From 598702944157dc8dc18de643696b80b850687f17 Mon Sep 17 00:00:00 2001 From: Jens Steube Date: Sun, 15 May 2016 01:13:09 +0200 Subject: [PATCH] Added SIMD code for DCC2 --- OpenCL/common.c | 332 ++++++++++++++++++++++++ OpenCL/m02100.cl | 507 ++++++++++++++++++++++++------------- include/kernel_functions.c | 8 + src/hashcat.c | 3 +- 4 files changed, 679 insertions(+), 171 deletions(-) diff --git a/OpenCL/common.c b/OpenCL/common.c index 0d99f6124..7ccbf6564 100644 --- a/OpenCL/common.c +++ b/OpenCL/common.c @@ -6052,6 +6052,76 @@ inline void overwrite_at_be_4x4 (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], * vector functions as scalar (for outer loop usage) */ +inline void append_0x01_1x4_S (u32 w0[4], const u32 offset) +{ + switch (offset) + { + case 0: + w0[0] = 0x01; + break; + + case 1: + w0[0] = w0[0] | 0x0100; + break; + + case 2: + w0[0] = w0[0] | 0x010000; + break; + + case 3: + w0[0] = w0[0] | 0x01000000; + break; + + case 4: + w0[1] = 0x01; + break; + + case 5: + w0[1] = w0[1] | 0x0100; + break; + + case 6: + w0[1] = w0[1] | 0x010000; + break; + + case 7: + w0[1] = w0[1] | 0x01000000; + break; + + case 8: + w0[2] = 0x01; + break; + + case 9: + w0[2] = w0[2] | 0x0100; + break; + + case 10: + w0[2] = w0[2] | 0x010000; + break; + + case 11: + w0[2] = w0[2] | 0x01000000; + break; + + case 12: + w0[3] = 0x01; + break; + + case 13: + w0[3] = w0[3] | 0x0100; + break; + + case 14: + w0[3] = w0[3] | 0x010000; + break; + + case 15: + w0[3] = w0[3] | 0x01000000; + break; + } +} + inline void append_0x01_2x4_S (u32 w0[4], u32 w1[4], const u32 offset) { switch (offset) @@ -6384,6 +6454,268 @@ inline void append_0x01_3x4_S (u32 w0[4], u32 w1[4], u32 w2[4], const u32 offset } } +inline void append_0x01_4x4_S (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 offset) +{ + switch (offset) + { + case 0: + w0[0] = 0x01; + break; + + case 1: + w0[0] = w0[0] | 0x0100; + break; + + case 2: + w0[0] = w0[0] | 0x010000; + break; + + case 3: + w0[0] = w0[0] | 0x01000000; + break; + + case 4: + w0[1] = 0x01; + break; + + case 5: + w0[1] = w0[1] | 0x0100; + break; + + case 6: + w0[1] = w0[1] | 0x010000; + break; + + case 7: + w0[1] = w0[1] | 0x01000000; + break; + + case 8: + w0[2] = 0x01; + break; + + case 9: + w0[2] = w0[2] | 0x0100; + break; + + case 10: + w0[2] = w0[2] | 0x010000; + break; + + case 11: + w0[2] = w0[2] | 0x01000000; + break; + + case 12: + w0[3] = 0x01; + break; + + case 13: + w0[3] = w0[3] | 0x0100; + break; + + case 14: + w0[3] = w0[3] | 0x010000; + break; + + case 15: + w0[3] = w0[3] | 0x01000000; + break; + + case 16: + w1[0] = 0x01; + break; + + case 17: + w1[0] = w1[0] | 0x0100; + break; + + case 18: + w1[0] = w1[0] | 0x010000; + break; + + case 19: + w1[0] = w1[0] | 0x01000000; + break; + + case 20: + w1[1] = 0x01; + break; + + case 21: + w1[1] = w1[1] | 0x0100; + break; + + case 22: + w1[1] = w1[1] | 0x010000; + break; + + case 23: + w1[1] = w1[1] | 0x01000000; + break; + + case 24: + w1[2] = 0x01; + break; + + case 25: + w1[2] = w1[2] | 0x0100; + break; + + case 26: + w1[2] = w1[2] | 0x010000; + break; + + case 27: + w1[2] = w1[2] | 0x01000000; + break; + + case 28: + w1[3] = 0x01; + break; + + case 29: + w1[3] = w1[3] | 0x0100; + break; + + case 30: + w1[3] = w1[3] | 0x010000; + break; + + case 31: + w1[3] = w1[3] | 0x01000000; + break; + + case 32: + w2[0] = 0x01; + break; + + case 33: + w2[0] = w2[0] | 0x0100; + break; + + case 34: + w2[0] = w2[0] | 0x010000; + break; + + case 35: + w2[0] = w2[0] | 0x01000000; + break; + + case 