You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-commits@hadoop.apache.org by ll...@apache.org on 2013/08/05 23:10:54 UTC
svn commit: r1510734 [1/2] - in
/hadoop/common/trunk/hadoop-common-project/hadoop-common: ./ src/
src/main/java/org/apache/hadoop/fs/
src/main/java/org/apache/hadoop/io/compress/
src/main/java/org/apache/hadoop/io/compress/lz4/ src/main/native/ src/mai...
Author: llu
Date: Mon Aug 5 21:10:54 2013
New Revision: 1510734
URL: http://svn.apache.org/r1510734
Log:
HADOOP-9319. Update bundled LZ4 source to r99. (Binglin Chang via llu)
Added:
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc.c
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc.h
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc_encoder.h
Modified:
hadoop/common/trunk/hadoop-common-project/hadoop-common/CHANGES.txt
hadoop/common/trunk/hadoop-common-project/hadoop-common/LICENSE.txt
hadoop/common/trunk/hadoop-common-project/hadoop-common/pom.xml
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/CMakeLists.txt
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/Lz4Codec.java
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/lz4/Lz4Compressor.java
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj.filters
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c
hadoop/common/trunk/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/compress/TestCodec.java
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/CHANGES.txt
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/CHANGES.txt?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/CHANGES.txt (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/CHANGES.txt Mon Aug 5 21:10:54 2013
@@ -283,6 +283,8 @@ Release 2.3.0 - UNRELEASED
IMPROVEMENTS
+ HADOOP-9319. Update bundled LZ4 source to r99. (Binglin Chang via llu)
+
HADOOP-9241. DU refresh interval is not configurable (harsh)
HADOOP-9417. Support for symlink resolution in LocalFileSystem /
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/LICENSE.txt
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/LICENSE.txt?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/LICENSE.txt (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/LICENSE.txt Mon Aug 5 21:10:54 2013
@@ -252,24 +252,26 @@ in src/main/native/src/org/apache/hadoop
* BSD-style license that can be found in the LICENSE file.
*/
- For src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c:
+For src/main/native/src/org/apache/hadoop/io/compress/lz4/{lz4.h,lz4.c,
+lz4_encoder.h,lz4hc.h,lz4hc.c,lz4hc_encoder.h},
/*
LZ4 - Fast LZ compression algorithm
- Copyright (C) 2011, Yann Collet.
- BSD License
+ Header File
+ Copyright (C) 2011-2013, Yann Collet.
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
-
+
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
-
+
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
@@ -281,4 +283,8 @@ in src/main/native/src/org/apache/hadoop
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+ - LZ4 source repository : http://code.google.com/p/lz4/
*/
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/pom.xml
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/pom.xml?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/pom.xml (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/pom.xml Mon Aug 5 21:10:54 2013
@@ -456,7 +456,12 @@
<exclude>src/test/empty-file</exclude>
<exclude>src/test/all-tests</exclude>
<exclude>src/test/resources/kdc/ldif/users.ldif</exclude>
+ <exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h</exclude>
<exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c</exclude>
+ <exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h</exclude>
+ <exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc.h</exclude>
+ <exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc.c</exclude>
+ <exclude>src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4hc_encoder.h</exclude>
<exclude>src/test/java/org/apache/hadoop/fs/test-untar.tgz</exclude>
</excludes>
</configuration>
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/CMakeLists.txt
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/CMakeLists.txt?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/CMakeLists.txt (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/CMakeLists.txt Mon Aug 5 21:10:54 2013
@@ -170,6 +170,7 @@ add_dual_library(hadoop
${D}/io/compress/lz4/Lz4Compressor.c
${D}/io/compress/lz4/Lz4Decompressor.c
${D}/io/compress/lz4/lz4.c
+ ${D}/io/compress/lz4/lz4hc.c
${SNAPPY_SOURCE_FILES}
${D}/io/compress/zlib/ZlibCompressor.c
${D}/io/compress/zlib/ZlibDecompressor.c
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/CommonConfigurationKeys.java Mon Aug 5 21:10:54 2013
@@ -96,7 +96,7 @@ public class CommonConfigurationKeys ext
public static final int IO_COMPRESSION_CODEC_SNAPPY_BUFFERSIZE_DEFAULT =
256 * 1024;
- /** Internal buffer size for Snappy compressor/decompressors */
+ /** Internal buffer size for Lz4 compressor/decompressors */
public static final String IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_KEY =
"io.compression.codec.lz4.buffersize";
@@ -104,6 +104,14 @@ public class CommonConfigurationKeys ext
public static final int IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_DEFAULT =
256 * 1024;
+ /** Use lz4hc(slow but with high compression ratio) for lz4 compression */
+ public static final String IO_COMPRESSION_CODEC_LZ4_USELZ4HC_KEY =
+ "io.compression.codec.lz4.use.lz4hc";
+
+ /** Default value for IO_COMPRESSION_CODEC_USELZ4HC_KEY */
+ public static final boolean IO_COMPRESSION_CODEC_LZ4_USELZ4HC_DEFAULT =
+ false;
+
/**
* Service Authorization
*/
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/Lz4Codec.java
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/Lz4Codec.java?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/Lz4Codec.java (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/Lz4Codec.java Mon Aug 5 21:10:54 2013
@@ -107,7 +107,7 @@ public class Lz4Codec implements Configu
CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_KEY,
CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_DEFAULT);
- int compressionOverhead = Math.max((int)(bufferSize * 0.01), 10);
+ int compressionOverhead = bufferSize/255 + 16;
return new BlockCompressorStream(out, compressor, bufferSize,
compressionOverhead);
@@ -140,7 +140,10 @@ public class Lz4Codec implements Configu
int bufferSize = conf.getInt(
CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_KEY,
CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_BUFFERSIZE_DEFAULT);
- return new Lz4Compressor(bufferSize);
+ boolean useLz4HC = conf.getBoolean(
+ CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_USELZ4HC_KEY,
+ CommonConfigurationKeys.IO_COMPRESSION_CODEC_LZ4_USELZ4HC_DEFAULT);
+ return new Lz4Compressor(bufferSize, useLz4HC);
}
/**
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/lz4/Lz4Compressor.java
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/lz4/Lz4Compressor.java?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/lz4/Lz4Compressor.java (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/io/compress/lz4/Lz4Compressor.java Mon Aug 5 21:10:54 2013
@@ -52,6 +52,7 @@ public class Lz4Compressor implements Co
private long bytesRead = 0L;
private long bytesWritten = 0L;
+ private final boolean useLz4HC;
static {
if (NativeCodeLoader.isNativeCodeLoaded()) {
@@ -72,8 +73,11 @@ public class Lz4Compressor implements Co
* Creates a new compressor.
*
* @param directBufferSize size of the direct buffer to be used.
+ * @param useLz4HC use high compression ratio version of lz4,
+ * which trades CPU for compression ratio.
*/
- public Lz4Compressor(int directBufferSize) {
+ public Lz4Compressor(int directBufferSize, boolean useLz4HC) {
+ this.useLz4HC = useLz4HC;
this.directBufferSize = directBufferSize;
uncompressedDirectBuf = ByteBuffer.allocateDirect(directBufferSize);
@@ -82,6 +86,15 @@ public class Lz4Compressor implements Co
}
/**
+ * Creates a new compressor.
+ *
+ * @param directBufferSize size of the direct buffer to be used.
+ */
+ public Lz4Compressor(int directBufferSize) {
+ this(directBufferSize, false);
+ }
+
+ /**
* Creates a new compressor with the default buffer size.
*/
public Lz4Compressor() {
@@ -227,7 +240,7 @@ public class Lz4Compressor implements Co
}
// Compress data
- n = compressBytesDirect();
+ n = useLz4HC ? compressBytesDirectHC() : compressBytesDirect();
compressedDirectBuf.limit(n);
uncompressedDirectBuf.clear(); // lz4 consumes all buffer input
@@ -297,5 +310,7 @@ public class Lz4Compressor implements Co
private native int compressBytesDirect();
+ private native int compressBytesDirectHC();
+
public native static String getLibraryName();
}
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj Mon Aug 5 21:10:54 2013
@@ -72,6 +72,7 @@
</ItemDefinitionGroup>
<ItemGroup>
<ClCompile Include="src\org\apache\hadoop\io\compress\lz4\lz4.c" />
+ <ClCompile Include="src\org\apache\hadoop\io\compress\lz4\lz4hc.c" />
<ClCompile Include="src\org\apache\hadoop\io\compress\lz4\Lz4Compressor.c" />
<ClCompile Include="src\org\apache\hadoop\io\compress\lz4\Lz4Decompressor.c" />
<ClCompile Include="src\org\apache\hadoop\io\nativeio\file_descriptor.c" />
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj.filters
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj.filters?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj.filters (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/native.vcxproj.filters Mon Aug 5 21:10:54 2013
@@ -51,6 +51,9 @@
<ClCompile Include="src\org\apache\hadoop\io\compress\lz4\lz4.c">
<Filter>Source Files</Filter>
</ClCompile>
+ <ClCompile Include="src\org\apache\hadoop\io\compress\lz4\lz4hc.c">
+ <Filter>Source Files</Filter>
+ </ClCompile>
<ClCompile Include="src\org\apache\hadoop\io\compress\lz4\Lz4Compressor.c">
<Filter>Source Files</Filter>
</ClCompile>
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Compressor.c Mon Aug 5 21:10:54 2013
@@ -23,21 +23,9 @@
#ifdef UNIX
#include "config.h"
#endif // UNIX
+#include "lz4.h"
+#include "lz4hc.h"
-//****************************
-// Simple Functions
-//****************************
-
-extern int LZ4_compress (const char* source, char* dest, int isize);
-
-/*
-LZ4_compress() :
- return : the number of bytes in compressed buffer dest
- note : destination buffer must be already allocated.
