/** * Author......: Jens Steube * License.....: MIT */ #define _KRB5PA_ #include "include/constants.h" #include "include/kernel_vendor.h" #ifdef VLIW1 #define VECT_SIZE1 #endif #ifdef VLIW4 #define VECT_SIZE1 #endif #ifdef VLIW5 #define VECT_SIZE1 #endif #define DGST_R0 3 #define DGST_R1 7 #define DGST_R2 2 #define DGST_R3 6 #include "include/kernel_functions.c" #include "types_amd.c" #include "common_amd.c" #include "include/rp_gpu.h" #include "rp_amd.c" typedef struct { u8 S[256]; u8 i; u8 j; } RC4_KEY; static 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; } static void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4]) { u32 i; #pragma unroll 256 for (i = 0; i < 256; i += 1) rc4_key->S[i] = i; u8 j = 0; #pragma unroll 16 for (i = 0; i < 256; i += 16) { u32 idx = i; 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++; } rc4_key->i = 0; rc4_key->j = 0; } static u32 rc4_next_4 (__local RC4_KEY *rc4_key, const u32 ct) { u8 idx; u32 xor4 = 0; u8 i = rc4_key->i; u8 j = rc4_key->j; 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; rc4_key->i = i; rc4_key->j = j; return ct ^ xor4; } static int decrypt_and_check (__local RC4_KEY *rc4_key, u32 data[4], u32 timestamp_ct[8]) { u32 pt; rc4_init_16 (rc4_key, data); pt = rc4_next_4 (rc4_key, timestamp_ct[0]); pt = rc4_next_4 (rc4_key, timestamp_ct[1]); pt = rc4_next_4 (rc4_key, timestamp_ct[2]); pt = rc4_next_4 (rc4_key, timestamp_ct[3]); if ((pt & 0xffff0000) != 0x30320000) return 0; pt = rc4_next_4 (rc4_key, timestamp_ct[4]); if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt = rc4_next_4 (rc4_key, timestamp_ct[5]); if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt = rc4_next_4 (rc4_key, timestamp_ct[6]); if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; pt >>= 8; if (((pt & 0xff) < '0') || ((pt & 0xff) > '9')) return 0; return 1; } static void md4_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4]) { u32x a = digest[0]; u32x b = digest[1]; u32x c = digest[2]; u32x d = digest[3]; MD4_STEP (MD4_Fo, a, b, c, d, w0[0], MD4C00, MD4S00); MD4_STEP (MD4_Fo, d, a, b, c, w0[1], MD4C00, MD4S01); MD4_STEP (MD4_Fo, c, d, a, b, w0[2], MD4C00, MD4S02); MD4_STEP (MD4_Fo, b, c, d, a, w0[3], MD4C00, MD4S03); MD4_STEP (MD4_Fo, a, b, c, d, w1[0], MD4C00, MD4S00); MD4_STEP (MD4_Fo, d, a, b, c, w1[1], MD4C00, MD4S01); MD4_STEP (MD4_Fo, c, d, a, b, w1[2], MD4C00, MD4S02); MD4_STEP (MD4_Fo, b, c, d, a, w1[3], MD4C00, MD4S03); MD4_STEP (MD4_Fo, a, b, c, d, w2[0], MD4C00, MD4S00); MD4_STEP (MD4_Fo, d, a, b, c, w2[1], MD4C00, MD4S01); MD4_STEP (MD4_Fo, c, d, a, b, w2[2], MD4C00, MD4S02); MD4_STEP (MD4_Fo, b, c, d, a, w2[3], MD4C00, MD4S03); MD4_STEP (MD4_Fo, a, b, c, d, w3[0], MD4C00, MD4S00); MD4_STEP (MD4_Fo, d, a, b, c, w3[1], MD4C00, MD4S01); MD4_STEP (MD4_Fo, c, d, a, b, w3[2], MD4C00, MD4S02); MD4_STEP (MD4_Fo, b, c, d, a, w3[3], MD4C00, MD4S03); MD4_STEP (MD4_Go, a, b, c, d, w0[0], MD4C01, MD4S10); MD4_STEP (MD4_Go, d, a, b, c, w1[0], MD4C01, MD4S11); MD4_STEP (MD4_Go, c, d, a, b, w2[0], MD4C01, MD4S12); MD4_STEP (MD4_Go, b, c, d, a, w3[0], MD4C01, MD4S13); MD4_STEP (MD4_Go, a, b, c, d, w0[1], MD4C01, MD4S10); MD4_STEP (MD4_Go, d, a, b, c, w1[1], MD4C01, MD4S11); MD4_STEP (MD4_Go, c, d, a, b, w2[1], MD4C01, MD4S12); MD4_STEP (MD4_Go, b, c, d, a, w3[1], MD4C01, MD4S13); MD4_STEP (MD4_Go, a, b, c, d, w0[2], MD4C01, MD4S10); MD4_STEP (MD4_Go, d, a, b, c, w1[2], MD4C01, MD4S11); MD4_STEP (MD4_Go, c, d, a, b, w2[2], MD4C01, MD4S12); MD4_STEP (MD4_Go, b, c, d, a, w3[2], MD4C01, MD4S13); MD4_STEP (MD4_Go, a, b, c, d, w0[3], MD4C01, MD4S10); MD4_STEP (MD4_Go, d, a, b, c, w1[3], MD4C01, MD4S11); MD4_STEP (MD4_Go, c, d, a, b, w2[3], MD4C01, MD4S12); MD4_STEP (MD4_Go, b, c, d, a, w3[3], MD4C01, MD4S13); MD4_STEP (MD4_H , a, b, c, d, w0[0], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2[0], MD4C02, MD4S21); MD4_STEP (MD4_H , c, d, a, b, w1[0], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[0], MD4C02, MD4S23); MD4_STEP (MD4_H , a, b, c, d, w0[2], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2[2], MD4C02, MD4S21); MD4_STEP (MD4_H , c, d, a, b, w1[2], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[2], MD4C02, MD4S23); MD4_STEP (MD4_H , a, b, c, d, w0[1], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2[1], MD4C02, MD4S21); MD4_STEP (MD4_H , c, d, a, b, w1[1], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[1], MD4C02, MD4S23); MD4_STEP (MD4_H , a, b, c, d, w0[3], MD4C02, MD4S20); MD4_STEP (MD4_H , d, a, b, c, w2[3], MD4C02, MD4S21); MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); digest[0] += a; digest[1] += b; digest[2] += c; digest[3] += d; } static void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4]) { u32x a = digest[0]; u32x b = digest[1]; u32x c = digest[2]; u32x d = digest[3]; u32x w0_t = w0[0]; u32x w1_t = w0[1]; u32x w2_t = w0[2]; u32x w3_t = w0[3]; u32x w4_t = w1[0]; u32x w5_t = w1[1]; u32x w6_t = w1[2]; u32x w7_t = w1[3]; u32x w8_t = w2[0]; u32x w9_t = w2[1]; u32x wa_t = w2[2]; u32x wb_t = w2[3]; u32x wc_t = w3[0]; u32x wd_t = w3[1]; u32x we_t = w3[2]; u32x wf_t = w3[3]; 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_H , a, b, c, d, w5_t, MD5C20, MD5S20); MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21); MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22); MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23); MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20); MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21); MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22); MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23); MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20); MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21); MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22); MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23); MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20); MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21); MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22); MD5_STEP (MD5_H , 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 hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4]) { w0[0] = w0[0] ^ 0x36363636; w0[1] = w0[1] ^ 0x36363636; w0[2] = w0[2] ^ 0x36363636; w0[3] = w0[3] ^ 0x36363636; w1[0] = w1[0] ^ 0x36363636; w1[1] = w1[1] ^ 0x36363636; w1[2] = w1[2] ^ 0x36363636; w1[3] = w1[3] ^ 0x36363636; w2[0] = w2[0] ^ 0x36363636; w2[1] = w2[1] ^ 0x36363636; w2[2] = w2[2] ^ 0x36363636; w2[3] = w2[3] ^ 0x36363636; w3[0] = w3[0] ^ 0x36363636; w3[1] = w3[1] ^ 0x36363636; w3[2] = w3[2] ^ 0x36363636; w3[3] = w3[3] ^ 0x36363636; ipad[0] = MD5M_A; ipad[1] = MD5M_B; ipad[2] = MD5M_C; ipad[3] = MD5M_D; md5_transform (w0, w1, w2, w3, ipad); w0[0] = w0[0] ^ 0x6a6a6a6a; w0[1] = w0[1] ^ 0x6a6a6a6a; w0[2] = w0[2] ^ 0x6a6a6a6a; w0[3] = w0[3] ^ 0x6a6a6a6a; w1[0] = w1[0] ^ 0x6a6a6a6a; w1[1] = w1[1] ^ 0x6a6a6a6a; w1[2] = w1[2] ^ 0x6a6a6a6a; w1[3] = w1[3] ^ 0x6a6a6a6a; w2[0] = w2[0] ^ 0x6a6a6a6a; w2[1] = w2[1] ^ 0x6a6a6a6a; w2[2] = w2[2] ^ 0x6a6a6a6a; w2[3] = w2[3] ^ 0x6a6a6a6a; w3[0] = w3[0] ^ 0x6a6a6a6a; w3[1] = w3[1] ^ 0x6a6a6a6a; w3[2] = w3[2] ^ 0x6a6a6a6a; w3[3] = w3[3] ^ 