/**
- * Author......: Jens Steube <jens.steube@gmail.com>
+ * Authors.....: Jens Steube <jens.steube@gmail.com>
+ * magnum <john.magnum@hushmail.com>
+ *
* License.....: MIT
*/
#define _SHA1_MD5_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#define DGST_R3 1
#include "include/kernel_functions.c"
-#undef _MD5_
-#include "types_ocl.c"
-#include "common.c"
-
-#define COMPARE_S "check_single_comp4.c"
-#define COMPARE_M "check_multi_comp4.c"
-
-#ifdef VECT_SIZE1
-#define uint_to_hex_lower8_le(i) l_bin2asc[(i)]
-#endif
-
-#ifdef VECT_SIZE2
-#define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
-#endif
-
-#ifdef VECT_SIZE4
-#define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
+#include "OpenCL/types_ocl.c"
+#include "OpenCL/common.c"
+#include "OpenCL/simd.c"
+
+#if VECT_SIZE == 1
+#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i)])
+#elif VECT_SIZE == 2
+#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
+#elif VECT_SIZE == 4
+#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
+#elif VECT_SIZE == 8
+#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7])
+#elif VECT_SIZE == 16
+#define uint_to_hex_lower8_le(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7], l_bin2asc[(i).s8], l_bin2asc[(i).s9], l_bin2asc[(i).sa], l_bin2asc[(i).sb], l_bin2asc[(i).sc], l_bin2asc[(i).sd], l_bin2asc[(i).se], l_bin2asc[(i).sf])
#endif
-static void m04700m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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, __local u32 l_bin2asc[256])
+static void m04700m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 il_cnt, const u32 digests_cnt, const u32 digests_offset, __local u32 l_bin2asc[256])
{
/**
* modifier
u32 w0l = w0[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 = bfs_buf[il_pos].i;
-
- w0[0] = w0l | w0r;
+ const u32x w0r = ix_create_bft (bfs_buf, il_pos);
+
+ const u32x w0lr = w0l | w0r;
+
+ u32x w0_t = w0lr;
+ 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];
/**
* md5
*/
- u32 a = MD5M_A;
- u32 b = MD5M_B;
- u32 c = MD5M_C;
- u32 d = MD5M_D;
-
- MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
-
- MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
-
- MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
-
- MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
+ u32x e = 0;
+
+ MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
+
+ MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
+
+ MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
+
+ MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
a += MD5M_A;
b += MD5M_B;
* sha1
*/
- u32 w0_t = uint_to_hex_lower8_le ((a >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((a >> 0) & 255) << 16;
- u32 w1_t = uint_to_hex_lower8_le ((a >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((a >> 16) & 255) << 16;
- u32 w2_t = uint_to_hex_lower8_le ((b >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((b >> 0) & 255) << 16;
- u32 w3_t = uint_to_hex_lower8_le ((b >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((b >> 16) & 255) << 16;
- u32 w4_t = uint_to_hex_lower8_le ((c >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((c >> 0) & 255) << 16;
- u32 w5_t = uint_to_hex_lower8_le ((c >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((c >> 16) & 255) << 16;
- u32 w6_t = uint_to_hex_lower8_le ((d >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((d >> 0) & 255) << 16;
- u32 w7_t = uint_to_hex_lower8_le ((d >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((d >> 16) & 255) << 16;
-
- u32 w8_t = 0x80000000;
- u32 w9_t = 0;
- u32 wa_t = 0;
- u32 wb_t = 0;
- u32 wc_t = 0;
- u32 wd_t = 0;
- u32 we_t = 0;
- u32 wf_t = 32 * 8;
-
- u32 e;
+ w0_t = uint_to_hex_lower8_le ((a >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((a >> 0) & 255) << 16;
+ w1_t = uint_to_hex_lower8_le ((a >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((a >> 16) & 255) << 16;
+ w2_t = uint_to_hex_lower8_le ((b >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((b >> 0) & 255) << 16;
+ w3_t = uint_to_hex_lower8_le ((b >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((b >> 16) & 255) << 16;
+ w4_t = uint_to_hex_lower8_le ((c >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((c >> 0) & 255) << 16;
+ w5_t = uint_to_hex_lower8_le ((c >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((c >> 16) & 255) << 16;
+ w6_t = uint_to_hex_lower8_le ((d >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((d >> 0) & 255) << 16;
+ w7_t = uint_to_hex_lower8_le ((d >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((d >> 16) & 255) << 16;
+ w8_t = 0x80000000;
+ w9_t = 0;
+ wa_t = 0;
+ wb_t = 0;
+ wc_t = 0;
+ wd_t = 0;
+ we_t = 0;
+ wf_t = 32 * 8;
a = SHA1M_A;
b = SHA1M_B;
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ COMPARE_M_SIMD (d, e, c, b);
}
}
-static void m04700s (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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, __local u32 l_bin2asc[256])
+static void m04700s (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 il_cnt, const u32 digests_cnt, const u32 digests_offset, __local u32 l_bin2asc[256])
{
/**
* modifier
* reverse
*/
- const u32 e_rev = rotl32 (search[1], 2u);
+ const u32 e_rev = rotl32_S (search[1], 2u);
/**
* loop
u32 w0l = w0[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 = bfs_buf[il_pos].i;
-
- w0[0] = w0l | w0r;
+ const u32x w0r = ix_create_bft (bfs_buf, il_pos);
