arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-29 11:28:15 +00:00
Code Issues Releases Wiki Activity

Get rid of extract_byte() in inc_cipher_twofish.cl

Browse Source
This commit is contained in:
Jens Steube 2019-04-23 15:16:42 +02:00
parent bf4b1a8e02
commit f10d27b2c9
1 changed files with 77 additions and 79 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

132
OpenCL/inc_cipher_twofish.cl
View File

@ -255,64 +255,62 @@ CONSTANT_AS u32a m_tab[4][256] =
0xECC94AEC, 0xFDD25EFD, 0xAB7FC1AB, 0xD8A8E0D8 } 0xECC94AEC, 0xFDD25EFD, 0xAB7FC1AB, 0xD8A8E0D8 }
}; };
#define extract_byte(x,n) (((x) >> (8 * (n))) & 0xff)
#define g1_fun128(x) \ #define g1_fun128(x) \
(mds (0, q20 (extract_byte (x, 3), sk)) ^ \ (mds (0, q20 (unpack_v8d_from_v32_S (x), sk)) ^ \
mds (1, q21 (extract_byte (x, 0), sk)) ^ \ mds (1, q21 (unpack_v8a_from_v32_S (x), sk)) ^ \
mds (2, q22 (extract_byte (x, 1), sk)) ^ \ mds (2, q22 (unpack_v8b_from_v32_S (x), sk)) ^ \
mds (3, q23 (extract_byte (x, 2), sk))) mds (3, q23 (unpack_v8c_from_v32_S (x), sk)))
#define g0_fun128(x) \ #define g0_fun128(x) \
(mds (0, q20 (extract_byte (x, 0), sk)) ^ \ (mds (0, q20 (unpack_v8a_from_v32_S (x), sk)) ^ \
mds (1, q21 (extract_byte (x, 1), sk)) ^ \ mds (1, q21 (unpack_v8b_from_v32_S (x), sk)) ^ \
mds (2, q22 (extract_byte (x, 2), sk)) ^ \ mds (2, q22 (unpack_v8c_from_v32_S (x), sk)) ^ \
mds (3, q23 (extract_byte (x, 3), sk))) mds (3, q23 (unpack_v8d_from_v32_S (x), sk)))
#define f_rnd128(i) \ #define f_rnd128(i) \
{ \ { \
u32 t0 = g0_fun128 (data[0]); \ const u32 t0 = g0_fun128 (data[0]); \
u32 t1 = g1_fun128 (data[1]); \ const u32 t1 = g1_fun128 (data[1]); \
data[2] = hc_rotr32_S (data[2] ^ (t0 + t1 + lk[4 * (i) + 8]), 1); \ data[2] = hc_rotr32_S (data[2] ^ (t0 + t1 + lk[4 * (i) + 8]), 1); \
data[3] = hc_rotl32_S (data[3], 1) ^ (t0 + 2 * t1 + lk[4 * (i) + 9]); \ data[3] = hc_rotl32_S (data[3], 1) ^ (t0 + 2 * t1 + lk[4 * (i) + 9]); \
u32 t2 = g0_fun128 (data[2]); \ const u32 t2 = g0_fun128 (data[2]); \
u32 t3 = g1_fun128 (data[3]); \ const u32 t3 = g1_fun128 (data[3]); \
data[0] = hc_rotr32_S (data[0] ^ (t2 + t3 + lk[4 * (i) + 10]), 1); \ data[0] = hc_rotr32_S (data[0] ^ (t2 + t3 + lk[4 * (i) + 10]), 1); \
data[1] = hc_rotl32_S (data[1], 1) ^ (t2 + 2 * t3 + lk[4 * (i) + 11]); \ data[1] = hc_rotl32_S (data[1], 1) ^ (t2 + 2 * t3 + lk[4 * (i) + 11]); \
