2 * Author......: Jens Steube <jens.steube@gmail.com>
10 #include "inc_vendor.cl"
11 #include "inc_hash_constants.h"
12 #include "inc_hash_functions.cl"
13 #include "inc_types.cl"
14 #include "inc_common.cl"
17 #include "inc_simd.cl"
19 __constant u64 keccakf_rndc[24] =
21 0x0000000000000001, 0x0000000000008082, 0x800000000000808a,
22 0x8000000080008000, 0x000000000000808b, 0x0000000080000001,
23 0x8000000080008081, 0x8000000000008009, 0x000000000000008a,
24 0x0000000000000088, 0x0000000080008009, 0x000000008000000a,
25 0x000000008000808b, 0x800000000000008b, 0x8000000000008089,
26 0x8000000000008003, 0x8000000000008002, 0x8000000000000080,
27 0x000000000000800a, 0x800000008000000a, 0x8000000080008081,
28 0x8000000000008080, 0x0000000080000001, 0x8000000080008008
32 #define KECCAK_ROUNDS 24
35 #define Theta1(s) (st[0 + s] ^ st[5 + s] ^ st[10 + s] ^ st[15 + s] ^ st[20 + s])
48 u32 j = keccakf_piln[s]; \
49 u32 k = keccakf_rotc[s]; \
51 st[j] = rotl64 (t, k); \
62 st[0 + s] ^= ~bc1 & bc2; \
63 st[1 + s] ^= ~bc2 & bc3; \
64 st[2 + s] ^= ~bc3 & bc4; \
65 st[3 + s] ^= ~bc4 & bc0; \
66 st[4 + s] ^= ~bc0 & bc1; \
69 __kernel void m05000_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)
75 const u32 lid = get_local_id (0);
81 const u32 gid = get_global_id (0);
83 if (gid >= gid_max) return;
88 pw_buf0[0] = pws[gid].i[0];
89 pw_buf0[1] = pws[gid].i[1];
90 pw_buf0[2] = pws[gid].i[2];
91 pw_buf0[3] = pws[gid].i[3];
92 pw_buf1[0] = pws[gid].i[4];
93 pw_buf1[1] = pws[gid].i[5];
94 pw_buf1[2] = pws[gid].i[6];
95 pw_buf1[3] = pws[gid].i[7];
97 const u32 pw_len = pws[gid].pw_len;
103 const u8 keccakf_rotc[24] =
105 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 2, 14,
106 27, 41, 56, 8, 25, 43, 62, 18, 39, 61, 20, 44
109 const u8 keccakf_piln[24] =
111 10, 7, 11, 17, 18, 3, 5, 16, 8, 21, 24, 4,
112 15, 23, 19, 13, 12, 2, 20, 14, 22, 9, 6, 1
116 * 0x80 keccak, very special
119 const u32 mdlen = salt_bufs[salt_pos].keccak_mdlen;
121 const u32 rsiz = 200 - (2 * mdlen);
123 const u32 add80w = (rsiz - 1) / 8;
129 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
136 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
138 append_0x01_2x4_VV (w0, w1, out_len);
146 st[ 0] = hl32_to_64 (w0[1], w0[0]);
147 st[ 1] = hl32_to_64 (w0[3], w0[2]);
148 st[ 2] = hl32_to_64 (w1[1], w1[0]);
149 st[ 3] = hl32_to_64 (w1[3], w1[2]);
172 st[add80w] |= 0x8000000000000000;
176 for (round = 0; round < KECCAK_ROUNDS; round++)
180 u64x bc0 = Theta1 (0);
181 u64x bc1 = Theta1 (1);
182 u64x bc2 = Theta1 (2);
183 u64x bc3 = Theta1 (3);
184 u64x bc4 = Theta1 (4);
188 t = bc4 ^ rotl64 (bc1, 1); Theta2 (0);
189 t = bc0 ^ rotl64 (bc2, 1); Theta2 (1);
190 t = bc1 ^ rotl64 (bc3, 1); Theta2 (2);
191 t = bc2 ^ rotl64 (bc4, 1); Theta2 (3);
