1 =============================================================================
2 v2.16 Intel Intelligent Storage Acceleration Library Release Notes
4 =============================================================================
6 =============================================================================
8 =============================================================================
11 3. CHANGE LOG & FEATURES ADDED
13 =============================================================================
15 =============================================================================
17 * Only erasure code unit included in open source version at this time.
19 * Perf tests do not run in Windows environment.
21 * 32-bit lib is not supported in Windows.
23 =============================================================================
25 =============================================================================
28 * Fix for windows register save in gf_6vect_mad_avx2.asm. Only affects windows
29 versions of ec_encode_data_update() running with AVX2. A GP register was not
30 properly restored resulting in corruption on return.
34 * Building in unit directories is no longer supported removing the issue of
35 leftover object files causing the top-level make build to fail.
39 * Fix for windows register save overlap in gf_{3-6}vect_dot_prod_sse.asm. Only
40 affects windows versions of erasure code. GP register saves/restore were
41 pushed to same stack area as XMM.
43 =============================================================================
44 3. CHANGE LOG & FEATURES ADDED
45 =============================================================================
48 * Units added: crc, raid, igzip (deflate compression).
52 * Erasure code updates. New AVX512 versions.
54 * Nasm support. ISA-L ported to build with nasm or yasm assembler.
56 * Windows DLL support. Windows builds DLL by default.
60 * Autoconf and autotools build allows easier porting to additional systems.
61 Previous make system still available to embedded users with Makefile.unx.
63 * Includes update for building on Mac OS X/darwin systems. Add --target=darwin
68 * Erasure code improvments
69 - 32-bit port of optimized gf_vect_dot_prod() functions. This makes
70 ec_encode_data() functions much faster on 32-bit processors.
71 - Avoton performance improvements. Performance on Avoton for
72 gf_vect_dot_prod() and ec_encode_data() can improve by as much as 20%.
76 * Incremental erasure code. New functions added to erasure code to handle
77 single source update of code blocks. The function ec_encode_data_update()
78 works with parameters similar to ec_encode_data() but are called incrementally
79 with each source block. These versions are useful when source blocks are not
80 all available at once.
84 * Erasure code updates
85 - New AVX and AVX2 support functions.
86 - Changes min len requirement on gf_vect_dot_prod() to 32 from 16.
87 - Tests include both source and parity recovery with ec_encode_data().
88 - New encoding examples with Vandermonde or Cauchy matrix.
92 * First open release of erasure code unit that is part of ISA-L.