/** * Author......: See docs/credits.txt * License.....: MIT */ //too much register pressure //#define NEW_SIMD_CODE #include "inc_vendor.cl" #include "inc_hash_constants.h" #include "inc_hash_functions.cl" #include "inc_types.cl" #include "inc_common.cl" #include "inc_simd.cl" #include "inc_hash_md5.cl" typedef struct { u8 S[256]; u32 wtf_its_faster; } RC4_KEY; DECLSPEC void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j) { u8 tmp; tmp = rc4_key->S[i]; rc4_key->S[i] = rc4_key->S[j]; rc4_key->S[j] = tmp; } DECLSPEC void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4]) { u32 v = 0x03020100; u32 a = 0x04040404; __local u32 *ptr = (__local u32 *) rc4_key->S; #ifdef _unroll #pragma unroll #endif for (u32 i = 0; i < 64; i++) { *ptr++ = v; v += a; } u32 j = 0; for (u32 i = 0; i < 16; i++) { u32 idx = i * 16; u32 v; v = data[0]; j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++; v = data[1]; j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++; v = data[2]; j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++; v = data[3]; j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++; j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++; } } DECLSPEC u8 rc4_next_16 (__local RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4]) { #ifdef _unroll #pragma unroll #endif for (u32 k = 0; k < 4; k++) { u32 xor4 = 0; u8 idx; i += 1; j += rc4_key->S[i]; swap (rc4_key, i, j); idx = rc4_key->S[i] + rc4_key->S[j]; xor4 |= rc4_key->S[idx] << 0; i += 1; j += rc4_key->S[i]; swap (rc4_key, i, j); idx = rc4_key->S[i] + rc4_key->S[j]; xor4 |= rc4_key->S[idx] << 8; i += 1; j += rc4_key->S[i]; swap (rc4_key, i, j); idx = rc4_key->S[i] + rc4_key->S[j]; xor4 |= rc4_key->S[idx] << 16; i += 1; j += rc4_key->S[i]; swap (rc4_key, i, j); idx = rc4_key->S[i] + rc4_key->S[j]; xor4 |= rc4_key->S[idx] << 24; out[k] = in[k] ^ xor4; } return j; } DECLSPEC void m09710m (__local RC4_KEY *rc4_keys, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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) { /** * modifier */ const u64 gid = get_global_id (0); const u64 lid = get_local_id (0); /** * shared */ __local RC4_KEY *rc4_key = &rc4_keys[lid]; /** * esalt */ const u32 version = oldoffice01_bufs[digests_offset].version; u32 encryptedVerifier[4]; encryptedVerifier[0] = oldoffice01_bufs[digests_offset].encryptedVerifier[0]; encryptedVerifier[1] = oldoffice01_bufs[digests_offset].encryptedVerifier[1]; encryptedVerifier[2] = oldoffice01_bufs[digests_offset].encryptedVerifier[2]; encryptedVerifier[3] = oldoffice01_bufs[digests_offset].encryptedVerifier[3]; /** * loop */ u32 w0l = w0[0]; for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) { const u32 w0r = ix_create_bft (bfs_buf, il_pos); const u32 w0lr = w0l | w0r; // first md5 to generate RC4 128 bit key u32 w0_t[4]; u32 w1_t[4]; u32 w2_t[4]; u32 w3_t[4]; w0_t[0] = w0lr; w0_t[1] = w0[1] & 0xff; w0_t[2] = 0x8000; w0_t[3] = 0; w1_t[0] = 0; w1_t[1] = 0; w1_t[2] = 0; w1_t[3] = 0; w2_t[0] = 0; w2_t[1] = 0; w2_t[2] = 0; w2_t[3] = 0; w3_t[0] = 0; w3_t[1] = 0; w3_t[2] = 9 * 8; w3_t[3] = 0; u32 digest[4]; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (w0_t, w1_t, w2_t, w3_t, digest); // now the RC4 part rc4_init_16 (rc4_key, digest); u32 out[4]; u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out); w0_t[0] = out[0]; w0_t[1] = out[1]; w0_t[2] = out[2]; w0_t[3] = out[3]; w1_t[0] = 0x80; w1_t[1] = 0; w1_t[2] = 0; w1_t[3] = 0; w2_t[0] = 0; w2_t[1] = 0; w2_t[2] = 0; w2_t[3] = 0; w3_t[0] = 0; w3_t[1] = 0; w3_t[2] = 16 * 8; w3_t[3] = 0; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (w0_t, w1_t, w2_t, w3_t, digest); rc4_next_16 (rc4_key, 16, j, digest, out); COMPARE_M_SIMD (out[0], out[1], out[2], out[3]); } } DECLSPEC