36: + w2[1] = 0x01; + break; + + case 37: + w2[1] = w2[1] | 0x0100; + break; + + case 38: + w2[1] = w2[1] | 0x010000; + break; + + case 39: + w2[1] = w2[1] | 0x01000000; + break; + + case 40: + w2[2] = 0x01; + break; + + case 41: + w2[2] = w2[2] | 0x0100; + break; + + case 42: + w2[2] = w2[2] | 0x010000; + break; + + case 43: + w2[2] = w2[2] | 0x01000000; + break; + + case 44: + w2[3] = 0x01; + break; + + case 45: + w2[3] = w2[3] | 0x0100; + break; + + case 46: + w2[3] = w2[3] | 0x010000; + break; + + case 47: + w2[3] = w2[3] | 0x01000000; + break; + + case 48: + w3[0] = 0x01; + break; + + case 49: + w3[0] = w3[0] | 0x0100; + break; + + case 50: + w3[0] = w3[0] | 0x010000; + break; + + case 51: + w3[0] = w3[0] | 0x01000000; + break; + + case 52: + w3[1] = 0x01; + break; + + case 53: + w3[1] = w3[1] | 0x0100; + break; + + case 54: + w3[1] = w3[1] | 0x010000; + break; + + case 55: + w3[1] = w3[1] | 0x01000000; + break; + + case 56: + w3[2] = 0x01; + break; + + case 57: + w3[2] = w3[2] | 0x0100; + break; + + case 58: + w3[2] = w3[2] | 0x010000; + break; + + case 59: + w3[2] = w3[2] | 0x01000000; + break; + + case 60: + w3[3] = 0x01; + break; + + case 61: + w3[3] = w3[3] | 0x0100; + break; + + case 62: + w3[3] = w3[3] | 0x010000; + break; + + case 63: + w3[3] = w3[3] | 0x01000000; + break; + } +} + inline void append_0x02_2x4_S (u32 w0[4], u32 w1[4], const u32 offset) { switch (offset) diff --git a/OpenCL/m02100.cl b/OpenCL/m02100.cl index 41d2abe5e..2e2a8d2ed 100644 --- a/OpenCL/m02100.cl +++ b/OpenCL/m02100.cl @@ -5,6 +5,8 @@ #define _DCC2_ +#define NEW_SIMD_CODE + #include "include/constants.h" #include "include/kernel_vendor.h" @@ -16,67 +18,68 @@ #include "include/kernel_functions.c" #include "OpenCL/types_ocl.c" #include "OpenCL/common.c" +#include "OpenCL/simd.c" #define COMPARE_S "OpenCL/check_single_comp4.c" #define COMPARE_M "OpenCL/check_multi_comp4.c" -void md4_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4]) +void md4_transform_S (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4]) { u32 a = digest[0]; u32 b = digest[1]; u32 c = digest[2]; u32 d = digest[3]; - MD4_STEP (MD4_Fo, a, b, c, d, w0[0], MD4C00, MD4S00); - MD4_STEP (MD4_Fo, d, a, b, c, w0[1], MD4C00, MD4S01); - MD4_STEP (MD4_Fo, c, d, a, b, w0[2], MD4C00, MD4S02); - MD4_STEP (MD4_Fo, b, c, d, a, w0[3], MD4C00, MD4S03); - MD4_STEP (MD4_Fo, a, b, c, d, w1[0], MD4C00, MD4S00); - MD4_STEP (MD4_Fo, d, a, b, c, w1[1], MD4C00, MD4S01); - MD4_STEP (MD4_Fo, c, d, a, b, w1[2], MD4C00, MD4S02); - MD4_STEP (MD4_Fo, b, c, d, a, w1[3], MD4C00, MD4S03); - MD4_STEP (MD4_Fo, a, b, c, d, w2[0], MD4C00, MD4S00); - MD4_STEP (MD4_Fo, d, a, b, c, w2[1], MD4C00, MD4S01); - MD4_STEP (MD4_Fo, c, d, a, b, w2[2], MD4C00, MD4S02); - MD4_STEP (MD4_Fo, b, c, d, a, w2[3], MD4C00, MD4S03); - MD4_STEP (MD4_Fo, a, b, c, d, w3[0], MD4C00, MD4S00); - MD4_STEP (MD4_Fo, d, a, b, c, w3[1], MD4C00, MD4S01); - MD4_STEP (MD4_Fo, c, d, a, b, w3[2], MD4C00, MD4S02); - MD4_STEP (MD4_Fo, b, c, d, a, w3[3], MD4C00, MD4S03); + MD4_STEP_S (MD4_Fo, a, b, c, d, w0[0], MD4C00, MD4S00); + MD4_STEP_S (MD4_Fo, d, a, b, c, w0[1], MD4C00, MD4S01); + MD4_STEP_S (MD4_Fo, c, d, a, b, w0[2], MD4C00, MD4S02); + MD4_STEP_S (MD4_Fo, b, c, d, a, w0[3], MD4C00, MD4S03); + MD4_STEP_S (MD4_Fo, a, b, c, d, w1[0], MD4C00, MD4S00); + MD4_STEP_S (MD4_Fo, d, a, b, c, w1[1], MD4C00, MD4S01); + MD4_STEP_S (MD4_Fo, c, d, a, b, w1[2], MD4C00, MD4S02); + MD4_STEP_S (MD4_Fo, b, c, d, a, w1[3], MD4C00, MD4S03); + MD4_STEP_S (MD4_Fo, a, b, c, d, w2[0], MD4C00, MD4S00); + MD4_STEP_S (MD4_Fo, d, a, b, c, w2[1], MD4C00, MD4S01); + MD4_STEP_S (MD4_Fo, c, d, a, b, w2[2], MD4C00, MD4S02); + MD4_STEP_S (MD4_Fo, b, c, d, a, w2[3], MD4C00, MD4S03); + MD4_STEP_S (MD4_Fo, a, b, c, d, w3[0], MD4C00, MD4S00); + MD4_STEP_S (MD4_Fo, d, a, b, c, w3[1], MD4C00, MD4S01); + MD4_STEP_S (MD4_Fo, c, d, a, b, w3[2], MD4C00, MD4S02); + MD4_STEP_S (MD4_Fo, b, c, d, a, w3[3], MD4C00, MD4S03); - MD4_STEP (MD4_Go, a, b, c, d, w0[0], MD4C01, MD4S10); - MD4_STEP (MD4_Go, d, a, b, c, w1[0], MD4C01, MD4S11); - MD4_STEP (MD4_Go, c, d, a, b, w2[0], MD4C01, MD4S12); - MD4_STEP (MD4_Go, b, c, d, a, w3[0], MD4C01, MD4S13); - MD4_STEP (MD4_Go, a, b, c, d, w0[1], MD4C01, MD4S10); - MD4_STEP (MD4_Go, d, a, b, c, w1[1], MD4C01, MD4S11); - MD4_STEP (MD4_Go, c, d, a, b, w2[1], MD4C01, MD4S12); - MD4_STEP (MD4_Go, b, c, d, a, w3[1], MD4C01, MD4S13); - MD4_STEP (MD4_Go, a, b, c, d, w0[2], MD4C01, MD4S10); - MD4_STEP (MD4_Go, d, a, b, c, w1[2], MD4C01, MD4S11); - MD4_STEP (MD4_Go, c, d, a, b, w2[2], MD4C01, MD4S12); - MD4_STEP (MD4_Go, b, c, d, a, w3[2], MD4C01, MD4S13); - MD4_STEP (MD4_Go, a, b, c, d, w0[3], MD4C01, MD4S10); - MD4_STEP (MD4_Go, d, a, b, c, w1[3], MD4C01, MD4S11); - MD4_STEP (MD4_Go, c, d, a, b, w2[3], MD4C01, MD4S12); - MD4_STEP (MD4_Go, b, c, d, a, w3[3], MD4C01, MD4S13); + MD4_STEP_S (MD4_Go, a, b, c, d, w0[0], MD4C01, MD4S10); + MD4_STEP_S (MD4_Go, d, a, b, c, w1[0], MD4C01, MD4S11); + MD4_STEP_S (MD4_Go, c, d, a, b, w2[0], MD4C01, MD4S12); + MD4_STEP_S (MD4_Go, b, c, d, a, w3[0], MD4C01, MD4S13); + MD4_STEP_S (MD4_Go, a, b, c, d, w0[1], MD4C01, MD4S10); + MD4_STEP_S (MD4_Go, d, a, b, c, w1[1], MD4C01, MD4S11); + MD4_STEP_S (MD4_Go, c, d, a, b, w2[1], MD4C01, MD4S12); + MD4_STEP_S (MD4_Go, b, c, d, a, w3[1], MD4C01, MD4S13); + MD4_STEP_S (MD4_Go, a, b, c, d, w0[2], MD4C01, MD4S10); + MD4_STEP_S (MD4_Go, d, a, b, c, w1[2], MD4C01, MD4S11); + MD4_STEP_S (MD4_Go, c, d, a, b, w2[2], MD4C01, MD4S12); + MD4_STEP_S (MD4_Go, b, c, d, a, w3[2], MD4C01, MD4S13); + MD4_STEP_S (MD4_Go, a, b, c, d, w0[3], MD4C01, MD4S10); + MD4_STEP_S (MD4_Go, d, a, b, c, w1[3], MD4C01, MD4S11); + MD4_STEP_S (MD4_Go, c, d, a, b, w2[3], MD4C01, MD4S12); + MD4_STEP_S (MD4_Go, b, c, d, a, w3[3], MD4C01, MD4S13); - MD4_STEP (MD4_H , a, b, c, d, w0[0], MD4C02, MD4S20); - MD4_STEP (MD4_H , d, a, b, c, w2[0], MD4C02, MD4S21); - MD4_STEP (MD4_H , c, d, a, b, w1[0], MD4C02, MD4S22); - MD4_STEP (MD4_H , b, c, d, a, w3[0], MD4C02, MD4S23); - MD4_STEP (MD4_H , a, b, c, d, w0[2], MD4C02, MD4S20); - MD4_STEP (MD4_H , d, a, b, c, w2[2], MD4C02, MD4S21); - MD4_STEP (MD4_H , c, d, a, b, w1[2], MD4C02, MD4S22); - MD4_STEP (MD4_H , b, c, d, a, w3[2], MD4C02, MD4S23); - MD4_STEP (MD4_H , a, b, c, d, w0[1], MD4C02, MD4S20); - MD4_STEP (MD4_H , d, a, b, c, w2[1], MD4C02, MD4S21); - MD4_STEP (MD4_H , c, d, a, b, w1[1], MD4C02, MD4S22); - MD4_STEP (MD4_H , b, c, d, a, w3[1], MD4C02, MD4S23); - MD4_STEP (MD4_H , a, b, c, d, w0[3], MD4C02, MD4S20); - MD4_STEP (MD4_H , d, a, b, c, w2[3], MD4C02, MD4S21); - MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); - MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); + MD4_STEP_S (MD4_H , a, b, c, d, w0[0], MD4C02, MD4S20); + MD4_STEP_S (MD4_H , d, a, b, c, w2[0], MD4C02, MD4S21); + MD4_STEP_S (MD4_H , c, d, a, b, w1[0], MD4C02, MD4S22); + MD4_STEP_S (MD4_H , b, c, d, a, w3[0], MD4C02, MD4S23); + MD4_STEP_S (MD4_H , a, b, c, d, w0[2], MD4C02, MD4S20); + MD4_STEP_S (MD4_H , d, a, b, c, w2[2], MD4C02, MD4S21); + MD4_STEP_S (MD4_H , c, d, a, b, w1[2], MD4C02, MD4S22); + MD4_STEP_S (MD4_H , b, c, d, a, w3[2], MD4C02, MD4S23); + MD4_STEP_S (MD4_H , a, b, c, d, w0[1], MD4C02, MD4S20); + MD4_STEP_S (MD4_H , d, a, b, c, w2[1], MD4C02, MD4S21); + MD4_STEP_S (MD4_H , c, d, a, b, w1[1], MD4C02, MD4S22); + MD4_STEP_S (MD4_H , b, c, d, a, w3[1], MD4C02, MD4S23); + MD4_STEP_S (MD4_H , a, b, c, d, w0[3], MD4C02, MD4S20); + MD4_STEP_S (MD4_H , d, a, b, c, w2[3], MD4C02, MD4S21); + MD4_STEP_S (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22); + MD4_STEP_S (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23); digest[0] += a; digest[1] += b; @@ -84,7 +87,7 @@ void md4_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 digest[3] += d; } -void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5]) +void sha1_transform_S (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5]) { u32 A = digest[0]; u32 B = digest[1]; @@ -112,6 +115,223 @@ void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u3 #undef K #define K SHA1C00 + SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, w0_t); + SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w1_t); + SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w2_t); + SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w3_t); + SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w4_t); + SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, w5_t); + SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w6_t); + SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w7_t); + SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w8_t); + SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w9_t); + SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, wa_t); + SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, wb_t); + SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, wc_t); + SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, wd_t); + SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, we_t); + SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, wf_t); + w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w0_t); + w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w1_t); + w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w2_t); + w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w3_t); + + #undef K + #define K SHA1C01 + + w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w4_t); + w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w5_t); + w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w6_t); + w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w7_t); + w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w8_t); + w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w9_t); + wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wa_t); + wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, wb_t); + wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, wc_t); + wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wd_t); + we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, we_t); + wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wf_t); + w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w0_t); + w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w1_t); + w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w2_t); + w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w3_t); + w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w4_t); + w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w5_t); + w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w6_t); + w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w7_t); + + #undef