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"
34 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
35 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
38 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
66 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
67 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
68 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
69 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
70 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
71 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
72 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
73 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
74 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
75 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
76 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
77 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
78 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
79 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
80 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
81 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
82 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
83 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
84 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
85 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
90 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
91 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
92 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
93 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
94 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
95 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
96 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
97 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
98 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
99 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
100 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
101 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
102 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
103 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
104 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
105 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
106 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
107 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
108 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
109 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
114 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
115 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
116 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
117 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
118 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
119 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
120 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
121 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
122 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
123 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
124 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
125 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
126 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
127 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
128 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
129 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
130 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
131 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
132 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
133 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
138 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
139 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
140 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
141 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
142 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
143 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
144 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
145 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
146 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
147 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
148 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
149 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
150 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
151 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
152 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
153 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
154 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
155 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
156 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
157 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
166 __device__ static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
168 w0[0] = w0[0] ^ 0x36363636;
169 w0[1] = w0[1] ^ 0x36363636;
170 w0[2] = w0[2] ^ 0x36363636;
171 w0[3] = w0[3] ^ 0x36363636;
172 w1[0] = w1[0] ^ 0x36363636;
173 w1[1] = w1[1] ^ 0x36363636;
174 w1[2] = w1[2] ^ 0x36363636;
175 w1[3] = w1[3] ^ 0x36363636;
176 w2[0] = w2[0] ^ 0x36363636;
177 w2[1] = w2[1] ^ 0x36363636;
178 w2[2] = w2[2] ^ 0x36363636;
179 w2[3] = w2[3] ^ 0x36363636;
180 w3[0] = w3[0] ^ 0x36363636;
181 w3[1] = w3[1] ^ 0x36363636;
182 w3[2] = w3[2] ^ 0x36363636;
183 w3[3] = w3[3] ^ 0x36363636;
191 sha1_transform (w0, w1, w2, w3, ipad);
193 w0[0] = w0[0] ^ 0x6a6a6a6a;
194 w0[1] = w0[1] ^ 0x6a6a6a6a;
195 w0[2] = w0[2] ^ 0x6a6a6a6a;
196 w0[3] = w0[3] ^ 0x6a6a6a6a;
197 w1[0] = w1[0] ^ 0x6a6a6a6a;
198 w1[1] = w1[1] ^ 0x6a6a6a6a;
199 w1[2] = w1[2] ^ 0x6a6a6a6a;
200 w1[3] = w1[3] ^ 0x6a6a6a6a;
201 w2[0] = w2[0] ^ 0x6a6a6a6a;
202 w2[1] = w2[1] ^ 0x6a6a6a6a;
203 w2[2] = w2[2] ^ 0x6a6a6a6a;
204 w2[3] = w2[3] ^ 0x6a6a6a6a;
205 w3[0] = w3[0] ^ 0x6a6a6a6a;
206 w3[1] = w3[1] ^ 0x6a6a6a6a;
207 w3[2] = w3[2] ^ 0x6a6a6a6a;
208 w3[3] = w3[3] ^ 0x6a6a6a6a;
216 sha1_transform (w0, w1, w2, w3, opad);
219 __device__ static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
227 sha1_transform (w0, w1, w2, w3, digest);
244 w3[3] = (64 + 20) * 8;
252 sha1_transform (w0, w1, w2, w3, digest);
255 __device__ __constant__ bf_t c_bfs[1024];
257 __device__ static void m07300m (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, 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 rakp_t *rakp_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)
263 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
264 const u32 lid = threadIdx.x;
270 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
278 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
280 const u32 w0r = c_bfs[il_pos].i;
319 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
321 int esalt_size = esalt_len;
326 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
328 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
329 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
330 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
331 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
332 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
333 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
334 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
335 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
336 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
337 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
338 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
339 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
