD_compressedBlockState_t* prevCBlock;
ZSTD_compressedBlockState_t* nextCBlock;
ZSTD_matchState_t matchState;
ZSTD_MatchState_t matchState;
} ZSTD_blockState_t;
typedef struct {
@ -290,7 +348,7 @@ typedef struct {
} ldmState_t;
typedef struct {
ZSTD_paramSwitch_e enableLdm; /* ZSTD_ps_enable to enable LDM. ZSTD_ps_auto by default */
ZSTD_ParamSwitch_e enableLdm; /* ZSTD_ps_enable to enable LDM. ZSTD_ps_auto by default */
U32 hashLog; /* Log size of hashTable */
U32 bucketSizeLog; /* Log bucket size for collision resolution, at most 8 */
U32 minMatchLength; /* Minimum match length */
@ -321,7 +379,7 @@ struct ZSTD_CCtx_params_s {
* There is no guarantee that hint is close to actual source size */
ZSTD_dictAttachPref_e attachDictPref;
ZSTD_paramSwitch_e literalCompressionMode;
ZSTD_ParamSwitch_e literalCompressionMode;
/* Multithreading: used to pass parameters to mtctx */
int nbWorkers;
@ -340,7 +398,7 @@ struct ZSTD_CCtx_params_s {
ZSTD_bufferMode_e outBufferMode;
/* Sequence compression API */
ZSTD_sequenceFormat_e blockDelimiters;
ZSTD_SequenceFormat_e blockDelimiters;
int validateSequences;
/* Block splitting
@ -353,14 +411,14 @@ struct ZSTD_CCtx_params_s {
* then levels are sorted in increasing cpu budget, from 2 (fastest) to 6 (slowest).
* Highest @preBlockSplitter_level combines well with @postBlockSplitter.
*/
ZSTD_paramSwitch_e postBlockSplitter;
ZSTD_ParamSwitch_e postBlockSplitter;
int preBlockSplitter_level;
/* Adjust the max block size*/
size_t maxBlockSize;
/* Param for deciding whether to use row-based matchfinder */
ZSTD_paramSwitch_e useRowMatchFinder;
ZSTD_ParamSwitch_e useRowMatchFinder;
/* Always load a dictionary in ext-dict mode (not prefix mode)? */
int deterministicRefPrefix;
@ -369,7 +427,7 @@ struct ZSTD_CCtx_params_s {
ZSTD_customMem customMem;
/* Controls prefetching in some dictMatchState matchfinders */
ZSTD_paramSwitch_e prefetchCDictTables;
ZSTD_ParamSwitch_e prefetchCDictTables;
/* Controls whether zstd will fall back to an internal matchfinder
* if the external matchfinder returns an error code. */
@ -382,7 +440,7 @@ struct ZSTD_CCtx_params_s {
ZSTD_sequenceProducer_F extSeqProdFunc;
/* Controls repcode search in external sequence parsing */
ZSTD_paramSwitch_e searchForExternalRepcodes;
ZSTD_ParamSwitch_e searchForExternalRepcodes;
}; /* typedef'd to ZSTD_CCtx_params within "zstd.h" */
#define COMPRESS_SEQUENCES_WORKSPACE_SIZE (sizeof(unsigned) * (MaxSeq + 2))
@ -405,11 +463,11 @@ typedef enum {
*/
#define ZSTD_MAX_NB_BLOCK_SPLITS 196
typedef struct {
seqStore_t fullSeqStoreChunk;
seqStore_t firstHalfSeqStore;
seqStore_t secondHalfSeqStore;
seqStore_t currSeqStore;
seqStore_t nextSeqStore;
SeqStore_t fullSeqStoreChunk;
SeqStore_t firstHalfSeqStore;
SeqStore_t secondHalfSeqStore;
SeqStore_t currSeqStore;
SeqStore_t nextSeqStore;
U32 partitions[ZSTD_MAX_NB_BLOCK_SPLITS];
ZSTD_entropyCTablesMetadata_t entropyMetadata;
@ -426,7 +484,7 @@ struct ZSTD_CCtx_s {
size_t dictContentSize;
ZSTD_cwksp workspace; /* manages buffer for dynamic allocations */
size_t blockSize;
size_t blockSizeMax;
unsigned long long pledgedSrcSizePlusOne; /* this way, 0 (default) == unknown */
unsigned long long consumedSrcSize;
unsigned long long producedCSize;
@ -438,11 +496,11 @@ struct ZSTD_CCtx_s {
int isFirstBlock;
int initialized;
seqStore_t seqStore; /* sequences storage ptrs */
SeqStore_t seqStore; /* sequences storage ptrs */
ldmState_t ldmState; /* long distance matching state */
rawSeq* ldmSequences; /* Storage for the ldm output sequences */
size_t maxNbLdmSequences;
rawSeqStore_t externSeqStore; /* Mutable reference to external sequences */
RawSeqStore_t externSeqStore; /* Mutable reference to external sequences */
ZSTD_blockState_t blockState;
void* tmpWorkspace; /* used as substitute of stack space - must be aligned for S64 type */
size_t tmpWkspSize;
@ -519,12 +577,12 @@ typedef enum {
* behavior of taking both the source size and the dict size into account
* when selecting and adjusting parameters.
*/
} ZSTD_cParamMode_e;
} ZSTD_CParamMode_e;
typedef size_t (*ZSTD_blockCompressor) (
ZSTD_matchState_t* bs, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
typedef size_t (*ZSTD_BlockCompressor_f) (
ZSTD_MatchState_t* bs, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_paramSwitch_e rowMatchfinderMode, ZSTD_dictMode_e dictMode);
ZSTD_BlockCompressor_f ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_ParamSwitch_e rowMatchfinderMode, ZSTD_dictMode_e dictMode);
MEM_STATIC U32 ZSTD_LLcode(U32 litLength)
@ -671,14 +729,55 @@ ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const iend, BYTE con
#define OFFBASE_TO_OFFSET(o) (assert(OFFBASE_IS_OFFSET(o)), (o) - ZSTD_REP_NUM)
#define OFFBASE_TO_REPCODE(o) (assert(OFFBASE_IS_REPCODE(o)), (o)) /* returns ID 1,2,3 */
/*! ZSTD_storeSeqOnly() :
* Store a sequence (litlen, litPtr, offBase and matchLength) into SeqStore_t.
* Literals themselves are not copied, but @litPtr is updated.
* @offBase : Users should employ macros REPCODE_TO_OFFBASE() and OFFSET_TO_OFFBASE().
* @matchLength : must be >= MINMATCH
*/
HINT_INLINE UNUSED_ATTR void
ZSTD_storeSeqOnly(SeqStore_t* seqStorePtr,
size_t litLength,
U32 offBase,
size_t matchLength)
{
assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq);
/* literal Length */
assert(litLength <= ZSTD_BLOCKSIZE_MAX);
if (UNLIKELY(litLength>0xFFFF)) {
assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */
seqStorePtr->longLengthType = ZSTD_llt_literalLength;
seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
}
seqStorePtr->sequences[0].litLength = (U16)litLength;
/* match offset */
seqStorePtr->sequences[0].offBase = offBase;
/* match Length */
assert(matchLength <= ZSTD_BLOCKSIZE_MAX);
assert(matchLength >= MINMATCH);
{ size_t const mlBase = matchLength - MINMATCH;
if (UNLIKELY(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++;
}
/*! ZSTD_storeSeq() :
* Store a sequence (litlen, litPtr, offBase and matchLength) into seqStore_t.
* Store a sequence (litlen, litPtr, offBase and matchLength) into SeqStore_t.
* @offBase : Users should employ macros REPCODE_TO_OFFBASE() and OFFSET_TO_OFFBASE().
* @matchLength : must be >= MINMATCH
* Allowed to over-read literals up to litLimit.
*/
HINT_INLINE UNUSED_ATTR void
ZSTD_storeSeq(seqStore_t* seqStorePtr,
ZSTD_storeSeq(SeqStore_t* seqStorePtr,
size_t litLength, const BYTE* literals, const BYTE* litLimit,
U32 offBase,
size_t matchLength)
@ -712,29 +811,7 @@ ZSTD_storeSeq(seqStore_t* seqStorePtr,
}
seqStorePtr->lit += litLength;
/* literal Length */
if (litLength>0xFFFF) {
assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */
seqStorePtr->longLengthType = ZSTD_llt_literalLength;
seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart);
}
seqStorePtr->sequences[0].litLength = (U16)litLength;
/* match offset */
seqStorePtr->sequences[0].offBase = offBase;
/* match Length */
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++;
ZSTD_storeSeqOnly(seqStorePtr, litLength, offBase, matchLength);
}
/* ZSTD_updateRep() :
@ -763,12 +840,12 @@ ZSTD_updateRep(U32 rep[ZSTD_REP_NUM], U32 const offBase, U32 const ll0)
typedef struct repcodes_s {
U32 rep[3];
} repcodes_t;
} Repcodes_t;
MEM_STATIC repcodes_t
MEM_STATIC Repcodes_t
ZSTD_newRep(U32 const rep[ZSTD_REP_NUM], U32 const offBase, U32 const ll0)
{
repcodes_t newReps;
Repcodes_t newReps;
ZSTD_memcpy(&newReps, rep, sizeof(newReps));
ZSTD_updateRep(newReps.rep, offBase, ll0);
return newReps;
@ -995,7 +1072,7 @@ MEM_STATIC U32 ZSTD_window_hasExtDict(ZSTD_window_t const window)
* Inspects the provided matchState and figures out what dictMode should be
* passed to the compressor.
*/
MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_matchState_t *ms)
MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_MatchState_t *ms)
{
return ZSTD_window_hasExtDict(ms->window) ?
ZSTD_extDict :
@ -1184,7 +1261,7 @@ ZSTD_window_enforceMaxDist(ZSTD_window_t* window,
const void* blockEnd,
U32 maxDist,
U32* loadedDictEndPtr,
const ZSTD_matchState_t** dictMatchStatePtr)
const ZSTD_MatchState_t** dictMatchStatePtr)
{
U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base);
U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0;
@ -1229,7 +1306,7 @@ ZSTD_checkDictValidity(const ZSTD_window_t* window,
const void* blockEnd,
U32 maxDist,
U32* loadedDictEndPtr,
const ZSTD_matchState_t** dictMatchStatePtr)
const ZSTD_MatchState_t** dictMatchStatePtr)
{
assert(loadedDictEndPtr != NULL);
assert(dictMatchStatePtr != NULL);
@ -1319,7 +1396,7 @@ U32 ZSTD_window_update(ZSTD_window_t* window,
/**
* Returns the lowest allowed match index. It may either be in the ext-dict or the prefix.
*/
MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog)
MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_MatchState_t* ms, U32 curr, unsigned windowLog)
{
U32 const maxDistance = 1U << windowLog;
U32 const lowestValid = ms->window.lowLimit;
@ -1336,7 +1413,7 @@ MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, u
/**
* Returns the lowest allowed match index in the prefix.
*/
MEM_STATIC U32 ZSTD_getLowestPrefixIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog)
MEM_STATIC U32 ZSTD_getLowestPrefixIndex(const ZSTD_MatchState_t* ms, U32 curr, unsigned windowLog)
{
U32 const maxDistance = 1U << windowLog;
U32 const lowestValid = ms->window.dictLimit;
@ -1445,6 +1522,18 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace,
void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs);
typedef struct {
U32 idx; /* Index in array of ZSTD_Sequence */
U32 posInSequence; /* Position within sequence at idx */
size_t posInSrc; /* Number of bytes given by sequences provided so far */
} ZSTD_SequencePosition;
/* for benchmark */
size_t ZSTD_convertBlockSequences(ZSTD_CCtx* cctx,
const ZSTD_Sequence* const inSeqs, size_t nbSequences,
int const repcodeResolution);
/* ==============================================================
* Private declarations
* These prototypes shall only be called from within lib/compress
@ -1456,7 +1545,7 @@ void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs);
* Note: srcSizeHint == 0 means 0!
*/
ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams(
const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode);
const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_CParamMode_e mode);
/*! ZSTD_initCStream_internal() :
* Private use only. Init streaming operation.
@ -1468,7 +1557,7 @@ size_t ZSTD_initCStream_internal(ZSTD_CStream* zcs,
const ZSTD_CDict* cdict,
const ZSTD_CCtx_params* params, unsigned long long pledgedSrcSize);
void ZSTD_resetSeqStore(seqStore_t* ssPtr);
void ZSTD_resetSeqStore(SeqStore_t* ssPtr);
/*! ZSTD_getCParamsFromCDict() :
* as the name implies */
@ -1521,33 +1610,6 @@ U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat);
*/
void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize);
/* 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.
* Note that the block delimiter must include the last literals of the block.
*/
size_t
ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx,
ZSTD_sequencePosition* seqPos,
const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
const void* src, size_t blockSize, ZSTD_paramSwitch_e externalRepSearch);
/* Returns the number of bytes to move the current read position back by.
* Only non-zero if we ended up splitting a sequence.
* Otherwise, it may return a ZSTD error if something went wrong.
*
* This function will attempt to scan through blockSize bytes
* represented by the sequences in @inSeqs,
* storing any (partial) sequences.
*
* Occasionally, we may want to change the actual number of bytes we consumed from inSeqs to
* 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.
*/
size_t
ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos,
const ZSTD_Sequence* const inSeqs, size_t inSeqsSize,
const void* src, size_t blockSize, ZSTD_paramSwitch_e externalRepSearch);
/* Returns 1 if an external sequence producer is registered, otherwise returns 0. */
MEM_STATIC int ZSTD_hasExtSeqProd(const ZSTD_CCtx_params* params) {
return params->extSeqProdFunc != NULL;

2
lib/compress/zstd_compress_literals.c
View File

@ -140,7 +140,7 @@ size_t ZSTD_compressLiterals (
size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB);
BYTE* const ostart = (BYTE*)dst;
U32 singleStream = srcSize < 256;
symbolEncodingType_e hType = set_compressed;
SymbolEncodingType_e hType = set_compressed;
size_t cLitSize;
DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i, srcSize=%u, dstCapacity=%zu)",

14
lib/compress/zstd_compress_sequences.c
View File

@ -153,13 +153,13 @@ size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog,
return cost >> 8;
}
symbolEncodingType_e
SymbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_DefaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy)
{
ZSTD_STATIC_ASSERT(ZSTD_defaultDisallowed == 0 && ZSTD_defaultAllowed != 0);
@ -241,7 +241,7 @@ typedef struct {
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
FSE_CTable* nextCTable, U32 FSELog, SymbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
@ -293,7 +293,7 @@ ZSTD_encodeSequences_body(
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
SeqDef const* sequences, size_t nbSeq, int longOffsets)
{
BIT_CStream_t blockStream;
FSE_CState_t stateMatchLength;
@ -387,7 +387,7 @@ ZSTD_encodeSequences_default(
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
SeqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
@ -405,7 +405,7 @@ ZSTD_encodeSequences_bmi2(
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets)
SeqDef const* sequences, size_t nbSeq, int longOffsets)
{
return ZSTD_encodeSequences_body(dst, dstCapacity,
CTable_MatchLength, mlCodeTable,
@ -421,7 +421,7 @@ size_t ZSTD_encodeSequences(
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2)
SeqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2)
{
DEBUGLOG(5, "ZSTD_encodeSequences: dstCapacity = %u", (unsigned)dstCapacity);
#if DYNAMIC_BMI2

13
lib/compress/zstd_compress_sequences.h
View File

@ -11,26 +11,27 @@
#ifndef ZSTD_COMPRESS_SEQUENCES_H
#define ZSTD_COMPRESS_SEQUENCES_H
#include "zstd_compress_internal.h" /* SeqDef */
#include "../common/fse.h" /* FSE_repeat, FSE_CTable */
#include "../common/zstd_internal.h" /* symbolEncodingType_e, ZSTD_strategy */
#include "../common/zstd_internal.h" /* SymbolEncodingType_e, ZSTD_strategy */
typedef enum {
ZSTD_defaultDisallowed = 0,
ZSTD_defaultAllowed = 1
} ZSTD_defaultPolicy_e;
} ZSTD_DefaultPolicy_e;
symbolEncodingType_e
SymbolEncodingType_e
ZSTD_selectEncodingType(
FSE_repeat* repeatMode, unsigned const* count, unsigned const max,
size_t const mostFrequent, size_t nbSeq, unsigned const FSELog,
FSE_CTable const* prevCTable,
short const* defaultNorm, U32 defaultNormLog,
ZSTD_defaultPolicy_e const isDefaultAllowed,
ZSTD_DefaultPolicy_e const isDefaultAllowed,
ZSTD_strategy const strategy);
size_t
ZSTD_buildCTable(void* dst, size_t dstCapacity,
FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type,
FSE_CTable* nextCTable, U32 FSELog, SymbolEncodingType_e type,
unsigned* count, U32 max,
const BYTE* codeTable, size_t nbSeq,
const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax,
@ -42,7 +43,7 @@ size_t ZSTD_encodeSequences(
FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable,
FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable,
FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable,
seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2);
SeqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2);
size_t ZSTD_fseBitCost(
FSE_CTable const* ctable,

