--- /dev/null
+/**
+ * Author......: Jens Steube <jens.steube@gmail.com>
+ * License.....: MIT
+ */
+
+#define _MD5_
+
+#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_rp.h"
+#include "inc_rp.cl"
+#include "inc_simd.cl"
+
+__kernel void m00000_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ u32x w0[4] = { 0 };
+ u32x w1[4] = { 0 };
+ u32x w2[4] = { 0 };
+ u32x w3[4] = { 0 };
+
+ const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+ append_0x80_2x4_VV (w0, w1, out_len);
+
+ w3[2] = out_len * 8;
+ w3[3] = 0;
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x 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);
+
+ COMPARE_M_SIMD (a, d, c, b);
+ }
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * digest
+ */
+
+ const u32 search[4] =
+ {
+ digests_buf[digests_offset].digest_buf[DGST_R0],
+ digests_buf[digests_offset].digest_buf[DGST_R1],
+ digests_buf[digests_offset].digest_buf[DGST_R2],
+ digests_buf[digests_offset].digest_buf[DGST_R3]
+ };
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ u32x w0[4] = { 0 };
+ u32x w1[4] = { 0 };
+ u32x w2[4] = { 0 };
+ u32x w3[4] = { 0 };
+
+ const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+ append_0x80_2x4_VV (w0, w1, out_len);
+
+ w3[2] = out_len * 8;
+ w3[3] = 0;
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x 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);
+
+ if (MATCHES_NONE_VS (a, search[0])) continue;
+
+ 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);
+
+ COMPARE_S_SIMD (a, d, c, b);
+ }
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+}
--- /dev/null
+/**
+ * Author......: Jens Steube <jens.steube@gmail.com>
+ * License.....: MIT
+ */
+
+#define _MD5_
+
+#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_simd.cl"
+
+__kernel void m00000_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_l_len = pws[gid].pw_len;
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_len = pw_l_len + pw_r_len;
+
+ /**
+ * concat password candidate
+ */
+
+ u32x wordl0[4] = { 0 };
+ u32x wordl1[4] = { 0 };
+ u32x wordl2[4] = { 0 };
+ u32x wordl3[4] = { 0 };
+
+ wordl0[0] = pw_buf0[0];
+ wordl0[1] = pw_buf0[1];
+ wordl0[2] = pw_buf0[2];
+ wordl0[3] = pw_buf0[3];
+ wordl1[0] = pw_buf1[0];
+ wordl1[1] = pw_buf1[1];
+ wordl1[2] = pw_buf1[2];
+ wordl1[3] = pw_buf1[3];
+
+ u32x wordr0[4] = { 0 };
+ u32x wordr1[4] = { 0 };
+ u32x wordr2[4] = { 0 };
+ u32x wordr3[4] = { 0 };
+
+ wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
+ wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
+ wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
+ wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
+ wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
+ wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
+ wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
+ wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
+
+ if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
+ {
+ switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ }
+ else
+ {
+ switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
+ }
+
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
+
+ w0[0] = wordl0[0] | wordr0[0];
+ w0[1] = wordl0[1] | wordr0[1];
+ w0[2] = wordl0[2] | wordr0[2];
+ w0[3] = wordl0[3] | wordr0[3];
+ w1[0] = wordl1[0] | wordr1[0];
+ w1[1] = wordl1[1] | wordr1[1];
+ w1[2] = wordl1[2] | wordr1[2];
+ w1[3] = wordl1[3] | wordr1[3];
+ w2[0] = wordl2[0] | wordr2[0];
+ w2[1] = wordl2[1] | wordr2[1];
+ w2[2] = wordl2[2] | wordr2[2];
+ w2[3] = wordl2[3] | wordr2[3];
+ w3[0] = wordl3[0] | wordr3[0];
+ w3[1] = wordl3[1] | wordr3[1];
+ w3[2] = pw_len * 8;
+ w3[3] = 0;
+
+ /**
+ * md5
+ */
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x 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);
+
+ COMPARE_M_SIMD (a, d, c, b);
+ }
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_l_len = pws[gid].pw_len;
+
+ /**
+ * digest
+ */
+
+ const u32 search[4] =
+ {
+ digests_buf[digests_offset].digest_buf[DGST_R0],
