/**
- * Author......: Jens Steube <jens.steube@gmail.com>
+ * Authors.....: Jens Steube <jens.steube@gmail.com>
+ * Gabriele Gristina <matrix@hashcat.net>
+ *
* License.....: MIT
*/
#define _SHA256_
-#include "include/constants.h"
-#include "include/kernel_vendor.h"
+#define NEW_SIMD_CODE
-#define DGST_R0 3
-#define DGST_R1 7
-#define DGST_R2 2
-#define DGST_R3 6
-
-#include "include/kernel_functions.c"
-#include "types_ocl.c"
-#include "common.c"
-
-#define COMPARE_S "check_single_comp4.c"
-#define COMPARE_M "check_multi_comp4.c"
+#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"
__constant u32 k_sha256[64] =
{
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
-static void sha256_transform (u32 digest[8], const u32 w[16])
+#define SHA256_S0_S(x) (rotl32_S ((x), 25u) ^ rotl32_S ((x), 14u) ^ SHIFT_RIGHT_32 ((x), 3u))
+#define SHA256_S1_S(x) (rotl32_S ((x), 15u) ^ rotl32_S ((x), 13u) ^ SHIFT_RIGHT_32 ((x), 10u))
+
+#define SHA256_EXPAND_S(x,y,z,w) (SHA256_S1_S (x) + y + SHA256_S0_S (z) + w)
+
+void sha256_transform (u32x digest[8], const u32x w[16])
{
- u32 a = digest[0];
- u32 b = digest[1];
- u32 c = digest[2];
- u32 d = digest[3];
- u32 e = digest[4];
- u32 f = digest[5];
- u32 g = digest[6];
- u32 h = digest[7];
-
- u32 w0_t = w[ 0];
- u32 w1_t = w[ 1];
- u32 w2_t = w[ 2];
- u32 w3_t = w[ 3];
- u32 w4_t = w[ 4];
- u32 w5_t = w[ 5];
- u32 w6_t = w[ 6];
- u32 w7_t = w[ 7];
- u32 w8_t = w[ 8];
- u32 w9_t = w[ 9];
- u32 wa_t = w[10];
- u32 wb_t = w[11];
- u32 wc_t = w[12];
- u32 wd_t = w[13];
- u32 we_t = w[14];
- u32 wf_t = w[15];
+ u32x a = digest[0];
+ u32x b = digest[1];
+ u32x c = digest[2];
+ u32x d = digest[3];
+ u32x e = digest[4];
+ u32x f = digest[5];
+ u32x g = digest[6];
+ u32x h = digest[7];
+
+ u32x w0_t = w[ 0];
+ u32x w1_t = w[ 1];
+ u32x w2_t = w[ 2];
+ u32x w3_t = w[ 3];
+ u32x w4_t = w[ 4];
+ u32x w5_t = w[ 5];
+ u32x w6_t = w[ 6];
+ u32x w7_t = w[ 7];
+ u32x w8_t = w[ 8];
+ u32x w9_t = w[ 9];
+ u32x wa_t = w[10];
+ u32x wb_t = w[11];
+ u32x wc_t = w[12];
+ u32x wd_t = w[13];
+ u32x we_t = w[14];
+ u32x wf_t = w[15];
#define ROUND_EXPAND() \
{ \
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 m08000m (u32 w[16], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
+void sha256_transform_z (u32x digest[8])
+{
+ u32x a = digest[0];
+ u32x b = digest[1];
+ u32x c = digest[2];
+ u32x d = digest[3];
+ u32x e = digest[4];
+ u32x f = digest[5];
+ u32x g = digest[6];
+ u32x h = digest[7];
+
+ #define ROUND_STEP_Z(i) \
+ { \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 0]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 1]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 2]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 3]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 4]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 5]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 6]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 7]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 8]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 9]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 10]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 11]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 12]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 13]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 14]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 15]); \
+ }
+
+ ROUND_STEP_Z (0);
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i += 16)
+ {
+ ROUND_STEP_Z (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;
+}
+
+void sha256_transform_s (u32x digest[8], __local u32 *w)
+{
+ u32x a = digest[0];
+ u32x b = digest[1];
+ u32x c = digest[2];
+ u32x d = digest[3];
+ u32x e = digest[4];
+ u32x f = digest[5];
+ u32x g = digest[6];
+ u32x h = digest[7];
+