K + #define K SHA1C02 + + w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w8_t); + w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w9_t); + wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, wa_t); + wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, wb_t); + wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, wc_t); + wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, wd_t); + we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, we_t); + wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, wf_t); + w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, w0_t); + w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, w1_t); + w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w2_t); + w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w3_t); + w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, w4_t); + w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, w5_t); + w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, w6_t); + w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w7_t); + w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w8_t); + w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, w9_t); + wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, wa_t); + wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, wb_t); + + #undef K + #define K SHA1C03 + + wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, wc_t); + wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wd_t); + we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, we_t); + wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, wf_t); + w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w0_t); + w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w1_t); + w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w2_t); + w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w3_t); + w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w4_t); + w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w5_t); + w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w6_t); + w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w7_t); + w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w8_t); + w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w9_t); + wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wa_t); + wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, wb_t); + wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wc_t); + wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, wd_t); + we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, we_t); + wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wf_t); + + digest[0] += A; + digest[1] += B; + digest[2] += C; + digest[3] += D; + digest[4] += E; +} + +void hmac_sha1_pad_S (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5]) +{ + w0[0] = w0[0] ^ 0x36363636; + w0[1] = w0[1] ^ 0x36363636; + w0[2] = w0[2] ^ 0x36363636; + w0[3] = w0[3] ^ 0x36363636; + w1[0] = w1[0] ^ 0x36363636; + w1[1] = w1[1] ^ 0x36363636; + w1[2] = w1[2] ^ 0x36363636; + w1[3] = w1[3] ^ 0x36363636; + w2[0] = w2[0] ^ 0x36363636; + w2[1] = w2[1] ^ 0x36363636; + w2[2] = w2[2] ^ 0x36363636; + w2[3] = w2[3] ^ 0x36363636; + w3[0] = w3[0] ^ 0x36363636; + w3[1] = w3[1] ^ 0x36363636; + w3[2] = w3[2] ^ 0x36363636; + w3[3] = w3[3] ^ 0x36363636; + + ipad[0] = SHA1M_A; + ipad[1] = SHA1M_B; + ipad[2] = SHA1M_C; + ipad[3] = SHA1M_D; + ipad[4] = SHA1M_E; + + sha1_transform_S (w0, w1, w2, w3, ipad); + + w0[0] = w0[0] ^ 