diff --git a/OpenCL/inc_cipher_aes.cl b/OpenCL/inc_cipher_aes.cl index 716addc39..ed98d58d9 100644 --- a/OpenCL/inc_cipher_aes.cl +++ b/OpenCL/inc_cipher_aes.cl @@ -1040,6 +1040,389 @@ DECLSPEC void aes128_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u out[3] = hc_swap32_S (out[3]); } +// 192 bit key + +DECLSPEC void aes192_ExpandKey (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3) +{ + ks[ 0] = ukey[0]; + ks[ 1] = ukey[1]; + ks[ 2] = ukey[2]; + ks[ 3] = ukey[3]; + ks[ 4] = ukey[4]; + ks[ 5] = ukey[5]; + ks[ 6] = ks[ 0] ^ 0x01000000 + ^ (s_te2[(ks[ 5] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[ 5] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[ 5] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[ 5] >> 24) & 0xff] & 0x000000ff); + ks[ 7] = ks[ 1] ^ ks[ 6]; + ks[ 8] = ks[ 2] ^ ks[ 7]; + ks[ 9] = ks[ 3] ^ ks[ 8]; + ks[10] = ks[ 4] ^ ks[ 9]; + ks[11] = ks[ 5] ^ ks[10]; + ks[12] = ks[ 6] ^ 0x02000000 + ^ (s_te2[(ks[11] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[11] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[11] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[11] >> 24) & 0xff] & 0x000000ff); + ks[13] = ks[ 7] ^ ks[12]; + ks[14] = ks[ 8] ^ ks[13]; + ks[15] = ks[ 9] ^ ks[14]; + ks[16] = ks[10] ^ ks[15]; + ks[17] = ks[11] ^ ks[16]; + ks[18] = ks[12] ^ 0x04000000 + ^ (s_te2[(ks[17] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[17] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[17] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[17] >> 24) & 0xff] & 0x000000ff); + ks[19] = ks[13] ^ ks[18]; + ks[20] = ks[14] ^ ks[19]; + ks[21] = ks[15] ^ ks[20]; + ks[22] = ks[16] ^ ks[21]; + ks[23] = ks[17] ^ ks[22]; + ks[24] = ks[18] ^ 0x08000000 + ^ (s_te2[(ks[23] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[23] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[23] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[23] >> 24) & 0xff] & 0x000000ff); + ks[25] = ks[19] ^ ks[24]; + ks[26] = ks[20] ^ ks[25]; + ks[27] = ks[21] ^ ks[26]; + ks[28] = ks[22] ^ ks[27]; + ks[29] = ks[23] ^ ks[28]; + ks[30] = ks[24] ^ 0x10000000 + ^ (s_te2[(ks[29] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[29] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[29] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[29] >> 24) & 0xff] & 0x000000ff); + ks[31] = ks[25] ^ ks[30]; + ks[32] = ks[26] ^ ks[31]; + ks[33] = ks[27] ^ ks[32]; + ks[34] = ks[28] ^ ks[33]; + ks[35] = ks[29] ^ ks[34]; + ks[36] = ks[30] ^ 0x20000000 + ^ (s_te2[(ks[35] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[35] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[35] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[35] >> 24) & 0xff] & 0x000000ff); + ks[37] = ks[31] ^ ks[36]; + ks[38] = ks[32] ^ ks[37]; + ks[39] = ks[33] ^ ks[38]; + ks[40] = ks[34] ^ ks[39]; + ks[41] = ks[35] ^ ks[40]; + ks[42] = ks[36] ^ 0x40000000 + ^ (s_te2[(ks[41] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[41] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[41] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[41] >> 24) & 0xff] & 0x000000ff); + ks[43] = ks[37] ^ ks[42]; + ks[44] = ks[38] ^ ks[43]; + ks[45] = ks[39] ^ ks[44]; + ks[46] = ks[40] ^ ks[45]; + ks[47] = ks[41] ^ ks[46]; + ks[48] = ks[42] ^ 0x80000000 + ^ (s_te2[(ks[47] >> 16) & 0xff] & 0xff000000) + ^ (s_te3[(ks[47] >> 8) & 0xff] & 0x00ff0000) + ^ (s_te0[(ks[47] >> 0) & 0xff] & 0x0000ff00) + ^ (s_te1[(ks[47] >> 24) & 0xff] & 0x000000ff); + ks[49] = ks[43] ^ ks[48]; + ks[50] = ks[44] ^ ks[49]; + ks[51] = ks[45] ^ ks[50]; +} + +DECLSPEC void aes192_InvertKey (u32 *ks, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3) +{ + u32 temp; + + temp = ks[ 0]; ks[ 0] = ks[48]; ks[48] = temp; + temp = ks[ 1]; ks[ 1] = ks[49]; ks[49] = temp; + temp = ks[ 2]; ks[ 2] = ks[50]; ks[50] = temp; + temp = ks[ 3]; ks[ 3] = ks[51]; ks[51] = temp; + temp = ks[ 4]; ks[ 4] = ks[44]; ks[44] = temp; + temp = ks[ 5]; ks[ 5] = ks[45]; ks[45] = temp; + temp = ks[ 6]; ks[ 6] = ks[46]; ks[46] = temp; + temp = ks[ 7]; ks[ 7] = ks[47]; ks[47] = temp; + temp = ks[ 8]; ks[ 8] = ks[40]; ks[40] = temp; + temp = ks[ 9]; ks[ 9] = ks[41]; ks[41] = temp; + temp = ks[10]; ks[10] = ks[42]; ks[42] = temp; + temp = ks[11]; ks[11] = ks[43]; ks[43] = temp; + temp = ks[12]; ks[12] = ks[36]; ks[36] = temp; + temp = ks[13]; ks[13] = ks[37]; ks[37] = temp; + temp = ks[14]; ks[14] = ks[38]; ks[38] = temp; + temp = ks[15]; ks[15] = ks[39]; ks[39] = temp; + temp = ks[16]; ks[16] = ks[32]; ks[32] = temp; + temp = ks[17]; ks[17] = ks[33]; ks[33] = temp; + temp = ks[18]; ks[18] = ks[34]; ks[34] = temp; + temp = ks[19]; ks[19] = ks[35]; ks[35] = temp; + temp = ks[20]; ks[20] = ks[28]; ks[28] = temp; + temp = ks[21]; ks[21] = ks[29]; ks[29] = temp; + temp = ks[22]; ks[22] = ks[30]; ks[30] = temp; + temp = ks[23]; ks[23] = ks[31]; ks[31] = temp; + + ks[ 4] = s_td0[s_te1[(ks[ 4] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 4] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 4] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 4] >> 0) & 0xff] & 0xff]; + ks[ 5] = s_td0[s_te1[(ks[ 5] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 5] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 5] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 5] >> 0) & 0xff] & 0xff]; + ks[ 6] = s_td0[s_te1[(ks[ 6] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 6] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 6] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 6] >> 0) & 0xff] & 0xff]; + ks[ 7] = s_td0[s_te1[(ks[ 7] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 7] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 7] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 7] >> 0) & 0xff] & 0xff]; + ks[ 8] = s_td0[s_te1[(ks[ 8] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 8] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 8] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 8] >> 0) & 0xff] & 0xff]; + ks[ 9] = s_td0[s_te1[(ks[ 9] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[ 9] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[ 9] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[ 9] >> 0) & 0xff] & 0xff]; + ks[10] = s_td0[s_te1[(ks[10] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[10] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[10] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[10] >> 0) & 0xff] & 0xff]; + ks[11] = s_td0[s_te1[(ks[11] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[11] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[11] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[11] >> 0) & 0xff] & 0xff]; + ks[12] = s_td0[s_te1[(ks[12] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[12] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[12] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[12] >> 0) & 0xff] & 0xff]; + ks[13] = s_td0[s_te1[(ks[13] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[13] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[13] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[13] >> 0) & 0xff] & 0xff]; + ks[14] = s_td0[s_te1[(ks[14] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[14] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[14] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[14] >> 0) & 0xff] & 0xff]; + ks[15] = s_td0[s_te1[(ks[15] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[15] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[15] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[15] >> 0) & 0xff] & 0xff]; + ks[16] = s_td0[s_te1[(ks[16] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[16] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[16] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[16] >> 0) & 0xff] & 0xff]; + ks[17] = s_td0[s_te1[(ks[17] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[17] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[17] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[17] >> 0) & 0xff] & 0xff]; + ks[18] = s_td0[s_te1[(ks[18] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[18] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[18] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[18] >> 0) & 0xff] & 0xff]; + ks[19] = s_td0[s_te1[(ks[19] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[19] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[19] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[19] >> 0) & 0xff] & 0xff]; + ks[20] = s_td0[s_te1[(ks[20] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[20] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[20] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[20] >> 0) & 0xff] & 0xff]; + ks[21] = s_td0[s_te1[(ks[21] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[21] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[21] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[21] >> 0) & 0xff] & 0xff]; + ks[22] = s_td0[s_te1[(ks[22] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[22] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[22] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[22] >> 0) & 0xff] & 0xff]; + ks[23] = s_td0[s_te1[(ks[23] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[23] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[23] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[23] >> 0) & 0xff] & 0xff]; + ks[24] = s_td0[s_te1[(ks[24] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[24] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[24] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[24] >> 0) & 0xff] & 0xff]; + ks[25] = s_td0[s_te1[(ks[25] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[25] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[25] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[25] >> 0) & 0xff] & 0xff]; + ks[26] = s_td0[s_te1[(ks[26] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[26] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[26] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[26] >> 0) & 0xff] & 0xff]; + ks[27] = s_td0[s_te1[(ks[27] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[27] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[27] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[27] >> 0) & 0xff] & 0xff]; + ks[28] = s_td0[s_te1[(ks[28] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[28] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[28] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[28] >> 0) & 0xff] & 0xff]; + ks[29] = s_td0[s_te1[(ks[29] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[29] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[29] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[29] >> 0) & 0xff] & 0xff]; + ks[30] = s_td0[s_te1[(ks[30] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[30] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[30] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[30] >> 0) & 0xff] & 0xff]; + ks[31] = s_td0[s_te1[(ks[31] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[31] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[31] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[31] >> 0) & 0xff] & 0xff]; + ks[32] = s_td0[s_te1[(ks[32] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[32] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[32] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[32] >> 0) & 0xff] & 0xff]; + ks[33] = s_td0[s_te1[(ks[33] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[33] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[33] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[33] >> 0) & 0xff] & 0xff]; + ks[34] = s_td0[s_te1[(ks[34] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[34] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[34] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[34] >> 0) & 0xff] & 0xff]; + ks[35] = s_td0[s_te1[(ks[35] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[35] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[35] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[35] >> 0) & 0xff] & 0xff]; + ks[36] = s_td0[s_te1[(ks[36] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[36] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[36] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[36] >> 0) & 0xff] & 0xff]; + ks[37] = s_td0[s_te1[(ks[37] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[37] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[37] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[37] >> 0) & 0xff] & 0xff]; + ks[38] = s_td0[s_te1[(ks[38] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[38] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[38] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[38] >> 0) & 0xff] & 0xff]; + ks[39] = s_td0[s_te1[(ks[39] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[39] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[39] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[39] >> 0) & 0xff] & 0xff]; + ks[40] = s_td0[s_te1[(ks[40] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[40] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[40] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[40] >> 0) & 0xff] & 0xff]; + ks[41] = s_td0[s_te1[(ks[41] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[41] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[41] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[41] >> 0) & 0xff] & 0xff]; + ks[42] = s_td0[s_te1[(ks[42] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[42] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[42] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[42] >> 0) & 0xff] & 0xff]; + ks[43] = s_td0[s_te1[(ks[43] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[43] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[43] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[43] >> 0) & 0xff] & 0xff]; + ks[44] = s_td0[s_te1[(ks[44] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[44] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[44] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[44] >> 0) & 0xff] & 0xff]; + ks[45] = s_td0[s_te1[(ks[45] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[45] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[45] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[45] >> 0) & 0xff] & 0xff]; + ks[46] = s_td0[s_te1[(ks[46] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[46] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[46] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[46] >> 0) & 0xff] & 0xff]; + ks[47] = s_td0[s_te1[(ks[47] >> 24) & 0xff] & 0xff] ^ s_td1[s_te1[(ks[47] >> 16) & 0xff] & 0xff] ^ s_td2[s_te1[(ks[47] >> 8) & 0xff] & 0xff] ^ s_td3[s_te1[(ks[47] >> 0) & 0xff] & 0xff]; +} + +DECLSPEC void aes192_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3) +{ + u32 ukey_s[6]; + + ukey_s[0] = hc_swap32_S (ukey[0]); + ukey_s[1] = hc_swap32_S (ukey[1]); + ukey_s[2] = hc_swap32_S (ukey[2]); + ukey_s[3] = hc_swap32_S (ukey[3]); + ukey_s[4] = hc_swap32_S (ukey[4]); + ukey_s[5] = hc_swap32_S (ukey[5]); + + aes192_ExpandKey (ks, ukey_s, s_te0, s_te1, s_te2, s_te3); +} + +DECLSPEC void aes192_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3) +{ + u32 ukey_s[6]; + + ukey_s[0] = hc_swap32_S (ukey[0]); + ukey_s[1] = hc_swap32_S (ukey[1]); + ukey_s[2] = hc_swap32_S (ukey[2]); + ukey_s[3] = hc_swap32_S (ukey[3]); + ukey_s[4] = hc_swap32_S (ukey[4]); + ukey_s[5] = hc_swap32_S (ukey[5]); + + aes192_ExpandKey (ks, ukey_s, s_te0, s_te1, s_te2, s_te3); + + aes192_InvertKey (ks, s_te1, s_td0, s_td1, s_td2, s_td3); +} + +DECLSPEC void aes192_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4) +{ + u32 in_s[4]; + + in_s[0] = hc_swap32_S (in[0]); + in_s[1] = hc_swap32_S (in[1]); + in_s[2] = hc_swap32_S (in[2]); + in_s[3] = hc_swap32_S (in[3]); + + u32 s0 = in_s[0] ^ ks[0]; + u32 s1 = in_s[1] ^ ks[1]; + u32 s2 = in_s[2] ^ ks[2]; + u32 s3 = in_s[3] ^ ks[3]; + + u32 t0; + u32 t1; + u32 t2; + u32 t3; + + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[ 4]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[ 5]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[ 6]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[ 7]; + s0 = s_te0[t0 >> 24] ^ s_te1[(t1 >> 16) & 0xff] ^ s_te2[(t2 >> 8) & 0xff] ^ s_te3[t3 & 0xff] ^ ks[ 8]; + s1 = s_te0[t1 >> 24] ^ s_te1[(t2 >> 16) & 0xff] ^ s_te2[(t3 >> 8) & 0xff] ^ s_te3[t0 & 0xff] ^ ks[ 9]; + s2 = s_te0[t2 >> 24] ^ s_te1[(t3 >> 16) & 0xff] ^ s_te2[(t0 >> 8) & 0xff] ^ s_te3[t1 & 0xff] ^ ks[10]; + s3 = s_te0[t3 >> 24] ^ s_te1[(t0 >> 16) & 0xff] ^ s_te2[(t1 >> 8) & 0xff] ^ s_te3[t2 & 0xff] ^ ks[11]; + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[12]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[13]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[14]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[15]; + s0 = s_te0[t0 >> 24] ^ s_te1[(t1 >> 16) & 0xff] ^ s_te2[(t2 >> 8) & 0xff] ^ s_te3[t3 & 0xff] ^ ks[16]; + s1 = s_te0[t1 >> 24] ^ s_te1[(t2 >> 16) & 0xff] ^ s_te2[(t3 >> 8) & 0xff] ^ s_te3[t0 & 0xff] ^ ks[17]; + s2 = s_te0[t2 >> 24] ^ s_te1[(t3 >> 16) & 0xff] ^ s_te2[(t0 >> 8) & 0xff] ^ s_te3[t1 & 0xff] ^ ks[18]; + s3 = s_te0[t3 >> 24] ^ s_te1[(t0 >> 16) & 0xff] ^ s_te2[(t1 >> 8) & 0xff] ^ s_te3[t2 & 0xff] ^ ks[19]; + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[20]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[21]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[22]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[23]; + s0 = s_te0[t0 >> 24] ^ s_te1[(t1 >> 16) & 0xff] ^ s_te2[(t2 >> 8) & 0xff] ^ s_te3[t3 & 0xff] ^ ks[24]; + s1 = s_te0[t1 >> 24] ^ s_te1[(t2 >> 16) & 0xff] ^ s_te2[(t3 >> 8) & 0xff] ^ s_te3[t0 & 0xff] ^ ks[25]; + s2 = s_te0[t2 >> 24] ^ s_te1[(t3 >> 16) & 0xff] ^ s_te2[(t0 >> 8) & 0xff] ^ s_te3[t1 & 0xff] ^ ks[26]; + s3 = s_te0[t3 >> 24] ^ s_te1[(t0 >> 16) & 0xff] ^ s_te2[(t1 >> 8) & 0xff] ^ s_te3[t2 & 0xff] ^ ks[27]; + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[28]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[29]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[30]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[31]; + s0 = s_te0[t0 >> 24] ^ s_te1[(t1 >> 16) & 0xff] ^ s_te2[(t2 >> 8) & 0xff] ^ s_te3[t3 & 0xff] ^ ks[32]; + s1 = s_te0[t1 >> 24] ^ s_te1[(t2 >> 16) & 0xff] ^ s_te2[(t3 >> 8) & 0xff] ^ s_te3[t0 & 0xff] ^ ks[33]; + s2 = s_te0[t2 >> 24] ^ s_te1[(t3 >> 16) & 0xff] ^ s_te2[(t0 >> 8) & 0xff] ^ s_te3[t1 & 0xff] ^ ks[34]; + s3 = s_te0[t3 >> 24] ^ s_te1[(t0 >> 16) & 0xff] ^ s_te2[(t1 >> 8) & 0xff] ^ s_te3[t2 & 0xff] ^ ks[35]; + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[36]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[37]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[38]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[39]; + s0 = s_te0[t0 >> 24] ^ s_te1[(t1 >> 16) & 0xff] ^ s_te2[(t2 >> 8) & 0xff] ^ s_te3[t3 & 0xff] ^ ks[40]; + s1 = s_te0[t1 >> 24] ^ s_te1[(t2 >> 16) & 0xff] ^ s_te2[(t3 >> 8) & 0xff] ^ s_te3[t0 & 0xff] ^ ks[41]; + s2 = s_te0[t2 >> 24] ^ s_te1[(t3 >> 16) & 0xff] ^ s_te2[(t0 >> 8) & 0xff] ^ s_te3[t1 & 0xff] ^ ks[42]; + s3 = s_te0[t3 >> 24] ^ s_te1[(t0 >> 16) & 0xff] ^ s_te2[(t1 >> 8) & 0xff] ^ s_te3[t2 & 0xff] ^ ks[43]; + t0 = s_te0[s0 >> 24] ^ s_te1[(s1 >> 16) & 0xff] ^ s_te2[(s2 >> 8) & 0xff] ^ s_te3[s3 & 0xff] ^ ks[44]; + t1 = s_te0[s1 >> 24] ^ s_te1[(s2 >> 16) & 0xff] ^ s_te2[(s3 >> 8) & 0xff] ^ s_te3[s0 & 0xff] ^ ks[45]; + t2 = s_te0[s2 >> 24] ^ s_te1[(s3 >> 16) & 0xff] ^ s_te2[(s0 >> 8) & 0xff] ^ s_te3[s1 & 0xff] ^ ks[46]; + t3 = s_te0[s3 >> 24] ^ s_te1[(s0 >> 16) & 0xff] ^ s_te2[(s1 >> 8) & 0xff] ^ s_te3[s2 & 0xff] ^ ks[47]; + + out[0] = (s_te4[(t0 >> 24) & 0xff] & 0xff000000) + ^ (s_te4[(t1 >> 16) & 0xff] & 0x00ff0000) + ^ (s_te4[(t2 >> 8) & 0xff] & 0x0000ff00) + ^ (s_te4[(t3 >> 0) & 0xff] & 0x000000ff) + ^ ks[48]; + + out[1] = (s_te4[(t1 >> 24) & 