diff --git a/OpenCL/m00670_a1.cl b/OpenCL/m00670_a1.cl
new file mode 100644
index 000000000..8b12f5bfe
--- /dev/null
+++ b/OpenCL/m00670_a1.cl
@@ -0,0 +1,381 @@
+/**
+ * Author......: See docs/credits.txt
+ * License.....: MIT
+ */
+
+#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"
+
+#define CHACHA_CONST_00 0x61707865
+#define CHACHA_CONST_01 0x3320646e
+#define CHACHA_CONST_02 0x79622d32
+#define CHACHA_CONST_03 0x6b206574
+
+#define QR(a, b, c, d)                \
+  do {                                \
+    x[a] = x[a] + x[b];               \
+    x[d] = rotl32(x[d] ^ x[a], 16);   \
+    x[c] = x[c] + x[d];               \
+    x[b] = rotl32(x[b] ^ x[c], 12);   \
+    x[a] = x[a] + x[b];               \
+    x[d] = rotl32(x[d] ^ x[a], 8);    \
+    x[c] = x[c] + x[d];               \
+    x[b] = rotl32(x[b] ^ x[c], 7);    \
+  } while (0);
+
+void chacha20_transform (const u32x w0[4], const u32x w1[4], const u32 position, const u32 iv[2], const u32 plain[4], u32x digest[4])
+{
+  u32x ctx[16];
+  
+  ctx[ 0] = CHACHA_CONST_00;
+  ctx[ 1] = CHACHA_CONST_01;
+  ctx[ 2] = CHACHA_CONST_02;
+  ctx[ 3] = CHACHA_CONST_03;
+  ctx[ 4] = w0[0]; 
+  ctx[ 5] = w0[1];
+  ctx[ 6] = w0[2];
+  ctx[ 7] = w0[3];
+  ctx[ 8] = w1[0];
+  ctx[ 9] = w1[1];
+  ctx[10] = w1[2];
+  ctx[11] = w1[3];
+  ctx[12] = 0;
+  ctx[13] = 0;
+  ctx[14] = iv[1];
+  ctx[15] = iv[0];
+
+  u32x x[16];
+
+  x[ 0] = ctx[ 0];
+  x[ 1] = ctx[ 1];
+  x[ 2] = ctx[ 2];
+  x[ 3] = ctx[ 3];
+  x[ 4] = ctx[ 4];
+  x[ 5] = ctx[ 5];
+  x[ 6] = ctx[ 6];
+  x[ 7] = ctx[ 7];
+  x[ 8] = ctx[ 8];
+  x[ 9] = ctx[ 9];
+  x[10] = ctx[10];
+  x[11] = ctx[11];
+  x[12] = ctx[12];
+  x[13] = ctx[13];
+  x[14] = ctx[14];
+  x[15] = ctx[15];
+
+  for (u8 i = 0; i < 10; ++i) {
+
+    /* Column round */
+    QR(0, 4, 8,  12);
+    QR(1, 5, 9,  13);
+    QR(2, 6, 10, 14);
+    QR(3, 7, 11, 15);
+
+    /* Diagonal round */
+    QR(0, 5, 10, 15);
+    QR(1, 6, 11, 12);
+    QR(2, 7, 8,  13);
+    QR(3, 4, 9,  14);
+  }
+
+  x[ 0] += ctx[ 0];
+  x[ 1] += ctx[ 1];
+  x[ 2] += ctx[ 2];
+  x[ 3] += ctx[ 3];
+  x[ 4] += ctx[ 4];
+  x[ 5] += ctx[ 5];
+  x[ 6] += ctx[ 6];
+  x[ 7] += ctx[ 7];
+  x[ 8] += ctx[ 8];
+  x[ 9] += ctx[ 9];
+  x[10] += ctx[10];
+  x[11] += ctx[11];
+  x[12] += ctx[12];
+  x[13] += ctx[13];
+  x[14] += ctx[14];
+  x[15] += ctx[15];
+
+  digest[1] = plain[0] ^ x[0];
+  digest[0] = plain[1] ^ x[1];
+  digest[3] = plain[2] ^ x[2];
+  digest[2] = plain[3] ^ x[3];
+}  
+
+__kernel void m00670_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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 gid = get_global_id (0);
+  const u32 lid = get_local_id (0);
+
+  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;
+
+  u32 iv[2]    = { 0 };
