2 * Author......: Jens Steube <jens.steube@gmail.com>
10 #include "include/constants.h"
11 #include "include/kernel_vendor.h"
18 #include "include/kernel_functions.c"
19 #include "OpenCL/types_ocl.c"
20 #include "OpenCL/common.c"
21 #include "OpenCL/simd.c"
23 #define uint_to_hex_lower8(i) l_bin2asc[(i)]
25 __kernel void m11100_m04 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
31 const u32 gid = get_global_id (0);
32 const u32 lid = get_local_id (0);
33 const u32 lsz = get_local_size (0);
39 __local u32 l_bin2asc[256];
41 for (u32 i = lid; i < 256; i += lsz)
43 const u32 i0 = (i >> 0) & 15;
44 const u32 i1 = (i >> 4) & 15;
46 l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
47 | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
50 barrier (CLK_LOCAL_MEM_FENCE);
52 if (gid >= gid_max) return;
61 pws0[0] = pws[gid].i[0];
62 pws0[1] = pws[gid].i[1];
63 pws0[2] = pws[gid].i[2];
64 pws0[3] = pws[gid].i[3];
65 pws1[0] = pws[gid].i[4];
66 pws1[1] = pws[gid].i[5];
67 pws1[2] = pws[gid].i[6];
68 pws1[3] = pws[gid].i[7];
70 const u32 pw_l_len = pws[gid].pw_len;
78 challenge = salt_bufs[salt_pos].salt_buf[0];
86 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 1]; // not a bug
87 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 2];
88 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 3];
89 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 4];
93 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 5];
94 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 6];
95 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 7];
96 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 8];
98 const u32 salt_len = salt_bufs[salt_pos].salt_len - 4;
104 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
106 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
108 const u32x pw_len = pw_l_len + pw_r_len;
110 u32x wordr0[4] = { 0 };
111 u32x wordr1[4] = { 0 };
112 u32x wordr2[4] = { 0 };
113 u32x wordr3[4] = { 0 };
115 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
116 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
117 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
118 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
119 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
120 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
121 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
122 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
124 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
126 switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
131 w0_t[0] = salt_buf0[0];
132 w0_t[1] = salt_buf0[1];
133 w0_t[2] = salt_buf0[2];
134 w0_t[3] = salt_buf0[3];
138 w1_t[0] = salt_buf1[0];
139 w1_t[1] = salt_buf1[1];
140 w1_t[2] = salt_buf1[2];
141 w1_t[3] = salt_buf1[3];
161 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, pw_len);
163 const u32x pw_salt_len = pw_len + salt_len;
165 w0_t[0] |= wordl0[0] | wordr0[0];
166 w0_t[1] |= wordl0[1] | wordr0[1];
167 w0_t[2] |= wordl0[2] | wordr0[2];
168 w0_t[3] |= wordl0[3] | wordr0[3];
170 w1_t[0] |= wordl1[0] | wordr1[0];
171 w1_t[1] |= wordl1[1] | wordr1[1];
172 w1_t[2] |= wordl1[2] | wordr1[2];
173 w1_t[3] |= wordl1[3] | wordr1[3];
175 w2_t[0] |= wordl2[0] | wordr2[0];
176 w2_t[1] |= wordl2[1] | wordr2[1];
177 w2_t[2] |= wordl2[2] | wordr2[2];
178 w2_t[3] |= wordl2[3] | wordr2[3];
180 w3_t[0] |= wordl3[0] | wordr3[0];
181 w3_t[1] |= wordl3[0] | wordr3[0];
182 w3_t[2] = pw_salt_len * 8;
194 MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
