/**
- * Author......: Jens Steube <jens.steube@gmail.com>
+ * Authors.....: Jens Steube <jens.steube@gmail.com>
+ * Gabriele Gristina <matrix@hashcat.net>
+ *
* License.....: MIT
*/
#define DGST_R3 3
#include "include/kernel_functions.c"
-#include "types_ocl.c"
-#include "common.c"
+#include "OpenCL/types_ocl.c"
+#include "OpenCL/common.c"
-#include "gpu_aes256_amd.c"
-#include "gpu_twofish256_amd.c"
-#include "gpu_serpent256_amd.c"
+#include "OpenCL/kernel_aes256.c"
+#include "OpenCL/kernel_twofish256.c"
+#include "OpenCL/kernel_serpent256.c"
#define R 10
-#define BOX(S,n,i) (u32) ((S)[(n)][(i)])
-
__constant u32 Ch[8][256] =
{
{
},
};
-#ifdef VECT_SIZE1
-#define BOX(S,n,i) (u32) ((S)[(n)][(i)])
-#endif
+#define BOX(S,n,i) (S)[(n)][(i)]
-static void whirlpool_transform (const u32 w[16], u32 dgst[16], __local u32 s_Ch[8][256], __local u32 s_Cl[8][256])
+static void whirlpool_transform (const u32 w[16], u32 dgst[16], __local u32 (*s_Ch)[256], __local u32 (*s_Cl)[256])
{
const u32 rch[R + 1] =
{
dgst[15] ^= statel[7] ^ w[15];
}
-static void hmac_run2 (const u32 w1[16], const u32 w2[16], const u32 ipad[16], const u32 opad[16], u32 dgst[16], __local u32 s_Ch[8][256], __local u32 s_Cl[8][256])
+static void hmac_run2 (const u32 w1[16], const u32 w2[16], const u32 ipad[16], const u32 opad[16], u32 dgst[16], __local u32 (*s_Ch)[256], __local u32 (*s_Cl)[256])
{
dgst[ 0] = ipad[ 0];
dgst[ 1] = ipad[ 1];
whirlpool_transform (w, dgst, s_Ch, s_Cl);
}
-static void hmac_init (u32 w[16], u32 ipad[16], u32 opad[16], __local u32 s_Ch[8][256], __local u32 s_Cl[8][256])
+static void hmac_init (u32 w[16], u32 ipad[16], u32 opad[16], __local u32 (*s_Ch)[256], __local u32 (*s_Cl)[256])
{
w[ 0] ^= 0x36363636;
w[ 1] ^= 0x36363636;
return r;
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06233_init (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m06233_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
/**
- * base
+ * modifier
*/
const u32 gid = get_global_id (0);
+ const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
+
+ /**
+ * shared
+ */
+
+ __local u32 s_Ch[8][256];
+ __local u32 s_Cl[8][256];
+
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ s_Ch[0][i] = Ch[0][i];
+ s_Ch[1][i] = Ch[1][i];
+ s_Ch[2][i] = Ch[2][i];
+ s_Ch[3][i] = Ch[3][i];
+ s_Ch[4][i] = Ch[4][i];
+ s_Ch[5][i] = Ch[5][i];
+ s_Ch[6][i] = Ch[6][i];
+ s_Ch[7][i] = Ch[7][i];
+
+ s_Cl[0][i] = Cl[0][i];
+ s_Cl[1][i] = Cl[1][i];
+ s_Cl[2][i] = Cl[2][i];
+ s_Cl[3][i] = Cl[3][i];
+ s_Cl[4][i] = Cl[4][i];
+ s_Cl[5][i] = Cl[5][i];
+ s_Cl[6][i] = Cl[6][i];
+ s_Cl[7][i] = Cl[7][i];
+ }
+
+ barrier (CLK_LOCAL_MEM_FENCE);
+
+ if (gid >= gid_max) return;
+
+ /**
+ * base
+ */
u32 w0[4];
w3[2] = u8add (w3[2], esalt_bufs[salt_pos].keyfile_buf[14]);
w3[3] = u8add (w3[3], esalt_bufs[salt_pos].keyfile_buf[15]);
- /**
- * shared mem
- */
-
- __local u32 s_Ch[8][256];
- __local u32 s_Cl[8][256];
-
- const u32 lid = get_local_id (0);
-
- const u32 lid4 = lid * 4;
-
- for (u32 i = 0; i < 8; i++)
- {
- s_Ch[i][lid4 + 0] = Ch[i][lid4 + 0];
- s_Ch[i][lid4 + 1] = Ch[i][lid4 + 1];
- s_Ch[i][lid4 + 2] = Ch[i][lid4 + 2];
- s_Ch[i][lid4 + 3] = Ch[i][lid4 + 3];
- s_Cl[i][lid4 + 0] = Cl[i][lid4 + 0];
- s_Cl[i][lid4 + 1] = Cl[i][lid4 + 1];
- s_Cl[i][lid4 + 2] = Cl[i][lid4 + 2];