0x6a6a6a6a; opad[0] = MD5M_A; opad[1] = MD5M_B; opad[2] = MD5M_C; opad[3] = MD5M_D; md5_transform (w0, w1, w2, w3, opad); } static void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4]) { digest[0] = ipad[0]; digest[1] = ipad[1]; digest[2] = ipad[2]; digest[3] = ipad[3]; md5_transform (w0, w1, w2, w3, digest); w0[0] = digest[0]; w0[1] = digest[1]; w0[2] = digest[2]; w0[3] = digest[3]; w1[0] = 0x80; w1[1] = 0; w1[2] = 0; w1[3] = 0; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; w3[0] = 0; w3[1] = 0; w3[2] = (64 + 16) * 8; w3[3] = 0; digest[0] = opad[0]; digest[1] = opad[1]; digest[2] = opad[2]; digest[3] = opad[3]; md5_transform (w0, w1, w2, w3, digest); } static void kerb_prepare (const u32x w0[4], const u32x w1[4], const u32 pw_len, const u32 checksum[4], u32x digest[4]) { /** * pads */ u32x w0_t[4]; u32x w1_t[4]; u32x w2_t[4]; u32x w3_t[4]; w0_t[0] = w0[0]; w0_t[1] = w0[1]; w0_t[2] = w0[2]; w0_t[3] = w0[3]; w1_t[0] = w1[0]; w1_t[1] = w1[1]; w1_t[2] = w1[2]; w1_t[3] = w1[3]; 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] = 0; w3_t[3] = 0; // K=MD4(Little_indian(UNICODE(pwd)) append_0x80_2 (w0_t, w1_t, pw_len); make_unicode (w1_t, w2_t, w3_t); make_unicode (w0_t, w0_t, w1_t); w3_t[2] = pw_len * 8 * 2; w3_t[3] = 0; digest[0] = MD4M_A; digest[1] = MD4M_B; digest[2] = MD4M_C; digest[3] = MD4M_D; md4_transform (w0_t, w1_t, w2_t, w3_t, digest); // K1=MD5_HMAC(K,1); with 1 encoded as little indian on 4 bytes (01000000 in hexa); w0_t[0] = digest[0]; w0_t[1] = digest[1]; w0_t[2] = digest[2]; w0_t[3] = digest[3]; 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] = 0; w3_t[3] = 0; u32x ipad[4]; u32x opad[4]; hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad); w0_t[0] = 1; w0_t[1] = 0x80; w0_t[2] = 0; 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] = (64 + 4) * 8; w3_t[3] = 0; hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest); // K3=MD5_HMAC(K1,checksum); w0_t[0] = digest[0]; w0_t[1] = digest[1]; w0_t[2] = digest[2]; w0_t[3] = digest[3]; 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] = 0; w3_t[3] = 0; hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad); w0_t[0] = checksum[0]; w0_t[1] = checksum[1]; w0_t[2] = checksum[2]; w0_t[3] = checksum[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] = (64 + 16) * 8; w3_t[3] = 0; hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest); } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_m04 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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 */ __local RC4_KEY rc4_keys[64]; const u32 lid = get_local_id (0); const u32 gid = get_global_id (0); if (gid >= gid_max) return; /** * base */ u32x pw_buf0[4]; pw_buf0[0] = pws[gid].i[ 0]; pw_buf0[1] = pws[gid].i[ 1]; pw_buf0[2] = pws[gid].i[ 2]; pw_buf0[3] = pws[gid].i[ 3]; u32x pw_buf1[4]; pw_buf1[0] = pws[gid].i[ 4]; pw_buf1[1] = pws[gid].i[ 5]; pw_buf1[2] = pws[gid].i[ 6]; pw_buf1[3] = pws[gid].i[ 7]; const u32 pw_len = pws[gid].pw_len; /** * salt */ u32 checksum[4]; checksum[0] = krb5pa_bufs[salt_pos].checksum[0]; checksum[1] = krb5pa_bufs[salt_pos].checksum[1]; checksum[2] = krb5pa_bufs[salt_pos].checksum[2]; checksum[3] = krb5pa_bufs[salt_pos].checksum[3]; u32 timestamp_ct[8]; timestamp_ct[0] = krb5pa_bufs[salt_pos].timestamp[0]; timestamp_ct[1] = krb5pa_bufs[salt_pos].timestamp[1]; timestamp_ct[2] = krb5pa_bufs[salt_pos].timestamp[2]; timestamp_ct[3] = krb5pa_bufs[salt_pos].timestamp[3]; timestamp_ct[4] = krb5pa_bufs[salt_pos].timestamp[4]; timestamp_ct[5] = krb5pa_bufs[salt_pos].timestamp[5]; timestamp_ct[6] = krb5pa_bufs[salt_pos].timestamp[6]; timestamp_ct[7] = krb5pa_bufs[salt_pos].timestamp[7]; /** * loop */ for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++) { u32x w0[4]; w0[0] = pw_buf0[0]; w0[1] = pw_buf0[1]; w0[2] = pw_buf0[2]; w0[3] = pw_buf0[3]; u32x w1[4]; w1[0] = pw_buf1[0]; w1[1] = pw_buf1[1]; w1[2] = pw_buf1[2]; w1[3] = pw_buf1[3]; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len); /** * kerberos */ u32x digest[4]; kerb_prepare (w0, w1, out_len, checksum, digest); u32 tmp[4]; #ifdef VECT_SIZE1 tmp[0] = digest[0]; tmp[1] = digest[1]; tmp[2] = digest[2]; tmp[3] = digest[3]; if (decrypt_and_check (&rc4_keys[lid], tmp, timestamp_ct) == 1) { mark_hash_s0 (plains_buf, hashes_shown, digests_offset, gid, il_pos); d_return_buf[lid] = 1; } #endif } } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_m08 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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) { } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_m16 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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) { } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_s04 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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 */ __local RC4_KEY rc4_keys[64]; const u32 lid = get_local_id (0); const u32 gid = get_global_id (0); if (gid >= gid_max) return; /** * base */ u32x pw_buf0[4]; pw_buf0[0] = pws[gid].i[ 0]; pw_buf0[1] = pws[gid].i[ 1]; pw_buf0[2] = pws[gid].i[ 2]; pw_buf0[3] = pws[gid].i[ 3]; u32x pw_buf1[4]; pw_buf1[0] = pws[gid].i[ 4]; pw_buf1[1] = pws[gid].i[ 5]; pw_buf1[2] = pws[gid].i[ 6]; pw_buf1[3] = pws[gid].i[ 7]; const u32 pw_len = pws[gid].pw_len; /** * salt */ u32 checksum[4]; checksum[0] = krb5pa_bufs[salt_pos].checksum[0]; checksum[1] = krb5pa_bufs[salt_pos].checksum[1]; checksum[2] = krb5pa_bufs[salt_pos].checksum[2]; checksum[3] = krb5pa_bufs[salt_pos].checksum[3]; u32 timestamp_ct[8]; timestamp_ct[0] = krb5pa_bufs[salt_pos].timestamp[0]; timestamp_ct[1] = krb5pa_bufs[salt_pos].timestamp[1]; timestamp_ct[2] = krb5pa_bufs[salt_pos].timestamp[2]; timestamp_ct[3] = krb5pa_bufs[salt_pos].timestamp[3]; timestamp_ct[4] = krb5pa_bufs[salt_pos].timestamp[4]; timestamp_ct[5] = krb5pa_bufs[salt_pos].timestamp[5]; timestamp_ct[6] = krb5pa_bufs[salt_pos].timestamp[6]; timestamp_ct[7] = krb5pa_bufs[salt_pos].timestamp[7]; /** * loop */ for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++) { u32x w0[4]; w0[0] = pw_buf0[0]; w0[1] = pw_buf0[1]; w0[2] = pw_buf0[2]; w0[3] = pw_buf0[3]; u32x w1[4]; w1[0] = pw_buf1[0]; w1[1] = pw_buf1[1]; w1[2] = pw_buf1[2]; w1[3] = pw_buf1[3]; u32x w2[4]; w2[0] = 0; w2[1] = 0; w2[2] = 0; w2[3] = 0; u32x w3[4]; w3[0] = 0; w3[1] = 0; w3[2] = 0; w3[3] = 0; const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len); /** * kerberos */ u32x digest[4]; kerb_prepare (w0, w1, out_len, checksum, digest); u32 tmp[4]; #ifdef VECT_SIZE1 tmp[0] = digest[0]; tmp[1] = digest[1]; tmp[2] = digest[2]; tmp[3] = digest[3]; if (decrypt_and_check (&rc4_keys[lid], tmp, timestamp_ct) == 1) { mark_hash_s0 (plains_buf, hashes_shown, digests_offset, gid, il_pos); d_return_buf[lid] = 1; } #endif } } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_s08 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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) { } __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07500_s16 (__global pw_t *pws, __global gpu_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 krb5pa_t *krb5pa_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) { }