0xff] & 0xff000000) + ^ (s_te4[(t2 >> 16) & 0xff] & 0x00ff0000) + ^ (s_te4[(t3 >> 8) & 0xff] & 0x0000ff00) + ^ (s_te4[(t0 >> 0) & 0xff] & 0x000000ff) + ^ ks[49]; + + out[2] = (s_te4[(t2 >> 24) & 0xff] & 0xff000000) + ^ (s_te4[(t3 >> 16) & 0xff] & 0x00ff0000) + ^ (s_te4[(t0 >> 8) & 0xff] & 0x0000ff00) + ^ (s_te4[(t1 >> 0) & 0xff] & 0x000000ff) + ^ ks[50]; + + out[3] = (s_te4[(t3 >> 24) & 0xff] & 0xff000000) + ^ (s_te4[(t0 >> 16) & 0xff] & 0x00ff0000) + ^ (s_te4[(t1 >> 8) & 0xff] & 0x0000ff00) + ^ (s_te4[(t2 >> 0) & 0xff] & 0x000000ff) + ^ ks[51]; + + out[0] = hc_swap32_S (out[0]); + out[1] = hc_swap32_S (out[1]); + out[2] = hc_swap32_S (out[2]); + out[3] = hc_swap32_S (out[3]); +} + +DECLSPEC void aes192_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4) +{ + u32 in_s[4]; + + in_s[0] = hc_swap32_S (in[0]); + in_s[1] = hc_swap32_S (in[1]); + in_s[2] = hc_swap32_S (in[2]); + in_s[3] = hc_swap32_S (in[3]); + + u32 s0 = in_s[0] ^ ks[0]; + u32 s1 = in_s[1] ^ ks[1]; + u32 s2 = in_s[2] ^ ks[2]; + u32 s3 = in_s[3] ^ ks[3]; + + u32 t0; + u32 t1; + u32 t2; + u32 t3; + + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[ 4]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[ 5]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[ 6]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[ 7]; + s0 = s_td0[t0 >> 24] ^ s_td1[(t3 >> 16) & 0xff] ^ s_td2[(t2 >> 8) & 0xff] ^ s_td3[t1 & 0xff] ^ ks[ 8]; + s1 = s_td0[t1 >> 24] ^ s_td1[(t0 >> 16) & 0xff] ^ s_td2[(t3 >> 8) & 0xff] ^ s_td3[t2 & 0xff] ^ ks[ 9]; + s2 = s_td0[t2 >> 24] ^ s_td1[(t1 >> 16) & 0xff] ^ s_td2[(t0 >> 8) & 0xff] ^ s_td3[t3 & 0xff] ^ ks[10]; + s3 = s_td0[t3 >> 24] ^ s_td1[(t2 >> 16) & 0xff] ^ s_td2[(t1 >> 8) & 0xff] ^ s_td3[t0 & 0xff] ^ ks[11]; + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[12]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[13]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[14]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[15]; + s0 = s_td0[t0 >> 24] ^ s_td1[(t3 >> 16) & 0xff] ^ s_td2[(t2 >> 8) & 0xff] ^ s_td3[t1 & 0xff] ^ ks[16]; + s1 = s_td0[t1 >> 24] ^ s_td1[(t0 >> 16) & 0xff] ^ s_td2[(t3 >> 8) & 0xff] ^ s_td3[t2 & 0xff] ^ ks[17]; + s2 = s_td0[t2 >> 24] ^ s_td1[(t1 >> 16) & 0xff] ^ s_td2[(t0 >> 8) & 0xff] ^ s_td3[t3 & 0xff] ^ ks[18]; + s3 = s_td0[t3 >> 24] ^ s_td1[(t2 >> 16) & 0xff] ^ s_td2[(t1 >> 8) & 0xff] ^ s_td3[t0 & 0xff] ^ ks[19]; + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[20]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[21]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[22]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[23]; + s0 = s_td0[t0 >> 24] ^ s_td1[(t3 >> 16) & 0xff] ^ s_td2[(t2 >> 8) & 0xff] ^ s_td3[t1 & 0xff] ^ ks[24]; + s1 = s_td0[t1 >> 24] ^ s_td1[(t0 >> 16) & 0xff] ^ s_td2[(t3 >> 8) & 0xff] ^ s_td3[t2 & 0xff] ^ ks[25]; + s2 = s_td0[t2 >> 24] ^ s_td1[(t1 >> 16) & 0xff] ^ s_td2[(t0 >> 8) & 0xff] ^ s_td3[t3 & 0xff] ^ ks[26]; + s3 = s_td0[t3 >> 24] ^ s_td1[(t2 >> 16) & 0xff] ^ s_td2[(t1 >> 8) & 0xff] ^ s_td3[t0 & 0xff] ^ ks[27]; + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[28]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[29]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[30]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[31]; + s0 = s_td0[t0 >> 24] ^ s_td1[(t3 >> 16) & 0xff] ^ s_td2[(t2 >> 8) & 0xff] ^ s_td3[t1 & 0xff] ^ ks[32]; + s1 = s_td0[t1 >> 24] ^ s_td1[(t0 >> 16) & 0xff] ^ s_td2[(t3 >> 8) & 0xff] ^ s_td3[t2 & 0xff] ^ ks[33]; + s2 = s_td0[t2 >> 24] ^ s_td1[(t1 >> 16) & 0xff] ^ s_td2[(t0 >> 8) & 0xff] ^ s_td3[t3 & 0xff] ^ ks[34]; + s3 = s_td0[t3 >> 24] ^ s_td1[(t2 >> 16) & 0xff] ^ s_td2[(t1 >> 8) & 0xff] ^ s_td3[t0 & 0xff] ^ ks[35]; + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[36]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[37]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[38]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[39]; + s0 = s_td0[t0 >> 24] ^ s_td1[(t3 >> 16) & 0xff] ^ s_td2[(t2 >> 8) & 0xff] ^ s_td3[t1 & 0xff] ^ ks[40]; + s1 = s_td0[t1 >> 24] ^ s_td1[(t0 >> 16) & 0xff] ^ s_td2[(t3 >> 8) & 0xff] ^ s_td3[t2 & 0xff] ^ ks[41]; + s2 = s_td0[t2 >> 24] ^ s_td1[(t1 >> 16) & 0xff] ^ s_td2[(t0 >> 8) & 0xff] ^ s_td3[t3 & 0xff] ^ ks[42]; + s3 = s_td0[t3 >> 24] ^ s_td1[(t2 >> 16) & 0xff] ^ s_td2[(t1 >> 8) & 0xff] ^ s_td3[t0 & 0xff] ^ ks[43]; + t0 = s_td0[s0 >> 24] ^ s_td1[(s3 >> 16) & 0xff] ^ s_td2[(s2 >> 8) & 0xff] ^ s_td3[s1 & 0xff] ^ ks[44]; + t1 = s_td0[s1 >> 24] ^ s_td1[(s0 >> 16) & 0xff] ^ s_td2[(s3 >> 8) & 0xff] ^ s_td3[s2 & 0xff] ^ ks[45]; + t2 = s_td0[s2 >> 24] ^ s_td1[(s1 >> 16) & 0xff] ^ s_td2[(s0 >> 8) & 0xff] ^ s_td3[s3 & 0xff] ^ ks[46]; + t3 = s_td0[s3 >> 24] ^ s_td1[(s2 >> 16) & 0xff] ^ s_td2[(s1 >> 8) & 0xff] ^ s_td3[s0 & 0xff] ^ ks[47]; + + out[0] = (s_td4[(t0 >> 24) & 0xff] & 0xff000000) + ^ (s_td4[(t3 >> 16) & 0xff] & 0x00ff0000) + ^ (s_td4[(t2 >> 8) & 0xff] & 0x0000ff00) + ^ (s_td4[(t1 >> 0) & 0xff] & 0x000000ff) + ^ ks[48]; + + out[1] = (s_td4[(t1 >> 24) & 0xff] & 0xff000000) + ^ (s_td4[(t0 >> 16) & 0xff] & 0x00ff0000) + ^ (s_td4[(t3 >> 8) & 0xff] & 0x0000ff00) + ^ (s_td4[(t2 >> 0) & 0xff] & 0x000000ff) + ^ ks[49]; + + out[2] = (s_td4[(t2 >> 24) & 0xff] & 0xff000000) + ^ (s_td4[(t1 >> 16) & 0xff] & 0x00ff0000) + ^ (s_td4[(t0 >> 8) & 0xff] & 0x0000ff00) + ^ (s_td4[(t3 >> 0) & 0xff] & 0x000000ff) + ^ ks[50]; + + out[3] = (s_td4[(t3 >> 24) & 0xff] & 0xff000000) + ^ (s_td4[(t2 >> 16) & 0xff] & 0x00ff0000) + ^ (s_td4[(t1 >> 8) & 0xff] & 0x0000ff00) + ^ (s_td4[(t0 >> 0) & 0xff] & 0x000000ff) + ^ ks[51]; + + out[0] = hc_swap32_S (out[0]); + out[1] = hc_swap32_S (out[1]); + out[2] = hc_swap32_S (out[2]); + out[3] = hc_swap32_S (out[3]); +} + // 256 bit key DECLSPEC void aes256_ExpandKey (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3) @@ -1547,6 +1930,72 @@ DECLSPEC void AES128_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u out[3] = hc_swap32_S (out_s[3]); } +DECLSPEC void AES192_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3) +{ + u32 ukey_s[6]; + + ukey_s[0] = hc_swap32_S (ukey[0]); + ukey_s[1] = hc_swap32_S (ukey[1]); + ukey_s[2] = hc_swap32_S (ukey[2]); + ukey_s[3] = hc_swap32_S (ukey[3]); + ukey_s[4] = hc_swap32_S (ukey[4]); + ukey_s[5] = hc_swap32_S (ukey[5]); + + aes192_set_encrypt_key (ks, ukey_s, s_te0, s_te1, s_te2, s_te3); +} + +DECLSPEC void AES192_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3) +{ + u32 ukey_s[6]; + + ukey_s[0] = hc_swap32_S (ukey[0]); + ukey_s[1] = hc_swap32_S (ukey[1]); + ukey_s[2] = hc_swap32_S (ukey[2]); + ukey_s[3] = hc_swap32_S (ukey[3]); + ukey_s[4] = hc_swap32_S (ukey[4]); + ukey_s[5] = hc_swap32_S (ukey[5]); + + aes192_set_decrypt_key (ks, ukey_s, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); +} + +DECLSPEC void AES192_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4) +{ + u32 in_s[4]; + + in_s[0] = hc_swap32_S (in[0]); + in_s[1] = hc_swap32_S (in[1]); + in_s[2] = hc_swap32_S (in[2]); + in_s[3] = hc_swap32_S (in[3]); + + u32 out_s[4]; + + aes192_encrypt (ks, in_s, out_s, s_te0, s_te1, s_te2, s_te3, s_te4); + + out[0] = hc_swap32_S (out_s[0]); + out[1] = hc_swap32_S (out_s[1]); + out[2] = hc_swap32_S (out_s[2]); + out[3] = hc_swap32_S (out_s[3]); +} + +DECLSPEC void AES192_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4) +{ + u32 in_s[4]; + + in_s[0] = hc_swap32_S (in[0]); + in_s[1] = hc_swap32_S (in[1]); + in_s[2] = hc_swap32_S (in[2]); + in_s[3] = hc_swap32_S (in[3]); + + u32 out_s[4]; + + aes192_decrypt (ks, in_s, out_s, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] = hc_swap32_S (out_s[0]); + out[1] = hc_swap32_S (out_s[1]); + out[2] = hc_swap32_S (out_s[2]); + out[3] = hc_swap32_S (out_s[3]); +} + DECLSPEC void AES256_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3) { u32 ukey_s[8]; diff --git a/OpenCL/inc_cipher_aes.h b/OpenCL/inc_cipher_aes.h index b0394e2a6..d2edaf3ae 100644 --- a/OpenCL/inc_cipher_aes.h +++ b/OpenCL/inc_cipher_aes.h @@ -12,6 +12,12 @@ DECLSPEC void aes128_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_ DECLSPEC void aes128_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); DECLSPEC void aes128_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4); DECLSPEC void aes128_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4); +DECLSPEC void aes192_ExpandKey (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); +DECLSPEC void aes192_InvertKey (u32 *ks, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); +DECLSPEC void aes192_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); +DECLSPEC void aes192_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); +DECLSPEC void aes192_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4); +DECLSPEC void aes192_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4); DECLSPEC void aes256_ExpandKey (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); DECLSPEC void aes256_InvertKey (u32 *ks, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); DECLSPEC void aes256_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); @@ -22,6 +28,10 @@ DECLSPEC void AES128_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_ DECLSPEC void AES128_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); DECLSPEC void AES128_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4); DECLSPEC void AES128_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4); +DECLSPEC void AES192_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); +DECLSPEC void AES192_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); +DECLSPEC void AES192_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4); +DECLSPEC void AES192_decrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3, SHM_TYPE u32 *s_td4); DECLSPEC void AES256_set_encrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3); DECLSPEC void AES256_set_decrypt_key (u32 *ks, const u32 *ukey, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_td0, SHM_TYPE u32 *s_td1, SHM_TYPE u32 *s_td2, SHM_TYPE u32 *s_td3); DECLSPEC void AES256_encrypt (const u32 *ks, const u32 *in, u32 *out, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4); diff --git a/OpenCL/m21000_a3-pure.cl b/OpenCL/m21000_a3-pure.cl index 87c56c54f..4fd40a511 100644 --- a/OpenCL/m21000_a3-pure.cl +++ b/OpenCL/m21000_a3-pure.cl @@ -66,22 +66,22 @@ KERNEL_FQ void m21000_mxx (KERN_ATTR_VECTOR ()) u32x final[32] = { 0 }; - final[ 0] = h32_from_64_S (ctx0.h[0]); - final[ 1] = l32_from_64_S (ctx0.h[0]); - final[ 2] = h32_from_64_S (ctx0.h[1]); - final[ 3] = l32_from_64_S (ctx0.h[1]); - final[ 4] = h32_from_64_S (ctx0.h[2]); - final[ 5] = l32_from_64_S (ctx0.h[2]); - final[ 6] = h32_from_64_S (ctx0.h[3]); - final[ 7] = l32_from_64_S (ctx0.h[3]); - final[ 8] = h32_from_64_S (ctx0.h[4]); - final[ 9] = l32_from_64_S (ctx0.h[4]); - final[10] = h32_from_64_S (ctx0.h[5]); - final[11] = l32_from_64_S (ctx0.h[5]); - final[12] = h32_from_64_S (ctx0.h[6]); - final[13] = l32_from_64_S (ctx0.h[6]); - final[14] = h32_from_64_S (ctx0.h[7]); - final[15] = l32_from_64_S (ctx0.h[7]); + final[ 0] = h32_from_64 (ctx0.h[0]); + final[ 1] = l32_from_64 (ctx0.h[0]); + final[ 2] = h32_from_64 (ctx0.h[1]); + final[ 3] = l32_from_64 (ctx0.h[1]); + final[ 4] = h32_from_64 (ctx0.h[2]); + final[ 5] = l32_from_64 (ctx0.h[2]); + final[ 6] = h32_from_64 (ctx0.h[3]); + final[ 7] = l32_from_64 (ctx0.h[3]); + final[ 8] = h32_from_64 (ctx0.h[4]); + final[ 9] = l32_from_64 (ctx0.h[4]); + final[10] = h32_from_64 (ctx0.h[5]); + final[11] = l32_from_64 (ctx0.h[5]); + final[12] = h32_from_64 (ctx0.h[6]); + final[13] = l32_from_64 (ctx0.h[6]); + final[14] = h32_from_64 (ctx0.h[7]); + final[15] = l32_from_64 (ctx0.h[7]); sha512_update_vector (&ctx, final, 64); @@ -160,22 +160,22 @@ KERNEL_FQ void m21000_sxx (KERN_ATTR_VECTOR ()) u32x final[32] = { 0 }; - final[ 0] = h32_from_64_S (ctx0.h[0]); - final[ 1] = l32_from_64_S (ctx0.h[0]); - final[ 2] = h32_from_64_S (ctx0.h[1]); - final[ 3] = l32_from_64_S (ctx0.h[1]); - final[ 4] = h32_from_64_S (ctx0.h[2]); - final[ 5] = l32_from_64_S (ctx0.h[2]); - final[ 6] = h32_from_64_S (ctx0.h[3]); - final[ 7] = l32_from_64_S (ctx0.h[3]); - final[ 8] = h32_from_64_S (ctx0.h[4]); - final[ 9] = l32_from_64_S (ctx0.h[4]); - final[10] = h32_from_64_S (ctx0.h[5]); - final[11] = l32_from_64_S (ctx0.h[5]); - final[12] = h32_from_64_S (ctx0.h[6]); - final[13] = l32_from_64_S (ctx0.h[6]); - final[14] = h32_from_64_S (ctx0.h[7]); - final[15] = l32_from_64_S (ctx0.h[7]); + final[ 0] = h32_from_64 (ctx0.h[0]); + final[ 1] = l32_from_64 (ctx0.h[0]); + final[ 2] = h32_from_64 (ctx0.h[1]); + final[ 3] = l32_from_64 (ctx0.h[1]); + final[ 4] = h32_from_64 (ctx0.h[2]); + final[ 5] = l32_from_64 (ctx0.h[2]); + final[ 6] = h32_from_64 (ctx0.h[3]); + final[ 7] = l32_from_64 (ctx0.h[3]); + final[ 8] = h32_from_64 (ctx0.h[4]); + final[ 9] = l32_from_64 (ctx0.h[4]); + final[10] = h32_from_64 (ctx0.h[5]); + final[11] = l32_from_64 (ctx0.h[5]); + final[12] = h32_from_64 (ctx0.h[6]); + final[13] = l32_from_64 (ctx0.h[6]); + final[14] = h32_from_64 (ctx0.h[7]); + final[15] = l32_from_64 (ctx0.h[7]); sha512_update_vector (&ctx, final, 64); diff --git a/OpenCL/m23001_a0-optimized.cl b/OpenCL/m23001_a0-optimized.cl new file mode 100644 index 000000000..38e36df2d --- /dev/null +++ b/OpenCL/m23001_a0-optimized.cl @@ -0,0 +1,695 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp_optimized.h" +#include "inc_rp_optimized.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23001_m04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_m08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_m16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_s04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_s08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_s16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23001_a0-pure.cl b/OpenCL/m23001_a0-pure.cl new file mode 100644 index 000000000..0ce3cf586 --- /dev/null +++ b/OpenCL/m23001_a0-pure.cl @@ -0,0 +1,411 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp.h" +#include "inc_rp.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23001_mxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_sxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23001_a1-optimized.cl b/OpenCL/m23001_a1-optimized.cl new file mode 100644 index 000000000..dcbcab1cc --- /dev/null +++ b/OpenCL/m23001_a1-optimized.cl @@ -0,0 +1,809 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23001_m04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_m08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_m16 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_s04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_s08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23001_s16 (KERN_ATTR_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23001_a1-pure.cl b/OpenCL/m23001_a1-pure.cl new file mode 100644 index 000000000..cf4f004c2 --- /dev/null +++ b/OpenCL/m23001_a1-pure.cl @@ -0,0 +1,405 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23001_mxx (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_sxx (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23001_a3-optimized.cl b/OpenCL/m23001_a3-optimized.cl new file mode 100644 index 000000000..04c4c264b --- /dev/null +++ b/OpenCL/m23001_a3-optimized.cl @@ -0,0 +1,1279 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +DECLSPEC void m23001m (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +DECLSPEC void m23001s (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_m04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23001_m08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23001_m16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23001_s04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23001_s08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23001_s16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23001s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} diff --git a/OpenCL/m23001_a3-pure.cl b/OpenCL/m23001_a3-pure.cl new file mode 100644 index 000000000..9875bdf79 --- /dev/null +++ b/OpenCL/m23001_a3-pure.cl @@ -0,0 +1,431 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23001_mxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23001_sxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[4]; + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 44 + + u32 ks[KEYLEN]; + + AES128_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23002_a0-optimized.cl b/OpenCL/m23002_a0-optimized.cl new file mode 100644 index 000000000..8ef0e36a9 --- /dev/null +++ b/OpenCL/m23002_a0-optimized.cl @@ -0,0 +1,801 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp_optimized.h" +#include "inc_rp_optimized.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23002_m04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_m08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_m16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_s04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_s08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_s16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23002_a0-pure.cl b/OpenCL/m23002_a0-pure.cl new file mode 100644 index 000000000..309f07cfa --- /dev/null +++ b/OpenCL/m23002_a0-pure.cl @@ -0,0 +1,517 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp.h" +#include "inc_rp.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23002_mxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_sxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23002_a1-optimized.cl b/OpenCL/m23002_a1-optimized.cl new file mode 100644 index 000000000..bf8e90201 --- /dev/null +++ b/OpenCL/m23002_a1-optimized.cl @@ -0,0 +1,915 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23002_m04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_m08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_m16 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_s04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_s08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23002_s16 (KERN_ATTR_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23002_a1-pure.cl b/OpenCL/m23002_a1-pure.cl new file mode 100644 index 000000000..6cad15fb1 --- /dev/null +++ b/OpenCL/m23002_a1-pure.cl @@ -0,0 +1,511 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23002_mxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_sxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23002_a3-optimized.cl b/OpenCL/m23002_a3-optimized.cl new file mode 100644 index 000000000..e81f891a9 --- /dev/null +++ b/OpenCL/m23002_a3-optimized.cl @@ -0,0 +1,1385 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +DECLSPEC void m23002m (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +DECLSPEC void m23002s (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_m04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23002_m08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23002_m16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23002_s04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23002_s08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23002_s16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23002s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} diff --git a/OpenCL/m23002_a3-pure.cl b/OpenCL/m23002_a3-pure.cl new file mode 100644 index 000000000..0dd5cf00b --- /dev/null +++ b/OpenCL/m23002_a3-pure.cl @@ -0,0 +1,537 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23002_mxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23002_sxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[6]; // 5 + 1 = 6 (20 bytes + 4 bytes = 24 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 52 + + u32 ks[KEYLEN]; + + AES192_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes192_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23003_a0-optimized.cl b/OpenCL/m23003_a0-optimized.cl new file mode 100644 index 000000000..b83e45769 --- /dev/null +++ b/OpenCL/m23003_a0-optimized.cl @@ -0,0 +1,805 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp_optimized.h" +#include "inc_rp_optimized.