- To avoid any problem, size it to handle worst cases situations (input data not compressible)
- Worst case size is : "inputsize + 0.4%", with "0.4%" being at least 8 bytes.
-
-*/
static jfieldID Lz4Compressor_clazz;
static jfieldID Lz4Compressor_uncompressedDirectBuf;
@@ -107,5 +95,45 @@ JNIEXPORT jstring JNICALL
Java_org_apache_hadoop_io_compress_lz4_Lz4Compressor_getLibraryName(
JNIEnv *env, jclass class
) {
- return (*env)->NewStringUTF(env, "revision:43");
+ return (*env)->NewStringUTF(env, "revision:99");
+}
+
+JNIEXPORT jint JNICALL Java_org_apache_hadoop_io_compress_lz4_Lz4Compressor_compressBytesDirectHC
+(JNIEnv *env, jobject thisj){
+ const char* uncompressed_bytes = NULL;
+ char* compressed_bytes = NULL;
+
+ // Get members of Lz4Compressor
+ jobject clazz = (*env)->GetStaticObjectField(env, thisj, Lz4Compressor_clazz);
+ jobject uncompressed_direct_buf = (*env)->GetObjectField(env, thisj, Lz4Compressor_uncompressedDirectBuf);
+ jint uncompressed_direct_buf_len = (*env)->GetIntField(env, thisj, Lz4Compressor_uncompressedDirectBufLen);
+ jobject compressed_direct_buf = (*env)->GetObjectField(env, thisj, Lz4Compressor_compressedDirectBuf);
+ jint compressed_direct_buf_len = (*env)->GetIntField(env, thisj, Lz4Compressor_directBufferSize);
+
+ // Get the input direct buffer
+ LOCK_CLASS(env, clazz, "Lz4Compressor");
+ uncompressed_bytes = (const char*)(*env)->GetDirectBufferAddress(env, uncompressed_direct_buf);
+ UNLOCK_CLASS(env, clazz, "Lz4Compressor");
+
+ if (uncompressed_bytes == 0) {
+ return (jint)0;
+ }
+
+ // Get the output direct buffer
+ LOCK_CLASS(env, clazz, "Lz4Compressor");
+ compressed_bytes = (char *)(*env)->GetDirectBufferAddress(env, compressed_direct_buf);
+ UNLOCK_CLASS(env, clazz, "Lz4Compressor");
+
+ if (compressed_bytes == 0) {
+ return (jint)0;
+ }
+
+ compressed_direct_buf_len = LZ4_compressHC(uncompressed_bytes, compressed_bytes, uncompressed_direct_buf_len);
+ if (compressed_direct_buf_len < 0){
+ THROW(env, "java/lang/InternalError", "LZ4_compressHC failed");
+ }
+
+ (*env)->SetIntField(env, thisj, Lz4Compressor_uncompressedDirectBufLen, 0);
+
+ return (jint)compressed_direct_buf_len;
}
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/Lz4Decompressor.c Mon Aug 5 21:10:54 2013
@@ -22,18 +22,7 @@
#ifdef UNIX
#include "config.h"
#endif // UNIX
-
-int LZ4_uncompress_unknownOutputSize(const char* source, char* dest, int isize, int maxOutputSize);
-
-/*
-LZ4_uncompress_unknownOutputSize() :
- isize : is the input size, therefore the compressed size
- maxOutputSize : is the size of the destination buffer (which must be already allocated)
- return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize)
- If the source stream is malformed, the function will stop decoding and return a negative result, indicating the byte position of the faulty instruction
- This version never writes beyond dest + maxOutputSize, and is therefore protected against malicious data packets
- note : This version is a bit slower than LZ4_uncompress
-*/
+#include "lz4.h"
static jfieldID Lz4Decompressor_clazz;
@@ -89,7 +78,7 @@ JNIEXPORT jint JNICALL Java_org_apache_h
return (jint)0;
}
- uncompressed_direct_buf_len = LZ4_uncompress_unknownOutputSize(compressed_bytes, uncompressed_bytes, compressed_direct_buf_len, uncompressed_direct_buf_len);
+ uncompressed_direct_buf_len = LZ4_decompress_safe(compressed_bytes, uncompressed_bytes, compressed_direct_buf_len, uncompressed_direct_buf_len);
if (uncompressed_direct_buf_len < 0) {
THROW(env, "java/lang/InternalError", "LZ4_uncompress_unknownOutputSize failed.");
}
Modified: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c?rev=1510734&r1=1510733&r2=1510734&view=diff
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c (original)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.c Mon Aug 5 21:10:54 2013
@@ -1,19 +1,19 @@
/*
LZ4 - Fast LZ compression algorithm
- Copyright (C) 2011, Yann Collet.
- BSD License
+ Copyright (C) 2011-2013, Yann Collet.
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
-
+
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following disclaimer
in the documentation and/or other materials provided with the
distribution.
-
+
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
@@ -25,621 +25,672 @@
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+ - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+
+/*
+Note : this source file requires "lz4_encoder.h"
*/
//**************************************
-// Copy from:
-// URL: http://lz4.googlecode.com/svn/trunk/lz4.c
-// Repository Root: http://lz4.googlecode.com/svn
-// Repository UUID: 650e7d94-2a16-8b24-b05c-7c0b3f6821cd
-// Revision: 43
-// Node Kind: file
-// Last Changed Author: yann.collet.73@gmail.com
-// Last Changed Rev: 43
-// Last Changed Date: 2011-12-16 15:41:46 -0800 (Fri, 16 Dec 2011)
-// Sha1: 9db7b2c57698c528d79572e6bce2e7dc33fa5998
+// Tuning parameters
//**************************************
+// MEMORY_USAGE :
+// Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.)
+// Increasing memory usage improves compression ratio
+// Reduced memory usage can improve speed, due to cache effect
+// Default value is 14, for 16KB, which nicely fits into Intel x86 L1 cache
+#define MEMORY_USAGE 14
+
+// HEAPMODE :
+// Select how default compression function will allocate memory for its hash table,
+// in memory stack (0:default, fastest), or in memory heap (1:requires memory allocation (malloc)).
+// Default allocation strategy is to use stack (HEAPMODE 0)
+// Note : explicit functions *_stack* and *_heap* are unaffected by this setting
+#define HEAPMODE 0
+
//**************************************
-// Compilation Directives
+// CPU Feature Detection
//**************************************
-#if __STDC_VERSION__ >= 199901L
- /* "restrict" is a known keyword */
+// 32 or 64 bits ?
+#if (defined(__x86_64__) || defined(_M_X64) || defined(_WIN64) \
+ || defined(__powerpc64__) || defined(__ppc64__) || defined(__PPC64__) \
+ || defined(__64BIT__) || defined(_LP64) || defined(__LP64__) \
+ || defined(__ia64) || defined(__itanium__) || defined(_M_IA64) ) // Detects 64 bits mode
+# define LZ4_ARCH64 1
#else
-#define restrict // Disable restrict
+# define LZ4_ARCH64 0
+#endif
+
+// Little Endian or Big Endian ?
+// Overwrite the #define below if you know your architecture endianess
+#if defined (__GLIBC__)
+# include <endian.h>
+# if (__BYTE_ORDER == __BIG_ENDIAN)
+# define LZ4_BIG_ENDIAN 1
+# endif
+#elif (defined(__BIG_ENDIAN__) || defined(__BIG_ENDIAN) || defined(_BIG_ENDIAN)) && !(defined(__LITTLE_ENDIAN__) || defined(__LITTLE_ENDIAN) || defined(_LITTLE_ENDIAN))
+# define LZ4_BIG_ENDIAN 1
+#elif defined(__sparc) || defined(__sparc__) \
+ || defined(__powerpc__) || defined(__ppc__) || defined(__PPC__) \
+ || defined(__hpux) || defined(__hppa) \
+ || defined(_MIPSEB) || defined(__s390__)
+# define LZ4_BIG_ENDIAN 1
+#else
+// Little Endian assumed. PDP Endian and other very rare endian format are unsupported.
+#endif
+
+// Unaligned memory access is automatically enabled for "common" CPU, such as x86.
