1
0
mirror of https://github.com/hashcat/hashcat.git synced 2024-12-22 22:58:30 +00:00
hashcat/OpenCL/m29200_a0-optimized.cl
2022-05-28 15:56:29 +02:00

1087 lines
51 KiB
Common Lisp

/**
* Author......: See docs/credits.txt
* License.....: MIT
* This algorithm for password-storage for the Radmin 3 software was analyzed and made public by synacktiv:
* https://www.synacktiv.com/publications/cracking-radmin-server-3-passwords.html
*/
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include M2S(INCLUDE_PATH/inc_vendor.h)
#include M2S(INCLUDE_PATH/inc_types.h)
#include M2S(INCLUDE_PATH/inc_platform.cl)
#include M2S(INCLUDE_PATH/inc_common.cl)
#include M2S(INCLUDE_PATH/inc_rp_optimized.h)
#include M2S(INCLUDE_PATH/inc_rp_optimized.cl)
#include M2S(INCLUDE_PATH/inc_simd.cl)
#include M2S(INCLUDE_PATH/inc_hash_sha1.cl)
#include M2S(INCLUDE_PATH/inc_bignum_operations.cl)
#include M2S(INCLUDE_PATH/inc_radmin3_constants.h)
#endif
typedef struct radmin3
{
u32 user[64];
u32 user_len;
u32 pre[PRECOMP_DATALEN]; // 38400 for PRECOMP_BITS = 4
} radmin3_t;
KERNEL_FQ void m29200_m04 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
const u64 gid = get_global_id (0);
const u64 lsz = get_local_size (0);
/**
* cache constant values to shared memory
*/
LOCAL_VK u32 m[64];
LOCAL_VK u32 r[64];
LOCAL_VK u32 fact[64];
for (u32 i = lid; i < 64; i += lsz)
{
m[i] = RADMIN3_M[i];
r[i] = RADMIN3_R[i];
fact[i] = RADMIN3_FACT[i];
}
SYNC_THREADS ();
if (gid >= GID_CNT) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
pw_buf0[0] = pws[gid].i[0];
pw_buf0[1] = pws[gid].i[1];
pw_buf0[2] = pws[gid].i[2];
pw_buf0[3] = pws[gid].i[3];
pw_buf1[0] = pws[gid].i[4];
pw_buf1[1] = pws[gid].i[5];
pw_buf1[2] = pws[gid].i[6];
pw_buf1[3] = pws[gid].i[7];
const u32 pw_len = pws[gid].pw_len & 63;
/**
* salt
*/
u32 salt1_buf0[4];
u32 salt1_buf1[4];
u32 salt1_buf2[4];
u32 salt1_buf3[4];
salt1_buf0[0] = salt_bufs[SALT_POS_HOST].salt_buf[0];
salt1_buf0[1] = salt_bufs[SALT_POS_HOST].salt_buf[1];
salt1_buf0[2] = salt_bufs[SALT_POS_HOST].salt_buf[2];
salt1_buf0[3] = salt_bufs[SALT_POS_HOST].salt_buf[3];
salt1_buf1[0] = salt_bufs[SALT_POS_HOST].salt_buf[4];
salt1_buf1[1] = salt_bufs[SALT_POS_HOST].salt_buf[5];
salt1_buf1[2] = salt_bufs[SALT_POS_HOST].salt_buf[6];
salt1_buf1[3] = salt_bufs[SALT_POS_HOST].salt_buf[7];
salt1_buf2[0] = 0;
salt1_buf2[1] = 0;
salt1_buf2[2] = 0;
salt1_buf2[3] = 0;
salt1_buf3[0] = 0;
salt1_buf3[1] = 0;
salt1_buf3[2] = 0;
salt1_buf3[3] = 0;
u32 salt2_buf0[4];
u32 salt2_buf1[4];
u32 salt2_buf2[4];
u32 salt2_buf3[4];
salt2_buf0[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 0]);
salt2_buf0[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 1]);
salt2_buf0[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 2]);
salt2_buf0[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 3]);
salt2_buf1[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 4]);
salt2_buf1[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 5]);
salt2_buf1[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 6]);
salt2_buf1[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 7]);
salt2_buf2[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 8]);
salt2_buf2[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 9]);
salt2_buf2[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[10]);
salt2_buf2[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[11]);
salt2_buf3[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[12]);
salt2_buf3[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[13]);
salt2_buf3[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[14]);
salt2_buf3[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[15]);
const u32 salt2_len = esalt_bufs[DIGESTS_OFFSET_HOST].user_len;
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos += VECT_SIZE)
{
u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
make_utf16le (w1, w2, w3);
make_utf16le (w0, w0, w1);
const u32x out_len2 = out_len * 2;
/**
* prepend salt
*/
const u32x out_salt_len = out_len2 + salt2_len;
switch_buffer_by_offset_le_VV (w0, w1, w2, w3, salt2_len);
w0[0] |= salt2_buf0[0];
w0[1] |= salt2_buf0[1];
w0[2] |= salt2_buf0[2];
w0[3] |= salt2_buf0[3];
w1[0] |= salt2_buf1[0];
w1[1] |= salt2_buf1[1];
w1[2] |= salt2_buf1[2];
w1[3] |= salt2_buf1[3];
w2[0] |= salt2_buf2[0];
w2[1] |= salt2_buf2[1];