+ digests_buf[digests_offset].digest_buf[DGST_R1],
+ digests_buf[digests_offset].digest_buf[DGST_R2],
+ digests_buf[digests_offset].digest_buf[DGST_R3]
+ };
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_len = pw_l_len + pw_r_len;
+
+ /**
+ * concat password candidate
+ */
+
+ u32x wordl0[4] = { 0 };
+ u32x wordl1[4] = { 0 };
+ u32x wordl2[4] = { 0 };
+ u32x wordl3[4] = { 0 };
+
+ wordl0[0] = pw_buf0[0];
+ wordl0[1] = pw_buf0[1];
+ wordl0[2] = pw_buf0[2];
+ wordl0[3] = pw_buf0[3];
+ wordl1[0] = pw_buf1[0];
+ wordl1[1] = pw_buf1[1];
+ wordl1[2] = pw_buf1[2];
+ wordl1[3] = pw_buf1[3];
+
+ u32x wordr0[4] = { 0 };
+ u32x wordr1[4] = { 0 };
+ u32x wordr2[4] = { 0 };
+ u32x wordr3[4] = { 0 };
+
+ wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
+ wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
+ wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
+ wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
+ wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
+ wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
+ wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
+ wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
+
+ if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
+ {
+ switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ }
+ else
+ {
+ switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
+ }
+
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
+
+ w0[0] = wordl0[0] | wordr0[0];
+ w0[1] = wordl0[1] | wordr0[1];
+ w0[2] = wordl0[2] | wordr0[2];
+ w0[3] = wordl0[3] | wordr0[3];
+ w1[0] = wordl1[0] | wordr1[0];
+ w1[1] = wordl1[1] | wordr1[1];
+ w1[2] = wordl1[2] | wordr1[2];
+ w1[3] = wordl1[3] | wordr1[3];
+ w2[0] = wordl2[0] | wordr2[0];
+ w2[1] = wordl2[1] | wordr2[1];
+ w2[2] = wordl2[2] | wordr2[2];
+ w2[3] = wordl2[3] | wordr2[3];
+ w3[0] = wordl3[0] | wordr3[0];
+ w3[1] = wordl3[1] | wordr3[1];
+ w3[2] = pw_len * 8;
+ w3[3] = 0;
+
+ /**
+ * md5
+ */
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x 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);
+
+ if (MATCHES_NONE_VS (a, search[0])) continue;
+
+ 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);
+
+ COMPARE_S_SIMD (a, d, c, b);
+ }
+}
+
+__kernel void m00000_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_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)
+{
+}
+
+__kernel void m00000_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_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)
+{
+}
--- /dev/null
+/**
+ * Author......: Jens Steube <jens.steube@gmail.com>
+ * License.....: MIT
+ */
+
+#define _MD5_
+
+#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_simd.cl"
+
+#define MD5_STEP_REV(f,a,b,c,d,x,t,s) \
+{ \
+ a -= b; \
+ a = rotr32_S (a, s); \
+ a -= f (b, c, d); \
+ a -= x; \
+ a -= t; \
+}
+
+#define MD5_STEP_REV1(f,a,b,c,d,x,t,s) \
+{ \
+ a -= b; \
+ a = rotr32_S (a, s); \
+ a -= x; \
+ a -= t; \
+}
+
+void m00000m (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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 gid = get_global_id (0);
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 F_w0c00 = 0 + MD5C00;
+ const u32 F_w1c01 = w[ 1] + MD5C01;
+ const u32 F_w2c02 = w[ 2] + MD5C02;
+ const u32 F_w3c03 = w[ 3] + MD5C03;
+ const u32 F_w4c04 = w[ 4] + MD5C04;
+ const u32 F_w5c05 = w[ 5] + MD5C05;
+ const u32 F_w6c06 = w[ 6] + MD5C06;
+ const u32 F_w7c07 = w[ 7] + MD5C07;
+ const u32 F_w8c08 = w[ 8] + MD5C08;
+ const u32 F_w9c09 = w[ 9] + MD5C09;
+ const u32 F_wac0a = w[10] + MD5C0a;
+ const u32 F_wbc0b = w[11] + MD5C0b;
+ const u32 F_wcc0c = w[12] + MD5C0c;
+ const u32 F_wdc0d = w[13] + MD5C0d;
+ const u32 F_wec0e = w[14] + MD5C0e;
+ const u32 F_wfc0f = w[15] + MD5C0f;
+
+ const u32 G_w1c10 = w[ 1] + MD5C10;
+ const u32 G_w6c11 = w[ 6] + MD5C11;
+ const u32 G_wbc12 = w[11] + MD5C12;
+ const u32 G_w0c13 = 0 + MD5C13;
+ const u32 G_w5c14 = w[ 5] + MD5C14;
+ const u32 G_wac15 = w[10] + MD5C15;
+ const u32 G_wfc16 = w[15] + MD5C16;
