DeforaNetworks/hashcat
3
0
Fork 0
Code Releases Activity

Mix in pure kernel functions in various optimized kernels

Browse Source
This commit is contained in:
jsteube
2017-08-10 12:04:07 +02:00
parent 4b2f3011e9
commit 560a786ea7
26 changed files with 445 additions and 3083 deletions
Download Patch File Download Diff File Expand all files Collapse all files

158
OpenCL/m01760_a1-optimized.cl
View File

@@ -11,117 +11,53 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@@ -157,7 +93,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@@ -185,7 +121,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@@ -221,7 +157,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@@ -249,7 +185,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
__kernel void m01760_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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 void *esalt_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 u32 gid_max)