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23003_m04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_m08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_m16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_s04 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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 & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + u32x w0[4] = { 0 }; + u32x w1[4] = { 0 }; + u32x w2[4] = { 0 }; + u32x w3[4] = { 0 }; + + const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1); + + append_0x80_2x4_VV (w0, w1, out_len); + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = out_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_s08 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_s16 (KERN_ATTR_RULES_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23003_a0-pure.cl b/OpenCL/m23003_a0-pure.cl new file mode 100644 index 000000000..62c15e642 --- /dev/null +++ b/OpenCL/m23003_a0-pure.cl @@ -0,0 +1,521 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_rp.h" +#include "inc_rp.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23003_mxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_sxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + COPY_PW (pws[gid]); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + pw_t tmp = PASTE_PW; + + tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len); + + sha1_ctx_t ctx; + + sha1_init (&ctx); + + sha1_update_swap (&ctx, tmp.i, tmp.pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23003_a1-optimized.cl b/OpenCL/m23003_a1-optimized.cl new file mode 100644 index 000000000..1e38939b1 --- /dev/null +++ b/OpenCL/m23003_a1-optimized.cl @@ -0,0 +1,919 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23003_m04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_m08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_m16 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_s04 (KERN_ATTR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 pw_buf0[4]; + u32 pw_buf1[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]; + 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_l_len = pws[gid].pw_len & 63; + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos) & 63; + + const u32x pw_len = (pw_l_len + pw_r_len) & 63; + + /** + * concat password candidate + */ + + u32x wordl0[4] = { 0 }; + u32x wordl1[4] = { 0 }; + u32x wordl2[4] = { 0 }; + u32x wordl3[4] = { 0 }; + + wordl0[0] = pw_buf0[0]; + wordl0[1] = pw_buf0[1]; + wordl0[2] = pw_buf0[2]; + wordl0[3] = pw_buf0[3]; + wordl1[0] = pw_buf1[0]; + wordl1[1] = pw_buf1[1]; + wordl1[2] = pw_buf1[2]; + wordl1[3] = pw_buf1[3]; + + u32x wordr0[4] = { 0 }; + u32x wordr1[4] = { 0 }; + u32x wordr2[4] = { 0 }; + u32x wordr3[4] = { 0 }; + + wordr0[0] = ix_create_combt (combs_buf, il_pos, 0); + wordr0[1] = ix_create_combt (combs_buf, il_pos, 1); + wordr0[2] = ix_create_combt (combs_buf, il_pos, 2); + wordr0[3] = ix_create_combt (combs_buf, il_pos, 3); + wordr1[0] = ix_create_combt (combs_buf, il_pos, 4); + wordr1[1] = ix_create_combt (combs_buf, il_pos, 5); + wordr1[2] = ix_create_combt (combs_buf, il_pos, 6); + wordr1[3] = ix_create_combt (combs_buf, il_pos, 7); + + if (combs_mode == COMBINATOR_MODE_BASE_LEFT) + { + switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len); + } + else + { + switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len); + } + + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; + + w0[0] = wordl0[0] | wordr0[0]; + w0[1] = wordl0[1] | wordr0[1]; + w0[2] = wordl0[2] | wordr0[2]; + w0[3] = wordl0[3] | wordr0[3]; + w1[0] = wordl1[0] | wordr1[0]; + w1[1] = wordl1[1] | wordr1[1]; + w1[2] = wordl1[2] | wordr1[2]; + w1[3] = wordl1[3] | wordr1[3]; + w2[0] = wordl2[0] | wordr2[0]; + w2[1] = wordl2[1] | wordr2[1]; + w2[2] = wordl2[2] | wordr2[2]; + w2[3] = wordl2[3] | wordr2[3]; + w3[0] = wordl3[0] | wordr3[0]; + w3[1] = wordl3[1] | wordr3[1]; + w3[2] = wordl3[2] | wordr3[2]; + w3[3] = wordl3[3] | wordr3[3]; + + /** + * sha1 + */ + + u32x w0_t = hc_swap32 (w0[0]); + u32x w1_t = hc_swap32 (w0[1]); + u32x w2_t = hc_swap32 (w0[2]); + u32x w3_t = hc_swap32 (w0[3]); + u32x w4_t = hc_swap32 (w1[0]); + u32x w5_t = hc_swap32 (w1[1]); + u32x w6_t = hc_swap32 (w1[2]); + u32x w7_t = hc_swap32 (w1[3]); + u32x w8_t = hc_swap32 (w2[0]); + u32x w9_t = hc_swap32 (w2[1]); + u32x wa_t = hc_swap32 (w2[2]); + u32x wb_t = hc_swap32 (w2[3]); + u32x wc_t = hc_swap32 (w3[0]); + u32x wd_t = hc_swap32 (w3[1]); + u32x we_t = 0; + u32x wf_t = pw_len * 8; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t); + w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t); + w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t); + w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t); + w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t); + w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t); + w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t); + w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t); + w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t); + w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t); + w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t); + wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t); + wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t); + wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t); + wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t); + we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t); + wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_s08 (KERN_ATTR_ESALT (securezip_t)) +{ +} + +KERNEL_FQ void m23003_s16 (KERN_ATTR_ESALT (securezip_t)) +{ +} diff --git a/OpenCL/m23003_a1-pure.cl b/OpenCL/m23003_a1-pure.cl new file mode 100644 index 000000000..30b4afd8f --- /dev/null +++ b/OpenCL/m23003_a1-pure.cl @@ -0,0 +1,515 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_scalar.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23003_mxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_sxx (KERN_ATTR_RULES_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + sha1_ctx_t ctx0; + + sha1_init (&ctx0); + + sha1_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len); + + /** + * loop + */ + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos++) + { + sha1_ctx_t ctx = ctx0; + + sha1_update_global_swap (&ctx, combs_buf[il_pos].i, combs_buf[il_pos].pw_len); + + sha1_final (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/OpenCL/m23003_a3-optimized.cl b/OpenCL/m23003_a3-optimized.cl new file mode 100644 index 000000000..7f6656935 --- /dev/null +++ b/OpenCL/m23003_a3-optimized.cl @@ -0,0 +1,1389 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +DECLSPEC void m23003m (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +DECLSPEC void m23003s (SHM_TYPE u32a *s_te0, SHM_TYPE u32a *s_te1, SHM_TYPE u32a *s_te2, SHM_TYPE u32a *s_te3, SHM_TYPE u32a *s_te4, SHM_TYPE u32a *s_td0, SHM_TYPE u32a *s_td1, SHM_TYPE u32a *s_td2, SHM_TYPE u32a *s_td3, SHM_TYPE u32a *s_td4, u32 *w, const u32 pw_len, KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * modifier + */ + + const u64 gid = get_global_id (0); + + /** + * base + */ + + const u32 c_16s = hc_rotl32_S ((w[13] ^ w[ 8] ^ w[ 2] ), 1u); + const u32 c_17s = hc_rotl32_S ((w[14] ^ w[ 9] ^ w[ 3] ^ w[ 1]), 1u); + const u32 c_18s = hc_rotl32_S ((w[15] ^ w[10] ^ w[ 4] ^ w[ 2]), 1u); + const u32 c_19s = hc_rotl32_S ((c_16s ^ w[11] ^ w[ 5] ^ w[ 3]), 1u); + const u32 c_20s = hc_rotl32_S ((c_17s ^ w[12] ^ w[ 6] ^ w[ 4]), 1u); + const u32 c_21s = hc_rotl32_S ((c_18s ^ w[13] ^ w[ 7] ^ w[ 5]), 1u); + const u32 c_22s = hc_rotl32_S ((c_19s ^ w[14] ^ w[ 8] ^ w[ 6]), 1u); + const u32 c_23s = hc_rotl32_S ((c_20s ^ w[15] ^ w[ 9] ^ w[ 7]), 1u); + const u32 c_24s = hc_rotl32_S ((c_21s ^ c_16s ^ w[10] ^ w[ 8]), 1u); + const u32 c_25s = hc_rotl32_S ((c_22s ^ c_17s ^ w[11] ^ w[ 9]), 1u); + const u32 c_26s = hc_rotl32_S ((c_23s ^ c_18s ^ w[12] ^ w[10]), 1u); + const u32 c_27s = hc_rotl32_S ((c_24s ^ c_19s ^ w[13] ^ w[11]), 1u); + const u32 c_28s = hc_rotl32_S ((c_25s ^ c_20s ^ w[14] ^ w[12]), 1u); + const u32 c_29s = hc_rotl32_S ((c_26s ^ c_21s ^ w[15] ^ w[13]), 1u); + const u32 c_30s = hc_rotl32_S ((c_27s ^ c_22s ^ c_16s ^ w[14]), 1u); + const u32 c_31s = hc_rotl32_S ((c_28s ^ c_23s ^ c_17s ^ w[15]), 1u); + const u32 c_32s = hc_rotl32_S ((c_29s ^ c_24s ^ c_18s ^ c_16s), 1u); + const u32 c_33s = hc_rotl32_S ((c_30s ^ c_25s ^ c_19s ^ c_17s), 1u); + const u32 c_34s = hc_rotl32_S ((c_31s ^ c_26s ^ c_20s ^ c_18s), 1u); + const u32 c_35s = hc_rotl32_S ((c_32s ^ c_27s ^ c_21s ^ c_19s), 1u); + const u32 c_36s = hc_rotl32_S ((c_33s ^ c_28s ^ c_22s ^ c_20s), 1u); + const u32 c_37s = hc_rotl32_S ((c_34s ^ c_29s ^ c_23s ^ c_21s), 1u); + const u32 c_38s = hc_rotl32_S ((c_35s ^ c_30s ^ c_24s ^ c_22s), 1u); + const u32 c_39s = hc_rotl32_S ((c_36s ^ c_31s ^ c_25s ^ c_23s), 1u); + const u32 c_40s = hc_rotl32_S ((c_37s ^ c_32s ^ c_26s ^ c_24s), 1u); + const u32 c_41s = hc_rotl32_S ((c_38s ^ c_33s ^ c_27s ^ c_25s), 1u); + const u32 c_42s = hc_rotl32_S ((c_39s ^ c_34s ^ c_28s ^ c_26s), 1u); + const u32 c_43s = hc_rotl32_S ((c_40s ^ c_35s ^ c_29s ^ c_27s), 1u); + const u32 c_44s = hc_rotl32_S ((c_41s ^ c_36s ^ c_30s ^ c_28s), 1u); + const u32 c_45s = hc_rotl32_S ((c_42s ^ c_37s ^ c_31s ^ c_29s), 1u); + const u32 c_46s = hc_rotl32_S ((c_43s ^ c_38s ^ c_32s ^ c_30s), 1u); + const u32 c_47s = hc_rotl32_S ((c_44s ^ c_39s ^ c_33s ^ c_31s), 1u); + const u32 c_48s = hc_rotl32_S ((c_45s ^ c_40s ^ c_34s ^ c_32s), 1u); + const u32 c_49s = hc_rotl32_S ((c_46s ^ c_41s ^ c_35s ^ c_33s), 1u); + const u32 c_50s = hc_rotl32_S ((c_47s ^ c_42s ^ c_36s ^ c_34s), 1u); + const u32 c_51s = hc_rotl32_S ((c_48s ^ c_43s ^ c_37s ^ c_35s), 1u); + const u32 c_52s = hc_rotl32_S ((c_49s ^ c_44s ^ c_38s ^ c_36s), 1u); + const u32 c_53s = hc_rotl32_S ((c_50s ^ c_45s ^ c_39s ^ c_37s), 1u); + const u32 c_54s = hc_rotl32_S ((c_51s ^ c_46s ^ c_40s ^ c_38s), 1u); + const u32 c_55s = hc_rotl32_S ((c_52s ^ c_47s ^ c_41s ^ c_39s), 1u); + const u32 c_56s = hc_rotl32_S ((c_53s ^ c_48s ^ c_42s ^ c_40s), 1u); + const u32 c_57s = hc_rotl32_S ((c_54s ^ c_49s ^ c_43s ^ c_41s), 1u); + const u32 c_58s = hc_rotl32_S ((c_55s ^ c_50s ^ c_44s ^ c_42s), 1u); + const u32 c_59s = hc_rotl32_S ((c_56s ^ c_51s ^ c_45s ^ c_43s), 1u); + const u32 c_60s = hc_rotl32_S ((c_57s ^ c_52s ^ c_46s ^ c_44s), 1u); + const u32 c_61s = hc_rotl32_S ((c_58s ^ c_53s ^ c_47s ^ c_45s), 1u); + const u32 c_62s = hc_rotl32_S ((c_59s ^ c_54s ^ c_48s ^ c_46s), 1u); + const u32 c_63s = hc_rotl32_S ((c_60s ^ c_55s ^ c_49s ^ c_47s), 1u); + const u32 c_64s = hc_rotl32_S ((c_61s ^ c_56s ^ c_50s ^ c_48s), 1u); + const u32 c_65s = hc_rotl32_S ((c_62s ^ c_57s ^ c_51s ^ c_49s), 1u); + const u32 c_66s = hc_rotl32_S ((c_63s ^ c_58s ^ c_52s ^ c_50s), 1u); + const u32 c_67s = hc_rotl32_S ((c_64s ^ c_59s ^ c_53s ^ c_51s), 1u); + const u32 c_68s = hc_rotl32_S ((c_65s ^ c_60s ^ c_54s ^ c_52s), 1u); + const u32 c_69s = hc_rotl32_S ((c_66s ^ c_61s ^ c_55s ^ c_53s), 1u); + const u32 c_70s = hc_rotl32_S ((c_67s ^ c_62s ^ c_56s ^ c_54s), 1u); + const u32 c_71s = hc_rotl32_S ((c_68s ^ c_63s ^ c_57s ^ c_55s), 1u); + const u32 c_72s = hc_rotl32_S ((c_69s ^ c_64s ^ c_58s ^ c_56s), 1u); + const u32 c_73s = hc_rotl32_S ((c_70s ^ c_65s ^ c_59s ^ c_57s), 1u); + const u32 c_74s = hc_rotl32_S ((c_71s ^ c_66s ^ c_60s ^ c_58s), 1u); + const u32 c_75s = hc_rotl32_S ((c_72s ^ c_67s ^ c_61s ^ c_59s), 1u); + + const u32 c_17sK = c_17s + SHA1C00; + const u32 c_18sK = c_18s + SHA1C00; + const u32 c_20sK = c_20s + SHA1C01; + const u32 c_21sK = c_21s + SHA1C01; + const u32 c_23sK = c_23s + SHA1C01; + const u32 c_26sK = c_26s + SHA1C01; + const u32 c_27sK = c_27s + SHA1C01; + const u32 c_29sK = c_29s + SHA1C01; + const u32 c_33sK = c_33s + SHA1C01; + const u32 c_39sK = c_39s + SHA1C01; + const u32 c_41sK = c_41s + SHA1C02; + const u32 c_45sK = c_45s + SHA1C02; + const u32 c_53sK = c_53s + SHA1C02; + const u32 c_65sK = c_65s + SHA1C03; + const u32 c_69sK = c_69s + SHA1C03; + + /** + * loop + */ + + u32 w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + const u32x w0s01 = hc_rotl32 (w0, 1u); + const u32x w0s02 = hc_rotl32 (w0, 2u); + const u32x w0s03 = hc_rotl32 (w0, 3u); + const u32x w0s04 = hc_rotl32 (w0, 4u); + const u32x w0s05 = hc_rotl32 (w0, 5u); + const u32x w0s06 = hc_rotl32 (w0, 6u); + const u32x w0s07 = hc_rotl32 (w0, 7u); + const u32x w0s08 = hc_rotl32 (w0, 8u); + const u32x w0s09 = hc_rotl32 (w0, 9u); + const u32x w0s10 = hc_rotl32 (w0, 10u); + const u32x w0s11 = hc_rotl32 (w0, 11u); + const u32x w0s12 = hc_rotl32 (w0, 12u); + const u32x w0s13 = hc_rotl32 (w0, 13u); + const u32x w0s14 = hc_rotl32 (w0, 14u); + const u32x w0s15 = hc_rotl32 (w0, 15u); + const u32x w0s16 = hc_rotl32 (w0, 16u); + const u32x w0s17 = hc_rotl32 (w0, 17u); + const u32x w0s18 = hc_rotl32 (w0, 18u); + const u32x w0s19 = hc_rotl32 (w0, 19u); + const u32x w0s20 = hc_rotl32 (w0, 20u); + + const u32x w0s04___w0s06 = w0s04 ^ w0s06; + const u32x w0s04___w0s08 = w0s04 ^ w0s08; + const u32x w0s08___w0s12 = w0s08 ^ w0s12; + const u32x w0s04___w0s06___w0s07 = w0s04___w0s06 ^ w0s07; + + u32x a = SHA1M_A; + u32x b = SHA1M_B; + u32x c = SHA1M_C; + u32x d = SHA1M_D; + u32x e = SHA1M_E; + + #undef K + #define K SHA1C00 + + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 1]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 2]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 3]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 4]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[ 5]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[ 6]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[ 7]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[ 8]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[ 9]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[10]); + SHA1_STEP (SHA1_F0o, e, a, b, c, d, w[11]); + SHA1_STEP (SHA1_F0o, d, e, a, b, c, w[12]); + SHA1_STEP (SHA1_F0o, c, d, e, a, b, w[13]); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, w[14]); + SHA1_STEP (SHA1_F0o, a, b, c, d, e, w[15]); + + SHA1_STEP (SHA1_F0o, e, a, b, c, d, (c_16s ^ w0s01)); + SHA1_STEPX(SHA1_F0o, d, e, a, b, c, (c_17sK)); + SHA1_STEPX(SHA1_F0o, c, d, e, a, b, (c_18sK)); + SHA1_STEP (SHA1_F0o, b, c, d, e, a, (c_19s ^ w0s02)); + + #undef K + #define K SHA1C01 + + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_20sK)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_21sK)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_22s ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_23sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_24s ^ w0s02)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_25s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , e, a, b, c, d, (c_26sK)); + SHA1_STEPX(SHA1_F1 , d, e, a, b, c, (c_27sK)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_28s ^ w0s05)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_29sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_30s ^ w0s02 ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_31s ^ w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_32s ^ w0s02 ^ w0s03)); + SHA1_STEPX(SHA1_F1 , c, d, e, a, b, (c_33sK)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_34s ^ w0s07)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_35s ^ w0s04)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_36s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_37s ^ w0s08)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_38s ^ w0s04)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_39sK)); + + #undef K + #define K SHA1C02 + + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_40s ^ w0s04 ^ w0s09)); + SHA1_STEPX(SHA1_F2o, e, a, b, c, d, (c_41sK)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_42s ^ w0s06 ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_43s ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_44s ^ w0s03 ^ w0s06 ^ w0s07)); + SHA1_STEPX(SHA1_F2o, a, b, c, d, e, (c_45sK)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_46s ^ w0s04 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_47s ^ w0s04___w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_48s ^ w0s03 ^ w0s04___w0s08 ^ w0s05 ^ w0s10)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_49s ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_50s ^ w0s08)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_51s ^ w0s04___w0s06)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_52s ^ w0s04___w0s08 ^ w0s13)); + SHA1_STEPX(SHA1_F2o, c, d, e, a, b, (c_53sK)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_54s ^ w0s07 ^ w0s10 ^ w0s12)); + SHA1_STEP (SHA1_F2o, a, b, c, d, e, (c_55s ^ w0s14)); + SHA1_STEP (SHA1_F2o, e, a, b, c, d, (c_56s ^ w0s04___w0s06___w0s07 ^ w0s10 ^ w0s11)); + SHA1_STEP (SHA1_F2o, d, e, a, b, c, (c_57s ^ w0s08)); + SHA1_STEP (SHA1_F2o, c, d, e, a, b, (c_58s ^ w0s04___w0s08 ^ w0s15)); + SHA1_STEP (SHA1_F2o, b, c, d, e, a, (c_59s ^ w0s08___w0s12)); + + #undef K + #define K SHA1C03 + + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_60s ^ w0s04 ^ w0s08___w0s12 ^ w0s07 ^ w0s14)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_61s ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_62s ^ w0s04___w0s06 ^ w0s08___w0s12)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_63s ^ w0s08)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_64s ^ w0s04___w0s06___w0s07 ^ w0s08___w0s12 ^ w0s17)); + SHA1_STEPX(SHA1_F1 , a, b, c, d, e, (c_65sK)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_66s ^ w0s14 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_67s ^ w0s08 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_68s ^ w0s11 ^ w0s14 ^ w0s15)); + SHA1_STEPX(SHA1_F1 , b, c, d, e, a, (c_69sK)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_70s ^ w0s12 ^ w0s19)); + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_71s ^ w0s12 ^ w0s16)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_72s ^ w0s05 ^ w0s11 ^ w0s12 ^ w0s13 ^ w0s16 ^ w0s18)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_73s ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_74s ^ w0s08 ^ w0s16)); + SHA1_STEP (SHA1_F1 , a, b, c, d, e, (c_75s ^ w0s06 ^ w0s12 ^ w0s14)); + + const u32x c_76s = hc_rotl32 ((c_73s ^ c_68s ^ c_62s ^ c_60s), 1u); + const u32x c_77s = hc_rotl32 ((c_74s ^ c_69s ^ c_63s ^ c_61s), 1u); + const u32x c_78s = hc_rotl32 ((c_75s ^ c_70s ^ c_64s ^ c_62s), 1u); + const u32x c_79s = hc_rotl32 ((c_76s ^ c_71s ^ c_65s ^ c_63s), 1u); + + const u32x w0s21 = hc_rotl32 (w0, 21u); + const u32x w0s22 = hc_rotl32 (w0, 22U); + + SHA1_STEP (SHA1_F1 , e, a, b, c, d, (c_76s ^ w0s07 ^ w0s08___w0s12 ^ w0s16 ^ w0s21)); + SHA1_STEP (SHA1_F1 , d, e, a, b, c, (c_77s)); + SHA1_STEP (SHA1_F1 , c, d, e, a, b, (c_78s ^ w0s07 ^ w0s08 ^ w0s15 ^ w0s18 ^ w0s20)); + SHA1_STEP (SHA1_F1 , b, c, d, e, a, (c_79s ^ w0s08 ^ w0s22)); + + a += SHA1M_A; + b += SHA1M_B; + c += SHA1M_C; + d += SHA1M_D; + e += SHA1M_E; + + u32 t0[4]; + + t0[0] = 0x36363636 ^ a; + t0[1] = 0x36363636 ^ b; + t0[2] = 0x36363636 ^ c; + t0[3] = 0x36363636 ^ d; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ e; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ a; + t0[1] = 0x5c5c5c5c ^ b; + t0[2] = 0x5c5c5c5c ^ c; + t0[3] = 0x5c5c5c5c ^ d; + + t1[0] = 0x5c5c5c5c ^ e; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_m04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23003_m08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23003_m16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003m (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23003_s04 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = 0; + w[ 5] = 0; + w[ 6] = 0; + w[ 7] = 0; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23003_s08 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = 0; + w[ 9] = 0; + w[10] = 0; + w[11] = 0; + w[12] = 0; + w[13] = 0; + w[14] = 0; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} + +KERNEL_FQ void m23003_s16 (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + /** + * base + */ + + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + u32 w[16]; + + w[ 0] = pws[gid].i[ 0]; + w[ 1] = pws[gid].i[ 1]; + w[ 2] = pws[gid].i[ 2]; + w[ 3] = pws[gid].i[ 3]; + w[ 4] = pws[gid].i[ 4]; + w[ 5] = pws[gid].i[ 5]; + w[ 6] = pws[gid].i[ 6]; + w[ 7] = pws[gid].i[ 7]; + w[ 8] = pws[gid].i[ 8]; + w[ 9] = pws[gid].i[ 9]; + w[10] = pws[gid].i[10]; + w[11] = pws[gid].i[11]; + w[12] = pws[gid].i[12]; + w[13] = pws[gid].i[13]; + w[14] = pws[gid].i[14]; + w[15] = pws[gid].i[15]; + + const u32 pw_len = pws[gid].pw_len & 63; + + /** + * main + */ + + m23003s (s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, 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, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max); +} diff --git a/OpenCL/m23003_a3-pure.cl b/OpenCL/m23003_a3-pure.cl new file mode 100644 index 000000000..d1522701c --- /dev/null +++ b/OpenCL/m23003_a3-pure.cl @@ -0,0 +1,541 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +//#define NEW_SIMD_CODE + +#ifdef KERNEL_STATIC +#include "inc_vendor.h" +#include "inc_types.h" +#include "inc_platform.cl" +#include "inc_common.cl" +#include "inc_simd.cl" +#include "inc_hash_sha1.cl" +#include "inc_cipher_aes.cl" +#endif + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +KERNEL_FQ void m23003_mxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} + +KERNEL_FQ void m23003_sxx (KERN_ATTR_VECTOR_ESALT (securezip_t)) +{ + const u64 gid = get_global_id (0); + const u64 lid = get_local_id (0); + const u64 lsz = get_local_size (0); + + /** + * aes shared + */ + + #ifdef REAL_SHM + + LOCAL_VK u32 s_td0[256]; + LOCAL_VK u32 s_td1[256]; + LOCAL_VK u32 s_td2[256]; + LOCAL_VK u32 s_td3[256]; + LOCAL_VK u32 s_td4[256]; + + LOCAL_VK u32 s_te0[256]; + LOCAL_VK u32 s_te1[256]; + LOCAL_VK u32 s_te2[256]; + LOCAL_VK u32 s_te3[256]; + LOCAL_VK u32 s_te4[256]; + + for (u32 i = lid; i < 256; i += lsz) + { + s_td0[i] = td0[i]; + s_td1[i] = td1[i]; + s_td2[i] = td2[i]; + s_td3[i] = td3[i]; + s_td4[i] = td4[i]; + + s_te0[i] = te0[i]; + s_te1[i] = te1[i]; + s_te2[i] = te2[i]; + s_te3[i] = te3[i]; + s_te4[i] = te4[i]; + } + + SYNC_THREADS (); + + #else + + CONSTANT_AS u32a *s_td0 = td0; + CONSTANT_AS u32a *s_td1 = td1; + CONSTANT_AS u32a *s_td2 = td2; + CONSTANT_AS u32a *s_td3 = td3; + CONSTANT_AS u32a *s_td4 = td4; + + CONSTANT_AS u32a *s_te0 = te0; + CONSTANT_AS u32a *s_te1 = te1; + CONSTANT_AS u32a *s_te2 = te2; + CONSTANT_AS u32a *s_te3 = te3; + CONSTANT_AS u32a *s_te4 = te4; + + #endif + + if (gid >= gid_max) return; + + /** + * base + */ + + const u32 pw_len = pws[gid].pw_len; + + u32x w[64] = { 0 }; + + for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1) + { + w[idx] = pws[gid].i[idx]; + } + + /** + * loop + */ + + u32x w0l = w[0]; + + for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE) + { + const u32x w0r = words_buf_r[il_pos / VECT_SIZE]; + + const u32x w0 = w0l | w0r; + + w[0] = w0; + + sha1_ctx_vector_t ctx; + + sha1_init_vector (&ctx); + + sha1_update_vector (&ctx, w, pw_len); + + sha1_final_vector (&ctx); + + u32 t0[4]; + + t0[0] = 0x36363636 ^ ctx.h[0]; + t0[1] = 0x36363636 ^ ctx.h[1]; + t0[2] = 0x36363636 ^ ctx.h[2]; + t0[3] = 0x36363636 ^ ctx.h[3]; + + u32 t1[4]; + + t1[0] = 0x36363636 ^ ctx.h[4]; + t1[1] = 0x36363636; + t1[2] = 0x36363636; + t1[3] = 0x36363636; + + u32 t2[4]; + + t2[0] = 0x36363636; + t2[1] = 0x36363636; + t2[2] = 0x36363636; + t2[3] = 0x36363636; + + u32 t3[4]; + + t3[0] = 0x36363636; + t3[1] = 0x36363636; + t3[2] = 0x36363636; + t3[3] = 0x36363636; + + u32 digest[5]; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + u32 key[8]; // 5 + 3 = 8 (20 bytes + 12 bytes = 32 bytes for the key) + + key[0] = digest[0]; + key[1] = digest[1]; + key[2] = digest[2]; + key[3] = digest[3]; + key[4] = digest[4]; + + t0[0] = 0x5c5c5c5c ^ ctx.h[0]; + t0[1] = 0x5c5c5c5c ^ ctx.h[1]; + t0[2] = 0x5c5c5c5c ^ ctx.h[2]; + t0[3] = 0x5c5c5c5c ^ ctx.h[3]; + + t1[0] = 0x5c5c5c5c ^ ctx.h[4]; + t1[1] = 0x5c5c5c5c; + t1[2] = 0x5c5c5c5c; + t1[3] = 0x5c5c5c5c; + + t2[0] = 0x5c5c5c5c; + t2[1] = 0x5c5c5c5c; + t2[2] = 0x5c5c5c5c; + t2[3] = 0x5c5c5c5c; + + t3[0] = 0x5c5c5c5c; + t3[1] = 0x5c5c5c5c; + t3[2] = 0x5c5c5c5c; + t3[3] = 0x5c5c5c5c; + + digest[0] = SHA1M_A; + digest[1] = SHA1M_B; + digest[2] = SHA1M_C; + digest[3] = SHA1M_D; + digest[4] = SHA1M_E; + + sha1_transform (t0, t1, t2, t3, digest); + + t0[0] = 0x80000000; + t0[1] = 0; + t0[2] = 0; + t0[3] = 0; + + t1[0] = 0; + t1[1] = 0; + t1[2] = 0; + t1[3] = 0; + + t2[0] = 0; + t2[1] = 0; + t2[2] = 0; + t2[3] = 0; + + t3[0] = 0; + t3[1] = 0; + t3[2] = 0; + t3[3] = 64 * 8; + + sha1_transform (t0, t1, t2, t3, digest); + + key[5] = digest[0]; + key[6] = digest[1]; + key[7] = digest[2]; + + u32 iv[4]; + + iv[0] = esalt_bufs[digests_offset].data[28]; + iv[1] = esalt_bufs[digests_offset].data[29]; + iv[2] = esalt_bufs[digests_offset].data[30]; + iv[3] = esalt_bufs[digests_offset].data[31]; + + u32 data[4]; + + data[0] = esalt_bufs[digests_offset].data[32]; + data[1] = esalt_bufs[digests_offset].data[33]; + data[2] = esalt_bufs[digests_offset].data[34]; + data[3] = esalt_bufs[digests_offset].data[35]; + + #define KEYLEN 60 + + u32 ks[KEYLEN]; + + AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3); + + u32 out[4]; + + aes256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4); + + out[0] ^= iv[0]; + out[1] ^= iv[1]; + out[2] ^= iv[2]; + out[3] ^= iv[3]; + + if ((out[0] == 0x10101010) && + (out[1] == 0x10101010) && + (out[2] == 0x10101010) && + (out[3] == 0x10101010)) + { + if (atomic_inc (&hashes_shown[digests_offset]) == 0) + { + mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0); + } + } + } +} diff --git a/docs/changes.txt b/docs/changes.txt index cb0de7412..62e9732a7 100644 --- a/docs/changes.txt +++ b/docs/changes.txt @@ -64,6 +64,9 @@ - Added hash-mode: sha256($salt.$pass.$salt) - Added hash-mode: sha256(sha256_bin($pass)) - Added hash-mode: sha256(sha256($pass).$salt) +- Added hash-mode: SecureZIP AES-128 +- Added hash-mode: SecureZIP AES-192 +- Added hash-mode: SecureZIP AES-256 - Added hash-mode: SolarWinds Orion - Added hash-mode: Telegram Desktop App Passcode (PBKDF2-HMAC-SHA1) - Added hash-mode: Telegram Mobile App Passcode (SHA256) @@ -78,8 +81,9 @@ - Fixed buffer overflow in build_plain() function - Fixed buffer overflow in mp_add_cs_buf() function -- Fixed copy/paste error leading to invalid "Integer overflow detected in keyspace of mask" in attack-mode 6 and 7 - Fixed calculation of brain-session ID, only the first hash of the hashset was taken into account +- Fixed cleanup of password candidate buffers on GPU set from autotune in case -n was used +- Fixed copy/paste error leading to invalid "Integer overflow detected in keyspace of mask" in attack-mode 6 and 7 - Fixed cracking multiple Office hashes (modes 9500, 9600) with the same salt - Fixed cracking of Blockchain, My Wallet (V1 and V2) hashes with unexpected decrypted data - Fixed cracking of Cisco-PIX and Cisco-ASA MD5 passwords in mask-attack mode if mask > length 16 @@ -101,9 +105,9 @@ - Fixed race condition in maskfile mode by using a dedicated flag for restore execution - Fixed some memory leaks in case hashcat is shutting down due to some file error - Fixed some memory leaks in case mask-files are used in optimized mode +- Fixed --status-json to correctly escape certain characters in hashes - Fixed the 7-Zip parser to allow the entire supported range of encrypted and decrypted data lengths - Fixed the validation of the --brain-client-features command line argument (only values 1, 2 or 3 are allowed) -- Fixed --status-json to correctly escape certain characters in hashes ## ## Improvements @@ -118,7 +122,7 @@ - File handling: Print a truncation warning when an oversized line is detected - My Wallet: Added additional plaintext pattern used in newer versions - Office cracking: Support hash format with second block data for 40-bit oldoffice files (eliminates false positives) -- OpenCL Runtime: Added a warning if OpenCL runtime NEO, Beignet, POCL or MESA is detected and skip associated devices (override with --force) +- OpenCL Runtime: Added a warning if OpenCL runtime NEO, Beignet, POCL (v1.4 or older) or MESA is detected and skip associated devices (override with --force) - OpenCL Runtime: Disable OpenCL kernel cache on Apple for Intel CPU (throws CL_BUILD_PROGRAM_FAILURE for no reason) - OpenCL Runtime: Do not run shared- and constant-memory size checks if their memory type is of type global memory (typically CPU) - OpenCL Runtime: Improve ROCm detection and make sure to not confuse with recent AMDGPU drivers diff --git a/docs/readme.txt b/docs/readme.txt index 8c8febbd7..4787ba01a 100644 --- a/docs/readme.txt +++ b/docs/readme.txt @@ -297,6 +297,9 @@ NVIDIA GPUs require "NVIDIA Driver" (440.64 or later) and "CUDA Toolkit" (9.0 or - PKZIP (Uncompressed) - PKZIP Master Key - PKZIP Master Key (6 byte optimization) +- SecureZIP AES-128 +- SecureZIP AES-192 +- SecureZIP AES-256 - iTunes backup < 10.0 - iTunes backup >= 10.0 - WinZip @@ -344,6 +347,7 @@ NVIDIA GPUs require "NVIDIA Driver" (440.64 or later) and "CUDA Toolkit" (9.0 or - Apple - Intel - NVidia +- POCL ## ## Supported OpenCL device types diff --git a/include/types.h b/include/types.h index b610b46fa..2dc786fdb 100644 --- a/include/types.h +++ b/include/types.h @@ -539,6 +539,7 @@ typedef enum parser_rc PARSER_BLOCK_SIZE = -39, PARSER_CIPHER = -40, PARSER_FILE_SIZE = -41, + PARSER_HAVE_ERRNO = -100, PARSER_UNKNOWN_ERROR = -255 } parser_rc_t; diff --git a/src/autotune.c b/src/autotune.c index bcf1ac33b..0af2d3841 100644 --- a/src/autotune.c +++ b/src/autotune.c @@ -136,233 +136,227 @@ static int autotune (hashcat_ctx_t *hashcat_ctx, hc_device_param_t *device_param } #endif - - device_param->kernel_accel = kernel_accel; - device_param->kernel_loops = kernel_loops; - - const u32 kernel_power = device_param->hardware_power * device_param->kernel_accel; - - device_param->kernel_power = kernel_power; - - return 0; - } - - // from here it's clear we are allowed to autotune - // so let's init some fake words - - const u32 kernel_power_max = device_param->hardware_power * kernel_accel_max; - - int CL_rc; - int CU_rc; - - if (device_param->is_cuda == true) - { - CU_rc = run_cuda_kernel_atinit (hashcat_ctx, device_param, device_param->cuda_d_pws_buf, kernel_power_max); - - if (CU_rc == -1) return -1; - } - - if (device_param->is_opencl == true) - { - CL_rc = run_opencl_kernel_atinit (hashcat_ctx, device_param, device_param->opencl_d_pws_buf, kernel_power_max); - - if (CL_rc == -1) return -1; - } - - if (user_options->slow_candidates == true) - { } else { - if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) + // from here it's clear we are allowed to autotune + // so let's init some fake words + + const u32 kernel_power_max = device_param->hardware_power * kernel_accel_max; + + int CL_rc; + int CU_rc; + + if (device_param->is_cuda == true) { - if (straight_ctx->kernel_rules_cnt > 1) - { - if (device_param->is_cuda == true) - { - CU_rc = hc_cuMemcpyDtoD (hashcat_ctx, device_param->cuda_d_rules_c, device_param->cuda_d_rules, MIN (kernel_loops_max, KERNEL_RULES) * sizeof (kernel_rule_t)); + CU_rc = run_cuda_kernel_atinit (hashcat_ctx, device_param, device_param->cuda_d_pws_buf, kernel_power_max); - if (CU_rc == -1) return -1; - } - - if (device_param->is_opencl == true) - { - CL_rc = hc_clEnqueueCopyBuffer (hashcat_ctx, device_param->opencl_command_queue, device_param->opencl_d_rules, device_param->opencl_d_rules_c, 0, 0, MIN (kernel_loops_max, KERNEL_RULES) * sizeof (kernel_rule_t), 0, NULL, NULL); - - if (CL_rc == -1) return -1; - } - } - } - } - - // Do a pre-autotune test run to find out if kernel runtime is above some TDR limit - - u32 kernel_loops_max_reduced = kernel_loops_max; - - if (true) - { - double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops_min); - - if (exec_msec > 2000) - { - event_log_error (hashcat_ctx, "Kernel minimum runtime larger than default TDR"); - - return -1; + if (CU_rc == -1) return -1; } - exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops_min); - - const u32 mm = kernel_loops_max / kernel_loops_min; - - if ((exec_msec * mm) > target_msec) + if (device_param->is_opencl == true) { - const u32 loops_valid = (const u32) (target_msec / exec_msec); + CL_rc = run_opencl_kernel_atinit (hashcat_ctx, device_param, device_param->opencl_d_pws_buf, kernel_power_max); - kernel_loops_max_reduced = kernel_loops_min * loops_valid; + if (CL_rc == -1) return -1; } - } - // first find out highest kernel-loops that stays below target_msec - - if (kernel_loops_min < kernel_loops_max) - { - for (kernel_loops = kernel_loops_max; kernel_loops > kernel_loops_min; kernel_loops >>= 1) + if (user_options->slow_candidates == true) { - if (kernel_loops > kernel_loops_max_reduced) continue; - - double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops); - - if (exec_msec < target_msec) break; - } - } - - // now the same for kernel-accel but with the new kernel-loops from previous loop set - - #define STEPS_CNT 16 - - if (kernel_accel_min < kernel_accel_max) - { - for (int i = 0; i < STEPS_CNT; i++) - { - const u32 kernel_accel_try = 1U << i; - - if (kernel_accel_try < kernel_accel_min) continue; - if (kernel_accel_try > kernel_accel_max) break; - - double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_try, kernel_loops); - - if (exec_msec > target_msec) break; - - kernel_accel = kernel_accel_try; - } - } - - // now find the middle balance between kernel_accel and kernel_loops - // while respecting allowed ranges at the same time - - if (kernel_accel < kernel_loops) - { - const u32 kernel_accel_orig = kernel_accel; - const u32 kernel_loops_orig = kernel_loops; - - double exec_msec_prev = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); - - for (int i = 1; i < STEPS_CNT; i++) - { - const u32 kernel_accel_try = kernel_accel_orig * (1U << i); - const u32 kernel_loops_try = kernel_loops_orig / (1U << i); - - if (kernel_accel_try < kernel_accel_min) continue; - if (kernel_accel_try > kernel_accel_max) break; - - if (kernel_loops_try > kernel_loops_max) continue; - if (kernel_loops_try < kernel_loops_min) break; - - // do a real test - - const double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_try, kernel_loops_try); - - if (exec_msec_prev < exec_msec) break; - - exec_msec_prev = exec_msec; - - // so far, so good! save - - kernel_accel = kernel_accel_try; - kernel_loops = kernel_loops_try; - - // too much if the next test is true - - if (kernel_loops_try < kernel_accel_try) break; - } - } - - double exec_msec_pre_final = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); - - const u32 exec_left = (const u32) (target_msec / exec_msec_pre_final); - - const u32 accel_left = kernel_accel_max / kernel_accel; - - const u32 exec_accel_min = MIN (exec_left, accel_left); // we want that to be int - - if (exec_accel_min >= 1) - { - // this is safe to not overflow kernel_accel_max because of accel_left - - kernel_accel *= exec_accel_min; - } - - // start finding best thread count is easier. - // it's either the preferred or the maximum thread count - - /* - const u32 kernel_threads_min = device_param->kernel_threads_min; - const u32 kernel_threads_max = device_param->kernel_threads_max; - - if (kernel_threads_min < kernel_threads_max) - { - const double exec_msec_max = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); - - u32 preferred_threads = 0; - - if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) - { - if (hashconfig->opti_type & OPTI_TYPE_OPTIMIZED_KERNEL) - { - preferred_threads = device_param->kernel_preferred_wgs_multiple1; - } - else - { - preferred_threads = device_param->kernel_preferred_wgs_multiple4; - } } else { - preferred_threads = device_param->kernel_preferred_wgs_multiple2; - } - - if ((preferred_threads >= kernel_threads_min) && (preferred_threads <= kernel_threads_max)) - { - const double exec_msec_preferred = try_run_preferred (hashcat_ctx, device_param, kernel_accel, kernel_loops); - - if (exec_msec_preferred < exec_msec_max) + if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) { - device_param->kernel_threads = preferred_threads; + if (straight_ctx->kernel_rules_cnt > 1) + { + if (device_param->is_cuda == true) + { + CU_rc = hc_cuMemcpyDtoD (hashcat_ctx, device_param->cuda_d_rules_c, device_param->cuda_d_rules, MIN (kernel_loops_max, KERNEL_RULES) * sizeof (kernel_rule_t)); + + if (CU_rc == -1) return -1; + } + + if (device_param->is_opencl == true) + { + CL_rc = hc_clEnqueueCopyBuffer (hashcat_ctx, device_param->opencl_command_queue, device_param->opencl_d_rules, device_param->opencl_d_rules_c, 0, 0, MIN (kernel_loops_max, KERNEL_RULES) * sizeof (kernel_rule_t), 0, NULL, NULL); + + if (CL_rc == -1) return -1; + } + } } } + + // Do a pre-autotune test run to find out if kernel runtime is above some TDR limit + + u32 kernel_loops_max_reduced = kernel_loops_max; + + if (true) + { + double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops_min); + + if (exec_msec > 2000) + { + event_log_error (hashcat_ctx, "Kernel minimum runtime larger than default TDR"); + + return -1; + } + + exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops_min); + + const u32 mm = kernel_loops_max / kernel_loops_min; + + if ((exec_msec * mm) > target_msec) + { + const u32 loops_valid = (const u32) (target_msec / exec_msec); + + kernel_loops_max_reduced = kernel_loops_min * loops_valid; + } + } + + // first find out highest kernel-loops that stays below target_msec + + if (kernel_loops_min < kernel_loops_max) + { + for (kernel_loops = kernel_loops_max; kernel_loops > kernel_loops_min; kernel_loops >>= 1) + { + if (kernel_loops > kernel_loops_max_reduced) continue; + + double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_min, kernel_loops); + + if (exec_msec < target_msec) break; + } + } + + // now the same for kernel-accel but with the new kernel-loops from previous loop set + + #define STEPS_CNT 16 + + if (kernel_accel_min < kernel_accel_max) + { + for (int i = 0; i < STEPS_CNT; i++) + { + const u32 kernel_accel_try = 1U << i; + + if (kernel_accel_try < kernel_accel_min) continue; + if (kernel_accel_try > kernel_accel_max) break; + + double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_try, kernel_loops); + + if (exec_msec > target_msec) break; + + kernel_accel = kernel_accel_try; + } + } + + // now find the middle balance between kernel_accel and kernel_loops + // while respecting allowed ranges at the same time + + if (kernel_accel < kernel_loops) + { + const u32 kernel_accel_orig = kernel_accel; + const u32 kernel_loops_orig = kernel_loops; + + double exec_msec_prev = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); + + for (int i = 1; i < STEPS_CNT; i++) + { + const u32 kernel_accel_try = kernel_accel_orig * (1U << i); + const u32 kernel_loops_try = kernel_loops_orig / (1U << i); + + if (kernel_accel_try < kernel_accel_min) continue; + if (kernel_accel_try > kernel_accel_max) break; + + if (kernel_loops_try > kernel_loops_max) continue; + if (kernel_loops_try < kernel_loops_min) break; + + // do a real test + + const double exec_msec = try_run (hashcat_ctx, device_param, kernel_accel_try, kernel_loops_try); + + if (exec_msec_prev < exec_msec) break; + + exec_msec_prev = exec_msec; + + // so far, so good! save + + kernel_accel = kernel_accel_try; + kernel_loops = kernel_loops_try; + + // too much if the next test is true + + if (kernel_loops_try < kernel_accel_try) break; + } + } + + double exec_msec_pre_final = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); + + const u32 exec_left = (const u32) (target_msec / exec_msec_pre_final); + + const u32 accel_left = kernel_accel_max / kernel_accel; + + const u32 exec_accel_min = MIN (exec_left, accel_left); // we want that to be int + + if (exec_accel_min >= 1) + { + // this is safe to not overflow kernel_accel_max because of accel_left + + kernel_accel *= exec_accel_min; + } + + // start finding best thread count is easier. + // it's either the preferred or the maximum thread count + + /* + const u32 kernel_threads_min = device_param->kernel_threads_min; + const u32 kernel_threads_max = device_param->kernel_threads_max; + + if (kernel_threads_min < kernel_threads_max) + { + const double exec_msec_max = try_run (hashcat_ctx, device_param, kernel_accel, kernel_loops); + + u32 preferred_threads = 0; + + if (hashconfig->attack_exec == ATTACK_EXEC_INSIDE_KERNEL) + { + if (hashconfig->opti_type & OPTI_TYPE_OPTIMIZED_KERNEL) + { + preferred_threads = device_param->kernel_preferred_wgs_multiple1; + } + else + { + preferred_threads = device_param->kernel_preferred_wgs_multiple4; + } + } + else + { + preferred_threads = device_param->kernel_preferred_wgs_multiple2; + } + + if ((preferred_threads >= kernel_threads_min) && (preferred_threads <= kernel_threads_max)) + { + const double exec_msec_preferred = try_run_preferred (hashcat_ctx, device_param, kernel_accel, kernel_loops); + + if (exec_msec_preferred < exec_msec_max) + { + device_param->kernel_threads = preferred_threads; + } + } + } + */ } - */ + + // reset them fake words + // reset other buffers in case autotune cracked something if (device_param->is_cuda == true) { - // reset them fake words + int CU_rc; CU_rc = run_cuda_kernel_memset (hashcat_ctx, device_param, device_param->cuda_d_pws_buf, 0, device_param->size_pws); if (CU_rc == -1) return -1; - // reset other buffers in case autotune cracked something - CU_rc = run_cuda_kernel_memset (hashcat_ctx, device_param, device_param->cuda_d_plain_bufs, 0, device_param->size_plains); if (CU_rc == -1) return -1; @@ -378,14 +372,12 @@ static int autotune (hashcat_ctx_t *hashcat_ctx, hc_device_param_t *device_param if (device_param->is_opencl == true) { - // reset them fake words + int CL_rc; CL_rc = run_opencl_kernel_memset (hashcat_ctx, device_param, device_param->opencl_d_pws_buf, 0, device_param->size_pws); if (CL_rc == -1) return -1; - // reset other buffers in case autotune cracked something - CL_rc = run_opencl_kernel_memset (hashcat_ctx, device_param, device_param->opencl_d_plain_bufs, 0, device_param->size_plains); if (CL_rc == -1) return -1; diff --git a/src/backend.c b/src/backend.c index 8feaa2b4e..94f2df4ef 100644 --- a/src/backend.c +++ b/src/backend.c @@ -54,22 +54,16 @@ static int backend_ctx_find_alias_devices (hashcat_ctx_t *hashcat_ctx) { backend_ctx_t *backend_ctx = hashcat_ctx->backend_ctx; + // first identify all aliases + for (int backend_devices_cnt_src = 0; backend_devices_cnt_src < backend_ctx->backend_devices_cnt; backend_devices_cnt_src++) { hc_device_param_t *device_param_src = &backend_ctx->devices_param[backend_devices_cnt_src]; - if (device_param_src->skipped == true) continue; - - if (device_param_src->skipped_warning == true) continue; - for (int backend_devices_cnt_dst = backend_devices_cnt_src + 1; backend_devices_cnt_dst < backend_ctx->backend_devices_cnt; backend_devices_cnt_dst++) { hc_device_param_t *device_param_dst = &backend_ctx->devices_param[backend_devices_cnt_dst]; - if (device_param_dst->skipped == true) continue; - - if (device_param_dst->skipped_warning == true) continue; - if (is_same_device (device_param_src, device_param_dst) == false) continue; device_param_src->device_id_alias_buf[device_param_src->device_id_alias_cnt] = device_param_dst->device_id; @@ -77,18 +71,45 @@ static int backend_ctx_find_alias_devices (hashcat_ctx_t *hashcat_ctx) device_param_dst->device_id_alias_buf[device_param_dst->device_id_alias_cnt] = device_param_src->device_id; device_param_dst->device_id_alias_cnt++; + } + } - if (device_param_dst->is_opencl == true) + // find the alias to skip + + for (int backend_devices_pos = 0; backend_devices_pos < backend_ctx->backend_devices_cnt; backend_devices_pos++) + { + hc_device_param_t *backend_device = &backend_ctx->devices_param[backend_devices_pos]; + + if (backend_device->skipped == true) continue; + + if (backend_device->skipped_warning == true) continue; + + for (int device_id_alias_pos = 0; device_id_alias_pos < backend_device->device_id_alias_cnt; device_id_alias_pos++) + { + const int alias_pos = backend_device->device_id_alias_buf[device_id_alias_pos]; + + hc_device_param_t *alias_device = &backend_ctx->devices_param[alias_pos]; + + if (alias_device->skipped == true) continue; + + if (alias_device->skipped_warning == true) continue; + + // this lets CUDA devices survive over OpenCL + + if (alias_device->is_cuda == true) continue; + + // this lets native OpenCL runtime survive over generic OpenCL runtime + + if (alias_device->opencl_device_type & CL_DEVICE_TYPE_CPU) { - if (device_param_dst->skipped == false) - { - device_param_dst->skipped = true; - - backend_ctx->opencl_devices_active--; - - backend_ctx->backend_devices_active--; - } + if (alias_device->opencl_platform_vendor_id == alias_device->opencl_device_vendor_id) continue; } + + alias_device->skipped = true; + + backend_ctx->opencl_devices_active--; + + backend_ctx->backend_devices_active--; } } @@ -5972,23 +5993,61 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime) device_param->device_local_mem_size = device_local_mem_size; - // If there's both an Intel CPU and an AMD OpenCL runtime it's a tricky situation - // Both platforms support CPU device types and therefore both will try to use 100% of the physical resources - // This results in both utilizing it for 50% - // However, Intel has much better SIMD control over their own hardware - // It makes sense to give them full control over their own hardware + // older POCL version and older LLVM versions are known to fail compiling kernels + // we need to inform the user to update + // https://github.com/hashcat/hashcat/issues/2344 - if (opencl_device_type & CL_DEVICE_TYPE_CPU) + if (opencl_platform_vendor_id == VENDOR_ID_POCL) { - if (device_param->opencl_device_vendor_id == VENDOR_ID_AMD_USE_INTEL) - { - if (user_options->force == false) - { - if (user_options->quiet == false) event_log_warning (hashcat_ctx, "* Device #%u: Not a native Intel OpenCL runtime. Expect massive speed loss.", device_id + 1); - if (user_options->quiet == false) event_log_warning (hashcat_ctx, " You can use --force to override, but do not report related errors."); - if (user_options->quiet == false) event_log_warning (hashcat_ctx, NULL); + char *pocl_version_ptr = strstr (opencl_platform_version, "pocl "); + char *llvm_version_ptr = strstr (opencl_platform_version, "LLVM "); - device_param->skipped = true; + if ((pocl_version_ptr != NULL) && (llvm_version_ptr != NULL)) + { + bool pocl_skip = false; + + int pocl_maj = 0; + int pocl_min = 0; + + const int res1 = sscanf (pocl_version_ptr, "pocl %d.%d", &pocl_maj, &pocl_min); + + if (res1 == 2) + { + const int pocl_version = (pocl_maj * 100) + pocl_min; + + if (pocl_version < 105) + { + pocl_skip = true; + } + } + + int llvm_maj = 0; + int llvm_min = 0; + + const int res2 = sscanf (llvm_version_ptr, "LLVM %d.%d", &llvm_maj, &llvm_min); + + if (res2 == 2) + { + const int llvm_version = (llvm_maj * 100) + llvm_min; + + if (llvm_version < 900) + { + pocl_skip = true; + } + } + + if (pocl_skip == true) + { + if (user_options->force == false) + { + event_log_error (hashcat_ctx, "* Device #%u: Outdated POCL OpenCL driver detected!", device_id + 1); + + if (user_options->quiet == false) event_log_warning (hashcat_ctx, "This OpenCL driver has been marked as likely to fail kernel compilation or to produce false negatives."); + if (user_options->quiet == false) event_log_warning (hashcat_ctx, "You can use --force to override this, but do not report related errors."); + if (user_options->quiet == false) event_log_warning (hashcat_ctx, NULL); + + device_param->skipped = true; + } } } } @@ -6001,14 +6060,11 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime) || (strstr (opencl_device_version_lower, " neo")) || (strstr (opencl_device_version_lower, "beignet ")) || (strstr (opencl_device_version_lower, " beignet")) - || (strstr (opencl_device_version_lower, "pocl ")) - || (strstr (opencl_device_version_lower, " pocl")) || (strstr (opencl_device_version_lower, "mesa ")) || (strstr (opencl_device_version_lower, " mesa"))) { // NEO: https://github.com/hashcat/hashcat/issues/2342 // BEIGNET: https://github.com/hashcat/hashcat/issues/2243 - // POCL: https://github.com/hashcat/hashcat/issues/2344 // MESA: https://github.com/hashcat/hashcat/issues/2269 if (user_options->force == false) @@ -6262,7 +6318,9 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime) if (r == 2) { - if (version_maj >= 440) + // nvidia 441.x looks ok + + if (version_maj == 440) { if (version_min >= 64) { @@ -6541,10 +6599,10 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime) backend_ctx->backend_devices_cnt = cuda_devices_cnt + opencl_devices_cnt; backend_ctx->backend_devices_active = cuda_devices_active + opencl_devices_active; - // find duplicate devices (typically CUDA and OpenCL) + // find duplicate devices - if ((cuda_devices_cnt > 0) && (opencl_devices_cnt > 0)) - { + //if ((cuda_devices_cnt > 0) && (opencl_devices_cnt > 0)) + //{ // using force here enables both devices, which is the worst possible outcome // many users force by default, so this is not a good idea @@ -6552,7 +6610,7 @@ int backend_ctx_devices_init (hashcat_ctx_t *hashcat_ctx, const int comptime) //{ backend_ctx_find_alias_devices (hashcat_ctx); //{ - } + //} if (backend_ctx->backend_devices_active == 0) { diff --git a/src/brain.c b/src/brain.c index b39c4f16d..ed0fad804 100644 --- a/src/brain.c +++ b/src/brain.c @@ -1486,12 +1486,14 @@ bool brain_server_read_hash_dumps (brain_server_dbs_t *brain_server_dbs, const c { brain_server_dbs->hash_cnt = 0; + /* temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 if (chdir (path) == -1) { brain_logging (stderr, 0, "%s: %s\n", path, strerror (errno)); return false; } + */ DIR *dirp = opendir (path); @@ -1683,12 +1685,14 @@ bool brain_server_read_attack_dumps (brain_server_dbs_t *brain_server_dbs, const { brain_server_dbs->attack_cnt = 0; + /* temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 if (chdir (path) == -1) { brain_logging (stderr, 0, "%s: %s\n", path, strerror (errno)); return false; } + */ DIR *dirp = opendir (path); diff --git a/src/filehandling.c b/src/filehandling.c index 22a9e2aa6..598881077 100644 --- a/src/filehandling.c +++ b/src/filehandling.c @@ -72,7 +72,7 @@ bool hc_fopen (HCFILE *fp, const char *path, char *mode) { lseek (fd_tmp, 0, SEEK_SET); - if (read (fd_tmp, check, sizeof(check)) > 0) + if (read (fd_tmp, check, sizeof (check)) > 0) { if (check[0] == 0x1f && check[1] == 0x8b && check[2] == 0x08 && check[3] == 0x08) fp->is_gzip = true; if (check[0] == 0x50 && check[1] == 0x4b && check[2] == 0x03 && check[3] == 0x04) fp->is_zip = true; @@ -131,7 +131,47 @@ size_t hc_fread (void *ptr, size_t size, size_t nmemb, HCFILE *fp) } else { + #if defined (_WIN) + + // 4 GB fread () limit for windows systems ? + // see: https://social.msdn.microsoft.com/Forums/vstudio/en-US/7c913001-227e-439b-bf07-54369ba07994/fwrite-issues-with-large-data-write + + #define GIGABYTE (1024u * 1024u * 1024u) + + if (((size * nmemb) / GIGABYTE) < 4) + { + n = fread (ptr, size, nmemb, fp->pfp); + } + else + { + if ((size / GIGABYTE) > 3) return -1; + + size_t elements_max = (3u * GIGABYTE) / size; + size_t elements_left = nmemb; + + size_t off = 0; + + n = 0; + + while (elements_left > 0) + { + size_t elements_cur = elements_max; + + if (elements_left < elements_max) elements_cur = elements_left; + + size_t ret = fread (ptr + off, size, elements_cur, fp->pfp); + + if (ret != elements_cur) return -1; + + n += ret; + off += ret * size; + + elements_left -= ret; + } + } + #else n = fread (ptr, size, nmemb, fp->pfp); + #endif } return n; @@ -152,10 +192,48 @@ size_t hc_fwrite (const void *ptr, size_t size, size_t nmemb, HCFILE *fp) } else { - n = fwrite (ptr, size, nmemb, fp->pfp); - } + #if defined (_WIN) - if (n != nmemb) return -1; + // 4 GB fwrite () limit for windows systems ? + // see: https://social.msdn.microsoft.com/Forums/vstudio/en-US/7c913001-227e-439b-bf07-54369ba07994/fwrite-issues-with-large-data-write + + #define GIGABYTE (1024u * 1024u * 1024u) + + if (((size * nmemb) / GIGABYTE) < 4) + { + n = fwrite (ptr, size, nmemb, fp->pfp); + } + else + { + if ((size / GIGABYTE) > 3) return -1; + + size_t elements_max = (3u * GIGABYTE) / size; + size_t elements_left = nmemb; + + size_t off = 0; + + n = 0; + + while (elements_left > 0) + { + size_t elements_cur = elements_max; + + if (elements_left < elements_max) elements_cur = elements_left; + + size_t ret = fwrite (ptr + off, size, elements_cur, fp->pfp); + + if (ret != elements_cur) return -1; + + n += ret; + off += ret * size; + + elements_left -= ret; + } + } + #else + n = fwrite (ptr, size, nmemb, fp->pfp); + #endif + } return n; } diff --git a/src/folder.c b/src/folder.c index 9f1130f05..84fad8ef6 100644 --- a/src/folder.c +++ b/src/folder.c @@ -258,6 +258,7 @@ int folder_config_init (hashcat_ctx_t *hashcat_ctx, MAYBE_UNUSED const char *ins * A workaround is to chdir() to the OpenCL folder, * then compile the kernels, * then chdir() back to where we came from so we need to save it first + * - temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 */ char *cwd = (char *) hcmalloc (HCBUFSIZ_TINY); @@ -450,6 +451,7 @@ int folder_config_init (hashcat_ctx_t *hashcat_ctx, MAYBE_UNUSED const char *ins hcfree (cpath); //if (getenv ("TMP") == NULL) + /* temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 if (true) { char *tmp; @@ -458,6 +460,7 @@ int folder_config_init (hashcat_ctx_t *hashcat_ctx, MAYBE_UNUSED const char *ins putenv (tmp); } + */ #if defined (_WIN) diff --git a/src/hashes.c b/src/hashes.c index ab1d7c692..d8dd14d29 100644 --- a/src/hashes.c +++ b/src/hashes.c @@ -182,8 +182,8 @@ int save_hash (hashcat_ctx_t *hashcat_ctx) { event_log_error (hashcat_ctx, "%s: %s", new_hashfile, strerror (errno)); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return -1; } @@ -194,8 +194,8 @@ int save_hash (hashcat_ctx_t *hashcat_ctx) event_log_error (hashcat_ctx, "%s: %s", new_hashfile, strerror (errno)); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return -1; } @@ -256,8 +256,8 @@ int save_hash (hashcat_ctx_t *hashcat_ctx) event_log_error (hashcat_ctx, "%s: %s", new_hashfile, strerror (errno)); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return -1; } @@ -270,8 +270,8 @@ int save_hash (hashcat_ctx_t *hashcat_ctx) { event_log_error (hashcat_ctx, "Rename file '%s' to '%s': %s", hashfile, old_hashfile, strerror (errno)); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return -1; } @@ -282,16 +282,16 @@ int save_hash (hashcat_ctx_t *hashcat_ctx) { event_log_error (hashcat_ctx, "Rename file '%s' to '%s': %s", new_hashfile, hashfile, strerror (errno)); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return -1; } unlink (old_hashfile); - free (new_hashfile); - free (old_hashfile); + hcfree (new_hashfile); + hcfree (old_hashfile); return 0; } @@ -735,21 +735,28 @@ int hashes_init_stage1 (hashcat_ctx_t *hashcat_ctx) { const int binary_count = module_ctx->module_hash_binary_count (hashes); - if (binary_count == 0) + if (binary_count > 0) + { + hashes_avail = binary_count; + } + else if (binary_count == 0) { event_log_error (hashcat_ctx, "No hashes loaded."); return -1; } - - if (binary_count == -1) + else if (binary_count == PARSER_HAVE_ERRNO) { event_log_error (hashcat_ctx, "%s: %s", hashes->hashfile, strerror (errno)); return -1; } + else + { + event_log_error (hashcat_ctx, "%s: %s", hashes->hashfile, strerror (binary_count)); - hashes_avail = binary_count; + return -1; + } } else { @@ -1289,6 +1296,10 @@ int hashes_init_stage1 (hashcat_ctx_t *hashcat_ctx) hashes_cnt = hashes_parsed; } else if (hashes_parsed == 0) + { + event_log_warning (hashcat_ctx, "No hashes loaded."); + } + else if (hashes_parsed == PARSER_HAVE_ERRNO) { event_log_warning (hashcat_ctx, "Hashfile '%s': %s", hashes->hashfile, strerror (errno)); } diff --git a/src/hlfmt.c b/src/hlfmt.c index 68ea27c82..0e803445b 100644 --- a/src/hlfmt.c +++ b/src/hlfmt.c @@ -374,7 +374,7 @@ u32 hlfmt_detect (hashcat_ctx_t *hashcat_ctx, HCFILE *fp, u32 max_check) hashlist_format = i; } - free (formats_cnt); + hcfree (formats_cnt); return hashlist_format; } diff --git a/src/modules/module_02500.c b/src/modules/module_02500.c index 753636dc1..5ae58066e 100644 --- a/src/modules/module_02500.c +++ b/src/modules/module_02500.c @@ -376,7 +376,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); char *in = (char *) hcmalloc (sizeof (hccapx_t)); @@ -467,7 +467,7 @@ int module_hash_binary_count (MAYBE_UNUSED const hashes_t *hashes) { struct stat st; - if (stat (hashes->hashfile, &st) == -1) return -1; + if (stat (hashes->hashfile, &st) == -1) return (PARSER_HAVE_ERRNO); return st.st_size / sizeof (hccapx_t); } diff --git a/src/modules/module_02501.c b/src/modules/module_02501.c index 41aa86acd..0a5f349f8 100644 --- a/src/modules/module_02501.c +++ b/src/modules/module_02501.c @@ -351,7 +351,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); char *in = (char *) hcmalloc (sizeof (hccapx_t)); @@ -442,7 +442,7 @@ int module_hash_binary_count (MAYBE_UNUSED const hashes_t *hashes) { struct stat st; - if (stat (hashes->hashfile, &st) == -1) return -1; + if (stat (hashes->hashfile, &st) == -1) return (PARSER_HAVE_ERRNO); return st.st_size / sizeof (hccapx_t); } diff --git a/src/modules/module_06211.c b/src/modules/module_06211.c index ef63c4be4..f2502bac7 100644 --- a/src/modules/module_06211.c +++ b/src/modules/module_06211.c @@ -155,7 +155,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -199,7 +199,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06212.c b/src/modules/module_06212.c index a2f6fc4bf..6773c3d96 100644 --- a/src/modules/module_06212.c +++ b/src/modules/module_06212.c @@ -155,7 +155,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -199,7 +199,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06213.c b/src/modules/module_06213.c index 0cdbd43c5..98f68977a 100644 --- a/src/modules/module_06213.c +++ b/src/modules/module_06213.c @@ -153,7 +153,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -197,7 +197,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06221.c b/src/modules/module_06221.c index 2744b6b37..fb0f5ee41 100644 --- a/src/modules/module_06221.c +++ b/src/modules/module_06221.c @@ -137,7 +137,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -181,7 +181,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06222.c b/src/modules/module_06222.c index d4314e47f..30e283d7c 100644 --- a/src/modules/module_06222.c +++ b/src/modules/module_06222.c @@ -137,7 +137,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -181,7 +181,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06223.c b/src/modules/module_06223.c index ec19b3c56..7632a605e 100644 --- a/src/modules/module_06223.c +++ b/src/modules/module_06223.c @@ -135,7 +135,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -179,7 +179,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06231.c b/src/modules/module_06231.c index e86e09b69..8285937b2 100644 --- a/src/modules/module_06231.c +++ b/src/modules/module_06231.c @@ -136,7 +136,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -180,7 +180,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06232.c b/src/modules/module_06232.c index 847c2c3e9..abae8da11 100644 --- a/src/modules/module_06232.c +++ b/src/modules/module_06232.c @@ -136,7 +136,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -180,7 +180,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06233.c b/src/modules/module_06233.c index 080976adf..58464a6a0 100644 --- a/src/modules/module_06233.c +++ b/src/modules/module_06233.c @@ -136,7 +136,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -180,7 +180,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06241.c b/src/modules/module_06241.c index 4ff957c71..0f7c6dd3b 100644 --- a/src/modules/module_06241.c +++ b/src/modules/module_06241.c @@ -149,7 +149,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -193,7 +193,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06242.c b/src/modules/module_06242.c index 048df78b1..f9c0c59b8 100644 --- a/src/modules/module_06242.c +++ b/src/modules/module_06242.c @@ -149,7 +149,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -193,7 +193,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_06243.c b/src/modules/module_06243.c index 2ee532815..1a57a9a84 100644 --- a/src/modules/module_06243.c +++ b/src/modules/module_06243.c @@ -149,7 +149,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define TC_HEADER_SIZE 512 @@ -193,7 +193,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13711.c b/src/modules/module_13711.c index 2096bd458..711671689 100644 --- a/src/modules/module_13711.c +++ b/src/modules/module_13711.