340 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
341 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
342 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
343 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
345 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
348 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
349 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
350 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
351 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
352 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
353 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
354 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
355 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
356 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
357 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
358 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
359 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
360 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
361 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
363 w3_t[3] = (64 + esalt_size) * 8;
367 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
369 const u32x r0 = digest[3];
370 const u32x r1 = digest[4];
371 const u32x r2 = digest[2];
372 const u32x r3 = digest[1];
374 #include VECT_COMPARE_M
378 __device__ static void m07300s (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, 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 rakp_t *rakp_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)
384 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
385 const u32 lid = threadIdx.x;
391 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
397 const u32 search[4] =
399 digests_buf[digests_offset].digest_buf[DGST_R0],
400 digests_buf[digests_offset].digest_buf[DGST_R1],
401 digests_buf[digests_offset].digest_buf[DGST_R2],
402 digests_buf[digests_offset].digest_buf[DGST_R3]
411 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
413 const u32 w0r = c_bfs[il_pos].i;
452 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
454 int esalt_size = esalt_len;
459 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
461 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
462 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
463 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
464 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
465 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
466 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
467 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
468 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
469 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
470 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
471 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
472 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
473 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
474 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
475 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
476 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
478 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
481 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
482 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
483 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
484 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
485 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
486 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
487 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
488 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
489 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
490 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
491 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
492 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
493 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
494 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
496 w3_t[3] = (64 + esalt_size) * 8;
500 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
502 const u32x r0 = digest[3];
503 const u32x r1 = digest[4];
504 const u32x r2 = digest[2];
505 const u32x r3 = digest[1];
507 #include VECT_COMPARE_S
511 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
517 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
519 if (gid >= gid_max) return;
523 w0[0] = pws[gid].i[ 0];
524 w0[1] = pws[gid].i[ 1];
525 w0[2] = pws[gid].i[ 2];
526 w0[3] = pws[gid].i[ 3];
549 const u32 pw_len = pws[gid].pw_len;
555 m07300m (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, rakp_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);
558 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
564 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
566 if (gid >= gid_max) return;
570 w0[0] = pws[gid].i[ 0];
571 w0[1] = pws[gid].i[ 1];
572 w0[2] = pws[gid].i[ 2];
573 w0[3] = pws[gid].i[ 3];
577 w1[0] = pws[gid].i[ 4];
578 w1[1] = pws[gid].i[ 5];
579 w1[2] = pws[gid].i[ 6];
580 w1[3] = pws[gid].i[ 7];
596 const u32 pw_len = pws[gid].pw_len;
602 m07300m (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, rakp_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);
605 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
611 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
613 if (gid >= gid_max) return;
617 w0[0] = pws[gid].i[ 0];
618 w0[1] = pws[gid].i[ 1];
619 w0[2] = pws[gid].i[ 2];
620 w0[3] = pws[gid].i[ 3];
624 w1[0] = pws[gid].i[ 4];
625 w1[1] = pws[gid].i[ 5];
626 w1[2] = pws[gid].i[ 6];
627 w1[3] = pws[gid].i[ 7];
631 w2[0] = pws[gid].i[ 8];
632 w2[1] = pws[gid].i[ 9];
633 w2[2] = pws[gid].i[10];
634 w2[3] = pws[gid].i[11];
638 w3[0] = pws[gid].i[12];
639 w3[1] = pws[gid].i[13];
643 const u32 pw_len = pws[gid].pw_len;
649 m07300m (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, rakp_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);
652 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
658 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
660 if (gid >= gid_max) return;
664 w0[0] = pws[gid].i[ 0];
665 w0[1] = pws[gid].i[ 1];
666 w0[2] = pws[gid].i[ 2];
667 w0[3] = pws[gid].i[ 3];
690 const u32 pw_len = pws[gid].pw_len;
696 m07300s (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, rakp_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);
699 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
705 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
707 if (gid >= gid_max) return;
711 w0[0] = pws[gid].i[ 0];
712 w0[1] = pws[gid].i[ 1];
713 w0[2] = pws[gid].i[ 2];
714 w0[3] = pws[gid].i[ 3];
718 w1[0] = pws[gid].i[ 4];
719 w1[1] = pws[gid].i[ 5];
720 w1[2] = pws[gid].i[ 6];
721 w1[3] = pws[gid].i[ 7];
737 const u32 pw_len = pws[gid].pw_len;
743 m07300s (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, rakp_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);
746 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
752 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
754 if (gid >= gid_max) return;
758 w0[0] = pws[gid].i[ 0];
759 w0[1] = pws[gid].i[ 1];
760 w0[2] = pws[gid].i[ 2];
761 w0[3] = pws[gid].i[ 3];
765 w1[0] = pws[gid].i[ 4];
766 w1[1] = pws[gid].i[ 5];
767 w1[2] = pws[gid].i[ 6];
768 w1[3] = pws[gid].i[ 7];
772 w2[0] = pws[gid].i[ 8];
773 w2[1] = pws[gid].i[ 9];
774 w2[2] = pws[gid].i[10];
775 w2[3] = pws[gid].i[11];
779 w3[0] = pws[gid].i[12];
780 w3[1] = pws[gid].i[13];
784 const u32 pw_len = pws[gid].pw_len;
790 m07300s (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, rakp_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);