2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
31 #include "include/rp_gpu.h"
35 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
36 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
40 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
41 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
45 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
46 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
49 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
77 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
78 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
79 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
80 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
81 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
82 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
83 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
84 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
85 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
86 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
87 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
88 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
89 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
90 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
91 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
92 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
93 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
94 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
95 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
96 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
101 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
102 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
103 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
104 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
105 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
106 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
107 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
108 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
109 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
110 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
111 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
112 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
113 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
114 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
115 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
116 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
117 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
118 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
119 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
120 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
125 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
126 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
127 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
128 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
129 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
130 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
131 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
132 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
133 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
134 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
135 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
136 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
137 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
138 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
139 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
140 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
141 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
142 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
143 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
144 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
149 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
150 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
151 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
152 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
153 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
154 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
155 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
156 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
157 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
158 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
159 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
160 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
161 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
162 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
163 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
164 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
165 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
166 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
167 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
168 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
177 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
183 const u32 lid = get_local_id (0);
189 const u32 gid = get_global_id (0);
191 if (gid >= gid_max) return;
195 pw_buf0[0] = pws[gid].i[ 0];
196 pw_buf0[1] = pws[gid].i[ 1];
197 pw_buf0[2] = pws[gid].i[ 2];
198 pw_buf0[3] = pws[gid].i[ 3];
202 pw_buf1[0] = pws[gid].i[ 4];
203 pw_buf1[1] = pws[gid].i[ 5];
204 pw_buf1[2] = pws[gid].i[ 6];
205 pw_buf1[3] = pws[gid].i[ 7];
207 const u32 pw_len = pws[gid].pw_len;
215 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
216 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
217 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
218 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
220 const u32 salt_len = 16;
222 const u32 pw_salt_len = pw_len + salt_len;
228 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
258 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
260 const u32 pw_salt_len = (out_len * 2) + salt_len;
262 append_0x80_2 (w0, w1, out_len);
269 make_unicode (w0, w0_t, w1_t);
270 make_unicode (w1, w2_t, w3_t);
272 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
274 w0_t[0] = salt_buf[0];
275 w0_t[1] = salt_buf[1];
276 w0_t[2] = salt_buf[2];
277 w0_t[3] = salt_buf[3];
278 w1_t[0] = swap_workaround (w1_t[0]);
279 w1_t[1] = swap_workaround (w1_t[1]);
280 w1_t[2] = swap_workaround (w1_t[2]);
281 w1_t[3] = swap_workaround (w1_t[3]);
282 w2_t[0] = swap_workaround (w2_t[0]);
283 w2_t[1] = swap_workaround (w2_t[1]);
284 w2_t[2] = swap_workaround (w2_t[2]);
285 w2_t[3] = swap_workaround (w2_t[3]);
286 w3_t[0] = swap_workaround (w3_t[0]);
287 w3_t[1] = swap_workaround (w3_t[1]);
289 w3_t[3] = pw_salt_len * 8;
299 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
307 w1_t[2] = 0x80000000;
316 w3_t[3] = (20 + 4) * 8;
324 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
326 u32x a = swap_workaround (digest[0]);
327 u32x b = swap_workaround (digest[1]) & 0xff;
334 #include VECT_COMPARE_M
338 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
342 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
346 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
352 const u32 lid = get_local_id (0);
358 const u32 gid = get_global_id (0);
360 if (gid >= gid_max) return;
364 pw_buf0[0] = pws[gid].i[ 0];
365 pw_buf0[1] = pws[gid].i[ 1];
366 pw_buf0[2] = pws[gid].i[ 2];
367 pw_buf0[3] = pws[gid].i[ 3];
371 pw_buf1[0] = pws[gid].i[ 4];
372 pw_buf1[1] = pws[gid].i[ 5];
373 pw_buf1[2] = pws[gid].i[ 6];
374 pw_buf1[3] = pws[gid].i[ 7];
376 const u32 pw_len = pws[gid].pw_len;
382 const u32 search[4] =
384 digests_buf[digests_offset].digest_buf[DGST_R0],
385 digests_buf[digests_offset].digest_buf[DGST_R1],
386 digests_buf[digests_offset].digest_buf[DGST_R2],
387 digests_buf[digests_offset].digest_buf[DGST_R3]
396 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
397 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
398 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
399 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
401 const u32 salt_len = 16;
403 const u32 pw_salt_len = pw_len + salt_len;
409 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
439 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
441 const u32 pw_salt_len = (out_len * 2) + salt_len;
443 append_0x80_2 (w0, w1, out_len);
450 make_unicode (w0, w0_t, w1_t);
451 make_unicode (w1, w2_t, w3_t);
453 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
455 w0_t[0] = salt_buf[0];
456 w0_t[1] = salt_buf[1];
457 w0_t[2] = salt_buf[2];
458 w0_t[3] = salt_buf[3];
459 w1_t[0] = swap_workaround (w1_t[0]);
460 w1_t[1] = swap_workaround (w1_t[1]);
461 w1_t[2] = swap_workaround (w1_t[2]);
462 w1_t[3] = swap_workaround (w1_t[3]);
463 w2_t[0] = swap_workaround (w2_t[0]);
464 w2_t[1] = swap_workaround (w2_t[1]);
465 w2_t[2] = swap_workaround (w2_t[2]);
466 w2_t[3] = swap_workaround (w2_t[3]);
467 w3_t[0] = swap_workaround (w3_t[0]);
468 w3_t[1] = swap_workaround (w3_t[1]);
470 w3_t[3] = pw_salt_len * 8;
480 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
488 w1_t[2] = 0x80000000;
497 w3_t[3] = (20 + 4) * 8;
505 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
507 u32x a = swap_workaround (digest[0]);
508 u32x b = swap_workaround (digest[1]) & 0xff;
515 #include VECT_COMPARE_S
519 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
523 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)