30
lib/compress/zstd_compress_superblock.c
View File

@ -51,7 +51,7 @@ ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable,
BYTE* const oend = ostart + dstSize;
BYTE* op = ostart + lhSize;
U32 const singleStream = lhSize == 3;
symbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat;
SymbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat;
size_t cLitSize = 0;
DEBUGLOG(5, "ZSTD_compressSubBlock_literal (litSize=%zu, lhSize=%zu, writeEntropy=%d)", litSize, lhSize, writeEntropy);
@ -126,15 +126,15 @@ ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable,
}
static size_t
ZSTD_seqDecompressedSize(seqStore_t const* seqStore,
const seqDef* sequences, size_t nbSeqs,
ZSTD_seqDecompressedSize(SeqStore_t const* seqStore,
const SeqDef* sequences, size_t nbSeqs,
size_t litSize, int lastSubBlock)
{
size_t matchLengthSum = 0;
size_t litLengthSum = 0;
size_t n;
for (n=0; n<nbSeqs; n++) {
const ZSTD_sequenceLength seqLen = ZSTD_getSequenceLength(seqStore, sequences+n);
const ZSTD_SequenceLength seqLen = ZSTD_getSequenceLength(seqStore, sequences+n);
litLengthSum += seqLen.litLength;
matchLengthSum += seqLen.matchLength;
}
@ -162,7 +162,7 @@ ZSTD_seqDecompressedSize(seqStore_t const* seqStore,
static size_t
ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables,
const ZSTD_fseCTablesMetadata_t* fseMetadata,
const seqDef* sequences, size_t nbSeq,
const SeqDef* sequences, size_t nbSeq,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
void* dst, size_t dstCapacity,
@ -262,7 +262,7 @@ ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables,
* Or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock(const ZSTD_entropyCTables_t* entropy,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
const seqDef* sequences, size_t nbSeq,
const SeqDef* sequences, size_t nbSeq,
const BYTE* literals, size_t litSize,
const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode,
const ZSTD_CCtx_params* cctxParams,
@ -327,7 +327,7 @@ static size_t ZSTD_estimateSubBlockSize_literal(const BYTE* literals, size_t lit
return 0;
}
static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type,
static size_t ZSTD_estimateSubBlockSize_symbolType(SymbolEncodingType_e type,
const BYTE* codeTable, unsigned maxCode,
size_t nbSeq, const FSE_CTable* fseCTable,
const U8* additionalBits,
@ -426,7 +426,7 @@ static int ZSTD_needSequenceEntropyTables(ZSTD_fseCTablesMetadata_t const* fseMe
return 0;
}
static size_t countLiterals(seqStore_t const* seqStore, const seqDef* sp, size_t seqCount)
static size_t countLiterals(SeqStore_t const* seqStore, const SeqDef* sp, size_t seqCount)
{
size_t n, total = 0;
assert(sp != NULL);
@ -439,7 +439,7 @@ static size_t countLiterals(seqStore_t const* seqStore, const seqDef* sp, size_t
#define BYTESCALE 256
static size_t sizeBlockSequences(const seqDef* sp, size_t nbSeqs,
static size_t sizeBlockSequences(const SeqDef* sp, size_t nbSeqs,
size_t targetBudget, size_t avgLitCost, size_t avgSeqCost,
int firstSubBlock)
{
@ -476,7 +476,7 @@ static size_t sizeBlockSequences(const seqDef* sp, size_t nbSeqs,
* Sub-blocks are all compressed, except the last one when beneficial.
* @return : compressed size of the super block (which features multiple ZSTD blocks)
* or 0 if it failed to compress. */
static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
static size_t ZSTD_compressSubBlock_multi(const SeqStore_t* seqStorePtr,
const ZSTD_compressedBlockState_t* prevCBlock,
ZSTD_compressedBlockState_t* nextCBlock,
const ZSTD_entropyCTablesMetadata_t* entropyMetadata,
@ -486,9 +486,9 @@ static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
const int bmi2, U32 lastBlock,
void* workspace, size_t wkspSize)
{
const seqDef* const sstart = seqStorePtr->sequencesStart;
const seqDef* const send = seqStorePtr->sequences;
const seqDef* sp = sstart; /* tracks progresses within seqStorePtr->sequences */
const SeqDef* const sstart = seqStorePtr->sequencesStart;
const SeqDef* const send = seqStorePtr->sequences;
const SeqDef* sp = sstart; /* tracks progresses within seqStorePtr->sequences */
size_t const nbSeqs = (size_t)(send - sstart);
const BYTE* const lstart = seqStorePtr->litStart;
const BYTE* const lend = seqStorePtr->lit;
@ -647,8 +647,8 @@ static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr,
op += cSize;
/* We have to regenerate the repcodes because we've skipped some sequences */
if (sp < send) {
const seqDef* seq;
repcodes_t rep;
const SeqDef* seq;
Repcodes_t rep;
ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep));
for (seq = sstart; seq < sp; ++seq) {
ZSTD_updateRep(rep.rep, seq->offBase, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0);

22
lib/compress/zstd_double_fast.c
View File

@ -15,7 +15,7 @@
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_fillDoubleHashTableForCDict(ZSTD_matchState_t* ms,
void ZSTD_fillDoubleHashTableForCDict(ZSTD_MatchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
@ -53,7 +53,7 @@ void ZSTD_fillDoubleHashTableForCDict(ZSTD_matchState_t* ms,
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
void ZSTD_fillDoubleHashTableForCCtx(ZSTD_MatchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
@ -87,7 +87,7 @@ void ZSTD_fillDoubleHashTableForCCtx(ZSTD_matchState_t* ms,
} }
}
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
void ZSTD_fillDoubleHashTable(ZSTD_MatchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
@ -103,7 +103,7 @@ void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_doubleFast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls /* template */)
{
ZSTD_compressionParameters const* cParams = &ms->cParams;
@ -326,7 +326,7 @@ _match_stored:
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
@ -347,7 +347,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic(
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_MatchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams;
const U32* const dictHashLong = dms->hashTable;
const U32* const dictHashSmall = dms->chainTable;
@ -548,7 +548,7 @@ _match_stored:
#define ZSTD_GEN_DFAST_FN(dictMode, mls) \
static size_t ZSTD_compressBlock_doubleFast_##dictMode##_##mls( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_doubleFast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls); \
@ -566,7 +566,7 @@ ZSTD_GEN_DFAST_FN(dictMatchState, 7)
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
@ -586,7 +586,7 @@ size_t ZSTD_compressBlock_doubleFast(
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
const U32 mls = ms->cParams.minMatch;
@ -608,7 +608,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState(
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_doubleFast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls /* template */)
{
@ -757,7 +757,7 @@ ZSTD_GEN_DFAST_FN(extDict, 6)
ZSTD_GEN_DFAST_FN(extDict, 7)
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;

8
lib/compress/zstd_double_fast.h
View File

@ -20,18 +20,18 @@ extern "C" {
#ifndef ZSTD_EXCLUDE_DFAST_BLOCK_COMPRESSOR
void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms,
void ZSTD_fillDoubleHashTable(ZSTD_MatchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_doubleFast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_doubleFast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_DOUBLEFAST ZSTD_compressBlock_doubleFast

22
lib/compress/zstd_fast.c
View File

@ -13,7 +13,7 @@
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_fillHashTableForCDict(ZSTD_matchState_t* ms,
void ZSTD_fillHashTableForCDict(ZSTD_MatchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
@ -50,7 +50,7 @@ void ZSTD_fillHashTableForCDict(ZSTD_matchState_t* ms,
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_fillHashTableForCCtx(ZSTD_matchState_t* ms,
void ZSTD_fillHashTableForCCtx(ZSTD_MatchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm)
{
@ -84,7 +84,7 @@ void ZSTD_fillHashTableForCCtx(ZSTD_matchState_t* ms,
} } } }
}
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
void ZSTD_fillHashTable(ZSTD_MatchState_t* ms,
const void* const end,
ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp)
@ -190,7 +190,7 @@ ZSTD_match4Found_branch(const BYTE* currentPtr, const BYTE* matchAddress, U32 ma
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_fast_noDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize,
U32 const mls, int useCmov)
{
@ -424,7 +424,7 @@ _match: /* Requires: ip0, match0, offcode */
#define ZSTD_GEN_FAST_FN(dictMode, mml, cmov) \
static size_t ZSTD_compressBlock_fast_##dictMode##_##mml##_##cmov( \
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \
void const* src, size_t srcSize) \
{ \
return ZSTD_compressBlock_fast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mml, cmov); \
@ -441,7 +441,7 @@ ZSTD_GEN_FAST_FN(noDict, 6, 0)
ZSTD_GEN_FAST_FN(noDict, 7, 0)
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mml = ms->cParams.minMatch;
@ -481,7 +481,7 @@ size_t ZSTD_compressBlock_fast(
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_fast_dictMatchState_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
@ -500,7 +500,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic(
const BYTE* const ilimit = iend - HASH_READ_SIZE;
U32 offset_1=rep[0], offset_2=rep[1];
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_MatchState_t* const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dictCParams = &dms->cParams ;
const U32* const dictHashTable = dms->hashTable;
const U32 dictStartIndex = dms->window.dictLimit;
@ -684,7 +684,7 @@ ZSTD_GEN_FAST_FN(dictMatchState, 6, 0)
ZSTD_GEN_FAST_FN(dictMatchState, 7, 0)
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;
@ -707,7 +707,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState(
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_fast_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize, U32 const mls, U32 const hasStep)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
@ -965,7 +965,7 @@ ZSTD_GEN_FAST_FN(extDict, 6, 0)
ZSTD_GEN_FAST_FN(extDict, 7, 0)
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
U32 const mls = ms->cParams.minMatch;

8
lib/compress/zstd_fast.h
View File

@ -18,17 +18,17 @@ extern "C" {
#include "../common/mem.h" /* U32 */
#include "zstd_compress_internal.h"
void ZSTD_fillHashTable(ZSTD_matchState_t* ms,
void ZSTD_fillHashTable(ZSTD_MatchState_t* ms,
void const* end, ZSTD_dictTableLoadMethod_e dtlm,
ZSTD_tableFillPurpose_e tfp);
size_t ZSTD_compressBlock_fast(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_fast_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#if defined (__cplusplus)

106
lib/compress/zstd_lazy.c
View File

@ -26,7 +26,7 @@
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_updateDUBT(ZSTD_matchState_t* ms,
void ZSTD_updateDUBT(ZSTD_MatchState_t* ms,
const BYTE* ip, const BYTE* iend,
U32 mls)
{
@ -71,7 +71,7 @@ void ZSTD_updateDUBT(ZSTD_matchState_t* ms,
* doesn't fail */
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_insertDUBT1(const ZSTD_matchState_t* ms,
void ZSTD_insertDUBT1(const ZSTD_MatchState_t* ms,
U32 curr, const BYTE* inputEnd,
U32 nbCompares, U32 btLow,
const ZSTD_dictMode_e dictMode)
@ -162,7 +162,7 @@ void ZSTD_insertDUBT1(const ZSTD_matchState_t* ms,
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_DUBT_findBetterDictMatch (
const ZSTD_matchState_t* ms,
const ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
size_t* offsetPtr,
size_t bestLength,
@ -170,7 +170,7 @@ size_t ZSTD_DUBT_findBetterDictMatch (
U32 const mls,
const ZSTD_dictMode_e dictMode)
{
const ZSTD_matchState_t * const dms = ms->dictMatchState;
const ZSTD_MatchState_t * const dms = ms->dictMatchState;
const ZSTD_compressionParameters* const dmsCParams = &dms->cParams;
const U32 * const dictHashTable = dms->hashTable;
U32 const hashLog = dmsCParams->hashLog;
@ -240,7 +240,7 @@ size_t ZSTD_DUBT_findBetterDictMatch (
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms,
size_t ZSTD_DUBT_findBestMatch(ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
size_t* offBasePtr,
U32 const mls,
@ -392,7 +392,7 @@ size_t ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms,
/** ZSTD_BtFindBestMatch() : Tree updater, providing best match */
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_BtFindBestMatch( ZSTD_matchState_t* ms,
size_t ZSTD_BtFindBestMatch( ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iLimit,
size_t* offBasePtr,
const U32 mls /* template */,
@ -408,7 +408,7 @@ size_t ZSTD_BtFindBestMatch( ZSTD_matchState_t* ms,
* Dedicated dict search
***********************************/
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip)
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_MatchState_t* ms, const BYTE* const ip)
{
const BYTE* const base = ms->window.base;
U32 const target = (U32)(ip - base);
@ -527,7 +527,7 @@ void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const B
*/
FORCE_INLINE_TEMPLATE
size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nbAttempts,
const ZSTD_matchState_t* const dms,
const ZSTD_MatchState_t* const dms,
const BYTE* const ip, const BYTE* const iLimit,
const BYTE* const prefixStart, const U32 curr,
const U32 dictLimit, const size_t ddsIdx) {
@ -630,7 +630,7 @@ size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nb
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
U32 ZSTD_insertAndFindFirstIndex_internal(
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
const ZSTD_compressionParameters* const cParams,
const BYTE* ip, U32 const mls, U32 const lazySkipping)
{
@ -656,7 +656,7 @@ U32 ZSTD_insertAndFindFirstIndex_internal(
return hashTable[ZSTD_hashPtr(ip, hashLog, mls)];
}
U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) {
U32 ZSTD_insertAndFindFirstIndex(ZSTD_MatchState_t* ms, const BYTE* ip) {
const ZSTD_compressionParameters* const cParams = &ms->cParams;
return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, ms->cParams.minMatch, /* lazySkipping*/ 0);
}
@ -665,7 +665,7 @@ U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) {
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_HcFindBestMatch(
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iLimit,
size_t* offsetPtr,
const U32 mls, const ZSTD_dictMode_e dictMode)
@ -689,7 +689,7 @@ size_t ZSTD_HcFindBestMatch(
U32 nbAttempts = 1U << cParams->searchLog;
size_t ml=4-1;
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_MatchState_t* const dms = ms->dictMatchState;
const U32 ddsHashLog = dictMode == ZSTD_dedicatedDictSearch
? dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG : 0;
const size_t ddsIdx = dictMode == ZSTD_dedicatedDictSearch
@ -834,7 +834,7 @@ FORCE_INLINE_TEMPLATE void ZSTD_row_prefetch(U32 const* hashTable, BYTE const* t
*/
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_row_fillHashCache(ZSTD_matchState_t* ms, const BYTE* base,
void ZSTD_row_fillHashCache(ZSTD_MatchState_t* ms, const BYTE* base,
U32 const rowLog, U32 const mls,
U32 idx, const BYTE* const iLimit)
{
@ -882,7 +882,7 @@ U32 ZSTD_row_nextCachedHash(U32* cache, U32 const* hashTable,
*/
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_row_update_internalImpl(ZSTD_matchState_t* ms,
void ZSTD_row_update_internalImpl(ZSTD_MatchState_t* ms,
U32 updateStartIdx, U32 const updateEndIdx,
U32 const mls, U32 const rowLog,
U32 const rowMask, U32 const useCache)
@ -913,7 +913,7 @@ void ZSTD_row_update_internalImpl(ZSTD_matchState_t* ms,
*/
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_row_update_internal(ZSTD_matchState_t* ms, const BYTE* ip,
void ZSTD_row_update_internal(ZSTD_MatchState_t* ms, const BYTE* ip,
U32 const mls, U32 const rowLog,
U32 const rowMask, U32 const useCache)
{
@ -946,7 +946,7 @@ void ZSTD_row_update_internal(ZSTD_matchState_t* ms, const BYTE* ip,
* External wrapper for ZSTD_row_update_internal(). Used for filling the hashtable during dictionary
* processing.
*/
void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip) {
void ZSTD_row_update(ZSTD_MatchState_t* const ms, const BYTE* ip) {
const U32 rowLog = BOUNDED(4, ms->cParams.searchLog, 6);
const U32 rowMask = (1u << rowLog) - 1;
const U32 mls = MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */);
@ -1139,7 +1139,7 @@ ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 headGr
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_RowFindBestMatch(
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iLimit,
size_t* offsetPtr,
const U32 mls, const ZSTD_dictMode_e dictMode,
@ -1171,7 +1171,7 @@ size_t ZSTD_RowFindBestMatch(
U32 hash;
/* DMS/DDS variables that may be referenced laster */
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_MatchState_t* const dms = ms->dictMatchState;
/* Initialize the following variables to satisfy static analyzer */
size_t ddsIdx = 0;
@ -1340,7 +1340,7 @@ size_t ZSTD_RowFindBestMatch(
* ZSTD_searchMax() dispatches to the correct implementation function.
*
* TODO: The start of the search function involves loading and calculating a
* bunch of constants from the ZSTD_matchState_t. These computations could be
* bunch of constants from the ZSTD_MatchState_t. These computations could be
* done in an initialization function, and saved somewhere in the match state.
* Then we could pass a pointer to the saved state instead of the match state,
* and avoid duplicate computations.
@ -1364,7 +1364,7 @@ size_t ZSTD_RowFindBestMatch(
#define GEN_ZSTD_BT_SEARCH_FN(dictMode, mls) \
ZSTD_SEARCH_FN_ATTRS size_t ZSTD_BT_SEARCH_FN(dictMode, mls)( \
ZSTD_matchState_t* ms, \
ZSTD_MatchState_t* ms, \
const BYTE* ip, const BYTE* const iLimit, \
size_t* offBasePtr) \
{ \
@ -1374,7 +1374,7 @@ size_t ZSTD_RowFindBestMatch(
#define GEN_ZSTD_HC_SEARCH_FN(dictMode, mls) \
ZSTD_SEARCH_FN_ATTRS size_t ZSTD_HC_SEARCH_FN(dictMode, mls)( \
ZSTD_matchState_t* ms, \
ZSTD_MatchState_t* ms, \
const BYTE* ip, const BYTE* const iLimit, \
size_t* offsetPtr) \
{ \
@ -1384,7 +1384,7 @@ size_t ZSTD_RowFindBestMatch(
#define GEN_ZSTD_ROW_SEARCH_FN(dictMode, mls, rowLog) \
ZSTD_SEARCH_FN_ATTRS size_t ZSTD_ROW_SEARCH_FN(dictMode, mls, rowLog)( \
ZSTD_matchState_t* ms, \
ZSTD_MatchState_t* ms, \
const BYTE* ip, const BYTE* const iLimit, \
size_t* offsetPtr) \
{ \
@ -1485,7 +1485,7 @@ typedef enum { search_hashChain=0, search_binaryTree=1, search_rowHash=2 } searc
* If a match is found its offset is stored in @p offsetPtr.
*/
FORCE_INLINE_TEMPLATE size_t ZSTD_searchMax(
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
const BYTE* ip,
const BYTE* iend,
size_t* offsetPtr,
@ -1514,7 +1514,7 @@ FORCE_INLINE_TEMPLATE size_t ZSTD_searchMax(
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_lazy_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore,
ZSTD_MatchState_t* ms, SeqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize,
const searchMethod_e searchMethod, const U32 depth,
@ -1537,7 +1537,7 @@ size_t ZSTD_compressBlock_lazy_generic(
const int isDMS = dictMode == ZSTD_dictMatchState;
const int isDDS = dictMode == ZSTD_dedicatedDictSearch;
const int isDxS = isDMS || isDDS;
const ZSTD_matchState_t* const dms = ms->dictMatchState;
const ZSTD_MatchState_t* const dms = ms->dictMatchState;
const U32 dictLowestIndex = isDxS ? dms->window.dictLimit : 0;
const BYTE* const dictBase = isDxS ? dms->window.base : NULL;
const BYTE* const dictLowest = isDxS ? dictBase + dictLowestIndex : NULL;
@ -1782,42 +1782,42 @@ _storeSequence:
#ifndef ZSTD_EXCLUDE_GREEDY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_noDict);
}
size_t ZSTD_compressBlock_greedy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dedicatedDictSearch);
}
size_t ZSTD_compressBlock_greedy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_noDict);
}
size_t ZSTD_compressBlock_greedy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dedicatedDictSearch);
@ -1826,42 +1826,42 @@ size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
#ifndef ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dedicatedDictSearch);
}
size_t ZSTD_compressBlock_lazy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dedicatedDictSearch);
@ -1870,42 +1870,42 @@ size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
#ifndef ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dedicatedDictSearch);
}
size_t ZSTD_compressBlock_lazy2_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_noDict);
}
size_t ZSTD_compressBlock_lazy2_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dedicatedDictSearch);
@ -1914,14 +1914,14 @@ size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
#ifndef ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2, ZSTD_noDict);
}
size_t ZSTD_compressBlock_btlazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2, ZSTD_dictMatchState);
@ -1935,7 +1935,7 @@ size_t ZSTD_compressBlock_btlazy2_dictMatchState(
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_compressBlock_lazy_extDict_generic(
ZSTD_matchState_t* ms, seqStore_t* seqStore,
ZSTD_MatchState_t* ms, SeqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize,
const searchMethod_e searchMethod, const U32 depth)
@ -2139,14 +2139,14 @@ _storeSequence:
#ifndef ZSTD_EXCLUDE_GREEDY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0);
}
size_t ZSTD_compressBlock_greedy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0);
@ -2155,7 +2155,7 @@ size_t ZSTD_compressBlock_greedy_extDict_row(
#ifndef ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
@ -2163,7 +2163,7 @@ size_t ZSTD_compressBlock_lazy_extDict(
}
size_t ZSTD_compressBlock_lazy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
@ -2173,7 +2173,7 @@ size_t ZSTD_compressBlock_lazy_extDict_row(
#ifndef ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
@ -2181,7 +2181,7 @@ size_t ZSTD_compressBlock_lazy2_extDict(
}
size_t ZSTD_compressBlock_lazy2_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{
return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2);
@ -2190,7 +2190,7 @@ size_t ZSTD_compressBlock_lazy2_extDict_row(
#ifndef ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize)
{

