Go to file
Takayuki MATSUOKA 41b6ed3c5b Replace fseek with _fseeki64 to avoid MSVC's 2GiB barrier
Since MSVC fseek()'s SEEK_CUR mode has 2GiB barrier, our fseek() calling in program/lz4io.c/selectDecoder() will fail for large (>2GiB) .lz4 file.
This commit just replace fseek with _fseeki64 by macro. Second argument is automatically casted to __int64 (signed 64bit integer). Other things are completely same.
2015-02-14 12:48:11 +09:00
cmake_unofficial New : lz4frame integrated into liblz4 (v1.5.0) 2014-12-13 15:05:46 +01:00
examples Replace fseek with _fseeki64 to avoid MSVC's 2GiB barrier 2015-02-14 12:48:11 +09:00
images Added : LZ4 Framing Format, new version (v1.4.1) 2014-08-18 16:00:00 +01:00
lib Added : -pedantic compilation option 2014-12-17 12:32:49 +01:00
programs Replace fseek with _fseeki64 to avoid MSVC's 2GiB barrier 2015-02-14 12:48:11 +09:00
visual/2012 Added : Visual project directory 2015-01-13 05:36:58 +01:00
.gitattributes Added : Visual project directory 2015-01-13 05:36:58 +01:00
.travis.yml fixed travis tests 2014-12-13 16:41:05 +01:00
lz4_block_format.txt Clarified some file names 2014-12-01 00:53:20 +01:00
LZ4_Frame_Format.html Clarified some file names 2014-12-01 00:53:20 +01:00
Makefile fixed travis tests 2014-12-13 16:41:05 +01:00
NEWS Added : -pedantic compilation option 2014-12-17 12:32:49 +01:00
README.md clarified "dev" branch policy 2014-08-08 12:48:45 +01:00

LZ4 - Extremely fast compression

LZ4 is lossless compression algorithm, providing compression speed at 400 MB/s per core, scalable with multi-cores CPU. It also features an extremely fast decoder, with speed in multiple GB/s per core, typically reaching RAM speed limits on multi-core systems. A high compression derivative, called LZ4_HC, is also provided. It trades CPU time for compression ratio.

Branch Status
master Build Status
dev Build Status

This is an official mirror of LZ4 project, hosted on Google Code. The intention is to offer github's capabilities to lz4 users, such as cloning, branch, pull requests or source download.

The "master" branch will reflect, the status of lz4 at its official homepage. The "dev" branch is the one where all contributions will be merged. If you plan to propose a patch, please commit into the "dev" branch. Direct commit to "master" are not permitted. Feature branches will also exist, typically to introduce new requirements, and be temporarily available for testing before merge into "dev" branch.

Benchmarks

The benchmark uses the Open-Source Benchmark program by m^2 (v0.14.2) compiled with GCC v4.6.1 on Linux Ubuntu 64-bits v11.10, The reference system uses a Core i5-3340M @2.7GHz. Benchmark evaluates the compression of reference Silesia Corpus in single-thread mode.

CompressorRatioCompressionDecompression
LZ4 (r101)2.084422 MB/s1820 MB/s
LZO 2.062.106414 MB/s600 MB/s
QuickLZ 1.5.1b62.237373 MB/s420 MB/s
Snappy 1.1.02.091323 MB/s1070 MB/s
LZF2.077270 MB/s570 MB/s
zlib 1.2.8 -12.73065 MB/s280 MB/s
LZ4 HC (r101)2.72025 MB/s2080 MB/s
zlib 1.2.8 -63.09921 MB/s300 MB/s