From dbf74b68b29738f91f26b20eb66a4706211bd7ed Mon Sep 17 00:00:00 2001
From: jsteube <jens.steube@gmail.com>
Date: Sun, 16 Jul 2017 19:38:38 +0200
Subject: [PATCH] Add -L kernel for Domain Cached Credentials (DCC), MS Cache

---
 OpenCL/m01100_a0-pure.cl | 190 +++++++++++++++++++++++++++++++++++++
 OpenCL/m01100_a1-pure.cl | 166 ++++++++++++++++++++++++++++++++
 OpenCL/m01100_a3-pure.cl | 200 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 556 insertions(+)
 create mode 100644 OpenCL/m01100_a0-pure.cl
 create mode 100644 OpenCL/m01100_a1-pure.cl
 create mode 100644 OpenCL/m01100_a3-pure.cl

diff --git a/OpenCL/m01100_a0-pure.cl b/OpenCL/m01100_a0-pure.cl
new file mode 100644
index 000000000..e257ed697
--- /dev/null
+++ b/OpenCL/m01100_a0-pure.cl
@@ -0,0 +1,190 @@
+/**
+ * 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_scalar.cl"
+#include "inc_hash_md4.cl"
+
+__kernel void m01100_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32 w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    // todo: add rules engine
+
+    md4_ctx_t ctx0;
+
+    md4_init (&ctx0);
+
+    md4_update_utf16le (&ctx0, w, pw_len);
+
+    md4_final (&ctx0);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SCALAR (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m01100_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * 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]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32 w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    // todo: add rules engine
+
+    md4_ctx_t ctx0;
+
+    md4_init (&ctx0);
+
+    md4_update_utf16le (&ctx0, w, pw_len);
+
+    md4_final (&ctx0);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SCALAR (r0, r1, r2, r3);
+  }
+}
diff --git a/OpenCL/m01100_a1-pure.cl b/OpenCL/m01100_a1-pure.cl
new file mode 100644
index 000000000..56fa3bfda
--- /dev/null
+++ b/OpenCL/m01100_a1-pure.cl
@@ -0,0 +1,166 @@
+/**
+ * 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_scalar.cl"
+#include "inc_hash_md4.cl"
+
+__kernel void m01100_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  md4_ctx_t ctx0;
+
+  md4_init (&ctx0);
+
+  md4_update_global_utf16le (&ctx0, pws[gid].i, pws[gid].pw_len);
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    md4_ctx_t ctx1 = ctx0;
+
+    md4_update_global_utf16le (&ctx1, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
+
+    md4_final (&ctx1);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx1.h[0];
+    ctx.w0[1] = ctx1.h[1];
+    ctx.w0[2] = ctx1.h[2];
+    ctx.w0[3] = ctx1.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SCALAR (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m01100_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * 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]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  md4_ctx_t ctx0;
+
+  md4_init (&ctx0);
+
+  md4_update_global_utf16le (&ctx0, pws[gid].i, pws[gid].pw_len);
+
+  /**
+   * loop
+   */
+
+  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
+  {
+    md4_ctx_t ctx1 = ctx0;
+
+    md4_update_global_utf16le (&ctx1, combs_buf[il_pos].i, combs_buf[il_pos].pw_len);
+
+    md4_final (&ctx1);
+
+    md4_ctx_t ctx;
+
+    md4_init (&ctx);
+
+    ctx.w0[0] = ctx1.h[0];
+    ctx.w0[1] = ctx1.h[1];
+    ctx.w0[2] = ctx1.h[2];
+    ctx.w0[3] = ctx1.h[3];
+
+    ctx.len = 16;
+
+    md4_update_utf16le (&ctx, s, salt_len);
+
+    md4_final (&ctx);
+
+    const u32 r0 = ctx.h[DGST_R0];
+    const u32 r1 = ctx.h[DGST_R1];
+    const u32 r2 = ctx.h[DGST_R2];
+    const u32 r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SCALAR (r0, r1, r2, r3);
+  }
+}
diff --git a/OpenCL/m01100_a3-pure.cl b/OpenCL/m01100_a3-pure.cl
new file mode 100644
index 000000000..461b834a8
--- /dev/null
+++ b/OpenCL/m01100_a3-pure.cl
@@ -0,0 +1,200 @@
+/**
+ * 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"
+#include "inc_hash_md4.cl"
+
+__kernel void m01100_mxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant 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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32x w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  /**
+   * loop
+   */
+
+  u32x 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;
+
+    w[0] = w0;
+
+    md4_ctx_vector_t ctx0;
+
+    md4_init_vector (&ctx0);
+
+    md4_update_vector_utf16le (&ctx0, w, pw_len);
+
+    md4_final_vector (&ctx0);
+
+    md4_ctx_vector_t ctx;
+
+    md4_init_vector (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_vector_utf16le (&ctx, s, salt_len);
+
+    md4_final_vector (&ctx);
+
+    const u32x r0 = ctx.h[DGST_R0];
+    const u32x r1 = ctx.h[DGST_R1];
+    const u32x r2 = ctx.h[DGST_R2];
+    const u32x r3 = ctx.h[DGST_R3];
+
+    COMPARE_M_SIMD (r0, r1, r2, r3);
+  }
+}
+
+__kernel void m01100_sxx (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __constant 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 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);
+  const u32 gid = get_global_id (0);
+
+  if (gid >= gid_max) return;
+
+  /**
+   * 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]
+  };
+
+  /**
+   * base
+   */
+
+  const u32 pw_len = pws[gid].pw_len;
+
+  const u32 pw_lenv = ceil ((float) pw_len / 4);
+
+  u32x w[64] = { 0 };
+
+  for (int idx = 0; idx < pw_lenv; idx++)
+  {
+    w[idx] = pws[gid].i[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  const u32 salt_len = salt_bufs[salt_pos].salt_len;
+
+  const u32 salt_lenv = ceil ((float) salt_len / 4);
+
+  u32x s[64] = { 0 };
+
+  for (int idx = 0; idx < salt_lenv; idx++)
+  {
+    s[idx] = salt_bufs[salt_pos].salt_buf[idx];
+
+    barrier (CLK_GLOBAL_MEM_FENCE);
+  }
+
+  /**
+   * loop
+   */
+
+  u32x 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;
+
+    w[0] = w0;
+
+    md4_ctx_vector_t ctx0;
+
+    md4_init_vector (&ctx0);
+
+    md4_update_vector_utf16le (&ctx0, w, pw_len);
+
+    md4_final_vector (&ctx0);
+
+    md4_ctx_vector_t ctx;
+
+    md4_init_vector (&ctx);
+
+    ctx.w0[0] = ctx0.h[0];
+    ctx.w0[1] = ctx0.h[1];
+    ctx.w0[2] = ctx0.h[2];
+    ctx.w0[3] = ctx0.h[3];
+
+    ctx.len = 16;
+
+    md4_update_vector_utf16le (&ctx, s, salt_len);
+
+    md4_final_vector (&ctx);
+
+    const u32x r0 = ctx.h[DGST_R0];
+    const u32x r1 = ctx.h[DGST_R1];
+    const u32x r2 = ctx.h[DGST_R2];
+    const u32x r3 = ctx.h[DGST_R3];
+
+    COMPARE_S_SIMD (r0, r1, r2, r3);
+  }
+}