if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
if (gid >= gid_max) return;
- u32 pws0[4] = { 0 };
- u32 pws1[4] = { 0 };
-
- pws0[0] = pws[gid].i[0];
- pws0[1] = pws[gid].i[1];
- pws0[2] = pws[gid].i[2];
- pws0[3] = pws[gid].i[3];
- pws1[0] = pws[gid].i[4];
- pws1[1] = pws[gid].i[5];
- pws1[2] = pws[gid].i[6];
- pws1[3] = pws[gid].i[7];
+ 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;
u32x wordl2[4] = { 0 };
u32x wordl3[4] = { 0 };
- wordl0[0] = pws0[0];
- wordl0[1] = pws0[1];
- wordl0[2] = pws0[2];
- wordl0[3] = pws0[3];
- wordl1[0] = pws1[0];
- wordl1[1] = pws1[1];
- wordl1[2] = pws1[2];
- wordl1[3] = pws1[3];
+ 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 };
if (gid >= gid_max) return;
- u32 pws0[4] = { 0 };
- u32 pws1[4] = { 0 };
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- pws0[0] = pws[gid].i[0];
- pws0[1] = pws[gid].i[1];
- pws0[2] = pws[gid].i[2];
- pws0[3] = pws[gid].i[3];
- pws1[0] = pws[gid].i[4];
- pws1[1] = pws[gid].i[5];
- pws1[2] = pws[gid].i[6];
- pws1[3] = pws[gid].i[7];
+ 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;
u32x wordl2[4] = { 0 };
u32x wordl3[4] = { 0 };
- wordl0[0] = pws0[0];
- wordl0[1] = pws0[1];
- wordl0[2] = pws0[2];
- wordl0[3] = pws0[3];
- wordl1[0] = pws1[0];
- wordl1[1] = pws1[1];
- wordl1[2] = pws1[2];
- wordl1[3] = pws1[3];
+ 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 };
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
*/
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32x s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32x s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32x s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
switch_buffer_by_offset_le (s0, s1, s2, s3, out_len);
w0[1] |= s0[1];
w0[2] |= s0[2];
w0[3] |= s0[3];
-
w1[0] |= s1[0];
w1[1] |= s1[1];
w1[2] |= s1[2];
w1[3] |= s1[3];
-
w2[0] |= s2[0];
w2[1] |= s2[1];
w2[2] |= s2[2];
w2[3] |= s2[3];
-
w3[0] |= s3[0];
w3[1] |= s3[1];
w3[2] |= s3[2];
w3[3] |= s3[3];
- append_0x80_4x4 (w0, w1, w2, w3, pw_salt_len);
-
/**
* sha1
*/
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
*/
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32x s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32x s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32x s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
switch_buffer_by_offset_le (s0, s1, s2, s3, out_len);
w0[1] |= s0[1];
w0[2] |= s0[2];
w0[3] |= s0[3];
-
w1[0] |= s1[0];
w1[1] |= s1[1];
w1[2] |= s1[2];
w1[3] |= s1[3];
-
w2[0] |= s2[0];
w2[1] |= s2[1];
w2[2] |= s2[2];
w2[3] |= s2[3];
-
w3[0] |= s3[0];
w3[1] |= s3[1];
w3[2] |= s3[2];
w3[3] |= s3[3];
- append_0x80_4x4 (w0, w1, w2, w3, pw_salt_len);
-
/**
* sha1
*/
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
+#include "OpenCL/simd.c"
__kernel void m00110_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)
{
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr1[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
-
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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];
/**
* append salt
*/
- u32 s0[4];
+ u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32 s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32 s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32 s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, pw_len);
-
- const u32 pw_salt_len = pw_len + salt_len;
-
- u32 w0[4];
-
- w0[0] = wordl0[0] | wordr0[0] | s0[0];
- w0[1] = wordl0[1] | wordr0[1] | s0[1];
- w0[2] = wordl0[2] | wordr0[2] | s0[2];
- w0[3] = wordl0[3] | wordr0[3] | s0[3];
-
- u32 w1[4];
-
- w1[0] = wordl1[0] | wordr1[0] | s1[0];
- w1[1] = wordl1[1] | wordr1[1] | s1[1];
- w1[2] = wordl1[2] | wordr1[2] | s1[2];
- w1[3] = wordl1[3] | wordr1[3] | s1[3];
-
- u32 w2[4];
-
- w2[0] = wordl2[0] | wordr2[0] | s2[0];
- w2[1] = wordl2[1] | wordr2[1] | s2[1];
- w2[2] = wordl2[2] | wordr2[2] | s2[2];
- w2[3] = wordl2[3] | wordr2[3] | s2[3];
-
- u32 w3[4];
-
- w3[0] = wordl3[0] | wordr3[0] | s3[0];
- w3[1] = wordl3[1] | wordr3[1] | s3[1];
- w3[2] = 0;
- w3[3] = 0;
-
- append_0x80_4x4 (w0, w1, w2, w3, pw_salt_len);
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
+
+ const u32x pw_salt_len = pw_len + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- u32 w0_t = swap32 (w0[0]);
- u32 w1_t = swap32 (w0[1]);
- u32 w2_t = swap32 (w0[2]);
- u32 w3_t = swap32 (w0[3]);
- u32 w4_t = swap32 (w1[0]);
- u32 w5_t = swap32 (w1[1]);
- u32 w6_t = swap32 (w1[2]);
- u32 w7_t = swap32 (w1[3]);
- u32 w8_t = swap32 (w2[0]);
- u32 w9_t = swap32 (w2[1]);
- u32 wa_t = swap32 (w2[2]);
- u32 wb_t = swap32 (w2[3]);
- u32 wc_t = swap32 (w3[0]);
- u32 wd_t = swap32 (w3[1]);
- u32 we_t = 0;
- u32 wf_t = pw_salt_len * 8;
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ COMPARE_M_SIMD (d, e, c, b);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* reverse
*/
- const u32 e_rev = rotl32 (search[1], 2u);
+ const u32 e_rev = rotl32_S (search[1], 2u);
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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];
/**
* append salt
*/
- u32 s0[4];
+ u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32 s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32 s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32 s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, pw_len);
-
- const u32 pw_salt_len = pw_len + salt_len;
-
- u32 w0[4];
-
- w0[0] = wordl0[0] | wordr0[0] | s0[0];
- w0[1] = wordl0[1] | wordr0[1] | s0[1];
- w0[2] = wordl0[2] | wordr0[2] | s0[2];
- w0[3] = wordl0[3] | wordr0[3] | s0[3];
-
- u32 w1[4];
-
- w1[0] = wordl1[0] | wordr1[0] | s1[0];
- w1[1] = wordl1[1] | wordr1[1] | s1[1];
- w1[2] = wordl1[2] | wordr1[2] | s1[2];
- w1[3] = wordl1[3] | wordr1[3] | s1[3];
-
- u32 w2[4];
-
- w2[0] = wordl2[0] | wordr2[0] | s2[0];
- w2[1] = wordl2[1] | wordr2[1] | s2[1];
- w2[2] = wordl2[2] | wordr2[2] | s2[2];
- w2[3] = wordl2[3] | wordr2[3] | s2[3];
-
- u32 w3[4];
-
- w3[0] = wordl3[0] | wordr3[0] | s3[0];
- w3[1] = wordl3[1] | wordr3[1] | s3[1];
- w3[2] = 0;
- w3[3] = 0;
-
- append_0x80_4x4 (w0, w1, w2, w3, pw_salt_len);
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
+
+ const u32x pw_salt_len = pw_len + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- u32 w0_t = swap32 (w0[0]);
- u32 w1_t = swap32 (w0[1]);
- u32 w2_t = swap32 (w0[2]);
- u32 w3_t = swap32 (w0[3]);
- u32 w4_t = swap32 (w1[0]);
- u32 w5_t = swap32 (w1[1]);
- u32 w6_t = swap32 (w1[2]);
- u32 w7_t = swap32 (w1[3]);
- u32 w8_t = swap32 (w2[0]);
- u32 w9_t = swap32 (w2[1]);
- u32 wa_t = swap32 (w2[2]);
- u32 wb_t = swap32 (w2[3]);
- u32 wc_t = swap32 (w3[0]);
- u32 wd_t = swap32 (w3[1]);
- u32 we_t = 0;
- u32 wf_t = pw_salt_len * 8;
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
- if (allx (e != e_rev)) continue;
+ if (MATCHES_NONE_VS (e, e_rev)) continue;
wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ COMPARE_S_SIMD (d, e, c, b);
}
}
*/
u32 salt_buf0[4];
+ u32 salt_buf1[4];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
-
- u32 salt_buf2[4];
-
salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
- u32 salt_buf3[4];
-
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
switch_buffer_by_offset_le_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, pw_len);
