/**
- * Author......: Jens Steube <jens.steube@gmail.com>
+ * Authors.....: Jens Steube <jens.steube@gmail.com>
+ * Gabriele Gristina <matrix@hashcat.net>
+ *
* License.....: MIT
*/
#define _ANDROIDFDE_
-#include "include/constants.h"
-#include "include/kernel_vendor.h"
+#include "inc_hash_constants.h"
+#include "inc_vendor.cl"
#define DGST_R0 0
#define DGST_R1 1
#define DGST_R2 2
#define DGST_R3 3
-#include "include/kernel_functions.c"
-#include "types_ocl.c"
-#include "common.c"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
__constant u32 te0[256] =
{
0x1b000000, 0x36000000,
};
-static void AES128_ExpandKey (u32 *userkey, u32 *rek, __local u32 s_te0[256], __local u32 s_te1[256], __local u32 s_te2[256], __local u32 s_te3[256], __local u32 s_te4[256])
+void AES128_ExpandKey (u32 *userkey, u32 *rek, __local u32 *s_te0, __local u32 *s_te1, __local u32 *s_te2, __local u32 *s_te3, __local u32 *s_te4)
{
rek[0] = userkey[0];
rek[1] = userkey[1];
}
}
-static void AES128_InvertKey (u32 *rdk, __local u32 s_td0[256], __local u32 s_td1[256], __local u32 s_td2[256], __local u32 s_td3[256], __local u32 s_td4[256], __local u32 s_te0[256], __local u32 s_te1[256], __local u32 s_te2[256], __local u32 s_te3[256], __local u32 s_te4[256])
+void AES128_InvertKey (u32 *rdk, __local u32 *s_td0, __local u32 *s_td1, __local u32 *s_td2, __local u32 *s_td3, __local u32 *s_td4, __local u32 *s_te0, __local u32 *s_te1, __local u32 *s_te2, __local u32 *s_te3, __local u32 *s_te4)
{
for (u32 i = 0, j = 40; i < j; i += 4, j -= 4)
{
}
}
-static void AES128_decrypt (const u32 *in, u32 *out, const u32 *rdk, __local u32 s_td0[256], __local u32 s_td1[256], __local u32 s_td2[256], __local u32 s_td3[256], __local u32 s_td4[256])
+void AES128_decrypt (const u32 *in, u32 *out, const u32 *rdk, __local u32 *s_td0, __local u32 *s_td1, __local u32 *s_td2, __local u32 *s_td3, __local u32 *s_td4)
{
u32 s0 = in[0] ^ rdk[0];
u32 s1 = in[1] ^ rdk[1];
^ rdk[43];
}
-static void AES256_ExpandKey (u32 *userkey, u32 *rek, __local u32 s_te0[256], __local u32 s_te1[256], __local u32 s_te2[256], __local u32 s_te3[256], __local u32 s_te4[256])
+void AES256_ExpandKey (u32 *userkey, u32 *rek, __local u32 *s_te0, __local u32 *s_te1, __local u32 *s_te2, __local u32 *s_te3, __local u32 *s_te4)
{
rek[0] = userkey[0];
rek[1] = userkey[1];
}
}
-static void AES256_InvertKey (u32 *rdk, __local u32 s_td0[256], __local u32 s_td1[256], __local u32 s_td2[256], __local u32 s_td3[256], __local u32 s_td4[256], __local u32 s_te0[256], __local u32 s_te1[256], __local u32 s_te2[256], __local u32 s_te3[256], __local u32 s_te4[256])
+void AES256_InvertKey (u32 *rdk, __local u32 *s_td0, __local u32 *s_td1, __local u32 *s_td2, __local u32 *s_td3, __local u32 *s_td4, __local u32 *s_te0, __local u32 *s_te1, __local u32 *s_te2, __local u32 *s_te3, __local u32 *s_te4)
{
for (u32 i = 0, j = 56; i < j; i += 4, j -= 4)
{
}
}
-static void AES256_decrypt (const u32 *in, u32 *out, const u32 *rdk, __local u32 s_td0[256], __local u32 s_td1[256], __local u32 s_td2[256], __local u32 s_td3[256], __local u32 s_td4[256])
+void AES256_decrypt (const u32 *in, u32 *out, const u32 *rdk, __local u32 *s_td0, __local u32 *s_td1, __local u32 *s_td2, __local u32 *s_td3, __local u32 *s_td4)