+// For others CPU, such as ARM, the compiler may be more cautious, inserting unnecessary extra code to ensure aligned access property
+// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance
+#if defined(__ARM_FEATURE_UNALIGNED)
+# define LZ4_FORCE_UNALIGNED_ACCESS 1
+#endif
+
+// Define this parameter if your target system or compiler does not support hardware bit count
+#if defined(_MSC_VER) && defined(_WIN32_WCE) // Visual Studio for Windows CE does not support Hardware bit count
+# define LZ4_FORCE_SW_BITCOUNT
#endif
+// BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE :
+// This option may provide a small boost to performance for some big endian cpu, although probably modest.
+// You may set this option to 1 if data will remain within closed environment.
+// This option is useless on Little_Endian CPU (such as x86)
+//#define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1
+
//**************************************
-// Includes
+// Compiler Options
//**************************************
-#include <stdlib.h> // for malloc
-#include <string.h> // for memset
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) // C99
+/* "restrict" is a known keyword */
+#else
+# define restrict // Disable restrict
+#endif
+
+#ifdef _MSC_VER // Visual Studio
+# define forceinline static __forceinline
+# include <intrin.h> // For Visual 2005
+# if LZ4_ARCH64 // 64-bits
+# pragma intrinsic(_BitScanForward64) // For Visual 2005
+# pragma intrinsic(_BitScanReverse64) // For Visual 2005
+# else // 32-bits
+# pragma intrinsic(_BitScanForward) // For Visual 2005
+# pragma intrinsic(_BitScanReverse) // For Visual 2005
+# endif
+# pragma warning(disable : 4127) // disable: C4127: conditional expression is constant
+#else
+# ifdef __GNUC__
+# define forceinline static inline __attribute__((always_inline))
+# else
+# define forceinline static inline
+# endif
+#endif
+
+#ifdef _MSC_VER
+# define lz4_bswap16(x) _byteswap_ushort(x)
+#else
+# define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | (((x) & 0xffu) << 8)))
+#endif
+
+#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
+
+#if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__)
+# define expect(expr,value) (__builtin_expect ((expr),(value)) )
+#else
+# define expect(expr,value) (expr)
+#endif
+
+#define likely(expr) expect((expr) != 0, 1)
+#define unlikely(expr) expect((expr) != 0, 0)
//**************************************
-// Performance parameter
+// Includes
//**************************************
-// Increasing this value improves compression ratio
-// Lowering this value reduces memory usage
-// Lowering may also improve speed, typically on reaching cache size limits (L1 32KB for Intel, 64KB for AMD)
-// Memory usage formula for 32 bits systems : N->2^(N+2) Bytes (examples : 17 -> 512KB ; 12 -> 16KB)
-#define HASH_LOG 12
+#include <stdlib.h> // for malloc
+#include <string.h> // for memset
+#include "lz4.h"
//**************************************
// Basic Types
//**************************************
-#if defined(_MSC_VER) // Visual Studio does not support 'stdint' natively
-#define BYTE unsigned __int8
-#define U16 unsigned __int16
-#define U32 unsigned __int32
-#define S32 __int32
-#else
-#include <stdint.h>
-#define BYTE uint8_t
-#define U16 uint16_t
-#define U32 uint32_t
-#define S32 int32_t
+#if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L // C99
+# include <stdint.h>
+ typedef uint8_t BYTE;
+ typedef uint16_t U16;
+ typedef uint32_t U32;
+ typedef int32_t S32;
+ typedef uint64_t U64;
+#else
+ typedef unsigned char BYTE;
+ typedef unsigned short U16;
+ typedef unsigned int U32;
+ typedef signed int S32;
+ typedef unsigned long long U64;
#endif
+#if defined(__GNUC__) && !defined(LZ4_FORCE_UNALIGNED_ACCESS)
+# define _PACKED __attribute__ ((packed))
+#else
+# define _PACKED
+#endif
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+# ifdef __IBMC__
+# pragma pack(1)
+# else
+# pragma pack(push, 1)
+# endif
+#endif
+
+typedef struct { U16 v; } _PACKED U16_S;
+typedef struct { U32 v; } _PACKED U32_S;
+typedef struct { U64 v; } _PACKED U64_S;
+typedef struct {size_t v;} _PACKED size_t_S;
+
+#if !defined(LZ4_FORCE_UNALIGNED_ACCESS) && !defined(__GNUC__)
+# pragma pack(pop)
+#endif
+
+#define A16(x) (((U16_S *)(x))->v)
+#define A32(x) (((U32_S *)(x))->v)
+#define A64(x) (((U64_S *)(x))->v)
+#define AARCH(x) (((size_t_S *)(x))->v)
+
//**************************************
// Constants
//**************************************
+#define HASHTABLESIZE (1 << MEMORY_USAGE)
+
#define MINMATCH 4
-#define SKIPSTRENGTH 6
-#define STACKLIMIT 13
-#define HEAPMODE (HASH_LOG>STACKLIMIT) // Defines if memory is allocated into the stack (local variable), or into the heap (malloc()).
-#define COPYTOKEN 4
+
#define COPYLENGTH 8
#define LASTLITERALS 5
#define MFLIMIT (COPYLENGTH+MINMATCH)
#define MINLENGTH (MFLIMIT+1)
+#define LZ4_64KLIMIT ((1<<16) + (MFLIMIT-1))
+#define SKIPSTRENGTH 6 // Increasing this value will make the compression run slower on incompressible data
+
#define MAXD_LOG 16
#define MAX_DISTANCE ((1 << MAXD_LOG) - 1)
-#define HASHTABLESIZE (1 << HASH_LOG)
-#define HASH_MASK (HASHTABLESIZE - 1)
-
-#define ML_BITS 4
-#define ML_MASK ((1U<<ML_BITS)-1)
+#define ML_BITS 4
+#define ML_MASK ((1U<<ML_BITS)-1)
#define RUN_BITS (8-ML_BITS)
#define RUN_MASK ((1U<<RUN_BITS)-1)
//**************************************
-// Local structures
+// Architecture-specific macros
//**************************************
-struct refTables
-{
- const BYTE* hashTable[HASHTABLESIZE];
-};
-
-#ifdef __GNUC__
-# define _PACKED __attribute__ ((packed))
-#else
-# define _PACKED
+#define STEPSIZE sizeof(size_t)
+#define LZ4_COPYSTEP(s,d) { AARCH(d) = AARCH(s); d+=STEPSIZE; s+=STEPSIZE; }
+#define LZ4_COPY8(s,d) { LZ4_COPYSTEP(s,d); if (STEPSIZE<8) LZ4_COPYSTEP(s,d); }
+#define LZ4_SECURECOPY(s,d,e) { if ((STEPSIZE==8)&&(d<e)) LZ4_WILDCOPY(s,d,e); }
+
+#if LZ4_ARCH64 // 64-bit
+# define HTYPE U32
+# define INITBASE(base) const BYTE* const base = ip
+#else // 32-bit
+# define HTYPE const BYTE*
+# define INITBASE(base) const int base = 0
+#endif
+
+#if (defined(LZ4_BIG_ENDIAN) && !defined(BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE))
+# define LZ4_READ_LITTLEENDIAN_16(d,s,p) { U16 v = A16(p); v = lz4_bswap16(v); d = (s) - v; }
+# define LZ4_WRITE_LITTLEENDIAN_16(p,i) { U16 v = (U16)(i); v = lz4_bswap16(v); A16(p) = v; p+=2; }
+#else // Little Endian
+# define LZ4_READ_LITTLEENDIAN_16(d,s,p) { d = (s) - A16(p); }
+# define LZ4_WRITE_LITTLEENDIAN_16(p,v) { A16(p) = v; p+=2; }
#endif
-typedef struct _U32_S
-{
- U32 v;
-} _PACKED U32_S;
-
-typedef struct _U16_S
-{
- U16 v;
-} _PACKED U16_S;
-
-#define A32(x) (((U32_S *)(x))->v)
-#define A16(x) (((U16_S *)(x))->v)
-
//**************************************
// Macros
//**************************************
-#define LZ4_HASH_FUNCTION(i) (((i) * 2654435761U) >> ((MINMATCH*8)-HASH_LOG))
-#define LZ4_HASH_VALUE(p) LZ4_HASH_FUNCTION(A32(p))
-#define LZ4_COPYPACKET(s,d) A32(d) = A32(s); d+=4; s+=4; A32(d) = A32(s); d+=4; s+=4;
-#define LZ4_WILDCOPY(s,d,e) do { LZ4_COPYPACKET(s,d) } while (d<e);
-#define LZ4_BLINDCOPY(s,d,l) { BYTE* e=d+l; LZ4_WILDCOPY(s,d,e); d=e; }
-
+#define LZ4_WILDCOPY(s,d,e) { do { LZ4_COPY8(s,d) } while (d<e); }
+#define LZ4_BLINDCOPY(s,d,l) { BYTE* e=(d)+(l); LZ4_WILDCOPY(s,d,e); d=e; }
//****************************
-// Compression CODE
+// Private functions
//****************************
+#if LZ4_ARCH64
+
+forceinline int LZ4_NbCommonBytes (register U64 val)
+{
+# if defined(LZ4_BIG_ENDIAN)
+# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ unsigned long r = 0;
+ _BitScanReverse64( &r, val );
+ return (int)(r>>3);
+# elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (__builtin_clzll(val) >> 3);
+# else
+ int r;
+ if (!(val>>32)) { r=4; } else { r=0; val>>=32; }
+ if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; }
+ r += (!val);
+ return r;
+# endif
+# else
+# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ unsigned long r = 0;
+ _BitScanForward64( &r, val );
+ return (int)(r>>3);
+# elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (__builtin_ctzll(val) >> 3);
+# else
+ static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 };
+ return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58];
+# endif
+# endif
+}
-int LZ4_compressCtx(void** ctx,
- char* source,
- char* dest,
- int isize)
-{
-#if HEAPMODE
- struct refTables *srt = (struct refTables *) (*ctx);
- const BYTE** HashTable;
#else
- const BYTE* HashTable[HASHTABLESIZE] = {0};
-#endif
- const BYTE* ip = (BYTE*) source;
- const BYTE* anchor = ip;
- const BYTE* const iend = ip + isize;
- const BYTE* const mflimit = iend - MFLIMIT;
-#define matchlimit (iend - LASTLITERALS)
+forceinline int LZ4_NbCommonBytes (register U32 val)
+{
+# if defined(LZ4_BIG_ENDIAN)
+# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ unsigned long r = 0;
+ _BitScanReverse( &r, val );
+ return (int)(r>>3);
+# elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (__builtin_clz(val) >> 3);
+# else
+ int r;
+ if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; }
+ r += (!val);
+ return r;
+# endif
+# else
+# if defined(_MSC_VER) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ unsigned long r;
+ _BitScanForward( &r, val );
+ return (int)(r>>3);
+# elif defined(__GNUC__) && (GCC_VERSION >= 304) && !defined(LZ4_FORCE_SW_BITCOUNT)
+ return (__builtin_ctz(val) >> 3);
+# else
+ static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
+ return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27];
+# endif
+# endif
+}
- BYTE* op = (BYTE*) dest;
-
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- const size_t DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
#endif
- int len, length;
- const int skipStrength = SKIPSTRENGTH;
- U32 forwardH;
- // Init
- if (isize<MINLENGTH) goto _last_literals;
-#if HEAPMODE
- if (*ctx == NULL)
- {
- srt = (struct refTables *) malloc ( sizeof(struct refTables) );
- *ctx = (void*) srt;
- }
- HashTable = srt->hashTable;
- memset((void*)HashTable, 0, sizeof(srt->hashTable));
-#else
- (void) ctx;
-#endif
-
-
- // First Byte
- HashTable[LZ4_HASH_VALUE(ip)] = ip;
- ip++; forwardH = LZ4_HASH_VALUE(ip);
-
- // Main Loop
- for ( ; ; )
- {
- int findMatchAttempts = (1U << skipStrength) + 3;
- const BYTE* forwardIp = ip;
- const BYTE* ref;
- BYTE* token;
-
- // Find a match
- do {
- U32 h = forwardH;
- int step = findMatchAttempts++ >> skipStrength;
- ip = forwardIp;
- forwardIp = ip + step;
-
- if (forwardIp > mflimit) { goto _last_literals; }
-
- forwardH = LZ4_HASH_VALUE(forwardIp);
- ref = HashTable[h];
- HashTable[h] = ip;
-
- } while ((ref < ip - MAX_DISTANCE) || (A32(ref) != A32(ip)));
-
- // Catch up
- while ((ip>anchor) && (ref>(BYTE*)source) && (ip[-1]==ref[-1])) { ip--; ref--; }
-
- // Encode Literal length
- length = ip - anchor;
- token = op++;
- if (length>=(int)RUN_MASK) { *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *op++ = 255; *op++ = (BYTE)len; }
- else *token = (length<<ML_BITS);
-
- // Copy Literals
- LZ4_BLINDCOPY(anchor, op, length);
-
-
-_next_match:
- // Encode Offset
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- A16(op) = (ip-ref); op+=2;
-#else
- { int delta = ip-ref; *op++ = delta; *op++ = delta>>8; }
-#endif
-
- // Start Counting
- ip+=MINMATCH; ref+=MINMATCH; // MinMatch verified
- anchor = ip;
- while (ip<matchlimit-3)
- {
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- int diff = A32(ref) ^ A32(ip);
- if (!diff) { ip+=4; ref+=4; continue; }
- ip += DeBruijnBytePos[((U32)((diff & -diff) * 0x077CB531U)) >> 27];
-#else
- if (A32(ref) == A32(ip)) { ip+=4; ref+=4; continue; }
- if (A16(ref) == A16(ip)) { ip+=2; ref+=2; }
- if (*ref == *ip) ip++;
-#endif
- goto _endCount;
- }
- if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
- if ((ip<matchlimit) && (*ref == *ip)) ip++;
-_endCount:
- len = (ip - anchor);
-
- // Encode MatchLength
- if (len>=(int)ML_MASK) { *token+=ML_MASK; len-=ML_MASK; for(; len > 509 ; len-=510) { *op++ = 255; *op++ = 255; } if (len > 254) { len-=255; *op++ = 255; } *op++ = (BYTE)len; }
- else *token += len;
-
- // Test end of chunk
- if (ip > mflimit) { anchor = ip; break; }
-
- // Fill table
- HashTable[LZ4_HASH_VALUE(ip-2)] = ip-2;
-
- // Test next position
- ref = HashTable[LZ4_HASH_VALUE(ip)];
- HashTable[LZ4_HASH_VALUE(ip)] = ip;
- if ((ref > ip - (MAX_DISTANCE + 1)) && (A32(ref) == A32(ip))) { token = op++; *token=0; goto _next_match; }
-
- // Prepare next loop
- anchor = ip++;
- forwardH = LZ4_HASH_VALUE(ip);
- }
-
-_last_literals:
- // Encode Last Literals
- {
- int lastRun = iend - anchor;
- if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun > 254 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
- else *op++ = (lastRun<<ML_BITS);
- memcpy(op, anchor, iend - anchor);
- op += iend-anchor;
- }
- // End
- return (int) (((char*)op)-dest);
-}
+//******************************
+// Compression functions
+//******************************
+/*
+int LZ4_compress_stack(
+ const char* source,
+ char* dest,
+ int inputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest'.
+Destination buffer must be already allocated, and sized at a minimum of LZ4_compressBound(inputSize).
+return : the number of bytes written in buffer 'dest'
+*/
+#define FUNCTION_NAME LZ4_compress_stack
+#include "lz4_encoder.h"
-// Note : this function is valid only if isize < LZ4_64KLIMIT
-#define LZ4_64KLIMIT ((1U<<16) + (MFLIMIT-1))
-#define HASHLOG64K (HASH_LOG+1)
-#define LZ4_HASH64K_FUNCTION(i) (((i) * 2654435761U) >> ((MINMATCH*8)-HASHLOG64K))
-#define LZ4_HASH64K_VALUE(p) LZ4_HASH64K_FUNCTION(A32(p))
-int LZ4_compress64kCtx(void** ctx,
- char* source,
- char* dest,
- int isize)
-{
-#if HEAPMODE
- struct refTables *srt = (struct refTables *) (*ctx);
- U16* HashTable;
-#else
- U16 HashTable[HASHTABLESIZE<<1] = {0};
-#endif
+/*
+int LZ4_compress_stack_limitedOutput(
+ const char* source,
+ char* dest,
+ int inputSize,
+ int maxOutputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+If it cannot achieve it, compression will stop, and result of the function will be zero.
+return : the number of bytes written in buffer 'dest', or 0 if the compression fails
+*/
+#define FUNCTION_NAME LZ4_compress_stack_limitedOutput
+#define LIMITED_OUTPUT
+#include "lz4_encoder.h"
- const BYTE* ip = (BYTE*) source;
- const BYTE* anchor = ip;
- const BYTE* const base = ip;
- const BYTE* const iend = ip + isize;
- const BYTE* const mflimit = iend - MFLIMIT;
-#define matchlimit (iend - LASTLITERALS)
- BYTE* op = (BYTE*) dest;
-
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- const size_t DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 };
-#endif
- int len, length;
- const int skipStrength = SKIPSTRENGTH;
- U32 forwardH;
+/*
+int LZ4_compress64k_stack(
+ const char* source,
+ char* dest,
+ int inputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest'.
+This function compresses better than LZ4_compress_stack(), on the condition that
+'inputSize' must be < to LZ4_64KLIMIT, or the function will fail.
+Destination buffer must be already allocated, and sized at a minimum of LZ4_compressBound(inputSize).