195 MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
196 MD5_STEP (MD5_Fo, c, d, a, b, w0_t[2], MD5C02, MD5S02);
197 MD5_STEP (MD5_Fo, b, c, d, a, w0_t[3], MD5C03, MD5S03);
198 MD5_STEP (MD5_Fo, a, b, c, d, w1_t[0], MD5C04, MD5S00);
199 MD5_STEP (MD5_Fo, d, a, b, c, w1_t[1], MD5C05, MD5S01);
200 MD5_STEP (MD5_Fo, c, d, a, b, w1_t[2], MD5C06, MD5S02);
201 MD5_STEP (MD5_Fo, b, c, d, a, w1_t[3], MD5C07, MD5S03);
202 MD5_STEP (MD5_Fo, a, b, c, d, w2_t[0], MD5C08, MD5S00);
203 MD5_STEP (MD5_Fo, d, a, b, c, w2_t[1], MD5C09, MD5S01);
204 MD5_STEP (MD5_Fo, c, d, a, b, w2_t[2], MD5C0a, MD5S02);
205 MD5_STEP (MD5_Fo, b, c, d, a, w2_t[3], MD5C0b, MD5S03);
206 MD5_STEP (MD5_Fo, a, b, c, d, w3_t[0], MD5C0c, MD5S00);
207 MD5_STEP (MD5_Fo, d, a, b, c, w3_t[1], MD5C0d, MD5S01);
208 MD5_STEP (MD5_Fo, c, d, a, b, w3_t[2], MD5C0e, MD5S02);
209 MD5_STEP (MD5_Fo, b, c, d, a, w3_t[3], MD5C0f, MD5S03);
211 MD5_STEP (MD5_Go, a, b, c, d, w0_t[1], MD5C10, MD5S10);
212 MD5_STEP (MD5_Go, d, a, b, c, w1_t[2], MD5C11, MD5S11);
213 MD5_STEP (MD5_Go, c, d, a, b, w2_t[3], MD5C12, MD5S12);
214 MD5_STEP (MD5_Go, b, c, d, a, w0_t[0], MD5C13, MD5S13);
215 MD5_STEP (MD5_Go, a, b, c, d, w1_t[1], MD5C14, MD5S10);
216 MD5_STEP (MD5_Go, d, a, b, c, w2_t[2], MD5C15, MD5S11);
217 MD5_STEP (MD5_Go, c, d, a, b, w3_t[3], MD5C16, MD5S12);
218 MD5_STEP (MD5_Go, b, c, d, a, w1_t[0], MD5C17, MD5S13);
219 MD5_STEP (MD5_Go, a, b, c, d, w2_t[1], MD5C18, MD5S10);
220 MD5_STEP (MD5_Go, d, a, b, c, w3_t[2], MD5C19, MD5S11);
221 MD5_STEP (MD5_Go, c, d, a, b, w0_t[3], MD5C1a, MD5S12);
222 MD5_STEP (MD5_Go, b, c, d, a, w2_t[0], MD5C1b, MD5S13);
223 MD5_STEP (MD5_Go, a, b, c, d, w3_t[1], MD5C1c, MD5S10);
224 MD5_STEP (MD5_Go, d, a, b, c, w0_t[2], MD5C1d, MD5S11);
225 MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
226 MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
228 MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
229 MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
230 MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
231 MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
232 MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
233 MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
234 MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
235 MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
236 MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
237 MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
238 MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
239 MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
240 MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
241 MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
242 MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
243 MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
245 MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
246 MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
247 MD5_STEP (MD5_I , c, d, a, b, w3_t[2], MD5C32, MD5S32);
248 MD5_STEP (MD5_I , b, c, d, a, w1_t[1], MD5C33, MD5S33);
249 MD5_STEP (MD5_I , a, b, c, d, w3_t[0], MD5C34, MD5S30);
250 MD5_STEP (MD5_I , d, a, b, c, w0_t[3], MD5C35, MD5S31);
251 MD5_STEP (MD5_I , c, d, a, b, w2_t[2], MD5C36, MD5S32);
252 MD5_STEP (MD5_I , b, c, d, a, w0_t[1], MD5C37, MD5S33);
253 MD5_STEP (MD5_I , a, b, c, d, w2_t[0], MD5C38, MD5S30);
254 MD5_STEP (MD5_I , d, a, b, c, w3_t[3], MD5C39, MD5S31);
255 MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32);
256 MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33);
