ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/liblzf/Changes
Revision: 1.38
Committed: Tue Nov 13 21:14:15 2007 UTC (16 years, 6 months ago) by root
Branch: MAIN
CVS Tags: rel-3_0
Changes since 1.37: +1 -1 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 3.0 Tue Nov 13 22:13:09 CET 2007
2 - switched to 2-clause bsd with "GPL v2 or any later version" option.
3 - speed up compression by ~10-15% in common cases
4 by some manual unrolling.
5 - import some compiler tricks from JSON::XS, for further speed-ups.
6 - tune hash functions depending on ULTRA_FAST or VERY_FAST settings.
7 - for typical binary data (e.g. /bin/bash, memory dumps,
8 canterbury corpus etc.), speed is now comparable to fastlz, but
9 with better compression ratio. with ULTRA_FAST, it's typically
10 3-15% faster than fastlz while still maintaining a similar ratio.
11 (amd64 and core 2 duo, ymmv). thanks a lot for the competition :)
12 - undo inline assembly in compressor, it is no longer helpful.
13 - no changes to the decompressor.
14 - use a HLOG of 16 by default now (formerly 15).
15
16 2.1 Fri Nov 2 13:34:42 CET 2007
17 - switched to a 2-clause bsd license with GPL exception.
18 - get rid of memcpy.
19 - tentatively use rep movsb on x86 and x86_64 (gcc only) for a
20 moderate speed improvement.
21 - applied patch by Kein-Hong Man to maske lzf.c compile under
22 the crippled mingw32 environment.
23
24 2.0 Fri Feb 16 23:11:18 CET 2007
25 - replaced lzf demo by industrial-strength lzf utility with behaviour
26 similar other compression utilities. Thanks for Stefan Traby for
27 rewriting it!
28 - fix state arg prototype.
29
30 1.7 Wed Sep 27 17:29:15 CEST 2006
31 - remove bogus "unlzf" patch.
32 note to self: never accept well-meant patches.
33 - make lzf more robust in presence of padding bytes or sudden eof.
34
35 1.6 Fri Jul 7 17:31:26 CEST 2006
36 - the lzf example utility will now uncompress if invoked
37 as "unlzf" (patch by Scott Feeney).
38 - add CHECK_INPUT option that adds more checks for input
39 data validity.
40 - help applications that do not pass in the correct length
41 (such as php) by returning either EINVAL or E2BIG.
42 - default HLOG size is now 15 (cpu caches have increased).
43 - documentation fixes.
44
45 1.51 Thu Apr 14 22:15:46 CEST 2005
46 - incorporated C♯ implementation of both the en- and decoder,
47 written by "Oren J. Maurice <oymaurice@hazorea.org.il>".
48 You can find it in the cs/ subdirectory.
49 - make FRST, NEXT IDX overridable if lzf_c.c is directly included
50 in the code.
51
52 1.5 Tue Mar 8 20:23:23 CET 2005
53 - incorporated improvements by Adam D. Moss,
54 which includes a new VERY_FAST mode which is
55 a bit slower than ULTRA_FAST but much better,
56 and enabled it as default.
57
58 1.401 Thu Mar 3 18:00:52 CET 2005
59 - use cstring in c++, not string.h.
60 - change of contact address.
61
62 1.4 Wed Dec 15 08:08:49 CET 2004
63 - very very slight tuning of the hashing function.
64
65 1.3 Thu Mar 25 15:41:17 CET 2004
66 - changed license of lzf core code to explicitly allow
67 relicensing under the GPLv2.
68 - added VPATH support as suggested by Björn Eriksson.
69
70 1.2 Mon Dec 29 13:47:28 CET 2003
71 - avoid spurious memory accesses after the to-be-compressed
72 memory region. originally reported by Michal Zalewski.
73 - flip LZF_STACK_ARG meaning (to be correct).
74
75 1.1 Tue Dec 23 05:48:32 CET 2003
76 - removed #warn directive, it's not worth the hassle.
77 - add LZF_STACK_ARG and AVOID_ERRNO configurations
78 for embedded systems.
79 - make it compile cleanly as c++.
80 - some small documentation and code fixes.
81
82 1.0 Sun Nov 17 12:37:37 CET 2002
83 - slightly better compression ratio, almost unmeasurably
84 slower.
85 - some documentation fixes.
86
87 0.4 Thu Jun 13 14:11:10 CEST 2002
88 - typoe fix.
89 - lzf demo program now properly decompresses small files.
90 - fix another 64 bit issue, found by Laurent Deniel.
91
92 0.3 Tue Jan 16 13:21:14 CET 2001
93 - fix silly beginners 32/64 bit mistake.
94
95 0.2 Thu Jan 4 05:56:42 CET 2001
96 - now totally independent of autoconfig, for
97 easy inclusion into other programs.
98 - much better fine-tuning, faster and better than 0.1.
99
100 0.1 2000
101 - initial release.