+return : the number of bytes written in buffer 'dest', or 0 if compression fails
+*/
+#define FUNCTION_NAME LZ4_compress64k_stack
+#define COMPRESS_64K
+#include "lz4_encoder.h"
- // Init
- if (isize<MINLENGTH) goto _last_literals;
-#if HEAPMODE
- if (*ctx == NULL)
- {
- srt = (struct refTables *) malloc ( sizeof(struct refTables) );
- *ctx = (void*) srt;
- }
- HashTable = (U16*)(srt->hashTable);
- memset((void*)HashTable, 0, sizeof(srt->hashTable));
-#else
- (void) ctx;
-#endif
-
-
- // First Byte
- ip++; forwardH = LZ4_HASH64K_VALUE(ip);
-
- // Main Loop
- for ( ; ; )
- {
- int findMatchAttempts = (1U << skipStrength) + 3;
- const BYTE* forwardIp = ip;
- const BYTE* ref;
- BYTE* token;
-
- // Find a match
- do {
- U32 h = forwardH;
- int step = findMatchAttempts++ >> skipStrength;
- ip = forwardIp;
- forwardIp = ip + step;
-
- if (forwardIp > mflimit) { goto _last_literals; }
-
- forwardH = LZ4_HASH64K_VALUE(forwardIp);
- ref = base + HashTable[h];
- HashTable[h] = ip - base;
-
- } while (A32(ref) != A32(ip));
-
- // Catch up
- while ((ip>anchor) && (ref>(BYTE*)source) && (ip[-1]==ref[-1])) { ip--; ref--; }
-
- // Encode Literal length
- length = ip - anchor;
- token = op++;
- if (length>=(int)RUN_MASK) { *token=(RUN_MASK<<ML_BITS); len = length-RUN_MASK; for(; len > 254 ; len-=255) *op++ = 255; *op++ = (BYTE)len; }
- else *token = (length<<ML_BITS);
-
- // Copy Literals
- LZ4_BLINDCOPY(anchor, op, length);
-
-
-_next_match:
- // Encode Offset
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- A16(op) = (ip-ref); op+=2;
-#else
- { int delta = ip-ref; *op++ = delta; *op++ = delta>>8; }
-#endif
-
- // Start Counting
- ip+=MINMATCH; ref+=MINMATCH; // MinMatch verified
- anchor = ip;
- while (ip<matchlimit-3)
- {
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- int diff = A32(ref) ^ A32(ip);
- if (!diff) { ip+=4; ref+=4; continue; }
- ip += DeBruijnBytePos[((U32)((diff & -diff) * 0x077CB531U)) >> 27];
-#else
- if (A32(ref) == A32(ip)) { ip+=4; ref+=4; continue; }
- if (A16(ref) == A16(ip)) { ip+=2; ref+=2; }
- if (*ref == *ip) ip++;
-#endif
- goto _endCount;
- }
- if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
- if ((ip<matchlimit) && (*ref == *ip)) ip++;
-_endCount:
- len = (ip - anchor);
-
- // Encode MatchLength
- if (len>=(int)ML_MASK) { *token+=ML_MASK; len-=ML_MASK; for(; len > 509 ; len-=510) { *op++ = 255; *op++ = 255; } if (len > 254) { len-=255; *op++ = 255; } *op++ = (BYTE)len; }
- else *token += len;
-
- // Test end of chunk
- if (ip > mflimit) { anchor = ip; break; }
-
- // Test next position
- ref = base + HashTable[LZ4_HASH64K_VALUE(ip)];
- HashTable[LZ4_HASH64K_VALUE(ip)] = ip - base;
- if (A32(ref) == A32(ip)) { token = op++; *token=0; goto _next_match; }
-
- // Prepare next loop
- anchor = ip++;
- forwardH = LZ4_HASH64K_VALUE(ip);
- }
-
-_last_literals:
- // Encode Last Literals
- {
- int lastRun = iend - anchor;
- if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun > 254 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
- else *op++ = (lastRun<<ML_BITS);
- memcpy(op, anchor, iend - anchor);
- op += iend-anchor;
- }
+/*
+int LZ4_compress64k_stack_limitedOutput(
+ const char* source,
+ char* dest,
+ int inputSize,
+ int maxOutputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+This function compresses better than LZ4_compress_stack_limitedOutput(), on the condition that
+'inputSize' must be < to LZ4_64KLIMIT, or the function will fail.
+If it cannot achieve it, compression will stop, and result of the function will be zero.
+return : the number of bytes written in buffer 'dest', or 0 if the compression fails
+*/
+#define FUNCTION_NAME LZ4_compress64k_stack_limitedOutput
+#define COMPRESS_64K
+#define LIMITED_OUTPUT
+#include "lz4_encoder.h"
+
+
+/*
+void* LZ4_createHeapMemory();
+int LZ4_freeHeapMemory(void* ctx);
+
+Used to allocate and free hashTable memory
+to be used by the LZ4_compress_heap* family of functions.
+LZ4_createHeapMemory() returns NULL is memory allocation fails.
+*/
+void* LZ4_create() { return malloc(HASHTABLESIZE); }
+int LZ4_free(void* ctx) { free(ctx); return 0; }
+
+
+/*
+int LZ4_compress_heap(
+ void* ctx,
+ const char* source,
+ char* dest,
+ int inputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest'.
+The memory used for compression must be created by LZ4_createHeapMemory() and provided by pointer 'ctx'.
+Destination buffer must be already allocated, and sized at a minimum of LZ4_compressBound(inputSize).
+return : the number of bytes written in buffer 'dest'
+*/
+#define FUNCTION_NAME LZ4_compress_heap
+#define USE_HEAPMEMORY
+#include "lz4_encoder.h"
- // End
- return (int) (((char*)op)-dest);
-}
+/*
+int LZ4_compress_heap_limitedOutput(
+ void* ctx,
+ const char* source,
+ char* dest,
+ int inputSize,
+ int maxOutputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+If it cannot achieve it, compression will stop, and result of the function will be zero.
+The memory used for compression must be created by LZ4_createHeapMemory() and provided by pointer 'ctx'.
+return : the number of bytes written in buffer 'dest', or 0 if the compression fails
+*/
+#define FUNCTION_NAME LZ4_compress_heap_limitedOutput
+#define LIMITED_OUTPUT
+#define USE_HEAPMEMORY
+#include "lz4_encoder.h"
+
+
+/*
+int LZ4_compress64k_heap(
+ void* ctx,
+ const char* source,
+ char* dest,
+ int inputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest'.
+The memory used for compression must be created by LZ4_createHeapMemory() and provided by pointer 'ctx'.
+'inputSize' must be < to LZ4_64KLIMIT, or the function will fail.
+Destination buffer must be already allocated, and sized at a minimum of LZ4_compressBound(inputSize).
+return : the number of bytes written in buffer 'dest'
+*/
+#define FUNCTION_NAME LZ4_compress64k_heap
+#define COMPRESS_64K
+#define USE_HEAPMEMORY
+#include "lz4_encoder.h"
-int LZ4_compress(char* source,
- char* dest,
- int isize)
+/*
+int LZ4_compress64k_heap_limitedOutput(
+ void* ctx,
+ const char* source,
+ char* dest,
+ int inputSize,
+ int maxOutputSize)
+
+Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+If it cannot achieve it, compression will stop, and result of the function will be zero.
+The memory used for compression must be created by LZ4_createHeapMemory() and provided by pointer 'ctx'.
+'inputSize' must be < to LZ4_64KLIMIT, or the function will fail.
+return : the number of bytes written in buffer 'dest', or 0 if the compression fails
+*/
+#define FUNCTION_NAME LZ4_compress64k_heap_limitedOutput
+#define COMPRESS_64K
+#define LIMITED_OUTPUT
+#define USE_HEAPMEMORY
+#include "lz4_encoder.h"
+
+
+int LZ4_compress(const char* source, char* dest, int inputSize)
{
#if HEAPMODE
- void* ctx = malloc(sizeof(struct refTables));
- int result;
- if (isize < LZ4_64KLIMIT)
- result = LZ4_compress64kCtx(&ctx, source, dest, isize);
- else result = LZ4_compressCtx(&ctx, source, dest, isize);
- free(ctx);
- return result;
+ void* ctx = LZ4_create();
+ int result;
+ if (ctx == NULL) return 0; // Failed allocation => compression not done
+ if (inputSize < LZ4_64KLIMIT)
+ result = LZ4_compress64k_heap(ctx, source, dest, inputSize);
+ else result = LZ4_compress_heap(ctx, source, dest, inputSize);
+ LZ4_free(ctx);
+ return result;
#else
- if (isize < (int)LZ4_64KLIMIT) return LZ4_compress64kCtx(NULL, source, dest, isize);
- return LZ4_compressCtx(NULL, source, dest, isize);
+ if (inputSize < (int)LZ4_64KLIMIT) return LZ4_compress64k_stack(source, dest, inputSize);
+ return LZ4_compress_stack(source, dest, inputSize);
#endif
}
+int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+#if HEAPMODE
+ void* ctx = LZ4_create();
+ int result;
+ if (ctx == NULL) return 0; // Failed allocation => compression not done
+ if (inputSize < LZ4_64KLIMIT)
+ result = LZ4_compress64k_heap_limitedOutput(ctx, source, dest, inputSize, maxOutputSize);
+ else result = LZ4_compress_heap_limitedOutput(ctx, source, dest, inputSize, maxOutputSize);
+ LZ4_free(ctx);
+ return result;
+#else
+ if (inputSize < (int)LZ4_64KLIMIT) return LZ4_compress64k_stack_limitedOutput(source, dest, inputSize, maxOutputSize);
+ return LZ4_compress_stack_limitedOutput(source, dest, inputSize, maxOutputSize);
+#endif
+}
//****************************
-// Decompression CODE
+// Decompression functions
//****************************
-// Note : The decoding functions LZ4_uncompress() and LZ4_uncompress_unknownOutputSize()
-// are safe against "buffer overflow" attack type
-// since they will *never* write outside of the provided output buffer :
-// they both check this condition *before* writing anything.
