2 * Author......: Jens Steube <jens.steube@gmail.com>
8 //incompatible data-dependant code
9 //#define NEW_SIMD_CODE
11 #include "inc_vendor.cl"
12 #include "inc_hash_constants.h"
13 #include "inc_hash_functions.cl"
14 #include "inc_types.cl"
15 #include "inc_common.cl"
18 #include "inc_simd.cl"
20 #define GETSHIFTEDINT(a,n) amd_bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], (n))
22 #define SETSHIFTEDINT(a,n,v) \
24 const u32 s = ((n) & 3) * 8; \
25 const u64 x = (u64) (v) << s; \
26 (a)[((n)/4)+0] |= x; \
27 (a)[((n)/4)+1] = x >> 32; \
30 __constant u32 theMagicArray[64] =
32 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
33 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
34 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
35 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
36 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
37 0, 0, 0, 0, 0, 0, 0, 0,
38 0, 0, 0, 0, 0, 0, 0, 0,
39 0, 0, 0, 0, 0, 0, 0, 0
42 void swap_buffer (u32 final[16])
44 final[ 0] = swap32 (final[ 0]);
45 final[ 1] = swap32 (final[ 1]);
46 final[ 2] = swap32 (final[ 2]);
47 final[ 3] = swap32 (final[ 3]);
48 final[ 4] = swap32 (final[ 4]);
49 final[ 5] = swap32 (final[ 5]);
50 final[ 6] = swap32 (final[ 6]);
51 final[ 7] = swap32 (final[ 7]);
52 final[ 8] = swap32 (final[ 8]);
53 final[ 9] = swap32 (final[ 9]);
54 final[10] = swap32 (final[10]);
55 final[11] = swap32 (final[11]);
56 final[12] = swap32 (final[12]);
57 final[13] = swap32 (final[13]);
58 final[14] = swap32 (final[14]);
59 final[15] = swap32 (final[15]);
62 void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
90 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
91 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
92 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
93 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
94 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
95 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
96 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
97 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
98 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
99 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
100 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
101 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
102 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
103 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
104 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
105 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
106 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
107 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
108 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
109 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
114 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
115 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
116 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
117 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
118 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
119 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
120 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
121 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
122 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
123 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
124 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
125 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
126 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
127 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
128 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
129 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
130 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
131 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
132 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
133 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
138 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
139 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
140 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
141 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
142 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
143 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
144 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
145 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
146 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
147 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
148 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
149 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
150 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
151 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
152 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
153 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
154 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
155 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
156 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
157 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
162 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
163 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
164 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
165 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
166 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
167 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
168 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
169 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
170 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
171 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
172 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
173 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
174 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
175 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
176 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
177 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
178 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
179 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
180 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
181 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
190 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
196 const u32 lid = get_local_id (0);
202 const u32 gid = get_global_id (0);
204 if (gid >= gid_max) return;
209 pw_buf0[0] = pws[gid].i[0];
210 pw_buf0[1] = pws[gid].i[1];
211 pw_buf0[2] = pws[gid].i[2];
212 pw_buf0[3] = pws[gid].i[3];
213 pw_buf1[0] = pws[gid].i[4];
214 pw_buf1[1] = pws[gid].i[5];
215 pw_buf1[2] = pws[gid].i[6];
216 pw_buf1[3] = pws[gid].i[7];
218 const u32 pw_len = pws[gid].pw_len;
226 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
227 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
228 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
229 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
230 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
231 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
232 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
233 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
235 const u32 salt_len = salt_bufs[salt_pos].salt_len;
241 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
248 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
276 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
278 const u32x pw_salt_len = out_len + salt_len;