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
+ /**
+ * 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
*/
const u32 out_salt_len = out_len + salt_len;
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- 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] = w3[2];
- w3_t[3] = w3[3];
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ append_0x80_4x4 (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_M_SIMD (d, e, c, b);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_salt_len = out_len + salt_len;
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- 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] = w3[2];
- w3_t[3] = w3[3];
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ append_0x80_4x4 (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
if (MATCHES_NONE_VS (e, e_rev)) continue;
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_S_SIMD (d, e, c, b);
}
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
+#include "OpenCL/simd.c"
__kernel void m00120_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)
{
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
+ /**
+ * 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 < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
- u32 wordr1[4];
- u32 wordr2[4];
- u32 wordr3[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 w3[4];
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
* prepend salt
*/
- const u32 pw_salt_len = pw_len + salt_len;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- 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] = w3[2];
- w3_t[3] = w3[3];
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ const u32x pw_salt_len = pw_len + salt_len;
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_M_SIMD (d, e, c, b);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* reverse
*/
- const u32 e_rev = rotl32 (search[1], 2u);
+ const u32 e_rev = rotl32_S (search[1], 2u);
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
- u32 wordr1[4];
- u32 wordr2[4];
- u32 wordr3[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 w3[4];
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
* prepend salt
*/
- const u32 pw_salt_len = pw_len + salt_len;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- 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] = w3[2];
- w3_t[3] = w3[3];
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ const u32x pw_salt_len = pw_len + salt_len;
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
-
- if (allx (e != e_rev)) continue;
-
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+
+ if (MATCHES_NONE_VS (e, e_rev)) continue;
+
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_S_SIMD (d, e, c, b);
}
}
*/
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];
-
u32 salt_buf2[4];
-
- salt_buf2[0] = 0;
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
u32 salt_buf3[4];
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* prepend salt
*/
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- w0_t[0] = swap32_S (w0[0]);
- w0_t[1] = swap32_S (w0[1]);
- w0_t[2] = swap32_S (w0[2]);
- w0_t[3] = swap32_S (w0[3]);
- w1_t[0] = swap32_S (w1[0]);
- w1_t[1] = swap32_S (w1[1]);
- w1_t[2] = swap32_S (w1[2]);
- w1_t[3] = swap32_S (w1[3]);
- w2_t[0] = swap32_S (w2[0]);
- w2_t[1] = swap32_S (w2[1]);
- w2_t[2] = swap32_S (w2[2]);
- w2_t[3] = swap32_S (w2[3]);
- w3_t[0] = swap32_S (w3[0]);
- w3_t[1] = swap32_S (w3[1]);
- w3_t[2] = swap32_S (w3[2]);
- w3_t[3] = swap32_S (w3[3]);
-
- switch_buffer_by_offset_le_S (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
- w2_t[0] |= salt_buf2[0];
- w2_t[1] |= salt_buf2[1];
- w2_t[2] |= salt_buf2[2];
- w2_t[3] |= salt_buf2[3];
- w3_t[0] |= salt_buf3[0];
- w3_t[1] |= salt_buf3[1];
- w3_t[2] |= salt_buf3[2];
- w3_t[3] |= salt_buf3[3];
-
- w0_t[0] = swap32_S (w0_t[0]);
- w0_t[1] = swap32_S (w0_t[1]);
- w0_t[2] = swap32_S (w0_t[2]);
- w0_t[3] = swap32_S (w0_t[3]);
- w1_t[0] = swap32_S (w1_t[0]);
- w1_t[1] = swap32_S (w1_t[1]);
- w1_t[2] = swap32_S (w1_t[2]);
- w1_t[3] = swap32_S (w1_t[3]);
- w2_t[0] = swap32_S (w2_t[0]);
- w2_t[1] = swap32_S (w2_t[1]);
- w2_t[2] = swap32_S (w2_t[2]);
- w2_t[3] = swap32_S (w2_t[3]);
- w3_t[0] = swap32_S (w3_t[0]);
- w3_t[1] = swap32_S (w3_t[1]);
- w3_t[2] = swap32_S (w3_t[2]);
- w3_t[3] = swap32_S (w3_t[3]);
+ const u32 w0l = w0[0];
+
+ switch_buffer_by_offset_be_S (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ u32x t0[4];
+ u32x t1[4];
+ u32x t2[4];
+ u32x t3[4];
+
+ t0[0] = w0[0];
+ t0[1] = w0[1];
+ t0[2] = w0[2];
+ t0[3] = w0[3];
+ t1[0] = w1[0];
+ t1[1] = w1[1];
+ t1[2] = w1[2];
+ t1[3] = w1[3];
+ t2[0] = w2[0];
+ t2[1] = w2[1];
+ t2[2] = w2[2];
+ t2[3] = w2[3];
+ t3[0] = w3[0];
+ t3[1] = w3[1];
+ t3[2] = w3[2];
+ t3[3] = w3[3];
/**
* loop
*/
- u32 w0l = w0[0];
-
for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
{
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
const u32x w0lr = w0l | w0r;
- u32x wx[16];
-
- wx[ 0] = w0_t[0];
- wx[ 1] = w0_t[1];
- wx[ 2] = w0_t[2];
- wx[ 3] = w0_t[3];
- wx[ 4] = w1_t[0];
- wx[ 5] = w1_t[1];
- wx[ 6] = w1_t[2];
- wx[ 7] = w1_t[3];
- wx[ 8] = w2_t[0];
- wx[ 9] = w2_t[1];
- wx[10] = w2_t[2];
- wx[11] = w2_t[3];
- wx[12] = w3_t[0];
- wx[13] = w3_t[1];
- wx[14] = w3_t[2];
- wx[15] = w3_t[3];
-
- overwrite_at_be (wx, w0lr, salt_len);
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- w0_t[0] = wx[ 0];
- w0_t[1] = wx[ 1];
- w0_t[2] = wx[ 2];
- w0_t[3] = wx[ 3];
- w1_t[0] = wx[ 4];
- w1_t[1] = wx[ 5];
- w1_t[2] = wx[ 6];
- w1_t[3] = wx[ 7];
- w2_t[0] = wx[ 8];
- w2_t[1] = wx[ 9];
- w2_t[2] = wx[10];
- w2_t[3] = wx[11];
- w3_t[0] = wx[12];
- w3_t[1] = wx[13];
- w3_t[2] = 0;
- w3_t[3] = pw_salt_len * 8;
+ overwrite_at_be_4x4 (t0, t1, t2, t3, w0lr, salt_len);
/**
* sha1
*/
+ u32x w0_t = t0[0];
+ u32x w1_t = t0[1];
+ u32x w2_t = t0[2];
+ u32x w3_t = t0[3];
+ u32x w4_t = t1[0];
+ u32x w5_t = t1[1];
+ u32x w6_t = t1[2];
+ u32x w7_t = t1[3];
+ u32x w8_t = t2[0];
+ u32x w9_t = t2[1];
+ u32x wa_t = t2[2];
+ u32x wb_t = t2[3];
+ u32x wc_t = t3[0];
+ u32x wd_t = t3[1];
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_M_SIMD (d, e, c, b);
}
*/
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];
-
u32 salt_buf2[4];
-
- salt_buf2[0] = 0;
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
u32 salt_buf3[4];
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* prepend salt
*/
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- w0_t[0] = swap32_S (w0[0]);
- w0_t[1] = swap32_S (w0[1]);
- w0_t[2] = swap32_S (w0[2]);
- w0_t[3] = swap32_S (w0[3]);
- w1_t[0] = swap32_S (w1[0]);
- w1_t[1] = swap32_S (w1[1]);
- w1_t[2] = swap32_S (w1[2]);
- w1_t[3] = swap32_S (w1[3]);
- w2_t[0] = swap32_S (w2[0]);
- w2_t[1] = swap32_S (w2[1]);
- w2_t[2] = swap32_S (w2[2]);
- w2_t[3] = swap32_S (w2[3]);
- w3_t[0] = swap32_S (w3[0]);
- w3_t[1] = swap32_S (w3[1]);
- w3_t[2] = swap32_S (w3[2]);
- w3_t[3] = swap32_S (w3[3]);
-
- switch_buffer_by_offset_le_S (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
- w2_t[0] |= salt_buf2[0];
- w2_t[1] |= salt_buf2[1];
- w2_t[2] |= salt_buf2[2];
- w2_t[3] |= salt_buf2[3];
- w3_t[0] |= salt_buf3[0];
- w3_t[1] |= salt_buf3[1];
- w3_t[2] |= salt_buf3[2];
- w3_t[3] |= salt_buf3[3];
-
- w0_t[0] = swap32_S (w0_t[0]);
- w0_t[1] = swap32_S (w0_t[1]);