60
lib/compress/zstd_lazy.h
View File

@ -31,38 +31,38 @@ extern "C" {
|| !defined(ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR) \
|| !defined(ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR) \
|| !defined(ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR)
U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip);
void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip);
U32 ZSTD_insertAndFindFirstIndex(ZSTD_MatchState_t* ms, const BYTE* ip);
void ZSTD_row_update(ZSTD_MatchState_t* const ms, const BYTE* ip);
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip);
void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_MatchState_t* ms, const BYTE* const ip);
void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */
#endif
#ifndef ZSTD_EXCLUDE_GREEDY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_greedy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_greedy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_GREEDY ZSTD_compressBlock_greedy
@ -86,28 +86,28 @@ size_t ZSTD_compressBlock_greedy_extDict_row(
#ifndef ZSTD_EXCLUDE_LAZY_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_LAZY ZSTD_compressBlock_lazy
@ -131,28 +131,28 @@ size_t ZSTD_compressBlock_lazy_extDict_row(
#ifndef ZSTD_EXCLUDE_LAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_lazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dictMatchState_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_lazy2_extDict_row(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_LAZY2 ZSTD_compressBlock_lazy2
@ -176,13 +176,13 @@ size_t ZSTD_compressBlock_lazy2_extDict_row(
#ifndef ZSTD_EXCLUDE_BTLAZY2_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btlazy2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btlazy2_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_BTLAZY2 ZSTD_compressBlock_btlazy2

22
lib/compress/zstd_ldm.c
View File

@ -234,7 +234,7 @@ static size_t ZSTD_ldm_countBackwardsMatch_2segments(
*
* The tables for the other strategies are filled within their
* block compressors. */
static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms,
static size_t ZSTD_ldm_fillFastTables(ZSTD_MatchState_t* ms,
void const* end)
{
const BYTE* const iend = (const BYTE*)end;
@ -314,7 +314,7 @@ void ZSTD_ldm_fillHashTable(
* Sets cctx->nextToUpdate to a position corresponding closer to anchor
* if it is far way
* (after a long match, only update tables a limited amount). */
static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor)
static void ZSTD_ldm_limitTableUpdate(ZSTD_MatchState_t* ms, const BYTE* anchor)
{
U32 const curr = (U32)(anchor - ms->window.base);
if (curr > ms->nextToUpdate + 1024) {
@ -326,7 +326,7 @@ static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor)
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t ZSTD_ldm_generateSequences_internal(
ldmState_t* ldmState, rawSeqStore_t* rawSeqStore,
ldmState_t* ldmState, RawSeqStore_t* rawSeqStore,
ldmParams_t const* params, void const* src, size_t srcSize)
{
/* LDM parameters */
@ -510,7 +510,7 @@ static void ZSTD_ldm_reduceTable(ldmEntry_t* const table, U32 const size,
}
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldmState, rawSeqStore_t* sequences,
ldmState_t* ldmState, RawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize)
{
U32 const maxDist = 1U << params->windowLog;
@ -587,7 +587,7 @@ size_t ZSTD_ldm_generateSequences(
}
void
ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch)
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;
@ -623,7 +623,7 @@ ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const min
* Returns the current sequence to handle, or if the rest of the block should
* be literals, it returns a sequence with offset == 0.
*/
static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore,
static rawSeq maybeSplitSequence(RawSeqStore_t* rawSeqStore,
U32 const remaining, U32 const minMatch)
{
rawSeq sequence = rawSeqStore->seq[rawSeqStore->pos];
@ -647,7 +647,7 @@ static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore,
return sequence;
}
void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
void ZSTD_ldm_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];
@ -664,14 +664,14 @@ void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) {
}
}
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_paramSwitch_e useRowMatchFinder,
size_t ZSTD_ldm_blockCompress(RawSeqStore_t* rawSeqStore,
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_ParamSwitch_e useRowMatchFinder,
void const* src, size_t srcSize)
{
const ZSTD_compressionParameters* const cParams = &ms->cParams;
unsigned const minMatch = cParams->minMatch;
ZSTD_blockCompressor const blockCompressor =
ZSTD_BlockCompressor_f const blockCompressor =
ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms));
/* Input bounds */
BYTE const* const istart = (BYTE const*)src;

12
lib/compress/zstd_ldm.h
View File

@ -43,7 +43,7 @@ void ZSTD_ldm_fillHashTable(
* sequences.
*/
size_t ZSTD_ldm_generateSequences(
ldmState_t* ldms, rawSeqStore_t* sequences,
ldmState_t* ldms, RawSeqStore_t* sequences,
ldmParams_t const* params, void const* src, size_t srcSize);
/**
@ -64,9 +64,9 @@ size_t ZSTD_ldm_generateSequences(
* two. We handle that case correctly, and update `rawSeqStore` appropriately.
* NOTE: This function does not return any errors.
*/
size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_paramSwitch_e useRowMatchFinder,
size_t ZSTD_ldm_blockCompress(RawSeqStore_t* rawSeqStore,
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_ParamSwitch_e useRowMatchFinder,
void const* src, size_t srcSize);
/**
@ -76,7 +76,7 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore,
* Avoids emitting matches less than `minMatch` bytes.
* Must be called for data that is not passed to ZSTD_ldm_blockCompress().
*/
void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
void ZSTD_ldm_skipSequences(RawSeqStore_t* rawSeqStore, size_t srcSize,
U32 const minMatch);
/* ZSTD_ldm_skipRawSeqStoreBytes():
@ -84,7 +84,7 @@ void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize,
* Not to be used in conjunction with ZSTD_ldm_skipSequences().
* Must be called for data with is not passed to ZSTD_ldm_blockCompress().
*/
void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes);
void ZSTD_ldm_skipRawSeqStoreBytes(RawSeqStore_t* rawSeqStore, size_t nbBytes);
/** ZSTD_ldm_getTableSize() :
* Estimate the space needed for long distance matching tables or 0 if LDM is

66
lib/compress/zstd_opt.c
View File

@ -408,7 +408,7 @@ MEM_STATIC U32 ZSTD_readMINMATCH(const void* memPtr, U32 length)
Assumption : always within prefix (i.e. not within extDict) */
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
U32 ZSTD_insertAndFindFirstIndexHash3 (const ZSTD_matchState_t* ms,
U32 ZSTD_insertAndFindFirstIndexHash3 (const ZSTD_MatchState_t* ms,
U32* nextToUpdate3,
const BYTE* const ip)
{
@ -440,7 +440,7 @@ U32 ZSTD_insertAndFindFirstIndexHash3 (const ZSTD_matchState_t* ms,
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
U32 ZSTD_insertBt1(
const ZSTD_matchState_t* ms,
const ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
U32 const target,
U32 const mls, const int extDict)
@ -560,7 +560,7 @@ U32 ZSTD_insertBt1(
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_updateTree_internal(
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
const BYTE* const ip, const BYTE* const iend,
const U32 mls, const ZSTD_dictMode_e dictMode)
{
@ -580,7 +580,7 @@ void ZSTD_updateTree_internal(
ms->nextToUpdate = target;
}
void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend) {
void ZSTD_updateTree(ZSTD_MatchState_t* ms, const BYTE* ip, const BYTE* iend) {
ZSTD_updateTree_internal(ms, ip, iend, ms->cParams.minMatch, ZSTD_noDict);
}
@ -589,7 +589,7 @@ ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
U32
ZSTD_insertBtAndGetAllMatches (
ZSTD_match_t* matches, /* store result (found matches) in this table (presumed large enough) */
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
U32* nextToUpdate3,
const BYTE* const ip, const BYTE* const iLimit,
const ZSTD_dictMode_e dictMode,
@ -625,7 +625,7 @@ ZSTD_insertBtAndGetAllMatches (
U32 mnum = 0;
U32 nbCompares = 1U << cParams->searchLog;
const ZSTD_matchState_t* dms = dictMode == ZSTD_dictMatchState ? ms->dictMatchState : NULL;
const ZSTD_MatchState_t* dms = dictMode == ZSTD_dictMatchState ? ms->dictMatchState : NULL;
const ZSTD_compressionParameters* const dmsCParams =
dictMode == ZSTD_dictMatchState ? &dms->cParams : NULL;
const BYTE* const dmsBase = dictMode == ZSTD_dictMatchState ? dms->window.base : NULL;
@ -819,7 +819,7 @@ ZSTD_insertBtAndGetAllMatches (
typedef U32 (*ZSTD_getAllMatchesFn)(
ZSTD_match_t*,
ZSTD_matchState_t*,
ZSTD_MatchState_t*,
U32*,
const BYTE*,
const BYTE*,
@ -831,7 +831,7 @@ FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
U32 ZSTD_btGetAllMatches_internal(
ZSTD_match_t* matches,
ZSTD_matchState_t* ms,
ZSTD_MatchState_t* ms,
U32* nextToUpdate3,
const BYTE* ip,
const BYTE* const iHighLimit,
@ -854,7 +854,7 @@ U32 ZSTD_btGetAllMatches_internal(
#define GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, mls) \
static U32 ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, mls)( \
ZSTD_match_t* matches, \
ZSTD_matchState_t* ms, \
ZSTD_MatchState_t* ms, \
U32* nextToUpdate3, \
const BYTE* ip, \
const BYTE* const iHighLimit, \
@ -886,7 +886,7 @@ GEN_ZSTD_BT_GET_ALL_MATCHES(dictMatchState)
}
static ZSTD_getAllMatchesFn
ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const dictMode)
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),
@ -905,7 +905,7 @@ ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const di
/* Struct containing info needed to make decision about ldm inclusion */
typedef struct {
rawSeqStore_t seqStore; /* External match candidates store for this block */
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 */
@ -915,7 +915,7 @@ typedef struct {
* 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) {
@ -1072,8 +1072,8 @@ listStats(const U32* table, int lastEltID)
FORCE_INLINE_TEMPLATE
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
size_t
ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
seqStore_t* seqStore,
ZSTD_compressBlock_opt_generic(ZSTD_MatchState_t* ms,
SeqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize,
const int optLevel,
@ -1227,13 +1227,13 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
&& (with1literal < opt[cur+1].price) ) {
/* update offset history - before it disappears */
U32 const prev = cur - prevMatch.mlen;
repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, prevMatch.off, opt[prev].litlen==0);
Repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, prevMatch.off, opt[prev].litlen==0);
assert(cur >= prevMatch.mlen);
DEBUGLOG(7, "==> match+1lit is cheaper (%.2f < %.2f) (hist:%u,%u,%u) !",
ZSTD_fCost(with1literal), ZSTD_fCost(withMoreLiterals),
newReps.rep[0], newReps.rep[1], newReps.rep[2] );
opt[cur+1] = prevMatch; /* mlen & offbase */
ZSTD_memcpy(opt[cur+1].rep, &newReps, sizeof(repcodes_t));
ZSTD_memcpy(opt[cur+1].rep, &newReps, sizeof(Repcodes_t));
opt[cur+1].litlen = 1;
opt[cur+1].price = with1literal;
if (last_pos < cur+1) last_pos = cur+1;
@ -1248,13 +1248,13 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms,
/* Offset history is not updated during match comparison.
* Do it here, now that the match is selected and confirmed.
*/
ZSTD_STATIC_ASSERT(sizeof(opt[cur].rep) == sizeof(repcodes_t));
ZSTD_STATIC_ASSERT(sizeof(opt[cur].rep) == sizeof(Repcodes_t));
assert(cur >= opt[cur].mlen);
if (opt[cur].litlen == 0) {
/* just finished a match => alter offset history */
U32 const prev = cur - opt[cur].mlen;
repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, opt[cur].off, opt[prev].litlen==0);
ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(repcodes_t));
Repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, opt[cur].off, opt[prev].litlen==0);
ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(Repcodes_t));
}
/* last match must start at a minimum distance of 8 from oend */
@ -1353,10 +1353,10 @@ _shortestPath: /* cur, last_pos, best_mlen, best_off have to be set */
/* Update offset history */
if (lastStretch.litlen == 0) {
/* finishing on a match : update offset history */
repcodes_t const reps = ZSTD_newRep(opt[cur].rep, lastStretch.off, opt[cur].litlen==0);
ZSTD_memcpy(rep, &reps, sizeof(repcodes_t));
Repcodes_t const reps = ZSTD_newRep(opt[cur].rep, lastStretch.off, opt[cur].litlen==0);
ZSTD_memcpy(rep, &reps, sizeof(Repcodes_t));
} else {
ZSTD_memcpy(rep, lastStretch.rep, sizeof(repcodes_t));
ZSTD_memcpy(rep, lastStretch.rep, sizeof(Repcodes_t));
assert(cur >= lastStretch.litlen);
cur -= lastStretch.litlen;
}
@ -1440,7 +1440,7 @@ _shortestPath: /* cur, last_pos, best_mlen, best_off have to be set */
#ifndef ZSTD_EXCLUDE_BTOPT_BLOCK_COMPRESSOR
static size_t ZSTD_compressBlock_opt0(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize, const ZSTD_dictMode_e dictMode)
{
return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 0 /* optLevel */, dictMode);
@ -1449,7 +1449,7 @@ static size_t ZSTD_compressBlock_opt0(
#ifndef ZSTD_EXCLUDE_BTULTRA_BLOCK_COMPRESSOR
static size_t ZSTD_compressBlock_opt2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize, const ZSTD_dictMode_e dictMode)
{
return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /* optLevel */, dictMode);
@ -1458,7 +1458,7 @@ static size_t ZSTD_compressBlock_opt2(
#ifndef ZSTD_EXCLUDE_BTOPT_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btopt(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
DEBUGLOG(5, "ZSTD_compressBlock_btopt");
@ -1477,8 +1477,8 @@ size_t ZSTD_compressBlock_btopt(
*/
static
ZSTD_ALLOW_POINTER_OVERFLOW_ATTR
void ZSTD_initStats_ultra(ZSTD_matchState_t* ms,
seqStore_t* seqStore,
void ZSTD_initStats_ultra(ZSTD_MatchState_t* ms,
SeqStore_t* seqStore,
U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
@ -1503,7 +1503,7 @@ void ZSTD_initStats_ultra(ZSTD_matchState_t* ms,
}
size_t ZSTD_compressBlock_btultra(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
DEBUGLOG(5, "ZSTD_compressBlock_btultra (srcSize=%zu)", srcSize);
@ -1511,7 +1511,7 @@ size_t ZSTD_compressBlock_btultra(
}
size_t ZSTD_compressBlock_btultra2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
U32 const curr = (U32)((const BYTE*)src - ms->window.base);
@ -1541,14 +1541,14 @@ size_t ZSTD_compressBlock_btultra2(
#ifndef ZSTD_EXCLUDE_BTOPT_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btopt_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
return ZSTD_compressBlock_opt0(ms, seqStore, rep, src, srcSize, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_btopt_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
return ZSTD_compressBlock_opt0(ms, seqStore, rep, src, srcSize, ZSTD_extDict);
@ -1557,14 +1557,14 @@ size_t ZSTD_compressBlock_btopt_extDict(
#ifndef ZSTD_EXCLUDE_BTULTRA_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btultra_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_dictMatchState);
}
size_t ZSTD_compressBlock_btultra_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
const void* src, size_t srcSize)
{
return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_extDict);

16
lib/compress/zstd_opt.h
View File

@ -21,18 +21,18 @@ extern "C" {
|| !defined(ZSTD_EXCLUDE_BTOPT_BLOCK_COMPRESSOR) \
|| !defined(ZSTD_EXCLUDE_BTULTRA_BLOCK_COMPRESSOR)
/* used in ZSTD_loadDictionaryContent() */
void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend);
void ZSTD_updateTree(ZSTD_MatchState_t* ms, const BYTE* ip, const BYTE* iend);
#endif
#ifndef ZSTD_EXCLUDE_BTOPT_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btopt(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btopt_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btopt_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_BTOPT ZSTD_compressBlock_btopt
@ -46,20 +46,20 @@ size_t ZSTD_compressBlock_btopt_extDict(
#ifndef ZSTD_EXCLUDE_BTULTRA_BLOCK_COMPRESSOR
size_t ZSTD_compressBlock_btultra(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra_dictMatchState(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
size_t ZSTD_compressBlock_btultra_extDict(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
/* note : no btultra2 variant for extDict nor dictMatchState,
* because btultra2 is not meant to work with dictionaries
* and is only specific for the first block (no prefix) */
size_t ZSTD_compressBlock_btultra2(
ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
ZSTD_MatchState_t* ms, SeqStore_t* seqStore, U32 rep[ZSTD_REP_NUM],
void const* src, size_t srcSize);
#define ZSTD_COMPRESSBLOCK_BTULTRA ZSTD_compressBlock_btultra