+ #define ROUND_STEP_S(i) \
+ { \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 0], k_sha256[i + 0]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 1], k_sha256[i + 1]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 2], k_sha256[i + 2]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 3], k_sha256[i + 3]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 4], k_sha256[i + 4]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 5], k_sha256[i + 5]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 6], k_sha256[i + 6]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 7], k_sha256[i + 7]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 8], k_sha256[i + 8]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 9], k_sha256[i + 9]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 10], k_sha256[i + 10]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 11], k_sha256[i + 11]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 12], k_sha256[i + 12]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 13], k_sha256[i + 13]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 14], k_sha256[i + 14]); \
+ SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 15], k_sha256[i + 15]); \
+ }
+
+ ROUND_STEP_S (0);
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i += 16)
+ {
+ ROUND_STEP_S (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;
+}
+
+void m08000m (__local u32 *w_s1, __local u32 *w_s2, u32 w[16], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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 gid_max)
{
/**
* modifier
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
/**
* salt
*/
- const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
- const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
- const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
+ const u32 salt_buf0 = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ const u32 salt_buf1 = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ const u32 salt_buf2 = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
+
+ /**
+ * precompute final msg blocks
+ */
+
+ for (u32 i = lid; i < 64; i += lsz)
+ {
+ w_s1[i] = 0;
+ w_s2[i] = 0;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (lid == 0)
+ {
+ w_s1[15] = 0 | salt_buf0 >> 16;
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i++)
+ {
+ w_s1[i] = SHA256_EXPAND_S (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
+ }
+
+ w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
+ w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
+ w_s2[ 2] = salt_buf2 << 16 | 0;
+ w_s2[15] = (510 + 8) * 8;
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i++)
+ {
+ w_s2[i] = SHA256_EXPAND_S (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
+ }
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
+
+ u32x w_t[16];
+
+ w_t[ 0] = w[ 0] >> 8;
+ w_t[ 1] = w[ 1] >> 8;
+ w_t[ 2] = w[ 2] >> 8;
+ w_t[ 3] = w[ 3] >> 8;
+ w_t[ 4] = w[ 4] >> 8;
+ w_t[ 5] = w[ 5] >> 8;
+ w_t[ 6] = w[ 6] >> 8;
+ w_t[ 7] = w[ 7] >> 8;
+ w_t[ 8] = w[ 8] >> 8;
+ w_t[ 9] = w[ 9] >> 8;
+ w_t[10] = w[10] >> 8;
+ w_t[11] = w[11] >> 8;
+ w_t[12] = w[12] >> 8;
+ w_t[13] = w[13] >> 8;
+ w_t[14] = w[14] >> 8;
+ w_t[15] = w[15] >> 8;
/**
* loop
u32 w0l = w[0];
- for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
{
- const u32 w0r = words_buf_r[il_pos];
-
- const u32 w0 = w0l | w0r;
-
- u32 w_t[16];
-
- w_t[ 0] = w0 >> 8;
- w_t[ 1] = w[ 1] >> 8;
- w_t[ 2] = w[ 2] >> 8;
- w_t[ 3] = w[ 3] >> 8;
- w_t[ 4] = w[ 4] >> 8;
- w_t[ 5] = w[ 5] >> 8;
- w_t[ 6] = w[ 6] >> 8;
- w_t[ 7] = w[ 7] >> 8;
- w_t[ 8] = w[ 8] >> 8;
- w_t[ 9] = w[ 9] >> 8;
- w_t[10] = w[10] >> 8;
- w_t[11] = w[11] >> 8;
- w_t[12] = w[12] >> 8;
- w_t[13] = w[13] >> 8;
- w_t[14] = w[14] >> 8;
- w_t[15] = w[15] >> 8;
-
- u32 digest[8];