- w0_t[2] = swap32_S (w0_t[2]);
- w0_t[3] = swap32_S (w0_t[3]);
- w1_t[0] = swap32_S (w1_t[0]);
- w1_t[1] = swap32_S (w1_t[1]);
- w1_t[2] = swap32_S (w1_t[2]);
- w1_t[3] = swap32_S (w1_t[3]);
- w2_t[0] = swap32_S (w2_t[0]);
- w2_t[1] = swap32_S (w2_t[1]);
- w2_t[2] = swap32_S (w2_t[2]);
- w2_t[3] = swap32_S (w2_t[3]);
- w3_t[0] = swap32_S (w3_t[0]);
- w3_t[1] = swap32_S (w3_t[1]);
- w3_t[2] = swap32_S (w3_t[2]);
- w3_t[3] = swap32_S (w3_t[3]);
+ const u32 w0l = w0[0];
+
+ switch_buffer_by_offset_be_S (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ u32x t0[4];
+ u32x t1[4];
+ u32x t2[4];
+ u32x t3[4];
+
+ t0[0] = w0[0];
+ t0[1] = w0[1];
+ t0[2] = w0[2];
+ t0[3] = w0[3];
+ t1[0] = w1[0];
+ t1[1] = w1[1];
+ t1[2] = w1[2];
+ t1[3] = w1[3];
+ t2[0] = w2[0];
+ t2[1] = w2[1];
+ t2[2] = w2[2];
+ t2[3] = w2[3];
+ t3[0] = w3[0];
+ t3[1] = w3[1];
+ t3[2] = w3[2];
+ t3[3] = w3[3];
/**
* loop
*/
- u32 w0l = w0[0];
-
for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
{
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
const u32x w0lr = w0l | w0r;
- u32x wx[16];
-
- wx[ 0] = w0_t[0];
- wx[ 1] = w0_t[1];
- wx[ 2] = w0_t[2];
- wx[ 3] = w0_t[3];
- wx[ 4] = w1_t[0];
- wx[ 5] = w1_t[1];
- wx[ 6] = w1_t[2];
- wx[ 7] = w1_t[3];
- wx[ 8] = w2_t[0];
- wx[ 9] = w2_t[1];
- wx[10] = w2_t[2];
- wx[11] = w2_t[3];
- wx[12] = w3_t[0];
- wx[13] = w3_t[1];
- wx[14] = w3_t[2];
- wx[15] = w3_t[3];
-
- overwrite_at_be (wx, w0lr, salt_len);
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- w0_t[0] = wx[ 0];
- w0_t[1] = wx[ 1];
- w0_t[2] = wx[ 2];
- w0_t[3] = wx[ 3];
- w1_t[0] = wx[ 4];
- w1_t[1] = wx[ 5];
- w1_t[2] = wx[ 6];
- w1_t[3] = wx[ 7];
- w2_t[0] = wx[ 8];
- w2_t[1] = wx[ 9];
- w2_t[2] = wx[10];
- w2_t[3] = wx[11];
- w3_t[0] = wx[12];
- w3_t[1] = wx[13];
- w3_t[2] = 0;
- w3_t[3] = pw_salt_len * 8;
+ overwrite_at_be_4x4 (t0, t1, t2, t3, w0lr, salt_len);
/**
* sha1
*/
+ u32x w0_t = t0[0];
+ u32x w1_t = t0[1];
+ u32x w2_t = t0[2];
+ u32x w3_t = t0[3];
+ u32x w4_t = t1[0];
+ u32x w5_t = t1[1];
+ u32x w6_t = t1[2];
+ u32x w7_t = t1[3];
+ u32x w8_t = t2[0];
+ u32x w9_t = t2[1];
+ u32x wa_t = t2[2];
+ u32x wb_t = t2[3];
+ u32x wc_t = t3[0];
+ u32x wd_t = t3[1];
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
if (MATCHES_NONE_VS (e, e_rev)) continue;
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_S_SIMD (d, e, c, b);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32 out_len2 = out_len * 2;
+
/**
* append salt
*/
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32x s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32x s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32x s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, (out_len * 2));
-
- const u32 out_salt_len = (out_len * 2) + salt_len;
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- w0_t[0] |= s0[0];
- w0_t[1] |= s0[1];
- w0_t[2] |= s0[2];
- w0_t[3] |= s0[3];
- w1_t[0] |= s1[0];
- w1_t[1] |= s1[1];
- w1_t[2] |= s1[2];
- w1_t[3] |= s1[3];
- w2_t[0] |= s2[0];
- w2_t[1] |= s2[1];
- w2_t[2] |= s2[2];
- w2_t[3] |= s2[3];
- w3_t[0] |= s3[0];
- w3_t[1] |= s3[1];
- w3_t[2] |= s3[2];
- w3_t[3] |= s3[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le (s0, s1, s2, s3, out_len2);
+
+ const u32 pw_salt_len = out_len2 + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_M_SIMD (d, e, c, b);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32 out_len2 = out_len * 2;
+
/**
* append salt
*/
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32x s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32x s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32x s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, (out_len * 2));
-
- const u32 out_salt_len = (out_len * 2) + salt_len;
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- w0_t[0] |= s0[0];
- w0_t[1] |= s0[1];
- w0_t[2] |= s0[2];
- w0_t[3] |= s0[3];
- w1_t[0] |= s1[0];
- w1_t[1] |= s1[1];
- w1_t[2] |= s1[2];
- w1_t[3] |= s1[3];
- w2_t[0] |= s2[0];
- w2_t[1] |= s2[1];
- w2_t[2] |= s2[2];
- w2_t[3] |= s2[3];
- w3_t[0] |= s3[0];
- w3_t[1] |= s3[1];
- w3_t[2] |= s3[2];
- w3_t[3] |= s3[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le (s0, s1, s2, s3, out_len2);
+
+ const u32 pw_salt_len = out_len2 + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
if (MATCHES_NONE_VS (e, e_rev)) continue;
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_S_SIMD (d, e, c, b);
}
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
+#include "OpenCL/simd.c"
// no unicode yet
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr0[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
+ /**
+ * concat password candidate
+ */
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ 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);
- u32 wordr2[4];
+ 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);
+ }
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
- u32 wordr3[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];
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
- if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
- {
- switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
- }
+ const u32x pw_len2 = pw_len * 2;
/**
* append salt
*/
- u32 s0[4];
+ u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32 s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32 s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32 s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, (pw_len * 2));
-
- const u32 pw_salt_len = (pw_len * 2) + salt_len;
-
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 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] = 0;
- w2[1] = 0;
- w2[2] = 0;
- w2[3] = 0;
- w3[0] = 0;
- w3[1] = 0;
- w3[2] = 0;
- w3[3] = 0;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- w0_t[0] |= s0[0];
- w0_t[1] |= s0[1];
- w0_t[2] |= s0[2];
- w0_t[3] |= s0[3];
- w1_t[0] |= s1[0];
- w1_t[1] |= s1[1];
- w1_t[2] |= s1[2];
- w1_t[3] |= s1[3];
- w2_t[0] |= s2[0];
- w2_t[1] |= s2[1];
- w2_t[2] |= s2[2];
- w2_t[3] |= s2[3];
- w3_t[0] |= s3[0];
- w3_t[1] |= s3[1];
- w3_t[2] |= s3[2];
- w3_t[3] |= s3[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len2);
+
+ const u32x pw_salt_len = pw_len2 + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_M_SIMD (d, e, c, b);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* reverse
*/
- const u32 e_rev = rotl32 (search[1], 2u);
+ const u32 e_rev = rotl32_S (search[1], 2u);
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- const u32 pw_len = pw_l_len + pw_r_len;
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
+ /**
+ * concat password candidate
+ */
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ 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);
- u32 wordr2[4];
+ 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);
+ }
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
- u32 wordr3[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];
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
- if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
- {
- switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
- }
+ const u32x pw_len2 = pw_len * 2;
/**
* append salt
*/
- u32 s0[4];
+ u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3];
-
- u32 s1[4];
-