w2[2] |= salt2_buf2[2];
w2[3] |= salt2_buf2[3];
w3[0] |= salt2_buf3[0];
w3[1] |= salt2_buf3[1];
w3[2] |= salt2_buf3[2];
w3[3] |= salt2_buf3[3];
append_0x80_4x4_VV (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
u32x w0_t = hc_swap32 (w0[0]);
u32x w1_t = hc_swap32 (w0[1]);
u32x w2_t = hc_swap32 (w0[2]);
u32x w3_t = hc_swap32 (w0[3]);
u32x w4_t = hc_swap32 (w1[0]);
u32x w5_t = hc_swap32 (w1[1]);
u32x w6_t = hc_swap32 (w1[2]);
u32x w7_t = hc_swap32 (w1[3]);
u32x w8_t = hc_swap32 (w2[0]);
u32x w9_t = hc_swap32 (w2[1]);
u32x wa_t = hc_swap32 (w2[2]);
u32x wb_t = hc_swap32 (w2[3]);
u32x wc_t = hc_swap32 (w3[0]);
u32x wd_t = hc_swap32 (w3[1]);
u32x we_t = 0;
u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
u32x d = SHA1M_D;
u32x e = SHA1M_E;
#undef K
#define K SHA1C00
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
a += make_u32x (SHA1M_A);
b += make_u32x (SHA1M_B);
c += make_u32x (SHA1M_C);
d += make_u32x (SHA1M_D);
e += make_u32x (SHA1M_E);
w0_t = salt1_buf0[0];
w1_t = salt1_buf0[1];
w2_t = salt1_buf0[2];
w3_t = salt1_buf0[3];
w4_t = salt1_buf1[0];
w5_t = salt1_buf1[1];
w6_t = salt1_buf1[2];
w7_t = salt1_buf1[3];
w8_t = a;
w9_t = b;
wa_t = c;
wb_t = d;
wc_t = e;
wd_t = 0x80000000;
we_t = 0;
wf_t = (32 + 20) * 8;
a = SHA1M_A;
b = SHA1M_B;
c = SHA1M_C;
d = SHA1M_D;
e = SHA1M_E;
#undef K
#define K SHA1C00
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
a += make_u32x (SHA1M_A);
b += make_u32x (SHA1M_B);
c += make_u32x (SHA1M_C);
d += make_u32x (SHA1M_D);
e += make_u32x (SHA1M_E);
u32 exponent[5] = { 0 };
exponent[0] = e;
exponent[1] = d;
exponent[2] = c;
exponent[3] = b;
exponent[4] = a;
u32 r_t[64] =
{
r[ 0], r[ 1], r[ 2], r[ 3], r[ 4], r[ 5], r[ 6], r[ 7],
r[ 8], r[ 9], r[10], r[11], r[12], r[13], r[14], r[15],
r[16], r[17], r[18], r[19], r[20], r[21], r[22], r[23],
r[24], r[25], r[26], r[27], r[28], r[29], r[30], r[31],
r[32], r[33], r[34], r[35], r[36], r[37], r[38], r[39],
r[40], r[41], r[42], r[43], r[44], r[45], r[46], r[47],
r[48], r[49], r[50], r[51], r[52], r[53], r[54], r[55],
r[56], r[57], r[58], r[59], r[60], r[61], r[62], r[63],
};
for (u32 i = 0, j = 0; i < PRECOMP_SLOTS; i += 1, j += PRECOMP_ENTRIES - 1)
{
const u32 div = (PRECOMP_BITS * i) / 32; // for 4 bits: (i / 8)
const u32 shift = (PRECOMP_BITS * i) % 32; // for 4 bits: (i % 8) * 4
// const
u32 cur_sel = (exponent[div] >> shift) & PRECOMP_MASK; // 0x0f == 0b1111 (4 bits)
// working with non-divisible u32 (see PRECOMP_BITS):
if (32 - shift < PRECOMP_BITS)
{
cur_sel |= (exponent[div + 1] << (32 - shift)) & PRECOMP_MASK;
}
if (cur_sel == 0) continue;
const u32 pre_idx = (j + cur_sel - 1) * PRECOMP_ENTRYLEN; // x * 64 is same as x << 6
// u32 p[64]; for (u32 i = 0; i < 64; i++) p[i] = esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + i];
const u32 p[64] =
{
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 0],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 1],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 2],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 3],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 4],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 5],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 6],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 7],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 8],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 9],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 10],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 11],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 12],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 13],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 14],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 15],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 16],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 17],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 18],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 19],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 20],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 21],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 22],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 23],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 24],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 