2 * Authors.....: Jens Steube <jens.steube@gmail.com>
3 * Gabriele Gristina <matrix@hashcat.net>
12 #include "include/constants.h"
13 #include "include/kernel_vendor.h"
20 #include "include/kernel_functions.c"
21 #include "OpenCL/types_ocl.c"
22 #include "OpenCL/common.c"
23 #include "OpenCL/simd.c"
25 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
53 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
54 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
55 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
56 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
57 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
58 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
59 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
60 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
61 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
62 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
63 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
64 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
65 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
66 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
67 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
68 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
69 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
70 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
71 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
72 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
77 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
78 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
79 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
80 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
81 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
82 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
83 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
84 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
85 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
86 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
87 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
88 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
89 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
90 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
91 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
92 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
93 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
94 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
95 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
96 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
101 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
102 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
103 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
104 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
105 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
106 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
107 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
108 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
109 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
110 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
111 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
112 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
113 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
114 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
115 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
116 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
117 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
118 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
119 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
120 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
125 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
126 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
127 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
128 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
129 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
130 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
131 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
132 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
133 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
134 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
135 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
136 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
137 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
138 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
139 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
140 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
141 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
142 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
143 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
144 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
153 static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
155 w0[0] = w0[0] ^ 0x36363636;
156 w0[1] = w0[1] ^ 0x36363636;
157 w0[2] = w0[2] ^ 0x36363636;
158 w0[3] = w0[3] ^ 0x36363636;
159 w1[0] = w1[0] ^ 0x36363636;
160 w1[1] = w1[1] ^ 0x36363636;
161 w1[2] = w1[2] ^ 0x36363636;
162 w1[3] = w1[3] ^ 0x36363636;
163 w2[0] = w2[0] ^ 0x36363636;
164 w2[1] = w2[1] ^ 0x36363636;
165 w2[2] = w2[2] ^ 0x36363636;
166 w2[3] = w2[3] ^ 0x36363636;
167 w3[0] = w3[0] ^ 0x36363636;
168 w3[1] = w3[1] ^ 0x36363636;
169 w3[2] = w3[2] ^ 0x36363636;
170 w3[3] = w3[3] ^ 0x36363636;
178 sha1_transform (w0, w1, w2, w3, ipad);
180 w0[0] = w0[0] ^ 0x6a6a6a6a;
181 w0[1] = w0[1] ^ 0x6a6a6a6a;
182 w0[2] = w0[2] ^ 0x6a6a6a6a;
183 w0[3] = w0[3] ^ 0x6a6a6a6a;
184 w1[0] = w1[0] ^ 0x6a6a6a6a;
185 w1[1] = w1[1] ^ 0x6a6a6a6a;
186 w1[2] = w1[2] ^ 0x6a6a6a6a;
187 w1[3] = w1[3] ^ 0x6a6a6a6a;
188 w2[0] = w2[0] ^ 0x6a6a6a6a;
189 w2[1] = w2[1] ^ 0x6a6a6a6a;
190 w2[2] = w2[2] ^ 0x6a6a6a6a;
191 w2[3] = w2[3] ^ 0x6a6a6a6a;
192 w3[0] = w3[0] ^ 0x6a6a6a6a;
193 w3[1] = w3[1] ^ 0x6a6a6a6a;
194 w3[2] = w3[2] ^ 0x6a6a6a6a;
195 w3[3] = w3[3] ^ 0x6a6a6a6a;
203 sha1_transform (w0, w1, w2, w3, opad);
206 static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
214 sha1_transform (w0, w1, w2, w3, digest);
231 w3[3] = (64 + 20) * 8;
239 sha1_transform (w0, w1, w2, w3, digest);
242 static void m05400m (__local u32 *w_s, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, __local u32 *s_msg_buf)
248 const u32 gid = get_global_id (0);
249 const u32 lid = get_local_id (0);
255 const u32 nr_len = ikepsk_bufs[salt_pos].nr_len;
256 const u32 msg_len = ikepsk_bufs[salt_pos].msg_len;
264 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
266 const u32x w0r = w0r_create_bft (bfs_buf, il_pos);
268 const u32x w0lr = w0l | w0r;
305 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
322 w3_t[3] = (64 + nr_len) * 8;
326 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
345 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
350 for (left = ikepsk_bufs[salt_pos].msg_len, off = 0; left >= 56; left -= 64, off += 16)
352 w0_t[0] = s_msg_buf[off + 0];
353 w0_t[1] = s_msg_buf[off + 1];
354 w0_t[2] = s_msg_buf[off + 2];
355 w0_t[3] = s_msg_buf[off + 3];
356 w1_t[0] = s_msg_buf[off + 4];
357 w1_t[1] = s_msg_buf[off + 5];
358 w1_t[2] = s_msg_buf[off + 6];
359 w1_t[3] = s_msg_buf[off + 7];
360 w2_t[0] = s_msg_buf[off + 8];
361 w2_t[1] = s_msg_buf[off + 9];
362 w2_t[2] = s_msg_buf[off + 10];
363 w2_t[3] = s_msg_buf[off + 11];
364 w3_t[0] = s_msg_buf[off + 12];
365 w3_t[1] = s_msg_buf[off + 13];
366 w3_t[2] = s_msg_buf[off + 14];
367 w3_t[3] = s_msg_buf[off + 15];
369 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
372 w0_t[0] = s_msg_buf[off + 0];
373 w0_t[1] = s_msg_buf[off + 1];
374 w0_t[2] = s_msg_buf[off + 2];
375 w0_t[3] = s_msg_buf[off + 3];
376 w1_t[0] = s_msg_buf[off + 4];
377 w1_t[1] = s_msg_buf[off + 5];
378 w1_t[2] = s_msg_buf[off + 6];
379 w1_t[3] = s_msg_buf[off + 7];
380 w2_t[0] = s_msg_buf[off + 8];