0x6a6a6a6a; + w0[1] = w0[1] ^ 0x6a6a6a6a; + w0[2] = w0[2] ^ 0x6a6a6a6a; + w0[3] = w0[3] ^ 0x6a6a6a6a; + w1[0] = w1[0] ^ 0x6a6a6a6a; + w1[1] = w1[1] ^ 0x6a6a6a6a; + w1[2] = w1[2] ^ 0x6a6a6a6a; + w1[3] = w1[3] ^ 0x6a6a6a6a; + w2[0] = w2[0] ^ 0x6a6a6a6a; + w2[1] = w2[1] ^ 0x6a6a6a6a; + w2[2] = w2[2] ^ 0x6a6a6a6a; + w2[3] = w2[3] ^ 0x6a6a6a6a; + w3[0] = w3[0] ^ 0x6a6a6a6a; + w3[1] = w3[1] ^ 0x6a6a6a6a; + w3[2] = w3[2] ^ 0x6a6a6a6a; + w3[3] = w3[3] ^ 0x6a6a6a6a; + + opad[0] = SHA1M_A; + opad[1] = SHA1M_B; + opad[2] = SHA1M_C; + opad[3] = SHA1M_D; + opad[4] = SHA1M_E; + + sha1_transform_S (w0, w1, w2, w3, opad); +} + +void hmac_sha1_run_S (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5]) +{ + digest[0] = ipad[0]; + digest[1] = ipad[1]; + digest[2] = ipad[2]; + digest[3] = ipad[3]; + digest[4] = ipad[4]; + + sha1_transform_S (w0, w1, w2, w3, digest); + + w0[0] = digest[0]; + w0[1] = digest[1]; + w0[2] = digest[2]; + w0[3] = digest[3]; + w1[0] = digest[4]; + w1[1] = 0x80000000; + w1[2] = 0; + w1[3] = 0; + w2[0] = 0; + w2[1] = 0; + w2[2] = 0; + w2[3] = 0; + w3[0] = 0; + w3[1] = 0; + w3[2] = 0; + w3[3] = (64 + 20) * 8; + + digest[0] = opad[0]; + digest[1] = opad[1]; + digest[2] = opad[2]; + digest[3] = opad[3]; + digest[4] = opad[4]; + + sha1_transform_S (w0, w1, w2, w3, digest); +} + +void sha1_transform_V (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5]) +{ + u32x A = digest[0]; + u32x B = digest[1]; + u32x C = digest[2]; + u32x D = digest[3]; + u32x E = digest[4]; + + u32x w0_t = w0[0]; + u32x w1_t = w0[1]; + u32x w2_t = w0[2]; + u32x w3_t = w0[3]; + u32x w4_t = w1[0]; + u32x w5_t = w1[1]; + u32x w6_t = w1[2]; + u32x w7_t = w1[3]; + u32x w8_t = w2[0]; + u32x w9_t = w2[1]; + u32x wa_t = w2[2]; + u32x wb_t = w2[3]; + u32x wc_t = w3[0]; + u32x wd_t = w3[1]; + u32x we_t = w3[2]; + u32x wf_t = w3[3]; + + #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); @@ -212,60 +432,7 @@ void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u3 digest[4] += E; } -void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5]) -{ - w0[0] = w0[0] ^ 0x36363636; - w0[1] = w0[1] ^ 0x36363636; - w0[2] = w0[2] ^ 0x36363636; - w0[3] = w0[3] ^ 0x36363636; - w1[0] = w1[0] ^ 0x36363636; - w1[1] = w1[1] ^ 0x36363636; - w1[2] = w1[2] ^ 0x36363636; - w1[3] = w1[3] ^ 0x36363636; - w2[0] = w2[0] ^ 0x36363636; - w2[1] = w2[1] ^ 0x36363636; - w2[2] = w2[2] ^ 0x36363636; - w2[3] = w2[3] ^ 0x36363636; - w3[0] = w3[0] ^ 0x36363636; - w3[1] = w3[1] ^ 0x36363636; - w3[2] = w3[2] ^ 0x36363636; - w3[3] = w3[3] ^ 0x36363636; - - ipad[0] = SHA1M_A; - ipad[1] = SHA1M_B; - ipad[2] = SHA1M_C; - ipad[3] = SHA1M_D; - ipad[4] = SHA1M_E; - - sha1_transform (w0, w1, w2, w3, ipad); - - w0[0] = w0[0] ^ 0x6a6a6a6a; - w0[1] = w0[1] ^ 0x6a6a6a6a; - w0[2] = w0[2] ^ 0x6a6a6a6a; - w0[3] = w0[3] ^ 0x6a6a6a6a; - w1[0] = w1[0] ^ 0x6a6a6a6a; - w1[1] = w1[1] ^ 0x6a6a6a6a; - w1[2] = w1[2] ^ 0x6a6a6a6a; - w1[3] = w1[3] ^ 0x6a6a6a6a; - w2[0] = w2[0] ^ 0x6a6a6a6a; - w2[1] = w2[1] ^ 0x6a6a6a6a; - w2[2] = w2[2] ^ 0x6a6a6a6a; - w2[3] = w2[3] ^ 0x6a6a6a6a; - w3[0] = w3[0] ^ 0x6a6a6a6a; - w3[1] = w3[1] ^ 0x6a6a6a6a; - w3[2] = w3[2] ^ 0x6a6a6a6a; - w3[3] = w3[3] ^ 0x6a6a6a6a; - - opad[0] = SHA1M_A; - opad[1] = SHA1M_B; - opad[2] = SHA1M_C; - opad[3] = SHA1M_D; - opad[4] = SHA1M_E; - - sha1_transform (w0, w1, w2, w3, opad); -} - -void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5]) +void hmac_sha1_run_V (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5]) { digest[0] = ipad[0]; digest[1] = ipad[1]; @@ -273,7 +440,7 @@ void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 digest[3] = ipad[3]; digest[4] = ipad[4]; - sha1_transform (w0, w1, w2, w3, digest); + sha1_transform_V (w0, w1, w2, w3, digest); w0[0] = digest[0]; w0[1] = digest[1]; @@ -298,7 +465,7 @@ void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 digest[3] = opad[3]; digest[4] = opad[4]; - sha1_transform (w0, w1, w2, w3, digest); + sha1_transform_V (w0, w1, w2, w3, digest); } __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) @@ -368,9 +535,9 @@ __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf * generate dcc */ - append_0x80_1x4 (w0, pw_len); + append_0x80_1x4_S (w0, pw_len); - make_unicode (w0, w0, w1); + make_unicode_S (w0, w0, w1); w3[2] = pw_len * 2 * 8; @@ -381,7 +548,7 @@ __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf digest_md4[2] = MD4M_C; digest_md4[3] = MD4M_D; - md4_transform (w0, w1, w2, w3, digest_md4); + md4_transform_S (w0, w1, w2, w3, digest_md4); w0[0] = digest_md4[0]; w0[1] = digest_md4[1]; @@ -400,23 +567,23 @@ __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf w3[2] = (16 + salt_len) * 8; w3[3] = 0; - append_0x80_4x4 (w0, w1, w2, w3, 16 + salt_len); + append_0x80_4x4_S (w0, w1, w2, w3, 16 + salt_len); digest_md4[0] = MD4M_A; digest_md4[1] = MD4M_B; digest_md4[2] = MD4M_C; digest_md4[3] = MD4M_D; - md4_transform (w0, w1, w2, w3, digest_md4); + md4_transform_S (w0, w1, w2, w3, digest_md4); /** * pads */ - w0[0] = swap32 (digest_md4[0]); - w0[1] = swap32 (digest_md4[1]); - w0[2] = swap32 (digest_md4[2]); - w0[3] = swap32 (digest_md4[3]); + w0[0] = swap32_S (digest_md4[0]); + w0[1] = swap32_S (digest_md4[1]); + w0[2] = swap32_S (digest_md4[2]); + w0[3] = swap32_S (digest_md4[3]); w1[0] = 0; w1[1] = 0; w1[2] = 0; @@ -433,7 +600,7 @@ __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf u32 ipad[5]; u32 opad[5]; - hmac_sha1_pad (w0, w1, w2, w3, ipad, opad); + hmac_sha1_pad_S (w0, w1, w2, w3, ipad, opad); tmps[gid].ipad[0] = ipad[0]; tmps[gid].ipad[1] = ipad[1]; @@ -468,27 +635,27 @@ __kernel void m02100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf w3[2] = 0; w3[3] = (64 + salt_len + 4) * 8; - append_0x01_4x4 (w0, w1, w2, w3, salt_len + 3); - append_0x80_4x4 (w0, w1, w2, w3, salt_len + 4); + append_0x01_4x4_S (w0, w1, w2, w3, salt_len + 3); + append_0x80_4x4_S (w0, w1, w2, w3, salt_len + 4); - w0[0] = swap32 (w0[0]); - w0[1] = swap32 (w0[1]); - w0[2] = swap32 (w0[2]); - w0[3] = swap32 (w0[3]); - w1[0] = swap32 (w1[0]); - w1[1] = swap32 (w1[1]); - w1[2] = swap32 (w1[2]); - w1[3] = swap32 (w1[3]); - w2[0] = swap32 (w2[0]); - w2[1] = swap32 (w2[1]); - w2[2] = swap32 (w2[2]); - w2[3] = swap32 (w2[3]); - w3[0] = swap32 (w3[0]); - w3[1] = swap32 (w3[1]); + w0[0] = swap32_S (w0[0]); + w0[1] = swap32_S (w0[1]); + w0[2] = swap32_S (w0[2]); + w0[3] = swap32_S (w0[3]); + w1[0] = swap32_S (w1[0]); + w1[1] = swap32_S (w1[1]); + w1[2] = swap32_S (w1[2]); + w1[3] = swap32_S (w1[3]); + w2[0] = swap32_S (w2[0]); + w2[1] = swap32_S (w2[1]); + w2[2] = swap32_S (w2[2]); + w2[3] = swap32_S (w2[3]); + w3[0] = swap32_S (w3[0]); + w3[1] = swap32_S (w3[1]); u32 digest[5]; - hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest); + hmac_sha1_run_S (w0, w1, w2, w3, ipad, opad, digest); tmps[gid].dgst[0] = digest[0]; tmps[gid].dgst[1] = digest[1]; @@ -512,45 +679,45 @@ __kernel void m02100_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf if (gid >= gid_max) return; - u32 ipad[5]; - u32 opad[5]; + u32x ipad[5]; + u32x