+ const u32 G_w4c17 = w[ 4] + MD5C17;
+ const u32 G_w9c18 = w[ 9] + MD5C18;
+ const u32 G_wec19 = w[14] + MD5C19;
+ const u32 G_w3c1a = w[ 3] + MD5C1a;
+ const u32 G_w8c1b = w[ 8] + MD5C1b;
+ const u32 G_wdc1c = w[13] + MD5C1c;
+ const u32 G_w2c1d = w[ 2] + MD5C1d;
+ const u32 G_w7c1e = w[ 7] + MD5C1e;
+ const u32 G_wcc1f = w[12] + MD5C1f;
+
+ const u32 H_w5c20 = w[ 5] + MD5C20;
+ const u32 H_w8c21 = w[ 8] + MD5C21;
+ const u32 H_wbc22 = w[11] + MD5C22;
+ const u32 H_wec23 = w[14] + MD5C23;
+ const u32 H_w1c24 = w[ 1] + MD5C24;
+ const u32 H_w4c25 = w[ 4] + MD5C25;
+ const u32 H_w7c26 = w[ 7] + MD5C26;
+ const u32 H_wac27 = w[10] + MD5C27;
+ const u32 H_wdc28 = w[13] + MD5C28;
+ const u32 H_w0c29 = 0 + MD5C29;
+ const u32 H_w3c2a = w[ 3] + MD5C2a;
+ const u32 H_w6c2b = w[ 6] + MD5C2b;
+ const u32 H_w9c2c = w[ 9] + MD5C2c;
+ const u32 H_wcc2d = w[12] + MD5C2d;
+ const u32 H_wfc2e = w[15] + MD5C2e;
+ const u32 H_w2c2f = w[ 2] + MD5C2f;
+
+ const u32 I_w0c30 = 0 + MD5C30;
+ const u32 I_w7c31 = w[ 7] + MD5C31;
+ const u32 I_wec32 = w[14] + MD5C32;
+ const u32 I_w5c33 = w[ 5] + MD5C33;
+ const u32 I_wcc34 = w[12] + MD5C34;
+ const u32 I_w3c35 = w[ 3] + MD5C35;
+ const u32 I_wac36 = w[10] + MD5C36;
+ const u32 I_w1c37 = w[ 1] + MD5C37;
+ const u32 I_w8c38 = w[ 8] + MD5C38;
+ const u32 I_wfc39 = w[15] + MD5C39;
+ const u32 I_w6c3a = w[ 6] + MD5C3a;
+ const u32 I_wdc3b = w[13] + MD5C3b;
+ const u32 I_w4c3c = w[ 4] + MD5C3c;
+ const u32 I_wbc3d = w[11] + MD5C3d;
+ const u32 I_w2c3e = w[ 2] + MD5C3e;
+ const u32 I_w9c3f = w[ 9] + MD5C3f;
+
+ /**
+ * loop
+ */
+
+ u32 w0l = w[0];
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+ const u32x w0 = w0l | w0r;
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
+
+ MD5_STEP (MD5_Fo, a, b, c, d, w0, F_w0c00, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w1c01, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_w2c02, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_w3c03, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_w4c04, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w5c05, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_w6c06, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_w7c07, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_w8c08, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w9c09, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_wac0a, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_wbc0b, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_wcc0c, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_wdc0d, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_wec0e, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_wfc0f, MD5S03);
+
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w1c10, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_w6c11, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_wbc12, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w0, G_w0c13, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w5c14, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_wac15, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_wfc16, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_w4c17, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w9c18, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_wec19, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_w3c1a, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_w8c1b, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_wdc1c, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_w2c1d, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_w7c1e, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_wcc1f, MD5S13);
+
