2 * Author......: Jens Steube <jens.steube@gmail.com>
9 #include "include/constants.h"
10 #include "include/kernel_vendor.h"
25 #include "include/kernel_functions.c"
27 #include "common_nv.c"
30 #define VECT_COMPARE_S "check_single_vect1_comp4_warp.c"
31 #define VECT_COMPARE_M "check_multi_vect1_comp4_warp.c"
35 #define VECT_COMPARE_S "check_single_vect2_comp4_warp.c"
36 #define VECT_COMPARE_M "check_multi_vect2_comp4_warp.c"
40 #define VECT_COMPARE_S "check_single_vect4_comp4_warp.c"
41 #define VECT_COMPARE_M "check_multi_vect4_comp4_warp.c"
44 __device__ __constant__ u32x c_bfs[1024];
46 __device__ __constant__ u64 k_sha384[80] =
48 SHA384C00, SHA384C01, SHA384C02, SHA384C03,
49 SHA384C04, SHA384C05, SHA384C06, SHA384C07,
50 SHA384C08, SHA384C09, SHA384C0a, SHA384C0b,
51 SHA384C0c, SHA384C0d, SHA384C0e, SHA384C0f,
52 SHA384C10, SHA384C11, SHA384C12, SHA384C13,
53 SHA384C14, SHA384C15, SHA384C16, SHA384C17,
54 SHA384C18, SHA384C19, SHA384C1a, SHA384C1b,
55 SHA384C1c, SHA384C1d, SHA384C1e, SHA384C1f,
56 SHA384C20, SHA384C21, SHA384C22, SHA384C23,
57 SHA384C24, SHA384C25, SHA384C26, SHA384C27,
58 SHA384C28, SHA384C29, SHA384C2a, SHA384C2b,
59 SHA384C2c, SHA384C2d, SHA384C2e, SHA384C2f,
60 SHA384C30, SHA384C31, SHA384C32, SHA384C33,
61 SHA384C34, SHA384C35, SHA384C36, SHA384C37,
62 SHA384C38, SHA384C39, SHA384C3a, SHA384C3b,
63 SHA384C3c, SHA384C3d, SHA384C3e, SHA384C3f,
64 SHA384C40, SHA384C41, SHA384C42, SHA384C43,
65 SHA384C44, SHA384C45, SHA384C46, SHA384C47,
66 SHA384C48, SHA384C49, SHA384C4a, SHA384C4b,
67 SHA384C4c, SHA384C4d, SHA384C4e, SHA384C4f,
70 __device__ static void sha384_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u64x digest[8])
72 u64x w0_t = hl32_to_64 (w0[0], w0[1]);
73 u64x w1_t = hl32_to_64 (w0[2], w0[3]);
74 u64x w2_t = hl32_to_64 (w1[0], w1[1]);
75 u64x w3_t = hl32_to_64 (w1[2], w1[3]);
76 u64x w4_t = hl32_to_64 (w2[0], w2[1]);
77 u64x w5_t = hl32_to_64 (w2[2], w2[3]);
78 u64x w6_t = hl32_to_64 (w3[0], w3[1]);
87 u64x wf_t = hl32_to_64 (w3[2], w3[3]);
98 #define ROUND_EXPAND() \
100 w0_t = SHA384_EXPAND (we_t, w9_t, w1_t, w0_t); \
101 w1_t = SHA384_EXPAND (wf_t, wa_t, w2_t, w1_t); \
102 w2_t = SHA384_EXPAND (w0_t, wb_t, w3_t, w2_t); \
103 w3_t = SHA384_EXPAND (w1_t, wc_t, w4_t, w3_t); \
104 w4_t = SHA384_EXPAND (w2_t, wd_t, w5_t, w4_t); \
105 w5_t = SHA384_EXPAND (w3_t, we_t, w6_t, w5_t); \
106 w6_t = SHA384_EXPAND (w4_t, wf_t, w7_t, w6_t); \
107 w7_t = SHA384_EXPAND (w5_t, w0_t, w8_t, w7_t); \
108 w8_t = SHA384_EXPAND (w6_t, w1_t, w9_t, w8_t); \
109 w9_t = SHA384_EXPAND (w7_t, w2_t, wa_t, w9_t); \
110 wa_t = SHA384_EXPAND (w8_t, w3_t, wb_t, wa_t); \
111 wb_t = SHA384_EXPAND (w9_t, w4_t, wc_t, wb_t); \
112 wc_t = SHA384_EXPAND (wa_t, w5_t, wd_t, wc_t); \
113 wd_t = SHA384_EXPAND (wb_t, w6_t, we_t, wd_t); \
114 we_t = SHA384_EXPAND (wc_t, w7_t, wf_t, we_t); \
115 wf_t = SHA384_EXPAND (wd_t, w8_t, w0_t, wf_t); \
118 #define ROUND_STEP(i) \
120 SHA384_STEP (SHA384_F0o, SHA384_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha384[i + 0]); \
121 SHA384_STEP (SHA384_F0o, SHA384_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha384[i + 1]); \
122 SHA384_STEP (SHA384_F0o, SHA384_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha384[i + 2]); \
123 SHA384_STEP (SHA384_F0o, SHA384_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha384[i + 3]); \