s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3];
-
- u32 s2[4];
-
- s2[0] = 0;
- s2[1] = 0;
- s2[2] = 0;
- s2[3] = 0;
-
- u32 s3[4];
-
- s3[0] = 0;
- s3[1] = 0;
- s3[2] = 0;
- s3[3] = 0;
-
- switch_buffer_by_offset_le (s0, s1, s2, s3, (pw_len * 2));
-
- const u32 pw_salt_len = (pw_len * 2) + salt_len;
-
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 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] = 0;
- w2[1] = 0;
- w2[2] = 0;
- w2[3] = 0;
- w3[0] = 0;
- w3[1] = 0;
- w3[2] = 0;
- w3[3] = 0;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- w0_t[0] |= s0[0];
- w0_t[1] |= s0[1];
- w0_t[2] |= s0[2];
- w0_t[3] |= s0[3];
- w1_t[0] |= s1[0];
- w1_t[1] |= s1[1];
- w1_t[2] |= s1[2];
- w1_t[3] |= s1[3];
- w2_t[0] |= s2[0];
- w2_t[1] |= s2[1];
- w2_t[2] |= s2[2];
- w2_t[3] |= s2[3];
- w3_t[0] |= s3[0];
- w3_t[1] |= s3[1];
- w3_t[2] |= s3[2];
- w3_t[3] |= s3[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ s2[0] = salt_buf2[0];
+ s2[1] = salt_buf2[1];
+ s2[2] = salt_buf2[2];
+ s2[3] = salt_buf2[3];
+ s3[0] = salt_buf3[0];
+ s3[1] = salt_buf3[1];
+ s3[2] = salt_buf3[2];
+ s3[3] = salt_buf3[3];
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len2);
+
+ const u32x pw_salt_len = pw_len2 + salt_len;
+
+ w0[0] |= s0[0];
+ w0[1] |= s0[1];
+ w0[2] |= s0[2];
+ w0[3] |= s0[3];
+ w1[0] |= s1[0];
+ w1[1] |= s1[1];
+ w1[2] |= s1[2];
+ w1[3] |= s1[3];
+ w2[0] |= s2[0];
+ w2[1] |= s2[1];
+ w2[2] |= s2[2];
+ w2[3] |= s2[3];
+ w3[0] |= s3[0];
+ w3[1] |= s3[1];
+ w3[2] |= s3[2];
+ w3[3] |= s3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
-
- if (allx (e != e_rev)) continue;
-
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+
+ if (MATCHES_NONE_VS (e, e_rev)) continue;
+
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_S_SIMD (d, e, c, b);
}
}
*/
u32 salt_buf0[4];
+ u32 salt_buf1[4];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
-
- u32 salt_buf2[4];
-
salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
- u32 salt_buf3[4];
-
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
switch_buffer_by_offset_le_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, pw_len);
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32 out_len2 = out_len * 2;
+
/**
* prepend salt
*/
- const u32 out_salt_len = (out_len * 2) + salt_len;
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ const u32 out_salt_len = out_len2 + salt_len;
+
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ append_0x80_4x4 (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_M_SIMD (d, e, c, b);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32 out_len2 = out_len * 2;
+
/**
* prepend salt
*/
- const u32 out_salt_len = (out_len * 2) + salt_len;
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, out_salt_len);
-
- w3_t[3] = out_salt_len * 8;
+ const u32 out_salt_len = out_len2 + salt_len;
+
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ append_0x80_4x4 (w0, w1, w2, w3, out_salt_len);
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = out_salt_len * 8;
u32x a = SHA1M_A;
u32x b = SHA1M_B;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
if (MATCHES_NONE_VS (e, e_rev)) continue;
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
- COMPARE_M_SIMD (d, e, c, b);
+ COMPARE_S_SIMD (d, e, c, b);
}
}
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
+#include "OpenCL/simd.c"
// no unicode yet
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- /**
- * prepend salt
- */
-
- const u32 pw_salt_len = (pw_len * 2) + salt_len;
-
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 w3[4];
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
- w2[0] = 0;
- w2[1] = 0;
- w2[2] = 0;
- w2[3] = 0;
- w3[0] = 0;
- w3[1] = 0;
- w3[2] = 0;
- w3[3] = 0;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ 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];
+
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32x pw_len2 = pw_len * 2;
+
+ /**
+ * prepend salt
+ */
+
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ const u32x pw_salt_len = pw_len2 + salt_len;
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_M_SIMD (d, e, c, b);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[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];
+ 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];
+ salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
+ salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
+ salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
+ salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
+ salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
+ salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
+ salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
+ salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* reverse
*/
- const u32 e_rev = rotl32 (search[1], 2u);
+ const u32 e_rev = rotl32_S (search[1], 2u);
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
-
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr2[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- /**
- * prepend salt
- */
-
- const u32 pw_salt_len = (pw_len * 2) + salt_len;
-
- u32 w0[4];
- u32 w1[4];
- u32 w2[4];
- u32 w3[4];
+ u32x w0[4];
+ u32x w1[4];
+ u32x w2[4];
+ u32x w3[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
- w2[0] = 0;
- w2[1] = 0;
- w2[2] = 0;
- w2[3] = 0;
- w3[0] = 0;
- w3[1] = 0;
- w3[2] = 0;
- w3[3] = 0;
-
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- make_unicode (w0, w0_t, w1_t);
- make_unicode (w1, w2_t, w3_t);
-
- switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
-
- append_0x80_4x4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
-
- w3_t[3] = pw_salt_len * 8;
+ 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];
+
+ make_unicode (w1, w2, w3);
+ make_unicode (w0, w0, w1);
+
+ const u32x pw_len2 = pw_len * 2;
+
+ /**
+ * prepend salt
+ */
+
+ switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
+
+ const u32x pw_salt_len = pw_len2 + salt_len;
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
/**
* sha1
*/
- w0_t[0] = swap32 (w0_t[0]);
- w0_t[1] = swap32 (w0_t[1]);
- w0_t[2] = swap32 (w0_t[2]);
- w0_t[3] = swap32 (w0_t[3]);
- w1_t[0] = swap32 (w1_t[0]);
- w1_t[1] = swap32 (w1_t[1]);
- w1_t[2] = swap32 (w1_t[2]);
- w1_t[3] = swap32 (w1_t[3]);
- w2_t[0] = swap32 (w2_t[0]);
- w2_t[1] = swap32 (w2_t[1]);
- w2_t[2] = swap32 (w2_t[2]);
- w2_t[3] = swap32 (w2_t[3]);
- w3_t[0] = swap32 (w3_t[0]);
- w3_t[1] = swap32 (w3_t[1]);
- //w3_t[2] = swap32 (w3_t[2]);
- //w3_t[3] = swap32 (w3_t[3]);
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
-
- if (allx (e != e_rev)) continue;
-
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
-
-
- const u32 r0 = d;
- const u32 r1 = e;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+
+ if (MATCHES_NONE_VS (e, e_rev)) continue;
+
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
+
+ COMPARE_S_SIMD (d, e, c, b);
}
}
*/
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];
-
u32 salt_buf2[4];
-
- salt_buf2[0] = 0;
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
u32 salt_buf3[4];
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* prepend salt
*/
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- w0_t[0] = swap32_S (w0[0]);
- w0_t[1] = swap32_S (w0[1]);
- w0_t[2] = swap32_S (w0[2]);