+  u32 plain[2] = { 0 };
+  u32 position = esalt_bufs->position;
+  u32 plain_length = esalt_bufs->plain_length;
+
+  iv[0] = esalt_bufs->iv[0];
+  iv[1] = esalt_bufs->iv[1];
+
+  plain[0] = esalt_bufs->plain[0];
+  plain[1] = esalt_bufs->plain[1];
+
+  /**
+   * 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 out_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] = wordl3[2] | wordr3[2];
+    w3[3] = wordl3[3] | wordr3[3];
+
+    u32x digest[4] = { 0 };
+
+    chacha20_transform (w0, w1, position, iv, plain, digest);
+
+    const u32x r0 = digest[0];
+    const u32x r1 = digest[1];
+    const u32x r2 = digest[2];
+    const u32x r3 = digest[3];
+
+    COMPARE_M_SIMD(r0, r1, r2, r3);
+  }
+}
+
+__kernel void m00670_m08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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 m00670_m16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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 m00670_s04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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);
+
+  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;
+
+  u32 iv[2]    = { 0 };
+  u32 plain[2] = { 0 };
+  u32 position = esalt_bufs->position;
+  u32 plain_length = esalt_bufs->plain_length;
+
+  iv[0] = esalt_bufs->iv[0];
+  iv[1] = esalt_bufs->iv[1];
+
+  plain[0] = esalt_bufs->plain[0];
+  plain[1] = esalt_bufs->plain[1];
+
+  /**
+   * 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 out_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] = wordl3[2] | wordr3[2];
+    w3[3] = wordl3[3] | wordr3[3];
+
+    u32x digest[4] = { 0 };
+
+    chacha20_transform (w0, w1, position, iv, plain, digest);
+
+    const u32x r0 = digest[0];
+    const u32x r1 = digest[1];
+    const u32x r2 = digest[2];
+    const u32x r3 = digest[3];
+
+    COMPARE_S_SIMD(r0, r1, r2, r3);
+  } 
+}
+
+__kernel void m00670_s08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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 m00670_s16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const chacha20_t *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)
+{
+}
diff --git a/OpenCL/m00670_a3.cl b/OpenCL/m00670_a3.cl
new file mode 100644
index 000000000..87b4cce91
--- /dev/null
+++ b/OpenCL/m00670_a3.cl
@@ -0,0 +1,356 @@
+/**
+ * Author......: See docs/credits.txt
+ * License.....: MIT
+ */
+
+#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 BLAKE2B_FINAL   1
+#define BLAKE2B_UPDATE  0
+
+#define BLAKE2B_G(r,i,a,b,c,d)                \
+  do {                                        \
+    a = a + b + m[blake2b_sigma[r][2*i+0]];   \
+    d = rotr64(d ^ a, 32);                    \
+    c = c + d;                                \