+ MD5_STEP0(MD5_H , a, b, c, d, H_w5c20, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_w8c21, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_wbc22, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_wec23, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_w1c24, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_w4c25, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_w7c26, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_wac27, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_wdc28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0, H_w0c29, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_w3c2a, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_w6c2b, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_w9c2c, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_wcc2d, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_wfc2e, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_w2c2f, MD5S23);
+
+ MD5_STEP (MD5_I , a, b, c, d, w0, I_w0c30, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_w7c31, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_wec32, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w5c33, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_wcc34, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_w3c35, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_wac36, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w1c37, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_w8c38, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_wfc39, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_w6c3a, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_wdc3b, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_w4c3c, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_wbc3d, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_w2c3e, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w9c3f, MD5S33);
+
+ COMPARE_M_SIMD (a, d, c, b);
+ }
+}
+
+void m00000s (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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 gid = get_global_id (0);
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 F_w0c00 = 0 + MD5C00;
+ const u32 F_w1c01 = w[ 1] + MD5C01;
+ const u32 F_w2c02 = w[ 2] + MD5C02;
+ const u32 F_w3c03 = w[ 3] + MD5C03;
+ const u32 F_w4c04 = w[ 4] + MD5C04;
+ const u32 F_w5c05 = w[ 5] + MD5C05;
+ const u32 F_w6c06 = w[ 6] + MD5C06;
+ const u32 F_w7c07 = w[ 7] + MD5C07;
+ const u32 F_w8c08 = w[ 8] + MD5C08;
+ const u32 F_w9c09 = w[ 9] + MD5C09;
+ const u32 F_wac0a = w[10] + MD5C0a;
+ const u32 F_wbc0b = w[11] + MD5C0b;
+ const u32 F_wcc0c = w[12] + MD5C0c;
+ const u32 F_wdc0d = w[13] + MD5C0d;
+ const u32 F_wec0e = w[14] + MD5C0e;
+ const u32 F_wfc0f = w[15] + MD5C0f;
+
+ const u32 G_w1c10 = w[ 1] + MD5C10;
+ const u32 G_w6c11 = w[ 6] + MD5C11;
+ const u32 G_wbc12 = w[11] + MD5C12;
+ const u32 G_w0c13 = 0 + MD5C13;
+ const u32 G_w5c14 = w[ 5] + MD5C14;
+ const u32 G_wac15 = w[10] + MD5C15;
+ const u32 G_wfc16 = w[15] + MD5C16;
+ const u32 G_w4c17 = w[ 4] + MD5C17;
+ const u32 G_w9c18 = w[ 9] + MD5C18;
+ const u32 G_wec19 = w[14] + MD5C19;
+ const u32 G_w3c1a = w[ 3] + MD5C1a;
+ const u32 G_w8c1b = w[ 8] + MD5C1b;
+ const u32 G_wdc1c = w[13] + MD5C1c;
+ const u32 G_w2c1d = w[ 2] + MD5C1d;
+ const u32 G_w7c1e = w[ 7] + MD5C1e;
+ const u32 G_wcc1f = w[12] + MD5C1f;
+
+ const u32 H_w5c20 = w[ 5] + MD5C20;
+ const u32 H_w8c21 = w[ 8] + MD5C21;
+ const u32 H_wbc22 = w[11] + MD5C22;
+ const u32 H_wec23 = w[14] + MD5C23;
+ const u32 H_w1c24 = w[ 1] + MD5C24;
+ const u32 H_w4c25 = w[ 4] + MD5C25;
+ const u32 H_w7c26 = w[ 7] + MD5C26;
+ const u32 H_wac27 = w[10] + MD5C27;
+ const u32 H_wdc28 = w[13] + MD5C28;
+ const u32 H_w0c29 = 0 + MD5C29;
+ const u32 H_w3c2a = w[ 3] + MD5C2a;
+ const u32 H_w6c2b = w[ 6] + MD5C2b;
+ const u32 H_w9c2c = w[ 9] + MD5C2c;
+ const u32 H_wcc2d = w[12] + MD5C2d;
+ const u32 H_wfc2e = w[15] + MD5C2e;
+ const u32 H_w2c2f = w[ 2] + MD5C2f;
+
+ const u32 I_w0c30 = 0 + MD5C30;
+ const u32 I_w7c31 = w[ 7] + MD5C31;
+ const u32 I_wec32 = w[14] + MD5C32;
+ const u32 I_w5c33 = w[ 5] + MD5C33;
+ const u32 I_wcc34 = w[12] + MD5C34;
+ const u32 I_w3c35 = w[ 3] + MD5C35;
+ const u32 I_wac36 = w[10] + MD5C36;
+ const u32 I_w1c37 = w[ 1] + MD5C37;
