X-Git-Url: https://www.flypig.org.uk/git/?a=blobdiff_plain;f=OpenCL%2Fm07300_a0.cl;h=07b7edc2a657691ad2553a2d4ab08851a427d7d7;hb=161a6eb4bc643d8e636e96eda613f5137d30da59;hp=1ca81d8c9fdc1a35cd01c613985768d95410e9b3;hpb=5c07a412ec247a9e2b32c4d79996c4db5768560c;p=hashcat.git

diff --git a/OpenCL/m07300_a0.cl b/OpenCL/m07300_a0.cl
index 1ca81d8..07b7edc 100644
--- a/OpenCL/m07300_a0.cl
+++ b/OpenCL/m07300_a0.cl
@@ -5,47 +5,41 @@
 
 #define _SHA1_
 
-#include "include/constants.h"
-#include "include/kernel_vendor.h"
-
-#define DGST_R0 3
-#define DGST_R1 4
-#define DGST_R2 2
-#define DGST_R3 1
-
-#include "include/kernel_functions.c"
-#include "OpenCL/types_ocl.c"
-#include "OpenCL/common.c"
-#include "include/rp_kernel.h"
-#include "OpenCL/rp.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
-
-static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
+#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 sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
 {
-  u32 A = digest[0];
-  u32 B = digest[1];
-  u32 C = digest[2];
-  u32 D = digest[3];
-  u32 E = digest[4];
-
-  u32 w0_t = w0[0];
-  u32 w1_t = w0[1];
-  u32 w2_t = w0[2];
-  u32 w3_t = w0[3];
-  u32 w4_t = w1[0];
-  u32 w5_t = w1[1];
-  u32 w6_t = w1[2];
-  u32 w7_t = w1[3];
-  u32 w8_t = w2[0];
-  u32 w9_t = w2[1];
-  u32 wa_t = w2[2];
-  u32 wb_t = w2[3];
-  u32 wc_t = w3[0];
-  u32 wd_t = w3[1];
-  u32 we_t = w3[2];
-  u32 wf_t = w3[3];
+  u32x A = digest[0];
+  u32x B = digest[1];
+  u32x C = digest[2];
+  u32x D = digest[3];
+  u32x E = digest[4];
+
+  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];
 
   #undef K
   #define K SHA1C00
@@ -150,7 +144,7 @@ static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], c
   digest[4] += E;
 }
 
-static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
+void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
 {
   w0[0] = w0[0] ^ 0x36363636;
   w0[1] = w0[1] ^ 0x36363636;
@@ -203,7 +197,7 @@ static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[
   sha1_transform (w0, w1, w2, w3, opad);
 }
 
-static void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5])
+void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
 {
   digest[0] = ipad[0];
   digest[1] = ipad[1];
@@ -239,7 +233,7 @@ static void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[
   sha1_transform (w0, w1, w2, w3, digest);
 }
 
-__kernel void m07300_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 rakp_t *rakp_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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
@@ -256,18 +250,16 @@ __kernel void m07300_m04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
   if (gid >= gid_max) return;
 
   u32 pw_buf0[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];
-
   u32 pw_buf1[4];
 
-  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];
+  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;
 
@@ -281,74 +273,32 @@ __kernel void m07300_m04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
    * loop
    */
 
-  for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
   {
-    u32 w0[4];
-
-    w0[0] = pw_buf0[0];
-    w0[1] = pw_buf0[1];
-    w0[2] = pw_buf0[2];
-    w0[3] = pw_buf0[3];
-
-    u32 w1[4];
-
-    w1[0] = pw_buf1[0];
-    w1[1] = pw_buf1[1];
-    w1[2] = pw_buf1[2];
-    w1[3] = pw_buf1[3];
-
-    u32 w2[4];
-
-    w2[0] = 0;
-    w2[1] = 0;
-    w2[2] = 0;
-    w2[3] = 0;
-
-    u32 w3[4];
-
-    w3[0] = 0;
-    w3[1] = 0;
-    w3[2] = 0;
-    w3[3] = 0;
-
-    const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
+    u32x w0[4] = { 0 };
+    u32x w1[4] = { 0 };
+    u32x w2[4] = { 0 };
+    u32x w3[4] = { 0 };
+
+    apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+    w0[0] = swap32 (w0[0]);
+    w0[1] = swap32 (w0[1]);
+    w0[2] = swap32 (w0[2]);
+    w0[3] = swap32 (w0[3]);
+    w1[0] = swap32 (w1[0]);
+    w1[1] = swap32 (w1[1]);
+    w1[2] = swap32 (w1[2]);
+    w1[3] = swap32 (w1[3]);
 
