/** * Author......: Jens Steube * License.....: MIT */ #define _MD5_ #include "include/constants.h" #include "include/kernel_vendor.h" #define DGST_R0 0 #define DGST_R1 1 #define DGST_R2 2 #define DGST_R3 3 #include "include/kernel_functions.c" #include "OpenCL/types_ocl.c" #include "OpenCL/common.c" #define COMPARE_S "OpenCL/check_single_comp4.c" #define COMPARE_M "OpenCL/check_multi_comp4.c" #define md5apr1_magic0 0x72706124 #define md5apr1_magic1 0x00002431 static void md5_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4]) { u32 a = digest[0]; u32 b = digest[1]; u32 c = digest[2]; u32 d = digest[3]; u32 w0_t = w0[0]; u32 w1_t = w0[1]; u32 w2_t = w0[2]; u32 w3_t = w0[3]; u32 w4_t = w1[0]; u32 w5_t = w1[1]; u32 w6_t = w1[2]; u32 w7_t = w1[3]; u32 w8_t = w2[0]; u32 w9_t = w2[1]; u32 wa_t = w2[2]; u32 wb_t = w2[3]; u32 wc_t = w3[0]; u32 wd_t = w3[1]; u32 we_t = w3[2]; u32 wf_t = 0; u32 tmp2; MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00); MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01); MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02); MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03); MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00); MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01); MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02); MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03); MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00); MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01); MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02); MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03); MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00); MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01); MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02); MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03); MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10); MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11); MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12); MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13); MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10); MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11); MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12); MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13); MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10); MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11); MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12); MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13); MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10); MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11); MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12); MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13); MD5_STEP (MD5_H1, a, b, c, d, w5_t, MD5C20, MD5S20); MD5_STEP (MD5_H2, d, a, b, c, w8_t, MD5C21, MD5S21); MD5_STEP (MD5_H1, c, d, a, b, wb_t, MD5C22, MD5S22); MD5_STEP (MD5_H2, b, c, d, a, we_t, MD5C23, MD5S23); MD5_STEP (MD5_H1, a, b, c, d, w1_t, MD5C24, MD5S20); MD5_STEP (MD5_H2, d, a, b, c, w4_t, MD5C25, MD5S21); MD5_STEP (MD5_H1, c, d, a, b, w7_t, MD5C26, MD5S22); MD5_STEP (MD5_H2, b, c, d, a, wa_t, MD5C27, MD5S23); MD5_STEP (MD5_H1, a, b, c, d, wd_t, MD5C28, MD5S20); MD5_STEP (MD5_H2, d, a, b, c, w0_t, MD5C29, MD5S21); MD5_STEP (MD5_H1, c, d, a, b, w3_t, MD5C2a, MD5S22); MD5_STEP (MD5_H2, b, c, d, a, w6_t, MD5C2b, MD5S23); MD5_STEP (MD5_H1, a, b, c, d, w9_t, MD5C2c, MD5S20); MD5_STEP (MD5_H2, d, a, b, c, wc_t, MD5C2d, MD5S21); MD5_STEP (MD5_H1, c, d, a, b, wf_t, MD5C2e, MD5S22); MD5_STEP (MD5_H2, b, c, d, a, w2_t, MD5C2f, MD5S23); MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30); MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31); MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33); MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30); MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31); MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33); MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30); MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31); MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32); MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33); MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30); MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31); MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32); MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33); digest[0] += a; digest[1] += b; digest[2] += c; digest[3] += d; } static void memcat16 (u32 block0[4], u32 block1[4], u32 block2[4], u32 block3[4], const u32 block_len, const u32 append[4]) { u32 tmp0; u32 tmp1; u32 tmp2; u32 tmp3; u32 tmp4; #ifdef IS_AMD const int offset_minus_4 = 4 - (block_len & 3); tmp0 = amd_bytealign (append[0], 0, offset_minus_4); tmp1 = amd_bytealign (append[1], append[0], offset_minus_4); tmp2 = amd_bytealign (append[2], append[1], offset_minus_4); tmp3 = amd_bytealign (append[3], append[2], offset_minus_4); tmp4 = amd_bytealign ( 0, append[3], offset_minus_4); const u32 mod = block_len & 3; if (mod == 0) { tmp0 = tmp1; tmp1 = tmp2; tmp2 = tmp3; tmp3 = tmp4; tmp4 = 0; } #endif #ifdef IS_NV const int offset_minus_4 = 4 - (block_len & 3); const int selector = (0x76543210 >> (offset_minus_4 * 4)) & 0xffff; tmp0 = __byte_perm ( 0, append[0], selector); tmp1 = __byte_perm (append[0], append[1], selector); tmp2 = __byte_perm (append[1], append[2], selector); tmp3 = __byte_perm (append[2], append[3], selector); tmp4 = __byte_perm (append[3], 0, selector); #endif const u32 div = block_len / 4; switch (div) { case 0: block0[0] |= tmp0; block0[1] = tmp1; block0[2] = tmp2; block0[3] = tmp3; block1[0] = tmp4; break; case 1: block0[1] |= tmp0; block0[2] = tmp1; block0[3] = tmp2; block1[0] = tmp3; block1[1] = tmp4; break; case 2: block0[2] |= tmp0; block0[3] = tmp1; block1[0] = tmp2; block1[1] = tmp3; block1[2] = tmp4; break; case 3: block0[3] |= tmp0; block1[0] = tmp1; block1[1] = tmp2; block1[2] = tmp3; block1[3] = tmp4; break; case 4: block1[0] |= tmp0; block1[1] = tmp1; block1[2] = tmp2; block1[3] = tmp3; block2[0] = tmp4; break; case 5: block1[1] |= tmp0; block1[2] = tmp1; block1[3] = tmp2; block2[0] = tmp3; block2[1] = tmp4; break; case 6: block1[2] |= tmp0; block1[3] = tmp1; block2[0] = tmp2; block2[1] = tmp3; block2[2] = tmp4; break; case 7: block1[3] |= tmp0; block2[0] = tmp1; block2[1] = tmp2; block2[2] = tmp3; block2[3] = tmp4; break; case 8: block2[0] |= tmp0; block2[1] = tmp1; block2[2] = tmp2; block2[3] = tmp3; block3[0] = tmp4; break; case 9: block2[1] |= tmp0; block2[2] = tmp1; block2[3] = tmp2; block3[0] = tmp3; block3[1] = tmp4; break; } } static void memcat16_x80 (u32 block0[4], u32 block1[4], u32 block2[4], u32 block3[4], const u32 block_len, const u32 append[4]) { u32 tmp0; u32 tmp1; u32 tmp2; u32 tmp3; u32 tmp4; #ifdef IS_AMD const int offset_minus_4 = 4 - (block_len & 3); tmp0 = amd_bytealign (append[0], 0, offset_minus_4); tmp1 = amd_bytealign (append[1], append[0], offset_minus_4); tmp2 = amd_bytealign (append[2], append[1], offset_minus_4); tmp3 = amd_bytealign (append[3], append[2], offset_minus_4); tmp4 = amd_bytealign ( 0x80, append[3], offset_minus_4); const u32 mod = block_len & 3; if (mod == 0) { tmp0 = tmp1; tmp1 = tmp2; tmp2 = tmp3; tmp3 = tmp4; tmp4 = 0x80; } #endif #ifdef IS_NV const int offset_minus_4 = 4 - (block_len & 3); const int selector = (0x76543210 >> (offset_minus_4 * 4)) & 0xffff; tmp0 = __byte_perm ( 0, append[0], selector); tmp1 = __byte_perm (append[0], append[1], selector); tmp2 = __byte_perm (append[1], append[2], selector); tmp3 = __byte_perm (append[2], append[3], selector); tmp4 = __byte_perm (append[3], 0x80, selector); #endif const u32 div = block_len / 4; switch (div) { case 0: block0[0] |= tmp0; block0[1] = tmp1; block0[2] = tmp2; block0[3] = tmp3; block1[0] = tmp4; break; case 1: block0[1] |= tmp0; block0[2] = tmp1; block0[3] = tmp2; block1[0] = tmp3; block1[1] = tmp4; break; case 2: block0[2] |= tmp0; block0[3] = tmp1; block1[0] = tmp2; block1[1] = tmp3; block1[2] = tmp4; break; case 3: block0[3] |= tmp0; block1[0] = tmp1; block1[1] = tmp2; block1[2] = tmp3; block1[3] = tmp4; break; case 4: block1[0] |= tmp0; block1[1] = tmp1; block1[2] = tmp2; block1[3] = tmp3; block2[0] = tmp4; break; case 5: block1[1] |= tmp0; block1[2] = tmp1; block1[3] = tmp2; block2[0] = tmp3; block2[1] = tmp4; break; case 6: block1[2] |= tmp0; block1[3] = tmp1; block2[0] = tmp2; block2[1] = tmp3; block2[2] = tmp4; break; case 7: block1[3] |= tmp0; block2[0] = tmp1; block2[1] = tmp2; block2[2] = tmp3; block2[3] = tmp4; break; case 8: block2[0] |= tmp0; block2[1] = tmp1; block2[2] = tmp2; block2[3] = tmp3; block3[0] = tmp4; break; case 9: block2[1] |= tmp0; block2[2] = tmp1; block2[3] = tmp2; block3[0] = tmp3; block3[1] = tmp4; break; } } static void memcat8 (u32 block0[4], u32 block1[4], u32 block2[4], u32 block3[4], const u32 block_len, const u32 append[2]) { u32 tmp0; u32 tmp1; u32 tmp2; #ifdef IS_AMD const int offset_minus_4 = 4 - (block_len & 3); tmp0 = amd_bytealign (append[0], 0, offset_minus_4); tmp1 = amd_bytealign (append[1], append[0], offset_minus_4); tmp2 = amd_bytealign ( 0, append[1], offset_minus_4); const u32 mod = block_len & 3; if (mod == 0) { tmp0 = tmp1; tmp1 = tmp2; tmp2 = 0; } #endif #ifdef IS_NV const int offset_minus_4 = 4 - (block_len & 3); const int selector = (0x76543210 >> (offset_minus_4 * 4)) & 0xffff; tmp0 = __byte_perm ( 0, append[0], selector); tmp1 = __byte_perm (append[0], append[1], selector); tmp2 = __byte_perm (append[1], 0, selector); #endif const u32 div = block_len / 4; switch (div) { case 0: block0[0] |= tmp0; block0[1] = tmp1; block0[2] = tmp2; break; case 1: block0[1] |= tmp0; block0[2] = tmp1; block0[3] = tmp2; break; case 2: block0[2] |= tmp0; block0[3] = tmp1; block1[0] = tmp2; break; case 3: block0[3] |= tmp0; block1[0] = tmp1; block1[1] = tmp2; break; case 4: block1[0] |= tmp0; block1[1] = tmp1; block1[2] = tmp2; break; case 5: block1[1] |= tmp0; block1[2] = tmp1; block1[3] = tmp2; break; case 6: block1[2] |= tmp0; block1[3] = tmp1; block2[0] = tmp2; break; case 7: block1[3] |= tmp0; block2[0] = tmp1; block2[1] = tmp2; break; case 8: block2[0] |= tmp0; block2[1] = tmp1; block2[2] = tmp2; break; case 9: block2[1] |= tmp0; block2[2] = tmp1; block2[3] = tmp2; break; case 10: block2[2] |= tmp0; block2[3] = tmp1; block3[0] = tmp2; break; case 11: block2[3] |= tmp0; block3[0] = tmp1; block3[1] = tmp2; break; } } static void append_sign (u32 block0[4], u32 block1[4], const u32 block_len) { switch (block_len) { case 0: block0[0] = md5apr1_magic0; block0[1] = md5apr1_magic1; break; case 1: block0[0] = block0[0] | md5apr1_magic0 << 8; block0[1] = md5apr1_magic0 >> 24 | md5apr1_magic1 << 8; block0[2] = md5apr1_magic1 >> 24; break; case 2: block0[0] = block0[0] | md5apr1_magic0 << 16; block0[1] = md5apr1_magic0 >> 16 | md5apr1_magic1 << 16; block0[2] = md5apr1_magic1 >> 16; break; case 3: block0[0] = block0[0] | md5apr1_magic0 << 24; block0[1] = md5apr1_magic0 >> 8 | md5apr1_magic1 << 24; block0[2] = md5apr1_magic1 >> 8; break; case 4: block0[1] = md5apr1_magic0; block0[2] = md5apr1_magic1; break; case 5: block0[1] = block0[1] | md5apr1_magic0 << 8; block0[2] = md5apr1_magic0 >> 24 | md5apr1_magic1 << 8; block0[3] = md5apr1_magic1 >> 24; break; case 6: block0[1] = block0[1] | md5apr1_magic0 << 16; block0[2] = md5apr1_magic0 >> 16 | md5apr1_magic1 << 16; block0[3] = md5apr1_magic1 >> 16; break; case 7: block0[1] = block0[1] | md5apr1_magic0 << 24; block0[2] = md5apr1_magic0 >> 8 | md5apr1_magic1 << 24; block0[3] = md5apr1_magic1 >> 8; break; case 8: block0[2] = md5apr1_magic0; block0[3] = md5apr1_magic1; break; case 9: block0[2] = block0[2] | md5apr1_magic0 << 8; block0[3] = md5apr1_magic0 >> 24 | md5apr1_magic1 << 8; block1[0] = md5apr1_magic1 >> 24; break; case 10: block0[2] = block0[2] | md5apr1_magic0 << 16; block0[3] = md5apr1_magic0 >> 16 | md5apr1_magic1 << 16; block1[0] = md5apr1_magic1 >> 16; break; case 11: block0[2] = block0[2] | md5apr1_magic0 << 24; block0[3] = md5apr1_magic0 >> 8 | md5apr1_magic1 << 24; block1[0] = md5apr1_magic1 >> 8; break; case 12: block0[3] = md5apr1_magic0; block1[0] = md5apr1_magic1; break; case 13: block0[3] = block0[3] | md5apr1_magic0 << 8; block1[0] = md5apr1_magic0 >> 24 | md5apr1_magic1 << 8; block1[1] = md5apr1_magic1 >> 24; break; case 14: block0[3] = block0[3] | md5apr1_magic0 << 16; block1[0] = md5apr1_magic0 >> 16 | md5apr1_magic1 << 16; block1[1] = md5apr1_magic1 >> 16; break; case 15: block0[3] = block0[3] | md5apr1_magic0 << 24; block1[0] = md5apr1_magic0 >> 8 | md5apr1_magic1 << 24; block1[1] = md5apr1_magic1 >> 8; break; } } static void append_1st (u32 block0[4], u32 block1[4], u32 block2[4], u32 block3[4], const u32 block_len, const u32 append) { switch (block_len) { case 0: block0[0] = append; break; case 1: block0[0] = block0[0] | append << 8; break; case 2: block0[0] = block0[0] | append << 16; break; case 3: block0[0] = block0[0] | append << 24; break; case 4: block0[1] = append; break; case 5: block0[1] = block0[1] | append << 8; break; case 6: block0[1] = block0[1] | append << 16; break; case 7: block0[1] = block0[1] | append << 24; break; case 8: block0[2] = append; break; case 9: block0[2] = block0[2] | append << 8; break; case 10: block0[2] = block0[2] | append << 16; break; case 11: block0[2] = block0[2] | append << 24; break; case 12: block0[3] = append; break; case 13: block0[3] = block0[3] | append << 8; break; case 14: block0[3] = block0[3] | append << 16; break; case 15: block0[3] = block0[3] | append << 24; break; case 16: block1[0] = append; break; case 17: block1[0] = block1[0] | append << 8; break; case 18: block1[0] = block1[0] | append << 16; break; case 19: block1[0] = block1[0] | append << 24; break; case 20: block1[1] = append; break; case 21: block1[1] = block1[1] | append << 8; break; case 22: block1[1] = block1[1] | append << 16; break; case 23: block1[1] = block1[1] | append << 24; break; case 24: block1[2] = append; break; case 25: block1[2] = block1[2] | append << 8; break; case 26: block1[2] = block1[2] | append << 16; break; case 27: block1[2] = block1[2] | append << 24; break; case 28: block1[3] = append; break; case 29: block1[3] = block1[3] | append << 8; break; case 30: block1[3] = block1[3] | append << 16; break; case 31: block1[3] = block1[3] | append << 24; break; case 32: block2[0] = append; break; case 33: block2[0] = block2[0] | append << 8; break; case 34: block2[0] = block2[0] | append << 16; break; case 35: block2[0] = block2[0] | append << 24; break; case 36: block2[1] = append; break; case 37: block2[1] = block2[1] | append << 8; break; case 38: block2[1] = block2[1] | append << 16; break; case 39: block2[1] = block2[1] | append << 24; break; case 40: block2[2] = append; break; case 41: block2[2] = block2[2] | append << 8; break; case 42: block2[2] = block2[2] | append << 16; break; case 43: block2[2] = block2[2] | append << 24; break; case 44: block2[3] = append; break; case 45: block2[3] = block2[3] | append << 8; break; case 46: block2[3] = block2[3] | append << 16; break; case 47: block2[3] = block2[3] | append << 24; break; case 48: block3[0] = append; break; case 49: block3[0] = block3[0] | append << 8; break; case 50: block3[0] = block3[0] | append << 16; break; case 51: block3[0] = block3[0] | append << 24; break; case 52: block3[1] = append; break; case 53: block3[1] = block3[1] | append << 8; break; case 54: block3[1] = block3[1] | append << 16; break; case 55: block3[1] = block3[1] | append << 24; break; case 56: block3[2] = append; break; } } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01600_init (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global md5crypt_tmp_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 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]; const u32 pw_len = pws[gid].pw_len; /** * salt */ u32 salt_buf[2]; salt_buf[0] = salt_bufs[salt_pos].salt_buf[0]; salt_buf[1] = salt_bufs[salt_pos].salt_buf[1]; const u32 salt_len = salt_bufs[salt_pos].salt_len; /** * init */ //memcat16 (block0, block1, block2, block3, block_len, w0); //block_len += pw_len; u32 block_len = pw_len; u32 block0[4]; block0[0] = w0[0]; block0[1] = w0[1]; block0[2] = w0[2]; block0[3] = w0[3]; u32 block1[4]; block1[0] = 0; block1[1] = 0; block1[2] = 0; block1[3] = 0; u32 block2[4]; block2[0] = 0; block2[1] = 0; block2[2] = 0; block2[3] = 0; u32 block3[4]; block3[0] = 0; block3[1] = 0; block3[2] = 0; block3[3] = 0; memcat8 (block0, block1, block2, block3, block_len, salt_buf); block_len += salt_len; memcat16 (block0, block1, block2, block3, block_len, w0); block_len += pw_len; append_0x80_4x4 (block0, block1, block2, block3, block_len); block3[2] = block_len * 8; u32 digest[4]; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (block0, block1, block2, block3, digest); /* The password first, since that is what is most unknown */ /* Then our magic string */ /* Then the raw salt */ /* Then just as many characters of the MD5(pw,salt,pw) */ //memcat16 (block0, block1, block2, block3, block_len, w); //block_len += pw_len; block_len = pw_len; block0[0] = w0[0]; block0[1] = w0[1]; block0[2] = w0[2]; block0[3] = w0[3]; block1[0] = 0; block1[1] = 0; block1[2] = 0; block1[3] = 0; block2[0] = 0; block2[1] = 0; block2[2] = 0; block2[3] = 0; block3[0] = 0; block3[1] = 0; block3[2] = 0; block3[3] = 0; append_sign (block0, block1, block_len); block_len += 6; memcat8 (block0, block1, block2, block3, block_len, salt_buf); block_len += salt_len; truncate_block (digest, pw_len); memcat16 (block0, block1, block2, block3, block_len, digest); block_len += pw_len; /* Then something really weird... */ u32 append = block0[0] & 0xFF; for (u32 j = pw_len; j; j >>= 1) { if ((j & 1) == 0) { append_1st (block0, block1, block2, block3, block_len, append); } block_len++; } append_0x80_4x4 (block0, block1, block2, block3, block_len); block3[2] = block_len * 8; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (block0, block1, block2, block3, digest); tmps[gid].digest_buf[0] = digest[0]; tmps[gid].digest_buf[1] = digest[1]; tmps[gid].digest_buf[2] = digest[2]; tmps[gid].digest_buf[3] = digest[3]; } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01600_loop (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global md5crypt_tmp_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * base */ const u32 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]; const u32 pw_len = pws[gid].pw_len; u32 w0_x80[4]; w0_x80[0] = w0[0]; w0_x80[1] = w0[1]; w0_x80[2] = w0[2]; w0_x80[3] = w0[3]; append_0x80_1x4 (w0_x80, pw_len); /** * salt */ u32 salt_buf[2]; salt_buf[0] = salt_bufs[salt_pos].salt_buf[0]; salt_buf[1] = salt_bufs[salt_pos].salt_buf[1]; const u32 salt_len = salt_bufs[salt_pos].salt_len; /** * digest */ u32 digest[4]; digest[0] = tmps[gid].digest_buf[0]; digest[1] = tmps[gid].digest_buf[1]; digest[2] = tmps[gid].digest_buf[2]; digest[3] = tmps[gid].digest_buf[3]; /** * loop */ /* and now, just to make sure things don't run too fast */ u32 block_len; u32 block0[4]; block0[0] = 0; block0[1] = 0; block0[2] = 0; block0[3] = 0; u32 block1[4]; block1[0] = 0; block1[1] = 0; block1[2] = 0; block1[3] = 0; u32 block2[4]; block2[0] = 0; block2[1] = 0; block2[2] = 0; block2[3] = 0; u32 block3[4]; block3[0] = 0; block3[1] = 0; block3[2] = 0; block3[3] = 0; for (u32 i = 0, j = loop_pos; i < loop_cnt; i++, j++) { block1[0] = 0; block1[1] = 0; block1[2] = 0; block1[3] = 0; block2[0] = 0; block2[1] = 0; block2[2] = 0; block2[3] = 0; block3[0] = 0; block3[1] = 0; const u32 j1 = (j & 1) ? 1 : 0; const u32 j3 = (j % 3) ? 1 : 0; const u32 j7 = (j % 7) ? 1 : 0; if (j1) { block0[0] = w0[0]; block0[1] = w0[1]; block0[2] = w0[2]; block0[3] = w0[3]; block_len = pw_len; if (j3) { memcat8 (block0, block1, block2, block3, block_len, salt_buf); block_len += salt_len; } if (j7) { memcat16 (block0, block1, block2, block3, block_len, w0); block_len += pw_len; } memcat16_x80 (block0, block1, block2, block3, block_len, digest); block_len += 16; } else { block0[0] = digest[0]; block0[1] = digest[1]; block0[2] = digest[2]; block0[3] = digest[3]; block_len = 16; if (j3 && j7) { block1[0] = salt_buf[0]; block1[1] = salt_buf[1]; block_len += salt_len; memcat16 (block0, block1, block2, block3, block_len, w0); block_len += pw_len; } else if (j3) { block1[0] = salt_buf[0]; block1[1] = salt_buf[1]; block_len += salt_len; } else if (j7) { block1[0] = w0[0]; block1[1] = w0[1]; block1[2] = w0[2]; block1[3] = w0[3]; block_len += pw_len; } memcat16 (block0, block1, block2, block3, block_len, w0_x80); block_len += pw_len; } block3[2] = block_len * 8; digest[0] = MD5M_A; digest[1] = MD5M_B; digest[2] = MD5M_C; digest[3] = MD5M_D; md5_transform (block0, block1, block2, block3, digest); } tmps[gid].digest_buf[0] = digest[0]; tmps[gid].digest_buf[1] = digest[1]; tmps[gid].digest_buf[2] = digest[2]; tmps[gid].digest_buf[3] = digest[3]; } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01600_comp (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global md5crypt_tmp_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) { /** * modifier */ const u32 gid = get_global_id (0); if (gid >= gid_max) return; const u32 lid = get_local_id (0); /** * digest */ const u32 r0 = tmps[gid].digest_buf[DGST_R0]; const u32 r1 = tmps[gid].digest_buf[DGST_R1]; const u32 r2 = tmps[gid].digest_buf[DGST_R2]; const u32 r3 = tmps[gid].digest_buf[DGST_R3]; #define il_pos 0 #include COMPARE_M }