16
lib/compress/zstdmt_compress.c
View File

@ -290,15 +290,15 @@ static size_t ZSTDMT_sizeof_seqPool(ZSTDMT_seqPool* seqPool)
return ZSTDMT_sizeof_bufferPool(seqPool);
}
static rawSeqStore_t bufferToSeq(buffer_t buffer)
static RawSeqStore_t bufferToSeq(buffer_t buffer)
{
rawSeqStore_t seq = kNullRawSeqStore;
RawSeqStore_t seq = kNullRawSeqStore;
seq.seq = (rawSeq*)buffer.start;
seq.capacity = buffer.capacity / sizeof(rawSeq);
return seq;
}
static buffer_t seqToBuffer(rawSeqStore_t seq)
static buffer_t seqToBuffer(RawSeqStore_t seq)
{
buffer_t buffer;
buffer.start = seq.seq;
@ -306,7 +306,7 @@ static buffer_t seqToBuffer(rawSeqStore_t seq)
return buffer;
}
static rawSeqStore_t ZSTDMT_getSeq(ZSTDMT_seqPool* seqPool)
static RawSeqStore_t ZSTDMT_getSeq(ZSTDMT_seqPool* seqPool)
{
if (seqPool->bufferSize == 0) {
return kNullRawSeqStore;
@ -315,13 +315,13 @@ static rawSeqStore_t ZSTDMT_getSeq(ZSTDMT_seqPool* seqPool)
}
#if ZSTD_RESIZE_SEQPOOL
static rawSeqStore_t ZSTDMT_resizeSeq(ZSTDMT_seqPool* seqPool, rawSeqStore_t seq)
static RawSeqStore_t ZSTDMT_resizeSeq(ZSTDMT_seqPool* seqPool, RawSeqStore_t seq)
{
return bufferToSeq(ZSTDMT_resizeBuffer(seqPool, seqToBuffer(seq)));
}
#endif
static void ZSTDMT_releaseSeq(ZSTDMT_seqPool* seqPool, rawSeqStore_t seq)
static void ZSTDMT_releaseSeq(ZSTDMT_seqPool* seqPool, RawSeqStore_t seq)
{
ZSTDMT_releaseBuffer(seqPool, seqToBuffer(seq));
}
@ -578,7 +578,7 @@ static void ZSTDMT_serialState_free(serialState_t* serialState)
}
static void ZSTDMT_serialState_update(serialState_t* serialState,
ZSTD_CCtx* jobCCtx, rawSeqStore_t seqStore,
ZSTD_CCtx* jobCCtx, RawSeqStore_t seqStore,
range_t src, unsigned jobID)
{
/* Wait for our turn */
@ -685,7 +685,7 @@ static void ZSTDMT_compressionJob(void* jobDescription)
ZSTDMT_jobDescription* const job = (ZSTDMT_jobDescription*)jobDescription;
ZSTD_CCtx_params jobParams = job->params; /* do not modify job->params ! copy it, modify the copy */
ZSTD_CCtx* const cctx = ZSTDMT_getCCtx(job->cctxPool);
rawSeqStore_t rawSeqStore = ZSTDMT_getSeq(job->seqPool);
RawSeqStore_t rawSeqStore = ZSTDMT_getSeq(job->seqPool);
buffer_t dstBuff = job->dstBuff;
size_t lastCBlockSize = 0;

10
lib/decompress/zstd_decompress_block.c
View File

@ -139,7 +139,7 @@ static size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx,
RETURN_ERROR_IF(srcSize < MIN_CBLOCK_SIZE, corruption_detected, "");
{ const BYTE* const istart = (const BYTE*) src;
symbolEncodingType_e const litEncType = (symbolEncodingType_e)(istart[0] & 3);
SymbolEncodingType_e const litEncType = (SymbolEncodingType_e)(istart[0] & 3);
size_t const blockSizeMax = ZSTD_blockSizeMax(dctx);
switch(litEncType)
@ -645,7 +645,7 @@ void ZSTD_buildFSETable(ZSTD_seqSymbol* dt,
* @return : nb bytes read from src,
* or an error code if it fails */
static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymbol** DTablePtr,
symbolEncodingType_e type, unsigned max, U32 maxLog,
SymbolEncodingType_e type, unsigned max, U32 maxLog,
const void* src, size_t srcSize,
const U32* baseValue, const U8* nbAdditionalBits,
const ZSTD_seqSymbol* defaultTable, U32 flagRepeatTable,
@ -728,9 +728,9 @@ size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr,
/* FSE table descriptors */
RETURN_ERROR_IF(ip+1 > iend, srcSize_wrong, ""); /* minimum possible size: 1 byte for symbol encoding types */
RETURN_ERROR_IF(*ip & 3, corruption_detected, ""); /* The last field, Reserved, must be all-zeroes. */
{ symbolEncodingType_e const LLtype = (symbolEncodingType_e)(*ip >> 6);
symbolEncodingType_e const OFtype = (symbolEncodingType_e)((*ip >> 4) & 3);
symbolEncodingType_e const MLtype = (symbolEncodingType_e)((*ip >> 2) & 3);
{ SymbolEncodingType_e const LLtype = (SymbolEncodingType_e)(*ip >> 6);
SymbolEncodingType_e const OFtype = (SymbolEncodingType_e)((*ip >> 4) & 3);
SymbolEncodingType_e const MLtype = (SymbolEncodingType_e)((*ip >> 2) & 3);
ip++;
/* Build DTables */

4
lib/dictBuilder/zdict.c
View File

@ -580,7 +580,7 @@ static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
if (ZSTD_isError(cSize)) { DISPLAYLEVEL(3, "warning : could not compress sample size %u \n", (unsigned)srcSize); return; }
if (cSize) { /* if == 0; block is not compressible */
const seqStore_t* const seqStorePtr = ZSTD_getSeqStore(esr.zc);
const SeqStore_t* const seqStorePtr = ZSTD_getSeqStore(esr.zc);
/* literals stats */
{ const BYTE* bytePtr;
@ -608,7 +608,7 @@ static void ZDICT_countEStats(EStats_ress_t esr, const ZSTD_parameters* params,
}
if (nbSeq >= 2) { /* rep offsets */
const seqDef* const seq = seqStorePtr->sequencesStart;
const SeqDef* const seq = seqStorePtr->sequencesStart;
U32 offset1 = seq[0].offBase - ZSTD_REP_NUM;
U32 offset2 = seq[1].offBase - ZSTD_REP_NUM;
if (offset1 >= MAXREPOFFSET) offset1 = 0;

4
lib/legacy/zstd_v01.c
View File

@ -1383,7 +1383,7 @@ typedef struct {
BYTE* matchLength;
BYTE* dumpsStart;
BYTE* dumps;
} seqStore_t;
} SeqStore_t;
typedef struct ZSTD_Cctx_s
@ -1391,7 +1391,7 @@ typedef struct ZSTD_Cctx_s
const BYTE* base;
U32 current;
U32 nextUpdate;
seqStore_t seqStore;
SeqStore_t seqStore;
#ifdef __AVX2__
__m256i hashTable[HASH_TABLESIZE>>3];
#else

2
lib/legacy/zstd_v02.c
View File

@ -2722,7 +2722,7 @@ typedef struct {
BYTE* matchLength;
BYTE* dumpsStart;
BYTE* dumps;
} seqStore_t;
} SeqStore_t;
/* *************************************

2
lib/legacy/zstd_v03.c
View File

@ -2362,7 +2362,7 @@ typedef struct {
BYTE* matchLength;
BYTE* dumpsStart;
BYTE* dumps;
} seqStore_t;
} SeqStore_t;
/* *************************************

2
lib/legacy/zstd_v05.c
View File

@ -491,7 +491,7 @@ typedef struct {
U32 litLengthSum;
U32 litSum;
U32 offCodeSum;
} seqStore_t;
} SeqStore_t;

4
lib/legacy/zstd_v06.c
View File

@ -552,9 +552,9 @@ typedef struct {
U32 cachedLitLength;
const BYTE* cachedLiterals;
ZSTDv06_stats_t stats;
} seqStore_t;
} SeqStore_t;
void ZSTDv06_seqToCodes(const seqStore_t* seqStorePtr, size_t const nbSeq);
void ZSTDv06_seqToCodes(const SeqStore_t* seqStorePtr, size_t const nbSeq);
#endif /* ZSTDv06_CCOMMON_H_MODULE */

4
lib/legacy/zstd_v07.c
View File

@ -2787,9 +2787,9 @@ typedef struct {
U32 cachedLitLength;
const BYTE* cachedLiterals;
ZSTDv07_stats_t stats;
} seqStore_t;
} SeqStore_t;
void ZSTDv07_seqToCodes(const seqStore_t* seqStorePtr, size_t const nbSeq);
void ZSTDv07_seqToCodes(const SeqStore_t* seqStorePtr, size_t const nbSeq);
/* custom memory allocation functions */
static const ZSTDv07_customMem defaultCustomMem = { ZSTDv07_defaultAllocFunction, ZSTDv07_defaultFreeFunction, NULL };

117
lib/zstd.h
View File

@ -1328,7 +1328,7 @@ typedef struct {
*
* Note: This field is optional. ZSTD_generateSequences() will calculate the value of
* 'rep', but repeat offsets do not necessarily need to be calculated from an external
* sequence provider's perspective. For example, ZSTD_compressSequences() does not
* sequence provider perspective. For example, ZSTD_compressSequences() does not
* use this 'rep' field at all (as of now).
*/
} ZSTD_Sequence;
@ -1433,14 +1433,15 @@ typedef enum {
} ZSTD_literalCompressionMode_e;
typedef enum {
/* Note: This enum controls features which are conditionally beneficial. Zstd typically will make a final
* decision on whether or not to enable the feature (ZSTD_ps_auto), but setting the switch to ZSTD_ps_enable
* or ZSTD_ps_disable allow for a force enable/disable the feature.
/* Note: This enum controls features which are conditionally beneficial.
* Zstd can take a decision on whether or not to enable the feature (ZSTD_ps_auto),
* but setting the switch to ZSTD_ps_enable or ZSTD_ps_disable force enable/disable the feature.
*/
ZSTD_ps_auto = 0, /* Let the library automatically determine whether the feature shall be enabled */
ZSTD_ps_enable = 1, /* Force-enable the feature */
ZSTD_ps_disable = 2 /* Do not use the feature */
} ZSTD_paramSwitch_e;
} ZSTD_ParamSwitch_e;
#define ZSTD_paramSwitch_e ZSTD_ParamSwitch_e /* old name */
/***************************************
* Frame header and size functions
@ -1560,9 +1561,10 @@ ZSTDLIB_STATIC_API size_t ZSTD_decompressionMargin(const void* src, size_t srcSi
))
typedef enum {
ZSTD_sf_noBlockDelimiters = 0, /* Representation of ZSTD_Sequence has no block delimiters, sequences only */
ZSTD_sf_explicitBlockDelimiters = 1 /* Representation of ZSTD_Sequence contains explicit block delimiters */
} ZSTD_sequenceFormat_e;
ZSTD_sf_noBlockDelimiters = 0, /* ZSTD_Sequence[] has no block delimiters, just sequences */
ZSTD_sf_explicitBlockDelimiters = 1 /* ZSTD_Sequence[] contains explicit block delimiters */
} ZSTD_SequenceFormat_e;
#define ZSTD_sequenceFormat_e ZSTD_SequenceFormat_e /* old name */
/*! ZSTD_sequenceBound() :
* `srcSize` : size of the input buffer
@ -1586,7 +1588,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_sequenceBound(size_t srcSize);
* @param zc The compression context to be used for ZSTD_compress2(). Set any
* compression parameters you need on this context.
* @param outSeqs The output sequences buffer of size @p outSeqsSize
* @param outSeqsSize The size of the output sequences buffer.
* @param outSeqsCapacity The size of the output sequences buffer.
* ZSTD_sequenceBound(srcSize) is an upper bound on the number
* of sequences that can be generated.
* @param src The source buffer to generate sequences from of size @p srcSize.
@ -1604,7 +1606,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_sequenceBound(size_t srcSize);
ZSTD_DEPRECATED("For debugging only, will be replaced by ZSTD_extractSequences()")
ZSTDLIB_STATIC_API size_t
ZSTD_generateSequences(ZSTD_CCtx* zc,
ZSTD_Sequence* outSeqs, size_t outSeqsSize,
ZSTD_Sequence* outSeqs, size_t outSeqsCapacity,
const void* src, size_t srcSize);
/*! ZSTD_mergeBlockDelimiters() :
@ -1624,7 +1626,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, si
* Compress an array of ZSTD_Sequence, associated with @src buffer, into dst.
* @src contains the entire input (not just the literals).
* If @srcSize > sum(sequence.length), the remaining bytes are considered all literals
* If a dictionary is included, then the cctx should reference the dict. (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.)
* If a dictionary is included, then the cctx should reference the dict (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.).
* The entire source is compressed into a single frame.
*
* The compression behavior changes based on cctx params. In particular:
@ -1633,11 +1635,17 @@ ZSTDLIB_STATIC_API size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, si
* the block size derived from the cctx, and sequences may be split. This is the default setting.
*
* If ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, the array of ZSTD_Sequence is expected to contain
* block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided.
* valid block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided.
*
* If ZSTD_c_validateSequences == 0, this function will blindly accept the sequences provided. Invalid sequences cause undefined
* behavior. If ZSTD_c_validateSequences == 1, then if sequence is invalid (see doc/zstd_compression_format.md for
* specifics regarding offset/matchlength requirements) then the function will bail out and return an error.
* When ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, it's possible to decide generating repcodes
* using the advanced parameter ZSTD_c_repcodeResolution. Repcodes will improve compression ratio, though the benefit
* can vary greatly depending on Sequences. On the other hand, repcode resolution is an expensive operation.
* By default, it's disabled at low (<10) compression levels, and enabled above the threshold (>=10).
* ZSTD_c_repcodeResolution makes it possible to directly manage this processing in either direction.
*
* If ZSTD_c_validateSequences == 0, this function blindly accepts the Sequences provided. Invalid Sequences cause undefined
* behavior. If ZSTD_c_validateSequences == 1, then the function will detect invalid Sequences (see doc/zstd_compression_format.md for
* specifics regarding offset/matchlength requirements) and then bail out and return an error.
*
* In addition to the two adjustable experimental params, there are other important cctx params.
* - ZSTD_c_minMatch MUST be set as less than or equal to the smallest match generated by the match finder. It has a minimum value of ZSTD_MINMATCH_MIN.
@ -1645,15 +1653,41 @@ ZSTDLIB_STATIC_API size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, si
* - ZSTD_c_windowLog affects offset validation: this function will return an error at higher debug levels if a provided offset
* is larger than what the spec allows for a given window log and dictionary (if present). See: doc/zstd_compression_format.md
*
* Note: Repcodes are, as of now, always re-calculated within this function, so ZSTD_Sequence::rep is unused.
* Note 2: Once we integrate ability to ingest repcodes, the explicit block delims mode must respect those repcodes exactly,
* and cannot emit an RLE block that disagrees with the repcode history
* Note: Repcodes are, as of now, always re-calculated within this function, ZSTD_Sequence.rep is effectively unused.
* Dev Note: Once ability to ingest repcodes become available, the explicit block delims mode must respect those repcodes exactly,
* and cannot emit an RLE block that disagrees with the repcode history.
* @return : final compressed size, or a ZSTD error code.
*/
ZSTDLIB_STATIC_API size_t
ZSTD_compressSequences( ZSTD_CCtx* cctx, void* dst, size_t dstSize,
const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
const void* src, size_t srcSize);
ZSTD_compressSequences(ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const ZSTD_Sequence* inSeqs, size_t inSeqsSize,
const void* src, size_t srcSize);
/*! ZSTD_compressSequencesAndLiterals() :
* This is a variant of ZSTD_compressSequences() which,
* instead of receiving (src,srcSize) as input parameter, receives (literals,litSize),
* aka all the literals, already extracted and laid out into a single continuous buffer.
* This can be useful if the process generating the sequences also happens to generate the buffer of literals,
* thus skipping an extraction + caching stage.
* It's a speed optimization, useful when the right conditions are met,
* but it also features the following limitations:
* - Only supports explicit delimiter mode
* - Currently does not support Sequences validation (so input Sequences are trusted)
* - Not compatible with frame checksum, which must be disabled
* - If any block is incompressible, will fail and return an error
* - @litSize must be == sum of all @.litLength fields in @inSeqs. Any discrepancy will generate an error.
* - the buffer @literals must have a size @litCapacity which is larger than @litSize by at least 8 bytes.
* - @decompressedSize must be correct, and correspond to the sum of all Sequences. Any discrepancy will generate an error.
* @return : final compressed size, or a ZSTD error code.
*/
ZSTDLIB_STATIC_API size_t
ZSTD_compressSequencesAndLiterals(ZSTD_CCtx* cctx,
void* dst, size_t dstCapacity,
const ZSTD_Sequence* inSeqs, size_t nbSequences,
const void* literals, size_t litSize, size_t litCapacity,
size_t decompressedSize);
/*! ZSTD_writeSkippableFrame() :
@ -2005,7 +2039,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
* See the comments on that enum for an explanation of the feature. */
#define ZSTD_c_forceAttachDict ZSTD_c_experimentalParam4
/* Controlled with ZSTD_paramSwitch_e enum.
/* Controlled with ZSTD_ParamSwitch_e enum.
* Default is ZSTD_ps_auto.
* Set to ZSTD_ps_disable to never compress literals.
* Set to ZSTD_ps_enable to always compress literals. (Note: uncompressed literals
@ -2146,17 +2180,17 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
/* ZSTD_c_validateSequences
* Default is 0 == disabled. Set to 1 to enable sequence validation.
*
* For use with sequence compression API: ZSTD_compressSequences().
* Designates whether or not we validate sequences provided to ZSTD_compressSequences()
* For use with sequence compression API: ZSTD_compressSequences*().
* Designates whether or not provided sequences are validated within ZSTD_compressSequences*()
* during function execution.
*
* Without validation, providing a sequence that does not conform to the zstd spec will cause
* undefined behavior, and may produce a corrupted block.
* When Sequence validation is disabled (default), Sequences are compressed as-is,
* so they must correct, otherwise it would result in a corruption error.
*
* With validation enabled, if sequence is invalid (see doc/zstd_compression_format.md for
* Sequence validation adds some protection, by ensuring that all values respect boundary conditions.
* If a Sequence is detected invalid (see doc/zstd_compression_format.md for
* specifics regarding offset/matchlength requirements) then the function will bail out and
* return an error.
*
*/
#define ZSTD_c_validateSequences ZSTD_c_experimentalParam12
@ -2196,7 +2230,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
#define ZSTD_c_splitAfterSequences ZSTD_c_experimentalParam13
/* ZSTD_c_useRowMatchFinder
* Controlled with ZSTD_paramSwitch_e enum.
* Controlled with ZSTD_ParamSwitch_e enum.
* Default is ZSTD_ps_auto.
* Set to ZSTD_ps_disable to never use row-based matchfinder.
* Set to ZSTD_ps_enable to force usage of row-based matchfinder.
@ -2228,7 +2262,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
#define ZSTD_c_deterministicRefPrefix ZSTD_c_experimentalParam15
/* ZSTD_c_prefetchCDictTables
* Controlled with ZSTD_paramSwitch_e enum. Default is ZSTD_ps_auto.
* Controlled with ZSTD_ParamSwitch_e enum. Default is ZSTD_ps_auto.
*
* In some situations, zstd uses CDict tables in-place rather than copying them
* into the working context. (See docs on ZSTD_dictAttachPref_e above for details).
@ -2275,15 +2309,18 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
*/
#define ZSTD_c_maxBlockSize ZSTD_c_experimentalParam18
/* ZSTD_c_searchForExternalRepcodes
* This parameter affects how zstd parses external sequences, such as sequences
* provided through the compressSequences() API or from an external block-level
* sequence producer.
/* ZSTD_c_repcodeResolution
* This parameter only has an effect if ZSTD_c_blockDelimiters is
* set to ZSTD_sf_explicitBlockDelimiters (may change in the future).
*
* If set to ZSTD_ps_enable, the library will check for repeated offsets in
* This parameter affects how zstd parses external sequences,
* provided via the ZSTD_compressSequences*() API
* or from an external block-level sequence producer.
*
* If set to ZSTD_ps_enable, the library will check for repeated offsets within
* external sequences, even if those repcodes are not explicitly indicated in
* the "rep" field. Note that this is the only way to exploit repcode matches
* while using compressSequences() or an external sequence producer, since zstd
* while using compressSequences*() or an external sequence producer, since zstd
* currently ignores the "rep" field of external sequences.
*
* If set to ZSTD_ps_disable, the library will not exploit repeated offsets in
@ -2292,12 +2329,10 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo
* compression ratio.
*
* The default value is ZSTD_ps_auto, for which the library will enable/disable
* based on compression level.
*
* Note: for now, this param only has an effect if ZSTD_c_blockDelimiters is
* set to ZSTD_sf_explicitBlockDelimiters. That may change in the future.
* based on compression level (currently: level<10 disables, level>=10 enables).
*/
#define ZSTD_c_searchForExternalRepcodes ZSTD_c_experimentalParam19
#define ZSTD_c_repcodeResolution ZSTD_c_experimentalParam19
#define ZSTD_c_searchForExternalRepcodes ZSTD_c_experimentalParam19 /* older name */
/*! ZSTD_CCtx_getParameter() :

