/**
 * Author......: Jens Steube <jens.steube@gmail.com>
 * License.....: MIT
 */

#define _SHA256_

#include "include/constants.h"
#include "include/kernel_vendor.h"

#ifdef  VLIW1
#define VECT_SIZE1
#endif

#ifdef  VLIW2
#define VECT_SIZE1
#endif

#define DGST_R0 3
#define DGST_R1 7
#define DGST_R2 2
#define DGST_R3 6

#include "include/kernel_functions.c"
#include "types_nv.c"
#include "common_nv.c"

#ifdef  VECT_SIZE1
#define VECT_COMPARE_S "check_single_vect1_comp4.c"
#define VECT_COMPARE_M "check_multi_vect1_comp4.c"
#endif

#ifdef  VECT_SIZE2
#define VECT_COMPARE_S "check_single_vect2_comp4.c"
#define VECT_COMPARE_M "check_multi_vect2_comp4.c"
#endif

#ifdef  VECT_SIZE4
#define VECT_COMPARE_S "check_single_vect4_comp4.c"
#define VECT_COMPARE_M "check_multi_vect4_comp4.c"
#endif

__device__ __constant__ u32 k_sha256[64] =
{
  SHA256C00, SHA256C01, SHA256C02, SHA256C03,
  SHA256C04, SHA256C05, SHA256C06, SHA256C07,
  SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
  SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
  SHA256C10, SHA256C11, SHA256C12, SHA256C13,
  SHA256C14, SHA256C15, SHA256C16, SHA256C17,
  SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
  SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
  SHA256C20, SHA256C21, SHA256C22, SHA256C23,
  SHA256C24, SHA256C25, SHA256C26, SHA256C27,
  SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
  SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
  SHA256C30, SHA256C31, SHA256C32, SHA256C33,
  SHA256C34, SHA256C35, SHA256C36, SHA256C37,
  SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
  SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};

__device__ __constant__ bf_t c_bfs[1024];

__device__ static void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
  u32x a = digest[0];
  u32x b = digest[1];
  u32x c = digest[2];
  u32x d = digest[3];
  u32x e = digest[4];
  u32x f = digest[5];
  u32x g = digest[6];
  u32x h = digest[7];

  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];

  #define ROUND_EXPAND()                            \
  {                                                 \
    w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t);  \
    w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t);  \
    w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t);  \
    w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t);  \
    w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t);  \
    w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t);  \
    w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t);  \
    w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t);  \
    w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t);  \
    w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t);  \
    wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t);  \
    wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t);  \
    wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t);  \
    wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t);  \
    we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t);  \
    wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t);  \
  }

  #define ROUND_STEP(i)                                                                   \
  {                                                                                       \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i +  0]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i +  1]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i +  2]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i +  3]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i +  4]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i +  5]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i +  6]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i +  7]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i +  8]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i +  9]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
    SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
  }

  ROUND_STEP (0);

  for (int i = 16; i < 64; i += 16)
  {
    ROUND_EXPAND (); ROUND_STEP (i);
  }

  digest[0] += a;
  digest[1] += b;
  digest[2] += c;
  digest[3] += d;
  digest[4] += e;
  digest[5] += f;
  digest[6] += g;
  digest[7] += h;
}

__device__ static void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
  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] = SHA256M_A;
  ipad[1] = SHA256M_B;
  ipad[2] = SHA256M_C;
  ipad[3] = SHA256M_D;
  ipad[4] = SHA256M_E;
  ipad[5] = SHA256M_F;
  ipad[6] = SHA256M_G;
  ipad[7] = SHA256M_H;

  sha256_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] = SHA256M_A;
  opad[1] = SHA256M_B;
  opad[2] = SHA256M_C;
  opad[3] = SHA256M_D;
  opad[4] = SHA256M_E;
  opad[5] = SHA256M_F;
  opad[6] = SHA256M_G;
  opad[7] = SHA256M_H;