- w0_t[3] = swap32_S (w0[3]);
- w1_t[0] = swap32_S (w1[0]);
- w1_t[1] = swap32_S (w1[1]);
- w1_t[2] = swap32_S (w1[2]);
- w1_t[3] = swap32_S (w1[3]);
- w2_t[0] = swap32_S (w2[0]);
- w2_t[1] = swap32_S (w2[1]);
- w2_t[2] = swap32_S (w2[2]);
- w2_t[3] = swap32_S (w2[3]);
- w3_t[0] = swap32_S (w3[0]);
- w3_t[1] = swap32_S (w3[1]);
- w3_t[2] = swap32_S (w3[2]);
- w3_t[3] = swap32_S (w3[3]);
-
- switch_buffer_by_offset_le_S (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
- w2_t[0] |= salt_buf2[0];
- w2_t[1] |= salt_buf2[1];
- w2_t[2] |= salt_buf2[2];
- w2_t[3] |= salt_buf2[3];
- w3_t[0] |= salt_buf3[0];
- w3_t[1] |= salt_buf3[1];
- w3_t[2] |= salt_buf3[2];
- w3_t[3] |= salt_buf3[3];
-
- w0_t[0] = swap32_S (w0_t[0]);
- w0_t[1] = swap32_S (w0_t[1]);
- w0_t[2] = swap32_S (w0_t[2]);
- w0_t[3] = swap32_S (w0_t[3]);
- w1_t[0] = swap32_S (w1_t[0]);
- w1_t[1] = swap32_S (w1_t[1]);
- w1_t[2] = swap32_S (w1_t[2]);
- w1_t[3] = swap32_S (w1_t[3]);
- w2_t[0] = swap32_S (w2_t[0]);
- w2_t[1] = swap32_S (w2_t[1]);
- w2_t[2] = swap32_S (w2_t[2]);
- w2_t[3] = swap32_S (w2_t[3]);
- w3_t[0] = swap32_S (w3_t[0]);
- w3_t[1] = swap32_S (w3_t[1]);
- w3_t[2] = swap32_S (w3_t[2]);
- w3_t[3] = swap32_S (w3_t[3]);
+ const u32 w0l = w0[0];
+
+ switch_buffer_by_offset_be_S (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ u32x t0[4];
+ u32x t1[4];
+ u32x t2[4];
+ u32x t3[4];
+
+ t0[0] = w0[0];
+ t0[1] = w0[1];
+ t0[2] = w0[2];
+ t0[3] = w0[3];
+ t1[0] = w1[0];
+ t1[1] = w1[1];
+ t1[2] = w1[2];
+ t1[3] = w1[3];
+ t2[0] = w2[0];
+ t2[1] = w2[1];
+ t2[2] = w2[2];
+ t2[3] = w2[3];
+ t3[0] = w3[0];
+ t3[1] = w3[1];
+ t3[2] = w3[2];
+ t3[3] = w3[3];
/**
* loop
*/
- u32 w0l = w0[0];
-
for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
{
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
const u32x w0lr = w0l | w0r;
- u32x wx[16];
-
- wx[ 0] = w0_t[0];
- wx[ 1] = w0_t[1];
- wx[ 2] = w0_t[2];
- wx[ 3] = w0_t[3];
- wx[ 4] = w1_t[0];
- wx[ 5] = w1_t[1];
- wx[ 6] = w1_t[2];
- wx[ 7] = w1_t[3];
- wx[ 8] = w2_t[0];
- wx[ 9] = w2_t[1];
- wx[10] = w2_t[2];
- wx[11] = w2_t[3];
- wx[12] = w3_t[0];
- wx[13] = w3_t[1];
- wx[14] = w3_t[2];
- wx[15] = w3_t[3];
-
- overwrite_at_be (wx, w0lr, salt_len);
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- w0_t[0] = wx[ 0];
- w0_t[1] = wx[ 1];
- w0_t[2] = wx[ 2];
- w0_t[3] = wx[ 3];
- w1_t[0] = wx[ 4];
- w1_t[1] = wx[ 5];
- w1_t[2] = wx[ 6];
- w1_t[3] = wx[ 7];
- w2_t[0] = wx[ 8];
- w2_t[1] = wx[ 9];
- w2_t[2] = wx[10];
- w2_t[3] = wx[11];
- w3_t[0] = wx[12];
- w3_t[1] = wx[13];
- w3_t[2] = 0;
- w3_t[3] = pw_salt_len * 8;
+ overwrite_at_be_4x4 (t0, t1, t2, t3, w0lr, salt_len);
/**
* sha1
*/
+ u32x w0_t = t0[0];
+ u32x w1_t = t0[1];
+ u32x w2_t = t0[2];
+ u32x w3_t = t0[3];
+ u32x w4_t = t1[0];
+ u32x w5_t = t1[1];
+ u32x w6_t = t1[2];
+ u32x w7_t = t1[3];
+ u32x w8_t = t2[0];
+ u32x w9_t = t2[1];
+ u32x wa_t = t2[2];
+ u32x wb_t = t2[3];
+ u32x wc_t = t3[0];
+ u32x wd_t = t3[1];
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_M_SIMD (d, e, c, b);
}
*/
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];
-
u32 salt_buf2[4];
-
- salt_buf2[0] = 0;
- salt_buf2[1] = 0;
- salt_buf2[2] = 0;
- salt_buf2[3] = 0;
-
u32 salt_buf3[4];
- salt_buf3[0] = 0;
- salt_buf3[1] = 0;
- salt_buf3[2] = 0;
- salt_buf3[3] = 0;
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* prepend salt
*/
- u32 w0_t[4];
- u32 w1_t[4];
- u32 w2_t[4];
- u32 w3_t[4];
-
- w0_t[0] = swap32_S (w0[0]);
- w0_t[1] = swap32_S (w0[1]);
- w0_t[2] = swap32_S (w0[2]);
- w0_t[3] = swap32_S (w0[3]);
- w1_t[0] = swap32_S (w1[0]);
- w1_t[1] = swap32_S (w1[1]);
- w1_t[2] = swap32_S (w1[2]);
- w1_t[3] = swap32_S (w1[3]);
- w2_t[0] = swap32_S (w2[0]);
- w2_t[1] = swap32_S (w2[1]);
- w2_t[2] = swap32_S (w2[2]);
- w2_t[3] = swap32_S (w2[3]);
- w3_t[0] = swap32_S (w3[0]);
- w3_t[1] = swap32_S (w3[1]);
- w3_t[2] = swap32_S (w3[2]);
- w3_t[3] = swap32_S (w3[3]);
-
- switch_buffer_by_offset_le_S (w0_t, w1_t, w2_t, w3_t, salt_len);
-
- w0_t[0] |= salt_buf0[0];
- w0_t[1] |= salt_buf0[1];
- w0_t[2] |= salt_buf0[2];
- w0_t[3] |= salt_buf0[3];
- w1_t[0] |= salt_buf1[0];
- w1_t[1] |= salt_buf1[1];
- w1_t[2] |= salt_buf1[2];
- w1_t[3] |= salt_buf1[3];
- w2_t[0] |= salt_buf2[0];
- w2_t[1] |= salt_buf2[1];
- w2_t[2] |= salt_buf2[2];
- w2_t[3] |= salt_buf2[3];
- w3_t[0] |= salt_buf3[0];
- w3_t[1] |= salt_buf3[1];
- w3_t[2] |= salt_buf3[2];
- w3_t[3] |= salt_buf3[3];
-
- w0_t[0] = swap32_S (w0_t[0]);
- w0_t[1] = swap32_S (w0_t[1]);
- w0_t[2] = swap32_S (w0_t[2]);
- w0_t[3] = swap32_S (w0_t[3]);
- w1_t[0] = swap32_S (w1_t[0]);
- w1_t[1] = swap32_S (w1_t[1]);
- w1_t[2] = swap32_S (w1_t[2]);
- w1_t[3] = swap32_S (w1_t[3]);
- w2_t[0] = swap32_S (w2_t[0]);
- w2_t[1] = swap32_S (w2_t[1]);
- w2_t[2] = swap32_S (w2_t[2]);
- w2_t[3] = swap32_S (w2_t[3]);
- w3_t[0] = swap32_S (w3_t[0]);
- w3_t[1] = swap32_S (w3_t[1]);
- w3_t[2] = swap32_S (w3_t[2]);
- w3_t[3] = swap32_S (w3_t[3]);
+ const u32 w0l = w0[0];
+
+ switch_buffer_by_offset_be_S (w0, w1, w2, w3, salt_len);
+
+ w0[0] |= salt_buf0[0];
+ w0[1] |= salt_buf0[1];
+ w0[2] |= salt_buf0[2];
+ w0[3] |= salt_buf0[3];
+ w1[0] |= salt_buf1[0];
+ w1[1] |= salt_buf1[1];
+ w1[2] |= salt_buf1[2];
+ w1[3] |= salt_buf1[3];
+ w2[0] |= salt_buf2[0];
+ w2[1] |= salt_buf2[1];
+ w2[2] |= salt_buf2[2];
+ w2[3] |= salt_buf2[3];
+ w3[0] |= salt_buf3[0];
+ w3[1] |= salt_buf3[1];
+ w3[2] |= salt_buf3[2];
+ w3[3] |= salt_buf3[3];
+
+ u32x t0[4];
+ u32x t1[4];
+ u32x t2[4];
+ u32x t3[4];
+
+ t0[0] = w0[0];
+ t0[1] = w0[1];
+ t0[2] = w0[2];
+ t0[3] = w0[3];
+ t1[0] = w1[0];
+ t1[1] = w1[1];
+ t1[2] = w1[2];
+ t1[3] = w1[3];
+ t2[0] = w2[0];
+ t2[1] = w2[1];
+ t2[2] = w2[2];
+ t2[3] = w2[3];
+ t3[0] = w3[0];
+ t3[1] = w3[1];
+ t3[2] = w3[2];
+ t3[3] = w3[3];
/**
* loop
*/
- u32 w0l = w0[0];
-
for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos += VECT_SIZE)
{
const u32x w0r = ix_create_bft (bfs_buf, il_pos);
const u32x w0lr = w0l | w0r;
- u32x wx[16];
-
- wx[ 0] = w0_t[0];
- wx[ 1] = w0_t[1];
- wx[ 2] = w0_t[2];
- wx[ 3] = w0_t[3];
- wx[ 4] = w1_t[0];
- wx[ 5] = w1_t[1];
- wx[ 6] = w1_t[2];
- wx[ 7] = w1_t[3];
- wx[ 8] = w2_t[0];
- wx[ 9] = w2_t[1];
- wx[10] = w2_t[2];
- wx[11] = w2_t[3];
- wx[12] = w3_t[0];
- wx[13] = w3_t[1];
- wx[14] = w3_t[2];
- wx[15] = w3_t[3];
-
- overwrite_at_be (wx, w0lr, salt_len);
-
- u32x w0_t[4];
- u32x w1_t[4];
- u32x w2_t[4];
- u32x w3_t[4];
-
- w0_t[0] = wx[ 0];
- w0_t[1] = wx[ 1];
- w0_t[2] = wx[ 2];
- w0_t[3] = wx[ 3];
- w1_t[0] = wx[ 4];
- w1_t[1] = wx[ 5];
- w1_t[2] = wx[ 6];
- w1_t[3] = wx[ 7];
- w2_t[0] = wx[ 8];
- w2_t[1] = wx[ 9];
- w2_t[2] = wx[10];
- w2_t[3] = wx[11];
- w3_t[0] = wx[12];
- w3_t[1] = wx[13];
- w3_t[2] = 0;
- w3_t[3] = pw_salt_len * 8;
+ overwrite_at_be_4x4 (t0, t1, t2, t3, w0lr, salt_len);
/**
* sha1
*/
+ u32x w0_t = t0[0];
+ u32x w1_t = t0[1];
+ u32x w2_t = t0[2];
+ u32x w3_t = t0[3];
+ u32x w4_t = t1[0];
+ u32x w5_t = t1[1];
+ u32x w6_t = t1[2];
+ u32x w7_t = t1[3];
+ u32x w8_t = t2[0];
+ u32x w9_t = t2[1];
+ u32x wa_t = t2[2];
+ u32x wb_t = t2[3];
+ u32x wc_t = t3[0];
+ u32x wd_t = t3[1];
+ u32x we_t = 0;
+ u32x wf_t = pw_salt_len * 8;
+
u32x a = SHA1M_A;
u32x b = SHA1M_B;
u32x c = SHA1M_C;
#undef K
#define K SHA1C00