192 t = bc3 ^ rotl64 (bc0, 1); Theta2 (4);
233 st[0] ^= keccakf_rndc[round];
236 const u32x r0 = l32_from_64 (st[1]);
237 const u32x r1 = h32_from_64 (st[1]);
238 const u32x r2 = l32_from_64 (st[2]);
239 const u32x r3 = h32_from_64 (st[2]);
241 COMPARE_M_SIMD (r0, r1, r2, r3);
245 __kernel void m05000_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)
249 __kernel void m05000_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)
253 __kernel void m05000_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)
259 const u32 lid = get_local_id (0);
265 const u32 gid = get_global_id (0);
267 if (gid >= gid_max) return;
272 pw_buf0[0] = pws[gid].i[0];
273 pw_buf0[1] = pws[gid].i[1];
274 pw_buf0[2] = pws[gid].i[2];
275 pw_buf0[3] = pws[gid].i[3];
276 pw_buf1[0] = pws[gid].i[4];
277 pw_buf1[1] = pws[gid].i[5];
278 pw_buf1[2] = pws[gid].i[6];
279 pw_buf1[3] = pws[gid].i[7];
281 const u32 pw_len = pws[gid].pw_len;
287 const u8 keccakf_rotc[24] =
289 1, 3, 6, 10, 15, 21, 28, 36, 45, 55, 2, 14,
290 27, 41, 56, 8, 25, 43, 62, 18, 39, 61, 20, 44
293 const u8 keccakf_piln[24] =
295 10, 7, 11, 17, 18, 3, 5, 16, 8, 21, 24, 4,
296 15, 23, 19, 13, 12, 2, 20, 14, 22, 9, 6, 1
300 * 0x80 keccak, very special
303 const u32 mdlen = salt_bufs[salt_pos].keccak_mdlen;
305 const u32 rsiz = 200 - (2 * mdlen);
307 const u32 add80w = (rsiz - 1) / 8;
313 const u32 search[4] =
315 digests_buf[digests_offset].digest_buf[DGST_R0],
316 digests_buf[digests_offset].digest_buf[DGST_R1],
317 digests_buf[digests_offset].digest_buf[DGST_R2],
318 digests_buf[digests_offset].digest_buf[DGST_R3]
325 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
332 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
334 append_0x01_2x4_VV (w0, w1, out_len);
342 st[ 0] = hl32_to_64 (w0[1], w0[0]);
343 st[ 1] = hl32_to_64 (w0[3], w0[2]);
344 st[ 2] = hl32_to_64 (w1[1], w1[0]);
345 st[ 3] = hl32_to_64 (w1[3], w1[2]);
368 st[add80w] |= 0x8000000000000000;
372 for (round = 0; round < KECCAK_ROUNDS; round++)
376 u64x bc0 = Theta1 (0);
377 u64x bc1 = Theta1 (1);
378 u64x bc2 = Theta1 (2);
379 u64x bc3 = Theta1 (3);
380 u64x bc4 = Theta1 (4);
384 t = bc4 ^ rotl64 (bc1, 1); Theta2 (0);
385 t = bc0 ^ rotl64 (bc2, 1); Theta2 (1);
386 t = bc1 ^ rotl64 (bc3, 1); Theta2 (2);
387 t = bc2 ^ rotl64 (bc4, 1); Theta2 (3);
388 t = bc3 ^ rotl64 (bc0, 1); Theta2 (4);
429 st[0] ^= keccakf_rndc[round];
432 const u32x r0 = l32_from_64 (st[1]);
433 const u32x r1 = h32_from_64 (st[1]);
434 const u32x r2 = l32_from_64 (st[2]);
435 const u32x r3 = h32_from_64 (st[2]);
437 COMPARE_S_SIMD (r0, r1, r2, r3);
441 __kernel void m05000_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)
445 __kernel void m05000_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)