emo/busybox
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

sha3: code shrink (and speedup for SHA3_SMALL=0)

Browse Source 
function                                             old     new   delta
sha3_process_block72                                1454    1359     -95

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
This commit is contained in:
Denys Vlasenko 2014-07-30 14:18:57 +02:00
parent 6c6d37ee45
commit 4ff933c0e7
1 changed files with 66 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

82
libbb/hash_md5_sha.c
View File

@ -1141,22 +1141,30 @@ static void sha3_process_block72(uint64_t *state)
#undef RhoPi #undef RhoPi
} }
/* Chi */ /* Chi */
for (x = 0; x <= 20; x += 5) { for (x = 0; x <= 40;) {
/* uint32_t BC0, BC1, BC2, BC3, BC4;
* Can write this in terms of uint32 too, BC0 = s32[x + 0*2];
* but why? compiler does it automatically. BC1 = s32[x + 1*2];
*/ BC2 = s32[x + 2*2];
uint64_t BC0, BC1, BC2, BC3, BC4; s32[x + 0*2] = BC0 ^ ((~BC1) & BC2);
BC0 = state[x + 0]; BC3 = s32[x + 3*2];
BC1 = state[x + 1]; s32[x + 1*2] = BC1 ^ ((~BC2) & BC3);
BC2 = state[x + 2]; BC4 = s32[x + 4*2];
state[x + 0] = BC0 ^ ((~BC1) & BC2); s32[x + 2*2] = BC2 ^ ((~BC3) & BC4);
BC3 = state[x + 3]; s32[x + 3*2] = BC3 ^ ((~BC4) & BC0);
state[x + 1] = BC1 ^ ((~BC2) & BC3); s32[x + 4*2] = BC4 ^ ((~BC0) & BC1);
BC4 = state[x + 4]; x++;
state[x + 2] = BC2 ^ ((~BC3) & BC4); BC0 = s32[x + 0*2];
state[x + 3] = BC3 ^ ((~BC4) & BC0); BC1 = s32[x + 1*2];
state[x + 4] = BC4 ^ ((~BC0) & BC1); BC2 = s32[x + 2*2];
s32[x + 0*2] = BC0 ^ ((~BC1) & BC2);
BC3 = s32[x + 3*2];
s32[x + 1*2] = BC1 ^ ((~BC2) & BC3);
BC4 = s32[x + 4*2];
s32[x + 2*2] = BC2 ^ ((~BC3) & BC4);
s32[x + 3*2] = BC3 ^ ((~BC4) & BC0);
s32[x + 4*2] = BC4 ^ ((~BC0) & BC1);
x += 9;
} }
/* Iota */ /* Iota */
s32[0] ^= IOTA_CONST_0bits & 1; s32[0] ^= IOTA_CONST_0bits & 1;
@ -1275,6 +1283,7 @@ static void sha3_process_block72(uint64_t *state)
#undef RhoPi_twice #undef RhoPi_twice
} }
/* Chi */ /* Chi */
#if LONG_MAX > 0x7fffffff
for (x = 0; x <= 20; x += 5) { for (x = 0; x <= 20; x += 5) {
uint64_t BC0, BC1, BC2, BC3, BC4; uint64_t BC0, BC1, BC2, BC3, BC4;
BC0 = state[x + 0]; BC0 = state[x + 0];
@ -1288,6 +1297,47 @@ static void sha3_process_block72(uint64_t *state)
state[x + 3] = BC3 ^ ((~BC4) & BC0); state[x + 3] = BC3 ^ ((~BC4) & BC0);
state[x + 4] = BC4 ^ ((~BC0) & BC1); state[x + 4] = BC4 ^ ((~BC0) & BC1);
} }
#else
/* Reduced register pressure version
* for register-starved 32-bit arches
* (i386: -95 bytes, and it is _faster_)
*/
for (x = 0; x <= 40;) {
uint32_t BC0, BC1, BC2, BC3, BC4;
uint32_t *const s32 = (uint32_t*)state;
# if SHA3_SMALL
do_half:
#endif
BC0 = s32[x + 0*2];
BC1 = s32[x + 1*2];
BC2 = s32[x + 2*2];
s32[x + 0*2] = BC0 ^ ((~BC1) & BC2);
BC3 = s32[x + 3*2];
s32[x + 1*2] = BC1 ^ ((~BC2) & BC3);
BC4 = s32[x + 4*2];
s32[x + 2*2] = BC2 ^ ((~BC3) & BC4);
s32[x + 3*2] = BC3 ^ ((~BC4) & BC0);
s32[x + 4*2] = BC4 ^ ((~BC0) & BC1);
x++;
# if SHA3_SMALL
if (x & 1)
goto do_half;
x += 8;
# else
BC0 = s32[x + 0*2];
BC1 = s32[x + 1*2];
BC2 = s32[x + 2*2];
s32[x + 0*2] = BC0 ^ ((~BC1) & BC2);
BC3 = s32[x + 3*2];
s32[x + 1*2] = BC1 ^ ((~BC2) & BC3);
BC4 = s32[x + 4*2];
s32[x + 2*2] = BC2 ^ ((~BC3) & BC4);
s32[x + 3*2] = BC3 ^ ((~BC4) & BC0);
s32[x + 4*2] = BC4 ^ ((~BC0) & BC1);
x += 9;
# endif
}
#endif
/* Iota */ /* Iota */
state[0] ^= IOTA_CONST[round] state[0] ^= IOTA_CONST[round]
| (uint32_t)((IOTA_CONST_bit31 << round) & 0x80000000) | (uint32_t)((IOTA_CONST_bit31 << round) & 0x80000000)