2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
31 #include "include/rp_gpu.h"
35 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
36 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
39 #define GETSHIFTEDINT(a,n) amd_bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], (n))
41 #define SETSHIFTEDINT(a,n,v) \
43 const u32 s = ((n) & 3) * 8; \
44 const u64 x = (u64) (v) << s; \
45 (a)[((n)/4)+0] |= x; \
46 (a)[((n)/4)+1] = x >> 32; \
49 __constant u32 theMagicArray[64] =
51 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
52 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
53 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
54 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
55 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
56 0, 0, 0, 0, 0, 0, 0, 0,
57 0, 0, 0, 0, 0, 0, 0, 0,
58 0, 0, 0, 0, 0, 0, 0, 0
61 static void swap_buffer (u32x final[16])
63 final[ 0] = swap_workaround (final[ 0]);
64 final[ 1] = swap_workaround (final[ 1]);
65 final[ 2] = swap_workaround (final[ 2]);
66 final[ 3] = swap_workaround (final[ 3]);
67 final[ 4] = swap_workaround (final[ 4]);
68 final[ 5] = swap_workaround (final[ 5]);
69 final[ 6] = swap_workaround (final[ 6]);
70 final[ 7] = swap_workaround (final[ 7]);
71 final[ 8] = swap_workaround (final[ 8]);
72 final[ 9] = swap_workaround (final[ 9]);
73 final[10] = swap_workaround (final[10]);
74 final[11] = swap_workaround (final[11]);
75 final[12] = swap_workaround (final[12]);
76 final[13] = swap_workaround (final[13]);
77 final[14] = swap_workaround (final[14]);
78 final[15] = swap_workaround (final[15]);
81 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
109 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
110 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
111 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
112 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
113 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
114 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
115 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
116 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
117 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
118 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
119 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
120 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
121 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
122 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
123 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
124 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
125 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
126 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
127 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
128 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
133 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
134 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
135 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
136 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
137 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
138 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
139 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
140 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
141 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
142 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
143 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
144 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
145 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
146 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
147 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
148 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
149 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
150 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
151 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
152 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
157 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
158 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
159 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
160 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
161 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
162 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
163 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
164 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
165 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
166 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
167 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
168 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
169 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
170 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
171 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
172 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
173 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
174 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
175 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
176 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
181 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
182 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
183 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
184 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
185 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
186 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
187 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
188 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
189 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
190 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
191 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
192 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
193 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
194 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
195 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
196 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
197 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
198 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
199 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
200 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
209 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_m04 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
215 const u32 lid = get_local_id (0);
221 const u32 gid = get_global_id (0);
223 if (gid >= gid_max) return;
227 pw_buf0[0] = pws[gid].i[ 0];
228 pw_buf0[1] = pws[gid].i[ 1];
229 pw_buf0[2] = pws[gid].i[ 2];
230 pw_buf0[3] = pws[gid].i[ 3];
234 pw_buf1[0] = pws[gid].i[ 4];
235 pw_buf1[1] = pws[gid].i[ 5];
236 pw_buf1[2] = pws[gid].i[ 6];
237 pw_buf1[3] = pws[gid].i[ 7];
239 const u32 pw_len = pws[gid].pw_len;
247 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
248 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
249 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
250 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
251 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
252 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
253 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
254 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
256 const u32 salt_len = salt_bufs[salt_pos].salt_len;
262 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
292 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
326 switch_buffer_by_offset (s0, s1, s2, s3, out_len);
328 const u32 pw_salt_len = out_len + salt_len;
336 final[ 0] = swap_workaround (w0[0] | s0[0]);
337 final[ 1] = swap_workaround (w0[1] | s0[1]);
338 final[ 2] = swap_workaround (w0[2] | s0[2]);
339 final[ 3] = swap_workaround (w0[3] | s0[3]);
340 final[ 4] = swap_workaround (w1[0] | s1[0]);
341 final[ 5] = swap_workaround (w1[1] | s1[1]);
342 final[ 6] = swap_workaround (w1[2] | s1[2]);