+    b = rotr64(b ^ c, 24);                  \
+    a = a + b + m[blake2b_sigma[r][2*i+1]];   \
+    d = rotr64(d ^ a, 16);                  \
+    c = c + d;                                \
+    b = rotr64(b ^ c, 63);                  \
+  } while(0)
+
+#define BLAKE2B_ROUND(r)                    \
+  do {                                      \
+    BLAKE2B_G(r,0,v[ 0],v[ 4],v[ 8],v[12]); \
+    BLAKE2B_G(r,1,v[ 1],v[ 5],v[ 9],v[13]); \
+    BLAKE2B_G(r,2,v[ 2],v[ 6],v[10],v[14]); \
+    BLAKE2B_G(r,3,v[ 3],v[ 7],v[11],v[15]); \
+    BLAKE2B_G(r,4,v[ 0],v[ 5],v[10],v[15]); \
+    BLAKE2B_G(r,5,v[ 1],v[ 6],v[11],v[12]); \
+    BLAKE2B_G(r,6,v[ 2],v[ 7],v[ 8],v[13]); \
+    BLAKE2B_G(r,7,v[ 3],v[ 4],v[ 9],v[14]); \
+} while(0)
+
+void blake2b_transform(u64x h[8], u64x t[2], u64x f[2], u64x m[16], u64x v[16], const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], const u32x out_len, const u8 isFinal)
+{
+  if (isFinal)
+    f[0] = -1;
+
+  t[0] += hl32_to_64(0, out_len);
+
+  m[0] = hl32_to_64(w0[1], w0[0]);
+  m[1] = hl32_to_64(w0[3], w0[2]);
+  m[2] = hl32_to_64(w1[1], w1[0]);
+  m[3] = hl32_to_64(w1[3], w1[2]);
+  m[4] = hl32_to_64(w2[1], w2[0]);
+  m[5] = hl32_to_64(w2[3], w2[2]);
+  m[6] = hl32_to_64(w3[1], w3[0]);
+  m[7] = hl32_to_64(w3[3], w3[2]);
+  m[8] = 0;
+  m[9] = 0;
+  m[10] = 0;
+  m[11] = 0;
+  m[12] = 0;
+  m[13] = 0;
+  m[14] = 0;
+  m[15] = 0;
+
+  v[ 0] = h[0];
+  v[ 1] = h[1];
+  v[ 2] = h[2];
+  v[ 3] = h[3];
+  v[ 4] = h[4];
+  v[ 5] = h[5];
+  v[ 6] = h[6];
+  v[ 7] = h[7];
+  v[ 8] = BLAKE2B_IV_00;
+  v[ 9] = BLAKE2B_IV_01;
+  v[10] = BLAKE2B_IV_02;
+  v[11] = BLAKE2B_IV_03;
+  v[12] = BLAKE2B_IV_04 ^ t[0];
+  v[13] = BLAKE2B_IV_05 ^ t[1];
+  v[14] = BLAKE2B_IV_06 ^ f[0];
+  v[15] = BLAKE2B_IV_07 ^ f[1];
+
+  const u8 blake2b_sigma[12][16] =
+  {
+    {  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15 } ,
+    { 14, 10,  4,  8,  9, 15, 13,  6,  1, 12,  0,  2, 11,  7,  5,  3 } ,
+    { 11,  8, 12,  0,  5,  2, 15, 13, 10, 14,  3,  6,  7,  1,  9,  4 } ,
+    {  7,  9,  3,  1, 13, 12, 11, 14,  2,  6,  5, 10,  4,  0, 15,  8 } ,
+    {  9,  0,  5,  7,  2,  4, 10, 15, 14,  1, 11, 12,  6,  8,  3, 13 } ,
+    {  2, 12,  6, 10,  0, 11,  8,  3,  4, 13,  7,  5, 15, 14,  1,  9 } ,
+    { 12,  5,  1, 15, 14, 13,  4, 10,  0,  7,  6,  3,  9,  2,  8, 11 } ,
+    { 13, 11,  7, 14, 12,  1,  3,  9,  5,  0, 15,  4,  8,  6,  2, 10 } ,
+    {  6, 15, 14,  9, 11,  3,  0,  8, 12,  2, 13,  7,  1,  4, 10,  5 } ,
+    { 10,  2,  8,  4,  7,  6,  1,  5, 15, 11,  9, 14,  3, 12, 13 , 0 } ,
+    {  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15 } ,
+    { 14, 10,  4,  8,  9, 15, 13,  6,  1, 12,  0,  2, 11,  7,  5,  3 }
+  };
+
+  BLAKE2B_ROUND( 0);
+  BLAKE2B_ROUND( 1);
+  BLAKE2B_ROUND( 2);