     /**
-     * pads
+     * RAKP
      */
 
-    u32 w0_t[4];
-
-    w0_t[0] = swap32 (w0[0]);
-    w0_t[1] = swap32 (w0[1]);
-    w0_t[2] = swap32 (w0[2]);
-    w0_t[3] = swap32 (w0[3]);
-
-    u32 w1_t[4];
-
-    w1_t[0] = swap32 (w1[0]);
-    w1_t[1] = swap32 (w1[1]);
-    w1_t[2] = swap32 (w1[2]);
-    w1_t[3] = swap32 (w1[3]);
-
-    u32 w2_t[4];
+    u32x ipad[5];
+    u32x opad[5];
 
-    w2_t[0] = 0;
-    w2_t[1] = 0;
-    w2_t[2] = 0;
-    w2_t[3] = 0;
-
-    u32 w3_t[4];
-
-    w3_t[0] = 0;
-    w3_t[1] = 0;
-    w3_t[2] = 0;
-    w3_t[3] = 0;
-
-    u32 ipad[5];
-    u32 opad[5];
-
-    hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
+    hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
 
     int esalt_size = esalt_len;
 
@@ -357,65 +307,60 @@ __kernel void m07300_m04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
 
     for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
     {
-      w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
-      w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
-      w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
-      w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
-      w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
-      w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
-      w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
-      w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
-      w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
-      w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
-      w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
-      w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
-      w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
-      w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
-      w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
-      w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
-
-      sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
+      w0[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
+      w0[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
+      w0[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
+      w0[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
+      w1[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
+      w1[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
+      w1[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
+      w1[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
+      w2[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
+      w2[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
+      w2[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
+      w2[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
+      w3[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
+      w3[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
+      w3[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
+      w3[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
+
+      sha1_transform (w0, w1, w2, w3, ipad);
     }
 
-    w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
-    w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
-    w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
-    w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
-    w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
-    w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
-    w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
-    w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
-    w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
-    w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
-    w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
-    w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
-    w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
-    w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
-    w3_t[2] = 0;
-    w3_t[3] = (64 + esalt_size) * 8;
-
-    u32 digest[5];
-
-    hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
-    const u32 r0 = digest[3];
-    const u32 r1 = digest[4];
-    const u32 r2 = digest[2];
-    const u32 r3 = digest[1];
-
-    #include COMPARE_M
+    w0[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
+    w0[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
+    w0[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
+    w0[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
+    w1[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
+    w1[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
+    w1[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
+    w1[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
+    w2[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
+    w2[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
+    w2[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
+    w2[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
+    w3[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
+    w3[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
+    w3[2] = 0;
+    w3[3] = (64 + esalt_size) * 8;
+
+    u32x digest[5];
+
+    hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+    COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
   }
 }
 
-__kernel void m07300_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 rakp_t *rakp_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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 m07300_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 rakp_t *rakp_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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 m07300_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 rakp_t *rakp_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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
@@ -432,18 +377,16 @@ __kernel void m07300_s04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
   if (gid >= gid_max) return;
 
   u32 pw_buf0[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];
-
   u32 pw_buf1[4];
 
-  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];
+  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;
 
@@ -469,74 +412,32 @@ __kernel void m07300_s04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
    * loop
    */
 