257 MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30);
258 MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31);
259 MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
260 MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
267 w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0
268 | uint_to_hex_lower8 ((a >> 8) & 255) << 16;
269 w0_t[1] = uint_to_hex_lower8 ((a >> 16) & 255) << 0
270 | uint_to_hex_lower8 ((a >> 24) & 255) << 16;
271 w0_t[2] = uint_to_hex_lower8 ((b >> 0) & 255) << 0
272 | uint_to_hex_lower8 ((b >> 8) & 255) << 16;
273 w0_t[3] = uint_to_hex_lower8 ((b >> 16) & 255) << 0
274 | uint_to_hex_lower8 ((b >> 24) & 255) << 16;
275 w1_t[0] = uint_to_hex_lower8 ((c >> 0) & 255) << 0
276 | uint_to_hex_lower8 ((c >> 8) & 255) << 16;
277 w1_t[1] = uint_to_hex_lower8 ((c >> 16) & 255) << 0
278 | uint_to_hex_lower8 ((c >> 24) & 255) << 16;
279 w1_t[2] = uint_to_hex_lower8 ((d >> 0) & 255) << 0
280 | uint_to_hex_lower8 ((d >> 8) & 255) << 16;
281 w1_t[3] = uint_to_hex_lower8 ((d >> 16) & 255) << 0
282 | uint_to_hex_lower8 ((d >> 24) & 255) << 16;
284 // add the 4 byte challenge here
287 w2_t[1] = 0x00000080;
293 w3_t[2] = (32 + 4) * 8;
297 * md5 ($hash.$challenge)
305 MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
306 MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
307 MD5_STEP (MD5_Fo, c, d, a, b, w0_t[2], MD5C02, MD5S02);
308 MD5_STEP (MD5_Fo, b, c, d, a, w0_t[3], MD5C03, MD5S03);
309 MD5_STEP (MD5_Fo, a, b, c, d, w1_t[0], MD5C04, MD5S00);
310 MD5_STEP (MD5_Fo, d, a, b, c, w1_t[1], MD5C05, MD5S01);
311 MD5_STEP (MD5_Fo, c, d, a, b, w1_t[2], MD5C06, MD5S02);
312 MD5_STEP (MD5_Fo, b, c, d, a, w1_t[3], MD5C07, MD5S03);
313 MD5_STEP (MD5_Fo, a, b, c, d, w2_t[0], MD5C08, MD5S00);
314 MD5_STEP (MD5_Fo, d, a, b, c, w2_t[1], MD5C09, MD5S01);
315 MD5_STEP (MD5_Fo, c, d, a, b, w2_t[2], MD5C0a, MD5S02);
316 MD5_STEP (MD5_Fo, b, c, d, a, w2_t[3], MD5C0b, MD5S03);
317 MD5_STEP (MD5_Fo, a, b, c, d, w3_t[0], MD5C0c, MD5S00);
318 MD5_STEP (MD5_Fo, d, a, b, c, w3_t[1], MD5C0d, MD5S01);
319 MD5_STEP (MD5_Fo, c, d, a, b, w3_t[2], MD5C0e, MD5S02);
320 MD5_STEP (MD5_Fo, b, c, d, a, w3_t[3], MD5C0f, MD5S03);
322 MD5_STEP (MD5_Go, a, b, c, d, w0_t[1], MD5C10, MD5S10);
323 MD5_STEP (MD5_Go, d, a, b, c, w1_t[2], MD5C11, MD5S11);
324 MD5_STEP (MD5_Go, c, d, a, b, w2_t[3], MD5C12, MD5S12);
325 MD5_STEP (MD5_Go, b, c, d, a, w0_t[0], MD5C13, MD5S13);
326 MD5_STEP (MD5_Go, a, b, c, d, w1_t[1], MD5C14, MD5S10);
327 MD5_STEP (MD5_Go, d, a, b, c, w2_t[2], MD5C15, MD5S11);
328 MD5_STEP (MD5_Go, c, d, a, b, w3_t[3], MD5C16, MD5S12);
329 MD5_STEP (MD5_Go, b, c, d, a, w1_t[0], MD5C17, MD5S13);
330 MD5_STEP (MD5_Go, a, b, c, d, w2_t[1], MD5C18, MD5S10);
331 MD5_STEP (MD5_Go, d, a, b, c, w3_t[2], MD5C19, MD5S11);
332 MD5_STEP (MD5_Go, c, d, a, b, w0_t[3], MD5C1a, MD5S12);
333 MD5_STEP (MD5_Go, b, c, d, a, w2_t[0], MD5C1b, MD5S13);
334 MD5_STEP (MD5_Go, a, b, c, d, w3_t[1], MD5C1c, MD5S10);
335 MD5_STEP (MD5_Go, d, a, b, c, w0_t[2], MD5C1d, MD5S11);
336 MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
337 MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
339 MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
340 MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
341 MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
342 MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
343 MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
344 MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