+ const u32 I_w8c38 = w[ 8] + MD5C38;
+ const u32 I_wfc39 = w[15] + MD5C39;
+ const u32 I_w6c3a = w[ 6] + MD5C3a;
+ const u32 I_wdc3b = w[13] + MD5C3b;
+ const u32 I_w4c3c = w[ 4] + MD5C3c;
+ const u32 I_wbc3d = w[11] + MD5C3d;
+ const u32 I_w2c3e = w[ 2] + MD5C3e;
+ const u32 I_w9c3f = w[ 9] + MD5C3f;
+
+ /**
+ * digest
+ */
+
+ const u32 search[4] =
+ {
+ digests_buf[digests_offset].digest_buf[DGST_R0],
+ digests_buf[digests_offset].digest_buf[DGST_R1],
+ digests_buf[digests_offset].digest_buf[DGST_R2],
+ digests_buf[digests_offset].digest_buf[DGST_R3]
+ };
+
+ /**
+ * reverse
+ */
+
+ u32 a_rev = digests_buf[digests_offset].digest_buf[0];
+ u32 b_rev = digests_buf[digests_offset].digest_buf[1];
+ u32 c_rev = digests_buf[digests_offset].digest_buf[2];
+ u32 d_rev = digests_buf[digests_offset].digest_buf[3];
+
+ MD5_STEP_REV (MD5_I_S, b_rev, c_rev, d_rev, a_rev, w[ 9], MD5C3f, MD5S33);
+ MD5_STEP_REV (MD5_I_S, c_rev, d_rev, a_rev, b_rev, w[ 2], MD5C3e, MD5S32);
+ MD5_STEP_REV (MD5_I_S, d_rev, a_rev, b_rev, c_rev, w[11], MD5C3d, MD5S31);
+ MD5_STEP_REV (MD5_I_S, a_rev, b_rev, c_rev, d_rev, w[ 4], MD5C3c, MD5S30);
+ MD5_STEP_REV (MD5_I_S, b_rev, c_rev, d_rev, a_rev, w[13], MD5C3b, MD5S33);
+ MD5_STEP_REV (MD5_I_S, c_rev, d_rev, a_rev, b_rev, w[ 6], MD5C3a, MD5S32);
+ MD5_STEP_REV (MD5_I_S, d_rev, a_rev, b_rev, c_rev, w[15], MD5C39, MD5S31);
+ MD5_STEP_REV (MD5_I_S, a_rev, b_rev, c_rev, d_rev, w[ 8], MD5C38, MD5S30);
+ MD5_STEP_REV (MD5_I_S, b_rev, c_rev, d_rev, a_rev, w[ 1], MD5C37, MD5S33);
+ MD5_STEP_REV (MD5_I_S, c_rev, d_rev, a_rev, b_rev, w[10], MD5C36, MD5S32);
+ MD5_STEP_REV (MD5_I_S, d_rev, a_rev, b_rev, c_rev, w[ 3], MD5C35, MD5S31);
+ MD5_STEP_REV (MD5_I_S, a_rev, b_rev, c_rev, d_rev, w[12], MD5C34, MD5S30);
+ MD5_STEP_REV (MD5_I_S, b_rev, c_rev, d_rev, a_rev, w[ 5], MD5C33, MD5S33);
+ MD5_STEP_REV (MD5_I_S, c_rev, d_rev, a_rev, b_rev, w[14], MD5C32, MD5S32);
+ MD5_STEP_REV (MD5_I_S, d_rev, a_rev, b_rev, c_rev, w[ 7], MD5C31, MD5S31);
+ MD5_STEP_REV (MD5_I_S, a_rev, b_rev, c_rev, d_rev, 0, MD5C30, MD5S30);
+
+ const u32 pre_cd = c_rev ^ d_rev;
+
+ MD5_STEP_REV1(MD5_H_S, b_rev, c_rev, d_rev, a_rev, w[ 2], MD5C2f, MD5S23);
+ MD5_STEP_REV1(MD5_H_S, c_rev, d_rev, a_rev, b_rev, w[15], MD5C2e, MD5S22);
+
+ /**
+ * loop
+ */
+
+ u32 w0l = w[0];
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ const u32x w0r = words_buf_r[il_pos / VECT_SIZE];
+
+ const u32x w0 = w0l | w0r;
+
+ const u32x pre_d = d_rev;
+ const u32x pre_a = a_rev - w0;
+ const u32x pre_b = b_rev - (pre_a ^ pre_cd);
+ const u32x pre_c = c_rev - (pre_a ^ pre_b ^ pre_d);
+
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
+
+ MD5_STEP (MD5_Fo, a, b, c, d, w0, F_w0c00, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w1c01, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_w2c02, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_w3c03, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_w4c04, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w5c05, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_w6c06, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_w7c07, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_w8c08, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_w9c09, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_wac0a, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_wbc0b, MD5S03);
+ MD5_STEP0(MD5_Fo, a, b, c, d, F_wcc0c, MD5S00);
+ MD5_STEP0(MD5_Fo, d, a, b, c, F_wdc0d, MD5S01);
+ MD5_STEP0(MD5_Fo, c, d, a, b, F_wec0e, MD5S02);
+ MD5_STEP0(MD5_Fo, b, c, d, a, F_wfc0f, MD5S03);
+
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w1c10, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_w6c11, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_wbc12, MD5S12);