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t[0]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[1]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[2]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[3]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w1_t[0]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w1_t[1]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t[2]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t[3]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t[0]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w2_t[1]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w2_t[2]);
- SHA1_STEP (SHA1_F0o, e, a, b, c, d, w2_t[3]);
- SHA1_STEP (SHA1_F0o, d, e, a, b, c, w3_t[0]);
- SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t[1]);
- SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t[2]);
- SHA1_STEP (SHA1_F0o, a, b, c, d, e, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0_t[3]);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w1_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w2_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w3_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w4_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, w5_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, w6_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, w7_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, w8_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, w9_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wa_t);
+ SHA1_STEP (SHA1_F0o, e, a, b, c, d, wb_t);
+ SHA1_STEP (SHA1_F0o, d, e, a, b, c, wc_t);
+ SHA1_STEP (SHA1_F0o, c, d, e, a, b, wd_t);
+ SHA1_STEP (SHA1_F0o, b, c, d, e, a, we_t);
+ SHA1_STEP (SHA1_F0o, a, b, c, d, e, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[3]);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2_t[3]);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wb_t);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3_t[0]);
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[3]);
- w0_t[0] = rotl32 ((w3_t[1] ^ w2_t[0] ^ w0_t[2] ^ w0_t[0]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t[0]);
- w0_t[1] = rotl32 ((w3_t[2] ^ w2_t[1] ^ w0_t[3] ^ w0_t[1]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w0_t[1]);
- w0_t[2] = rotl32 ((w3_t[3] ^ w2_t[2] ^ w1_t[0] ^ w0_t[2]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w0_t[2]);
- w0_t[3] = rotl32 ((w0_t[0] ^ w2_t[3] ^ w1_t[1] ^ w0_t[3]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w0_t[3]);
- w1_t[0] = rotl32 ((w0_t[1] ^ w3_t[0] ^ w1_t[2] ^ w1_t[0]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1_t[0]);
- w1_t[1] = rotl32 ((w0_t[2] ^ w3_t[1] ^ w1_t[3] ^ w1_t[1]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1_t[1]);
- w1_t[2] = rotl32 ((w0_t[3] ^ w3_t[2] ^ w2_t[0] ^ w1_t[2]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t[2]);
- w1_t[3] = rotl32 ((w1_t[0] ^ w3_t[3] ^ w2_t[1] ^ w1_t[3]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1_t[3]);
- w2_t[0] = rotl32 ((w1_t[1] ^ w0_t[0] ^ w2_t[2] ^ w2_t[0]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w2_t[0]);
- w2_t[1] = rotl32 ((w1_t[2] ^ w0_t[1] ^ w2_t[3] ^ w2_t[1]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w2_t[1]);
- w2_t[2] = rotl32 ((w1_t[3] ^ w0_t[2] ^ w3_t[0] ^ w2_t[2]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w2_t[2]);
- w2_t[3] = rotl32 ((w2_t[0] ^ w0_t[3] ^ w3_t[1] ^ w2_t[3]), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w2_t[3]);
- w3_t[0] = rotl32 ((w2_t[1] ^ w1_t[0] ^ w3_t[2] ^ w3_t[0]), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3_t[0]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, wf_t);
+ w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w0_t);
+ w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1_t);
+ w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w2_t);
+ w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t);
+ w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4_t);
+ w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w5_t);
+ w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w6_t);
+ w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w7_t);
+ w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w8_t);
+ w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w9_t);
+ wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wa_t);
+ wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, wb_t);
if (MATCHES_NONE_VS (e, e_rev)) continue;
- w3_t[1] = rotl32 ((w2_t[2] ^ w1_t[1] ^ w3_t[3] ^ w3_t[1]), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3_t[1]);
- w3_t[2] = rotl32 ((w2_t[3] ^ w1_t[2] ^ w0_t[0] ^ w3_t[2]), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3_t[2]);
- w3_t[3] = rotl32 ((w3_t[0] ^ w1_t[3] ^ w0_t[1] ^ w3_t[3]), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w3_t[3]);
+ wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, wc_t);
+ wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, wd_t);
+ we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, we_t);
+ wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, wf_t);
COMPARE_S_SIMD (d, e, c, b);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+
/**
* pads
*/
- u32x w0_t[4];
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
-
- u32x w1_t[4];
-
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[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[5];
u32x opad[5];
- hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = (64 + salt_len) * 8;
+ hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
+
+ w0[0] = salt_buf0[0];
+ w0[1] = salt_buf0[1];
+ w0[2] = salt_buf0[2];
+ w0[3] = salt_buf0[3];
+ w1[0] = salt_buf1[0];
+ w1[1] = salt_buf1[1];
+ w1[2] = salt_buf1[2];
+ w1[3] = salt_buf1[3];
+ w2[0] = salt_buf2[0];
+ w2[1] = salt_buf2[1];
+ w2[2] = salt_buf2[2];
+ w2[3] = salt_buf2[3];
+ w3[0] = salt_buf3[0];
+ w3[1] = salt_buf3[1];
+ w3[2] = 0;
+ w3[3] = (64 + salt_len) * 8;
u32x digest[5];
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+
/**
* pads
*/
- u32x w0_t[4];
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
-
- u32x w1_t[4];
-
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[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[5];
u32x opad[5];
- hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = (64 + salt_len) * 8;
+ hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
+
+ w0[0] = salt_buf0[0];
+ w0[1] = salt_buf0[1];
+ w0[2] = salt_buf0[2];
+ w0[3] = salt_buf0[3];
+ w1[0] = salt_buf1[0];
+ w1[1] = salt_buf1[1];
+ w1[2] = salt_buf1[2];
+ w1[3] = salt_buf1[3];
+ w2[0] = salt_buf2[0];
+ w2[1] = salt_buf2[1];
+ w2[2] = salt_buf2[2];
+ w2[3] = salt_buf2[3];
+ w3[0] = salt_buf3[0];
+ w3[1] = salt_buf3[1];
+ w3[2] = 0;
+ w3[3] = (64 + salt_len) * 8;
u32x digest[5];
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
+#include "OpenCL/simd.c"
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
-
-static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
+static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
- u32 A = digest[0];
- u32 B = digest[1];
- u32 C = digest[2];
- u32 D = digest[3];
- u32 E = digest[4];
-
- u32 w0_t = w0[0];
- u32 w1_t = w0[1];
- u32 w2_t = w0[2];
- u32 w3_t = w0[3];
- u32 w4_t = w1[0];
- u32 w5_t = w1[1];
- u32 w6_t = w1[2];
- u32 w7_t = w1[3];
- u32 w8_t = w2[0];
- u32 w9_t = w2[1];
- u32 wa_t = w2[2];