345 MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
346 MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
347 MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
348 MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
349 MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
350 MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
351 MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
352 MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
353 MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
354 MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
356 MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
357 MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
358 MD5_STEP (MD5_I , c, d, a, b, w3_t[2], MD5C32, MD5S32);
359 MD5_STEP (MD5_I , b, c, d, a, w1_t[1], MD5C33, MD5S33);
360 MD5_STEP (MD5_I , a, b, c, d, w3_t[0], MD5C34, MD5S30);
361 MD5_STEP (MD5_I , d, a, b, c, w0_t[3], MD5C35, MD5S31);
362 MD5_STEP (MD5_I , c, d, a, b, w2_t[2], MD5C36, MD5S32);
363 MD5_STEP (MD5_I , b, c, d, a, w0_t[1], MD5C37, MD5S33);
364 MD5_STEP (MD5_I , a, b, c, d, w2_t[0], MD5C38, MD5S30);
365 MD5_STEP (MD5_I , d, a, b, c, w3_t[3], MD5C39, MD5S31);
366 MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32);
367 MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33);
368 MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30);
369 MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31);
370 MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
371 MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
372 COMPARE_M_SIMD (a, d, c, b);
376 __kernel void m11100_m08 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
380 __kernel void m11100_m16 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
384 __kernel void m11100_s04 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
390 const u32 gid = get_global_id (0);
391 const u32 lid = get_local_id (0);
392 const u32 lsz = get_local_size (0);
398 __local u32 l_bin2asc[256];
400 for (u32 i = lid; i < 256; i += lsz)
402 const u32 i0 = (i >> 0) & 15;
403 const u32 i1 = (i >> 4) & 15;
405 l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
406 | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
409 barrier (CLK_LOCAL_MEM_FENCE);
411 if (gid >= gid_max) return;
420 pws0[0] = pws[gid].i[0];
421 pws0[1] = pws[gid].i[1];
422 pws0[2] = pws[gid].i[2];
423 pws0[3] = pws[gid].i[3];
424 pws1[0] = pws[gid].i[4];
425 pws1[1] = pws[gid].i[5];
426 pws1[2] = pws[gid].i[6];
427 pws1[3] = pws[gid].i[7];
429 const u32 pw_l_len = pws[gid].pw_len;
437 challenge = salt_bufs[salt_pos].salt_buf[0];
445 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 1]; // not a bug
446 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 2];
447 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 3];
448 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 4];
452 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 5];
453 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 6];
454 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 7];
455 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 8];
457 const u32 salt_len = salt_bufs[salt_pos].salt_len - 4;
463 const u32 search[4] =
465 digests_buf[digests_offset].digest_buf[DGST_R0],
466 digests_buf[digests_offset].digest_buf[DGST_R1],
467 digests_buf[digests_offset].digest_buf[DGST_R2],
468 digests_buf[digests_offset].digest_buf[DGST_R3]
475 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
477 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