+ MD5_STEP (MD5_Go, b, c, d, a, w0, G_w0c13, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w5c14, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_wac15, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_wfc16, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_w4c17, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_w9c18, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_wec19, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_w3c1a, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_w8c1b, MD5S13);
+ MD5_STEP0(MD5_Go, a, b, c, d, G_wdc1c, MD5S10);
+ MD5_STEP0(MD5_Go, d, a, b, c, G_w2c1d, MD5S11);
+ MD5_STEP0(MD5_Go, c, d, a, b, G_w7c1e, MD5S12);
+ MD5_STEP0(MD5_Go, b, c, d, a, G_wcc1f, MD5S13);
+
+ MD5_STEP0(MD5_H , a, b, c, d, H_w5c20, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_w8c21, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_wbc22, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_wec23, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_w1c24, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_w4c25, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_w7c26, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_wac27, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_wdc28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0, H_w0c29, MD5S21);
+ MD5_STEP0(MD5_H , c, d, a, b, H_w3c2a, MD5S22);
+
+ if (MATCHES_NONE_VV (pre_c, c)) continue;
+
+ MD5_STEP0(MD5_H , b, c, d, a, H_w6c2b, MD5S23);
+ MD5_STEP0(MD5_H , a, b, c, d, H_w9c2c, MD5S20);
+ MD5_STEP0(MD5_H , d, a, b, c, H_wcc2d, MD5S21);
+
+ if (MATCHES_NONE_VV (pre_d, d)) continue;
+
+ MD5_STEP0(MD5_H , c, d, a, b, H_wfc2e, MD5S22);
+ MD5_STEP0(MD5_H , b, c, d, a, H_w2c2f, MD5S23);
+
+ MD5_STEP (MD5_I , a, b, c, d, w0, I_w0c30, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_w7c31, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_wec32, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w5c33, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_wcc34, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_w3c35, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_wac36, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w1c37, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_w8c38, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_wfc39, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_w6c3a, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_wdc3b, MD5S33);
+ MD5_STEP0(MD5_I , a, b, c, d, I_w4c3c, MD5S30);
+ MD5_STEP0(MD5_I , d, a, b, c, I_wbc3d, MD5S31);
+ MD5_STEP0(MD5_I , c, d, a, b, I_w2c3e, MD5S32);
+ MD5_STEP0(MD5_I , b, c, d, a, I_w9c3f, MD5S33);
+
+ COMPARE_S_SIMD (a, d, c, b);
+ }
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = 0;
+ w[ 5] = 0;
+ w[ 6] = 0;
+ w[ 7] = 0;
+ w[ 8] = 0;
+ w[ 9] = 0;
+ w[10] = 0;
+ w[11] = 0;
+ w[12] = 0;
+ w[13] = 0;
+ w[14] = pws[gid].i[14];
+ w[15] = 0;
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000m (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);
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = pws[gid].i[ 4];
+ w[ 5] = pws[gid].i[ 5];
+ w[ 6] = pws[gid].i[ 6];
+ w[ 7] = pws[gid].i[ 7];
+ w[ 8] = 0;
+ w[ 9] = 0;
+ w[10] = 0;
+ w[11] = 0;
+ w[12] = 0;
+ w[13] = 0;
+ w[14] = pws[gid].i[14];
+ w[15] = 0;
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000m (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);
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = pws[gid].i[ 4];
+ w[ 5] = pws[gid].i[ 5];
+ w[ 6] = pws[gid].i[ 6];
+ w[ 7] = pws[gid].i[ 7];
+ w[ 8] = pws[gid].i[ 8];
+ w[ 9] = pws[gid].i[ 9];
+ w[10] = pws[gid].i[10];
+ w[11] = pws[gid].i[11];
+ w[12] = pws[gid].i[12];
+ w[13] = pws[gid].i[13];
+ w[14] = pws[gid].i[14];
+ w[15] = pws[gid].i[15];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000m (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);
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = 0;
+ w[ 5] = 0;
+ w[ 6] = 0;
+ w[ 7] = 0;
+ w[ 8] = 0;
+ w[ 9] = 0;
+ w[10] = 0;
+ w[11] = 0;
+ w[12] = 0;
+ w[13] = 0;
+ w[14] = pws[gid].i[14];
+ w[15] = 0;
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000s (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);