- s_Cl[i][lid4 + 3] = Cl[i][lid4 + 3];
- }
-
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
-
/**
* salt
*/
u32 salt_buf1[16];
- salt_buf1[ 0] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 0]);
- salt_buf1[ 1] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 1]);
- salt_buf1[ 2] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 2]);
- salt_buf1[ 3] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 3]);
- salt_buf1[ 4] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 4]);
- salt_buf1[ 5] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 5]);
- salt_buf1[ 6] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 6]);
- salt_buf1[ 7] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 7]);
- salt_buf1[ 8] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 8]);
- salt_buf1[ 9] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 9]);
- salt_buf1[10] = swap_workaround (esalt_bufs[salt_pos].salt_buf[10]);
- salt_buf1[11] = swap_workaround (esalt_bufs[salt_pos].salt_buf[11]);
- salt_buf1[12] = swap_workaround (esalt_bufs[salt_pos].salt_buf[12]);
- salt_buf1[13] = swap_workaround (esalt_bufs[salt_pos].salt_buf[13]);
- salt_buf1[14] = swap_workaround (esalt_bufs[salt_pos].salt_buf[14]);
- salt_buf1[15] = swap_workaround (esalt_bufs[salt_pos].salt_buf[15]);
+ salt_buf1[ 0] = swap32 (esalt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf1[ 1] = swap32 (esalt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf1[ 2] = swap32 (esalt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf1[ 3] = swap32 (esalt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[ 4] = swap32 (esalt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[ 5] = swap32 (esalt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[ 6] = swap32 (esalt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[ 7] = swap32 (esalt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf1[ 8] = swap32 (esalt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf1[ 9] = swap32 (esalt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf1[10] = swap32 (esalt_bufs[salt_pos].salt_buf[10]);
+ salt_buf1[11] = swap32 (esalt_bufs[salt_pos].salt_buf[11]);
+ salt_buf1[12] = swap32 (esalt_bufs[salt_pos].salt_buf[12]);
+ salt_buf1[13] = swap32 (esalt_bufs[salt_pos].salt_buf[13]);
+ salt_buf1[14] = swap32 (esalt_bufs[salt_pos].salt_buf[14]);
+ salt_buf1[15] = swap32 (esalt_bufs[salt_pos].salt_buf[15]);
u32 salt_buf2[16];
u32 w[16];
- w[ 0] = swap_workaround (w0[0]);
- w[ 1] = swap_workaround (w0[1]);
- w[ 2] = swap_workaround (w0[2]);
- w[ 3] = swap_workaround (w0[3]);
- w[ 4] = swap_workaround (w1[0]);
- w[ 5] = swap_workaround (w1[1]);
- w[ 6] = swap_workaround (w1[2]);
- w[ 7] = swap_workaround (w1[3]);
- w[ 8] = swap_workaround (w2[0]);
- w[ 9] = swap_workaround (w2[1]);
- w[10] = swap_workaround (w2[2]);
- w[11] = swap_workaround (w2[3]);
- w[12] = swap_workaround (w3[0]);
- w[13] = swap_workaround (w3[1]);
- w[14] = swap_workaround (w3[2]);
- w[15] = swap_workaround (w3[3]);
+ w[ 0] = swap32 (w0[0]);
+ w[ 1] = swap32 (w0[1]);
+ w[ 2] = swap32 (w0[2]);
+ w[ 3] = swap32 (w0[3]);
+ w[ 4] = swap32 (w1[0]);
+ w[ 5] = swap32 (w1[1]);
+ w[ 6] = swap32 (w1[2]);
+ w[ 7] = swap32 (w1[3]);
+ w[ 8] = swap32 (w2[0]);