479 const u32x pw_len = pw_l_len + pw_r_len;
481 u32x wordr0[4] = { 0 };
482 u32x wordr1[4] = { 0 };
483 u32x wordr2[4] = { 0 };
484 u32x wordr3[4] = { 0 };
486 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
487 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
488 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
489 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
490 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
491 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
492 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
493 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
495 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
497 switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
502 w0_t[0] = salt_buf0[0];
503 w0_t[1] = salt_buf0[1];
504 w0_t[2] = salt_buf0[2];
505 w0_t[3] = salt_buf0[3];
509 w1_t[0] = salt_buf1[0];
510 w1_t[1] = salt_buf1[1];
511 w1_t[2] = salt_buf1[2];
512 w1_t[3] = salt_buf1[3];
532 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, pw_len);
534 const u32x pw_salt_len = pw_len + salt_len;
536 w0_t[0] |= wordl0[0] | wordr0[0];
537 w0_t[1] |= wordl0[1] | wordr0[1];
538 w0_t[2] |= wordl0[2] | wordr0[2];
539 w0_t[3] |= wordl0[3] | wordr0[3];
541 w1_t[0] |= wordl1[0] | wordr1[0];
542 w1_t[1] |= wordl1[1] | wordr1[1];
543 w1_t[2] |= wordl1[2] | wordr1[2];
544 w1_t[3] |= wordl1[3] | wordr1[3];
546 w2_t[0] |= wordl2[0] | wordr2[0];
547 w2_t[1] |= wordl2[1] | wordr2[1];
548 w2_t[2] |= wordl2[2] | wordr2[2];
549 w2_t[3] |= wordl2[3] | wordr2[3];
551 w3_t[0] |= wordl3[0] | wordr3[0];
552 w3_t[1] |= wordl3[0] | wordr3[0];
553 w3_t[2] = pw_salt_len * 8;
565 MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
566 MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
567 MD5_STEP (MD5_Fo, c, d, a, b, w0_t[2], MD5C02, MD5S02);
568 MD5_STEP (MD5_Fo, b, c, d, a, w0_t[3], MD5C03, MD5S03);
569 MD5_STEP (MD5_Fo, a, b, c, d, w1_t[0], MD5C04, MD5S00);
570 MD5_STEP (MD5_Fo, d, a, b, c, w1_t[1], MD5C05, MD5S01);
571 MD5_STEP (MD5_Fo, c, d, a, b, w1_t[2], MD5C06, MD5S02);
572 MD5_STEP (MD5_Fo, b, c, d, a, w1_t[3], MD5C07, MD5S03);
573 MD5_STEP (MD5_Fo, a, b, c, d, w2_t[0], MD5C08, MD5S00);
574 MD5_STEP (MD5_Fo, d, a, b, c, w2_t[1], MD5C09, MD5S01);
575 MD5_STEP (MD5_Fo, c, d, a, b, w2_t[2], MD5C0a, MD5S02);
576 MD5_STEP (MD5_Fo, b, c, d, a, w2_t[3], MD5C0b, MD5S03);
577 MD5_STEP (MD5_Fo, a, b, c, d, w3_t[0], MD5C0c, MD5S00);
578 MD5_STEP (MD5_Fo, d, a, b, c, w3_t[1], MD5C0d, MD5S01);
579 MD5_STEP (MD5_Fo, c, d, a, b, w3_t[2], MD5C0e, MD5S02);
580 MD5_STEP (MD5_Fo, b, c, d, a, w3_t[3], MD5C0f, MD5S03);
582 MD5_STEP (MD5_Go, a, b, c, d, w0_t[1], MD5C10, MD5S10);
583 MD5_STEP (MD5_Go, d, a, b, c, w1_t[2], MD5C11, MD5S11);
584 MD5_STEP (MD5_Go, c, d, a, b, w2_t[3], MD5C12, MD5S12);
585 MD5_STEP (MD5_Go, b, c, d, a, w0_t[0], MD5C13, MD5S13);
586 MD5_STEP (MD5_Go, a, b, c, d, w1_t[1], MD5C14, MD5S10);
587 MD5_STEP (MD5_Go, d, a, b, c, w2_t[2], MD5C15, MD5S11);
588 MD5_STEP (MD5_Go, c, d, a, b, w3_t[3], MD5C16, MD5S12);
589 MD5_STEP (MD5_Go, b, c, d, a, w1_t[0], MD5C17, MD5S13);
590 MD5_STEP (MD5_Go, a, b, c, d, w2_t[1], MD5C18, MD5S10);
591 MD5_STEP (MD5_Go, d, a, b, c, w3_t[2], MD5C19, MD5S11);
592 MD5_STEP (MD5_Go, c, d, a, b, w0_t[3], MD5C1a, MD5S12);
593 MD5_STEP (MD5_Go, b, c, d, a, w2_t[0], MD5C1b, MD5S13);
594 MD5_STEP (MD5_Go, a, b, c, d, w3_t[1], MD5C1c, MD5S10);
595 MD5_STEP (MD5_Go, d, a, b, c, w0_t[2], MD5C1d, MD5S11);
596 MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