void m09710s (__local RC4_KEY *rc4_keys, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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) { /** * modifier */ const u64 gid = get_global_id (0); const u64 lid = get_local_id (0); /** * shared */ __local RC4_KEY *rc4_key = &rc4_keys[lid]; /** * esalt */ const u32 version = oldoffice01_bufs[digests_offset].version; u32 encryptedVerifier[4]; encryptedVerifier[0] = oldoffice01_bufs[digests_offset].encryptedVerifier[0]; encryptedVerifier[1] = oldoffice01_bufs[digests_offset].encryptedVerifier[1]; encryptedVerifier[2] = oldoffice01_bufs[digests_offset].encryptedVerifier[2]; encryptedVerifier[3] = oldoffice01_bufs[digests_offset].encryptedVerifier[3]; /** * digest */ const u32 search[4] = { digests_buf[digests_offset].digest_buf[DGST_R0], digests_buf[digests_offset].digest_buf[DGST_R1], digests_buf[digests_offset].digest_buf[DGST_R2], digests_buf[digests_offset].digest_buf[DGST_R3] }; /** * loop */ u32 w0l = w0[0]; for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) { const u32 w0r = ix_create_bft (bfs_buf, il_pos); const u32 w0lr = w0l | w0r; // first md5 to generate RC4 128 bit key u32 w0_t[4]; u32 w1_t[4]; u32 w2_t[4]; u32 w3_t[4]; w0_t[0] = w0lr; w0_t[1] = w0[1] & 0xff; w0_t[2] = 0x8000; w0_t[3] = 0; w1_t[0] = 0; w1_t[1] = 0; w1_t[2] = 0; w1_t[3] = 0; w2_t[0] = 0; w2_t[1] = 0; w2_t[2] = 0; w2_t[3] = 0; w3_t[0] = 0; w3_t[1] = 0; w3_t[2] = 9 * 8; w3_t[3] = 0; u32 digest[4]; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (w0_t, w1_t, w2_t, w3_t, digest); // now the RC4 part rc4_init_16 (rc4_key, digest); u32 out[4]; u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out); w0_t[0] = out[0]; w0_t[1] = out[1]; w0_t[2] = out[2]; w0_t[3] = out[3]; w1_t[0] = 0x80; w1_t[1] = 0; w1_t[2] = 0; w1_t[3] = 0; w2_t[0] = 0; w2_t[1] = 0; w2_t[2] = 0; w2_t[3] = 0; w3_t[0] = 0; w3_t[1] = 0; w3_t[2] = 16 * 8; w3_t[3] = 0; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (w0_t, w1_t, w2_t, w3_t, digest); rc4_next_16 (rc4_key, 16, j, digest, out); COMPARE_S_SIMD (out[0], out[1], out[2], out[3]); } } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { /** * base */ const u64 gid = get_global_id (0); if (gid >= gid_max) return; u32 w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32 w1[4]; w1[0] = 0; w1[1] = 0; w1[2] = 0; w1[3] = 0; u32 w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32 w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ __local RC4_KEY rc4_keys[64]; m09710m (rc4_keys, 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, oldoffice01_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_m08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_m16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_s04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { /** * base */ const u64 gid = get_global_id (0); if (gid >= gid_max) return; u32 w0[4]; w0[0] = pws[gid].i[ 0]; w0[1] = pws[gid].i[ 1]; w0[2] = pws[gid].i[ 2]; w0[3] = pws[gid].i[ 3]; u32 w1[4]; w1[0] = 0; w1[1] = 0; w1[2] = 0; w1[3] = 0; u32 w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32 w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 pw_len = pws[gid].pw_len; /** * main */ __local RC4_KEY rc4_keys[64]; m09710s (rc4_keys, 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, oldoffice01_bufs, d_return_buf, d_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset); } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_s08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { } __kernel void __attribute__((reqd_work_group_size(64, 1, 1))) m09710_s16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global oldoffice01_t *oldoffice01_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 u64 gid_max) { }