-// A corrupted packet however can make them *read* within the first 64K before the output buffer.
-
-int LZ4_uncompress(char* source,
- char* dest,
- int osize)
-{
- // Local Variables
- const BYTE* restrict ip = (const BYTE*) source;
- const BYTE* restrict ref;
-
- BYTE* restrict op = (BYTE*) dest;
- BYTE* const oend = op + osize;
- BYTE* cpy;
-
- BYTE token;
-
- U32 dec[4]={0, 3, 2, 3};
- int len, length;
-
-
- // Main Loop
- while (1)
- {
- // get runlength
- token = *ip++;
- if ((length=(token>>ML_BITS)) == RUN_MASK) { for (;(len=*ip++)==255;length+=255){} length += len; }
-
- // copy literals
- cpy = op+length;
- if (cpy>oend-COPYLENGTH)
- {
- if (cpy > oend) goto _output_error;
- memcpy(op, ip, length);
- ip += length;
- break; // Necessarily EOF
- }
- LZ4_WILDCOPY(ip, op, cpy); ip -= (op-cpy); op = cpy;
-
-
- // get offset
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- ref = cpy - A16(ip); ip+=2;
-#else
- { int delta = *ip++; delta += *ip++ << 8; ref = cpy - delta; }
-#endif
-
- // get matchlength
- if ((length=(token&ML_MASK)) == ML_MASK) { for (;*ip==255;length+=255) {ip++;} length += *ip++; }
-
- // copy repeated sequence
- if (op-ref<COPYTOKEN)
- {
- *op++ = *ref++;
- *op++ = *ref++;
- *op++ = *ref++;
- *op++ = *ref++;
- ref -= dec[op-ref];
- A32(op)=A32(ref);
- } else { A32(op)=A32(ref); op+=4; ref+=4; }
- cpy = op + length;
- if (cpy > oend-COPYLENGTH)
- {
- if (cpy > oend) goto _output_error;
- LZ4_WILDCOPY(ref, op, (oend-COPYLENGTH));
- while(op<cpy) *op++=*ref++;
- op=cpy;
- if (op == oend) break; // Check EOF (should never happen, since last 5 bytes are supposed to be literals)
- continue;
- }
- LZ4_WILDCOPY(ref, op, cpy);
- op=cpy; // correction
- }
-
- // end of decoding
- return (int) (((char*)ip)-source);
+typedef enum { noPrefix = 0, withPrefix = 1 } prefix64k_directive;
+typedef enum { endOnOutputSize = 0, endOnInputSize = 1 } endCondition_directive;
+typedef enum { full = 0, partial = 1 } earlyEnd_directive;
+
+
+// This generic decompression function cover all use cases.
+// It shall be instanciated several times, using different sets of directives
+// Note that it is essential this generic function is really inlined,
+// in order to remove useless branches during compilation optimisation.
+forceinline int LZ4_decompress_generic(
+ const char* source,
+ char* dest,
+ int inputSize, //
+ int outputSize, // OutputSize must be != 0; if endOnInput==endOnInputSize, this value is the max size of Output Buffer.
+
+ int endOnInput, // endOnOutputSize, endOnInputSize
+ int prefix64k, // noPrefix, withPrefix
+ int partialDecoding, // full, partial
+ int targetOutputSize // only used if partialDecoding==partial
+ )
+{
+ // Local Variables
+ const BYTE* restrict ip = (const BYTE*) source;
+ const BYTE* ref;
+ const BYTE* const iend = ip + inputSize;
+
+ BYTE* op = (BYTE*) dest;
+ BYTE* const oend = op + outputSize;
+ BYTE* cpy;
+ BYTE* oexit = op + targetOutputSize;
+
+ size_t dec32table[] = {0, 3, 2, 3, 0, 0, 0, 0};
+#if LZ4_ARCH64
+ size_t dec64table[] = {0, 0, 0, (size_t)-1, 0, 1, 2, 3};
+#endif
+
+
+ // Special cases
+ if ((partialDecoding) && (oexit> oend-MFLIMIT)) oexit = oend-MFLIMIT; // targetOutputSize too high => decode everything
+ if ((endOnInput) && unlikely(outputSize==0)) return ((inputSize==1) && (*ip==0)) ? 0 : -1; // Empty output buffer
+ if ((!endOnInput) && unlikely(outputSize==0)) return (*ip==0?1:-1);
+
+
+ // Main Loop
+ while (1)
+ {
+ unsigned token;
+ size_t length;
+
+ // get runlength
+ token = *ip++;
+ if ((length=(token>>ML_BITS)) == RUN_MASK)
+ {
+ unsigned s=255;
+ while (((endOnInput)?ip<iend:1) && (s==255))
+ {
+ s = *ip++;
+ length += s;
+ }
+ }
+
+ // copy literals
+ cpy = op+length;
+ if (((endOnInput) && ((cpy>(partialDecoding?oexit:oend-MFLIMIT)) || (ip+length>iend-(2+1+LASTLITERALS))) )
+ || ((!endOnInput) && (cpy>oend-COPYLENGTH)))
+ {
+ if (partialDecoding)
+ {
+ if (cpy > oend) goto _output_error; // Error : write attempt beyond end of output buffer
+ if ((endOnInput) && (ip+length > iend)) goto _output_error; // Error : read attempt beyond end of input buffer
+ }
+ else
+ {
+ if ((!endOnInput) && (cpy != oend)) goto _output_error; // Error : block decoding must stop exactly there
+ if ((endOnInput) && ((ip+length != iend) || (cpy > oend))) goto _output_error; // Error : input must be consumed
+ }
+ memcpy(op, ip, length);
+ ip += length;
+ op += length;
+ break; // Necessarily EOF, due to parsing restrictions
+ }
+ LZ4_WILDCOPY(ip, op, cpy); ip -= (op-cpy); op = cpy;
+
+ // get offset
+ LZ4_READ_LITTLEENDIAN_16(ref,cpy,ip); ip+=2;
+ if ((prefix64k==noPrefix) && unlikely(ref < (BYTE* const)dest)) goto _output_error; // Error : offset outside destination buffer
+
+ // get matchlength
+ if ((length=(token&ML_MASK)) == ML_MASK)
+ {
+ for ( ; (!endOnInput) || (ip<iend-(LASTLITERALS+1)) ; ) // Ensure enough bytes remain for LASTLITERALS + token
+ {
+ unsigned s = *ip++;
+ length += s;
+ if (s==255) continue;
+ break;
+ }
+ }
+
+ // copy repeated sequence
+ if unlikely((op-ref)<(int)STEPSIZE)
+ {
+#if LZ4_ARCH64
+ size_t dec64 = dec64table[op-ref];
+#else
+ const size_t dec64 = 0;
+#endif
+ op[0] = ref[0];
+ op[1] = ref[1];
+ op[2] = ref[2];
+ op[3] = ref[3];
+ op += 4, ref += 4; ref -= dec32table[op-ref];
+ A32(op) = A32(ref);
+ op += STEPSIZE-4; ref -= dec64;
+ } else { LZ4_COPYSTEP(ref,op); }
+ cpy = op + length - (STEPSIZE-4);
+
+ if unlikely(cpy>oend-(COPYLENGTH)-(STEPSIZE-4))
+ {
+ if (cpy > oend-LASTLITERALS) goto _output_error; // Error : last 5 bytes must be literals
+ LZ4_SECURECOPY(ref, op, (oend-COPYLENGTH));
+ while(op<cpy) *op++=*ref++;
+ op=cpy;
+ continue;
+ }
+ LZ4_WILDCOPY(ref, op, cpy);
+ op=cpy; // correction
+ }
+
+ // end of decoding
+ if (endOnInput)
+ return (int) (((char*)op)-dest); // Nb of output bytes decoded
+ else
+ return (int) (((char*)ip)-source); // Nb of input bytes read
- // write overflow error detected
+ // Overflow error detected
_output_error:
- return (int) (-(((char*)ip)-source));
+ return (int) (-(((char*)ip)-source))-1;
}
-int LZ4_uncompress_unknownOutputSize(
- char* source,
- char* dest,
- int isize,
- int maxOutputSize)
-{
- // Local Variables
- const BYTE* restrict ip = (const BYTE*) source;
- const BYTE* const iend = ip + isize;
- const BYTE* restrict ref;
-
- BYTE* restrict op = (BYTE*) dest;
- BYTE* const oend = op + maxOutputSize;
- BYTE* cpy;
-
- BYTE token;
-
- U32 dec[4]={0, 3, 2, 3};
- int len, length;
-
-
- // Main Loop
- while (ip<iend)
- {
- // get runlength
- token = *ip++;
- if ((length=(token>>ML_BITS)) == RUN_MASK) { for (;(len=*ip++)==255;length+=255){} length += len; }
-
- // copy literals
- cpy = op+length;
- if (cpy>oend-COPYLENGTH)
- {
- if (cpy > oend) goto _output_error;
- memcpy(op, ip, length);
- op += length;
- break; // Necessarily EOF
- }
- LZ4_WILDCOPY(ip, op, cpy); ip -= (op-cpy); op = cpy;
- if (ip>=iend) break; // check EOF
-
- // get offset
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
- ref = cpy - A16(ip); ip+=2;
-#else
- { int delta = *ip++; delta += *ip++ << 8; ref = cpy - delta; }
-#endif
-
- // get matchlength
- if ((length=(token&ML_MASK)) == ML_MASK) { for (;(len=*ip++)==255;length+=255){} length += len; }
-
- // copy repeated sequence
- if (op-ref<COPYTOKEN)
- {
- *op++ = *ref++;
- *op++ = *ref++;
- *op++ = *ref++;
- *op++ = *ref++;
- ref -= dec[op-ref];
- A32(op)=A32(ref);
- } else { A32(op)=A32(ref); op+=4; ref+=4; }
- cpy = op + length;
- if (cpy>oend-COPYLENGTH)
- {
- if (cpy > oend) goto _output_error;
- LZ4_WILDCOPY(ref, op, (oend-COPYLENGTH));
- while(op<cpy) *op++=*ref++;
- op=cpy;
- if (op == oend) break; // Check EOF (should never happen, since last 5 bytes are supposed to be literals)