+
+ const u32x w0lr = w0l | w0r;
+
+ u32x w0_t = w0lr;
+ 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];
/**
* md5
*/
- u32 a = MD5M_A;
- u32 b = MD5M_B;
- u32 c = MD5M_C;
- u32 d = MD5M_D;
-
- MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
- MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
- MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
- MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
- MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
-
- MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
- MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
- MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
- MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
- MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
-
- MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
-
- MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
- MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
- MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
- MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
- MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
+ u32x e = 0;
+
+ MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
+ MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
+ MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
+ MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
+ MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
+
+ MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
+ MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
+ MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
+ MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
+
+ MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
+
+ MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
+ MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
+ MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
+ MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
+ MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
a += MD5M_A;
b += MD5M_B;
* sha1
*/
- u32 w0_t = uint_to_hex_lower8_le ((a >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((a >> 0) & 255) << 16;
- u32 w1_t = uint_to_hex_lower8_le ((a >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((a >> 16) & 255) << 16;
- u32 w2_t = uint_to_hex_lower8_le ((b >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((b >> 0) & 255) << 16;
- u32 w3_t = uint_to_hex_lower8_le ((b >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((b >> 16) & 255) << 16;
- u32 w4_t = uint_to_hex_lower8_le ((c >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((c >> 0) & 255) << 16;
- u32 w5_t = uint_to_hex_lower8_le ((c >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((c >> 16) & 255) << 16;
- u32 w6_t = uint_to_hex_lower8_le ((d >> 8) & 255) << 0
- | uint_to_hex_lower8_le ((d >> 0) & 255) << 16;
- u32 w7_t = uint_to_hex_lower8_le ((d >> 24) & 255) << 0
- | uint_to_hex_lower8_le ((d >> 16) & 255) << 16;
-
- u32 w8_t = 0x80000000;
- u32 w9_t = 0;
- u32 wa_t = 0;
- u32 wb_t = 0;
- u32 wc_t = 0;
- u32 wd_t = 0;
- u32 we_t = 0;
- u32 wf_t = 32 * 8;
-
- u32 e;
+ w0_t = uint_to_hex_lower8_le ((a >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((a >> 0) & 255) << 16;
+ w1_t = uint_to_hex_lower8_le ((a >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((a >> 16) & 255) << 16;
+ w2_t = uint_to_hex_lower8_le ((b >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((b >> 0) & 255) << 16;
+ w3_t = uint_to_hex_lower8_le ((b >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((b >> 16) & 255) << 16;
+ w4_t = uint_to_hex_lower8_le ((c >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((c >> 0) & 255) << 16;
+ w5_t = uint_to_hex_lower8_le ((c >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((c >> 16) & 255) << 16;
+ w6_t = uint_to_hex_lower8_le ((d >> 8) & 255) << 0
+ | uint_to_hex_lower8_le ((d >> 0) & 255) << 16;
+ w7_t = uint_to_hex_lower8_le ((d >> 24) & 255) << 0
+ | uint_to_hex_lower8_le ((d >> 16) & 255) << 16;
+ w8_t = 0x80000000;
+ w9_t = 0;
+ wa_t = 0;
+ wb_t = 0;
+ wc_t = 0;
+ wd_t = 0;
+ we_t = 0;
+ wf_t = 32 * 8;
a = SHA1M_A;
b = SHA1M_B;
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
- if (allx (e != e_rev)) continue;
+ if (MATCHES_NONE_VS (e, e_rev)) continue;
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ COMPARE_S_SIMD (d, e, c, b);
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m04700_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 m04700_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __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 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);
+ const u32 lsz = get_local_size (0);
/**
- * modifier
+ * bin2asc table
*/
- const u32 lid = get_local_id (0);
+ __local u32 l_bin2asc[256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
+
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 0
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 8;
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * modifier
+ */
u32 w0[4];
const u32 pw_len = pws[gid].pw_len;
- /**
- * bin2asc table
- */
-
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* main
*/
- m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, l_bin2asc);
+ m04700s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, 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, il_cnt, digests_cnt, digests_offset, l_bin2asc);
}