1
lib/zstd_errors.h
View File

@ -76,6 +76,7 @@ typedef enum {
ZSTD_error_tableLog_tooLarge = 44,
ZSTD_error_maxSymbolValue_tooLarge = 46,
ZSTD_error_maxSymbolValue_tooSmall = 48,
ZSTD_error_cannotProduce_uncompressedBlock = 49,
ZSTD_error_stabilityCondition_notRespected = 50,
ZSTD_error_stage_wrong = 60,
ZSTD_error_init_missing = 62,

2
programs/benchzstd.h
View File

@ -109,7 +109,7 @@ typedef struct {
int ldmHashLog;
int ldmBucketSizeLog;
int ldmHashRateLog;
ZSTD_paramSwitch_e literalCompressionMode;
ZSTD_ParamSwitch_e literalCompressionMode;
int useRowMatchFinder; /* use row-based matchfinder if possible */
} BMK_advancedParams_t;

4
programs/fileio.c
View File

@ -423,7 +423,7 @@ void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode) {
void FIO_setLiteralCompressionMode(
FIO_prefs_t* const prefs,
ZSTD_paramSwitch_e mode) {
ZSTD_ParamSwitch_e mode) {
prefs->literalCompressionMode = mode;
}
@ -485,7 +485,7 @@ void FIO_setPassThroughFlag(FIO_prefs_t* const prefs, int value) {
prefs->passThrough = (value != 0);
}
void FIO_setMMapDict(FIO_prefs_t* const prefs, ZSTD_paramSwitch_e value)
void FIO_setMMapDict(FIO_prefs_t* const prefs, ZSTD_ParamSwitch_e value)
{
prefs->mmapDict = value;
}

4
programs/fileio.h
View File

@ -95,7 +95,7 @@ void FIO_setSrcSizeHint(FIO_prefs_t* const prefs, size_t srcSizeHint);
void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode);
void FIO_setLiteralCompressionMode(
FIO_prefs_t* const prefs,
ZSTD_paramSwitch_e mode);
ZSTD_ParamSwitch_e mode);
void FIO_setProgressSetting(FIO_progressSetting_e progressSetting);
void FIO_setNotificationLevel(int level);
@ -106,7 +106,7 @@ void FIO_setContentSize(FIO_prefs_t* const prefs, int value);
void FIO_displayCompressionParameters(const FIO_prefs_t* prefs);
void FIO_setAsyncIOFlag(FIO_prefs_t* const prefs, int value);
void FIO_setPassThroughFlag(FIO_prefs_t* const prefs, int value);
void FIO_setMMapDict(FIO_prefs_t* const prefs, ZSTD_paramSwitch_e value);
void FIO_setMMapDict(FIO_prefs_t* const prefs, ZSTD_ParamSwitch_e value);
/* FIO_ctx_t functions */
void FIO_setNbFilesTotal(FIO_ctx_t* const fCtx, int value);

4
programs/fileio_types.h
View File

@ -53,7 +53,7 @@ typedef struct FIO_prefs_s {
size_t targetCBlockSize;
int srcSizeHint;
int testMode;
ZSTD_paramSwitch_e literalCompressionMode;
ZSTD_ParamSwitch_e literalCompressionMode;
/* IO preferences */
int removeSrcFile;
@ -69,7 +69,7 @@ typedef struct FIO_prefs_s {
int contentSize;
int allowBlockDevices;
int passThrough;
ZSTD_paramSwitch_e mmapDict;
ZSTD_ParamSwitch_e mmapDict;
} FIO_prefs_t;
typedef enum {FIO_mallocDict, FIO_mmapDict} FIO_dictBufferType_t;

8
programs/zstdcli.c
View File

@ -635,8 +635,6 @@ static unsigned parseCompressionParameters(const char* stringPtr, ZSTD_compressi
return 0;
}
DISPLAYLEVEL(4, "windowLog=%d, chainLog=%d, hashLog=%d, searchLog=%d \n", params->windowLog, params->chainLog, params->hashLog, params->searchLog);
DISPLAYLEVEL(4, "minMatch=%d, targetLength=%d, strategy=%d \n", params->minMatch, params->targetLength, params->strategy);
if (stringPtr[0] != 0) return 0; /* check the end of string */
return 1;
}
@ -851,8 +849,8 @@ int main(int argCount, const char* argv[])
ultra=0,
contentSize=1,
removeSrcFile=0;
ZSTD_paramSwitch_e mmapDict=ZSTD_ps_auto;
ZSTD_paramSwitch_e useRowMatchFinder = ZSTD_ps_auto;
ZSTD_ParamSwitch_e mmapDict=ZSTD_ps_auto;
ZSTD_ParamSwitch_e useRowMatchFinder = ZSTD_ps_auto;
FIO_compressionType_t cType = FIO_zstdCompression;
int nbWorkers = -1; /* -1 means unset */
double compressibility = -1.0; /* lorem ipsum generator */
@ -893,7 +891,7 @@ int main(int argCount, const char* argv[])
#ifndef ZSTD_NOBENCH
BMK_advancedParams_t benchParams = BMK_initAdvancedParams();
#endif
ZSTD_paramSwitch_e literalCompressionMode = ZSTD_ps_auto;
ZSTD_ParamSwitch_e literalCompressionMode = ZSTD_ps_auto;
/* init */
checkLibVersion();

13
tests/Makefile
View File

@ -26,6 +26,9 @@ export ZSTD_LEGACY_SUPPORT
DEBUGLEVEL ?= 2
export DEBUGLEVEL # transmit value to sub-makefiles
.PHONY: default
default: fullbench
LIBZSTD_MK_DIR := ../lib
include $(LIBZSTD_MK_DIR)/libzstd.mk
@ -78,9 +81,6 @@ FUZZERTEST ?= -T200s
ZSTDRTTEST = --test-large-data
DECODECORPUS_TESTTIME ?= -T30
.PHONY: default
default: fullbench
.PHONY: all
all: fullbench fuzzer zstreamtest paramgrill datagen decodecorpus roundTripCrash poolTests
@ -148,13 +148,14 @@ fullbench32: CPPFLAGS += -m32
$(FULLBENCHS) : CPPFLAGS += $(MULTITHREAD_CPP) -Wno-deprecated-declarations
$(FULLBENCHS) : LDFLAGS += $(MULTITHREAD_LD)
$(FULLBENCHS) : DEBUGFLAGS = -DNDEBUG # turn off assert() for speed measurements
$(FULLBENCHS) : DEBUGLEVEL ?= 0 # turn off assert() for speed measurements
$(FULLBENCHS) : $(ZSTD_FILES)
$(FULLBENCHS) : $(PRGDIR)/datagen.c $(PRGDIR)/util.c $(PRGDIR)/timefn.c $(PRGDIR)/benchfn.c fullbench.c
$(FULLBENCHS) : $(PRGDIR)/datagen.c $(PRGDIR)/lorem.c $(PRGDIR)/util.c $(PRGDIR)/timefn.c $(PRGDIR)/benchfn.c fullbench.c
$(LINK.c) $^ -o $@$(EXT)
CLEAN += fullbench-lib
fullbench-lib : CPPFLAGS += -DXXH_NAMESPACE=ZSTD_
fullbench-lib : $(PRGDIR)/datagen.c $(PRGDIR)/util.c $(PRGDIR)/timefn.c $(PRGDIR)/benchfn.c $(LIB_SRCDIR)/libzstd.a fullbench.c
fullbench-lib : $(PRGDIR)/datagen.c $(PRGDIR)/lorem.c $(PRGDIR)/util.c $(PRGDIR)/timefn.c $(PRGDIR)/benchfn.c $(LIB_SRCDIR)/libzstd.a fullbench.c
$(LINK.c) $^ -o $@$(EXT)
# note : broken : requires symbols unavailable from dynamic library
@ -271,7 +272,7 @@ clean:
#----------------------------------------------------------------------------------
# valgrind tests validated only for some posix platforms
#----------------------------------------------------------------------------------
UNAME := $(shell sh -c 'MSYSTEM="MSYS" uname')
UNAME := $(shell sh -c 'MSYSTEM="MSYS" uname')
ifneq (,$(filter Linux Darwin GNU/kFreeBSD GNU OpenBSD FreeBSD NetBSD DragonFly SunOS AIX CYGWIN_NT%,$(UNAME)))
HOST_OS = POSIX

14
tests/decodecorpus.c
View File

@ -182,7 +182,7 @@ BYTE CONTENT_BUFFER[MAX_DECOMPRESSED_SIZE];
BYTE FRAME_BUFFER[MAX_DECOMPRESSED_SIZE * 2];
BYTE LITERAL_BUFFER[ZSTD_BLOCKSIZE_MAX];
seqDef SEQUENCE_BUFFER[MAX_NB_SEQ];
SeqDef SEQUENCE_BUFFER[MAX_NB_SEQ];
BYTE SEQUENCE_LITERAL_BUFFER[ZSTD_BLOCKSIZE_MAX]; /* storeSeq expects a place to copy literals to */
BYTE SEQUENCE_LLCODE[ZSTD_BLOCKSIZE_MAX];
BYTE SEQUENCE_MLCODE[ZSTD_BLOCKSIZE_MAX];
@ -505,7 +505,7 @@ static size_t writeLiteralsBlockCompressed(U32* seed, frame_t* frame, size_t con
size_t compressedSize = 0;
size_t maxLitSize = MIN(contentSize-3, g_maxBlockSize);
symbolEncodingType_e hType;
SymbolEncodingType_e hType;
if (contentSize < 64) {
/* make sure we get reasonably-sized literals for compression */
@ -657,7 +657,7 @@ static size_t writeLiteralsBlock(U32* seed, frame_t* frame, size_t contentSize)
}
}
static inline void initSeqStore(seqStore_t *seqStore) {
static inline void initSeqStore(SeqStore_t *seqStore) {
seqStore->maxNbSeq = MAX_NB_SEQ;
seqStore->maxNbLit = ZSTD_BLOCKSIZE_MAX;
seqStore->sequencesStart = SEQUENCE_BUFFER;
@ -671,7 +671,7 @@ static inline void initSeqStore(seqStore_t *seqStore) {
/* Randomly generate sequence commands */
static U32
generateSequences(U32* seed, frame_t* frame, seqStore_t* seqStore,
generateSequences(U32* seed, frame_t* frame, SeqStore_t* seqStore,
size_t contentSize, size_t literalsSize, dictInfo info)
{
/* The total length of all the matches */
@ -832,7 +832,7 @@ static int isSymbolSubset(const BYTE* symbols, size_t len, const BYTE* set, BYTE
return 1;
}
static size_t writeSequences(U32* seed, frame_t* frame, seqStore_t* seqStorePtr,
static size_t writeSequences(U32* seed, frame_t* frame, SeqStore_t* seqStorePtr,
size_t nbSeq)
{
/* This code is mostly copied from ZSTD_compressSequences in zstd_compress.c */
@ -842,7 +842,7 @@ static size_t writeSequences(U32* seed, frame_t* frame, seqStore_t* seqStorePtr,
FSE_CTable* CTable_OffsetBits = frame->stats.offcodeCTable;
FSE_CTable* CTable_MatchLength = frame->stats.matchlengthCTable;
U32 LLtype, Offtype, MLtype; /* compressed, raw or rle */
const seqDef* const sequences = seqStorePtr->sequencesStart;
const SeqDef* const sequences = seqStorePtr->sequencesStart;
const BYTE* const ofCodeTable = seqStorePtr->ofCode;
const BYTE* const llCodeTable = seqStorePtr->llCode;
const BYTE* const mlCodeTable = seqStorePtr->mlCode;
@ -1028,7 +1028,7 @@ static size_t writeSequences(U32* seed, frame_t* frame, seqStore_t* seqStorePtr,
static size_t writeSequencesBlock(U32* seed, frame_t* frame, size_t contentSize,
size_t literalsSize, dictInfo info)
{
seqStore_t seqStore;
SeqStore_t seqStore;
size_t numSequences;