-  for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
   {
-    u32 w0[4];
-
-    w0[0] = pw_buf0[0];
-    w0[1] = pw_buf0[1];
-    w0[2] = pw_buf0[2];
-    w0[3] = pw_buf0[3];
-
-    u32 w1[4];
-
-    w1[0] = pw_buf1[0];
-    w1[1] = pw_buf1[1];
-    w1[2] = pw_buf1[2];
-    w1[3] = pw_buf1[3];
-
-    u32 w2[4];
-
-    w2[0] = 0;
-    w2[1] = 0;
-    w2[2] = 0;
-    w2[3] = 0;
-
-    u32 w3[4];
-
-    w3[0] = 0;
-    w3[1] = 0;
-    w3[2] = 0;
-    w3[3] = 0;
-
-    const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
+    u32x w0[4] = { 0 };
+    u32x w1[4] = { 0 };
+    u32x w2[4] = { 0 };
+    u32x w3[4] = { 0 };
+
+    apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+
+    w0[0] = swap32 (w0[0]);
+    w0[1] = swap32 (w0[1]);
+    w0[2] = swap32 (w0[2]);
+    w0[3] = swap32 (w0[3]);
+    w1[0] = swap32 (w1[0]);
+    w1[1] = swap32 (w1[1]);
+    w1[2] = swap32 (w1[2]);
+    w1[3] = swap32 (w1[3]);
 
     /**
-     * pads
+     * RAKP
      */
 
-    u32 w0_t[4];
+    u32x ipad[5];
+    u32x opad[5];
 
-    w0_t[0] = swap32 (w0[0]);
-    w0_t[1] = swap32 (w0[1]);
-    w0_t[2] = swap32 (w0[2]);
-    w0_t[3] = swap32 (w0[3]);
-
-    u32 w1_t[4];
-
-    w1_t[0] = swap32 (w1[0]);
-    w1_t[1] = swap32 (w1[1]);
-    w1_t[2] = swap32 (w1[2]);
-    w1_t[3] = swap32 (w1[3]);
-
-    u32 w2_t[4];
-
-    w2_t[0] = 0;
-    w2_t[1] = 0;
-    w2_t[2] = 0;
-    w2_t[3] = 0;
-
-    u32 w3_t[4];
-
-    w3_t[0] = 0;
-    w3_t[1] = 0;
-    w3_t[2] = 0;
-    w3_t[3] = 0;
-
-    u32 ipad[5];
-    u32 opad[5];
-
-    hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
+    hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
 
     int esalt_size = esalt_len;
 
@@ -545,60 +446,55 @@ __kernel void m07300_s04 (__global pw_t *pws, __global kernel_rule_t *  rules_bu
 
     for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
     {
-      w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
-      w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
-      w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
-      w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
-      w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
-      w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
-      w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
-      w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
-      w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
-      w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
-      w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
-      w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
-      w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
-      w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
-      w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
-      w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
-
-      sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
+      w0[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
+      w0[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
+      w0[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
+      w0[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
+      w1[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
+      w1[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
+      w1[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
+      w1[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
+      w2[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
+      w2[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
+      w2[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
+      w2[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
+      w3[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
+      w3[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
+      w3[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
+      w3[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
+
+      sha1_transform (w0, w1, w2, w3, ipad);
     }
 
-    w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
-    w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
-    w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
-    w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
-    w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
-    w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
-    w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
-    w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
-    w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
-    w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
-    w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
-    w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
-    w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
-    w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
-    w3_t[2] = 0;
-    w3_t[3] = (64 + esalt_size) * 8;
-
-    u32 digest[5];
-
-    hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
-    const u32 r0 = digest[3];
-    const u32 r1 = digest[4];
-    const u32 r2 = digest[2];
-    const u32 r3 = digest[1];
-
-    #include COMPARE_S
+    w0[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  0];
+    w0[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  1];
+    w0[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  2];
+    w0[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  3];
+    w1[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  4];
+    w1[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  5];
+    w1[2] = rakp_bufs[salt_pos].salt_buf[esalt_off +  6];
+    w1[3] = rakp_bufs[salt_pos].salt_buf[esalt_off +  7];
+    w2[0] = rakp_bufs[salt_pos].salt_buf[esalt_off +  8];
+    w2[1] = rakp_bufs[salt_pos].salt_buf[esalt_off +  9];
+    w2[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
+    w2[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
+    w3[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
+    w3[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
+    w3[2] = 0;
+    w3[3] = (64 + esalt_size) * 8;
+
+    u32x digest[5];
+
+    hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+    COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
   }
 }
 
-__kernel void m07300_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 rakp_t *rakp_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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 m07300_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 rakp_t *rakp_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m07300_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 rakp_t *rakp_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)
 {
 }