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13712.c b/src/modules/module_13712.c index 8a5f1cf13..caa648c94 100644 --- a/src/modules/module_13712.c +++ b/src/modules/module_13712.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13713.c b/src/modules/module_13713.c index 7a5360837..87921b4a3 100644 --- a/src/modules/module_13713.c +++ b/src/modules/module_13713.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13721.c b/src/modules/module_13721.c index 9680018f1..090b148c9 100644 --- a/src/modules/module_13721.c +++ b/src/modules/module_13721.c @@ -166,7 +166,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -210,7 +210,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13722.c b/src/modules/module_13722.c index 88b7c0862..3329b4fbf 100644 --- a/src/modules/module_13722.c +++ b/src/modules/module_13722.c @@ -166,7 +166,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -210,7 +210,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13723.c b/src/modules/module_13723.c index bf90814ed..e47dcf5ce 100644 --- a/src/modules/module_13723.c +++ b/src/modules/module_13723.c @@ -166,7 +166,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -210,7 +210,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13731.c b/src/modules/module_13731.c index ab2885331..f153a5304 100644 --- a/src/modules/module_13731.c +++ b/src/modules/module_13731.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13732.c b/src/modules/module_13732.c index 7ba12cfca..abb36e7b5 100644 --- a/src/modules/module_13732.c +++ b/src/modules/module_13732.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13733.c b/src/modules/module_13733.c index 2dfac5939..a27b3be7c 100644 --- a/src/modules/module_13733.c +++ b/src/modules/module_13733.c @@ -165,7 +165,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -209,7 +209,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13741.c b/src/modules/module_13741.c index e819dc268..bdcb6287a 100644 --- a/src/modules/module_13741.c +++ b/src/modules/module_13741.c @@ -167,7 +167,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -211,7 +211,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13742.c b/src/modules/module_13742.c index 42f09c216..325217e78 100644 --- a/src/modules/module_13742.c +++ b/src/modules/module_13742.c @@ -167,7 +167,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -211,7 +211,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13743.c b/src/modules/module_13743.c index 0c3e07a8b..c60db177e 100644 --- a/src/modules/module_13743.c +++ b/src/modules/module_13743.c @@ -167,7 +167,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -211,7 +211,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13751.c b/src/modules/module_13751.c index df3dab805..63a9b46ae 100644 --- a/src/modules/module_13751.c +++ b/src/modules/module_13751.c @@ -184,7 +184,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -228,7 +228,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13752.c b/src/modules/module_13752.c index c2675a0e1..02e07adad 100644 --- a/src/modules/module_13752.c +++ b/src/modules/module_13752.c @@ -184,7 +184,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -228,7 +228,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13753.c b/src/modules/module_13753.c index df061adce..6bb9d3213 100644 --- a/src/modules/module_13753.c +++ b/src/modules/module_13753.c @@ -184,7 +184,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -228,7 +228,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13761.c b/src/modules/module_13761.c index d121c9928..a6e386955 100644 --- a/src/modules/module_13761.c +++ b/src/modules/module_13761.c @@ -186,7 +186,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -230,7 +230,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13762.c b/src/modules/module_13762.c index a772f9180..545b7eeaa 100644 --- a/src/modules/module_13762.c +++ b/src/modules/module_13762.c @@ -186,7 +186,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -230,7 +230,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13763.c b/src/modules/module_13763.c index 885e9c1c6..2323a8921 100644 --- a/src/modules/module_13763.c +++ b/src/modules/module_13763.c @@ -186,7 +186,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -230,7 +230,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13771.c b/src/modules/module_13771.c index 21ab8fc58..0f754cfa4 100644 --- a/src/modules/module_13771.c +++ b/src/modules/module_13771.c @@ -169,7 +169,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -213,7 +213,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13772.c b/src/modules/module_13772.c index ce9a1edd8..699467afe 100644 --- a/src/modules/module_13772.c +++ b/src/modules/module_13772.c @@ -169,7 +169,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -213,7 +213,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_13773.c b/src/modules/module_13773.c index ed658c123..656b9a854 100644 --- a/src/modules/module_13773.c +++ b/src/modules/module_13773.c @@ -169,7 +169,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); #define VC_HEADER_SIZE 512 @@ -213,7 +213,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE keyfile = strtok_r ((char *) NULL, ",", &saveptr); } - free (keyfiles); + hcfree (keyfiles); } // keyboard layout mapping diff --git a/src/modules/module_14000.c b/src/modules/module_14000.c index e2dabbed1..8ccf0cdc3 100644 --- a/src/modules/module_14000.c +++ b/src/modules/module_14000.c @@ -178,6 +178,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[2] = 0; salt->salt_buf[3] = 0; + salt->salt_len = 8; + salt->salt_buf_pc[0] = salt->salt_buf[0]; salt->salt_buf_pc[1] = salt->salt_buf[1]; diff --git a/src/modules/module_14100.c b/src/modules/module_14100.c index 3449b6844..77ffe0a61 100644 --- a/src/modules/module_14100.c +++ b/src/modules/module_14100.c @@ -139,6 +139,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[2] = 0; salt->salt_buf[3] = 0; + salt->salt_len = 8; + salt->salt_buf_pc[0] = salt->salt_buf[0]; salt->salt_buf_pc[1] = salt->salt_buf[1]; diff --git a/src/modules/module_14600.c b/src/modules/module_14600.c index f506ef58a..9e4c488e7 100644 --- a/src/modules/module_14600.c +++ b/src/modules/module_14600.c @@ -26,8 +26,8 @@ static const u64 OPTS_TYPE = OPTS_TYPE_PT_GENERATE_LE | OPTS_TYPE_SELF_TEST_DISABLE | OPTS_TYPE_BINARY_HASHFILE; static const u32 SALT_TYPE = SALT_TYPE_EMBEDDED; -static const char *ST_PASS = "hashcat"; -static const char *ST_HASH = "tbd"; +static const char *ST_PASS = NULL; +static const char *ST_HASH = NULL; u32 module_attack_exec (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ATTACK_EXEC; } u32 module_dgst_pos0 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS0; } @@ -377,7 +377,7 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE HCFILE fp; - if (hc_fopen (&fp, (const char *) line_buf, "rb") == false) return (PARSER_HASH_FILE); + if (hc_fopen (&fp, (const char *) line_buf, "rb") == false) return (PARSER_HAVE_ERRNO); struct luks_phdr hdr; diff --git a/src/modules/module_15500.c b/src/modules/module_15500.c index 6e714bce3..f61e3b289 100644 --- a/src/modules/module_15500.c +++ b/src/modules/module_15500.c @@ -197,9 +197,7 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE // alias - const u8 *alias_pos = token.buf[6]; - - strncpy ((char *) jks_sha1->alias, (const char *) alias_pos, 64); + memcpy ((char *) jks_sha1->alias, (const char *) token.buf[6], token.len[6]); // fake salt @@ -237,6 +235,10 @@ int module_hash_encode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE u8 *der = (u8 *) jks_sha1->der; + char alias[65] = { 0 }; + + memcpy (alias, (char *) jks_sha1->alias, 64); + const int line_len = snprintf (line_buf, line_size, "%s*%08X%08X%08X%08X%08X*%08X%08X%08X%08X%08X*%s*%02X*%02X%02X%02X%02X%02X%02X%02X%02X%02X%02X%02X%02X%02X%02X*%s", SIGNATURE_JKS_SHA1, byte_swap_32 (jks_sha1->checksum[0]), @@ -265,7 +267,7 @@ int module_hash_encode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE der[17], der[18], der[19], - (char *) jks_sha1->alias + alias ); return line_len; diff --git a/src/modules/module_19600.c b/src/modules/module_19600.c index a96be303b..5f686a5be 100644 --- a/src/modules/module_19600.c +++ b/src/modules/module_19600.c @@ -242,6 +242,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[1] = krb5tgs->checksum[1]; salt->salt_buf[2] = krb5tgs->checksum[2]; + salt->salt_len = 12; + salt->salt_iter = 4096 - 1; digest[0] = krb5tgs->checksum[0]; diff --git a/src/modules/module_19700.c b/src/modules/module_19700.c index 52a4c1ef3..f1602593d 100644 --- a/src/modules/module_19700.c +++ b/src/modules/module_19700.c @@ -242,6 +242,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[1] = krb5tgs->checksum[1]; salt->salt_buf[2] = krb5tgs->checksum[2]; + salt->salt_len = 12; + salt->salt_iter = 4096 - 1; digest[0] = krb5tgs->checksum[0]; diff --git a/src/modules/module_19800.c b/src/modules/module_19800.c index 5d92c6fde..3bca20227 100644 --- a/src/modules/module_19800.c +++ b/src/modules/module_19800.c @@ -188,6 +188,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[1] = krb5pa->checksum[1]; salt->salt_buf[2] = krb5pa->checksum[2]; + salt->salt_len = 12; + salt->salt_iter = 4096 - 1; digest[0] = krb5pa->checksum[0]; diff --git a/src/modules/module_19900.c b/src/modules/module_19900.c index 6e355c4ac..117887cb9 100644 --- a/src/modules/module_19900.c +++ b/src/modules/module_19900.c @@ -188,6 +188,8 @@ int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSE salt->salt_buf[1] = krb5pa->checksum[1]; salt->salt_buf[2] = krb5pa->checksum[2]; + salt->salt_len = 12; + salt->salt_iter = 4096 - 1; digest[0] = krb5pa->checksum[0]; diff --git a/src/modules/module_22000.c b/src/modules/module_22000.c index b914157e4..32b9b5ffe 100644 --- a/src/modules/module_22000.c +++ b/src/modules/module_22000.c @@ -171,7 +171,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); const bool r = is_hccapx (&fp); @@ -258,7 +258,7 @@ int module_hash_binary_count (MAYBE_UNUSED const hashes_t *hashes) HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); const bool r = is_hccapx (&fp); diff --git a/src/modules/module_22001.c b/src/modules/module_22001.c index 9a3465266..7937d5211 100644 --- a/src/modules/module_22001.c +++ b/src/modules/module_22001.c @@ -172,7 +172,7 @@ int module_hash_binary_parse (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); const bool r = is_hccapx (&fp); @@ -259,7 +259,7 @@ int module_hash_binary_count (MAYBE_UNUSED const hashes_t *hashes) HCFILE fp; - if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return -1; + if (hc_fopen (&fp, hashes->hashfile, "rb") == false) return (PARSER_HAVE_ERRNO); const bool r = is_hccapx (&fp); diff --git a/src/modules/module_23001.c b/src/modules/module_23001.c new file mode 100644 index 000000000..0b85f1538 --- /dev/null +++ b/src/modules/module_23001.c @@ -0,0 +1,315 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#include "common.h" +#include "types.h" +#include "modules.h" +#include "bitops.h" +#include "convert.h" +#include "shared.h" + +static const u32 ATTACK_EXEC = ATTACK_EXEC_INSIDE_KERNEL; +static const u32 DGST_POS0 = 0; +static const u32 DGST_POS1 = 1; +static const u32 DGST_POS2 = 2; +static const u32 DGST_POS3 = 3; +static const u32 DGST_SIZE = DGST_SIZE_4_4; +static const u32 HASH_CATEGORY = HASH_CATEGORY_ARCHIVE; +static const char *HASH_NAME = "SecureZIP AES-128"; +static const u64 KERN_TYPE = 23001; +static const u32 OPTI_TYPE = OPTI_TYPE_ZERO_BYTE + | OPTI_TYPE_PRECOMPUTE_INIT + | OPTI_TYPE_NOT_ITERATED + | OPTI_TYPE_NOT_SALTED; +static const u64 OPTS_TYPE = OPTS_TYPE_PT_GENERATE_BE + | OPTS_TYPE_PT_ADD80 + | OPTS_TYPE_PT_ADDBITS15; +static const u32 SALT_TYPE = SALT_TYPE_EMBEDDED; +static const char *ST_PASS = "hashcat"; +static const char *ST_HASH = "$zip3$*0*1*128*0*b4630625c92b6e7848f6fd86*df2f62611b3d02d2c7e05a48dad57c7d93b0bac1362261ab533807afb69db856676aa6e350320130b5cbf27c55a48c0f75739654ac312f1cf5c37149557fc88a92c7e3dde8d23edd2b839036e88092a708b7e818bf1b6de92f0efb5cce184cceb11db6b3ca0527d0bdf1f1137ee6660d9890928cd80542ac1f439515519147c14d965b5ba107c6227f971e3e115170bf*0*0*0*file.txt"; + +u32 module_attack_exec (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ATTACK_EXEC; } +u32 module_dgst_pos0 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS0; } +u32 module_dgst_pos1 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS1; } +u32 module_dgst_pos2 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS2; } +u32 module_dgst_pos3 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS3; } +u32 module_dgst_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_SIZE; } +u32 module_hash_category (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_CATEGORY; } +const char *module_hash_name (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_NAME; } +u64 module_kern_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return KERN_TYPE; } +u32 module_opti_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTI_TYPE; } +u64 module_opts_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTS_TYPE; } +u32 module_salt_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return SALT_TYPE; } +const char *module_st_hash (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_HASH; } +const char *module_st_pass (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_PASS; } + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +static const char *SIGNATURE_SECUREZIP = "$zip3$"; + +u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) +{ + const u64 esalt_size = (const u64) sizeof (securezip_t); + + return esalt_size; +} + +int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED void *digest_buf, MAYBE_UNUSED salt_t *salt, MAYBE_UNUSED void *esalt_buf, MAYBE_UNUSED void *hook_salt_buf, MAYBE_UNUSED hashinfo_t *hash_info, const char *line_buf, MAYBE_UNUSED const int line_len) +{ + u32 *digest = (u32 *) digest_buf; + + securezip_t *securezip = (securezip_t *) esalt_buf; + + token_t token; + + token.token_cnt = 11; + + token.signatures_cnt = 1; + token.signatures_buf[0] = SIGNATURE_SECUREZIP; + + token.len_min[0] = 6; + token.len_max[0] = 6; + token.sep[0] = '*'; + token.attr[0] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_SIGNATURE; + + token.len_min[1] = 1; + token.len_max[1] = 1; + token.sep[1] = '*'; + token.attr[1] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[2] = 1; + token.len_max[2] = 1; + token.sep[2] = '*'; + token.attr[2] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[3] = 3; + token.len_max[3] = 3; + token.sep[3] = '*'; + token.attr[3] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[4] = 1; + token.len_max[4] = 1; + token.sep[4] = '*'; + token.attr[4] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[5] = 0; + token.len_max[5] = 32; + token.sep[5] = '*'; + token.attr[5] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[6] = 288; + token.len_max[6] = 288; + token.sep[6] = '*'; + token.attr[6] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[7] = 1; + token.len_max[7] = 1; + token.sep[7] = '*'; + token.attr[7] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[8] = 1; + token.len_max[8] = 1; + token.sep[8] = '*'; + token.attr[8] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[9] = 1; + token.len_max[9] = 1; + token.sep[9] = '*'; + token.attr[9] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[10] = 0; + token.len_max[10] = 64; + token.sep[10] = '*'; + token.attr[10] = TOKEN_ATTR_VERIFY_LENGTH; + + const int rc_tokenizer = input_tokenizer ((const u8 *) line_buf, line_len, &token); + + if (rc_tokenizer != PARSER_OK) return (rc_tokenizer); + + const u8 *version_pos = token.buf[ 1]; + const u8 *type_pos = token.buf[ 2]; + const u8 *bit_len_pos = token.buf[ 3]; + const u8 *unused1_pos = token.buf[ 4]; + const u8 *iv_pos = token.buf[ 5]; + const u8 *data_pos = token.buf[ 6]; + const u8 *unused2_pos = token.buf[ 7]; + const u8 *unused3_pos = token.buf[ 8]; + const u8 *unused4_pos = token.buf[ 9]; + const u8 *file_pos = token.buf[10]; + + if (version_pos[0] != '0') return (PARSER_HASH_ENCODING); // version 0 + if (type_pos[0] != '1') return (PARSER_HASH_ENCODING); // AES + + const u32 bit_len = hc_strtoul ((const char *) bit_len_pos, NULL, 10); + + if (bit_len != 128) return (PARSER_HASH_ENCODING); + + if (unused1_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused2_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused3_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused4_pos[0] != '0') return (PARSER_HASH_ENCODING); + + // IV: + + u8 *iv = (u8 *) securezip->iv; + + securezip->iv_len = hex_decode (iv_pos, token.len[5], iv); + + // data: + + u32 *data = securezip->data; + + hex_decode (data_pos, token.len[6], (u8 *) data); + + // file: + + u8 *file = (u8 *) securezip->file; + + memcpy (file, file_pos, token.len[10]); + + file[63] = 0; + + // fake salt: + + salt->salt_buf[0] = iv[0]; + salt->salt_buf[1] = iv[1]; + salt->salt_buf[2] = iv[2]; + salt->salt_buf[3] = iv[3]; + + salt->salt_len = 16; + + // fake digest: + + digest[0] = data[0]; + digest[1] = data[1]; + digest[2] = data[2]; + digest[3] = data[3]; + + return (PARSER_OK); +} + +int module_hash_encode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const void *digest_buf, MAYBE_UNUSED const salt_t *salt, MAYBE_UNUSED const void *esalt_buf, MAYBE_UNUSED const void *hook_salt_buf, MAYBE_UNUSED const hashinfo_t *hash_info, char *line_buf, MAYBE_UNUSED const int line_size) +{ + const securezip_t *securezip = (const securezip_t *) esalt_buf; + + // IV: + + u8 iv[33] = { 0 }; + + hex_encode ((u8 *) securezip->iv, securezip->iv_len, iv); + + // data: + + u8 data[289] = { 0 }; + + hex_encode ((u8 *) securezip->data, 144, data); + + // file: + + const u8 *file_ptr = (const u8 *) securezip->file; + + u8 file[65] = { 0 }; + + memcpy (file, file_ptr, 64); + + int out_len = snprintf (line_buf, line_size, "%s*0*1*128*0*%s*%s*0*0*0*%s", + SIGNATURE_SECUREZIP, + iv, + data, + file + ); + + return out_len; +} + +void module_init (module_ctx_t *module_ctx) +{ + module_ctx->module_context_size = MODULE_CONTEXT_SIZE_CURRENT; + module_ctx->module_interface_version = MODULE_INTERFACE_VERSION_CURRENT; + + module_ctx->module_attack_exec = module_attack_exec; + module_ctx->module_benchmark_esalt = MODULE_DEFAULT; + module_ctx->module_benchmark_hook_salt = MODULE_DEFAULT; + module_ctx->module_benchmark_mask = MODULE_DEFAULT; + module_ctx->module_benchmark_salt = MODULE_DEFAULT; + module_ctx->module_build_plain_postprocess = MODULE_DEFAULT; + module_ctx->module_deep_comp_kernel = MODULE_DEFAULT; + module_ctx->module_dgst_pos0 = module_dgst_pos0; + module_ctx->module_dgst_pos1 = module_dgst_pos1; + module_ctx->module_dgst_pos2 = module_dgst_pos2; + module_ctx->module_dgst_pos3 = module_dgst_pos3; + module_ctx->module_dgst_size = module_dgst_size; + module_ctx->module_dictstat_disable = MODULE_DEFAULT; + module_ctx->module_esalt_size = module_esalt_size; + module_ctx->module_extra_buffer_size = MODULE_DEFAULT; + module_ctx->module_extra_tmp_size = MODULE_DEFAULT; + module_ctx->module_forced_outfile_format = MODULE_DEFAULT; + module_ctx->module_hash_binary_count = MODULE_DEFAULT; + module_ctx->module_hash_binary_parse = MODULE_DEFAULT; + module_ctx->module_hash_binary_save = MODULE_DEFAULT; + module_ctx->module_hash_decode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_decode_zero_hash = MODULE_DEFAULT; + module_ctx->module_hash_decode = module_hash_decode; + module_ctx->module_hash_encode_status = MODULE_DEFAULT; + module_ctx->module_hash_encode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_encode = module_hash_encode; + module_ctx->module_hash_init_selftest = MODULE_DEFAULT; + module_ctx->module_hash_mode = MODULE_DEFAULT; + module_ctx->module_hash_category = module_hash_category; + module_ctx->module_hash_name = module_hash_name; + module_ctx->module_hashes_count_min = MODULE_DEFAULT; + module_ctx->module_hashes_count_max = MODULE_DEFAULT; + module_ctx->module_hlfmt_disable = MODULE_DEFAULT; + module_ctx->module_hook12 = MODULE_DEFAULT; + module_ctx->module_hook23 = MODULE_DEFAULT; + module_ctx->module_hook_salt_size = MODULE_DEFAULT; + module_ctx->module_hook_size = MODULE_DEFAULT; + module_ctx->module_jit_build_options = MODULE_DEFAULT; + module_ctx->module_jit_cache_disable = MODULE_DEFAULT; + module_ctx->module_kernel_accel_max = MODULE_DEFAULT; + module_ctx->module_kernel_accel_min = MODULE_DEFAULT; + module_ctx->module_kernel_loops_max = MODULE_DEFAULT; + module_ctx->module_kernel_loops_min = MODULE_DEFAULT; + module_ctx->module_kernel_threads_max = MODULE_DEFAULT; + module_ctx->module_kernel_threads_min = MODULE_DEFAULT; + module_ctx->module_kern_type = module_kern_type; + module_ctx->module_kern_type_dynamic = MODULE_DEFAULT; + module_ctx->module_opti_type = module_opti_type; + module_ctx->module_opts_type = module_opts_type; + module_ctx->module_outfile_check_disable = MODULE_DEFAULT; + module_ctx->module_outfile_check_nocomp = MODULE_DEFAULT; + module_ctx->module_potfile_custom_check = MODULE_DEFAULT; + module_ctx->module_potfile_disable = MODULE_DEFAULT; + module_ctx->module_potfile_keep_all_hashes = MODULE_DEFAULT; + module_ctx->module_pwdump_column = MODULE_DEFAULT; + module_ctx->module_pw_max = MODULE_DEFAULT; + module_ctx->module_pw_min = MODULE_DEFAULT; + module_ctx->module_salt_max = MODULE_DEFAULT; + module_ctx->module_salt_min = MODULE_DEFAULT; + module_ctx->module_salt_type = module_salt_type; + module_ctx->module_separator = MODULE_DEFAULT; + module_ctx->module_st_hash = module_st_hash; + module_ctx->module_st_pass = module_st_pass; + module_ctx->module_tmp_size = MODULE_DEFAULT; + module_ctx->module_unstable_warning = MODULE_DEFAULT; + module_ctx->module_warmup_disable = MODULE_DEFAULT; +} diff --git a/src/modules/module_23002.c b/src/modules/module_23002.c new file mode 100644 index 000000000..e25208ec5 --- /dev/null +++ b/src/modules/module_23002.c @@ -0,0 +1,315 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#include "common.h" +#include "types.h" +#include "modules.h" +#include "bitops.h" +#include "convert.h" +#include "shared.h" + +static const u32 ATTACK_EXEC = ATTACK_EXEC_INSIDE_KERNEL; +static const u32 DGST_POS0 = 0; +static const u32 DGST_POS1 = 1; +static const u32 DGST_POS2 = 2; +static const u32 DGST_POS3 = 3; +static const u32 DGST_SIZE = DGST_SIZE_4_4; +static const u32 HASH_CATEGORY = HASH_CATEGORY_ARCHIVE; +static const char *HASH_NAME = "SecureZIP AES-192"; +static const u64 KERN_TYPE = 23002; +static const u32 OPTI_TYPE = OPTI_TYPE_ZERO_BYTE + | OPTI_TYPE_PRECOMPUTE_INIT + | OPTI_TYPE_NOT_ITERATED + | OPTI_TYPE_NOT_SALTED; +static const u64 OPTS_TYPE = OPTS_TYPE_PT_GENERATE_BE + | OPTS_TYPE_PT_ADD80 + | OPTS_TYPE_PT_ADDBITS15; +static const u32 SALT_TYPE = SALT_TYPE_EMBEDDED; +static const char *ST_PASS = "hashcat"; +static const char *ST_HASH = "$zip3$*0*1*192*0*53ff2de8c280778e1e0ab997*603eb37dbab9ea109e2c405e37d8cae1ec89e1e0d0b9ce5bf55d1b571c343b6a3df35fe381c30249cb0738a9b956ba8e52dfc5552894296300446a771032776c811ff8a71d9bb3c4d6c37016c027e41fea2d157d5b0ce17804b1d7c1606b7c1121d37851bd705e001f2cd755bbf305966d129a17c1d48ff8e87cfa41f479090cd456527db7d1d43f9020ad8e73f851a5*0*0*0*file.txt"; + +u32 module_attack_exec (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ATTACK_EXEC; } +u32 module_dgst_pos0 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS0; } +u32 module_dgst_pos1 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS1; } +u32 module_dgst_pos2 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS2; } +u32 module_dgst_pos3 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS3; } +u32 module_dgst_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_SIZE; } +u32 module_hash_category (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_CATEGORY; } +const char *module_hash_name (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_NAME; } +u64 module_kern_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return KERN_TYPE; } +u32 module_opti_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTI_TYPE; } +u64 module_opts_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTS_TYPE; } +u32 module_salt_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return SALT_TYPE; } +const char *module_st_hash (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_HASH; } +const char *module_st_pass (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_PASS; } + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +static const char *SIGNATURE_SECUREZIP = "$zip3$"; + +u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) +{ + const u64 esalt_size = (const u64) sizeof (securezip_t); + + return esalt_size; +} + +int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED void *digest_buf, MAYBE_UNUSED salt_t *salt, MAYBE_UNUSED void *esalt_buf, MAYBE_UNUSED void *hook_salt_buf, MAYBE_UNUSED hashinfo_t *hash_info, const char *line_buf, MAYBE_UNUSED const int line_len) +{ + u32 *digest = (u32 *) digest_buf; + + securezip_t *securezip = (securezip_t *) esalt_buf; + + token_t token; + + token.token_cnt = 11; + + token.signatures_cnt = 1; + token.signatures_buf[0] = SIGNATURE_SECUREZIP; + + token.len_min[0] = 6; + token.len_max[0] = 6; + token.sep[0] = '*'; + token.attr[0] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_SIGNATURE; + + token.len_min[1] = 1; + token.len_max[1] = 1; + token.sep[1] = '*'; + token.attr[1] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[2] = 1; + token.len_max[2] = 1; + token.sep[2] = '*'; + token.attr[2] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[3] = 3; + token.len_max[3] = 3; + token.sep[3] = '*'; + token.attr[3] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[4] = 1; + token.len_max[4] = 1; + token.sep[4] = '*'; + token.attr[4] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[5] = 0; + token.len_max[5] = 32; + token.sep[5] = '*'; + token.attr[5] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[6] = 288; + token.len_max[6] = 288; + token.sep[6] = '*'; + token.attr[6] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[7] = 1; + token.len_max[7] = 1; + token.sep[7] = '*'; + token.attr[7] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[8] = 1; + token.len_max[8] = 1; + token.sep[8] = '*'; + token.attr[8] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[9] = 1; + token.len_max[9] = 1; + token.sep[9] = '*'; + token.attr[9] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[10] = 0; + token.len_max[10] = 64; + token.sep[10] = '*'; + token.attr[10] = TOKEN_ATTR_VERIFY_LENGTH; + + const int rc_tokenizer = input_tokenizer ((const u8 *) line_buf, line_len, &token); + + if (rc_tokenizer != PARSER_OK) return (rc_tokenizer); + + const u8 *version_pos = token.buf[ 1]; + const u8 *type_pos = token.buf[ 2]; + const u8 *bit_len_pos = token.buf[ 3]; + const u8 *unused1_pos = token.buf[ 4]; + const u8 *iv_pos = token.buf[ 5]; + const u8 *data_pos = token.buf[ 6]; + const u8 *unused2_pos = token.buf[ 7]; + const u8 *unused3_pos = token.buf[ 8]; + const u8 *unused4_pos = token.buf[ 9]; + const u8 *file_pos = token.buf[10]; + + if (version_pos[0] != '0') return (PARSER_HASH_ENCODING); // version 0 + if (type_pos[0] != '1') return (PARSER_HASH_ENCODING); // AES + + const u32 bit_len = hc_strtoul ((const char *) bit_len_pos, NULL, 10); + + if (bit_len != 192) return (PARSER_HASH_ENCODING); + + if (unused1_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused2_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused3_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused4_pos[0] != '0') return (PARSER_HASH_ENCODING); + + // IV: + + u8 *iv = (u8 *) securezip->iv; + + securezip->iv_len = hex_decode (iv_pos, token.len[5], iv); + + // data: + + u32 *data = securezip->data; + + hex_decode (data_pos, token.len[6], (u8 *) data); + + // file: + + u8 *file = (u8 *) securezip->file; + + memcpy (file, file_pos, token.len[10]); + + file[63] = 0; + + // fake salt: + + salt->salt_buf[0] = iv[0]; + salt->salt_buf[1] = iv[1]; + salt->salt_buf[2] = iv[2]; + salt->salt_buf[3] = iv[3]; + + salt->salt_len = 16; + + // fake digest: + + digest[0] = data[0]; + digest[1] = data[1]; + digest[2] = data[2]; + digest[3] = data[3]; + + return (PARSER_OK); +} + +int module_hash_encode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const void *digest_buf, MAYBE_UNUSED const salt_t *salt, MAYBE_UNUSED const void *esalt_buf, MAYBE_UNUSED const void *hook_salt_buf, MAYBE_UNUSED const hashinfo_t *hash_info, char *line_buf, MAYBE_UNUSED const int line_size) +{ + const securezip_t *securezip = (const securezip_t *) esalt_buf; + + // IV: + + u8 iv[33] = { 0 }; + + hex_encode ((u8 *) securezip->iv, securezip->iv_len, iv); + + // data: + + u8 data[289] = { 0 }; + + hex_encode ((u8 *) securezip->data, 144, data); + + // file: + + const u8 *file_ptr = (const u8 *) securezip->file; + + u8 file[65] = { 0 }; + + memcpy (file, file_ptr, 64); + + int out_len = snprintf (line_buf, line_size, "%s*0*1*192*0*%s*%s*0*0*0*%s", + SIGNATURE_SECUREZIP, + iv, + data, + file + ); + + return out_len; +} + +void module_init (module_ctx_t *module_ctx) +{ + module_ctx->module_context_size = MODULE_CONTEXT_SIZE_CURRENT; + module_ctx->module_interface_version = MODULE_INTERFACE_VERSION_CURRENT; + + module_ctx->module_attack_exec = module_attack_exec; + module_ctx->module_benchmark_esalt = MODULE_DEFAULT; + module_ctx->module_benchmark_hook_salt = MODULE_DEFAULT; + module_ctx->module_benchmark_mask = MODULE_DEFAULT; + module_ctx->module_benchmark_salt = MODULE_DEFAULT; + module_ctx->module_build_plain_postprocess = MODULE_DEFAULT; + module_ctx->module_deep_comp_kernel = MODULE_DEFAULT; + module_ctx->module_dgst_pos0 = module_dgst_pos0; + module_ctx->module_dgst_pos1 = module_dgst_pos1; + module_ctx->module_dgst_pos2 = module_dgst_pos2; + module_ctx->module_dgst_pos3 = module_dgst_pos3; + module_ctx->module_dgst_size = module_dgst_size; + module_ctx->module_dictstat_disable = MODULE_DEFAULT; + module_ctx->module_esalt_size = module_esalt_size; + module_ctx->module_extra_buffer_size = MODULE_DEFAULT; + module_ctx->module_extra_tmp_size = MODULE_DEFAULT; + module_ctx->module_forced_outfile_format = MODULE_DEFAULT; + module_ctx->module_hash_binary_count = MODULE_DEFAULT; + module_ctx->module_hash_binary_parse = MODULE_DEFAULT; + module_ctx->module_hash_binary_save = MODULE_DEFAULT; + module_ctx->module_hash_decode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_decode_zero_hash = MODULE_DEFAULT; + module_ctx->module_hash_decode = module_hash_decode; + module_ctx->module_hash_encode_status = MODULE_DEFAULT; + module_ctx->module_hash_encode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_encode = module_hash_encode; + module_ctx->module_hash_init_selftest = MODULE_DEFAULT; + module_ctx->module_hash_mode = MODULE_DEFAULT; + module_ctx->module_hash_category = module_hash_category; + module_ctx->module_hash_name = module_hash_name; + module_ctx->module_hashes_count_min = MODULE_DEFAULT; + module_ctx->module_hashes_count_max = MODULE_DEFAULT; + module_ctx->module_hlfmt_disable = MODULE_DEFAULT; + module_ctx->module_hook12 = MODULE_DEFAULT; + module_ctx->module_hook23 = MODULE_DEFAULT; + module_ctx->module_hook_salt_size = MODULE_DEFAULT; + module_ctx->module_hook_size = MODULE_DEFAULT; + module_ctx->module_jit_build_options = MODULE_DEFAULT; + module_ctx->module_jit_cache_disable = MODULE_DEFAULT; + module_ctx->module_kernel_accel_max = MODULE_DEFAULT; + module_ctx->module_kernel_accel_min = MODULE_DEFAULT; + module_ctx->module_kernel_loops_max = MODULE_DEFAULT; + module_ctx->module_kernel_loops_min = MODULE_DEFAULT; + module_ctx->module_kernel_threads_max = MODULE_DEFAULT; + module_ctx->module_kernel_threads_min = MODULE_DEFAULT; + module_ctx->module_kern_type = module_kern_type; + module_ctx->module_kern_type_dynamic = MODULE_DEFAULT; + module_ctx->module_opti_type = module_opti_type; + module_ctx->module_opts_type = module_opts_type; + module_ctx->module_outfile_check_disable = MODULE_DEFAULT; + module_ctx->module_outfile_check_nocomp = MODULE_DEFAULT; + module_ctx->module_potfile_custom_check = MODULE_DEFAULT; + module_ctx->module_potfile_disable = MODULE_DEFAULT; + module_ctx->module_potfile_keep_all_hashes = MODULE_DEFAULT; + module_ctx->module_pwdump_column = MODULE_DEFAULT; + module_ctx->module_pw_max = MODULE_DEFAULT; + module_ctx->module_pw_min = MODULE_DEFAULT; + module_ctx->module_salt_max = MODULE_DEFAULT; + module_ctx->module_salt_min = MODULE_DEFAULT; + module_ctx->module_salt_type = module_salt_type; + module_ctx->module_separator = MODULE_DEFAULT; + module_ctx->module_st_hash = module_st_hash; + module_ctx->module_st_pass = module_st_pass; + module_ctx->module_tmp_size = MODULE_DEFAULT; + module_ctx->module_unstable_warning = MODULE_DEFAULT; + module_ctx->module_warmup_disable = MODULE_DEFAULT; +} diff --git a/src/modules/module_23003.c b/src/modules/module_23003.c new file mode 100644 index 000000000..f09a242bd --- /dev/null +++ b/src/modules/module_23003.c @@ -0,0 +1,315 @@ +/** + * Author......: See docs/credits.txt + * License.....: MIT + */ + +#include "common.h" +#include "types.h" +#include "modules.h" +#include "bitops.h" +#include "convert.h" +#include "shared.h" + +static const u32 ATTACK_EXEC = ATTACK_EXEC_INSIDE_KERNEL; +static const u32 DGST_POS0 = 0; +static const u32 DGST_POS1 = 1; +static const u32 DGST_POS2 = 2; +static const u32 DGST_POS3 = 3; +static const u32 DGST_SIZE = DGST_SIZE_4_4; +static const u32 HASH_CATEGORY = HASH_CATEGORY_ARCHIVE; +static const char *HASH_NAME = "SecureZIP AES-256"; +static const u64 KERN_TYPE = 23003; +static const u32 OPTI_TYPE = OPTI_TYPE_ZERO_BYTE + | OPTI_TYPE_PRECOMPUTE_INIT + | OPTI_TYPE_NOT_ITERATED + | OPTI_TYPE_NOT_SALTED; +static const u64 OPTS_TYPE = OPTS_TYPE_PT_GENERATE_BE + | OPTS_TYPE_PT_ADD80 + | OPTS_TYPE_PT_ADDBITS15; +static const u32 SALT_TYPE = SALT_TYPE_EMBEDDED; +static const char *ST_PASS = "hashcat"; +static const char *ST_HASH = "$zip3$*0*1*256*0*39bff47df6152a0214d7a967*65ff418ffb3b1198cccdef0327c03750f328d6dd5287e00e4c467f33b92a6ef40a74bb11b5afad61a6c3c9b279d8bd7961e96af7b470c36fc186fd3cfe059107021c9dea0cf206692f727eeca71f18f5b0b6ee1f702b648bba01aa21c7b7f3f0f7d547838aad46868155a04214f22feef7b31d7a15e1abe6dba5e569c62ee640783bb4a54054c2c69e93ece9f1a2af9d*0*0*0*file.txt"; + +u32 module_attack_exec (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ATTACK_EXEC; } +u32 module_dgst_pos0 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS0; } +u32 module_dgst_pos1 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS1; } +u32 module_dgst_pos2 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS2; } +u32 module_dgst_pos3 (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_POS3; } +u32 module_dgst_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return DGST_SIZE; } +u32 module_hash_category (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_CATEGORY; } +const char *module_hash_name (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return HASH_NAME; } +u64 module_kern_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return KERN_TYPE; } +u32 module_opti_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTI_TYPE; } +u64 module_opts_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return OPTS_TYPE; } +u32 module_salt_type (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return SALT_TYPE; } +const char *module_st_hash (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_HASH; } +const char *module_st_pass (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) { return ST_PASS; } + +typedef struct securezip +{ + u32 data[36]; + u32 file[16]; + u32 iv[4]; + u32 iv_len; + +} securezip_t; + +static const char *SIGNATURE_SECUREZIP = "$zip3$"; + +u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra) +{ + const u64 esalt_size = (const u64) sizeof (securezip_t); + + return esalt_size; +} + +int module_hash_decode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED void *digest_buf, MAYBE_UNUSED salt_t *salt, MAYBE_UNUSED void *esalt_buf, MAYBE_UNUSED void *hook_salt_buf, MAYBE_UNUSED hashinfo_t *hash_info, const char *line_buf, MAYBE_UNUSED const int line_len) +{ + u32 *digest = (u32 *) digest_buf; + + securezip_t *securezip = (securezip_t *) esalt_buf; + + token_t token; + + token.token_cnt = 11; + + token.signatures_cnt = 1; + token.signatures_buf[0] = SIGNATURE_SECUREZIP; + + token.len_min[0] = 6; + token.len_max[0] = 6; + token.sep[0] = '*'; + token.attr[0] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_SIGNATURE; + + token.len_min[1] = 1; + token.len_max[1] = 1; + token.sep[1] = '*'; + token.attr[1] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[2] = 1; + token.len_max[2] = 1; + token.sep[2] = '*'; + token.attr[2] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[3] = 3; + token.len_max[3] = 3; + token.sep[3] = '*'; + token.attr[3] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[4] = 1; + token.len_max[4] = 1; + token.sep[4] = '*'; + token.attr[4] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[5] = 0; + token.len_max[5] = 32; + token.sep[5] = '*'; + token.attr[5] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[6] = 288; + token.len_max[6] = 288; + token.sep[6] = '*'; + token.attr[6] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_HEX; + + token.len_min[7] = 1; + token.len_max[7] = 1; + token.sep[7] = '*'; + token.attr[7] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[8] = 1; + token.len_max[8] = 1; + token.sep[8] = '*'; + token.attr[8] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[9] = 1; + token.len_max[9] = 1; + token.sep[9] = '*'; + token.attr[9] = TOKEN_ATTR_VERIFY_LENGTH + | TOKEN_ATTR_VERIFY_DIGIT; + + token.len_min[10] = 0; + token.len_max[10] = 64; + token.sep[10] = '*'; + token.attr[10] = TOKEN_ATTR_VERIFY_LENGTH; + + const int rc_tokenizer = input_tokenizer ((const u8 *) line_buf, line_len, &token); + + if (rc_tokenizer != PARSER_OK) return (rc_tokenizer); + + const u8 *version_pos = token.buf[ 1]; + const u8 *type_pos = token.buf[ 2]; + const u8 *bit_len_pos = token.buf[ 3]; + const u8 *unused1_pos = token.buf[ 4]; + const u8 *iv_pos = token.buf[ 5]; + const u8 *data_pos = token.buf[ 6]; + const u8 *unused2_pos = token.buf[ 7]; + const u8 *unused3_pos = token.buf[ 8]; + const u8 *unused4_pos = token.buf[ 9]; + const u8 *file_pos = token.buf[10]; + + if (version_pos[0] != '0') return (PARSER_HASH_ENCODING); // version 0 + if (type_pos[0] != '1') return (PARSER_HASH_ENCODING); // AES + + const u32 bit_len = hc_strtoul ((const char *) bit_len_pos, NULL, 10); + + if (bit_len != 256) return (PARSER_HASH_ENCODING); + + if (unused1_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused2_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused3_pos[0] != '0') return (PARSER_HASH_ENCODING); + if (unused4_pos[0] != '0') return (PARSER_HASH_ENCODING); + + // IV: + + u8 *iv = (u8 *) securezip->iv; + + securezip->iv_len = hex_decode (iv_pos, token.len[5], iv); + + // data: + + u32 *data = securezip->data; + + hex_decode (data_pos, token.len[6], (u8 *) data); + + // file: + + u8 *file = (u8 *) securezip->file; + + memcpy (file, file_pos, token.len[10]); + + file[63] = 0; + + // fake salt: + + salt->salt_buf[0] = iv[0]; + salt->salt_buf[1] = iv[1]; + salt->salt_buf[2] = iv[2]; + salt->salt_buf[3] = iv[3]; + + salt->salt_len = 16; + + // fake digest: + + digest[0] = data[0]; + digest[1] = data[1]; + digest[2] = data[2]; + digest[3] = data[3]; + + return (PARSER_OK); +} + +int module_hash_encode (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const void *digest_buf, MAYBE_UNUSED const salt_t *salt, MAYBE_UNUSED const void *esalt_buf, MAYBE_UNUSED const void *hook_salt_buf, MAYBE_UNUSED const hashinfo_t *hash_info, char *line_buf, MAYBE_UNUSED const int line_size) +{ + const securezip_t *securezip = (const securezip_t *) esalt_buf; + + // IV: + + u8 iv[33] = { 0 }; + + hex_encode ((u8 *) securezip->iv, securezip->iv_len, iv); + + // data: + + u8 data[289] = { 0 }; + + hex_encode ((u8 *) securezip->data, 144, data); + + // file: + + const u8 *file_ptr = (const u8 *) securezip->file; + + u8 file[65] = { 0 }; + + memcpy (file, file_ptr, 64); + + int out_len = snprintf (line_buf, line_size, "%s*0*1*256*0*%s*%s*0*0*0*%s", + SIGNATURE_SECUREZIP, + iv, + data, + file + ); + + return out_len; +} + +void module_init (module_ctx_t *module_ctx) +{ + module_ctx->module_context_size = MODULE_CONTEXT_SIZE_CURRENT; + module_ctx->module_interface_version = MODULE_INTERFACE_VERSION_CURRENT; + + module_ctx->module_attack_exec = module_attack_exec; + module_ctx->module_benchmark_esalt = MODULE_DEFAULT; + module_ctx->module_benchmark_hook_salt = MODULE_DEFAULT; + module_ctx->module_benchmark_mask = MODULE_DEFAULT; + module_ctx->module_benchmark_salt = MODULE_DEFAULT; + module_ctx->module_build_plain_postprocess = MODULE_DEFAULT; + module_ctx->module_deep_comp_kernel = MODULE_DEFAULT; + module_ctx->module_dgst_pos0 = module_dgst_pos0; + module_ctx->module_dgst_pos1 = module_dgst_pos1; + module_ctx->module_dgst_pos2 = module_dgst_pos2; + module_ctx->module_dgst_pos3 = module_dgst_pos3; + module_ctx->module_dgst_size = module_dgst_size; + module_ctx->module_dictstat_disable = MODULE_DEFAULT; + module_ctx->module_esalt_size = module_esalt_size; + module_ctx->module_extra_buffer_size = MODULE_DEFAULT; + module_ctx->module_extra_tmp_size = MODULE_DEFAULT; + module_ctx->module_forced_outfile_format = MODULE_DEFAULT; + module_ctx->module_hash_binary_count = MODULE_DEFAULT; + module_ctx->module_hash_binary_parse = MODULE_DEFAULT; + module_ctx->module_hash_binary_save = MODULE_DEFAULT; + module_ctx->module_hash_decode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_decode_zero_hash = MODULE_DEFAULT; + module_ctx->module_hash_decode = module_hash_decode; + module_ctx->module_hash_encode_status = MODULE_DEFAULT; + module_ctx->module_hash_encode_potfile = MODULE_DEFAULT; + module_ctx->module_hash_encode = module_hash_encode; + module_ctx->module_hash_init_selftest = MODULE_DEFAULT; + module_ctx->module_hash_mode = MODULE_DEFAULT; + module_ctx->module_hash_category = module_hash_category; + module_ctx->module_hash_name = module_hash_name; + module_ctx->module_hashes_count_min = MODULE_DEFAULT; + module_ctx->module_hashes_count_max = MODULE_DEFAULT; + module_ctx->module_hlfmt_disable = MODULE_DEFAULT; + module_ctx->module_hook12 = MODULE_DEFAULT; + module_ctx->module_hook23 = MODULE_DEFAULT; + module_ctx->module_hook_salt_size = MODULE_DEFAULT; + module_ctx->module_hook_size = MODULE_DEFAULT; + module_ctx->module_jit_build_options = MODULE_DEFAULT; + module_ctx->module_jit_cache_disable = MODULE_DEFAULT; + module_ctx->module_kernel_accel_max = MODULE_DEFAULT; + module_ctx->module_kernel_accel_min = MODULE_DEFAULT; + module_ctx->module_kernel_loops_max = MODULE_DEFAULT; + module_ctx->module_kernel_loops_min = MODULE_DEFAULT; + module_ctx->module_kernel_threads_max = MODULE_DEFAULT; + module_ctx->module_kernel_threads_min = MODULE_DEFAULT; + module_ctx->module_kern_type = module_kern_type; + module_ctx->module_kern_type_dynamic = MODULE_DEFAULT; + module_ctx->module_opti_type = module_opti_type; + module_ctx->module_opts_type = module_opts_type; + module_ctx->module_outfile_check_disable = MODULE_DEFAULT; + module_ctx->module_outfile_check_nocomp = MODULE_DEFAULT; + module_ctx->module_potfile_custom_check = MODULE_DEFAULT; + module_ctx->module_potfile_disable = MODULE_DEFAULT; + module_ctx->module_potfile_keep_all_hashes = MODULE_DEFAULT; + module_ctx->module_pwdump_column = MODULE_DEFAULT; + module_ctx->module_pw_max = MODULE_DEFAULT; + module_ctx->module_pw_min = MODULE_DEFAULT; + module_ctx->module_salt_max = MODULE_DEFAULT; + module_ctx->module_salt_min = MODULE_DEFAULT; + module_ctx->module_salt_type = module_salt_type; + module_ctx->module_separator = MODULE_DEFAULT; + module_ctx->module_st_hash = module_st_hash; + module_ctx->module_st_pass = module_st_pass; + module_ctx->module_tmp_size = MODULE_DEFAULT; + module_ctx->module_unstable_warning = MODULE_DEFAULT; + module_ctx->module_warmup_disable = MODULE_DEFAULT; +} diff --git a/src/shared.c b/src/shared.c index 11fde9771..761ce03eb 100644 --- a/src/shared.c +++ b/src/shared.c @@ -511,7 +511,7 @@ void setup_environment_variables (const folder_config_t *folder_config) putenv (display); - free (display); + hcfree (display); } else { diff --git a/src/status.c b/src/status.c index 26407519e..ebae643cc 100644 --- a/src/status.c +++ b/src/status.c @@ -323,7 +323,7 @@ char *status_get_hash_target (const hashcat_ctx_t *hashcat_ctx) tmp_buf2[tmp_len] = 0; - free (tmp_buf); + hcfree (tmp_buf); return tmp_buf2; } @@ -343,7 +343,7 @@ char *status_get_hash_target (const hashcat_ctx_t *hashcat_ctx) char *tmp_buf2 = strdup (tmp_buf); - free (tmp_buf); + hcfree (tmp_buf); return tmp_buf2; } @@ -1097,14 +1097,14 @@ char *status_get_time_estimated_relative (const hashcat_ctx_t *hashcat_ctx) snprintf (display, HCBUFSIZ_TINY, "%s; Runtime limited: %s", tmp_display, display_left); - free (display_left); + hcfree (display_left); } else { snprintf (display, HCBUFSIZ_TINY, "%s; Runtime limit exceeded", tmp_display); } - free (tmp_display); + hcfree (tmp_display); } return display; diff --git a/src/terminal.c b/src/terminal.c index fb68608a7..411c5b36f 100644 --- a/src/terminal.c +++ b/src/terminal.c @@ -1068,7 +1068,7 @@ void status_display_status_json (hashcat_ctx_t *hashcat_ctx) printf (" \"rejected\": %" PRIu64 ",", hashcat_status->progress_rejected); printf (" \"devices\": ["); - free (target_json_encoded); + hcfree (target_json_encoded); int device_num = 0; diff --git a/src/user_options.c b/src/user_options.c index ec19a0354..544abfc0c 100644 --- a/src/user_options.c +++ b/src/user_options.c @@ -2760,12 +2760,14 @@ int user_options_check_files (hashcat_ctx_t *hashcat_ctx) * default building options */ + /* temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 if (chdir (folder_config->cpath_real) == -1) { event_log_error (hashcat_ctx, "%s: %s", folder_config->cpath_real, strerror (errno)); return -1; } + */ // include check // this test needs to be done manually because of macOS opencl runtime @@ -2792,9 +2794,13 @@ int user_options_check_files (hashcat_ctx_t *hashcat_ctx) for (int i = 0; files_names[i] != NULL; i++) { - if (hc_path_read (files_names[i]) == false) + char *temp_filename = NULL; + + hc_asprintf (&temp_filename, "%s/%s", folder_config->cpath_real, files_names[i]); + + if (hc_path_read (temp_filename) == false) { - event_log_error (hashcat_ctx, "%s: %s", files_names[i], strerror (errno)); + event_log_error (hashcat_ctx, "%s: %s", temp_filename, strerror (errno)); return -1; } @@ -2802,6 +2808,7 @@ int user_options_check_files (hashcat_ctx_t *hashcat_ctx) // return back to the folder we came from initially (workaround) + /* temporary disabled due to https://github.com/hashcat/hashcat/issues/2379 #if defined (_WIN) if (chdir ("..") == -1) { @@ -2817,6 +2824,7 @@ int user_options_check_files (hashcat_ctx_t *hashcat_ctx) return -1; } #endif + */ return 0; } diff --git a/tools/test_modules/m23001.pm b/tools/test_modules/m23001.pm new file mode 100644 index 000000000..836eaa33e --- /dev/null +++ b/tools/test_modules/m23001.pm @@ -0,0 +1,193 @@ +#!/usr/bin/env perl + +## +## Author......: See docs/credits.txt +## License.....: MIT +## + +use strict; +use warnings; + +use Digest::SHA qw (sha1); +use Crypt::CBC; + +sub module_constraints { [[0, 256], [-1, -1], [0, 55], [-1, -1], [-1, -1]] } + +sub module_generate_hash +{ + my $word = shift; + my $salt = shift; # unused since unsalted algo + my $iv = shift; + my $data = shift; + my $file = shift; + + my $bit_len = 128; + my $key_len = $bit_len / 8; + + my $is_decrypt = defined ($data); + + my $padding = "none"; # for decryption we need this to "keep" the padding bytes + + if ($is_decrypt == 0) + { + $padding = "standard"; + + # generate some additional random hash data: + + my $iv_len = random_number (1, 16); + + $iv = random_bytes ($iv_len); + + $data = random_bytes (128); + + $file = random_lowercase_string (random_number (1, 16)); + + $file .= ".txt"; + } + + my $iv_mod = $iv; + + $iv_mod .= "\x00" x (16 - length ($iv_mod)); + + + # start of main algo: + + my $digest = sha1 ($word); + + my $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x36")); # or just ^ 0x36 + } + + $buf .= "\x36" x 44; + + my $key = sha1 ($buf); + + + $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x5c")); # or just ^ 0x36 + } + + $buf .= "\x5c" x 44; + + # final key: + + $key = $key . sha1 ($buf); + + $key = substr ($key, 0, $key_len); + + my $aes = Crypt::CBC->new ({ + cipher => "Crypt::Rijndael", + key => $key, + iv => $iv_mod, + keysize => $key_len, + literal_key => 1, + header => "none", + padding => $padding, + }); + + if ($is_decrypt == 0) + { + $data = $aes->encrypt ($data); + } + else + { + my $data_decrypted = $aes->decrypt ($data); + + # the password is wrong if the decrypted data does not have the expected padding bytes: + + if (substr ($data_decrypted, -16) ne "\x10" x 16) + { + $data = "fake"; # fake data + } + } + + my $iv_padded = $iv; + + if (length ($iv_padded) < 12) + { + $iv_padded .= "\x00" x (12 - length ($iv_padded)); + } + + my $hash = sprintf ("\$zip3\$*0*1*%i*0*%s*%s*0*0*0*%s", $bit_len, unpack ("H*", $iv_padded), unpack ("H*", $data), $file); + + return $hash; +} + +sub module_verify_hash +{ + my $line = shift; + + return unless (substr ($line, 0, 11) eq "\$zip3\$*0*1*"); + + my $idx1 = index ($line, ":"); + + return unless ($idx1 >= 11); + + my $hash = substr ($line, 0, $idx1); + my $word = substr ($line, $idx1 + 1); + + # bit_len: + + $idx1 = index ($hash, "*", 11); + + return unless ($idx1 > 0); + + my $bit_len = substr ($hash, 11, $idx1 - 11); + + $bit_len = int ($bit_len); + + return unless ($bit_len == 128); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 2) eq "0*"); + + # iv: + + my $idx2 = index ($hash, "*", $idx1 + 3); + + return unless ($idx2 > 0); + + my $iv = substr ($hash, $idx1 + 3, $idx2 - $idx1 - 3); + + return unless ($iv =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($iv) % 2) == 0); + + # data: + + $idx1 = index ($hash, "*", $idx2 + 1); + + return unless ($idx1 > 0); + + my $data = substr ($hash, $idx2 + 1, $idx1 - $idx2 - 1); + + return unless ($data =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($data) % 2) == 0); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 6) eq "0*0*0*"); + + # file: + + my $file = substr ($hash, $idx1 + 7); + + # convert to hex: + + $iv = pack ("H*", $iv); + $data = pack ("H*", $data); + + my $word_packed = pack_if_HEX_notation ($word); + + my $new_hash = module_generate_hash ($word_packed, "", $iv, $data, $file); + + return ($new_hash, $word); +} + +1; diff --git a/tools/test_modules/m23002.pm b/tools/test_modules/m23002.pm new file mode 100644 index 000000000..882bdc3ea --- /dev/null +++ b/tools/test_modules/m23002.pm @@ -0,0 +1,193 @@ +#!/usr/bin/env perl + +## +## Author......: See docs/credits.txt +## License.....: MIT +## + +use strict; +use warnings; + +use Digest::SHA qw (sha1); +use Crypt::CBC; + +sub module_constraints { [[0, 256], [-1, -1], [0, 55], [-1, -1], [-1, -1]] } + +sub module_generate_hash +{ + my $word = shift; + my $salt = shift; # unused since unsalted algo + my $iv = shift; + my $data = shift; + my $file = shift; + + my $bit_len = 192; + my $key_len = $bit_len / 8; + + my $is_decrypt = defined ($data); + + my $padding = "none"; # for decryption we need this to "keep" the padding bytes + + if ($is_decrypt == 0) + { + $padding = "standard"; + + # generate some additional random hash data: + + my $iv_len = random_number (1, 16); + + $iv = random_bytes ($iv_len); + + $data = random_bytes (128); + + $file = random_lowercase_string (random_number (1, 16)); + + $file .= ".txt"; + } + + my $iv_mod = $iv; + + $iv_mod .= "\x00" x (16 - length ($iv_mod)); + + + # start of main algo: + + my $digest = sha1 ($word); + + my $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x36")); # or just ^ 0x36 + } + + $buf .= "\x36" x 44; + + my $key = sha1 ($buf); + + + $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x5c")); # or just ^ 0x36 + } + + $buf .= "\x5c" x 44; + + # final key: + + $key = $key . sha1 ($buf); + + $key = substr ($key, 0, $key_len); + + my $aes = Crypt::CBC->new ({ + cipher => "Crypt::Rijndael", + key => $key, + iv => $iv_mod, + keysize => $key_len, + literal_key => 1, + header => "none", + padding => $padding, + }); + + if ($is_decrypt == 0) + { + $data = $aes->encrypt ($data); + } + else + { + my $data_decrypted = $aes->decrypt ($data); + + # the password is wrong if the decrypted data does not have the expected padding bytes: + + if (substr ($data_decrypted, -16) ne "\x10" x 16) + { + $data = "fake"; # fake data + } + } + + my $iv_padded = $iv; + + if (length ($iv_padded) < 12) + { + $iv_padded .= "\x00" x (12 - length ($iv_padded)); + } + + my $hash = sprintf ("\$zip3\$*0*1*%i*0*%s*%s*0*0*0*%s", $bit_len, unpack ("H*", $iv_padded), unpack ("H*", $data), $file); + + return $hash; +} + +sub module_verify_hash +{ + my $line = shift; + + return unless (substr ($line, 0, 11) eq "\$zip3\$*0*1*"); + + my $idx1 = index ($line, ":"); + + return unless ($idx1 >= 11); + + my $hash = substr ($line, 0, $idx1); + my $word = substr ($line, $idx1 + 1); + + # bit_len: + + $idx1 = index ($hash, "*", 11); + + return unless ($idx1 > 0); + + my $bit_len = substr ($hash, 11, $idx1 - 11); + + $bit_len = int ($bit_len); + + return unless ($bit_len == 192); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 2) eq "0*"); + + # iv: + + my $idx2 = index ($hash, "*", $idx1 + 3); + + return unless ($idx2 > 0); + + my $iv = substr ($hash, $idx1 + 3, $idx2 - $idx1 - 3); + + return unless ($iv =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($iv) % 2) == 0); + + # data: + + $idx1 = index ($hash, "*", $idx2 + 1); + + return unless ($idx1 > 0); + + my $data = substr ($hash, $idx2 + 1, $idx1 - $idx2 - 1); + + return unless ($data =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($data) % 2) == 0); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 6) eq "0*0*0*"); + + # file: + + my $file = substr ($hash, $idx1 + 7); + + # convert to hex: + + $iv = pack ("H*", $iv); + $data = pack ("H*", $data); + + my $word_packed = pack_if_HEX_notation ($word); + + my $new_hash = module_generate_hash ($word_packed, "", $iv, $data, $file); + + return ($new_hash, $word); +} + +1; diff --git a/tools/test_modules/m23003.pm b/tools/test_modules/m23003.pm new file mode 100644 index 000000000..1155b1483 --- /dev/null +++ b/tools/test_modules/m23003.pm @@ -0,0 +1,193 @@ +#!/usr/bin/env perl + +## +## Author......: See docs/credits.txt +## License.....: MIT +## + +use strict; +use warnings; + +use Digest::SHA qw (sha1); +use Crypt::CBC; + +sub module_constraints { [[0, 256], [-1, -1], [0, 55], [-1, -1], [-1, -1]] } + +sub module_generate_hash +{ + my $word = shift; + my $salt = shift; # unused since unsalted algo + my $iv = shift; + my $data = shift; + my $file = shift; + + my $bit_len = 256; + my $key_len = $bit_len / 8; + + my $is_decrypt = defined ($data); + + my $padding = "none"; # for decryption we need this to "keep" the padding bytes + + if ($is_decrypt == 0) + { + $padding = "standard"; + + # generate some additional random hash data: + + my $iv_len = random_number (1, 16); + + $iv = random_bytes ($iv_len); + + $data = random_bytes (128); + + $file = random_lowercase_string (random_number (1, 16)); + + $file .= ".txt"; + } + + my $iv_mod = $iv; + + $iv_mod .= "\x00" x (16 - length ($iv_mod)); + + + # start of main algo: + + my $digest = sha1 ($word); + + my $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x36")); # or just ^ 0x36 + } + + $buf .= "\x36" x 44; + + my $key = sha1 ($buf); + + + $buf = ""; + + for (my $i = 0; $i < 20; $i++) + { + $buf .= chr (ord (substr ($digest, $i, 1)) ^ ord ("\x5c")); # or just ^ 0x36 + } + + $buf .= "\x5c" x 44; + + # final key: + + $key = $key . sha1 ($buf); + + $key = substr ($key, 0, $key_len); + + my $aes = Crypt::CBC->new ({ + cipher => "Crypt::Rijndael", + key => $key, + iv => $iv_mod, + keysize => $key_len, + literal_key => 1, + header => "none", + padding => $padding, + }); + + if ($is_decrypt == 0) + { + $data = $aes->encrypt ($data); + } + else + { + my $data_decrypted = $aes->decrypt ($data); + + # the password is wrong if the decrypted data does not have the expected padding bytes: + + if (substr ($data_decrypted, -16) ne "\x10" x 16) + { + $data = "fake"; # fake data + } + } + + my $iv_padded = $iv; + + if (length ($iv_padded) < 12) + { + $iv_padded .= "\x00" x (12 - length ($iv_padded)); + } + + my $hash = sprintf ("\$zip3\$*0*1*%i*0*%s*%s*0*0*0*%s", $bit_len, unpack ("H*", $iv_padded), unpack ("H*", $data), $file); + + return $hash; +} + +sub module_verify_hash +{ + my $line = shift; + + return unless (substr ($line, 0, 11) eq "\$zip3\$*0*1*"); + + my $idx1 = index ($line, ":"); + + return unless ($idx1 >= 11); + + my $hash = substr ($line, 0, $idx1); + my $word = substr ($line, $idx1 + 1); + + # bit_len: + + $idx1 = index ($hash, "*", 11); + + return unless ($idx1 > 0); + + my $bit_len = substr ($hash, 11, $idx1 - 11); + + $bit_len = int ($bit_len); + + return unless ($bit_len == 256); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 2) eq "0*"); + + # iv: + + my $idx2 = index ($hash, "*", $idx1 + 3); + + return unless ($idx2 > 0); + + my $iv = substr ($hash, $idx1 + 3, $idx2 - $idx1 - 3); + + return unless ($iv =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($iv) % 2) == 0); + + # data: + + $idx1 = index ($hash, "*", $idx2 + 1); + + return unless ($idx1 > 0); + + my $data = substr ($hash, $idx2 + 1, $idx1 - $idx2 - 1); + + return unless ($data =~ m/^[0-9a-fA-F]+$/); + return unless ((length ($data) % 2) == 0); + + # unused: + + return unless (substr ($hash, $idx1 + 1, 6) eq "0*0*0*"); + + # file: + + my $file = substr ($hash, $idx1 + 7); + + # convert to hex: + + $iv = pack ("H*", $iv); + $data = pack ("H*", $data); + + my $word_packed = pack_if_HEX_notation ($word); + + my $new_hash = module_generate_hash ($word_packed, "", $iv, $data, $file); + + return ($new_hash, $word); +} + +1;