124 SHA384_STEP (SHA384_F0o, SHA384_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha384[i + 4]); \
125 SHA384_STEP (SHA384_F0o, SHA384_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha384[i + 5]); \
126 SHA384_STEP (SHA384_F0o, SHA384_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha384[i + 6]); \
127 SHA384_STEP (SHA384_F0o, SHA384_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha384[i + 7]); \
128 SHA384_STEP (SHA384_F0o, SHA384_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha384[i + 8]); \
129 SHA384_STEP (SHA384_F0o, SHA384_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha384[i + 9]); \
130 SHA384_STEP (SHA384_F0o, SHA384_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha384[i + 10]); \
131 SHA384_STEP (SHA384_F0o, SHA384_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha384[i + 11]); \
132 SHA384_STEP (SHA384_F0o, SHA384_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha384[i + 12]); \
133 SHA384_STEP (SHA384_F0o, SHA384_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha384[i + 13]); \
134 SHA384_STEP (SHA384_F0o, SHA384_F1o, c, d, e, f, g, h, a, b, we_t, k_sha384[i + 14]); \
135 SHA384_STEP (SHA384_F0o, SHA384_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha384[i + 15]); \
140 for (int i = 16; i < 80; i += 16)
142 ROUND_EXPAND (); ROUND_STEP (i);
166 __device__ static void m10800m (u32 w[16], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
172 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
173 const u32 lid = threadIdx.x;
179 const u32 bf_loops = ceil ((float) bfs_cnt / VECT_DIV);
183 for (u32 il_pos = 0; il_pos < bf_loops; il_pos++)
185 const u32x w0r = c_bfs[il_pos];
187 const u32x w0 = w0l | w0r;
213 digest[0] = SHA384M_A;
214 digest[1] = SHA384M_B;
215 digest[2] = SHA384M_C;
216 digest[3] = SHA384M_D;
217 digest[4] = SHA384M_E;
218 digest[5] = SHA384M_F;
219 digest[6] = SHA384M_G;
220 digest[7] = SHA384M_H;
222 sha384_transform (w0_t, w1_t, w2_t, w3_t, digest);
224 const u32x r0 = l32_from_64 (digest[3]);
225 const u32x r1 = h32_from_64 (digest[3]);
226 const u32x r2 = l32_from_64 (digest[2]);
227 const u32x r3 = h32_from_64 (digest[2]);
229 #include VECT_COMPARE_M
233 __device__ static void m10800s (u32 w[16], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
239 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
240 const u32 lid = threadIdx.x;
246 const u32 search[4] =
248 digests_buf[digests_offset].digest_buf[DGST_R0],
249 digests_buf[digests_offset].digest_buf[DGST_R1],
250 digests_buf[digests_offset].digest_buf[DGST_R2],
251 digests_buf[digests_offset].digest_buf[DGST_R3]
258 const u32 bf_loops = ceil ((float) bfs_cnt / VECT_DIV);
262 for (u32 il_pos = 0; il_pos < bf_loops; il_pos++)
264 const u32x w0r = c_bfs[il_pos];
266 const u32x w0 = w0l | w0r;
292 digest[0] = SHA384M_A;
293 digest[1] = SHA384M_B;
294 digest[2] = SHA384M_C;
295 digest[3] = SHA384M_D;
296 digest[4] = SHA384M_E;
297 digest[5] = SHA384M_F;
298 digest[6] = SHA384M_G;
299 digest[7] = SHA384M_H;
301 sha384_transform (w0_t, w1_t, w2_t, w3_t, digest);
303 const u32x r0 = l32_from_64 (digest[3]);
304 const u32x r1 = h32_from_64 (digest[3]);
305 const u32x r2 = l32_from_64 (digest[2]);
306 const u32x r3 = h32_from_64 (digest[2]);
308 #include VECT_COMPARE_S