{
u32 s0 = in[0] ^ rdk[0];
u32 s1 = in[1] ^ rdk[1];
^ rdk[59];
}
-static void AES256_encrypt (const u32 *in, u32 *out, const u32 *rek, __local u32 s_te0[256], __local u32 s_te1[256], __local u32 s_te2[256], __local u32 s_te3[256], __local u32 s_te4[256])
+void AES256_encrypt (const u32 *in, u32 *out, const u32 *rek, __local u32 *s_te0, __local u32 *s_te1, __local u32 *s_te2, __local u32 *s_te3, __local u32 *s_te4)
{
u32 s0 = in[0] ^ rek[0];
u32 s1 = in[1] ^ rek[1];
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
-static void sha256_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[8])
+void sha256_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[8])
{
u32 a = digest[0];
u32 b = digest[1];
ROUND_STEP (0);
+ #ifdef _unroll
#pragma unroll
+ #endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
digest[7] += h;
}
-static 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 (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];
digest[4] += E;
}
-static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
+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;
sha1_transform (w0, w1, w2, w3, opad);
}
-static 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 (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];
sha1_transform (w0, w1, w2, w3, digest);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08800_init (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08800_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_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)
{
/**
* base
* pads
*/
- w0[0] = swap_workaround (w0[0]);
- w0[1] = swap_workaround (w0[1]);
- w0[2] = swap_workaround (w0[2]);
- w0[3] = swap_workaround (w0[3]);
- w1[0] = swap_workaround (w1[0]);
- w1[1] = swap_workaround (w1[1]);
- w1[2] = swap_workaround (w1[2]);
- w1[3] = swap_workaround (w1[3]);
- w2[0] = swap_workaround (w2[0]);
- w2[1] = swap_workaround (w2[1]);
- w2[2] = swap_workaround (w2[2]);
- w2[3] = swap_workaround (w2[3]);
- w3[0] = swap_workaround (w3[0]);
- w3[1] = swap_workaround (w3[1]);
- w3[2] = swap_workaround (w3[2]);
- w3[3] = swap_workaround (w3[3]);
+ 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]);
+ w3[2] = swap32 (w3[2]);
+ w3[3] = swap32 (w3[3]);
u32 ipad[5];
u32 opad[5];
append_0x80_3x4 (w0, w1, w2, salt_len + 4);
- w0[0] = swap_workaround (w0[0]);
- w0[1] = swap_workaround (w0[1]);
- w0[2] = swap_workaround (w0[2]);
- w0[3] = swap_workaround (w0[3]);
- w1[0] = swap_workaround (w1[0]);
- w1[1] = swap_workaround (w1[1]);
- w1[2] = swap_workaround (w1[2]);
- w1[3] = swap_workaround (w1[3]);
- w2[0] = swap_workaround (w2[0]);
- w2[1] = swap_workaround (w2[1]);
+ 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] = 0;
w2[3] = 0;
w3[0] = 0;
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08800_loop (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08800_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_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)
{
const u32 gid = get_global_id (0);