} }
#define i_rnd128(i) \ #define i_rnd128(i) \
{ \ { \
u32 t0 = g0_fun128 (data[0]); \ const u32 t0 = g0_fun128 (data[0]); \
u32 t1 = g1_fun128 (data[1]); \ const u32 t1 = g1_fun128 (data[1]); \
data[2] = hc_rotl32_S (data[2], 1) ^ (t0 + t1 + lk[4 * (i) + 10]); \ data[2] = hc_rotl32_S (data[2], 1) ^ (t0 + t1 + lk[4 * (i) + 10]); \
data[3] = hc_rotr32_S (data[3] ^ (t0 + 2 * t1 + lk[4 * (i) + 11]), 1); \ data[3] = hc_rotr32_S (data[3] ^ (t0 + 2 * t1 + lk[4 * (i) + 11]), 1); \
u32 t2 = g0_fun128 (data[2]); \ const u32 t2 = g0_fun128 (data[2]); \
u32 t3 = g1_fun128 (data[3]); \ const u32 t3 = g1_fun128 (data[3]); \
data[0] = hc_rotl32_S (data[0], 1) ^ (t2 + t3 + lk[4 * (i) + 8]); \ data[0] = hc_rotl32_S (data[0], 1) ^ (t2 + t3 + lk[4 * (i) + 8]); \
data[1] = hc_rotr32_S (data[1] ^ (t2 + 2 * t3 + lk[4 * (i) + 9]), 1); \ data[1] = hc_rotr32_S (data[1] ^ (t2 + 2 * t3 + lk[4 * (i) + 9]), 1); \
} }
#define f_rnd256(i) \ #define f_rnd256(i) \
{ \ { \
u32 t0 = g0_fun256 (data[0]); \ const u32 t0 = g0_fun256 (data[0]); \
u32 t1 = g1_fun256 (data[1]); \ const u32 t1 = g1_fun256 (data[1]); \
data[2] = hc_rotr32_S (data[2] ^ (t0 + t1 + lk[4 * (i) + 8]), 1); \ data[2] = hc_rotr32_S (data[2] ^ (t0 + t1 + lk[4 * (i) + 8]), 1); \
data[3] = hc_rotl32_S (data[3], 1) ^ (t0 + 2 * t1 + lk[4 * (i) + 9]); \ data[3] = hc_rotl32_S (data[3], 1) ^ (t0 + 2 * t1 + lk[4 * (i) + 9]); \
u32 t2 = g0_fun256 (data[2]); \ const u32 t2 = g0_fun256 (data[2]); \
u32 t3 = g1_fun256 (data[3]); \ const u32 t3 = g1_fun256 (data[3]); \
data[0] = hc_rotr32_S (data[0] ^ (t2 + t3 + lk[4 * (i) + 10]), 1); \ data[0] = hc_rotr32_S (data[0] ^ (t2 + t3 + lk[4 * (i) + 10]), 1); \
data[1] = hc_rotl32_S (data[1], 1) ^ (t2 + 2 * t3 + lk[4 * (i) + 11]); \ data[1] = hc_rotl32_S (data[1], 1) ^ (t2 + 2 * t3 + lk[4 * (i) + 11]); \
} }
#define i_rnd256(i) \ #define i_rnd256(i) \
{ \ { \
u32 t0 = g0_fun256 (data[0]); \ const u32 t0 = g0_fun256 (data[0]); \
u32 t1 = g1_fun256 (data[1]); \ const u32 t1 = g1_fun256 (data[1]); \
data[2] = hc_rotl32_S (data[2], 1) ^ (t0 + t1 + lk[4 * (i) + 10]); \ data[2] = hc_rotl32_S (data[2], 1) ^ (t0 + t1 + lk[4 * (i) + 10]); \
data[3] = hc_rotr32_S (data[3] ^ (t0 + 2 * t1 + lk[4 * (i) + 11]), 1); \ data[3] = hc_rotr32_S (data[3] ^ (t0 + 2 * t1 + lk[4 * (i) + 11]), 1); \