343 final[ 7] = swap_workaround (w1[3] | s1[3]);
344 final[ 8] = swap_workaround (w2[0] | s2[0]);
345 final[ 9] = swap_workaround (w2[1] | s2[1]);
346 final[10] = swap_workaround (w2[2] | s2[2]);
347 final[11] = swap_workaround (w2[3] | s2[3]);
348 final[12] = swap_workaround (w3[0] | s3[0]);
349 final[13] = swap_workaround (w3[1] | s3[1]);
351 final[15] = pw_salt_len * 8;
361 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
363 // prepare magic array range
365 u32x lengthMagicArray = 0x20;
366 u32x offsetMagicArray = 0;
368 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
369 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
370 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
371 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
372 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
373 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
374 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
375 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
376 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
377 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
378 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
379 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
380 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
381 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
382 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
383 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
384 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
385 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
386 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
387 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
398 for (int i = 0; i < 64; i++) final[i] = 0;
409 u32 final_len = out_len;
415 for (i = 0; i < lengthMagicArray - 4; i += 4)
417 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
419 SETSHIFTEDINT (final, final_len + i, tmp);
422 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
424 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
426 SETSHIFTEDINT (final, final_len + i, tmp);
428 final_len += lengthMagicArray;
432 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
434 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
436 SETSHIFTEDINT (final, final_len + i, tmp);
439 final_len += salt_len;
446 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
448 swap_buffer (&final[off]);
450 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
453 swap_buffer (&final[off]);
456 final[off + 15] = final_len * 8;
458 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
460 const u32x r0 = digest[3];
461 const u32x r1 = digest[4];
462 const u32x r2 = digest[2];
463 const u32x r3 = digest[1];
465 #include VECT_COMPARE_M
469 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_m08 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
473 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_m16 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
477 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_s04 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
483 const u32 lid = get_local_id (0);
489 const u32 gid = get_global_id (0);
491 if (gid >= gid_max) return;
495 pw_buf0[0] = pws[gid].i[ 0];
496 pw_buf0[1] = pws[gid].i[ 1];
497 pw_buf0[2] = pws[gid].i[ 2];
498 pw_buf0[3] = pws[gid].i[ 3];
502 pw_buf1[0] = pws[gid].i[ 4];
503 pw_buf1[1] = pws[gid].i[ 5];
504 pw_buf1[2] = pws[gid].i[ 6];
505 pw_buf1[3] = pws[gid].i[ 7];
507 const u32 pw_len = pws[gid].pw_len;
515 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
516 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
517 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
518 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
519 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
520 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
521 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
522 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
524 const u32 salt_len = salt_bufs[salt_pos].salt_len;
530 const u32 search[4] =
532 digests_buf[digests_offset].digest_buf[DGST_R0],
533 digests_buf[digests_offset].digest_buf[DGST_R1],
534 digests_buf[digests_offset].digest_buf[DGST_R2],
535 digests_buf[digests_offset].digest_buf[DGST_R3]
542 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
572 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
606 switch_buffer_by_offset (s0, s1, s2, s3, out_len);
608 const u32 pw_salt_len = out_len + salt_len;
616 final[ 0] = swap_workaround (w0[0] | s0[0]);
617 final[ 1] = swap_workaround (w0[1] | s0[1]);
618 final[ 2] = swap_workaround (w0[2] | s0[2]);
619 final[ 3] = swap_workaround (w0[3] | s0[3]);
620 final[ 4] = swap_workaround (w1[0] | s1[0]);
621 final[ 5] = swap_workaround (w1[1] | s1[1]);
622 final[ 6] = swap_workaround (w1[2] | s1[2]);
623 final[ 7] = swap_workaround (w1[3] | s1[3]);
624 final[ 8] = swap_workaround (w2[0] | s2[0]);
625 final[ 9] = swap_workaround (w2[1] | s2[1]);
626 final[10] = swap_workaround (w2[2] | s2[2]);
627 final[11] = swap_workaround (w2[3] | s2[3]);
628 final[12] = swap_workaround (w3[0] | s3[0]);
629 final[13] = swap_workaround (w3[1] | s3[1]);
631 final[15] = pw_salt_len * 8;
641 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
643 // prepare magic array range
645 u32x lengthMagicArray = 0x20;
646 u32x offsetMagicArray = 0;
648 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
649 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
650 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
651 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
652 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
653 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
654 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
655 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
656 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
657 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
658 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
659 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
660 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
661 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
662 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
663 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
664 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
665 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
666 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
667 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
678 for (int i = 0; i < 64; i++) final[i] = 0;
689 u32 final_len = out_len;
695 for (i = 0; i < lengthMagicArray - 4; i += 4)
697 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
699 SETSHIFTEDINT (final, final_len + i, tmp);
702 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
704 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
706 SETSHIFTEDINT (final, final_len + i, tmp);
708 final_len += lengthMagicArray;
712 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
714 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
716 SETSHIFTEDINT (final, final_len + i, tmp);
719 final_len += salt_len;
726 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
728 swap_buffer (&final[off]);
730 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
733 swap_buffer (&final[off]);
736 final[off + 15] = final_len * 8;
738 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
740 const u32x r0 = digest[3];
741 const u32x r1 = digest[4];
742 const u32x r2 = digest[2];
743 const u32x r3 = digest[1];
745 #include VECT_COMPARE_S
749 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_s08 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
753 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_s16 (__global pw_t *pws, __global gpu_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 void *esalt_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)