286 final[ 0] = swap32 (w0[0] | s0[0]);
287 final[ 1] = swap32 (w0[1] | s0[1]);
288 final[ 2] = swap32 (w0[2] | s0[2]);
289 final[ 3] = swap32 (w0[3] | s0[3]);
290 final[ 4] = swap32 (w1[0] | s1[0]);
291 final[ 5] = swap32 (w1[1] | s1[1]);
292 final[ 6] = swap32 (w1[2] | s1[2]);
293 final[ 7] = swap32 (w1[3] | s1[3]);
294 final[ 8] = swap32 (w2[0] | s2[0]);
295 final[ 9] = swap32 (w2[1] | s2[1]);
296 final[10] = swap32 (w2[2] | s2[2]);
297 final[11] = swap32 (w2[3] | s2[3]);
298 final[12] = swap32 (w3[0] | s3[0]);
299 final[13] = swap32 (w3[1] | s3[1]);
301 final[15] = pw_salt_len * 8;
311 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
313 // prepare magic array range
315 u32 lengthMagicArray = 0x20;
316 u32 offsetMagicArray = 0;
318 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
319 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
320 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
321 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
322 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
323 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
324 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
325 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
326 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
327 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
328 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
329 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
330 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
331 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
332 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
333 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
334 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
335 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
336 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
337 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
350 for (int i = 0; i < 32; i++) final[i] = 0;
361 u32 final_len = out_len;
367 for (i = 0; i < lengthMagicArray - 4; i += 4)
369 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
371 SETSHIFTEDINT (final, final_len + i, tmp);
374 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
376 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
378 SETSHIFTEDINT (final, final_len + i, tmp);
380 final_len += lengthMagicArray;
384 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
386 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
388 SETSHIFTEDINT (final, final_len + i, tmp);
391 final_len += salt_len;
398 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
400 swap_buffer (&final[off]);
402 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
405 swap_buffer (&final[off]);
408 final[off + 15] = final_len * 8;
410 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
412 COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
416 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
420 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
424 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
430 const u32 lid = get_local_id (0);
436 const u32 gid = get_global_id (0);
438 if (gid >= gid_max) return;
443 pw_buf0[0] = pws[gid].i[0];
444 pw_buf0[1] = pws[gid].i[1];
445 pw_buf0[2] = pws[gid].i[2];
446 pw_buf0[3] = pws[gid].i[3];
447 pw_buf1[0] = pws[gid].i[4];
448 pw_buf1[1] = pws[gid].i[5];
449 pw_buf1[2] = pws[gid].i[6];
450 pw_buf1[3] = pws[gid].i[7];
452 const u32 pw_len = pws[gid].pw_len;
460 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
461 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
462 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
463 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
464 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
465 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
466 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
467 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
469 const u32 salt_len = salt_bufs[salt_pos].salt_len;
475 const u32 search[4] =
477 digests_buf[digests_offset].digest_buf[DGST_R0],
478 digests_buf[digests_offset].digest_buf[DGST_R1],
479 digests_buf[digests_offset].digest_buf[DGST_R2],
480 digests_buf[digests_offset].digest_buf[DGST_R3]
487 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
494 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
522 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
524 const u32x pw_salt_len = out_len + salt_len;
532 final[ 0] = swap32 (w0[0] | s0[0]);
533 final[ 1] = swap32 (w0[1] | s0[1]);
534 final[ 2] = swap32 (w0[2] | s0[2]);
535 final[ 3] = swap32 (w0[3] | s0[3]);
536 final[ 4] = swap32 (w1[0] | s1[0]);
537 final[ 5] = swap32 (w1[1] | s1[1]);
538 final[ 6] = swap32 (w1[2] | s1[2]);
539 final[ 7] = swap32 (w1[3] | s1[3]);
540 final[ 8] = swap32 (w2[0] | s2[0]);
541 final[ 9] = swap32 (w2[1] | s2[1]);
542 final[10] = swap32 (w2[2] | s2[2]);
543 final[11] = swap32 (w2[3] | s2[3]);
544 final[12] = swap32 (w3[0] | s3[0]);
545 final[13] = swap32 (w3[1] | s3[1]);
547 final[15] = pw_salt_len * 8;
557 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
559 // prepare magic array range
561 u32 lengthMagicArray = 0x20;
562 u32 offsetMagicArray = 0;
564 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
565 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
566 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
567 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
568 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
569 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
570 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
571 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
572 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
573 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
574 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
575 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
576 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
577 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
578 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
579 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
580 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
581 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
582 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
583 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
596 for (int i = 0; i < 32; i++) final[i] = 0;
607 u32 final_len = out_len;
613 for (i = 0; i < lengthMagicArray - 4; i += 4)
615 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
617 SETSHIFTEDINT (final, final_len + i, tmp);
620 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
622 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
624 SETSHIFTEDINT (final, final_len + i, tmp);
626 final_len += lengthMagicArray;
630 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
632 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
634 SETSHIFTEDINT (final, final_len + i, tmp);
637 final_len += salt_len;
644 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
646 swap_buffer (&final[off]);
648 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
651 swap_buffer (&final[off]);
654 final[off + 15] = final_len * 8;
656 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
658 COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
662 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
666 __kernel void m07800_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)