u32 t2 = g0_fun256 (data[2]); \ const u32 t2 = g0_fun256 (data[2]); \
u32 t3 = g1_fun256 (data[3]); \ const u32 t3 = g1_fun256 (data[3]); \
data[0] = hc_rotl32_S (data[0], 1) ^ (t2 + t3 + lk[4 * (i) + 8]); \ data[0] = hc_rotl32_S (data[0], 1) ^ (t2 + t3 + lk[4 * (i) + 8]); \
data[1] = hc_rotr32_S (data[1] ^ (t2 + 2 * t3 + lk[4 * (i) + 9]), 1); \ data[1] = hc_rotr32_S (data[1] ^ (t2 + 2 * t3 + lk[4 * (i) + 9]), 1); \
} }
@ -321,15 +319,15 @@ CONSTANT_AS u32a m_tab[4][256] =
#define mds(n,x) m_tab[n][x] #define mds(n,x) m_tab[n][x]
#define q20(x,k) q (0, q (0, x) ^ extract_byte (k[1], 0)) ^ extract_byte (k[0], 0) #define q20(x,k) q (0, q (0, x) ^ unpack_v8a_from_v32_S (k[1])) ^ unpack_v8a_from_v32_S (k[0])
#define q21(x,k) q (0, q (1, x) ^ extract_byte (k[1], 1)) ^ extract_byte (k[0], 1) #define q21(x,k) q (0, q (1, x) ^ unpack_v8b_from_v32_S (k[1])) ^ unpack_v8b_from_v32_S (k[0])
#define q22(x,k) q (1, q (0, x) ^ extract_byte (k[1], 2)) ^ extract_byte (k[0], 2) #define q22(x,k) q (1, q (0, x) ^ unpack_v8c_from_v32_S (k[1])) ^ unpack_v8c_from_v32_S (k[0])
#define q23(x,k) q (1, q (1, x) ^ extract_byte (k[1], 3)) ^ extract_byte (k[0], 3) #define q23(x,k) q (1, q (1, x) ^ unpack_v8d_from_v32_S (k[1])) ^ unpack_v8d_from_v32_S (k[0])
#define q40(x,k) q (0, q (0, q (1, q (1, x) ^ extract_byte (k[3], 0)) ^ extract_byte (k[2], 0)) ^ extract_byte (k[1], 0)) ^ extract_byte (k[0], 0) #define q40(x,k) q (0, q (0, q (1, q (1, x) ^ unpack_v8a_from_v32_S (k[3])) ^ unpack_v8a_from_v32_S (k[2])) ^ unpack_v8a_from_v32_S (k[1])) ^ unpack_v8a_from_v32_S (k[0])
#define q41(x,k) q (0, q (1, q (1, q (0, x) ^ extract_byte (k[3], 1)) ^ extract_byte (k[2], 1)) ^ extract_byte (k[1], 1)) ^ extract_byte (k[0], 1) #define q41(x,k) q (0, q (1, q (1, q (0, x) ^ unpack_v8b_from_v32_S (k[3])) ^ unpack_v8b_from_v32_S (k[2])) ^ unpack_v8b_from_v32_S (k[1])) ^ unpack_v8b_from_v32_S (k[0])
#define q42(x,k) q (1, q (0, q (0, q (0, x) ^ extract_byte (k[3], 2)) ^ extract_byte (k[2], 2)) ^ extract_byte (k[1], 2)) ^ extract_byte (k[0], 2) #define q42(x,k) q (1, q (0, q (0, q (0, x) ^ unpack_v8c_from_v32_S (k[3])) ^ unpack_v8c_from_v32_S (k[2])) ^ unpack_v8c_from_v32_S (k[1])) ^ unpack_v8c_from_v32_S (k[0])