25],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 26],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 27],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 28],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 29],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 30],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 31],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 32],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 33],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 34],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 35],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 36],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 37],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 38],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 39],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 40],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 41],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 42],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 43],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 44],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 45],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 46],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 47],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 48],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 49],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 50],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 51],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 52],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 53],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 54],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 55],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 56],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 57],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 58],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 59],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 60],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 61],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 62],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 63],
};
mul_mod (r_t, p, m, fact); // r = (r * pre[n]) % m
}
const u32 r0 = r_t[0];
const u32 r1 = r_t[1];
const u32 r2 = r_t[2];
const u32 r3 = r_t[3];
COMPARE_M_SIMD (r0, r1, r2, r3);
}
}
KERNEL_FQ void m29200_m08 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
}
KERNEL_FQ void m29200_m16 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
}
KERNEL_FQ void m29200_s04 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
const u64 gid = get_global_id (0);
const u64 lsz = get_local_size (0);
/**
* cache constant values to shared memory
*/
LOCAL_VK u32 m[64];
LOCAL_VK u32 r[64];
LOCAL_VK u32 fact[64];
for (u32 i = lid; i < 64; i += lsz)
{
m[i] = RADMIN3_M[i];
r[i] = RADMIN3_R[i];
fact[i] = RADMIN3_FACT[i];
}
/**
* digest
*/
const u32 search[4] =
{
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R0],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R1],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R2],
digests_buf[DIGESTS_OFFSET_HOST].digest_buf[DGST_R3]
};
/**
* base
*/
SYNC_THREADS ();
if (gid >= GID_CNT) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
pw_buf0[0] = pws[gid].i[0];
pw_buf0[1] = pws[gid].i[1];
pw_buf0[2] = pws[gid].i[2];
pw_buf0[3] = pws[gid].i[3];
pw_buf1[0] = pws[gid].i[4];
pw_buf1[1] = pws[gid].i[5];
pw_buf1[2] = pws[gid].i[6];
pw_buf1[3] = pws[gid].i[7];
const u32 pw_len = pws[gid].pw_len & 63;
/**
* salt
*/
u32 salt1_buf0[4];
u32 salt1_buf1[4];
u32 salt1_buf2[4];
u32 salt1_buf3[4];
salt1_buf0[0] = salt_bufs[SALT_POS_HOST].salt_buf[0];
salt1_buf0[1] = salt_bufs[SALT_POS_HOST].salt_buf[1];
salt1_buf0[2] = salt_bufs[SALT_POS_HOST].salt_buf[2];
salt1_buf0[3] = salt_bufs[SALT_POS_HOST].salt_buf[3];
salt1_buf1[0] = salt_bufs[SALT_POS_HOST].salt_buf[4];
salt1_buf1[1] = salt_bufs[SALT_POS_HOST].salt_buf[5];
salt1_buf1[2] = salt_bufs[SALT_POS_HOST].salt_buf[6];
salt1_buf1[3] = salt_bufs[SALT_POS_HOST].salt_buf[7];
salt1_buf2[0] = 0;
salt1_buf2[1] = 0;
salt1_buf2[2] = 0;
salt1_buf2[3] = 0;
salt1_buf3[0] = 0;
salt1_buf3[1] = 0;
salt1_buf3[2] = 0;
salt1_buf3[3] = 0;
u32 salt2_buf0[4];
u32 salt2_buf1[4];
u32 salt2_buf2[4];
u32 salt2_buf3[4];