381 w2_t[1] = s_msg_buf[off + 9];
382 w2_t[2] = s_msg_buf[off + 10];
383 w2_t[3] = s_msg_buf[off + 11];
384 w3_t[0] = s_msg_buf[off + 12];
385 w3_t[1] = s_msg_buf[off + 13];
387 w3_t[3] = (64 + msg_len) * 8;
389 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
391 COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
395 static void m05400s (__local u32 *w_s, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, __local u32 *s_msg_buf)
401 const u32 gid = get_global_id (0);
402 const u32 lid = get_local_id (0);
408 const u32 nr_len = ikepsk_bufs[salt_pos].nr_len;
409 const u32 msg_len = ikepsk_bufs[salt_pos].msg_len;
415 const u32 search[4] =
417 digests_buf[digests_offset].digest_buf[DGST_R0],
418 digests_buf[digests_offset].digest_buf[DGST_R1],
419 digests_buf[digests_offset].digest_buf[DGST_R2],
420 digests_buf[digests_offset].digest_buf[DGST_R3]
429 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
431 const u32x w0r = w0r_create_bft (bfs_buf, il_pos);
433 const u32x w0lr = w0l | w0r;
470 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
487 w3_t[3] = (64 + nr_len) * 8;
491 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
510 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
515 for (left = ikepsk_bufs[salt_pos].msg_len, off = 0; left >= 56; left -= 64, off += 16)
517 w0_t[0] = s_msg_buf[off + 0];
518 w0_t[1] = s_msg_buf[off + 1];
519 w0_t[2] = s_msg_buf[off + 2];
520 w0_t[3] = s_msg_buf[off + 3];
521 w1_t[0] = s_msg_buf[off + 4];
522 w1_t[1] = s_msg_buf[off + 5];
523 w1_t[2] = s_msg_buf[off + 6];
524 w1_t[3] = s_msg_buf[off + 7];
525 w2_t[0] = s_msg_buf[off + 8];
526 w2_t[1] = s_msg_buf[off + 9];
527 w2_t[2] = s_msg_buf[off + 10];
528 w2_t[3] = s_msg_buf[off + 11];
529 w3_t[0] = s_msg_buf[off + 12];
530 w3_t[1] = s_msg_buf[off + 13];
531 w3_t[2] = s_msg_buf[off + 14];
532 w3_t[3] = s_msg_buf[off + 15];
534 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
537 w0_t[0] = s_msg_buf[off + 0];
538 w0_t[1] = s_msg_buf[off + 1];
539 w0_t[2] = s_msg_buf[off + 2];
540 w0_t[3] = s_msg_buf[off + 3];
541 w1_t[0] = s_msg_buf[off + 4];
542 w1_t[1] = s_msg_buf[off + 5];
543 w1_t[2] = s_msg_buf[off + 6];
544 w1_t[3] = s_msg_buf[off + 7];
545 w2_t[0] = s_msg_buf[off + 8];
546 w2_t[1] = s_msg_buf[off + 9];
547 w2_t[2] = s_msg_buf[off + 10];
548 w2_t[3] = s_msg_buf[off + 11];
549 w3_t[0] = s_msg_buf[off + 12];
550 w3_t[1] = s_msg_buf[off + 13];
552 w3_t[3] = (64 + msg_len) * 8;
554 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
556 COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
560 __kernel void m05400_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
566 const u32 gid = get_global_id (0);
567 const u32 lid = get_local_id (0);
568 const u32 lsz = get_local_size (0);
576 for (u32 i = lid; i < 16; i += lsz)
578 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
581 barrier (CLK_LOCAL_MEM_FENCE);
583 __local u32 s_msg_buf[128];
585 for (u32 i = lid; i < 128; i += lsz)
587 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
590 barrier (CLK_LOCAL_MEM_FENCE);
592 if (gid >= gid_max) return;
600 w0[0] = pws[gid].i[ 0];
601 w0[1] = pws[gid].i[ 1];
602 w0[2] = pws[gid].i[ 2];
603 w0[3] = pws[gid].i[ 3];
626 const u32 pw_len = pws[gid].pw_len;
632 m05400m (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);
635 __kernel void m05400_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
641 const u32 gid = get_global_id (0);
642 const u32 lid = get_local_id (0);
643 const u32 lsz = get_local_size (0);
651 for (u32 i = lid; i < 16; i += lsz)
653 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
656 barrier (CLK_LOCAL_MEM_FENCE);
658 __local u32 s_msg_buf[128];
660 for (u32 i = lid; i < 128; i += lsz)
662 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
665 barrier (CLK_LOCAL_MEM_FENCE);
667 if (gid >= gid_max) return;
675 w0[0] = pws[gid].i[ 0];
676 w0[1] = pws[gid].i[ 1];
677 w0[2] = pws[gid].i[ 2];
678 w0[3] = pws[gid].i[ 3];
682 w1[0] = pws[gid].i[ 4];
683 w1[1] = pws[gid].i[ 5];
