Commit Graph

25 Commits

Author SHA1 Message Date
Denys Vlasenko
09a0e2223f sha3: tweak comments and indentation
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2014-07-30 16:26:09 +02:00
Denys Vlasenko
4ff933c0e7 sha3: code shrink (and speedup for SHA3_SMALL=0)
function                                             old     new   delta
sha3_process_block72                                1454    1359     -95

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2014-07-30 14:18:57 +02:00
Denys Vlasenko
2a563ea49a sha3: add 32-bit optimized bit-sliced implementation
It is an interesting trick, but so far I only managed to make it work
correctly, not to make it faster and/or smaller.
The code is ifdefed out for now.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2014-07-25 17:24:13 +02:00
Denys Vlasenko
1f5e81f8f8 md5/sha512: a better fix for strict aliasing warnings
The locations *are* well-aligned for direct stores
on any architecture.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-06-27 01:03:19 +02:00
Mike Frysinger
f6205c6ab7 md5/sha512: fix strict aliasing warnings
If the target can tolerate these issues, then gcc is smart enough
to generate the same code (x86_64 produces the same code).  If the
target can't, then it needs the memcpy anyways.

libbb/hash_md5_sha.c: In function 'common64_end':
libbb/hash_md5_sha.c:87:4: warning:
  dereferencing type-punned pointer will break strict-aliasing rules
    *(uint64_t *) (&ctx->wbuffer[64 - 8]) = t;
libbb/hash_md5_sha.c: In function 'sha512_end':
libbb/hash_md5_sha.c:886:4: warning:
  dereferencing type-punned pointer will break strict-aliasing rules
    *(uint64_t *) (&ctx->wbuffer[128 - 8]) = t;
libbb/hash_md5_sha.c:889:4: warning:
  dereferencing type-punned pointer will break strict-aliasing rules
    *(uint64_t *) (&ctx->wbuffer[128 - 16]) = t;

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2013-06-19 11:49:15 -04:00
Denys Vlasenko
2cfcc9e9d7 sha3: code shrink
function                                             old     new   delta
sha3_hash                                            155     101     -54

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-20 00:38:09 +01:00
Denys Vlasenko
e4f0f26bad sha3: s/sha3_process_block76/sha3_process_block72/
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-16 12:23:23 +01:00
Denys Vlasenko
5368fe541c sha3: rename KeccakF->sha3_process_block76.
This brings the naming more in line with other hashes.
Pulled most statics and constants into it.
Also noticed that two byte arrays are 1 element too big.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-16 02:20:31 +01:00
Denys Vlasenko
970aa6b5bd sha3: cache ctx->bytes_queued
function                                             old     new   delta
sha3_hash                                            171     155     -16

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 22:19:24 +01:00
Denys Vlasenko
8fb3ab528e sha3: remove two "small code" codepaths: I can't reproduce code size win on them anymore
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 22:07:48 +01:00
Denys Vlasenko
8e7312e12f sha3: tweak choice of a fast code path for 64-bit
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 21:50:41 +01:00
Denys Vlasenko
5b7f50f372 sha3: cosmetic tweaks to various names, comments. No logic changes.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 19:52:30 +01:00
Denys Vlasenko
ac4100e103 sha3: code shrink
function                                             old     new   delta
KeccakF                                             1053    1078     +25
KeccakF_RoundConstants                               192      48    -144

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 16:27:39 +01:00
Denys Vlasenko
a55df27936 sha3: code shrink
function                                             old     new   delta
KeccakF                                             1064    1053     -11

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 15:22:30 +01:00
Denys Vlasenko
07a54e21dd sha3: another speedup for SHA3_SMALL=0 case
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 14:47:05 +01:00
Denys Vlasenko
30a8652fbf sha3: make size/speed optimization decision configurable
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-15 01:12:26 +01:00
Lauri Kasanen
b8173b603f sha3sum: new applet
function                                             old     new   delta
KeccakF                                                -     496    +496
KeccakF_RoundConstants                                 -     192    +192
sha3_hash                                              -     171    +171
sha3_end                                               -      40     +40
hash_file                                            274     299     +25
KeccakF_RotationConstants                              -      25     +25
KeccakF_PiLane                                         -      25     +25
packed_usage                                       29213   29232     +19
sha3_begin                                             -      18     +18
KeccakF_Mod5                                           -      10     +10
applet_names                                        2445    2453      +8
applet_main                                         1420    1424      +4
applet_nameofs                                       710     712      +2
------------------------------------------------------------------------------
(add/remove: 8/0 grow/shrink: 9/7 up/down: 1049/-54)         Total: ~995 bytes

Signed-off-by: Lauri Kasanen <curaga@operamail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2013-01-14 05:20:50 +01:00
Denys Vlasenko
522041ee7b regularize options which control size/speed trade
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2011-09-10 13:25:57 +02:00
Denys Vlasenko
fb132e4737 whitespace cleanup
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
2010-10-29 11:46:52 +02:00
Denys Vlasenko
03a5fe378e sha1: small tweak for clearer code, no logic changes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2010-10-24 20:51:28 +02:00
Denys Vlasenko
f4c93ab304 sha1: use Rob's code, it's smaller and faster
function                                             old     new   delta
static.rconsts                                         -      16     +16
sha1_process_block64                                 460     298    -162
------------------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 0/1 up/down: 16/-162)          Total: -146 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2010-10-24 19:27:30 +02:00
Denys Vlasenko
7ab94ca351 md5: remove outdated comment
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2010-10-19 02:33:39 +02:00
Denys Vlasenko
302ad1450e libbb/hash_md5_sha: use common ctx and code for md5 and sha1/256
function                                             old     new   delta
sha256_process_block64                               421     433     +12
md5_crypt                                            578     587      +9
md5_begin                                             43      50      +7
md5_hash                                              99      97      -2
sha1_end                                              85      82      -3
md5_end                                               36      31      -5
common64_end                                          93      86      -7
sha1_hash                                             97       -     -97

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2010-10-19 02:16:12 +02:00
Denys Vlasenko
c48a5c607d hash_md5_sha: use common finalization routine for MD5 and sha1/256. -15 bytes
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
2010-10-18 14:48:30 +02:00
Denys Vlasenko
b5aa1d95a1 libbb/hash_sha.c -> libbb/hash_md5_sha.c
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
2010-10-18 13:47:47 +02:00