salt2_buf0[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 0]);
salt2_buf0[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 1]);
salt2_buf0[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 2]);
salt2_buf0[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 3]);
salt2_buf1[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 4]);
salt2_buf1[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 5]);
salt2_buf1[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 6]);
salt2_buf1[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 7]);
salt2_buf2[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 8]);
salt2_buf2[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[ 9]);
salt2_buf2[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[10]);
salt2_buf2[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[11]);
salt2_buf3[0] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[12]);
salt2_buf3[1] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[13]);
salt2_buf3[2] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[14]);
salt2_buf3[3] = hc_swap32_S (esalt_bufs[DIGESTS_OFFSET_HOST].user[15]);
const u32 salt2_len = esalt_bufs[DIGESTS_OFFSET_HOST].user_len;
/**
* loop
*/
for (u32 il_pos = 0; il_pos < IL_CNT; il_pos += VECT_SIZE)
{
u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
make_utf16le (w1, w2, w3);
make_utf16le (w0, w0, w1);
const u32x out_len2 = out_len * 2;
/**
* prepend salt
*/
const u32x out_salt_len = out_len2 + salt2_len;
switch_buffer_by_offset_le_VV (w0, w1, w2, w3, salt2_len);
w0[0] |= salt2_buf0[0];
w0[1] |= salt2_buf0[1];
w0[2] |= salt2_buf0[2];
w0[3] |= salt2_buf0[3];
w1[0] |= salt2_buf1[0];
w1[1] |= salt2_buf1[1];
w1[2] |= salt2_buf1[2];
w1[3] |= salt2_buf1[3];
w2[0] |= salt2_buf2[0];
w2[1] |= salt2_buf2[1];
w2[2] |= salt2_buf2[2];
w2[3] |= salt2_buf2[3];
w3[0] |= salt2_buf3[0];
w3[1] |= salt2_buf3[1];
w3[2] |= salt2_buf3[2];
w3[3] |= salt2_buf3[3];
append_0x80_4x4_VV (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
u32x w0_t = hc_swap32 (w0[0]);
u32x w1_t = hc_swap32 (w0[1]);
u32x w2_t = hc_swap32 (w0[2]);
u32x w3_t = hc_swap32 (w0[3]);
u32x w4_t = hc_swap32 (w1[0]);
u32x w5_t = hc_swap32 (w1[1]);
u32x w6_t = hc_swap32 (w1[2]);
u32x w7_t = hc_swap32 (w1[3]);
u32x w8_t = hc_swap32 (w2[0]);
u32x w9_t = hc_swap32 (w2[1]);
u32x wa_t = hc_swap32 (w2[2]);
u32x wb_t = hc_swap32 (w2[3]);
u32x wc_t = hc_swap32 (w3[0]);
u32x wd_t = hc_swap32 (w3[1]);
u32x we_t = 0;
u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
u32x d = SHA1M_D;
u32x e = SHA1M_E;
#undef K
#define K SHA1C00
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
a += make_u32x (SHA1M_A);
b += make_u32x (SHA1M_B);
c += make_u32x (SHA1M_C);
d += make_u32x (SHA1M_D);
e += make_u32x (SHA1M_E);
w0_t = salt1_buf0[0];
w1_t = salt1_buf0[1];
w2_t = salt1_buf0[2];
w3_t = salt1_buf0[3];
w4_t = salt1_buf1[0];
w5_t = salt1_buf1[1];
w6_t = salt1_buf1[2];
w7_t = salt1_buf1[3];
w8_t = a;
w9_t = b;
wa_t = c;
wb_t = d;
wc_t = e;
wd_t = 0x80000000;
we_t = 0;
wf_t = (32 + 20) * 8;
a = SHA1M_A;
b = SHA1M_B;
c = SHA1M_C;
d = SHA1M_D;
e = SHA1M_E;
#undef K
#define K SHA1C00
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
a += make_u32x (SHA1M_A);
b += make_u32x (SHA1M_B);
c += make_u32x (SHA1M_C);
d += make_u32x (SHA1M_D);
e += make_u32x (SHA1M_E);
u32 exponent[5] = { 0 };
exponent[0] = e;
exponent[1] = d;
exponent[2] = c;
exponent[3] = b;
exponent[4] = a;
u32 r_t[64] =
{
r[ 0], r[ 1], r[ 2], r[ 3], r[ 4], r[ 5], r[ 6], r[ 7],
r[ 8], r[ 9], r[10], r[11], r[12], r[13], r[14], r[15],
r[16], r[17], r[18], r[19], r[20], r[21], r[22], r[23],
r[24], r[25], r[26], r[27], r[28], r[29], r[30], r[31],
r[32], r[33], r[34], r[35], r[36], r[37], r[38], r[39],
r[40], r[41], r[42], r[43], r[44], r[45], r[46], r[47],
r[48], r[49], r[50], r[51], r[52], r[53], r[54], r[55],
r[56], r[57], r[58], r[59], r[60], r[61], r[62], r[63],
};
for (u32 i = 0, j = 0; i < PRECOMP_SLOTS; i += 1, j += PRECOMP_ENTRIES - 1)
{
const u32 div = (PRECOMP_BITS * i) / 32; // for 4 bits: (i / 8)
const u32 shift = (PRECOMP_BITS * i) % 32; // for 4 bits: (i % 8) * 4
// const
u32 cur_sel = (exponent[div] >> shift) & PRECOMP_MASK; // 0x0f == 0b1111 (4 bits)
// working with non-divisible u32 (see PRECOMP_BITS):
if (32 - shift < PRECOMP_BITS)
{
cur_sel |= (exponent[div + 1] << (32 - shift)) & PRECOMP_MASK;