opad[5]; - ipad[0] = tmps[gid].ipad[0]; - ipad[1] = tmps[gid].ipad[1]; - ipad[2] = tmps[gid].ipad[2]; - ipad[3] = tmps[gid].ipad[3]; - ipad[4] = tmps[gid].ipad[4]; + ipad[0] = packv (tmps, ipad, gid, 0); + ipad[1] = packv (tmps, ipad, gid, 1); + ipad[2] = packv (tmps, ipad, gid, 2); + ipad[3] = packv (tmps, ipad, gid, 3); + ipad[4] = packv (tmps, ipad, gid, 4); - opad[0] = tmps[gid].opad[0]; - opad[1] = tmps[gid].opad[1]; - opad[2] = tmps[gid].opad[2]; - opad[3] = tmps[gid].opad[3]; - opad[4] = tmps[gid].opad[4]; + opad[0] = packv (tmps, opad, gid, 0); + opad[1] = packv (tmps, opad, gid, 1); + opad[2] = packv (tmps, opad, gid, 2); + opad[3] = packv (tmps, opad, gid, 3); + opad[4] = packv (tmps, opad, gid, 4); /** * iter1 */ - u32 dgst[5]; - u32 out[4]; + u32x dgst[5]; + u32x out[4]; - dgst[0] = tmps[gid].dgst[0]; - dgst[1] = tmps[gid].dgst[1]; - dgst[2] = tmps[gid].dgst[2]; - dgst[3] = tmps[gid].dgst[3]; - dgst[4] = tmps[gid].dgst[4]; + dgst[0] = packv (tmps, dgst, gid, 0); + dgst[1] = packv (tmps, dgst, gid, 1); + dgst[2] = packv (tmps, dgst, gid, 2); + dgst[3] = packv (tmps, dgst, gid, 3); + dgst[4] = packv (tmps, dgst, gid, 4); - out[0] = tmps[gid].out[0]; - out[1] = tmps[gid].out[1]; - out[2] = tmps[gid].out[2]; - out[3] = tmps[gid].out[3]; + out[0] = packv (tmps, out, gid, 0); + out[1] = packv (tmps, out, gid, 1); + out[2] = packv (tmps, out, gid, 2); + out[3] = packv (tmps, out, gid, 3); for (u32 i = 0; i < loop_cnt; i++) { - u32 w0[4]; - u32 w1[4]; - u32 w2[4]; - u32 w3[4]; + u32x w0[4]; + u32x w1[4]; + u32x w2[4]; + u32x w3[4]; w0[0] = dgst[0]; w0[1] = dgst[1]; @@ -569,7 +736,7 @@ __kernel void m02100_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf w3[2] = 0; w3[3] = (64 + 20) * 8; - hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst); + hmac_sha1_run_V (w0, w1, w2, w3, ipad, opad, dgst); out[0] ^= dgst[0]; out[1] ^= dgst[1]; @@ -577,16 +744,16 @@ __kernel void m02100_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf out[3] ^= dgst[3]; } - tmps[gid].dgst[0] = dgst[0]; - tmps[gid].dgst[1] = dgst[1]; - tmps[gid].dgst[2] = dgst[2]; - tmps[gid].dgst[3] = dgst[3]; - tmps[gid].dgst[4] = dgst[4]; + unpackv (tmps, dgst, gid, 0, dgst[0]); + unpackv (tmps, dgst, gid, 1, dgst[1]); + unpackv (tmps, dgst, gid, 2, dgst[2]); + unpackv (tmps, dgst, gid, 3, dgst[3]); + unpackv (tmps, dgst, gid, 4, dgst[4]); - tmps[gid].out[0] = out[0]; - tmps[gid].out[1] = out[1]; - tmps[gid].out[2] = out[2]; - tmps[gid].out[3] = out[3]; + unpackv (tmps, out, gid, 0, out[0]); + unpackv (tmps, out, gid, 1, out[1]); + unpackv (tmps, out, gid, 2, out[2]); + unpackv (tmps, out, gid, 3, out[3]); } __kernel void m02100_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max) diff --git a/include/kernel_functions.c b/include/kernel_functions.c index 258238c62..0c70f5eff 100644 --- a/include/kernel_functions.c +++ b/include/kernel_functions.c @@ -33,6 +33,14 @@ #define MD4_Go(x,y,z) (MD4_G((x), (y), (z))) #endif +#define MD4_STEP_S(f,a,b,c,d,x,K,s) \ +{ \ + a += K; \ + a += x; \ + a += f (b, c, d); \ + a = rotl32_S (a, s); \ +} + #define MD4_STEP(f,a,b,c,d,x,K,s) \ { \ a += K; \ diff --git a/src/hashcat.c b/src/hashcat.c index 1619c5041..375ab12a7 100644 --- a/src/hashcat.c +++ b/src/hashcat.c @@ -8113,7 +8113,8 @@ int main (int argc, char **argv) dgst_size = DGST_SIZE_4_4; parse_func = dcc2_parse_hash; sort_by_digest = sort_by_digest_4_4; - opti_type = OPTI_TYPE_ZERO_BYTE; + opti_type = OPTI_TYPE_ZERO_BYTE + | OPTI_TYPE_SLOW_HASH_SIMD; dgst_pos0 = 0; dgst_pos1 = 1; dgst_pos2 = 2;