+ const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+ const u32x w0lr = w0l | w0r;
+
+ w_t[0] = w0lr >> 8;
+
+ u32x digest[8];
digest[0] = SHA256M_A;
digest[1] = SHA256M_B;
digest[6] = SHA256M_G;
digest[7] = SHA256M_H;
- sha256_transform (digest, w_t); // 0 - 64
-
- w_t[ 0] = 0;
- w_t[ 1] = 0;
- w_t[ 2] = 0;
- w_t[ 3] = 0;
- w_t[ 4] = 0;
- w_t[ 5] = 0;
- w_t[ 6] = 0;
- w_t[ 7] = 0;
- w_t[ 8] = 0;
- w_t[ 9] = 0;
- w_t[10] = 0;
- w_t[11] = 0;
- w_t[12] = 0;
- w_t[13] = 0;
- w_t[14] = 0;
- w_t[15] = 0;
-
- sha256_transform (digest, w_t); // 64 - 128
- sha256_transform (digest, w_t); // 128 - 192
- sha256_transform (digest, w_t); // 192 - 256
- sha256_transform (digest, w_t); // 256 - 320
- sha256_transform (digest, w_t); // 320 - 384
- sha256_transform (digest, w_t); // 384 - 448
-
- w_t[15] = 0 | salt_buf0 >> 16;
-
- sha256_transform (digest, w_t); // 448 - 512
-
- w_t[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
- w_t[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
- w_t[ 2] = salt_buf2 << 16 | 0;
- w_t[15] = (510 + 8) * 8;
-
- sha256_transform (digest, w_t); // 512 - 576
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[7];
- const u32 r2 = digest[2];
- const u32 r3 = digest[6];
-
- #include COMPARE_M
+ sha256_transform (digest, w_t); // 0 - 64
+ sha256_transform_z (digest); // 64 - 128
+ sha256_transform_z (digest); // 128 - 192
+ sha256_transform_z (digest); // 192 - 256
+ sha256_transform_z (digest); // 256 - 320
+ sha256_transform_z (digest); // 320 - 384
+ sha256_transform_z (digest); // 384 - 448
+ sha256_transform_s (digest, w_s1); // 448 - 512
+ sha256_transform_s (digest, w_s2); // 512 - 576
+
+ COMPARE_M_SIMD (digest[3], digest[7], digest[2], digest[6]);
}
}
-static void m08000s (u32 w[16], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
+void m08000s (__local u32 *w_s1, __local u32 *w_s2, u32 w[16], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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 gid_max)
{
/**
* modifier
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
+
+ /**
+ * salt
+ */
+
+ const u32 salt_buf0 = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ const u32 salt_buf1 = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ const u32 salt_buf2 = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
+
+ /**
+ * precompute final msg blocks
+ */
+
+ for (u32 i = lid; i < 64; i += lsz)
+ {
+ w_s1[i] = 0;
+ w_s2[i] = 0;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (lid == 0)
+ {
+ w_s1[15] = 0 | salt_buf0 >> 16;
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i++)
+ {
+ w_s1[i] = SHA256_EXPAND_S (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
+ }
+
+ w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
+ w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
+ w_s2[ 2] = salt_buf2 << 16 | 0;
+ w_s2[15] = (510 + 8) * 8;
+
+ #ifdef _unroll
+ #pragma unroll
+ #endif
+ for (int i = 16; i < 64; i++)
+ {
+ w_s2[i] = SHA256_EXPAND_S (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
+ }
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
+
+ u32x w_t[16];
+
+ w_t[ 0] = w[ 0] >> 8;
+ w_t[ 1] = w[ 1] >> 8;
+ w_t[ 2] = w[ 2] >> 8;
+ w_t[ 3] = w[ 3] >> 8;
+ w_t[ 4] = w[ 4] >> 8;
+ w_t[ 5] = w[ 5] >> 8;
+ w_t[ 6] = w[ 6] >> 8;
+ w_t[ 7] = w[ 7] >> 8;
+ w_t[ 8] = w[ 8] >> 8;
+ w_t[ 9] = w[ 9] >> 8;
+ w_t[10] = w[10] >> 8;
+ w_t[11] = w[11] >> 8;
+ w_t[12] = w[12] >> 8;
+ w_t[13] = w[13] >> 8;
+ w_t[14] = w[14] >> 8;
+ w_t[15] = w[15] >> 8;
/**
* digest
digests_buf[digests_offset].digest_buf[DGST_R3]
};
- /**
- * salt
- */
-
- const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
- const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
- const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
-
/**
* loop
*/
u32 w0l = w[0];
- for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