- u32 wb_t = w2[3];
- u32 wc_t = w3[0];
- u32 wd_t = w3[1];
- u32 we_t = w3[2];
- u32 wf_t = w3[3];
+ u32x A = digest[0];
+ u32x B = digest[1];
+ u32x C = digest[2];
+ u32x D = digest[3];
+ u32x E = digest[4];
+
+ 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];
#undef K
#define K SHA1C00
digest[4] += E;
}
-static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
+static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
w0[0] = w0[0] ^ 0x36363636;
w0[1] = w0[1] ^ 0x36363636;
sha1_transform (w0, w1, w2, w3, opad);
}
-static void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5])
+static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
{
digest[0] = ipad[0];
digest[1] = ipad[1];
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr1[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
-
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+ w2[0] = swap32 (w2[0]);
+ w2[1] = swap32 (w2[1]);
+ w2[2] = swap32 (w2[2]);
+ w2[3] = swap32 (w2[3]);
+ w3[0] = swap32 (w3[0]);
+ w3[1] = swap32 (w3[1]);
+ w3[2] = swap32 (w3[2]);
+ w3[3] = swap32 (w3[3]);
+
/**
* pads
*/
- u32 w0_t[4];
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
-
- u32 w1_t[4];
-
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[3]);
-
- u32 w2_t[4];
-
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
-
- u32 w3_t[4];
-
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = 0;
-
- u32 ipad[5];
- u32 opad[5];
-
- hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = (64 + salt_len) * 8;
-
- u32 digest[5];
-
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[4];
- const u32 r2 = digest[2];
- const u32 r3 = digest[1];
-
- #include COMPARE_M
+ u32x ipad[5];
+ u32x opad[5];
+
+ hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
+
+ w0[0] = salt_buf0[0];
+ w0[1] = salt_buf0[1];
+ w0[2] = salt_buf0[2];
+ w0[3] = salt_buf0[3];
+ w1[0] = salt_buf1[0];
+ w1[1] = salt_buf1[1];
+ w1[2] = salt_buf1[2];
+ w1[3] = salt_buf1[3];
+ w2[0] = salt_buf2[0];
+ w2[1] = salt_buf2[1];
+ w2[2] = salt_buf2[2];
+ w2[3] = salt_buf2[3];
+ w3[0] = salt_buf3[0];
+ w3[1] = salt_buf3[1];
+ w3[2] = 0;
+ w3[3] = (64 + salt_len) * 8;
+
+ u32x digest[5];
+
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+ COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+ w2[0] = swap32 (w2[0]);
+ w2[1] = swap32 (w2[1]);
+ w2[2] = swap32 (w2[2]);
+ w2[3] = swap32 (w2[3]);
+ w3[0] = swap32 (w3[0]);
+ w3[1] = swap32 (w3[1]);
+ w3[2] = swap32 (w3[2]);
+ w3[3] = swap32 (w3[3]);
+
/**
* pads
*/
- u32 w0_t[4];
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
-
- u32 w1_t[4];
-
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[3]);
-
- u32 w2_t[4];
-
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
-
- u32 w3_t[4];
-
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = 0;
-
- u32 ipad[5];
- u32 opad[5];
-
- hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = (64 + salt_len) * 8;
-
- u32 digest[5];
-
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[4];
- const u32 r2 = digest[2];
- const u32 r3 = digest[1];
-
- #include COMPARE_S
+ u32x ipad[5];
+ u32x opad[5];
+
+ hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
+
+ w0[0] = salt_buf0[0];
+ w0[1] = salt_buf0[1];
+ w0[2] = salt_buf0[2];
+ w0[3] = salt_buf0[3];
+ w1[0] = salt_buf1[0];
+ w1[1] = salt_buf1[1];
+ w1[2] = salt_buf1[2];
+ w1[3] = salt_buf1[3];
+ w2[0] = salt_buf2[0];
+ w2[1] = salt_buf2[1];
+ w2[2] = salt_buf2[2];
+ w2[3] = salt_buf2[3];
+ w3[0] = salt_buf3[0];
+ w3[1] = salt_buf3[1];
+ w3[2] = 0;
+ w3[3] = (64 + salt_len) * 8;
+
+ u32x digest[5];
+
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+ COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
}
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
*/
u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
w0_t[0] = w0lr;
w0_t[1] = w0[1];
w0_t[2] = w0[2];
w0_t[3] = w0[3];
-
- u32x w1_t[4];
-
w1_t[0] = w1[0];
w1_t[1] = w1[1];
w1_t[2] = w1[2];
w1_t[3] = w1[3];
-
- u32x w2_t[4];
-
w2_t[0] = w2[0];
w2_t[1] = w2[1];
w2_t[2] = w2[2];
w2_t[3] = w2[3];
-
- u32x w3_t[4];
-
w3_t[0] = w3[0];
w3_t[1] = w3[1];
- w3_t[2] = 0;
- w3_t[3] = 0;
+ w3_t[2] = w3[2];
+ w3_t[3] = w3[3];
u32x ipad[5];
u32x opad[5];
hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
w3_t[2] = 0;
w3_t[3] = (64 + salt_len) * 8;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
*/
u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
w0_t[0] = w0lr;
w0_t[1] = w0[1];
w0_t[2] = w0[2];
w0_t[3] = w0[3];
-
- u32x w1_t[4];
-
w1_t[0] = w1[0];
w1_t[1] = w1[1];
w1_t[2] = w1[2];
w1_t[3] = w1[3];
-
- u32x w2_t[4];
-
w2_t[0] = w2[0];
w2_t[1] = w2[1];
w2_t[2] = w2[2];
w2_t[3] = w2[3];
-
- u32x w3_t[4];
-
w3_t[0] = w3[0];
w3_t[1] = w3[1];
- w3_t[2] = 0;
- w3_t[3] = 0;
+ w3_t[2] = w3[2];
+ w3_t[3] = w3[3];
u32x ipad[5];
u32x opad[5];
hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
- w3_t[0] = 0;
- w3_t[1] = 0;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
w3_t[2] = 0;
w3_t[3] = (64 + salt_len) * 8;
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
u32x w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
u32x w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (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;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
u32x ipad[5];
u32x opad[5];
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
u32x w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
u32x w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (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;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
u32x ipad[5];
u32x opad[5];
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
+#include "OpenCL/simd.c"
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
-
-static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
+static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
- u32 A = digest[0];
- u32 B = digest[1];
- u32 C = digest[2];
- u32 D = digest[3];
- u32 E = digest[4];
-
- u32 w0_t = w0[0];
- u32 w1_t = w0[1];
- u32 w2_t = w0[2];
- u32 w3_t = w0[3];
- u32 w4_t = w1[0];
- u32 w5_t = w1[1];
- u32 w6_t = w1[2];
- u32 w7_t = w1[3];
- u32 w8_t = w2[0];
- u32 w9_t = w2[1];
- u32 wa_t = w2[2];
- u32 wb_t = w2[3];
- u32 wc_t = w3[0];
- u32 wd_t = w3[1];
- u32 we_t = w3[2];
- u32 wf_t = w3[3];
+ u32x A = digest[0];
+ u32x B = digest[1];
+ u32x C = digest[2];
+ u32x D = digest[3];
+ u32x E = digest[4];
+
+ 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];
#undef K
#define K SHA1C00
digest[4] += E;
}
-static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
+static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
w0[0] = w0[0] ^ 0x36363636;
w0[1] = w0[1] ^ 0x36363636;
sha1_transform (w0, w1, w2, w3, opad);
}
-static void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5])
+static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
{
digest[0] = ipad[0];
digest[1] = ipad[1];
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
- u32 w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
- u32 w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
-
- u32 w2_t[4];
-
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
-
- u32 w3_t[4];
-
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = 0;
-
- u32 ipad[5];
- u32 opad[5];
