2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
29 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
30 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
33 __device__ static u32 bytealign (u32 src0, u32 src1, u32 src2)
35 return (u32) (((((u64)src0) << 32) | (u64)src1) >> ((src2 & 3)*8));
38 #if __CUDA_ARCH__ >= 350
39 #define GETSHIFTEDINT(a,n) __funnelshift_r ((a)[((n)/4)+0], (a)[((n)/4)+1], (n & 3) * 8)
40 #elif __CUDA_ARCH__ >= 200
41 #define GETSHIFTEDINT(a,n) __byte_perm ((a)[((n)/4)+0], (a)[((n)/4)+1], (0x76543210 >> ((n & 3) * 4)) & 0xffff)
43 #define GETSHIFTEDINT(a,n) bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], n)
46 #define SETSHIFTEDINT(a,n,v) \
48 const u32 s = ((n) & 3) * 8; \
49 const u64 x = (u64) (v) << s; \
50 (a)[((n)/4)+0] |= x; \
51 (a)[((n)/4)+1] = x >> 32; \
54 __device__ __constant__ u32 theMagicArray[64] =
56 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
57 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
58 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
59 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
60 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
61 0, 0, 0, 0, 0, 0, 0, 0,
62 0, 0, 0, 0, 0, 0, 0, 0,
63 0, 0, 0, 0, 0, 0, 0, 0
66 __device__ static void swap_buffer (u32x final[16])
68 final[ 0] = swap_workaround (final[ 0]);
69 final[ 1] = swap_workaround (final[ 1]);
70 final[ 2] = swap_workaround (final[ 2]);
71 final[ 3] = swap_workaround (final[ 3]);
72 final[ 4] = swap_workaround (final[ 4]);
73 final[ 5] = swap_workaround (final[ 5]);
74 final[ 6] = swap_workaround (final[ 6]);
75 final[ 7] = swap_workaround (final[ 7]);
76 final[ 8] = swap_workaround (final[ 8]);
77 final[ 9] = swap_workaround (final[ 9]);
78 final[10] = swap_workaround (final[10]);
79 final[11] = swap_workaround (final[11]);
80 final[12] = swap_workaround (final[12]);
81 final[13] = swap_workaround (final[13]);
82 final[14] = swap_workaround (final[14]);
83 final[15] = swap_workaround (final[15]);
86 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
114 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
115 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
116 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
117 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
118 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
119 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
120 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
121 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
122 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
123 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
124 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
125 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
126 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
127 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
128 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
129 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
130 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
131 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
132 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
133 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
138 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
139 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
140 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
141 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
142 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
143 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
144 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
145 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
146 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
147 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
148 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
149 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
150 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
151 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
152 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
153 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
154 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
155 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
156 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
157 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
162 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
163 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
164 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
165 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
166 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
167 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
168 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
169 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
170 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
171 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
172 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
173 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
174 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
175 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
176 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
177 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
178 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
179 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
180 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
181 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
186 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
187 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
188 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
189 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
190 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
191 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
192 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
193 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