+  BLAKE2B_ROUND( 3);
+  BLAKE2B_ROUND( 4);
+  BLAKE2B_ROUND( 5);
+  BLAKE2B_ROUND( 6);
+  BLAKE2B_ROUND( 7);
+  BLAKE2B_ROUND( 8);
+  BLAKE2B_ROUND( 9);
+  BLAKE2B_ROUND(10);
+  BLAKE2B_ROUND(11);
+
+  h[0] = h[0] ^ v[0] ^ v[ 8];
+  h[1] = h[1] ^ v[1] ^ v[ 9];
+  h[2] = h[2] ^ v[2] ^ v[10];
+  h[3] = h[3] ^ v[3] ^ v[11];
+  h[4] = h[4] ^ v[4] ^ v[12];
+  h[5] = h[5] ^ v[5] ^ v[13];
+  h[6] = h[6] ^ v[6] ^ v[14];
+  h[7] = h[7] ^ v[7] ^ v[15];
+}
+
+__kernel void m00600_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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 gid = get_global_id (0);
+  const u32 lid = get_local_id (0);
+
+  u64 tmp_h[8];
+  u64 tmp_t[2];
+  u64 tmp_f[2];
+
+  tmp_h[0] = esalt_bufs->h[0];
+  tmp_h[1] = esalt_bufs->h[1];
+  tmp_h[2] = esalt_bufs->h[2];
+  tmp_h[3] = esalt_bufs->h[3];
+  tmp_h[4] = esalt_bufs->h[4];
+  tmp_h[5] = esalt_bufs->h[5];
+  tmp_h[6] = esalt_bufs->h[6];
+  tmp_h[7] = esalt_bufs->h[7];
+
+  tmp_t[0] = esalt_bufs->t[0];
+  tmp_t[1] = esalt_bufs->t[1];
+  tmp_f[0] = esalt_bufs->f[0];
+  tmp_f[1] = esalt_bufs->f[1];
+
+  /**
+   * loop
+   */
+
+  u32 w0l = pws[gid].i[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 w0x = w0l | w0r;
+
+    u32x w0[4];
+    u32x w1[4];
+    u32x w2[4];
+    u32x w3[4];
+
+    w0[0] = w0x;
+    w0[1] = pws[gid].i[ 1];
+    w0[2] = pws[gid].i[ 2];
+    w0[3] = pws[gid].i[ 3];
+    w1[0] = pws[gid].i[ 4];
+    w1[1] = pws[gid].i[ 5];
+    w1[2] = pws[gid].i[ 6];
+    w1[3] = pws[gid].i[ 7];
+    w2[0] = pws[gid].i[ 8];
+    w2[1] = pws[gid].i[ 9];
+    w2[2] = pws[gid].i[10];
+    w2[3] = pws[gid].i[11];
+    w3[0] = pws[gid].i[12];
+    w3[1] = pws[gid].i[13];
+    w3[2] = pws[gid].i[14];
+    w3[3] = pws[gid].i[15];
+
+    u32x out_len = pws[gid].pw_len;
+
+    u64x digest[8];
+    u64x m[16];
+    u64x v[16];
+
+    u64x h[8];
+    u64x t[2];
+    u64x f[2];
+
+    h[0] = tmp_h[0];
+    h[1] = tmp_h[1];
+    h[2] = tmp_h[2];
+    h[3] = tmp_h[3];
+    h[4] = tmp_h[4];
+    h[5] = tmp_h[5];
+    h[6] = tmp_h[6];
+    h[7] = tmp_h[7];
+
+    t[0] = tmp_t[0];
+    t[1] = tmp_t[1];
+    f[0] = tmp_f[0];
+    f[1] = tmp_f[1];
+
+    blake2b_transform(h, t, f, m, v, w0, w1, w2, w3, out_len, BLAKE2B_FINAL);
+
+    digest[0] = h[0];
+    digest[1] = h[1];
+    digest[2] = h[2];
+    digest[3] = h[3];
+    digest[4] = h[4];
+    digest[5] = h[5];
+    digest[6] = h[6];
+    digest[7] = h[7];
+
+    const u32x r0 = h32_from_64(digest[0]);
+    const u32x r1 = l32_from_64(digest[0]);
+    const u32x r2 = h32_from_64(digest[1]);
+    const u32x r3 = l32_from_64(digest[1]);
+
+    COMPARE_M_SIMD(r0, r1, r2, r3);
+  }
+}  
+