666
tests/fullbench.c
View File

@ -12,14 +12,16 @@
/*_************************************
* Includes
**************************************/
#define _CRT_SECURE_NO_WARNINGS /* disable Visual warning that it doesn't like fopen() */
#define ZSTD_DISABLE_DEPRECATE_WARNINGS /* No deprecation warnings, we still bench some deprecated functions */
#include <limits.h>
#include "util.h" /* Compiler options, UTIL_GetFileSize */
#include <stdlib.h> /* malloc */
#include <stdio.h> /* fprintf, fopen, ftello64 */
#include <assert.h>
#include "timefn.h" /* UTIL_clockSpanNano, UTIL_getTime */
#include "mem.h" /* U32 */
#include "compress/zstd_compress_internal.h"
#ifndef ZSTD_DLL_IMPORT
#include "zstd_internal.h" /* ZSTD_decodeSeqHeaders, ZSTD_blockHeaderSize, ZSTD_getcBlockSize, blockType_e, KB, MB */
#include "decompress/zstd_decompress_internal.h" /* ZSTD_DCtx struct */
@ -33,10 +35,10 @@
#include "zstd.h" /* ZSTD_versionString */
#include "util.h" /* time functions */
#include "datagen.h"
#include "lorem.h"
#include "benchfn.h" /* CustomBench */
#include "benchzstd.h" /* MB_UNIT */
/*_************************************
* Constants
**************************************/
@ -51,7 +53,7 @@
#define DEFAULT_CLEVEL 1
#define COMPRESSIBILITY_DEFAULT 0.50
#define COMPRESSIBILITY_DEFAULT (-1.0)
static const size_t kSampleSizeDefault = 10000000;
#define TIMELOOP_NANOSEC (1*1000000000ULL) /* 1 second */
@ -64,6 +66,7 @@ static const size_t kSampleSizeDefault = 10000000;
#define CONTROL(c) { if (!(c)) { abort(); } } /* like assert(), but cannot be disabled */
/*_************************************
* Benchmark Parameters
**************************************/
@ -97,7 +100,6 @@ static size_t BMK_findMaxMem(U64 requiredMem)
*********************************************************/
static ZSTD_CCtx* g_zcc = NULL;
static size_t
local_ZSTD_compress(const void* src, size_t srcSize,
void* dst, size_t dstSize,
@ -129,40 +131,88 @@ local_ZSTD_compress_freshCCtx(const void* src, size_t srcSize,
}
}
static size_t g_cSize = 0;
typedef struct {
void* prepBuffer;
size_t prepSize;
void* dst;
size_t dstCapacity;
size_t fixedOrigSize; /* optional, 0 means "no modification" */
} PrepResult;
#define PREPRESULT_INIT { NULL, 0, NULL, 0, 0 }
static PrepResult prepDecompress(const void* src, size_t srcSize, int cLevel)
{
size_t prepCapacity = ZSTD_compressBound(srcSize);
void* prepBuffer = malloc(prepCapacity);
size_t cSize = ZSTD_compress(prepBuffer, prepCapacity, src, srcSize, cLevel);
void* dst = malloc(srcSize);
PrepResult r = PREPRESULT_INIT;
assert(dst != NULL);
r.prepBuffer = prepBuffer;
r.prepSize = cSize;
r.dst = dst;
r.dstCapacity = srcSize;
return r;
}
static size_t local_ZSTD_decompress(const void* src, size_t srcSize,
void* dst, size_t dstSize,
void* buff2)
void* unused)
{
(void)src; (void)srcSize;
return ZSTD_decompress(dst, dstSize, buff2, g_cSize);
(void)unused;
return ZSTD_decompress(dst, dstSize, src, srcSize);
}
static ZSTD_DCtx* g_zdc = NULL; /* will be initialized within benchMem */
static size_t local_ZSTD_decompressDCtx(const void* src, size_t srcSize,
void* dst, size_t dstSize,
void* buff2)
void* unused)
{
(void)src; (void)srcSize;
return ZSTD_decompressDCtx(g_zdc, dst, dstSize, buff2, g_cSize);
(void)unused;
return ZSTD_decompressDCtx(g_zdc, dst, dstSize, src, srcSize);
}
#ifndef ZSTD_DLL_IMPORT
static PrepResult prepLiterals(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
size_t dstCapacity = srcSize;
void* dst = malloc(dstCapacity);
void* prepBuffer;
size_t prepSize = ZSTD_compress(dst, dstCapacity, src, srcSize, cLevel);
size_t frameHeaderSize = ZSTD_frameHeaderSize(dst, ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
CONTROL(!ZSTD_isError(frameHeaderSize));
/* check block is compressible, hence contains a literals section */
{ blockProperties_t bp;
ZSTD_getcBlockSize((char*)dst+frameHeaderSize, dstCapacity, &bp); /* Get 1st block type */
if (bp.blockType != bt_compressed) {
DISPLAY("no compressed literals\n");
return r;
} }
{ size_t const skippedSize = frameHeaderSize + ZSTD_blockHeaderSize;
prepSize -= skippedSize;
prepBuffer = malloc(prepSize);
CONTROL(prepBuffer != NULL);
memmove(prepBuffer, (char*)dst+skippedSize, prepSize);
}
ZSTD_decompressBegin(g_zdc);
r.prepBuffer = prepBuffer;
r.prepSize = prepSize;
r.dst = dst;
r.dstCapacity = dstCapacity;
r.fixedOrigSize = srcSize > 128 KB ? 128 KB : srcSize; /* speed relative to block */
return r;
}
extern size_t ZSTD_decodeLiteralsBlock_wrapper(ZSTD_DCtx* dctx,
const void* src, size_t srcSize,
void* dst, size_t dstCapacity);
static size_t local_ZSTD_decodeLiteralsBlock(const void* src, size_t srcSize, void* dst, size_t dstSize, void* buff2)
static size_t
local_ZSTD_decodeLiteralsBlock(const void* src, size_t srcSize, void* dst, size_t dstCapacity, void* unused)
{
(void)src; (void)srcSize; (void)dst; (void)dstSize;
return ZSTD_decodeLiteralsBlock_wrapper(g_zdc, buff2, g_cSize, dst, dstSize);
}
static size_t local_ZSTD_decodeSeqHeaders(const void* src, size_t srcSize, void* dst, size_t dstSize, void* buff2)
{
int nbSeq;
(void)src; (void)srcSize; (void)dst; (void)dstSize;
return ZSTD_decodeSeqHeaders(g_zdc, &nbSeq, buff2, g_cSize);
(void)unused;
return ZSTD_decodeLiteralsBlock_wrapper(g_zdc, src, srcSize, dst, dstCapacity);
}
FORCE_NOINLINE size_t ZSTD_decodeLiteralsHeader(ZSTD_DCtx* dctx, void const* src, size_t srcSize)
@ -170,7 +220,7 @@ FORCE_NOINLINE size_t ZSTD_decodeLiteralsHeader(ZSTD_DCtx* dctx, void const* src
RETURN_ERROR_IF(srcSize < MIN_CBLOCK_SIZE, corruption_detected, "");
{
BYTE const* istart = (BYTE const*)src;
symbolEncodingType_e const litEncType = (symbolEncodingType_e)(istart[0] & 3);
SymbolEncodingType_e const litEncType = (SymbolEncodingType_e)(istart[0] & 3);
if (litEncType == set_compressed) {
RETURN_ERROR_IF(srcSize < 5, corruption_detected, "srcSize >= MIN_CBLOCK_SIZE == 3; here we need up to 5 for case 3");
{
@ -219,11 +269,59 @@ FORCE_NOINLINE size_t ZSTD_decodeLiteralsHeader(ZSTD_DCtx* dctx, void const* src
return 0;
}
static size_t local_ZSTD_decodeLiteralsHeader(const void* src, size_t srcSize, void* dst, size_t dstSize, void* buff2)
static size_t
local_ZSTD_decodeLiteralsHeader(const void* src, size_t srcSize, void* dst, size_t dstCapacity, void* unused)
{
(void)dst, (void)dstSize, (void)src, (void)srcSize;
return ZSTD_decodeLiteralsHeader(g_zdc, buff2, g_cSize);
(void)dst; (void)dstCapacity; (void)unused;
return ZSTD_decodeLiteralsHeader(g_zdc, src, srcSize);
}
static PrepResult prepSequences1stBlock(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
size_t const dstCapacity = srcSize;
void* dst = malloc(dstCapacity);
const BYTE* ip = dst;
const BYTE* iend;
{ size_t const cSize = ZSTD_compress(dst, dstCapacity, src, srcSize, cLevel);
CONTROL(cSize > ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
}
/* Skip frame Header */
{ size_t const frameHeaderSize = ZSTD_frameHeaderSize(dst, ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
CONTROL(!ZSTD_isError(frameHeaderSize));
ip += frameHeaderSize;
}
/* Find end of block */
{ blockProperties_t bp;
size_t const cBlockSize = ZSTD_getcBlockSize(ip, dstCapacity, &bp); /* Get 1st block type */
if (bp.blockType != bt_compressed) {
DISPLAY("no compressed sequences\n");
return r;
}
iend = ip + ZSTD_blockHeaderSize + cBlockSize; /* End of first block */
}
ip += ZSTD_blockHeaderSize; /* skip block header */
ZSTD_decompressBegin(g_zdc);
CONTROL(iend > ip);
ip += ZSTD_decodeLiteralsBlock_wrapper(g_zdc, ip, (size_t)(iend-ip), dst, dstCapacity); /* skip literal segment */
r.prepSize = (size_t)(iend-ip);
r.prepBuffer = malloc(r.prepSize);
CONTROL(r.prepBuffer != NULL);
memmove(r.prepBuffer, ip, r.prepSize); /* copy rest of block (it starts by SeqHeader) */
r.dst = dst;
r.dstCapacity = dstCapacity;
r.fixedOrigSize = srcSize > 128 KB ? 128 KB : srcSize; /* speed relative to block */
return r;
}
static size_t
local_ZSTD_decodeSeqHeaders(const void* src, size_t srcSize, void* dst, size_t dstCapacity, void* unused)
{
int nbSeq;
(void)unused; (void)dst; (void)dstCapacity;
return ZSTD_decodeSeqHeaders(g_zdc, &nbSeq, src, srcSize);
}
#endif
static ZSTD_CStream* g_cstream= NULL;
@ -346,23 +444,22 @@ static ZSTD_DStream* g_dstream= NULL;
static size_t
local_ZSTD_decompressStream(const void* src, size_t srcSize,
void* dst, size_t dstCapacity,
void* buff2)
void* unused)
{
ZSTD_outBuffer buffOut;
ZSTD_inBuffer buffIn;
(void)src; (void)srcSize;
(void)unused;
ZSTD_initDStream(g_dstream);
buffOut.dst = dst;
buffOut.size = dstCapacity;
buffOut.pos = 0;
buffIn.src = buff2;
buffIn.size = g_cSize;
buffIn.src = src;
buffIn.size = srcSize;
buffIn.pos = 0;
ZSTD_decompressStream(g_dstream, &buffOut, &buffIn);
return buffOut.pos;
}
#ifndef ZSTD_DLL_IMPORT
static size_t local_ZSTD_compressContinue(const void* src, size_t srcSize,
void* dst, size_t dstCapacity,
void* payload)
@ -408,15 +505,15 @@ local_ZSTD_compressContinue_extDict(const void* src, size_t srcSize,
static size_t local_ZSTD_decompressContinue(const void* src, size_t srcSize,
void* dst, size_t dstCapacity,
void* buff2)
void* unused)
{
size_t regeneratedSize = 0;
const BYTE* ip = (const BYTE*)buff2;
const BYTE* const iend = ip + g_cSize;
const BYTE* ip = (const BYTE*)src;
const BYTE* const iend = ip + srcSize;
BYTE* op = (BYTE*)dst;
size_t remainingCapacity = dstCapacity;
(void)src; (void)srcSize; /* unused */
(void)unused;
ZSTD_decompressBegin(g_zdc);
while (ip < iend) {
size_t const iSize = ZSTD_nextSrcSizeToDecompress(g_zdc);
@ -429,99 +526,257 @@ static size_t local_ZSTD_decompressContinue(const void* src, size_t srcSize,
return regeneratedSize;
}
#endif
static PrepResult prepSequences(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
size_t const dstCapacity = ZSTD_compressBound(srcSize);
void* const dst = malloc(dstCapacity);
size_t const prepCapacity = dstCapacity * 4;
void* prepBuffer = malloc(prepCapacity);
void* sequencesStart = (char*)prepBuffer + 2*sizeof(unsigned);
ZSTD_Sequence* const seqs = sequencesStart;
size_t const seqsCapacity = prepCapacity / sizeof(ZSTD_Sequence);
size_t nbSeqs;
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_compressionLevel, cLevel);
nbSeqs = ZSTD_generateSequences(g_zcc, seqs, seqsCapacity, src, srcSize);
CONTROL(srcSize < UINT_MAX);
MEM_write32(prepBuffer, (U32)srcSize);
MEM_write32((char*)prepBuffer+4, (U32)nbSeqs);
memcpy(seqs + nbSeqs, src, srcSize);
r.prepBuffer = prepBuffer;
r.prepSize = 8 + sizeof(ZSTD_Sequence)*nbSeqs + srcSize;
r.dst = dst;
r.dstCapacity = dstCapacity;
return r;
}
static size_t local_compressSequences(const void* input, size_t inputSize,
void* dst, size_t dstCapacity,
void* payload)
{
const char* ip = input;
size_t srcSize = MEM_read32(ip);
size_t nbSeqs = MEM_read32(ip+=4);
const ZSTD_Sequence* seqs = (const ZSTD_Sequence*)(const void*)(ip+=4);
const void* src = (ip+=nbSeqs * sizeof(ZSTD_Sequence));
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_blockDelimiters, ZSTD_sf_explicitBlockDelimiters);
assert(8 + nbSeqs * sizeof(ZSTD_Sequence) + srcSize == inputSize); (void)inputSize;
(void)payload;
return ZSTD_compressSequences(g_zcc, dst, dstCapacity, seqs, nbSeqs, src, srcSize);
}
static PrepResult prepSequencesAndLiterals(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
size_t const dstCapacity = ZSTD_compressBound(srcSize);
void* const dst = malloc(dstCapacity);
size_t const prepCapacity = dstCapacity * 4;
void* prepBuffer = malloc(prepCapacity);
void* sequencesStart = (char*)prepBuffer + 3*sizeof(unsigned);
ZSTD_Sequence* const seqs = sequencesStart;
size_t const seqsCapacity = prepCapacity / sizeof(ZSTD_Sequence);
size_t nbSeqs;
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_compressionLevel, cLevel);
nbSeqs = ZSTD_generateSequences(g_zcc, seqs, seqsCapacity, src, srcSize);
CONTROL(srcSize < UINT_MAX);
MEM_write32(prepBuffer, (U32)srcSize);
MEM_write32((char*)prepBuffer+4, (U32)nbSeqs);
/* copy literals */
{ char* const litStart = (char*)(seqs + nbSeqs);
size_t nbLiterals = 0;
const char* ip = src;
size_t n;
for (n=0; n<nbSeqs; n++) {
size_t const litSize = seqs[n].litLength;
memcpy(litStart + nbLiterals, ip, litSize);
ip += litSize + seqs[n].matchLength;
nbLiterals += litSize;
}
MEM_write32((char*)prepBuffer+8, (U32)nbLiterals);
r.prepBuffer = prepBuffer;
r.prepSize = 12 + sizeof(ZSTD_Sequence)*nbSeqs + nbLiterals;
r.dst = dst;
r.dstCapacity = dstCapacity;
}
return r;
}
static size_t
local_compressSequencesAndLiterals(const void* input, size_t inputSize,
void* dst, size_t dstCapacity,
void* payload)
{
const char* ip = input;
size_t decompressedSize = MEM_read32(ip);
size_t nbSeqs = MEM_read32(ip+=4);
size_t nbLiterals = MEM_read32(ip+=4);
const ZSTD_Sequence* seqs = (const ZSTD_Sequence*)(const void*)(ip+=4);
const void* literals = (ip+=nbSeqs * sizeof(ZSTD_Sequence));
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_blockDelimiters, ZSTD_sf_explicitBlockDelimiters);
# if 0 /* for tests */
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_repcodeResolution, ZSTD_ps_enable);
#endif
assert(12 + nbSeqs * sizeof(ZSTD_Sequence) + nbLiterals == inputSize); (void)inputSize;
(void)payload;
return ZSTD_compressSequencesAndLiterals(g_zcc, dst, dstCapacity, seqs, nbSeqs, literals, nbLiterals, nbLiterals + 8, decompressedSize);
}
static PrepResult prepConvertSequences(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
size_t const prepCapacity = srcSize * 4;
void* prepBuffer = malloc(prepCapacity);
void* sequencesStart = (char*)prepBuffer + 2*sizeof(unsigned);
ZSTD_Sequence* const seqs = sequencesStart;
size_t const seqsCapacity = prepCapacity / sizeof(ZSTD_Sequence);
size_t totalNbSeqs, nbSeqs, blockSize=0;
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_compressionLevel, cLevel);
totalNbSeqs = ZSTD_generateSequences(g_zcc, seqs, seqsCapacity, src, srcSize);
CONTROL(!ZSTD_isError(totalNbSeqs));
/* find nb sequences in first block */
{ size_t n;
for (n=0; n<totalNbSeqs; n++) {
if (seqs[n].matchLength == 0) break;
blockSize += seqs[n].litLength + seqs[n].matchLength;
}
blockSize += seqs[n].litLength;
nbSeqs = n+1;
#if 0
printf("found %zu sequences representing a first block of size %zu\n", nbSeqs, blockSize);
#endif
}
/* generate benchmarked input */
CONTROL(blockSize < UINT_MAX);
MEM_write32(prepBuffer, (U32)blockSize);
MEM_write32((char*)prepBuffer+4, (U32)nbSeqs);
memcpy(seqs + nbSeqs, src, srcSize);
r.prepBuffer = prepBuffer;
r.prepSize = 8 + sizeof(ZSTD_Sequence) * nbSeqs;
r.fixedOrigSize = blockSize;
return r;
}
static size_t
local_convertSequences(const void* input, size_t inputSize,
void* dst, size_t dstCapacity,
void* payload)
{
const char* ip = input;
size_t const blockSize = MEM_read32(ip);
size_t const nbSeqs = MEM_read32(ip+=4);
const ZSTD_Sequence* seqs = (const ZSTD_Sequence*)(const void*)(ip+=4);
ZSTD_CCtx_reset(g_zcc, ZSTD_reset_session_and_parameters);
ZSTD_resetSeqStore(&g_zcc->seqStore);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_blockDelimiters, ZSTD_sf_explicitBlockDelimiters);
# if 0 /* for tests */
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_repcodeResolution, ZSTD_ps_enable);
#endif
assert(8 + nbSeqs * sizeof(ZSTD_Sequence) == inputSize); (void)inputSize;
(void)dst; (void)dstCapacity;
(void)payload; (void)blockSize;
(void)ZSTD_convertBlockSequences(g_zcc, seqs, nbSeqs, 0);
return nbSeqs;
}
static PrepResult prepCopy(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = PREPRESULT_INIT;
(void)cLevel;
r.prepSize = srcSize;
r.prepBuffer = malloc(srcSize);
CONTROL(r.prepBuffer != NULL);
memcpy(r.prepBuffer, src, srcSize);
r.dstCapacity = ZSTD_compressBound(srcSize);
r.dst = malloc(r.dstCapacity);
CONTROL(r.dst != NULL);
return r;
}
static PrepResult prepShorterDstCapacity(const void* src, size_t srcSize, int cLevel)
{
PrepResult r = prepCopy(src, srcSize, cLevel);
assert(r.dstCapacity > 1);
r.dstCapacity -= 1;
return r;
}
/*_*******************************************************
* Bench functions
* List of Scenarios
*********************************************************/
static int benchMem(unsigned benchNb,
const void* src, size_t srcSize,
/* if PrepFunction_f returns 0, benchmarking is cancelled */
typedef PrepResult (*PrepFunction_f)(const void* src, size_t srcSize, int cLevel);
typedef size_t (*BenchedFunction_f)(const void* src, size_t srcSize, void* dst, size_t dstSize, void* opaque);
typedef struct {
const char* name;
PrepFunction_f preparation_f;
BenchedFunction_f benchedFunction;
} BenchScenario;
static BenchScenario kScenarios[] = {
{ "compress", NULL, local_ZSTD_compress },
{ "decompress", prepDecompress, local_ZSTD_decompress },
{ "compress_freshCCtx", NULL, local_ZSTD_compress_freshCCtx },
{ "decompressDCtx", prepDecompress, local_ZSTD_decompressDCtx },
{ "compressContinue", NULL, local_ZSTD_compressContinue },
{ "compressContinue_extDict", NULL, local_ZSTD_compressContinue_extDict },
{ "decompressContinue", prepDecompress, local_ZSTD_decompressContinue },
{ "compressStream", NULL, local_ZSTD_compressStream },
{ "compressStream_freshCCtx", NULL, local_ZSTD_compressStream_freshCCtx },
{ "decompressStream", prepDecompress, local_ZSTD_decompressStream },
{ "compress2", NULL, local_ZSTD_compress2 },
{ "compressStream2, end", NULL, local_ZSTD_compressStream2_end },
{ "compressStream2, end & short", prepShorterDstCapacity, local_ZSTD_compressStream2_end },
{ "compressStream2, continue", NULL, local_ZSTD_compressStream2_continue },
{ "compressStream2, -T2, continue", NULL, local_ZSTD_compress_generic_T2_continue },
{ "compressStream2, -T2, end", NULL, local_ZSTD_compress_generic_T2_end },
{ "compressSequences", prepSequences, local_compressSequences },
{ "compressSequencesAndLiterals", prepSequencesAndLiterals, local_compressSequencesAndLiterals },
{ "convertSequences (1st block)", prepConvertSequences, local_convertSequences },
#ifndef ZSTD_DLL_IMPORT
{ "decodeLiteralsHeader (1st block)", prepLiterals, local_ZSTD_decodeLiteralsHeader },
{ "decodeLiteralsBlock (1st block)", prepLiterals, local_ZSTD_decodeLiteralsBlock },
{ "decodeSeqHeaders (1st block)", prepSequences1stBlock, local_ZSTD_decodeSeqHeaders },
#endif
};
#define NB_SCENARIOS (sizeof(kScenarios) / sizeof(kScenarios[0]))
/*_*******************************************************
* Bench loop
*********************************************************/
static int benchMem(unsigned scenarioID,
const void* origSrc, size_t origSrcSize,
int cLevel, ZSTD_compressionParameters cparams)
{
size_t dstBuffSize = ZSTD_compressBound(srcSize);
BYTE* dstBuff;
void* dstBuff2;
size_t dstCapacity = 0;
void* dst = NULL;
void* prepBuff = NULL;
size_t prepBuffSize = 0;
void* payload;
const char* benchName;
BMK_benchFn_t benchFunction;
PrepFunction_f prep_f;
int errorcode = 0;
/* Selection */
switch(benchNb)
{
case 1:
benchFunction = local_ZSTD_compress; benchName = "compress";
break;
case 2:
benchFunction = local_ZSTD_decompress; benchName = "decompress";
break;
case 3:
benchFunction = local_ZSTD_compress_freshCCtx; benchName = "compress_freshCCtx";
break;
case 4:
benchFunction = local_ZSTD_decompressDCtx; benchName = "decompressDCtx";
break;
#ifndef ZSTD_DLL_IMPORT
case 11:
benchFunction = local_ZSTD_compressContinue; benchName = "compressContinue";
break;
case 12:
benchFunction = local_ZSTD_compressContinue_extDict; benchName = "compressContinue_extDict";
break;
case 13:
benchFunction = local_ZSTD_decompressContinue; benchName = "decompressContinue";
break;
case 30:
benchFunction = local_ZSTD_decodeLiteralsHeader; benchName = "decodeLiteralsHeader";
break;
case 31:
benchFunction = local_ZSTD_decodeLiteralsBlock; benchName = "decodeLiteralsBlock";
break;
case 32:
benchFunction = local_ZSTD_decodeSeqHeaders; benchName = "decodeSeqHeaders";
break;
#endif
case 41:
benchFunction = local_ZSTD_compressStream; benchName = "compressStream";
break;
case 42:
benchFunction = local_ZSTD_decompressStream; benchName = "decompressStream";
break;
case 43:
benchFunction = local_ZSTD_compressStream_freshCCtx; benchName = "compressStream_freshCCtx";
break;
case 50:
benchFunction = local_ZSTD_compress2; benchName = "compress2";
break;
case 51:
benchFunction = local_ZSTD_compressStream2_end; benchName = "compressStream2, end";
break;
case 52:
benchFunction = local_ZSTD_compressStream2_end; benchName = "compressStream2, end & short";
break;
case 53:
benchFunction = local_ZSTD_compressStream2_continue; benchName = "compressStream2, continue";
break;
case 61:
benchFunction = local_ZSTD_compress_generic_T2_continue; benchName = "compress_generic, -T2, continue";
break;
case 62:
benchFunction = local_ZSTD_compress_generic_T2_end; benchName = "compress_generic, -T2, end";
break;
default :
return 0;
}
if (scenarioID >= NB_SCENARIOS) return 0; /* scenario doesn't exist */
/* Allocation */
dstBuff = (BYTE*)malloc(dstBuffSize);
dstBuff2 = malloc(dstBuffSize);
if ((!dstBuff) || (!dstBuff2)) {
DISPLAY("\nError: not enough memory!\n");
free(dstBuff); free(dstBuff2);
return 12;
}
payload = dstBuff2;
benchName = kScenarios[scenarioID].name;
benchFunction = kScenarios[scenarioID].benchedFunction;
prep_f = kScenarios[scenarioID].preparation_f;
if (prep_f == NULL) prep_f = prepCopy; /* default */
/* Initialization */
if (g_zcc==NULL) g_zcc = ZSTD_createCCtx();
if (g_zdc==NULL) g_zdc = ZSTD_createDCtx();
if (g_cstream==NULL) g_cstream = ZSTD_createCStream();
@ -538,7 +793,7 @@ static int benchMem(unsigned benchNb,
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_searchLog, (int)cparams.searchLog);
ZSTD_CCtx_setParameter(g_zcc, ZSTD_c_minMatch, (int)cparams.minMatch);
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_zcc, ZSTD_c_strategy, (int)cparams.strategy);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_compressionLevel, cLevel);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_windowLog, (int)cparams.windowLog);
@ -547,114 +802,26 @@ static int benchMem(unsigned benchNb,
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_searchLog, (int)cparams.searchLog);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_minMatch, (int)cparams.minMatch);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_targetLength, (int)cparams.targetLength);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_strategy, cparams.strategy);
ZSTD_CCtx_setParameter(g_cstream, ZSTD_c_strategy, (int)cparams.strategy);
/* Preparation */
switch(benchNb)
{
case 1:
payload = &cparams;
break;
case 2:
g_cSize = ZSTD_compress(dstBuff2, dstBuffSize, src, srcSize, cLevel);
break;
case 3:
payload = &cparams;
break;
case 4:
g_cSize = ZSTD_compress(dstBuff2, dstBuffSize, src, srcSize, cLevel);
break;
#ifndef ZSTD_DLL_IMPORT
case 11:
payload = &cparams;
break;
case 12:
payload = &cparams;
break;
case 13 :
g_cSize = ZSTD_compress(dstBuff2, dstBuffSize, src, srcSize, cLevel);
break;
case 30: /* ZSTD_decodeLiteralsHeader */
/* fall-through */
case 31: /* ZSTD_decodeLiteralsBlock : starts literals block in dstBuff2 */
{ size_t frameHeaderSize;
g_cSize = ZSTD_compress(dstBuff, dstBuffSize, src, srcSize, cLevel);
frameHeaderSize = ZSTD_frameHeaderSize(dstBuff, ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
CONTROL(!ZSTD_isError(frameHeaderSize));
/* check block is compressible, hence contains a literals section */
{ blockProperties_t bp;
ZSTD_getcBlockSize(dstBuff+frameHeaderSize, dstBuffSize, &bp); /* Get 1st block type */
if (bp.blockType != bt_compressed) {
DISPLAY("ZSTD_decodeLiteralsBlock : impossible to test on this sample (not compressible)\n");
goto _cleanOut;
} }
{ size_t const skippedSize = frameHeaderSize + ZSTD_blockHeaderSize;
memcpy(dstBuff2, dstBuff+skippedSize, g_cSize-skippedSize);
}
srcSize = srcSize > 128 KB ? 128 KB : srcSize; /* speed relative to block */
ZSTD_decompressBegin(g_zdc);
break;
}
case 32: /* ZSTD_decodeSeqHeaders */
{ blockProperties_t bp;
const BYTE* ip = dstBuff;
const BYTE* iend;
{ size_t const cSize = ZSTD_compress(dstBuff, dstBuffSize, src, srcSize, cLevel);
CONTROL(cSize > ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
}
/* Skip frame Header */
{ size_t const frameHeaderSize = ZSTD_frameHeaderSize(dstBuff, ZSTD_FRAMEHEADERSIZE_PREFIX(ZSTD_f_zstd1));
CONTROL(!ZSTD_isError(frameHeaderSize));
ip += frameHeaderSize;
}
/* Find end of block */
{ size_t const cBlockSize = ZSTD_getcBlockSize(ip, dstBuffSize, &bp); /* Get 1st block type */
if (bp.blockType != bt_compressed) {
DISPLAY("ZSTD_decodeSeqHeaders : impossible to test on this sample (not compressible)\n");
goto _cleanOut;
}
iend = ip + ZSTD_blockHeaderSize + cBlockSize; /* End of first block */
}
ip += ZSTD_blockHeaderSize; /* skip block header */
ZSTD_decompressBegin(g_zdc);
CONTROL(iend > ip);
ip += ZSTD_decodeLiteralsBlock_wrapper(g_zdc, ip, (size_t)(iend-ip), dstBuff, dstBuffSize); /* skip literal segment */
g_cSize = (size_t)(iend-ip);
memcpy(dstBuff2, ip, g_cSize); /* copy rest of block (it starts by SeqHeader) */
srcSize = srcSize > 128 KB ? 128 KB : srcSize; /* speed relative to block */
break;
}
#else
case 31:
goto _cleanOut;
#endif
case 41 :
payload = &cparams;
break;
case 42 :
g_cSize = ZSTD_compress(payload, dstBuffSize, src, srcSize, cLevel);
break;
case 43 :
payload = &cparams;
break;
case 52 :
/* compressStream2, short dstCapacity */
dstBuffSize--;
break;
/* test functions */
/* convention: test functions have ID > 100 */
default : ;
payload = &cparams;
{ PrepResult pr = prep_f(origSrc, origSrcSize, cLevel);
dst = pr.dst;
dstCapacity = pr.dstCapacity;
prepBuff = pr.prepBuffer;
prepBuffSize = pr.prepSize;
if (pr.fixedOrigSize) origSrcSize = pr.fixedOrigSize;
}
if (prepBuffSize==0) goto _cleanOut; /* failed preparation */
/* warming up dstBuff */
{ size_t i; for (i=0; i<dstBuffSize; i++) dstBuff[i]=(BYTE)i; }
{ size_t i; for (i=0; i<dstCapacity; i++) ((BYTE*)dst)[i]=(BYTE)i; }
/* benchmark loop */
{ BMK_timedFnState_t* const tfs = BMK_createTimedFnState(g_nbIterations * 1000, 1000);
void* const avoidStrictAliasingPtr = &dstBuff;
void* const avoidStrictAliasingPtr = &dst;
const void* prepSrc = prepBuff;
BMK_benchParams_t bp;
BMK_runTime_t bestResult;
bestResult.sumOfReturn = 0;
@ -667,18 +834,18 @@ static int benchMem(unsigned benchNb,
bp.initPayload = NULL;
bp.errorFn = ZSTD_isError;
bp.blockCount = 1;
bp.srcBuffers = &src;
bp.srcSizes = &srcSize;
bp.srcBuffers = &prepSrc;
bp.srcSizes = &prepBuffSize;
bp.dstBuffers = (void* const*) avoidStrictAliasingPtr; /* circumvent strict aliasing warning on gcc-8,
* because gcc considers that `void* const *` and `void**` are 2 different types */
bp.dstCapacities = &dstBuffSize;
bp.dstCapacities = &dstCapacity;
bp.blockResults = NULL;
for (;;) {
BMK_runOutcome_t const bOutcome = BMK_benchTimedFn(tfs, bp);
if (!BMK_isSuccessful_runOutcome(bOutcome)) {
DISPLAY("ERROR benchmarking function ! ! \n");
DISPLAY("ERROR: Scenario %u: %s \n", scenarioID, ZSTD_getErrorName(BMK_extract_errorResult(bOutcome)));
errorcode = 1;
goto _cleanOut;
}
@ -686,9 +853,9 @@ static int benchMem(unsigned benchNb,
{ BMK_runTime_t const newResult = BMK_extract_runTime(bOutcome);
if (newResult.nanoSecPerRun < bestResult.nanoSecPerRun )
bestResult.nanoSecPerRun = newResult.nanoSecPerRun;
DISPLAY("\r%2u#%-29.29s:%8.1f MB/s (%8u) ",
benchNb, benchName,
(double)srcSize * TIMELOOP_NANOSEC / bestResult.nanoSecPerRun / MB_UNIT,
DISPLAY("\r%2u#%-31.31s:%8.1f MB/s (%8u) ",
scenarioID, benchName,
(double)origSrcSize * TIMELOOP_NANOSEC / bestResult.nanoSecPerRun / MB_UNIT,
(unsigned)newResult.sumOfReturn );
}
@ -699,8 +866,8 @@ static int benchMem(unsigned benchNb,
DISPLAY("\n");
_cleanOut:
free(dstBuff);
free(dstBuff2);
free(prepBuff);
free(dst);
ZSTD_freeCCtx(g_zcc); g_zcc=NULL;
ZSTD_freeDCtx(g_zdc); g_zdc=NULL;
ZSTD_freeCStream(g_cstream); g_cstream=NULL;
@ -709,7 +876,13 @@ _cleanOut:
}
static int benchSample(U32 benchNb,
#define BENCH_ALL_SCENARIOS 999
/*
* if @compressibility < 0.0, use Lorem Ipsum generator
* otherwise, @compressibility is expected to be between 0.0 and 1.0
* if scenarioID == BENCH_ALL_SCENARIOS, all scenarios will be run on the sample
*/
static int benchSample(U32 scenarioID,
size_t benchedSize, double compressibility,
int cLevel, ZSTD_compressionParameters cparams)
{
@ -718,24 +891,30 @@ static int benchSample(U32 benchNb,
if (!origBuff) { DISPLAY("\nError: not enough memory!\n"); return 12; }
/* Fill buffer */
RDG_genBuffer(origBuff, benchedSize, compressibility, 0.0, 0);
if (compressibility < 0.0) {
LOREM_genBuffer(origBuff, benchedSize, 0);
} else {
RDG_genBuffer(origBuff, benchedSize, compressibility, 0.0, 0);
}
/* bench */
DISPLAY("\r%70s\r", "");
DISPLAY(" Sample %u bytes : \n", (unsigned)benchedSize);
if (benchNb) {
benchMem(benchNb, origBuff, benchedSize, cLevel, cparams);
} else { /* 0 == run all tests */
for (benchNb=0; benchNb<100; benchNb++) {
benchMem(benchNb, origBuff, benchedSize, cLevel, cparams);
} }
if (scenarioID == BENCH_ALL_SCENARIOS) {
for (scenarioID=0; scenarioID<100; scenarioID++) {
benchMem(scenarioID, origBuff, benchedSize, cLevel, cparams);
}
} else {
benchMem(scenarioID, origBuff, benchedSize, cLevel, cparams);
}
free(origBuff);
return 0;
}
static int benchFiles(U32 benchNb,
static int benchFiles(U32 scenarioID,
const char** fileNamesTable, const int nbFiles,
int cLevel, ZSTD_compressionParameters cparams)
{
@ -781,13 +960,12 @@ static int benchFiles(U32 benchNb,
/* bench */
DISPLAY("\r%70s\r", ""); /* blank line */
DISPLAY(" %s : \n", inFileName);
if (benchNb) {
benchMem(benchNb, origBuff, benchedSize, cLevel, cparams);
} else {
for (benchNb=0; benchNb<100; benchNb++) {
benchMem(benchNb, origBuff, benchedSize, cLevel, cparams);
if (scenarioID == BENCH_ALL_SCENARIOS) {
for (scenarioID=0; scenarioID<100; scenarioID++) {
benchMem(scenarioID, origBuff, benchedSize, cLevel, cparams);
}
benchNb = 0;
} else {
benchMem(scenarioID, origBuff, benchedSize, cLevel, cparams);
}
free(origBuff);
@ -877,7 +1055,7 @@ int main(int argc, const char** argv)
int argNb, filenamesStart=0, result;
const char* const exename = argv[0];
const char* input_filename = NULL;
U32 benchNb = 0, main_pause = 0;
U32 scenarioID = BENCH_ALL_SCENARIOS, main_pause = 0;
int cLevel = DEFAULT_CLEVEL;
ZSTD_compressionParameters cparams = ZSTD_getCParams(cLevel, 0, 0);
size_t sampleSize = kSampleSizeDefault;
@ -928,7 +1106,7 @@ int main(int argc, const char** argv)
/* Select specific algorithm to bench */
case 'b':
argument++;
benchNb = readU32FromChar(&argument);
scenarioID = readU32FromChar(&argument);
break;
/* Select compression level to use */
@ -970,9 +1148,9 @@ int main(int argc, const char** argv)
if (filenamesStart==0) /* no input file */
result = benchSample(benchNb, sampleSize, compressibility, cLevel, cparams);
result = benchSample(scenarioID, sampleSize, compressibility, cLevel, cparams);
else
result = benchFiles(benchNb, argv+filenamesStart, argc-filenamesStart, cLevel, cparams);
result = benchFiles(scenarioID, argv+filenamesStart, argc-filenamesStart, cLevel, cparams);
if (main_pause) { int unused; printf("press enter...\n"); unused = getchar(); (void)unused; }