}
if (cur_sel == 0) continue;
const u32 pre_idx = (j + cur_sel - 1) * PRECOMP_ENTRYLEN; // x * 64 is same as x << 6
// u32 p[64]; for (u32 i = 0; i < 64; i++) p[i] = esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + i];
const u32 p[64] =
{
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 0],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 1],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 2],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 3],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 4],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 5],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 6],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 7],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 8],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 9],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 10],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 11],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 12],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 13],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 14],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 15],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 16],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 17],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 18],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 19],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 20],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 21],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 22],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 23],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 24],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 25],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 26],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 27],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 28],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 29],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 30],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 31],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 32],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 33],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 34],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 35],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 36],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 37],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 38],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 39],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 40],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 41],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 42],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 43],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 44],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 45],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 46],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 47],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 48],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 49],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 50],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 51],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 52],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 53],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 54],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 55],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 56],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 57],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 58],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 59],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 60],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 61],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 62],
esalt_bufs[DIGESTS_OFFSET_HOST].pre[pre_idx + 63],
};
mul_mod (r_t, p, m, fact); // r = (r * pre[n]) % m
}
const u32 r0 = r_t[0];
const u32 r1 = r_t[1];
const u32 r2 = r_t[2];
const u32 r3 = r_t[3];
COMPARE_S_SIMD (r0, r1, r2, r3);
}
}
KERNEL_FQ void m29200_s08 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
}
KERNEL_FQ void m29200_s16 (KERN_ATTR_RULES_ESALT (radmin3_t))
{
}