  sha256_transform (w0, w1, w2, w3, opad);
}

__device__ static void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
{
  digest[0] = ipad[0];
  digest[1] = ipad[1];
  digest[2] = ipad[2];
  digest[3] = ipad[3];
  digest[4] = ipad[4];
  digest[5] = ipad[5];
  digest[6] = ipad[6];
  digest[7] = ipad[7];

  sha256_transform (w0, w1, w2, w3, digest);

  w0[0] = digest[0];
  w0[1] = digest[1];
  w0[2] = digest[2];
  w0[3] = digest[3];
  w1[0] = digest[4];
  w1[1] = digest[5];
  w1[2] = digest[6];
  w1[3] = digest[7];
  w2[0] = 0x80000000;
  w2[1] = 0;
  w2[2] = 0;
  w2[3] = 0;
  w3[0] = 0;
  w3[1] = 0;
  w3[2] = 0;
  w3[3] = (64 + 32) * 8;

  digest[0] = opad[0];
  digest[1] = opad[1];
  digest[2] = opad[2];
  digest[3] = opad[3];
  digest[4] = opad[4];
  digest[5] = opad[5];
  digest[6] = opad[6];
  digest[7] = opad[7];

  sha256_transform (w0, w1, w2, w3, digest);
}

__device__ static void m01460m (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
{
  /**
   * modifier
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
  const u32 lid = threadIdx.x;

  /**
   * salt
   */

  u32 salt_buf0[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];

  u32 salt_buf1[4];

  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];

  /**
   * pads
   */

  u32x w0_t[4];

  w0_t[0] = swap_workaround (salt_buf0[0]);
  w0_t[1] = swap_workaround (salt_buf0[1]);
  w0_t[2] = swap_workaround (salt_buf0[2]);
  w0_t[3] = swap_workaround (salt_buf0[3]);

  u32x w1_t[4];

  w1_t[0] = swap_workaround (salt_buf1[0]);
  w1_t[1] = swap_workaround (salt_buf1[1]);
  w1_t[2] = swap_workaround (salt_buf1[2]);
  w1_t[3] = swap_workaround (salt_buf1[3]);

  u32x w2_t[4];

  w2_t[0] = 0;
  w2_t[1] = 0;
  w2_t[2] = 0;
  w2_t[3] = 0;

  u32x w3_t[4];

  w3_t[0] = 0;
  w3_t[1] = 0;
  w3_t[2] = 0;
  w3_t[3] = 0;

  u32x ipad[8];
  u32x opad[8];

  hmac_sha256_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);

  /**
   * loop
   */

  u32x w0l = w0[0];

  for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
  {
    const u32 w0r = c_bfs[il_pos].i;

    w0[0] = w0l | w0r;

    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] = w2[0];
    w2_t[1] = w2[1];
    w2_t[2] = w2[2];
    w2_t[3] = w2[3];
    w3_t[0] = w3[0];
    w3_t[1] = w3[1];
    w3_t[2] = 0;
    w3_t[3] = (64 + pw_len) * 8;

    u32x digest[8];

    hmac_sha256_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);

    const u32x r0 = digest[3];
    const u32x r1 = digest[7];
    const u32x r2 = digest[2];
    const u32x r3 = digest[6];

    #include VECT_COMPARE_M
  }
}

__device__ static void m01460s (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
{
  /**
   * modifier
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
  const u32 lid = threadIdx.x;

  /**
   * salt
   */

  u32 salt_buf0[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];

  u32 salt_buf1[4];

  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];

  /**
   * pads
   */

  u32x w0_t[4];

  w0_t[0] = swap_workaround (salt_buf0[0]);
  w0_t[1] = swap_workaround (salt_buf0[1]);
  w0_t[2] = swap_workaround (salt_buf0[2]);
  w0_t[3] = swap_workaround (salt_buf0[3]);

  u32x w1_t[4];

  w1_t[0] = swap_workaround (salt_buf1[0]);
  w1_t[1] = swap_workaround (salt_buf1[1]);
  w1_t[2] = swap_workaround (salt_buf1[2]);
  w1_t[3] = swap_workaround (salt_buf1[3]);

  u32x w2_t[4];

  w2_t[0] = 0;
  w2_t[1] = 0;
  w2_t[2] = 0;
  w2_t[3] = 0;

  u32x w3_t[4];

  w3_t[0] = 0;
  w3_t[1] = 0;
  w3_t[2] = 0;
  w3_t[3] = 0;

  u32x ipad[8];
  u32x opad[8];

  hmac_sha256_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
   */

  u32x w0l = w0[0];

  for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
  {
    const u32 w0r = c_bfs[il_pos].i;

    w0[0] = w0l | w0r;