312 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
318 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
320 if (gid >= gid_max) return;
324 w[ 0] = pws[gid].i[ 0];
325 w[ 1] = pws[gid].i[ 1];
326 w[ 2] = pws[gid].i[ 2];
327 w[ 3] = pws[gid].i[ 3];
339 w[15] = pws[gid].i[15];
341 const u32 pw_len = pws[gid].pw_len;
347 m10800m (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);
350 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
356 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
358 if (gid >= gid_max) return;
362 w[ 0] = pws[gid].i[ 0];
363 w[ 1] = pws[gid].i[ 1];
364 w[ 2] = pws[gid].i[ 2];
365 w[ 3] = pws[gid].i[ 3];
366 w[ 4] = pws[gid].i[ 4];
367 w[ 5] = pws[gid].i[ 5];
368 w[ 6] = pws[gid].i[ 6];
369 w[ 7] = pws[gid].i[ 7];
377 w[15] = pws[gid].i[15];
379 const u32 pw_len = pws[gid].pw_len;
385 m10800m (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);
388 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
394 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
396 if (gid >= gid_max) return;
400 w[ 0] = pws[gid].i[ 0];
401 w[ 1] = pws[gid].i[ 1];
402 w[ 2] = pws[gid].i[ 2];
403 w[ 3] = pws[gid].i[ 3];
404 w[ 4] = pws[gid].i[ 4];
405 w[ 5] = pws[gid].i[ 5];
406 w[ 6] = pws[gid].i[ 6];
407 w[ 7] = pws[gid].i[ 7];
408 w[ 8] = pws[gid].i[ 8];
409 w[ 9] = pws[gid].i[ 9];
410 w[10] = pws[gid].i[10];
411 w[11] = pws[gid].i[11];
412 w[12] = pws[gid].i[12];
413 w[13] = pws[gid].i[13];
414 w[14] = pws[gid].i[14];
415 w[15] = pws[gid].i[15];
417 const u32 pw_len = pws[gid].pw_len;
423 m10800m (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);
426 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
432 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
434 if (gid >= gid_max) return;
438 w[ 0] = pws[gid].i[ 0];
439 w[ 1] = pws[gid].i[ 1];
440 w[ 2] = pws[gid].i[ 2];
441 w[ 3] = pws[gid].i[ 3];
453 w[15] = pws[gid].i[15];
455 const u32 pw_len = pws[gid].pw_len;
461 m10800s (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);
464 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
470 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
472 if (gid >= gid_max) return;
476 w[ 0] = pws[gid].i[ 0];
477 w[ 1] = pws[gid].i[ 1];
478 w[ 2] = pws[gid].i[ 2];
479 w[ 3] = pws[gid].i[ 3];
480 w[ 4] = pws[gid].i[ 4];
481 w[ 5] = pws[gid].i[ 5];
482 w[ 6] = pws[gid].i[ 6];
483 w[ 7] = pws[gid].i[ 7];
491 w[15] = pws[gid].i[15];
493 const u32 pw_len = pws[gid].pw_len;
499 m10800s (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);
502 extern "C" __global__ void __launch_bounds__ (256, 1) m10800_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
508 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
510 if (gid >= gid_max) return;
514 w[ 0] = pws[gid].i[ 0];
515 w[ 1] = pws[gid].i[ 1];
516 w[ 2] = pws[gid].i[ 2];
517 w[ 3] = pws[gid].i[ 3];
518 w[ 4] = pws[gid].i[ 4];
519 w[ 5] = pws[gid].i[ 5];
520 w[ 6] = pws[gid].i[ 6];
521 w[ 7] = pws[gid].i[ 7];
522 w[ 8] = pws[gid].i[ 8];
523 w[ 9] = pws[gid].i[ 9];
524 w[10] = pws[gid].i[10];
525 w[11] = pws[gid].i[11];
526 w[12] = pws[gid].i[12];
527 w[13] = pws[gid].i[13];
528 w[14] = pws[gid].i[14];
529 w[15] = pws[gid].i[15];
531 const u32 pw_len = pws[gid].pw_len;
537 m10800s (w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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);