194 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
195 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
196 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
197 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
198 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
199 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
200 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
201 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
202 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
203 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
204 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
205 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
214 __device__ __constant__ comb_t c_combs[1024];
216 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
222 const u32 lid = threadIdx.x;
228 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
230 if (gid >= gid_max) return;
234 wordl0[0] = pws[gid].i[ 0];
235 wordl0[1] = pws[gid].i[ 1];
236 wordl0[2] = pws[gid].i[ 2];
237 wordl0[3] = pws[gid].i[ 3];
241 wordl1[0] = pws[gid].i[ 4];
242 wordl1[1] = pws[gid].i[ 5];
243 wordl1[2] = pws[gid].i[ 6];
244 wordl1[3] = pws[gid].i[ 7];
260 const u32 pw_l_len = pws[gid].pw_len;
262 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
264 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
273 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
274 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
275 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
276 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
277 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
278 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
279 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
280 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
282 const u32 salt_len = salt_bufs[salt_pos].salt_len;
288 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
290 const u32 pw_r_len = c_combs[il_pos].pw_len;
292 const u32 pw_len = pw_l_len + pw_r_len;
296 wordr0[0] = c_combs[il_pos].i[0];
297 wordr0[1] = c_combs[il_pos].i[1];
298 wordr0[2] = c_combs[il_pos].i[2];
299 wordr0[3] = c_combs[il_pos].i[3];
303 wordr1[0] = c_combs[il_pos].i[4];
304 wordr1[1] = c_combs[il_pos].i[5];
305 wordr1[2] = c_combs[il_pos].i[6];
306 wordr1[3] = c_combs[il_pos].i[7];
322 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
324 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
359 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
361 const u32 pw_salt_len = pw_len + salt_len;
365 w0[0] = wordl0[0] | wordr0[0];
366 w0[1] = wordl0[1] | wordr0[1];
367 w0[2] = wordl0[2] | wordr0[2];
368 w0[3] = wordl0[3] | wordr0[3];
372 w1[0] = wordl1[0] | wordr1[0];
373 w1[1] = wordl1[1] | wordr1[1];
374 w1[2] = wordl1[2] | wordr1[2];
375 w1[3] = wordl1[3] | wordr1[3];
379 w2[0] = wordl2[0] | wordr2[0];
380 w2[1] = wordl2[1] | wordr2[1];
381 w2[2] = wordl2[2] | wordr2[2];
382 w2[3] = wordl2[3] | wordr2[3];
386 w3[0] = wordl3[0] | wordr3[0];
387 w3[1] = wordl3[1] | wordr3[1];
397 final[ 0] = swap_workaround (w0[0] | s0[0]);
398 final[ 1] = swap_workaround (w0[1] | s0[1]);
399 final[ 2] = swap_workaround (w0[2] | s0[2]);
400 final[ 3] = swap_workaround (w0[3] | s0[3]);
401 final[ 4] = swap_workaround (w1[0] | s1[0]);
402 final[ 5] = swap_workaround (w1[1] | s1[1]);
403 final[ 6] = swap_workaround (w1[2] | s1[2]);
404 final[ 7] = swap_workaround (w1[3] | s1[3]);
405 final[ 8] = swap_workaround (w2[0] | s2[0]);
406 final[ 9] = swap_workaround (w2[1] | s2[1]);
407 final[10] = swap_workaround (w2[2] | s2[2]);
408 final[11] = swap_workaround (w2[3] | s2[3]);
409 final[12] = swap_workaround (w3[0] | s3[0]);
410 final[13] = swap_workaround (w3[1] | s3[1]);
412 final[15] = pw_salt_len * 8;
422 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
424 // prepare magic array range
426 u32x lengthMagicArray = 0x20;
427 u32x offsetMagicArray = 0;
429 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
430 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
431 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
432 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
433 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
434 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
435 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
436 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
437 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
438 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
439 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
440 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
441 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
442 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
443 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
444 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
445 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
446 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
447 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
448 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
459 for (int i = 0; i < 64; i++) final[i] = 0;
470 u32 final_len = pw_len;
476 for (i = 0; i < lengthMagicArray - 4; i += 4)
478 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
480 SETSHIFTEDINT (final, final_len + i, tmp);
483 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
485 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
487 SETSHIFTEDINT (final, final_len + i, tmp);
489 final_len += lengthMagicArray;
493 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
495 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
497 SETSHIFTEDINT (final, final_len + i, tmp);
500 final_len += salt_len;
507 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
509 swap_buffer (&final[off]);
511 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
514 swap_buffer (&final[off]);
517 final[off + 15] = final_len * 8;
519 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
521 const u32x r0 = digest[3];
