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"
33 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
37 #define GETSHIFTEDINT(a,n) amd_bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], (n))
39 #define SETSHIFTEDINT(a,n,v) \
41 const u32 s = ((n) & 3) * 8; \
42 const u64 x = (u64) (v) << s; \
43 (a)[((n)/4)+0] |= x; \
44 (a)[((n)/4)+1] = x >> 32; \
47 __constant u32 theMagicArray[64] =
49 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
50 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
51 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
52 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
53 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
54 0, 0, 0, 0, 0, 0, 0, 0,
55 0, 0, 0, 0, 0, 0, 0, 0,
56 0, 0, 0, 0, 0, 0, 0, 0
59 static void swap_buffer (u32x final[16])
61 final[ 0] = swap_workaround (final[ 0]);
62 final[ 1] = swap_workaround (final[ 1]);
63 final[ 2] = swap_workaround (final[ 2]);
64 final[ 3] = swap_workaround (final[ 3]);
65 final[ 4] = swap_workaround (final[ 4]);
66 final[ 5] = swap_workaround (final[ 5]);
67 final[ 6] = swap_workaround (final[ 6]);
68 final[ 7] = swap_workaround (final[ 7]);
69 final[ 8] = swap_workaround (final[ 8]);
70 final[ 9] = swap_workaround (final[ 9]);
71 final[10] = swap_workaround (final[10]);
72 final[11] = swap_workaround (final[11]);
73 final[12] = swap_workaround (final[12]);
74 final[13] = swap_workaround (final[13]);
75 final[14] = swap_workaround (final[14]);
76 final[15] = swap_workaround (final[15]);
79 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
107 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
108 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
109 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
110 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
111 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
112 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
113 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
114 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
115 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
116 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
117 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
118 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
119 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
120 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
121 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
122 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
123 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
124 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
125 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
126 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
131 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
132 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
133 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
134 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
135 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
136 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
137 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
138 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
139 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
140 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
141 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
142 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
143 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
144 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
145 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
146 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
147 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
148 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
149 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
150 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
155 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
156 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
157 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
158 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
159 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
160 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
161 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
162 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
163 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
164 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
165 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
166 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
167 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
168 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
169 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
170 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
171 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
172 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
173 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
174 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
179 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
180 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
181 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
182 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
183 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
184 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
185 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
186 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
187 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
188 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
189 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
190 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
191 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
192 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
193 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
194 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
195 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
196 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
197 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
198 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
207 static void m07800m (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
213 const u32 gid = get_global_id (0);
214 const u32 lid = get_local_id (0);
216 w0[0] = swap_workaround (w0[0]);
217 w0[1] = swap_workaround (w0[1]);
218 w0[2] = swap_workaround (w0[2]);
219 w0[3] = swap_workaround (w0[3]);
221 w1[0] = swap_workaround (w1[0]);
222 w1[1] = swap_workaround (w1[1]);
223 w1[2] = swap_workaround (w1[2]);
224 w1[3] = swap_workaround (w1[3]);
232 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
233 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
234 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
235 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
236 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
237 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
238 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
239 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
241 const u32 salt_len = salt_bufs[salt_pos].salt_len;
271 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
273 const u32 pw_salt_len = pw_len + salt_len;
281 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
283 const u32 w0r = swap_workaround (bfs_buf[il_pos].i);
293 final[ 0] = swap_workaround (w0[0] | s0[0]);
294 final[ 1] = swap_workaround (w0[1] | s0[1]);
295 final[ 2] = swap_workaround (w0[2] | s0[2]);
296 final[ 3] = swap_workaround (w0[3] | s0[3]);
297 final[ 4] = swap_workaround (w1[0] | s1[0]);
298 final[ 5] = swap_workaround (w1[1] | s1[1]);
299 final[ 6] = swap_workaround (w1[2] | s1[2]);
300 final[ 7] = swap_workaround (w1[3] | s1[3]);
301 final[ 8] = swap_workaround (w2[0] | s2[0]);
302 final[ 9] = swap_workaround (w2[1] | s2[1]);
303 final[10] = swap_workaround (w2[2] | s2[2]);
304 final[11] = swap_workaround (w2[3] | s2[3]);
305 final[12] = swap_workaround (w3[0] | s3[0]);
306 final[13] = swap_workaround (w3[1] | s3[1]);
308 final[15] = pw_salt_len * 8;
318 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
320 // prepare magic array range
322 u32x lengthMagicArray = 0x20;
323 u32x offsetMagicArray = 0;
325 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
326 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
327 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
328 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
329 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