597 MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
599 MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
600 MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
601 MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
602 MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
603 MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
604 MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
605 MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
606 MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
607 MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
608 MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
609 MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
610 MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
611 MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
612 MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
613 MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
614 MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
616 MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
617 MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
618 MD5_STEP (MD5_I , c, d, a, b, w3_t[2], MD5C32, MD5S32);
619 MD5_STEP (MD5_I , b, c, d, a, w1_t[1], MD5C33, MD5S33);
620 MD5_STEP (MD5_I , a, b, c, d, w3_t[0], MD5C34, MD5S30);
621 MD5_STEP (MD5_I , d, a, b, c, w0_t[3], MD5C35, MD5S31);
622 MD5_STEP (MD5_I , c, d, a, b, w2_t[2], MD5C36, MD5S32);
623 MD5_STEP (MD5_I , b, c, d, a, w0_t[1], MD5C37, MD5S33);
624 MD5_STEP (MD5_I , a, b, c, d, w2_t[0], MD5C38, MD5S30);
625 MD5_STEP (MD5_I , d, a, b, c, w3_t[3], MD5C39, MD5S31);
626 MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32);
627 MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33);
628 MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30);
629 MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31);
630 MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
631 MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
638 w0_t[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0
639 | uint_to_hex_lower8 ((a >> 8) & 255) << 16;
640 w0_t[1] = uint_to_hex_lower8 ((a >> 16) & 255) << 0
641 | uint_to_hex_lower8 ((a >> 24) & 255) << 16;
642 w0_t[2] = uint_to_hex_lower8 ((b >> 0) & 255) << 0
643 | uint_to_hex_lower8 ((b >> 8) & 255) << 16;
644 w0_t[3] = uint_to_hex_lower8 ((b >> 16) & 255) << 0
645 | uint_to_hex_lower8 ((b >> 24) & 255) << 16;
646 w1_t[0] = uint_to_hex_lower8 ((c >> 0) & 255) << 0
647 | uint_to_hex_lower8 ((c >> 8) & 255) << 16;
648 w1_t[1] = uint_to_hex_lower8 ((c >> 16) & 255) << 0
649 | uint_to_hex_lower8 ((c >> 24) & 255) << 16;
650 w1_t[2] = uint_to_hex_lower8 ((d >> 0) & 255) << 0
651 | uint_to_hex_lower8 ((d >> 8) & 255) << 16;
652 w1_t[3] = uint_to_hex_lower8 ((d >> 16) & 255) << 0
653 | uint_to_hex_lower8 ((d >> 24) & 255) << 16;
655 // add the 4 byte challenge here
658 w2_t[1] = 0x00000080;
664 w3_t[2] = (32 + 4) * 8;
668 * md5 ($hash.$challenge)
676 MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
677 MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
678 MD5_STEP (MD5_Fo, c, d, a, b, w0_t[2], MD5C02, MD5S02);
679 MD5_STEP (MD5_Fo, b, c, d, a, w0_t[3], MD5C03, MD5S03);
680 MD5_STEP (MD5_Fo, a, b, c, d, w1_t[0], MD5C04, MD5S00);
681 MD5_STEP (MD5_Fo, d, a, b, c, w1_t[1], MD5C05, MD5S01);
682 MD5_STEP (MD5_Fo, c, d, a, b, w1_t[2], MD5C06, MD5S02);
683 MD5_STEP (MD5_Fo, b, c, d, a, w1_t[3], MD5C07, MD5S03);
684 MD5_STEP (MD5_Fo, a, b, c, d, w2_t[0], MD5C08, MD5S00);