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = pws[gid].i[ 4];
+ w[ 5] = pws[gid].i[ 5];
+ w[ 6] = pws[gid].i[ 6];
+ w[ 7] = pws[gid].i[ 7];
+ w[ 8] = 0;
+ w[ 9] = 0;
+ w[10] = 0;
+ w[11] = 0;
+ w[12] = 0;
+ w[13] = 0;
+ w[14] = pws[gid].i[14];
+ w[15] = 0;
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000s (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);
+}
+
+__kernel void m00000_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];
+ w[ 1] = pws[gid].i[ 1];
+ w[ 2] = pws[gid].i[ 2];
+ w[ 3] = pws[gid].i[ 3];
+ w[ 4] = pws[gid].i[ 4];
+ w[ 5] = pws[gid].i[ 5];
+ w[ 6] = pws[gid].i[ 6];
+ w[ 7] = pws[gid].i[ 7];
+ w[ 8] = pws[gid].i[ 8];
+ w[ 9] = pws[gid].i[ 9];
+ w[10] = pws[gid].i[10];
+ w[11] = pws[gid].i[11];
+ w[12] = pws[gid].i[12];
+ w[13] = pws[gid].i[13];
+ w[14] = pws[gid].i[14];
+ w[15] = pws[gid].i[15];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * main
+ */
+
+ m00000s (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);
+}
--- /dev/null
+/**
+ * Author......: Jens Steube <jens.steube@gmail.com>
+ * License.....: MIT
+ */
+
+#define _MD5_
+
+#define NEW_SIMD_CODE
+
+#include "inc_vendor.cl"
+#include "inc_hash_constants.h"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_rp.h"
+#include "inc_rp.cl"
+#include "inc_simd.cl"
+
+void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
+{
+ u32x a = digest[0];
+ u32x b = digest[1];
+ u32x c = digest[2];
+ u32x d = digest[3];
+
+ u32x w0_t = w0[0];
+ 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_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);
+
+ digest[0] += a;
+ digest[1] += b;
+ digest[2] += c;
+ digest[3] += d;
+}
+
+void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
+{
+ w0[0] = w0[0] ^ 0x36363636;
+ w0[1] = w0[1] ^ 0x36363636;
+ w0[2] = w0[2] ^ 0x36363636;
+ w0[3] = w0[3] ^ 0x36363636;
+ w1[0] = w1[0] ^ 0x36363636;
+ w1[1] = w1[1] ^ 0x36363636;
+ w1[2] = w1[2] ^ 0x36363636;
+ w1[3] = w1[3] ^ 0x36363636;
+ w2[0] = w2[0] ^ 0x36363636;
+ w2[1] = w2[1] ^ 0x36363636;
+ w2[2] = w2[2] ^ 0x36363636;
+ w2[3] = w2[3] ^ 0x36363636;
+ w3[0] = w3[0] ^ 0x36363636;
+ w3[1] = w3[1] ^ 0x36363636;
+ w3[2] = w3[2] ^ 0x36363636;
+ w3[3] = w3[3] ^ 0x36363636;
+
+ ipad[0] = MD5M_A;
+ ipad[1] = MD5M_B;
+ ipad[2] = MD5M_C;
+ ipad[3] = MD5M_D;
+
+ md5_transform (w0, w1, w2, w3, ipad);
+
+ w0[0] = w0[0] ^ 0x6a6a6a6a;
+ w0[1] = w0[1] ^ 0x6a6a6a6a;
+ w0[2] = w0[2] ^ 0x6a6a6a6a;
+ w0[3] = w0[3] ^ 0x6a6a6a6a;
+ w1[0] = w1[0] ^ 0x6a6a6a6a;
+ w1[1] = w1[1] ^ 0x6a6a6a6a;
+ w1[2] = w1[2] ^ 0x6a6a6a6a;
+ w1[3] = w1[3] ^ 0x6a6a6a6a;
+ w2[0] = w2[0] ^ 0x6a6a6a6a;
+ w2[1] = w2[1] ^ 0x6a6a6a6a;
+ w2[2] = w2[2] ^ 0x6a6a6a6a;
+ w2[3] = w2[3] ^ 0x6a6a6a6a;
+ w3[0] = w3[0] ^ 0x6a6a6a6a;
+ w3[1] = w3[1] ^ 0x6a6a6a6a;
+ w3[2] = w3[2] ^ 0x6a6a6a6a;
+ w3[3] = w3[3] ^ 0x6a6a6a6a;
+
+ opad[0] = MD5M_A;
+ opad[1] = MD5M_B;
+ opad[2] = MD5M_C;
+ opad[3] = MD5M_D;
+
+ md5_transform (w0, w1, w2, w3, opad);
+}
+
+void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
+{
+ digest[0] = ipad[0];
+ digest[1] = ipad[1];
+ digest[2] = ipad[2];
+ digest[3] = ipad[3];
+
+ md5_transform (w0, w1, w2, w3, digest);
+
+ w0[0] = digest[0];
+ w0[1] = digest[1];
+ w0[2] = digest[2];
+ w0[3] = digest[3];
+ w1[0] = 0x80;
+ w1[1] = 0;
+ w1[2] = 0;
+ w1[3] = 0;
+ w2[0] = 0;
+ w2[1] = 0;
+ w2[2] = 0;
+ w2[3] = 0;
+ w3[0] = 0;
+ w3[1] = 0;
+ w3[2] = (64 + 16) * 8;
+ w3[3] = 0;
+
+ digest[0] = opad[0];
+ digest[1] = opad[1];
+ digest[2] = opad[2];
+ digest[3] = opad[3];
+
+ md5_transform (w0, w1, w2, w3, digest);
+}
+
+__kernel void m00060_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * salt
+ */
+
+ u32 salt_buf0[4];
+ u32 salt_buf1[4];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
+ salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
+ salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
+ salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
+ salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
+ salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
+ salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
+ salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+ /**
+ * pads
+ */
+
+ u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
+
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
+
+ u32x ipad[4];
+ u32x opad[4];
+
+ hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ u32x w0[4] = { 0 };
+ u32x w1[4] = { 0 };
+ u32x w2[4] = { 0 };
+ u32x w3[4] = { 0 };
+
+ const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+ append_0x80_2x4_VV (w0, w1, out_len);
+
+ w0_t[0] = w0[0];
+ w0_t[1] = w0[1];
+ w0_t[2] = w0[2];
+ w0_t[3] = w0[3];
+ w1_t[0] = w1[0];
+ w1_t[1] = w1[1];
+ w1_t[2] = w1[2];
+ w1_t[3] = w1[3];
+ w2_t[0] = 0;
+ w2_t[1] = 0;
+ w2_t[2] = 0;
+ w2_t[3] = 0;
+ w3_t[0] = 0;
+ w3_t[1] = 0;
+ w3_t[2] = (64 + out_len) * 8;
+ w3_t[3] = 0;
+
+ u32x digest[4];
+
+ hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
+
+ COMPARE_M_SIMD (digest[0], digest[3], digest[2], digest[1]);
+ }
+}
+
+__kernel void m00060_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_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)
+{
+}
+
+__kernel void m00060_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_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)
+{
+}
+
+__kernel void m00060_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_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)
+{
+ /**
+ * modifier
+ */
+
+ const u32 lid = get_local_id (0);
+
+ /**
+ * base
+ */
+
+ const u32 gid = get_global_id (0);
+
+ if (gid >= gid_max) return;
+
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
+
+ pw_buf0[0] = pws[gid].i[0];
+ pw_buf0[1] = pws[gid].i[1];
+ pw_buf0[2] = pws[gid].i[2];
+ pw_buf0[3] = pws[gid].i[3];
+ pw_buf1[0] = pws[gid].i[4];
+ pw_buf1[1] = pws[gid].i[5];
+ pw_buf1[2] = pws[gid].i[6];
+ pw_buf1[3] = pws[gid].i[7];
+
+ const u32 pw_len = pws[gid].pw_len;
+
+ /**
+ * salt
+ */
+
+ u32 salt_buf0[4];
+ u32 salt_buf1[4];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
+ salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
+ salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
+ salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
+ salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
+ salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
+ salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
+ salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+ /**
+ * pads
+ */
+
+ u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
+
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
+
+ u32x ipad[4];
+ u32x opad[4];
+
+ hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
+
+ /**
+ * digest
+ */
+
+ const u32 search[4] =
+ {
+ digests_buf[digests_offset].digest_buf[DGST_R0],
+ digests_buf[digests_offset].digest_buf[DGST_R1],
+ digests_buf[digests_offset].digest_buf[DGST_R2],
+ digests_buf[digests_offset].digest_buf[DGST_R3]
+ };
+
+ /**
+ * loop
+ */
+
+ for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
+ {
+ u32x w0[4] = { 0 };
+ u32x w1[4] = { 0 };
+ u32x w2[4] = { 0 };
+ u32x w3[4] = { 0 };
+
+ const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+ append_0x80_2x4_VV (w0, w1, out_len);
+
+ w0_t[0] = w0[0];
+ w0_t[1] = w0[1];
+ w0_t[2] = w0[2];
+ w0_t[3] = w0[3];
+ w1_t[0] = w1[0];
+ w1_t[1] = w1[1];
+ w1_t[2] = w1[2];
+ w1_t[3] = w1[3];
+ w2_t[0] = 0;
+ w2_t[1] = 0;
+ w2_t[2] = 0;
+ w2_t[3] = 0;
+ w3_t[0] = 0;
+ w3_t[1] = 0;
+ w3_t[2] = (64 + out_len) * 8;
+ w3_t[3] = 0;
+
+ u32x digest[4];
+
+ hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
+
+ COMPARE_S_SIMD (digest[0], digest[3], digest[2], digest[1]);
+ }
+}
+
+__kernel void m00060_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_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)
+{
+}
+
+__kernel void m00060_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_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)
+{
+}