- continue;
- }
- LZ4_WILDCOPY(ref, op, cpy);
- op=cpy; // correction
- }
+int LZ4_decompress_safe(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+ return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, noPrefix, full, 0);
+}
- // end of decoding
- return (int) (((char*)op)-dest);
+int LZ4_decompress_fast(const char* source, char* dest, int outputSize)
+{
+ return LZ4_decompress_generic(source, dest, 0, outputSize, endOnOutputSize, noPrefix, full, 0);
+}
- // write overflow error detected
-_output_error:
- return (int) (-(((char*)ip)-source));
+int LZ4_decompress_safe_withPrefix64k(const char* source, char* dest, int inputSize, int maxOutputSize)
+{
+ return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, withPrefix, full, 0);
+}
+
+int LZ4_decompress_fast_withPrefix64k(const char* source, char* dest, int outputSize)
+{
+ return LZ4_decompress_generic(source, dest, 0, outputSize, endOnOutputSize, withPrefix, full, 0);
+}
+
+int LZ4_decompress_safe_partial(const char* source, char* dest, int inputSize, int targetOutputSize, int maxOutputSize)
+{
+ return LZ4_decompress_generic(source, dest, inputSize, maxOutputSize, endOnInputSize, noPrefix, partial, targetOutputSize);
}
Added: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h?rev=1510734&view=auto
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h (added)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4.h Mon Aug 5 21:10:54 2013
@@ -0,0 +1,179 @@
+/*
+ LZ4 - Fast LZ compression algorithm
+ Header File
+ Copyright (C) 2011-2013, Yann Collet.
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+ - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+
+/*
+ Copied from:
+ URL: http://lz4.googlecode.com/svn/trunk
+ Repository Root: http://lz4.googlecode.com/svn
+ Repository UUID: 650e7d94-2a16-8b24-b05c-7c0b3f6821cd
+ Revision: 99
+ Node Kind: directory
+ Schedule: normal
+ Last Changed Author: yann.collet.73@gmail.com
+ Last Changed Rev: 99
+ Last Changed Date: 2013-07-27 19:19:31 +0800 (Sat, 27 Jul 2013)
+ */
+
+#pragma once
+
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
+
+//**************************************
+// Compiler Options
+//**************************************
+#if defined(_MSC_VER) && !defined(__cplusplus) // Visual Studio
+# define inline __inline // Visual C is not C99, but supports some kind of inline
+#endif
+
+
+//****************************
+// Simple Functions
+//****************************
+
+int LZ4_compress (const char* source, char* dest, int inputSize);
+int LZ4_decompress_safe (const char* source, char* dest, int inputSize, int maxOutputSize);
+
+/*
+LZ4_compress() :
+ Compresses 'inputSize' bytes from 'source' into 'dest'.
+ Destination buffer must be already allocated,
+ and must be sized to handle worst cases situations (input data not compressible)
+ Worst case size evaluation is provided by function LZ4_compressBound()
+ inputSize : Max supported value is ~1.9GB
+ return : the number of bytes written in buffer dest
+ or 0 if the compression fails
+
+LZ4_decompress_safe() :
+ maxOutputSize : is the size of the destination buffer (which must be already allocated)
+ return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize)
+ If the source stream is detected malformed, the function will stop decoding and return a negative result.
+ This function is protected against buffer overflow exploits (never writes outside of output buffer, and never reads outside of input buffer). Therefore, it is protected against malicious data packets
+*/
+
+
+//****************************
+// Advanced Functions
+//****************************
+
+static inline int LZ4_compressBound(int isize) { return ((isize) + ((isize)/255) + 16); }
+#define LZ4_COMPRESSBOUND( isize) ((isize) + ((isize)/255) + 16)
+
+/*
+LZ4_compressBound() :
+ Provides the maximum size that LZ4 may output in a "worst case" scenario (input data not compressible)
+ primarily useful for memory allocation of output buffer.
+ inline function is recommended for the general case,
+ macro is also provided when result needs to be evaluated at compilation (such as table size allocation).
+
+ isize : is the input size. Max supported value is ~1.9GB
+ return : maximum output size in a "worst case" scenario
+ note : this function is limited by "int" range (2^31-1)
+*/
+
+
+int LZ4_compress_limitedOutput (const char* source, char* dest, int inputSize, int maxOutputSize);
+
+/*
+LZ4_compress_limitedOutput() :
+ Compress 'inputSize' bytes from 'source' into an output buffer 'dest' of maximum size 'maxOutputSize'.
+ If it cannot achieve it, compression will stop, and result of the function will be zero.
+ This function never writes outside of provided output buffer.
+
+ inputSize : Max supported value is ~1.9GB
+ maxOutputSize : is the size of the destination buffer (which must be already allocated)
+ return : the number of bytes written in buffer 'dest'
+ or 0 if the compression fails
+*/
+
+
+int LZ4_decompress_fast (const char* source, char* dest, int outputSize);
+
+/*
+LZ4_decompress_fast() :
+ outputSize : is the original (uncompressed) size
+ return : the number of bytes read from the source buffer (in other words, the compressed size)
+ If the source stream is malformed, the function will stop decoding and return a negative result.
+ note : This function is a bit faster than LZ4_decompress_safe()
+ This function never writes outside of output buffers, and never read before input buffer, but may read beyond input buffer (since it doesn't know its size) in case of malicious data packet.
+ Use this function preferably into a trusted environment (data to decode comes from a trusted source).
+ Destination buffer must be already allocated. Its size must be a minimum of 'outputSize' bytes.
+*/
+
+int LZ4_decompress_safe_partial (const char* source, char* dest, int inputSize, int targetOutputSize, int maxOutputSize);
+
+/*
+LZ4_decompress_safe_partial() :
+ This function decompress a compressed block of size 'inputSize' at position 'source'
+ into output buffer 'dest' of size 'maxOutputSize'.
+ The function stops decompressing operation as soon as 'targetOutputSize' has been reached,
+ reducing decompression time.
+ return : the number of bytes decoded in the destination buffer (necessarily <= maxOutputSize)
+ Note : this number can be < 'targetOutputSize' should the compressed block to decode be smaller.
+ Always control how many bytes were decoded.
+ If the source stream is malformed, the function will stop decoding and return a negative result.
+ This function never writes outside of output buffer, and never reads outside of input buffer. It is therefore protected against malicious data packets
+*/
+
+
+int LZ4_decompress_safe_withPrefix64k (const char* source, char* dest, int inputSize, int maxOutputSize);
+int LZ4_decompress_fast_withPrefix64k (const char* source, char* dest, int outputSize);
+
+/*
+*_withPrefix64k() :
+ These decoding functions work the same as their "normal name" versions,
+ but will potentially use up to 64KB of data in front of 'char* dest'.
+ These functions are used for decoding inter-dependant blocks.
+*/
+
+
+//****************************
+// Obsolete Functions
+//****************************
+
+static inline int LZ4_uncompress (const char* source, char* dest, int outputSize) { return LZ4_decompress_fast(source, dest, outputSize); }
+static inline int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize) { return LZ4_decompress_safe(source, dest, isize, maxOutputSize); }
+
+/*
+These functions are deprecated and should no longer be used.
+They are provided here for compatibility with existing user programs.