13
tests/fuzz/Makefile
View File

@ -28,6 +28,9 @@ LIBZSTD_MK_DIR = ../../lib
DEBUGLEVEL ?= 2
ZSTD_LEGACY_SUPPORT ?= 1
.PHONY: default
default: all
include $(LIBZSTD_MK_DIR)/libzstd.mk
PRGDIR = ../../programs
@ -101,10 +104,6 @@ FUZZ_RT_OBJ9 := $(FUZZ_RT_OBJ8:.c=.o)
FUZZ_RT_OBJ10 := $(THIRD_PARTY_SEQ_PROD_OBJ) $(FUZZ_RT_OBJ9)
FUZZ_ROUND_TRIP_OBJ := $(FUZZ_RT_OBJ10:.S=.o)
.PHONY: default all clean cleanall
default: all
FUZZ_TARGETS := \
simple_round_trip \
stream_round_trip \
@ -128,6 +127,7 @@ FUZZ_TARGETS := \
decompress_cross_format \
generate_sequences
.PHONY: all clean cleanall
all: libregression.a $(FUZZ_TARGETS)
rt_lib_common_%.o: $(LIB_SRCDIR)/common/%.c
@ -263,9 +263,10 @@ corpora: $(patsubst %,corpora/%,$(FUZZ_TARGETS))
.PHONY: seedcorpora
seedcorpora: $(patsubst %,corpora/%_seed_corpus.zip,$(FUZZ_TARGETS))
REGRESSION_TARGET ?= all
regressiontest: corpora
CC="$(CC)" CXX="$(CXX)" CFLAGS="$(CFLAGS)" CXXFLAGS="$(CXXFLAGS)" LDFLAGS="$(LDFLAGS)" $(PYTHON) ./fuzz.py build all --debug=$(DEBUGLEVEL)
$(PYTHON) ./fuzz.py regression all
CC="$(CC)" CXX="$(CXX)" CFLAGS="$(CFLAGS)" CXXFLAGS="$(CXXFLAGS)" LDFLAGS="$(LDFLAGS)" $(PYTHON) ./fuzz.py build $(REGRESSION_TARGET) --debug=$(DEBUGLEVEL)
$(PYTHON) ./fuzz.py regression $(REGRESSION_TARGET)
clean:
@$(RM) *.a *.o $(FUZZ_TARGETS)

