wbrawner/openssl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
9506 commits 20 branches 302 tags 153 MiB
Commit graph

29 commits

Author SHA1 Message Date
Andy Polyakov
8525950e7e ghash-x86.pl: "528B" variant of gcm_ghash_4bit_mmx gives 20-40% 
improvement.
 2010-06-04 13:21:01 +00:00
Andy Polyakov
e747f4d418 gcm128.c: P.-M. Hager has tipped about possibility to fold reductions 
in gcm_ghash_4bit. Taking the idea a step further I've added extra
256+16 bytes of per-key storage, so that one can speak about 3rd variant
in addition to "256B" and "4KB": "528B" one. Commonly it should be
~50% faster than "256B" implementation or ~25% slower than "4KB" one.
 2010-05-26 21:36:36 +00:00
Andy Polyakov
07e29c1234 ghash-x86.pl: MMX optimization (+20-40%) and commentary update. 2010-05-23 12:37:01 +00:00
Andy Polyakov
fb2d5a91e9 gcm128.c: commentary update. 2010-05-23 12:35:41 +00:00
Andy Polyakov
1aa8a6297c ghash-x86[_64].pl: add due credit. 2010-05-13 17:21:52 +00:00
Andy Polyakov
c1f092d14e GCM "jumbo" update: 
- gcm128.c: support for Intel PCLMULQDQ, readability improvements;
- asm/ghash-x86.pl: splitted vanilla, MMX, PCLMULQDQ subroutines;
- asm/ghash-x86_64.pl: add PCLMULQDQ implementations.
 2010-05-13 15:32:43 +00:00
Andy Polyakov
f472ec8c2f "Jumbo" update for crypto/modes: 
- introduce common modes_lcl.h;
- ctr128.c: implement additional CRYPTO_ctr128_encrypt_ctr32 interface;
- gcm128.c: add omitted ARM initialization, remove ctx.ctr;
 2010-05-04 19:23:02 +00:00
Andy Polyakov
8a682556b4 Add ghash-armv4.pl. 2010-05-03 18:23:29 +00:00
Ben Laurie
19f7e5e24a Missing declarations, no assembler in PEDANTIC. 2010-05-01 14:41:25 +00:00
Andy Polyakov
5e19ee96f6 Add ghash-parisc.pl. 2010-04-28 18:51:45 +00:00
Andy Polyakov
8a1c92ce49 Take gcm128.c and ghash assembler modules into the build loop. 2010-04-22 21:36:26 +00:00
Andy Polyakov
4f39edbff1 gcm128.c and assembler modules: change argument order for gcm_ghash_4bit. 
ghash-x86*.pl: fix performance numbers for Core2, as it turned out
previous ones were "tainted" by variable clock frequency.
 2010-04-14 19:04:51 +00:00
Andy Polyakov
6a1823b371 [co]fb128.c: fix "n=0" bug. 2010-04-14 07:47:28 +00:00
Andy Polyakov
a595baff9f gcm128.c: commentary and formatting updates. 2010-04-10 14:02:26 +00:00
Andy Polyakov
67a315b60b cts128.c: add support for NIST "Ciphertext Stealing" proposal. 2010-04-10 14:01:02 +00:00
Andy Polyakov
f62df694ad ctr129.c: fix typo, simplify ctr128_inc and fix "n=0" bug. 2010-04-10 13:46:53 +00:00
Andy Polyakov
42feba4797 Add ghash-alpha.pl assembler module. 2010-04-10 13:44:20 +00:00
Andy Polyakov
c3473126b1 GHASH assembler: new ghash-sparcv9.pl module and saner descriptions. 2010-03-22 17:24:18 +00:00
Andy Polyakov
480cd6ab6e ghash-ia64.pl: new file, GHASH for Itanium. 
ghash-x86_64.pl: minimize stack frame usage.
ghash-x86.pl: modulo-scheduling MMX loop in respect to input vector
results in up to 10% performance improvement.
 2010-03-15 19:07:52 +00:00
Andy Polyakov
f093794e55 Add GHASH x86_64 assembler. 2010-03-11 16:19:46 +00:00
Andy Polyakov
e3a510f8a6 Add GHASH x86 assembler. 2010-03-09 23:03:33 +00:00
Andy Polyakov
2262beef2e gcm128.c: add option for streamed GHASH, simple benchmark, minor naming 
change.
 2010-03-08 22:44:37 +00:00
Andy Polyakov
e7f5b1cd42 Initial version of Galois Counter Mode implementation. Interface is still 
subject to change...
 2010-03-02 16:33:25 +00:00
Dr. Stephen Henson
918a5d04e4 PR: 2164 
Submitted by: "Noszticzius, Istvan" <inoszticzius@rightnow.com>

Don't clear the output buffer: ciphers should correctly the same input
and output buffers.
 2010-02-15 19:00:12 +00:00
Andy Polyakov
5cabcf96e7 Fix "possible loss of data" Win64 compiler warnings. 2008-12-29 12:35:49 +00:00
Dr. Stephen Henson
2d1cbc85c8 Add standard .cvsignore file. 2008-12-29 00:27:06 +00:00
Andy Polyakov
fe150ac25d Add modes/cts128.c, Ciphertext Stealing implementation. 2008-12-27 13:40:45 +00:00
Andy Polyakov
63fc7f848d crypto/modes: make modes.h selfsufficient and rename block_f to block128_t. 2008-12-23 11:18:45 +00:00
Andy Polyakov
85b2c0ce7f 128-bit block cipher modes consolidation. As consolidated functions 
rely on indirect call to block functions, they are not as fast as
non-consolidated routines. However, performance loss(*) is within
measurement error and consolidation advantages are considered to
outweigh it.

(*) actually one can observe performance *improvement* on e.g.
    CBC benchmarks thanks to optimization, which also becomes
    shared among ciphers.
 2008-12-16 08:39:21 +00:00