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08800_comp (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08800_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global androidfde_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 androidfde_t *androidfde_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)
{
/**
* base
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
-
- u32 rek[60];
- u32 rdk[60];
-
- u32 data[4];
- u32 iv[4];
- u32 out[4];
+ const u32 lsz = get_local_size (0);
/**
* aes shared
*/
- const u32 lid4 = lid * 4;
-
__local u32 s_td0[256];
__local u32 s_td1[256];
__local u32 s_td2[256];
__local u32 s_te3[256];
__local u32 s_te4[256];
- s_td0[lid4 + 0] = td0[lid4 + 0];
- s_td0[lid4 + 1] = td0[lid4 + 1];
- s_td0[lid4 + 2] = td0[lid4 + 2];
- s_td0[lid4 + 3] = td0[lid4 + 3];
-
- s_td1[lid4 + 0] = td1[lid4 + 0];
- s_td1[lid4 + 1] = td1[lid4 + 1];
- s_td1[lid4 + 2] = td1[lid4 + 2];
- s_td1[lid4 + 3] = td1[lid4 + 3];
-
- s_td2[lid4 + 0] = td2[lid4 + 0];
- s_td2[lid4 + 1] = td2[lid4 + 1];
- s_td2[lid4 + 2] = td2[lid4 + 2];
- s_td2[lid4 + 3] = td2[lid4 + 3];
-
- s_td3[lid4 + 0] = td3[lid4 + 0];
- s_td3[lid4 + 1] = td3[lid4 + 1];
- s_td3[lid4 + 2] = td3[lid4 + 2];
- s_td3[lid4 + 3] = td3[lid4 + 3];
-
- s_td4[lid4 + 0] = td4[lid4 + 0];
- s_td4[lid4 + 1] = td4[lid4 + 1];
- s_td4[lid4 + 2] = td4[lid4 + 2];
- s_td4[lid4 + 3] = td4[lid4 + 3];
-
- s_te0[lid4 + 0] = te0[lid4 + 0];
- s_te0[lid4 + 1] = te0[lid4 + 1];
- s_te0[lid4 + 2] = te0[lid4 + 2];
- s_te0[lid4 + 3] = te0[lid4 + 3];
-
- s_te1[lid4 + 0] = te1[lid4 + 0];
- s_te1[lid4 + 1] = te1[lid4 + 1];
- s_te1[lid4 + 2] = te1[lid4 + 2];
- s_te1[lid4 + 3] = te1[lid4 + 3];
-
- s_te2[lid4 + 0] = te2[lid4 + 0];
- s_te2[lid4 + 1] = te2[lid4 + 1];
- s_te2[lid4 + 2] = te2[lid4 + 2];
- s_te2[lid4 + 3] = te2[lid4 + 3];
-
- s_te3[lid4 + 0] = te3[lid4 + 0];
- s_te3[lid4 + 1] = te3[lid4 + 1];
- s_te3[lid4 + 2] = te3[lid4 + 2];
- s_te3[lid4 + 3] = te3[lid4 + 3];
-
- s_te4[lid4 + 0] = te4[lid4 + 0];
- s_te4[lid4 + 1] = te4[lid4 + 1];
- s_te4[lid4 + 2] = te4[lid4 + 2];
- s_te4[lid4 + 3] = te4[lid4 + 3];
+ 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];
+ }
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
+ /**
+ * base
+ */
+
+ u32 rek[60];
+ u32 rdk[60];
+
+ u32 data[4];
+ u32 iv[4];
+ u32 out[4];
+
/**
* aes init
*/
u32 r2 = out[2] ^ iv[2];
u32 r3 = out[3] ^ iv[3];
- // rotate 3 byte (static in fat!)
+ // rotate 3 byte (in fat!)
r0 = r1 << 8 | r0 >> 24;
r1 = r2 << 8 | r1 >> 24;
// MSDOS5.0
if ((r0 == 0x4f44534d) && (r1 == 0x302e3553))
{
- mark_hash (plains_buf, hashes_shown, digests_offset + 0, gid, 0);
-
- d_return_buf[lid] = 1;
+ mark_hash (plains_buf, d_return_buf, salt_pos, 0, digests_offset + 0, gid, 0);
}
}
}
// we need just a few swapped, because we do not access the others
- r[ 5] = swap_workaround (r[ 5]);
- r[ 6] = swap_workaround (r[ 6]);
- r[14] = swap_workaround (r[14]);
+ r[ 5] = swap32 (r[ 5]);
+ r[ 6] = swap32 (r[ 6]);
+ r[14] = swap32 (r[14]);
// superblock not on id 0 or 1
// assumes max block size is 32MiB
if ((r[5] < 2) && (r[6] < 16) && ((r[14] & 0xffff) == 0xEF53))
{
- mark_hash (plains_buf, hashes_shown, digests_offset + 0, gid, 0);
-
- d_return_buf[lid] = 1;
+ mark_hash (plains_buf, d_return_buf, salt_pos, 0, digests_offset + 0, gid, 0);
}
}
}