#define q43(x,k) q (1, q (1, q (0, q (1, x) ^ extract_byte (k[3], 3)) ^ extract_byte (k[2], 3)) ^ extract_byte (k[1], 3)) ^ extract_byte (k[0], 3) #define q43(x,k) q (1, q (1, q (0, q (1, x) ^ unpack_v8d_from_v32_S (k[3])) ^ unpack_v8d_from_v32_S (k[2])) ^ unpack_v8d_from_v32_S (k[1])) ^ unpack_v8d_from_v32_S (k[0])
DECLSPEC u32 mds_rem (u32 p0, u32 p1) DECLSPEC u32 mds_rem (u32 p0, u32 p1)
{ {
@ -337,7 +335,7 @@ DECLSPEC u32 mds_rem (u32 p0, u32 p1)
#define MDS_REM_ROUND() \ #define MDS_REM_ROUND() \
{ \ { \
u32 t = p1 >> 24; \ const u32 t = p1 >> 24; \
p1 = (p1 << 8) | (p0 >> 24); \ p1 = (p1 << 8) | (p0 >> 24); \
p0 <<= 8; \ p0 <<= 8; \
u32 u = (t << 1); \ u32 u = (t << 1); \
@ -366,15 +364,15 @@ DECLSPEC u32 h_fun128 (const u32 x, const u32 *key)
{ {
u32 b0, b1, b2, b3; u32 b0, b1, b2, b3;
b0 = extract_byte (x, 0); b0 = unpack_v8a_from_v32_S (x);
b1 = extract_byte (x, 1); b1 = unpack_v8b_from_v32_S (x);
b2 = extract_byte (x, 2); b2 = unpack_v8c_from_v32_S (x);
b3 = extract_byte (x, 3); b3 = unpack_v8d_from_v32_S (x);
b0 = q (0, (q (0, b0) ^ extract_byte (key[1], 0))) ^ extract_byte (key[0], 0); b0 = q (0, (q (0, b0) ^ unpack_v8a_from_v32_S (key[1]))) ^ unpack_v8a_from_v32_S (key[0]);
b1 = q (0, (q (1, b1) ^ extract_byte (key[1], 1))) ^ extract_byte (key[0], 1); b1 = q (0, (q (1, b1) ^ unpack_v8b_from_v32_S (key[1]))) ^ unpack_v8b_from_v32_S (key[0]);
b2 = q (1, (q (0, b2) ^ extract_byte (key[1], 2))) ^ extract_byte (key[0], 2); b2 = q (1, (q (0, b2) ^ unpack_v8c_from_v32_S (key[1]))) ^ unpack_v8c_from_v32_S (key[0]);
b3 = q (1, (q (1, b3) ^ extract_byte (key[1], 3))) ^ extract_byte (key[0], 3); b3 = q (1, (q (1, b3) ^ unpack_v8d_from_v32_S (key[1]))) ^ unpack_v8d_from_v32_S (key[0]);
return mds (0, b0) ^ mds (1, b1) ^ mds (2, b2) ^ mds (3, b3); return mds (0, b0) ^ mds (1, b1) ^ mds (2, b2) ^ mds (3, b3);
} }
@ -460,40 +458,40 @@ DECLSPEC void twofish128_decrypt (const u32 *sk, const u32 *lk, const u32 *in, u
// 256 bit key // 256 bit key
#define g1_fun256(x) \ #define g1_fun256(x) \
(mds (0, q40 (extract_byte (x, 3), sk)) ^ \ (mds (0, q40 (unpack_v8d_from_v32_S (x), sk)) ^ \
mds (1, q41 (extract_byte (x, 0), sk)) ^ \ mds (1, q41 (unpack_v8a_from_v32_S (x), sk)) ^ \
mds (2, q42 (extract_byte (x, 1), sk)) ^ \ mds (2, q42 (unpack_v8b_from_v32_S (x), sk)) ^ \
mds (3, q43 (extract_byte (x, 2), sk))) mds (3, q43 (unpack_v8c_from_v32_S (x), sk)))