522 const u32x r1 = digest[4];
523 const u32x r2 = digest[2];
524 const u32x r3 = digest[1];
526 #include VECT_COMPARE_M
530 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
534 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
538 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
544 const u32 lid = threadIdx.x;
550 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
552 if (gid >= gid_max) return;
556 wordl0[0] = pws[gid].i[ 0];
557 wordl0[1] = pws[gid].i[ 1];
558 wordl0[2] = pws[gid].i[ 2];
559 wordl0[3] = pws[gid].i[ 3];
563 wordl1[0] = pws[gid].i[ 4];
564 wordl1[1] = pws[gid].i[ 5];
565 wordl1[2] = pws[gid].i[ 6];
566 wordl1[3] = pws[gid].i[ 7];
582 const u32 pw_l_len = pws[gid].pw_len;
584 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
586 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
589 const u32 search[4] =
591 digests_buf[digests_offset].digest_buf[DGST_R0],
592 digests_buf[digests_offset].digest_buf[DGST_R1],
593 digests_buf[digests_offset].digest_buf[DGST_R2],
594 digests_buf[digests_offset].digest_buf[DGST_R3]
603 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
604 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
605 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
606 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
607 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
608 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
609 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
610 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
612 const u32 salt_len = salt_bufs[salt_pos].salt_len;
618 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
620 const u32 pw_r_len = c_combs[il_pos].pw_len;
622 const u32 pw_len = pw_l_len + pw_r_len;
626 wordr0[0] = c_combs[il_pos].i[0];
627 wordr0[1] = c_combs[il_pos].i[1];
628 wordr0[2] = c_combs[il_pos].i[2];
629 wordr0[3] = c_combs[il_pos].i[3];
633 wordr1[0] = c_combs[il_pos].i[4];
634 wordr1[1] = c_combs[il_pos].i[5];
635 wordr1[2] = c_combs[il_pos].i[6];
636 wordr1[3] = c_combs[il_pos].i[7];
652 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
654 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
689 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
691 const u32 pw_salt_len = pw_len + salt_len;
695 w0[0] = wordl0[0] | wordr0[0];
696 w0[1] = wordl0[1] | wordr0[1];
697 w0[2] = wordl0[2] | wordr0[2];
698 w0[3] = wordl0[3] | wordr0[3];
702 w1[0] = wordl1[0] | wordr1[0];
703 w1[1] = wordl1[1] | wordr1[1];
704 w1[2] = wordl1[2] | wordr1[2];
705 w1[3] = wordl1[3] | wordr1[3];
709 w2[0] = wordl2[0] | wordr2[0];
710 w2[1] = wordl2[1] | wordr2[1];
711 w2[2] = wordl2[2] | wordr2[2];
712 w2[3] = wordl2[3] | wordr2[3];
716 w3[0] = wordl3[0] | wordr3[0];
717 w3[1] = wordl3[1] | wordr3[1];
727 final[ 0] = swap_workaround (w0[0] | s0[0]);
728 final[ 1] = swap_workaround (w0[1] | s0[1]);
729 final[ 2] = swap_workaround (w0[2] | s0[2]);
730 final[ 3] = swap_workaround (w0[3] | s0[3]);
731 final[ 4] = swap_workaround (w1[0] | s1[0]);
732 final[ 5] = swap_workaround (w1[1] | s1[1]);
733 final[ 6] = swap_workaround (w1[2] | s1[2]);
734 final[ 7] = swap_workaround (w1[3] | s1[3]);
735 final[ 8] = swap_workaround (w2[0] | s2[0]);
736 final[ 9] = swap_workaround (w2[1] | s2[1]);
737 final[10] = swap_workaround (w2[2] | s2[2]);
738 final[11] = swap_workaround (w2[3] | s2[3]);
739 final[12] = swap_workaround (w3[0] | s3[0]);
740 final[13] = swap_workaround (w3[1] | s3[1]);
742 final[15] = pw_salt_len * 8;
752 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
754 // prepare magic array range
756 u32x lengthMagicArray = 0x20;
757 u32x offsetMagicArray = 0;
759 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
760 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
761 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
762 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
763 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
764 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
765 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
766 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
767 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
768 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
769 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
770 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
771 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
772 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
773 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
774 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
775 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
776 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
777 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
778 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
789 for (int i = 0; i < 64; i++) final[i] = 0;
800 u32 final_len = pw_len;
806 for (i = 0; i < lengthMagicArray - 4; i += 4)
808 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
810 SETSHIFTEDINT (final, final_len + i, tmp);
813 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
815 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
817 SETSHIFTEDINT (final, final_len + i, tmp);
819 final_len += lengthMagicArray;
823 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
825 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
827 SETSHIFTEDINT (final, final_len + i, tmp);
830 final_len += salt_len;
837 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
839 swap_buffer (&final[off]);
841 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
844 swap_buffer (&final[off]);
847 final[off + 15] = final_len * 8;
849 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
851 const u32x r0 = digest[3];
852 const u32x r1 = digest[4];
853 const u32x r2 = digest[2];
854 const u32x r3 = digest[1];
856 #include VECT_COMPARE_S
860 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
864 extern "C" __global__ void __launch_bounds__ (256, 1) m07800_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)