330 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
331 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
332 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
333 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
334 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
335 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
336 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
337 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
338 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
339 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
340 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
341 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
342 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
343 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
344 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
355 for (int i = 0; i < 64; i++) final[i] = 0;
366 u32 final_len = pw_len;
372 for (i = 0; i < lengthMagicArray - 4; i += 4)
374 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
376 SETSHIFTEDINT (final, final_len + i, tmp);
379 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
381 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
383 SETSHIFTEDINT (final, final_len + i, tmp);
385 final_len += lengthMagicArray;
389 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
391 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
393 SETSHIFTEDINT (final, final_len + i, tmp);
396 final_len += salt_len;
403 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
405 swap_buffer (&final[off]);
407 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
410 swap_buffer (&final[off]);
413 final[off + 15] = final_len * 8;
415 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
417 const u32x r0 = digest[3];
418 const u32x r1 = digest[4];
419 const u32x r2 = digest[2];
420 const u32x r3 = digest[1];
422 #include VECT_COMPARE_M
426 static void m07800s (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
432 const u32 gid = get_global_id (0);
433 const u32 lid = get_local_id (0);
435 w0[0] = swap_workaround (w0[0]);
436 w0[1] = swap_workaround (w0[1]);
437 w0[2] = swap_workaround (w0[2]);
438 w0[3] = swap_workaround (w0[3]);
440 w1[0] = swap_workaround (w1[0]);
441 w1[1] = swap_workaround (w1[1]);
442 w1[2] = swap_workaround (w1[2]);
443 w1[3] = swap_workaround (w1[3]);
451 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
452 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
453 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
454 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
455 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
456 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
457 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
458 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
460 const u32 salt_len = salt_bufs[salt_pos].salt_len;
490 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
492 const u32 pw_salt_len = pw_len + salt_len;
498 const u32 search[4] =
500 digests_buf[digests_offset].digest_buf[DGST_R0],
501 digests_buf[digests_offset].digest_buf[DGST_R1],
502 digests_buf[digests_offset].digest_buf[DGST_R2],
503 digests_buf[digests_offset].digest_buf[DGST_R3]
512 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
514 const u32 w0r = swap_workaround (bfs_buf[il_pos].i);
524 final[ 0] = swap_workaround (w0[0] | s0[0]);
525 final[ 1] = swap_workaround (w0[1] | s0[1]);
526 final[ 2] = swap_workaround (w0[2] | s0[2]);
527 final[ 3] = swap_workaround (w0[3] | s0[3]);
528 final[ 4] = swap_workaround (w1[0] | s1[0]);
529 final[ 5] = swap_workaround (w1[1] | s1[1]);
530 final[ 6] = swap_workaround (w1[2] | s1[2]);
531 final[ 7] = swap_workaround (w1[3] | s1[3]);
532 final[ 8] = swap_workaround (w2[0] | s2[0]);
533 final[ 9] = swap_workaround (w2[1] | s2[1]);
534 final[10] = swap_workaround (w2[2] | s2[2]);
535 final[11] = swap_workaround (w2[3] | s2[3]);
536 final[12] = swap_workaround (w3[0] | s3[0]);
537 final[13] = swap_workaround (w3[1] | s3[1]);
539 final[15] = pw_salt_len * 8;
549 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
551 // prepare magic array range
553 u32x lengthMagicArray = 0x20;
554 u32x offsetMagicArray = 0;
556 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
557 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
558 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
559 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
560 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
561 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
562 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
563 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
564 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
565 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
566 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
567 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
568 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
569 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
570 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
571 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
572 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
573 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
574 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
575 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
586 for (int i = 0; i < 64; i++) final[i] = 0;
597 u32 final_len = pw_len;
603 for (i = 0; i < lengthMagicArray - 4; i += 4)
605 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
607 SETSHIFTEDINT (final, final_len + i, tmp);
610 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
612 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
614 SETSHIFTEDINT (final, final_len + i, tmp);
616 final_len += lengthMagicArray;
620 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
622 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
624 SETSHIFTEDINT (final, final_len + i, tmp);
627 final_len += salt_len;
634 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
636 swap_buffer (&final[off]);
638 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
641 swap_buffer (&final[off]);
644 final[off + 15] = final_len * 8;
646 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
648 const u32x r0 = digest[3];
649 const u32x r1 = digest[4];
650 const u32x r2 = digest[2];
651 const u32x r3 = digest[1];
653 #include VECT_COMPARE_S
657 __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
663 const u32 gid = get_global_id (0);
665 if (gid >= gid_max) return;
671 const u32 lid = get_local_id (0);
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];
701 const u32 pw_len = pws[gid].pw_len;
707 m07800m (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, esalt_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);
710 __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 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);
718 if (gid >= gid_max) return;
724 const u32 lid = get_local_id (0);
728 w0[0] = pws[gid].i[ 0];
729 w0[1] = pws[gid].i[ 1];
730 w0[2] = pws[gid].i[ 2];
731 w0[3] = pws[gid].i[ 3];
735 w1[0] = pws[gid].i[ 4];
736 w1[1] = pws[gid].i[ 5];
737 w1[2] = pws[gid].i[ 6];
738 w1[3] = pws[gid].i[ 7];
754 const u32 pw_len = pws[gid].pw_len;
760 m07800m (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, esalt_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);
763 __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
767 __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
773 const u32 gid = get_global_id (0);
775 if (gid >= gid_max) return;
781 const u32 lid = get_local_id (0);
785 w0[0] = pws[gid].i[ 0];
786 w0[1] = pws[gid].i[ 1];
787 w0[2] = pws[gid].i[ 2];
788 w0[3] = pws[gid].i[ 3];
811 const u32 pw_len = pws[gid].pw_len;
817 m07800s (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, esalt_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);
820 __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
826 const u32 gid = get_global_id (0);
828 if (gid >= gid_max) return;
834 const u32 lid = get_local_id (0);
838 w0[0] = pws[gid].i[ 0];
839 w0[1] = pws[gid].i[ 1];
840 w0[2] = pws[gid].i[ 2];
841 w0[3] = pws[gid].i[ 3];
845 w1[0] = pws[gid].i[ 4];
846 w1[1] = pws[gid].i[ 5];
847 w1[2] = pws[gid].i[ 6];
848 w1[3] = pws[gid].i[ 7];
864 const u32 pw_len = pws[gid].pw_len;
870 m07800s (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, esalt_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);
873 __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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)