Upgrade kernel to support dynamic local work sizes

[hashcat.git] / OpenCL / m03000_a1.cl

diff --git a/OpenCL/m03000_a1.cl b/OpenCL/m03000_a1.cl
index 20be419..817c779 100644 (file)
--- a/OpenCL/m03000_a1.cl
+++ b/OpenCL/m03000_a1.cl
@@ -469,20 +469,20 @@ static void transform_netntlmv1_key (const u32 w0, const u32 w1, u32 out[2])
   out[1] = as_uint (k1);
 }
 
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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)
 {
   /**
    * modifier
    */
 
+  const u32 gid = get_global_id (0);
   const u32 lid = get_local_id (0);
+  const u32 lsz = get_local_size (0);
 
   /**
    * base
    */
 
-  const u32 gid = get_global_id (0);
-
   u32 wordl0[4];
 
   wordl0[0] = pws[gid].i[ 0];
@@ -523,26 +523,28 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_m04 (__glo
    */
 
   __local u32 s_SPtrans[8][64];
-
   __local u32 s_skb[8][64];
 
-  s_SPtrans[0][lid] = c_SPtrans[0][lid];
-  s_SPtrans[1][lid] = c_SPtrans[1][lid];
-  s_SPtrans[2][lid] = c_SPtrans[2][lid];
-  s_SPtrans[3][lid] = c_SPtrans[3][lid];
-  s_SPtrans[4][lid] = c_SPtrans[4][lid];
-  s_SPtrans[5][lid] = c_SPtrans[5][lid];
-  s_SPtrans[6][lid] = c_SPtrans[6][lid];
-  s_SPtrans[7][lid] = c_SPtrans[7][lid];
-
-  s_skb[0][lid] = c_skb[0][lid];
-  s_skb[1][lid] = c_skb[1][lid];
-  s_skb[2][lid] = c_skb[2][lid];
-  s_skb[3][lid] = c_skb[3][lid];
-  s_skb[4][lid] = c_skb[4][lid];
-  s_skb[5][lid] = c_skb[5][lid];
-  s_skb[6][lid] = c_skb[6][lid];
-  s_skb[7][lid] = c_skb[7][lid];
+  for (u32 i = lid; i < 64; i += lsz)
+  {
+    s_SPtrans[0][i] = c_SPtrans[0][i];
+    s_SPtrans[1][i] = c_SPtrans[1][i];
+    s_SPtrans[2][i] = c_SPtrans[2][i];
+    s_SPtrans[3][i] = c_SPtrans[3][i];
+    s_SPtrans[4][i] = c_SPtrans[4][i];
+    s_SPtrans[5][i] = c_SPtrans[5][i];
+    s_SPtrans[6][i] = c_SPtrans[6][i];
+    s_SPtrans[7][i] = c_SPtrans[7][i];
+
+    s_skb[0][i] = c_skb[0][i];
+    s_skb[1][i] = c_skb[1][i];
+    s_skb[2][i] = c_skb[2][i];
+    s_skb[3][i] = c_skb[3][i];
+    s_skb[4][i] = c_skb[4][i];
+    s_skb[5][i] = c_skb[5][i];
+    s_skb[6][i] = c_skb[6][i];
+    s_skb[7][i] = c_skb[7][i];
+  }
 
   barrier (CLK_LOCAL_MEM_FENCE);
 
@@ -651,28 +653,28 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_m04 (__glo
   }
 }
 
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_m08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 __attribute__((reqd_work_group_size (64, 1, 1))) m03000_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_m16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 __attribute__((reqd_work_group_size (64, 1, 1))) m03000_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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)
 {
   /**
    * modifier
    */
 
+  const u32 gid = get_global_id (0);
   const u32 lid = get_local_id (0);
+  const u32 lsz = get_local_size (0);
 
   /**
    * base
    */
 
-  const u32 gid = get_global_id (0);
-
   u32 wordl0[4];
 
   wordl0[0] = pws[gid].i[ 0];
@@ -713,26 +715,28 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_s04 (__glo
    */
 
   __local u32 s_SPtrans[8][64];
-
   __local u32 s_skb[8][64];
 
-  s_SPtrans[0][lid] = c_SPtrans[0][lid];
-  s_SPtrans[1][lid] = c_SPtrans[1][lid];
-  s_SPtrans[2][lid] = c_SPtrans[2][lid];
-  s_SPtrans[3][lid] = c_SPtrans[3][lid];
-  s_SPtrans[4][lid] = c_SPtrans[4][lid];
-  s_SPtrans[5][lid] = c_SPtrans[5][lid];
-  s_SPtrans[6][lid] = c_SPtrans[6][lid];
-  s_SPtrans[7][lid] = c_SPtrans[7][lid];
-
-  s_skb[0][lid] = c_skb[0][lid];
-  s_skb[1][lid] = c_skb[1][lid];
-  s_skb[2][lid] = c_skb[2][lid];
-  s_skb[3][lid] = c_skb[3][lid];
-  s_skb[4][lid] = c_skb[4][lid];
-  s_skb[5][lid] = c_skb[5][lid];
-  s_skb[6][lid] = c_skb[6][lid];
-  s_skb[7][lid] = c_skb[7][lid];
+  for (u32 i = lid; i < 64; i += lsz)
+  {
+    s_SPtrans[0][i] = c_SPtrans[0][i];
+    s_SPtrans[1][i] = c_SPtrans[1][i];
+    s_SPtrans[2][i] = c_SPtrans[2][i];
+    s_SPtrans[3][i] = c_SPtrans[3][i];
+    s_SPtrans[4][i] = c_SPtrans[4][i];
+    s_SPtrans[5][i] = c_SPtrans[5][i];
+    s_SPtrans[6][i] = c_SPtrans[6][i];
+    s_SPtrans[7][i] = c_SPtrans[7][i];
+
+    s_skb[0][i] = c_skb[0][i];
+    s_skb[1][i] = c_skb[1][i];
+    s_skb[2][i] = c_skb[2][i];
+    s_skb[3][i] = c_skb[3][i];
+    s_skb[4][i] = c_skb[4][i];
+    s_skb[5][i] = c_skb[5][i];
+    s_skb[6][i] = c_skb[6][i];
+    s_skb[7][i] = c_skb[7][i];
+  }
 
   barrier (CLK_LOCAL_MEM_FENCE);
 
@@ -853,10 +857,10 @@ __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_s04 (__glo
   }
 }
 
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m03000_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_s08 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 __attribute__((reqd_work_group_size (64, 1, 1))) m03000_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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 m03000_s16 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_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)
 {
 }