#define g0_fun256(x) \ #define g0_fun256(x) \
(mds (0, q40 (extract_byte (x, 0), sk)) ^ \ (mds (0, q40 (unpack_v8a_from_v32_S (x), sk)) ^ \
mds (1, q41 (extract_byte (x, 1), sk)) ^ \ mds (1, q41 (unpack_v8b_from_v32_S (x), sk)) ^ \
mds (2, q42 (extract_byte (x, 2), sk)) ^ \ mds (2, q42 (unpack_v8c_from_v32_S (x), sk)) ^ \
mds (3, q43 (extract_byte (x, 3), sk))) mds (3, q43 (unpack_v8d_from_v32_S (x), sk)))
DECLSPEC u32 h_fun256 (const u32 x, const u32 *key) DECLSPEC u32 h_fun256 (const u32 x, const u32 *key)
{ {
u32 b0, b1, b2, b3; u32 b0, b1, b2, b3;
b0 = extract_byte (x, 0); b0 = unpack_v8a_from_v32_S (x);
b1 = extract_byte (x, 1); b1 = unpack_v8b_from_v32_S (x);
b2 = extract_byte (x, 2); b2 = unpack_v8c_from_v32_S (x);
b3 = extract_byte (x, 3); b3 = unpack_v8d_from_v32_S (x);
b0 = q (1, b0) ^ extract_byte (key[3], 0); b0 = q (1, b0) ^ unpack_v8a_from_v32_S (key[3]);
b1 = q (0, b1) ^ extract_byte (key[3], 1); b1 = q (0, b1) ^ unpack_v8b_from_v32_S (key[3]);
b2 = q (0, b2) ^ extract_byte (key[3], 2); b2 = q (0, b2) ^ unpack_v8c_from_v32_S (key[3]);
b3 = q (1, b3) ^ extract_byte (key[3], 3); b3 = q (1, b3) ^ unpack_v8d_from_v32_S (key[3]);
b0 = q (1, b0) ^ extract_byte (key[2], 0); b0 = q (1, b0) ^ unpack_v8a_from_v32_S (key[2]);
b1 = q (1, b1) ^ extract_byte (key[2], 1); b1 = q (1, b1) ^ unpack_v8b_from_v32_S (key[2]);
b2 = q (0, b2) ^ extract_byte (key[2], 2); b2 = q (0, b2) ^ unpack_v8c_from_v32_S (key[2]);
b3 = q (0, b3) ^ extract_byte (key[2], 3); b3 = q (0, b3) ^ unpack_v8d_from_v32_S (key[2]);
b0 = q (0, (q (0, b0) ^ extract_byte (key[1], 0))) ^ extract_byte (key[0], 0); b0 = q (0, (q (0, b0) ^ unpack_v8a_from_v32_S (key[1]))) ^ unpack_v8a_from_v32_S (key[0]);
b1 = q (0, (q (1, b1) ^ extract_byte (key[1], 1))) ^ extract_byte (key[0], 1); b1 = q (0, (q (1, b1) ^ unpack_v8b_from_v32_S (key[1]))) ^ unpack_v8b_from_v32_S (key[0]);
b2 = q (1, (q (0, b2) ^ extract_byte (key[1], 2))) ^ extract_byte (key[0], 2); b2 = q (1, (q (0, b2) ^ unpack_v8c_from_v32_S (key[1]))) ^ unpack_v8c_from_v32_S (key[0]);
b3 = q (1, (q (1, b3) ^ extract_byte (key[1], 3))) ^ extract_byte (key[0], 3); b3 = q (1, (q (1, b3) ^ unpack_v8d_from_v32_S (key[1]))) ^ unpack_v8d_from_v32_S (key[0]);
return mds (0, b0) ^ mds (1, b1) ^ mds (2, b2) ^ mds (3, b3); return mds (0, b0) ^ mds (1, b1) ^ mds (2, b2) ^ mds (3, b3);
} }