+*/
+
+
+
+#if defined (__cplusplus)
+}
+#endif
Added: hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h
URL: http://svn.apache.org/viewvc/hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h?rev=1510734&view=auto
==============================================================================
--- hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h (added)
+++ hadoop/common/trunk/hadoop-common-project/hadoop-common/src/main/native/src/org/apache/hadoop/io/compress/lz4/lz4_encoder.h Mon Aug 5 21:10:54 2013
@@ -0,0 +1,258 @@
+/*
+ LZ4 Encoder - Part of LZ4 compression algorithm
+ Copyright (C) 2011-2013, Yann Collet.
+ BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions are
+ met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above
+ copyright notice, this list of conditions and the following disclaimer
+ in the documentation and/or other materials provided with the
+ distribution.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ You can contact the author at :
+ - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html
+ - LZ4 source repository : http://code.google.com/p/lz4/
+*/
+
+/* lz4_encoder.h must be included into lz4.c
+ The objective of this file is to create a single LZ4 compression function source
+ which will be instanciated multiple times with minor variations
+ depending on a set of #define.
+*/
+
+
+
+//****************************
+// Check required defines
+//****************************
+
+#ifndef FUNCTION_NAME
+# error "FUNTION_NAME is not defined"
+#endif
+
+
+//****************************
+// Local definitions
+//****************************
+
+#ifdef COMPRESS_64K
+# define HASHLOG (MEMORY_USAGE-1)
+# define CURRENT_H_TYPE U16
+# define CURRENTBASE(base) const BYTE* const base = ip
+#else
+# define HASHLOG (MEMORY_USAGE-2)
+# define CURRENT_H_TYPE HTYPE
+# define CURRENTBASE(base) INITBASE(base)
+#endif
+
+#define HASHTABLE_NBCELLS (1U<<HASHLOG)
+#define LZ4_HASH(i) (((i) * 2654435761U) >> ((MINMATCH*8)-HASHLOG))
+#define LZ4_HASHVALUE(p) LZ4_HASH(A32(p))
+
+
+
+//****************************
+// Function code
+//****************************
+
+int FUNCTION_NAME(
+#ifdef USE_HEAPMEMORY
+ void* ctx,
+#endif
+ const char* source,
+ char* dest,
+ int inputSize
+#ifdef LIMITED_OUTPUT
+ ,int maxOutputSize
+#endif
+ )
+{
+#ifdef USE_HEAPMEMORY
+ CURRENT_H_TYPE* HashTable = (CURRENT_H_TYPE*)ctx;
+#else
+ CURRENT_H_TYPE HashTable[HASHTABLE_NBCELLS] = {0};
+#endif
+
+ const BYTE* ip = (BYTE*) source;
+ CURRENTBASE(base);
+ const BYTE* anchor = ip;
+ const BYTE* const iend = ip + inputSize;
+ const BYTE* const mflimit = iend - MFLIMIT;
+#define matchlimit (iend - LASTLITERALS)
+
+ BYTE* op = (BYTE*) dest;
+#ifdef LIMITED_OUTPUT
+ BYTE* const oend = op + maxOutputSize;
+#endif
+
+ int length;
+ const int skipStrength = SKIPSTRENGTH;
+ U32 forwardH;
+
+
+ // Init
+ if (inputSize<MINLENGTH) goto _last_literals;
+#ifdef COMPRESS_64K
+ if (inputSize>=LZ4_64KLIMIT) return 0; // Size too large (not within 64K limit)
+#endif
+#ifdef USE_HEAPMEMORY
+ memset((void*)HashTable, 0, HASHTABLESIZE);
+#endif
+
+ // First Byte
+ HashTable[LZ4_HASHVALUE(ip)] = (CURRENT_H_TYPE)(ip - base);
+ ip++; forwardH = LZ4_HASHVALUE(ip);
+
+ // Main Loop
+ for ( ; ; )
+ {
+ int findMatchAttempts = (1U << skipStrength) + 3;
+ const BYTE* forwardIp = ip;
+ const BYTE* ref;
+ BYTE* token;
+
+ // Find a match
+ do {
+ U32 h = forwardH;
+ int step = findMatchAttempts++ >> skipStrength;
+ ip = forwardIp;
+ forwardIp = ip + step;
+
+ if unlikely(forwardIp > mflimit) { goto _last_literals; }
+
+ forwardH = LZ4_HASHVALUE(forwardIp);
+ ref = base + HashTable[h];
+ HashTable[h] = (CURRENT_H_TYPE)(ip - base);
+
+ } while ((ref < ip - MAX_DISTANCE) || (A32(ref) != A32(ip)));
+
+ // Catch up
+ while ((ip>anchor) && (ref>(BYTE*)source) && unlikely(ip[-1]==ref[-1])) { ip--; ref--; }
+
+ // Encode Literal length
+ length = (int)(ip - anchor);
+ token = op++;
+#ifdef LIMITED_OUTPUT
+ if unlikely(op + length + (2 + 1 + LASTLITERALS) + (length>>8) > oend) return 0; // Check output limit
+#endif
+ if (length>=(int)RUN_MASK)
+ {
+ int len = length-RUN_MASK;
+ *token=(RUN_MASK<<ML_BITS);
+ for(; len >= 255 ; len-=255) *op++ = 255;
+ *op++ = (BYTE)len;
+ }
+ else *token = (BYTE)(length<<ML_BITS);
+
+ // Copy Literals
+ LZ4_BLINDCOPY(anchor, op, length);
+
+_next_match:
+ // Encode Offset
+ LZ4_WRITE_LITTLEENDIAN_16(op,(U16)(ip-ref));
+
+ // Start Counting
+ ip+=MINMATCH; ref+=MINMATCH; // MinMatch already verified
+ anchor = ip;
+ while likely(ip<matchlimit-(STEPSIZE-1))
+ {
+ size_t diff = AARCH(ref) ^ AARCH(ip);
+ if (!diff) { ip+=STEPSIZE; ref+=STEPSIZE; continue; }
+ ip += LZ4_NbCommonBytes(diff);
+ goto _endCount;
+ }
+ if (LZ4_ARCH64) if ((ip<(matchlimit-3)) && (A32(ref) == A32(ip))) { ip+=4; ref+=4; }
+ if ((ip<(matchlimit-1)) && (A16(ref) == A16(ip))) { ip+=2; ref+=2; }
+ if ((ip<matchlimit) && (*ref == *ip)) ip++;
+_endCount:
+
+ // Encode MatchLength
+ length = (int)(ip - anchor);
+#ifdef LIMITED_OUTPUT
+ if unlikely(op + (1 + LASTLITERALS) + (length>>8) > oend) return 0; // Check output limit
+#endif
+ if (length>=(int)ML_MASK)
+ {
+ *token += ML_MASK;
+ length -= ML_MASK;
+ for (; length > 509 ; length-=510) { *op++ = 255; *op++ = 255; }
+ if (length >= 255) { length-=255; *op++ = 255; }
+ *op++ = (BYTE)length;
+ }
+ else *token += (BYTE)(length);
+
+ // Test end of chunk
+ if (ip > mflimit) { anchor = ip; break; }
+
+ // Fill table
+ HashTable[LZ4_HASHVALUE(ip-2)] = (CURRENT_H_TYPE)(ip - 2 - base);
+
+ // Test next position
+ ref = base + HashTable[LZ4_HASHVALUE(ip)];
+ HashTable[LZ4_HASHVALUE(ip)] = (CURRENT_H_TYPE)(ip - base);
+ if ((ref >= ip - MAX_DISTANCE) && (A32(ref) == A32(ip))) { token = op++; *token=0; goto _next_match; }
+
+ // Prepare next loop
+ anchor = ip++;
+ forwardH = LZ4_HASHVALUE(ip);
+ }
+
+_last_literals:
+ // Encode Last Literals
+ {
+ int lastRun = (int)(iend - anchor);
+#ifdef LIMITED_OUTPUT
+ if (((char*)op - dest) + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > (U32)maxOutputSize) return 0; // Check output limit
+#endif
+ if (lastRun>=(int)RUN_MASK) { *op++=(RUN_MASK<<ML_BITS); lastRun-=RUN_MASK; for(; lastRun >= 255 ; lastRun-=255) *op++ = 255; *op++ = (BYTE) lastRun; }
+ else *op++ = (BYTE)(lastRun<<ML_BITS);
+ memcpy(op, anchor, iend - anchor);
+ op += iend-anchor;
+ }
+
+ // End
+ return (int) (((char*)op)-dest);
+}
+
+
+
+//****************************
+// Clean defines
+//****************************
+
+// Required defines
+#undef FUNCTION_NAME
+
+// Locally Generated
+#undef HASHLOG
+#undef HASHTABLE_NBCELLS
+#undef LZ4_HASH
+#undef LZ4_HASHVALUE
+#undef CURRENT_H_TYPE
+#undef CURRENTBASE
+
+// Optional defines
+#ifdef LIMITED_OUTPUT
+#undef LIMITED_OUTPUT
+#endif
+
+#ifdef USE_HEAPMEMORY
+#undef USE_HEAPMEMORY
+#endif