1
0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-01 13:09:11 +00:00
hashcat/OpenCL/m14621-pure.cl

343 lines
12 KiB
Common Lisp

/**
* Author......: See docs/credits.txt
* License.....: MIT
*/
#define NEW_SIMD_CODE
#include "inc_vendor.cl"
#include "inc_hash_constants.h"
#include "inc_hash_functions.cl"
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha256.cl"
#include "inc_cipher_aes.cl"
#include "inc_luks_af.cl"
#include "inc_luks_essiv.cl"
#include "inc_luks_xts.cl"
#include "inc_luks_aes.cl"
#define COMPARE_S "inc_comp_single.cl"
#define COMPARE_M "inc_comp_multi.cl"
#define MAX_ENTROPY 7.0
DECLSPEC void hmac_sha256_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest)
{
digest[0] = ipad[0];
digest[1] = ipad[1];
digest[2] = ipad[2];
digest[3] = ipad[3];
digest[4] = ipad[4];
digest[5] = ipad[5];
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform_vector (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] = digest[5];
w1[2] = digest[6];
w1[3] = digest[7];
w2[0] = 0x80000000;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (64 + 32) * 8;
digest[0] = opad[0];
digest[1] = opad[1];
digest[2] = opad[2];
digest[3] = opad[3];
digest[4] = opad[4];
digest[5] = opad[5];
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m14621_init (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global luks_tmp_t *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const luks_t *luks_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 u64 gid_max)
{
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
sha256_hmac_ctx_t sha256_hmac_ctx;
sha256_hmac_init_global_swap (&sha256_hmac_ctx, pws[gid].i, pws[gid].pw_len);
tmps[gid].ipad32[0] = sha256_hmac_ctx.ipad.h[0];
tmps[gid].ipad32[1] = sha256_hmac_ctx.ipad.h[1];
tmps[gid].ipad32[2] = sha256_hmac_ctx.ipad.h[2];
tmps[gid].ipad32[3] = sha256_hmac_ctx.ipad.h[3];
tmps[gid].ipad32[4] = sha256_hmac_ctx.ipad.h[4];
tmps[gid].ipad32[5] = sha256_hmac_ctx.ipad.h[5];
tmps[gid].ipad32[6] = sha256_hmac_ctx.ipad.h[6];
tmps[gid].ipad32[7] = sha256_hmac_ctx.ipad.h[7];
tmps[gid].opad32[0] = sha256_hmac_ctx.opad.h[0];
tmps[gid].opad32[1] = sha256_hmac_ctx.opad.h[1];
tmps[gid].opad32[2] = sha256_hmac_ctx.opad.h[2];
tmps[gid].opad32[3] = sha256_hmac_ctx.opad.h[3];
tmps[gid].opad32[4] = sha256_hmac_ctx.opad.h[4];
tmps[gid].opad32[5] = sha256_hmac_ctx.opad.h[5];
tmps[gid].opad32[6] = sha256_hmac_ctx.opad.h[6];
tmps[gid].opad32[7] = sha256_hmac_ctx.opad.h[7];
sha256_hmac_update_global_swap (&sha256_hmac_ctx, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len);
const u32 key_size = luks_bufs[digests_offset].key_size;
for (u32 i = 0, j = 1; i < ((key_size / 8) / 4); i += 8, j += 1)
{
sha256_hmac_ctx_t sha256_hmac_ctx2 = sha256_hmac_ctx;
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
w0[0] = j;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
sha256_hmac_update_64 (&sha256_hmac_ctx2, w0, w1, w2, w3, 4);
sha256_hmac_final (&sha256_hmac_ctx2);
tmps[gid].dgst32[i + 0] = sha256_hmac_ctx2.opad.h[0];
tmps[gid].dgst32[i + 1] = sha256_hmac_ctx2.opad.h[1];
tmps[gid].dgst32[i + 2] = sha256_hmac_ctx2.opad.h[2];
tmps[gid].dgst32[i + 3] = sha256_hmac_ctx2.opad.h[3];
tmps[gid].dgst32[i + 4] = sha256_hmac_ctx2.opad.h[4];
tmps[gid].dgst32[i + 5] = sha256_hmac_ctx2.opad.h[5];
tmps[gid].dgst32[i + 6] = sha256_hmac_ctx2.opad.h[6];
tmps[gid].dgst32[i + 7] = sha256_hmac_ctx2.opad.h[7];
tmps[gid].out32[i + 0] = tmps[gid].dgst32[i + 0];
tmps[gid].out32[i + 1] = tmps[gid].dgst32[i + 1];
tmps[gid].out32[i + 2] = tmps[gid].dgst32[i + 2];
tmps[gid].out32[i + 3] = tmps[gid].dgst32[i + 3];
tmps[gid].out32[i + 4] = tmps[gid].dgst32[i + 4];
tmps[gid].out32[i + 5] = tmps[gid].dgst32[i + 5];
tmps[gid].out32[i + 6] = tmps[gid].dgst32[i + 6];
tmps[gid].out32[i + 7] = tmps[gid].dgst32[i + 7];
}
}
__kernel void m14621_loop (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global luks_tmp_t *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const luks_t *luks_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 u64 gid_max)
{