+ w[ 9] = swap32 (w2[1]);
+ w[10] = swap32 (w2[2]);
+ w[11] = swap32 (w2[3]);
+ w[12] = swap32 (w3[0]);
+ w[13] = swap32 (w3[1]);
+ w[14] = swap32 (w3[2]);
+ w[15] = swap32 (w3[3]);
u32 ipad[16];
u32 opad[16];
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06233_loop (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m06233_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
- const u32 truecrypt_mdlen = salt_bufs[0].truecrypt_mdlen;
-
- __local u32 s_Ch[8][256];
- __local u32 s_Cl[8][256];
+ /**
+ * modifier
+ */
const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
+
+ /**
+ * shared
+ */
- const u32 lid4 = lid * 4;
+ __local u32 s_Ch[8][256];
+ __local u32 s_Cl[8][256];
- for (u32 i = 0; i < 8; i++)
+ for (u32 i = lid; i < 256; i += lsz)
{
- s_Ch[i][lid4 + 0] = Ch[i][lid4 + 0];
- s_Ch[i][lid4 + 1] = Ch[i][lid4 + 1];
- s_Ch[i][lid4 + 2] = Ch[i][lid4 + 2];
- s_Ch[i][lid4 + 3] = Ch[i][lid4 + 3];
- s_Cl[i][lid4 + 0] = Cl[i][lid4 + 0];
- s_Cl[i][lid4 + 1] = Cl[i][lid4 + 1];
- s_Cl[i][lid4 + 2] = Cl[i][lid4 + 2];
- s_Cl[i][lid4 + 3] = Cl[i][lid4 + 3];
+ s_Ch[0][i] = Ch[0][i];
+ s_Ch[1][i] = Ch[1][i];
+ s_Ch[2][i] = Ch[2][i];
+ s_Ch[3][i] = Ch[3][i];
+ s_Ch[4][i] = Ch[4][i];
+ s_Ch[5][i] = Ch[5][i];
+ s_Ch[6][i] = Ch[6][i];
+ s_Ch[7][i] = Ch[7][i];
+
+ s_Cl[0][i] = Cl[0][i];
+ s_Cl[1][i] = Cl[1][i];
+ s_Cl[2][i] = Cl[2][i];
+ s_Cl[3][i] = Cl[3][i];
+ s_Cl[4][i] = Cl[4][i];
+ s_Cl[5][i] = Cl[5][i];
+ s_Cl[6][i] = Cl[6][i];
+ s_Cl[7][i] = Cl[7][i];
}
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
+ const u32 truecrypt_mdlen = salt_bufs[0].truecrypt_mdlen;
+
u32 ipad[16];
ipad[ 0] = tmps[gid].ipad[ 0];
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06233_comp (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
+__kernel void m06233_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global tc_tmp_t *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 tc_t *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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
/**
* base
u32 ukey1[8];
- ukey1[0] = swap_workaround (tmps[gid].out[ 0]);
- ukey1[1] = swap_workaround (tmps[gid].out[ 1]);
- ukey1[2] = swap_workaround (tmps[gid].out[ 2]);
- ukey1[3] = swap_workaround (tmps[gid].out[ 3]);
- ukey1[4] = swap_workaround (tmps[gid].out[ 4]);
- ukey1[5] = swap_workaround (tmps[gid].out[ 5]);
- ukey1[6] = swap_workaround (tmps[gid].out[ 6]);
- ukey1[7] = swap_workaround (tmps[gid].out[ 7]);
+ ukey1[0] = swap32 (tmps[gid].out[ 0]);
+ ukey1[1] = swap32 (tmps[gid].out[ 1]);
+ ukey1[2] = swap32 (tmps[gid].out[ 2]);
+ ukey1[3] = swap32 (tmps[gid].out[ 3]);
+ ukey1[4] = swap32 (tmps[gid].out[ 4]);
+ ukey1[5] = swap32 (tmps[gid].out[ 5]);
+ ukey1[6] = swap32 (tmps[gid].out[ 6]);
+ ukey1[7] = swap32 (tmps[gid].out[ 7]);
u32 ukey2[8];
- ukey2[0] = swap_workaround (tmps[gid].out[ 8]);
- ukey2[1] = swap_workaround (tmps[gid].out[ 9]);
- ukey2[2] = swap_workaround (tmps[gid].out[10]);
- ukey2[3] = swap_workaround (tmps[gid].out[11]);
- ukey2[4] = swap_workaround (tmps[gid].out[12]);
- ukey2[5] = swap_workaround (tmps[gid].out[13]);
- ukey2[6] = swap_workaround (tmps[gid].out[14]);
- ukey2[7] = swap_workaround (tmps[gid].out[15]);
+ ukey2[0] = swap32 (tmps[gid].out[ 8]);
+ ukey2[1] = swap32 (tmps[gid].out[ 9]);