684 w1[2] = pws[gid].i[ 6];
685 w1[3] = pws[gid].i[ 7];
701 const u32 pw_len = pws[gid].pw_len;
707 m05400m (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);
710 __kernel void m05400_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
716 const u32 gid = get_global_id (0);
717 const u32 lid = get_local_id (0);
718 const u32 lsz = get_local_size (0);
726 for (u32 i = lid; i < 16; i += lsz)
728 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
731 barrier (CLK_LOCAL_MEM_FENCE);
733 __local u32 s_msg_buf[128];
735 for (u32 i = lid; i < 128; i += lsz)
737 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
740 barrier (CLK_LOCAL_MEM_FENCE);
742 if (gid >= gid_max) return;
750 w0[0] = pws[gid].i[ 0];
751 w0[1] = pws[gid].i[ 1];
752 w0[2] = pws[gid].i[ 2];
753 w0[3] = pws[gid].i[ 3];
757 w1[0] = pws[gid].i[ 4];
758 w1[1] = pws[gid].i[ 5];
759 w1[2] = pws[gid].i[ 6];
760 w1[3] = pws[gid].i[ 7];
764 w2[0] = pws[gid].i[ 8];
765 w2[1] = pws[gid].i[ 9];
766 w2[2] = pws[gid].i[10];
767 w2[3] = pws[gid].i[11];
771 w3[0] = pws[gid].i[12];
772 w3[1] = pws[gid].i[13];
776 const u32 pw_len = pws[gid].pw_len;
782 m05400m (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);
785 __kernel void m05400_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
791 const u32 gid = get_global_id (0);
792 const u32 lid = get_local_id (0);
793 const u32 lsz = get_local_size (0);
801 for (u32 i = lid; i < 16; i += lsz)
803 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
806 barrier (CLK_LOCAL_MEM_FENCE);
808 __local u32 s_msg_buf[128];
810 for (u32 i = lid; i < 128; i += lsz)
812 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
815 barrier (CLK_LOCAL_MEM_FENCE);
817 if (gid >= gid_max) return;
825 w0[0] = pws[gid].i[ 0];
826 w0[1] = pws[gid].i[ 1];
827 w0[2] = pws[gid].i[ 2];
828 w0[3] = pws[gid].i[ 3];
851 const u32 pw_len = pws[gid].pw_len;
857 m05400s (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);
860 __kernel void m05400_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
866 const u32 gid = get_global_id (0);
867 const u32 lid = get_local_id (0);
868 const u32 lsz = get_local_size (0);
876 for (u32 i = lid; i < 16; i += lsz)
878 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
881 barrier (CLK_LOCAL_MEM_FENCE);
883 __local u32 s_msg_buf[128];
885 for (u32 i = lid; i < 128; i += lsz)
887 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
890 barrier (CLK_LOCAL_MEM_FENCE);
892 if (gid >= gid_max) return;
900 w0[0] = pws[gid].i[ 0];
901 w0[1] = pws[gid].i[ 1];
902 w0[2] = pws[gid].i[ 2];
903 w0[3] = pws[gid].i[ 3];
907 w1[0] = pws[gid].i[ 4];
908 w1[1] = pws[gid].i[ 5];
909 w1[2] = pws[gid].i[ 6];
910 w1[3] = pws[gid].i[ 7];
926 const u32 pw_len = pws[gid].pw_len;
932 m05400s (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);
935 __kernel void m05400_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global ikepsk_t *ikepsk_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
941 const u32 gid = get_global_id (0);
942 const u32 lid = get_local_id (0);
943 const u32 lsz = get_local_size (0);
951 for (u32 i = lid; i < 16; i += lsz)
953 w_s[i] = swap32_S (ikepsk_bufs[salt_pos].nr_buf[i]);
956 barrier (CLK_LOCAL_MEM_FENCE);
958 __local u32 s_msg_buf[128];
960 for (u32 i = lid; i < 128; i += lsz)
962 s_msg_buf[i] = swap32_S (ikepsk_bufs[salt_pos].msg_buf[i]);
965 barrier (CLK_LOCAL_MEM_FENCE);
967 if (gid >= gid_max) return;
975 w0[0] = pws[gid].i[ 0];
976 w0[1] = pws[gid].i[ 1];
977 w0[2] = pws[gid].i[ 2];
978 w0[3] = pws[gid].i[ 3];
982 w1[0] = pws[gid].i[ 4];
983 w1[1] = pws[gid].i[ 5];
984 w1[2] = pws[gid].i[ 6];
985 w1[3] = pws[gid].i[ 7];
989 w2[0] = pws[gid].i[ 8];
990 w2[1] = pws[gid].i[ 9];
991 w2[2] = pws[gid].i[10];
992 w2[3] = pws[gid].i[11];
996 w3[0] = pws[gid].i[12];
997 w3[1] = pws[gid].i[13];
1001 const u32 pw_len = pws[gid].pw_len;
1007 m05400s (w_s, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, ikepsk_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset, s_msg_buf);