const u64 gid = get_global_id (0);
if ((gid * VECT_SIZE) >= gid_max) return;
u32x ipad[8];
u32x opad[8];
ipad[0] = packv (tmps, ipad32, gid, 0);
ipad[1] = packv (tmps, ipad32, gid, 1);
ipad[2] = packv (tmps, ipad32, gid, 2);
ipad[3] = packv (tmps, ipad32, gid, 3);
ipad[4] = packv (tmps, ipad32, gid, 4);
ipad[5] = packv (tmps, ipad32, gid, 5);
ipad[6] = packv (tmps, ipad32, gid, 6);
ipad[7] = packv (tmps, ipad32, gid, 7);
opad[0] = packv (tmps, opad32, gid, 0);
opad[1] = packv (tmps, opad32, gid, 1);
opad[2] = packv (tmps, opad32, gid, 2);
opad[3] = packv (tmps, opad32, gid, 3);
opad[4] = packv (tmps, opad32, gid, 4);
opad[5] = packv (tmps, opad32, gid, 5);
opad[6] = packv (tmps, opad32, gid, 6);
opad[7] = packv (tmps, opad32, gid, 7);
u32 key_size = luks_bufs[digests_offset].key_size;
for (u32 i = 0; i < ((key_size / 8) / 4); i += 8)
{
u32x dgst[8];
u32x out[8];
dgst[0] = packv (tmps, dgst32, gid, i + 0);
dgst[1] = packv (tmps, dgst32, gid, i + 1);
dgst[2] = packv (tmps, dgst32, gid, i + 2);
dgst[3] = packv (tmps, dgst32, gid, i + 3);
dgst[4] = packv (tmps, dgst32, gid, i + 4);
dgst[5] = packv (tmps, dgst32, gid, i + 5);
dgst[6] = packv (tmps, dgst32, gid, i + 6);
dgst[7] = packv (tmps, dgst32, gid, i + 7);
out[0] = packv (tmps, out32, gid, i + 0);
out[1] = packv (tmps, out32, gid, i + 1);
out[2] = packv (tmps, out32, gid, i + 2);
out[3] = packv (tmps, out32, gid, i + 3);
out[4] = packv (tmps, out32, gid, i + 4);
out[5] = packv (tmps, out32, gid, i + 5);
out[6] = packv (tmps, out32, gid, i + 6);
out[7] = packv (tmps, out32, gid, i + 7);
for (u32 j = 0; j < loop_cnt; j++)
{
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
w0[0] = dgst[0];
w0[1] = dgst[1];
w0[2] = dgst[2];
w0[3] = dgst[3];
w1[0] = dgst[4];
w1[1] = dgst[5];
w1[2] = dgst[6];
w1[3] = dgst[7];
w2[0] = 0x80000000;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (64 + 32) * 8;
hmac_sha256_run_V (w0, w1, w2, w3, ipad, opad, dgst);
out[0] ^= dgst[0];
out[1] ^= dgst[1];
out[2] ^= dgst[2];
out[3] ^= dgst[3];
out[4] ^= dgst[4];
out[5] ^= dgst[5];
out[6] ^= dgst[6];
out[7] ^= dgst[7];
}
unpackv (tmps, dgst32, gid, i + 0, dgst[0]);
unpackv (tmps, dgst32, gid, i + 1, dgst[1]);
unpackv (tmps, dgst32, gid, i + 2, dgst[2]);
unpackv (tmps, dgst32, gid, i + 3, dgst[3]);
unpackv (tmps, dgst32, gid, i + 4, dgst[4]);
unpackv (tmps, dgst32, gid, i + 5, dgst[5]);
unpackv (tmps, dgst32, gid, i + 6, dgst[6]);
unpackv (tmps, dgst32, gid, i + 7, dgst[7]);
unpackv (tmps, out32, gid, i + 0, out[0]);
unpackv (tmps, out32, gid, i + 1, out[1]);
unpackv (tmps, out32, gid, i + 2, out[2]);
unpackv (tmps, out32, gid, i + 3, out[3]);
unpackv (tmps, out32, gid, i + 4, out[4]);
unpackv (tmps, out32, gid, i + 5, out[5]);
unpackv (tmps, out32, gid, i + 6, out[6]);
unpackv (tmps, out32, gid, i + 7, out[7]);
}
}
__kernel void m14621_comp (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global luks_tmp_t *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const luks_t *luks_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 u64 gid_max)
{
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 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];
for (MAYBE_VOLATILE 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);
#else
__constant u32a *s_td0 = td0;
__constant u32a *s_td1 = td1;
__constant u32a *s_td2 = td2;
__constant u32a *s_td3 = td3;
__constant u32a *s_td4 = td4;
__constant u32a *s_te0 = te0;
__constant u32a *s_te1 = te1;
__constant u32a *s_te2 = te2;
__constant u32a *s_te3 = te3;
__constant u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
// decrypt AF with first pbkdf2 result
// merge AF to masterkey
// decrypt first payload sector with masterkey
u32 pt_buf[128];
luks_af_sha256_then_aes_decrypt (&luks_bufs[digests_offset], &tmps[gid], pt_buf, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4);
// check entropy
const float entropy = get_entropy (pt_buf, 128);
if (entropy < MAX_ENTROPY)
{
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, 0, gid, 0);
}
}
}