{
- const u32 w0r = words_buf_r[il_pos];
-
- const u32 w0 = w0l | w0r;
-
- u32 w_t[16];
-
- w_t[ 0] = w0 >> 8;
- w_t[ 1] = w[ 1] >> 8;
- w_t[ 2] = w[ 2] >> 8;
- w_t[ 3] = w[ 3] >> 8;
- w_t[ 4] = w[ 4] >> 8;
- w_t[ 5] = w[ 5] >> 8;
- w_t[ 6] = w[ 6] >> 8;
- w_t[ 7] = w[ 7] >> 8;
- w_t[ 8] = w[ 8] >> 8;
- w_t[ 9] = w[ 9] >> 8;
- w_t[10] = w[10] >> 8;
- w_t[11] = w[11] >> 8;
- w_t[12] = w[12] >> 8;
- w_t[13] = w[13] >> 8;
- w_t[14] = w[14] >> 8;
- w_t[15] = w[15] >> 8;
-
- u32 digest[8];
+ const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+ const u32x w0lr = w0l | w0r;
+
+ w_t[0] = w0lr >> 8;
+
+ u32x digest[8];
digest[0] = SHA256M_A;
digest[1] = SHA256M_B;
digest[6] = SHA256M_G;
digest[7] = SHA256M_H;
- sha256_transform (digest, w_t); // 0 - 64
-
- w_t[ 0] = 0;
- w_t[ 1] = 0;
- w_t[ 2] = 0;
- w_t[ 3] = 0;
- w_t[ 4] = 0;
- w_t[ 5] = 0;
- w_t[ 6] = 0;
- w_t[ 7] = 0;
- w_t[ 8] = 0;
- w_t[ 9] = 0;
- w_t[10] = 0;
- w_t[11] = 0;
- w_t[12] = 0;
- w_t[13] = 0;
- w_t[14] = 0;
- w_t[15] = 0;
-
- sha256_transform (digest, w_t); // 64 - 128
- sha256_transform (digest, w_t); // 128 - 192
- sha256_transform (digest, w_t); // 192 - 256
- sha256_transform (digest, w_t); // 256 - 320
- sha256_transform (digest, w_t); // 320 - 384
- sha256_transform (digest, w_t); // 384 - 448
-
- w_t[15] = 0 | salt_buf0 >> 16;
-
- sha256_transform (digest, w_t); // 448 - 512
-
- w_t[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
- w_t[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
- w_t[ 2] = salt_buf2 << 16 | 0;
- w_t[15] = (510 + 8) * 8;
-
- sha256_transform (digest, w_t); // 512 - 576
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[7];
- const u32 r2 = digest[2];
- const u32 r3 = digest[6];
-
- #include COMPARE_S
+ sha256_transform (digest, w_t); // 0 - 64
+ sha256_transform_z (digest); // 64 - 128
+ sha256_transform_z (digest); // 128 - 192
+ sha256_transform_z (digest); // 192 - 256
+ sha256_transform_z (digest); // 256 - 320
+ sha256_transform_z (digest); // 320 - 384
+ sha256_transform_z (digest); // 384 - 448
+ sha256_transform_s (digest, w_s1); // 448 - 512
+ sha256_transform_s (digest, w_s2); // 512 - 576
+
+ COMPARE_S_SIMD (digest[3], digest[7], digest[2], digest[6]);
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000m (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000m (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000m (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000m (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000m (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000m (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000s (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000s (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000s (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000s (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global u32 * words_buf_r, __global void *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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m08000_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __constant u32x * words_buf_r, __global void *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_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)
{
/**
* base
const u32 gid = get_global_id (0);
- if (gid >= gid_max) return;
-
u32 w[16];
w[ 0] = pws[gid].i[ 0];
const u32 pw_len = pws[gid].pw_len;
+ __local u32 w_s1[64];
+ __local u32 w_s2[64];
+
/**
* main
*/
- m08000s (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_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
+ m08000s (w_s1, w_s2, 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_scryptV0_buf, d_scryptV1_buf, d_scryptV2_buf, d_scryptV3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, gid_max);
}
projects / hashcat.git / blobdiff