    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] = w2[0];
    w2_t[1] = w2[1];
    w2_t[2] = w2[2];
    w2_t[3] = w2[3];
    w3_t[0] = w3[0];
    w3_t[1] = w3[1];
    w3_t[2] = 0;
    w3_t[3] = (64 + pw_len) * 8;

    u32x digest[8];

    hmac_sha256_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);

    const u32x r0 = digest[3];
    const u32x r1 = digest[7];
    const u32x r2 = digest[2];
    const u32x r3 = digest[6];

    #include VECT_COMPARE_S
  }
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = 0;
  w1[1] = 0;
  w1[2] = 0;
  w1[3] = 0;

  u32x w2[4];

  w2[0] = 0;
  w2[1] = 0;
  w2[2] = 0;
  w2[3] = 0;

  u32x w3[4];

  w3[0] = 0;
  w3[1] = 0;
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = pws[gid].i[ 4];
  w1[1] = pws[gid].i[ 5];
  w1[2] = pws[gid].i[ 6];
  w1[3] = pws[gid].i[ 7];

  u32x w2[4];

  w2[0] = 0;
  w2[1] = 0;
  w2[2] = 0;
  w2[3] = 0;

  u32x w3[4];

  w3[0] = 0;
  w3[1] = 0;
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = pws[gid].i[ 4];
  w1[1] = pws[gid].i[ 5];
  w1[2] = pws[gid].i[ 6];
  w1[3] = pws[gid].i[ 7];

  u32x w2[4];

  w2[0] = pws[gid].i[ 8];
  w2[1] = pws[gid].i[ 9];
  w2[2] = pws[gid].i[10];
  w2[3] = pws[gid].i[11];

  u32x w3[4];

  w3[0] = pws[gid].i[12];
  w3[1] = pws[gid].i[13];
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = 0;
  w1[1] = 0;
  w1[2] = 0;
  w1[3] = 0;

  u32x w2[4];

  w2[0] = 0;
  w2[1] = 0;
  w2[2] = 0;
  w2[3] = 0;

  u32x w3[4];

  w3[0] = 0;
  w3[1] = 0;
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = pws[gid].i[ 4];
  w1[1] = pws[gid].i[ 5];
  w1[2] = pws[gid].i[ 6];
  w1[3] = pws[gid].i[ 7];

  u32x w2[4];

  w2[0] = 0;
  w2[1] = 0;
  w2[2] = 0;
  w2[3] = 0;

  u32x w3[4];

  w3[0] = 0;
  w3[1] = 0;
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}

extern "C" __global__ void __launch_bounds__ (256, 1) m01460_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
  /**
   * base
   */

  const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;

  if (gid >= gid_max) return;

  u32x w0[4];

  w0[0] = pws[gid].i[ 0];
  w0[1] = pws[gid].i[ 1];
  w0[2] = pws[gid].i[ 2];
  w0[3] = pws[gid].i[ 3];

  u32x w1[4];

  w1[0] = pws[gid].i[ 4];
  w1[1] = pws[gid].i[ 5];
  w1[2] = pws[gid].i[ 6];
  w1[3] = pws[gid].i[ 7];

  u32x w2[4];

  w2[0] = pws[gid].i[ 8];
  w2[1] = pws[gid].i[ 9];
  w2[2] = pws[gid].i[10];
  w2[3] = pws[gid].i[11];

  u32x w3[4];

  w3[0] = pws[gid].i[12];
  w3[1] = pws[gid].i[13];
  w3[2] = 0;
  w3[3] = 0;

  const u32 pw_len = pws[gid].pw_len;

  /**
   * main
   */

  m01460s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
}