6
tests/fuzz/dictionary_loader.c
View File

@ -77,9 +77,9 @@ int LLVMFuzzerTestOneInput(const uint8_t *src, size_t size)
FUZZ_dataProducer_uint32Range(producer, 0, 2);
size = FUZZ_dataProducer_remainingBytes(producer);
DEBUGLOG(2, "Dict load method %d", dlm);
DEBUGLOG(2, "Dict content type %d", dct);
DEBUGLOG(2, "Dict size %u", (unsigned)size);
DEBUGLOG(4, "Dict load method %d", dlm);
DEBUGLOG(4, "Dict content type %d", dct);
DEBUGLOG(4, "Dict size %u", (unsigned)size);
void* const rBuf = FUZZ_malloc(size);
size_t const cBufSize = ZSTD_compressBound(size);

99
tests/fuzz/sequence_compression_api.c
View File

@ -16,6 +16,7 @@
*/
#define ZSTD_STATIC_LINKING_ONLY
#include "zstd_errors.h"
#include <stddef.h>
#include <stdlib.h>
@ -76,7 +77,7 @@ static char* generatePseudoRandomString(char* str, size_t size, FUZZ_dataProduce
static size_t decodeSequences(void* dst, size_t nbSequences,
size_t literalsSize,
const void* dict, size_t dictSize,
ZSTD_sequenceFormat_e mode)
ZSTD_SequenceFormat_e mode)
{
const uint8_t* litPtr = literalsBuffer;
const uint8_t* const litBegin = literalsBuffer;
@ -141,7 +142,7 @@ static size_t decodeSequences(void* dst, size_t nbSequences,
*/
static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
size_t literalsSizeLimit, size_t dictSize,
size_t windowLog, ZSTD_sequenceFormat_e mode)
size_t windowLog, ZSTD_SequenceFormat_e mode)
{
const uint32_t repCode = 0; /* not used by sequence ingestion api */
size_t windowSize = 1ULL << windowLog;
@ -155,7 +156,7 @@ static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
if (mode == ZSTD_sf_explicitBlockDelimiters) {
/* ensure that no sequence can be larger than one block */
literalsSizeLimit = MIN(literalsSizeLimit, blockSizeMax/2);
matchLengthMax = MIN(matchLengthMax, blockSizeMax/2);
matchLengthMax = MIN(matchLengthMax, (uint32_t)blockSizeMax/2);
}
while ( nbSeqGenerated < ZSTD_FUZZ_MAX_NBSEQ - 3 /* extra room for explicit delimiters */
@ -171,7 +172,7 @@ static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
if (bytesGenerated > ZSTD_FUZZ_GENERATED_SRC_MAXSIZE) {
break;
}
offsetBound = (bytesGenerated > windowSize) ? windowSize : bytesGenerated + (uint32_t)dictSize;
offsetBound = (bytesGenerated > windowSize) ? (uint32_t)windowSize : bytesGenerated + (uint32_t)dictSize;
offset = FUZZ_dataProducer_uint32Range(producer, 1, offsetBound);
if (dictSize > 0 && bytesGenerated <= windowSize) {
/* Prevent match length from being such that it would be associated with an offset too large
@ -180,7 +181,7 @@ static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
*/
const size_t bytesToReachWindowSize = windowSize - bytesGenerated;
if (bytesToReachWindowSize < ZSTD_MINMATCH_MIN) {
const uint32_t newOffsetBound = offsetBound > windowSize ? windowSize : offsetBound;
const uint32_t newOffsetBound = offsetBound > windowSize ? (uint32_t)windowSize : offsetBound;
offset = FUZZ_dataProducer_uint32Range(producer, 1, newOffsetBound);
} else {
matchBound = MIN(matchLengthMax, (uint32_t)bytesToReachWindowSize);
@ -201,14 +202,14 @@ static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
if (blockSize + seqSize > blockSizeMax) { /* reaching limit : must end block now */
const ZSTD_Sequence endBlock = {0, 0, 0, 0};
generatedSequences[nbSeqGenerated++] = endBlock;
blockSize = seqSize;
blockSize = (uint32_t)seqSize;
}
if (split) {
const ZSTD_Sequence endBlock = {0, lastLits, 0, 0};
generatedSequences[nbSeqGenerated++] = endBlock;
assert(lastLits <= seq.litLength);
seq.litLength -= lastLits;
blockSize = seqSize - lastLits;
blockSize = (uint32_t)(seqSize - lastLits);
} else {
blockSize += seqSize;
}
@ -227,12 +228,73 @@ static size_t generateRandomSequences(FUZZ_dataProducer_t* producer,
return nbSeqGenerated;
}
static size_t
transferLiterals(void* dst, size_t dstCapacity, const ZSTD_Sequence* seqs, size_t nbSeqs, const void* src, size_t srcSize)
{
size_t n;
char* op = dst;
char* const oend = op + dstCapacity;
const char* ip = src;
const char* const iend = ip + srcSize;
for (n=0; n<nbSeqs; n++) {
size_t litLen = seqs[n].litLength;
size_t mlen = seqs[n].matchLength;
assert(op + litLen < oend); (void)oend;
assert(ip + litLen + mlen <= iend); (void)iend;
memcpy(op, ip, litLen);
op += litLen;
ip += litLen + mlen;
}
assert(oend - op >= 8);
return (size_t)(op - (char*)dst);
}
static size_t roundTripTest_compressSequencesAndLiterals(
void* result, size_t resultCapacity,
void* compressed, size_t compressedCapacity,
const void* src, size_t srcSize,
const ZSTD_Sequence* seqs, size_t nbSeqs)
{
size_t const litCapacity = srcSize + 8;
void* literals = malloc(litCapacity);
size_t cSize, litSize;
assert(literals);
litSize = transferLiterals(literals, litCapacity, seqs, nbSeqs, src, srcSize);
cSize = ZSTD_compressSequencesAndLiterals(cctx,
compressed, compressedCapacity,
seqs, nbSeqs,
literals, litSize, litCapacity, srcSize);
free(literals);
if (ZSTD_getErrorCode(cSize) == ZSTD_error_cannotProduce_uncompressedBlock) {
/* Valid scenario : ZSTD_compressSequencesAndLiterals cannot generate uncompressed blocks */
return 0;
}
if (ZSTD_getErrorCode(cSize) == ZSTD_error_dstSize_tooSmall) {
/* Valid scenario : in explicit delimiter mode,
* it might be possible for the compressed size to outgrow dstCapacity.
* In which case, it's still a valid fuzzer scenario,
* but no roundtrip shall be possible */
return 0;
}
/* round-trip */
FUZZ_ZASSERT(cSize);
{ size_t const dSize = ZSTD_decompressDCtx(dctx, result, resultCapacity, compressed, cSize);
FUZZ_ZASSERT(dSize);
FUZZ_ASSERT_MSG(dSize == srcSize, "Incorrect regenerated size");
FUZZ_ASSERT_MSG(!FUZZ_memcmp(src, result, srcSize), "Corruption!");
return dSize;
}
}
static size_t roundTripTest(void* result, size_t resultCapacity,
void* compressed, size_t compressedCapacity,
const void* src, size_t srcSize,
const ZSTD_Sequence* seqs, size_t seqSize,
const ZSTD_Sequence* seqs, size_t nbSeqs,
unsigned hasDict,
ZSTD_sequenceFormat_e mode)
ZSTD_SequenceFormat_e mode)
{
size_t cSize;
size_t dSize;
@ -242,8 +304,17 @@ static size_t roundTripTest(void* result, size_t resultCapacity,
FUZZ_ZASSERT(ZSTD_DCtx_refDDict(dctx, ddict));
}
{ int blockMode, validation;
/* compressSequencesAndLiterals() only supports explicitBlockDelimiters and no validation */
FUZZ_ZASSERT(ZSTD_CCtx_getParameter(cctx, ZSTD_c_blockDelimiters, &blockMode));
FUZZ_ZASSERT(ZSTD_CCtx_getParameter(cctx, ZSTD_c_validateSequences, &validation));
if ((blockMode == ZSTD_sf_explicitBlockDelimiters) && (!validation)) {
FUZZ_ZASSERT(roundTripTest_compressSequencesAndLiterals(result, resultCapacity, compressed, compressedCapacity, src, srcSize, seqs, nbSeqs));
}
}
cSize = ZSTD_compressSequences(cctx, compressed, compressedCapacity,
seqs, seqSize,
seqs, nbSeqs,
src, srcSize);
if ( (ZSTD_getErrorCode(cSize) == ZSTD_error_dstSize_tooSmall)
&& (mode == ZSTD_sf_explicitBlockDelimiters) ) {
@ -276,7 +347,7 @@ int LLVMFuzzerTestOneInput(const uint8_t* src, size_t size)
unsigned hasDict;
unsigned wLog;
int cLevel;
ZSTD_sequenceFormat_e mode;
ZSTD_SequenceFormat_e mode;
FUZZ_dataProducer_t* const producer = FUZZ_dataProducer_create(src, size);
FUZZ_ASSERT(producer);
@ -293,15 +364,15 @@ int LLVMFuzzerTestOneInput(const uint8_t* src, size_t size)
/* Generate window log first so we don't generate offsets too large */
wLog = FUZZ_dataProducer_uint32Range(producer, ZSTD_WINDOWLOG_MIN, ZSTD_WINDOWLOG_MAX);
cLevel = FUZZ_dataProducer_int32Range(producer, -3, 22);
mode = (ZSTD_sequenceFormat_e)FUZZ_dataProducer_int32Range(producer, 0, 1);
mode = (ZSTD_SequenceFormat_e)FUZZ_dataProducer_int32Range(producer, 0, 1);
ZSTD_CCtx_reset(cctx, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_nbWorkers, 0);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_compressionLevel, cLevel);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_windowLog, wLog);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_windowLog, (int)wLog);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_minMatch, ZSTD_MINMATCH_MIN);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_validateSequences, 1);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_blockDelimiters, mode);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_blockDelimiters, (int)mode);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_forceAttachDict, ZSTD_dictForceAttach);
if (!literalsBuffer) {

2
tests/fuzz/zstd_helpers.c
View File

@ -146,7 +146,7 @@ void FUZZ_setRandomParameters(ZSTD_CCtx *cctx, size_t srcSize, FUZZ_dataProducer
setRand(cctx, ZSTD_c_prefetchCDictTables, 0, 2, producer);
setRand(cctx, ZSTD_c_maxBlockSize, ZSTD_BLOCKSIZE_MAX_MIN, ZSTD_BLOCKSIZE_MAX, producer);
setRand(cctx, ZSTD_c_validateSequences, 0, 1, producer);
setRand(cctx, ZSTD_c_searchForExternalRepcodes, 0, 2, producer);
setRand(cctx, ZSTD_c_repcodeResolution, 0, 2, producer);
if (FUZZ_dataProducer_uint32Range(producer, 0, 1) == 0) {
setRand(cctx, ZSTD_c_srcSizeHint, ZSTD_SRCSIZEHINT_MIN, 2 * srcSize, producer);
}

136
tests/fuzzer.c
View File

@ -40,7 +40,6 @@
#include "datagen.h" /* RDG_genBuffer */
#define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */
#include "xxhash.h" /* XXH64 */
#include "util.h"
#include "timefn.h" /* SEC_TO_MICRO, UTIL_time_t, UTIL_TIME_INITIALIZER, UTIL_clockSpanMicro, UTIL_getTime */
/* must be included after util.h, due to ERROR macro redefinition issue on Visual Studio */
#include "zstd_internal.h" /* ZSTD_WORKSPACETOOLARGE_MAXDURATION, ZSTD_WORKSPACETOOLARGE_FACTOR, KB, MB */
@ -310,7 +309,7 @@ static int FUZ_mallocTests(unsigned seed, double compressibility, unsigned part)
#endif
static void FUZ_decodeSequences(BYTE* dst, ZSTD_Sequence* seqs, size_t seqsSize,
BYTE* src, size_t size, ZSTD_sequenceFormat_e format)
BYTE* src, size_t size, ZSTD_SequenceFormat_e format)
{
size_t i;
size_t j;
@ -339,6 +338,35 @@ static void FUZ_decodeSequences(BYTE* dst, ZSTD_Sequence* seqs, size_t seqsSize,
}
}
static size_t FUZ_getLitSize(const ZSTD_Sequence* seqs, size_t nbSeqs)
{
size_t n, litSize = 0;
assert(seqs != NULL);
for (n=0; n<nbSeqs; n++) {
litSize += seqs[n].litLength;
}
return litSize;
}
static void
FUZ_transferLiterals(void* dst, size_t dstCapacity,
const void* src, size_t srcSize,
const ZSTD_Sequence* seqs, size_t nbSeqs)
{
size_t n;
const char* ip = (const char*)src;
char* op = (char*)dst;
size_t const litSize = FUZ_getLitSize(seqs, nbSeqs);
assert(litSize <= dstCapacity);
for (n=0; n<nbSeqs; n++) {
size_t const ll = seqs[n].litLength;
memcpy(op, ip, ll);
op += ll;
ip += ll + seqs[n].matchLength;
}
assert((size_t)(ip - (const char*)src) == srcSize);
}
#ifdef ZSTD_MULTITHREAD
typedef struct {
@ -3808,7 +3836,7 @@ static int basicUnitTests(U32 const seed, double compressibility)
if (seqs == NULL) goto _output_error;
assert(cctx != NULL);
/* Populate src with random data */
/* Populate src with compressible random data */
RDG_genBuffer(CNBuffer, srcSize, compressibility, 0., seed);
/* Roundtrip Test with block delimiters generated by ZSTD_generateSequences() */
@ -3850,6 +3878,108 @@ static int basicUnitTests(U32 const seed, double compressibility)
}
DISPLAYLEVEL(3, "OK \n");
DISPLAYLEVEL(3, "test%3i : ZSTD_compressSequencesAndLiterals : ", testNb++);
{
const size_t srcSize = 497000;
const BYTE* const src = (BYTE*)CNBuffer;
BYTE* const dst = (BYTE*)compressedBuffer;
const size_t dstCapacity = ZSTD_compressBound(srcSize);
const size_t decompressSize = srcSize;
char* const decompressBuffer = (char*)malloc(decompressSize);
char* const litBuffer = (char*)malloc(decompressSize);
size_t compressedSize;
ZSTD_CCtx* const cctx = ZSTD_createCCtx();
ZSTD_Sequence* const seqs = (ZSTD_Sequence*)malloc(srcSize * sizeof(ZSTD_Sequence));
size_t nbSeqs;
if (litBuffer == NULL) goto _output_error;
if (decompressBuffer == NULL) goto _output_error;
if (seqs == NULL) goto _output_error;
assert(cctx != NULL);
/* Populate src with compressible random data */
RDG_genBuffer(CNBuffer, srcSize, compressibility, 0., seed);
/* Roundtrip Test using the AndLiterals() variant */
nbSeqs = ZSTD_generateSequences(cctx, seqs, srcSize, src, srcSize);
ZSTD_CCtx_reset(cctx, ZSTD_reset_session_and_parameters);
ZSTD_CCtx_setParameter(cctx, ZSTD_c_blockDelimiters, ZSTD_sf_explicitBlockDelimiters);
{ size_t const litSize = FUZ_getLitSize(seqs, nbSeqs);
FUZ_transferLiterals(litBuffer, decompressSize, CNBuffer, srcSize, seqs, nbSeqs);
/* not enough literals: must fail */
compressedSize = ZSTD_compressSequencesAndLiterals(cctx, dst, dstCapacity, seqs, nbSeqs, src, litSize-1, decompressSize, srcSize);
if (!ZSTD_isError(compressedSize)) {
DISPLAY("ZSTD_compressSequencesAndLiterals() should have failed: not enough literals provided\n");
goto _output_error;
}
/* too many literals: must fail */
compressedSize = ZSTD_compressSequencesAndLiterals(cctx, dst, dstCapacity, seqs, nbSeqs, src, litSize+1, decompressSize, srcSize);
if (!ZSTD_isError(compressedSize)) {
DISPLAY("ZSTD_compressSequencesAndLiterals() should have failed: too many literals provided\n");
goto _output_error;
}
/* srcSize too large: must fail */
compressedSize = ZSTD_compressSequencesAndLiterals(cctx, dst, dstCapacity, seqs, nbSeqs, litBuffer, litSize, decompressSize, srcSize+1);
if (!ZSTD_isError(compressedSize)) {
DISPLAY("ZSTD_compressSequencesAndLiterals() should have failed: srcSize is too large\n");
goto _output_error;
}
/* srcSize too small: must fail */
compressedSize = ZSTD_compressSequencesAndLiterals(cctx, dst, dstCapacity, seqs, nbSeqs, litBuffer, litSize, decompressSize, srcSize-1);
if (!ZSTD_isError(compressedSize)) {
DISPLAY("ZSTD_compressSequencesAndLiterals() should have failed: srcSize is too small\n");
goto _output_error;
}
/* correct amount of literals: should compress successfully */
compressedSize = ZSTD_compressSequencesAndLiterals(cctx, dst, dstCapacity, seqs, nbSeqs, litBuffer, litSize, decompressSize, srcSize);
if (ZSTD_isError(compressedSize)) {
DISPLAY("Error in ZSTD_compressSequencesAndLiterals()\n");
goto _output_error;
}
}
{ ZSTD_frameHeader zfh;
size_t const zfhStatus = ZSTD_getFrameHeader(&zfh, dst, compressedSize);
if (zfhStatus != 0) {
DISPLAY("Error reading frame header\n");
goto _output_error;
}
if (zfh.frameContentSize != srcSize) {
DISPLAY("Error: ZSTD_compressSequencesAndLiterals() did not report srcSize in the frame header\n");
goto _output_error;
}
if (zfh.windowSize > srcSize) {
DISPLAY("Error: ZSTD_compressSequencesAndLiterals() did not resized window size to smaller contentSize\n");
goto _output_error;
}
}
{ size_t const dSize = ZSTD_decompress(decompressBuffer, decompressSize, dst, compressedSize);
if (ZSTD_isError(dSize)) {
DISPLAY("Error during decompression of frame produced by ZSTD_compressSequencesAndLiterals()\n");
goto _output_error;
}
if (dSize != srcSize) {
DISPLAY("Error: decompression of frame produced by ZSTD_compressSequencesAndLiterals() has different size\n");
goto _output_error;
}
if (memcmp(decompressBuffer, src, srcSize)) {
DISPLAY("Error: decompression of frame produced by ZSTD_compressSequencesAndLiterals() produces a different content (of same size)\n");
goto _output_error;
}
}
ZSTD_freeCCtx(cctx);
free(litBuffer);
free(decompressBuffer);
free(seqs);
}
DISPLAYLEVEL(3, "OK \n");
/* Multiple blocks of zeros test */
#define LONGZEROSLENGTH 1000000 /* 1MB of zeros */
DISPLAYLEVEL(3, "test%3i : compress %u zeroes : ", testNb++, LONGZEROSLENGTH);