-
Notifications
You must be signed in to change notification settings - Fork 300
Releases
Pablo de Lara edited this page Jan 19, 2024
·
23 revisions
-
API changes:
- gf_vect_mul_base() function now returns an integer, matching the return type of gf_vect_mul() function (not a breaking change).
-
Igzip compression improvements:
- Added compress/decompress with dictionary to perf test app.
- Zlib header can be now created on the fly when starting the compression.
- Added isal_zlib_hdr_init() function to initialize the zlib header to 0.
-
Zero-memory dectection improvements:
- Optimized AVX implementation.
- Added new AVX2 and AVX512 implementations.
-
Erasure coding improvements:
- Added new AVX512 and AVX2 implementations using GFNI instructions.
- Added new SVE implementation.
-
CRC improvements:
- Added new CRC64 Rocksoft algorithm.
- CRC x86 implementations optimized using ternary logic instructions and folding of bigger data on the last bytes.
- CRC16 T10dif aarch64 implementation improved.
- CRC aarch64 implementations optimized using XOR fusion feature.
-
Documentation:
- Added function overview documentation page.
- Added security file.
-
Performance apps:
- Changed performance tests to warm by default.
-
Example apps:
- Added CRC combine example
crc_combine_example
for multiple polynomials.
- Added CRC combine example
-
Fixes:
- Fixed various compilation issues/warnings for different platforms.
- Fixed documentation on xor/pq gen/check functions, with minimum number of vectors.
- Fixed potential out-of-bounds read on Adler32 Neon implementation.
- Fixed potential out-of-bounds read on gf_vect_mul Neon implementation.
- Fixed x86 load/store instructions in erasure coding functions (aligned moves that should be unaligned).
- Fixed memory leaks in unit tests.
- New optimized version of AES-CBC decode
- New AVX2, 8 lane version of multi-buffer SM3
- Added support for big-endian architectures
- Igzip compression/decompression enhancements.
- New functions for compression with dictionary acceleration.
- New static level 0 header decode tables. Accelerates decompressing small files.
- New feature for igzip cli tool: support for concatenated .gz files.
- CRC Improvements
- New vclmul version of crc32_iscsi().
- Updates for aarch64.
- Other fixes
- Intel CET support.
- Windows nasm support.
- AES-GCM, AES-XTS for next gen CPU
- New optimized versions of multi-buffer SM3 hashing.
- EC performance improvements for 6+ parity
- CRC32 and CRC16 next gen CPU
- CRC performance improvements for next gen CPU
- Multi-buffer SM3 functions. Experimental base functions only.
- Multi-architecture support
- Threaded igzip
- Adler32 function
- Multi-arch improvements
- Performance test improvements
- Various fixes for sanitizer warnings
- Compression on random file performance improvements
- Additional small file compression improvements
- Multi-arch autoconf
- Small file compression performance improvements
- igzip gzip/zlib header/trailer handling
- igzip userspace tool/example
- Zero detect memory function
- Decompression Performance Improvements (Multi Byte Decode)
- Level 3 Compression performance improvements
- Better EC Examples
- Level 2 & 3 Compression Level added. Level 3 is currently only optimized for processors with AVX512 instructions.
- New T10dif CRC and Copy function
- CRC32 iSCSI performance improvements
- Multi-buffer hash performance improvement for Intel Atom processors.
- New base functions for multi-buffer hashes.
- Multi Buffer Hashing : extend the max size to > 4 GB
- Add Non Temporal(NT) runtime versions of AES_GCM
- Add Saturation Benchmark for determining the saturation point for each algorithm like MBH, CRC, AES and EC.
- Rolling Hash
- Multi Buffer Hash improvements
- AES GCM : super small packet performance improvements
- Adler and zlib format for Inflate\Deflate
- New 2-pass fully-dynamic deflate compression (level 0-1).
- Base functions for multi-architecture support
- Multi-hash SHA-256 based
- RAID AVX512 versions
ISA-L v2.17 features summary
- Fast Decompression (Inflate) - New.
- Compression Improvements (Deflate) - speed, compression ratio improvements and new features.
- Fast Custom Huffman Code generation – creates optimized Huffman table based on some sample input.
- CRC64