+ ukey2[2] = swap32 (tmps[gid].out[10]);
+ ukey2[3] = swap32 (tmps[gid].out[11]);
+ ukey2[4] = swap32 (tmps[gid].out[12]);
+ ukey2[5] = swap32 (tmps[gid].out[13]);
+ ukey2[6] = swap32 (tmps[gid].out[14]);
+ ukey2[7] = swap32 (tmps[gid].out[15]);
u32 data[4];
u32 ukey3[8];
- ukey3[0] = swap_workaround (tmps[gid].out[16]);
- ukey3[1] = swap_workaround (tmps[gid].out[17]);
- ukey3[2] = swap_workaround (tmps[gid].out[18]);
- ukey3[3] = swap_workaround (tmps[gid].out[19]);
- ukey3[4] = swap_workaround (tmps[gid].out[20]);
- ukey3[5] = swap_workaround (tmps[gid].out[21]);
- ukey3[6] = swap_workaround (tmps[gid].out[22]);
- ukey3[7] = swap_workaround (tmps[gid].out[23]);
+ ukey3[0] = swap32 (tmps[gid].out[16]);
+ ukey3[1] = swap32 (tmps[gid].out[17]);
+ ukey3[2] = swap32 (tmps[gid].out[18]);
+ ukey3[3] = swap32 (tmps[gid].out[19]);
+ ukey3[4] = swap32 (tmps[gid].out[20]);
+ ukey3[5] = swap32 (tmps[gid].out[21]);
+ ukey3[6] = swap32 (tmps[gid].out[22]);
+ ukey3[7] = swap32 (tmps[gid].out[23]);
u32 ukey4[8];
- ukey4[0] = swap_workaround (tmps[gid].out[24]);
- ukey4[1] = swap_workaround (tmps[gid].out[25]);
- ukey4[2] = swap_workaround (tmps[gid].out[26]);
- ukey4[3] = swap_workaround (tmps[gid].out[27]);
- ukey4[4] = swap_workaround (tmps[gid].out[28]);
- ukey4[5] = swap_workaround (tmps[gid].out[29]);
- ukey4[6] = swap_workaround (tmps[gid].out[30]);
- ukey4[7] = swap_workaround (tmps[gid].out[31]);
+ ukey4[0] = swap32 (tmps[gid].out[24]);
+ ukey4[1] = swap32 (tmps[gid].out[25]);
+ ukey4[2] = swap32 (tmps[gid].out[26]);
+ ukey4[3] = swap32 (tmps[gid].out[27]);
+ ukey4[4] = swap32 (tmps[gid].out[28]);
+ ukey4[5] = swap32 (tmps[gid].out[29]);
+ ukey4[6] = swap32 (tmps[gid].out[30]);
+ ukey4[7] = swap32 (tmps[gid].out[31]);
{
tmp[0] = data[0];
u32 ukey5[8];
- ukey5[0] = swap_workaround (tmps[gid].out[32]);
- ukey5[1] = swap_workaround (tmps[gid].out[33]);
- ukey5[2] = swap_workaround (tmps[gid].out[34]);
- ukey5[3] = swap_workaround (tmps[gid].out[35]);
- ukey5[4] = swap_workaround (tmps[gid].out[36]);
- ukey5[5] = swap_workaround (tmps[gid].out[37]);
- ukey5[6] = swap_workaround (tmps[gid].out[38]);
- ukey5[7] = swap_workaround (tmps[gid].out[39]);
+ ukey5[0] = swap32 (tmps[gid].out[32]);
+ ukey5[1] = swap32 (tmps[gid].out[33]);
+ ukey5[2] = swap32 (tmps[gid].out[34]);
+ ukey5[3] = swap32 (tmps[gid].out[35]);
+ ukey5[4] = swap32 (tmps[gid].out[36]);
+ ukey5[5] = swap32 (tmps[gid].out[37]);
+ ukey5[6] = swap32 (tmps[gid].out[38]);
+ ukey5[7] = swap32 (tmps[gid].out[39]);
u32 ukey6[8];
- ukey6[0] = swap_workaround (tmps[gid].out[40]);
- ukey6[1] = swap_workaround (tmps[gid].out[41]);
- ukey6[2] = swap_workaround (tmps[gid].out[42]);
- ukey6[3] = swap_workaround (tmps[gid].out[43]);
- ukey6[4] = swap_workaround (tmps[gid].out[44]);
- ukey6[5] = swap_workaround (tmps[gid].out[45]);
- ukey6[6] = swap_workaround (tmps[gid].out[46]);
- ukey6[7] = swap_workaround (tmps[gid].out[47]);
+ ukey6[0] = swap32 (tmps[gid].out[40]);
+ ukey6[1] = swap32 (tmps[gid].out[41]);
+ ukey6[2] = swap32 (tmps[gid].out[42]);
+ ukey6[3] = swap32 (tmps[gid].out[43]);
+ ukey6[4] = swap32 (tmps[gid].out[44]);
+ ukey6[5] = swap32 (tmps[gid].out[45]);
+ ukey6[6] = swap32 (tmps[gid].out[46]);
+ ukey6[7] = swap32 (tmps[gid].out[47]);
{
tmp[0] = data[0];