+__kernel void m00600_m08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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 m00600_m16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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 m00600_s04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const u32x *words_buf_r, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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 gid = get_global_id (0);
+  const u32 lid = get_local_id (0);
+
+  u64 tmp_h[8];
+  u64 tmp_t[2];
+  u64 tmp_f[2];
+
+  tmp_h[0] = esalt_bufs->h[0];
+  tmp_h[1] = esalt_bufs->h[1];
+  tmp_h[2] = esalt_bufs->h[2];
+  tmp_h[3] = esalt_bufs->h[3];
+  tmp_h[4] = esalt_bufs->h[4];
+  tmp_h[5] = esalt_bufs->h[5];
+  tmp_h[6] = esalt_bufs->h[6];
+  tmp_h[7] = esalt_bufs->h[7];
+
+  tmp_t[0] = esalt_bufs->t[0];
+  tmp_t[1] = esalt_bufs->t[1];
+  tmp_f[0] = esalt_bufs->f[0];
+  tmp_f[1] = esalt_bufs->f[1];
+
+  /**
+   * 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
+   */
+
+  u32 w0l = pws[gid].i[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 w0x = w0l | w0r;
+    
+    u32x w0[4];
+    u32x w1[4];
+    u32x w2[4];
+    u32x w3[4];
+
+    w0[0] = w0x;
+    w0[1] = pws[gid].i[ 1];
+    w0[2] = pws[gid].i[ 2];
+    w0[3] = pws[gid].i[ 3];
+    w1[0] = pws[gid].i[ 4];
+    w1[1] = pws[gid].i[ 5];
+    w1[2] = pws[gid].i[ 6];
+    w1[3] = pws[gid].i[ 7];
+    w2[0] = pws[gid].i[ 8];
+    w2[1] = pws[gid].i[ 9];
+    w2[2] = pws[gid].i[10];
+    w2[3] = pws[gid].i[11];
+    w3[0] = pws[gid].i[12];
+    w3[1] = pws[gid].i[13];
+    w3[2] = pws[gid].i[14];
+    w3[3] = pws[gid].i[15];
+
+    u32x out_len = pws[gid].pw_len;
+
+    u64x digest[8];
+    u64x m[16];
+    u64x v[16];
+
+    u64x h[8];
+    u64x t[2];
+    u64x f[2];
+
+    h[0] = tmp_h[0];
+    h[1] = tmp_h[1];
+    h[2] = tmp_h[2];
+    h[3] = tmp_h[3];
+    h[4] = tmp_h[4];
+    h[5] = tmp_h[5];
+    h[6] = tmp_h[6];
+    h[7] = tmp_h[7];
+
+    t[0] = tmp_t[0];
+    t[1] = tmp_t[1];
+    f[0] = tmp_f[0];
+    f[1] = tmp_f[1];
+
+    blake2b_transform(h, t, f, m, v, w0, w1, w2, w3, out_len, BLAKE2B_FINAL);
+
+    digest[0] = h[0];
+    digest[1] = h[1];
+    digest[2] = h[2];
+    digest[3] = h[3];
+    digest[4] = h[4];
+    digest[5] = h[5];
+    digest[6] = h[6];
+    digest[7] = h[7];
+
+    const u32x r0 = h32_from_64(digest[0]);
+    const u32x r1 = l32_from_64(digest[0]);
+    const u32x r2 = h32_from_64(digest[1]);
+    const u32x r3 = l32_from_64(digest[1]);
+
+    COMPARE_S_SIMD(r0, r1, r2, r3);
+  }
+}  
+
+__kernel void m00600_s08 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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 m00600_s16 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const blake2_t *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)
+{
+}