685 MD5_STEP (MD5_Fo, d, a, b, c, w2_t[1], MD5C09, MD5S01);
686 MD5_STEP (MD5_Fo, c, d, a, b, w2_t[2], MD5C0a, MD5S02);
687 MD5_STEP (MD5_Fo, b, c, d, a, w2_t[3], MD5C0b, MD5S03);
688 MD5_STEP (MD5_Fo, a, b, c, d, w3_t[0], MD5C0c, MD5S00);
689 MD5_STEP (MD5_Fo, d, a, b, c, w3_t[1], MD5C0d, MD5S01);
690 MD5_STEP (MD5_Fo, c, d, a, b, w3_t[2], MD5C0e, MD5S02);
691 MD5_STEP (MD5_Fo, b, c, d, a, w3_t[3], MD5C0f, MD5S03);
693 MD5_STEP (MD5_Go, a, b, c, d, w0_t[1], MD5C10, MD5S10);
694 MD5_STEP (MD5_Go, d, a, b, c, w1_t[2], MD5C11, MD5S11);
695 MD5_STEP (MD5_Go, c, d, a, b, w2_t[3], MD5C12, MD5S12);
696 MD5_STEP (MD5_Go, b, c, d, a, w0_t[0], MD5C13, MD5S13);
697 MD5_STEP (MD5_Go, a, b, c, d, w1_t[1], MD5C14, MD5S10);
698 MD5_STEP (MD5_Go, d, a, b, c, w2_t[2], MD5C15, MD5S11);
699 MD5_STEP (MD5_Go, c, d, a, b, w3_t[3], MD5C16, MD5S12);
700 MD5_STEP (MD5_Go, b, c, d, a, w1_t[0], MD5C17, MD5S13);
701 MD5_STEP (MD5_Go, a, b, c, d, w2_t[1], MD5C18, MD5S10);
702 MD5_STEP (MD5_Go, d, a, b, c, w3_t[2], MD5C19, MD5S11);
703 MD5_STEP (MD5_Go, c, d, a, b, w0_t[3], MD5C1a, MD5S12);
704 MD5_STEP (MD5_Go, b, c, d, a, w2_t[0], MD5C1b, MD5S13);
705 MD5_STEP (MD5_Go, a, b, c, d, w3_t[1], MD5C1c, MD5S10);
706 MD5_STEP (MD5_Go, d, a, b, c, w0_t[2], MD5C1d, MD5S11);
707 MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
708 MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
710 MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
711 MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
712 MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
713 MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
714 MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
715 MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
716 MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
717 MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
718 MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
719 MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
720 MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
721 MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
722 MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
723 MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
724 MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
725 MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
727 MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
728 MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
729 MD5_STEP (MD5_I , c, d, a, b, w3_t[2], MD5C32, MD5S32);
730 MD5_STEP (MD5_I , b, c, d, a, w1_t[1], MD5C33, MD5S33);
731 MD5_STEP (MD5_I , a, b, c, d, w3_t[0], MD5C34, MD5S30);
732 MD5_STEP (MD5_I , d, a, b, c, w0_t[3], MD5C35, MD5S31);
733 MD5_STEP (MD5_I , c, d, a, b, w2_t[2], MD5C36, MD5S32);
734 MD5_STEP (MD5_I , b, c, d, a, w0_t[1], MD5C37, MD5S33);
735 MD5_STEP (MD5_I , a, b, c, d, w2_t[0], MD5C38, MD5S30);
736 MD5_STEP (MD5_I , d, a, b, c, w3_t[3], MD5C39, MD5S31);
737 MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32);
738 MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33);
739 MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30);
740 MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31);
741 MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
742 MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
743 COMPARE_S_SIMD (a, d, c, b);
747 __kernel void m11100_s08 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
751 __kernel void m11100_s16 (__global pw_t *pws, __global kernel_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)