+ u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
+
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
+
+ u32x ipad[5];
+ u32x opad[5];
hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
-
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
-
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
-
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
+ w3[2] = wordl3[2] | wordr3[2];
+ w3[3] = wordl3[3] | wordr3[3];
+
+ append_0x80_4x4_VV (w0, w1, w2, w3, pw_len);
+
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+ w2[0] = swap32 (w2[0]);
+ w2[1] = swap32 (w2[1]);
+ w2[2] = swap32 (w2[2]);
+ w2[3] = swap32 (w2[3]);
+ w3[0] = swap32 (w3[0]);
+ w3[1] = swap32 (w3[1]);
w3[2] = 0;
- w3[3] = 0;
-
- append_0x80_4x4 (w0, w1, w2, w3, pw_len);
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[3]);
- w2_t[0] = swap32 (w2[0]);
- w2_t[1] = swap32 (w2[1]);
- w2_t[2] = swap32 (w2[2]);
- w2_t[3] = swap32 (w2[3]);
- w3_t[0] = swap32 (w3[0]);
- w3_t[1] = swap32 (w3[1]);
- w3_t[2] = 0;
- w3_t[3] = (64 + pw_len) * 8;
-
- u32 digest[5];
-
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[4];
- const u32 r2 = digest[2];
- const u32 r3 = digest[1];
-
- #include COMPARE_M
+ w3[3] = (64 + pw_len) * 8;
+
+ u32x digest[5];
+
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+ COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* 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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
- u32 w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
- u32 w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (salt_buf1[3]);
-
- u32 w2_t[4];
-
- w2_t[0] = 0;
- w2_t[1] = 0;
- w2_t[2] = 0;
- w2_t[3] = 0;
-
- u32 w3_t[4];
-
- w3_t[0] = 0;
- w3_t[1] = 0;
- w3_t[2] = 0;
- w3_t[3] = 0;
-
- u32 ipad[5];
- u32 opad[5];
+ u32x w0_t[4];
+ u32x w1_t[4];
+ u32x w2_t[4];
+ u32x w3_t[4];
+
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
+
+ u32x ipad[5];
+ u32x opad[5];
hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
-
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
-
- u32 wordr2[4];
-
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
-
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+
+ const u32x pw_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 (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
+ w3[2] = wordl3[2] | wordr3[2];
+ w3[3] = wordl3[3] | wordr3[3];
+
+ append_0x80_4x4_VV (w0, w1, w2, w3, pw_len);
+
+ w0[0] = swap32 (w0[0]);
+ w0[1] = swap32 (w0[1]);
+ w0[2] = swap32 (w0[2]);
+ w0[3] = swap32 (w0[3]);
+ w1[0] = swap32 (w1[0]);
+ w1[1] = swap32 (w1[1]);
+ w1[2] = swap32 (w1[2]);
+ w1[3] = swap32 (w1[3]);
+ w2[0] = swap32 (w2[0]);
+ w2[1] = swap32 (w2[1]);
+ w2[2] = swap32 (w2[2]);
+ w2[3] = swap32 (w2[3]);
+ w3[0] = swap32 (w3[0]);
+ w3[1] = swap32 (w3[1]);
w3[2] = 0;
- w3[3] = 0;
-
- append_0x80_4x4 (w0, w1, w2, w3, pw_len);
-
- w0_t[0] = swap32 (w0[0]);
- w0_t[1] = swap32 (w0[1]);
- w0_t[2] = swap32 (w0[2]);
- w0_t[3] = swap32 (w0[3]);
- w1_t[0] = swap32 (w1[0]);
- w1_t[1] = swap32 (w1[1]);
- w1_t[2] = swap32 (w1[2]);
- w1_t[3] = swap32 (w1[3]);
- w2_t[0] = swap32 (w2[0]);
- w2_t[1] = swap32 (w2[1]);
- w2_t[2] = swap32 (w2[2]);
- w2_t[3] = swap32 (w2[3]);
- w3_t[0] = swap32 (w3[0]);
- w3_t[1] = swap32 (w3[1]);
- w3_t[2] = 0;
- w3_t[3] = (64 + pw_len) * 8;
-
- u32 digest[5];
-
- hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
-
- const u32 r0 = digest[3];
- const u32 r1 = digest[4];
- const u32 r2 = digest[2];
- const u32 r3 = digest[1];
-
- #include COMPARE_S
+ w3[3] = (64 + pw_len) * 8;
+
+ u32x digest[5];
+
+ hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
+
+ COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
}
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
u32x w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
u32x w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (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;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
u32x ipad[5];
u32x opad[5];
*/
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];
+ u32 salt_buf2[4];
+ u32 salt_buf3[4];
+
+ salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
+ salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
+ salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
+ salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
+ salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
+ salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
+ salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
+ salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
+ salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
+ salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
+ salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
+ salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
+ salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
+ salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
+ salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
+ salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
+
+ const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* pads
*/
u32x w0_t[4];
-
- w0_t[0] = swap32 (salt_buf0[0]);
- w0_t[1] = swap32 (salt_buf0[1]);
- w0_t[2] = swap32 (salt_buf0[2]);
- w0_t[3] = swap32 (salt_buf0[3]);
-
u32x w1_t[4];
-
- w1_t[0] = swap32 (salt_buf1[0]);
- w1_t[1] = swap32 (salt_buf1[1]);
- w1_t[2] = swap32 (salt_buf1[2]);
- w1_t[3] = swap32 (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;
+ w0_t[0] = salt_buf0[0];
+ w0_t[1] = salt_buf0[1];
+ w0_t[2] = salt_buf0[2];
+ w0_t[3] = salt_buf0[3];
+ w1_t[0] = salt_buf1[0];
+ w1_t[1] = salt_buf1[1];
+ w1_t[2] = salt_buf1[2];
+ w1_t[3] = salt_buf1[3];
+ w2_t[0] = salt_buf2[0];
+ w2_t[1] = salt_buf2[1];
+ w2_t[2] = salt_buf2[2];
+ w2_t[3] = salt_buf2[3];
+ w3_t[0] = salt_buf3[0];
+ w3_t[1] = salt_buf3[1];
+ w3_t[2] = salt_buf3[2];
+ w3_t[3] = salt_buf3[3];
u32x ipad[5];
u32x opad[5];
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
if (gid >= gid_max) return;
u32 pw_buf0[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];
-
u32 pw_buf1[4];
- 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];
+ 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_len = pws[gid].pw_len;
#define _SHA1_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-
-#define COMPARE_S "OpenCL/check_single_comp4.c"
-#define COMPARE_M "OpenCL/check_multi_comp4.c"
+#include "OpenCL/simd.c"
__kernel void m00190_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)
{
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- u32 wordl2[4];
-
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- append_0x80_2x4 (wordl0, wordl1, pw_l_len);
-
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
-
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr2[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
-
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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)
{
- append_0x80_2x4 (wordr0, wordr1, pw_r_len);
-
- switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
- w3[2] = 0;
- w3[3] = 0;
+ w3[2] = wordl3[2] | wordr3[2];
+ w3[3] = wordl3[3] | wordr3[3];
/**
* sha1
*/
- u32 w0_t = swap32 (w0[0]);
- u32 w1_t = swap32 (w0[1]);
- u32 w2_t = swap32 (w0[2]);
- u32 w3_t = swap32 (w0[3]);
- u32 w4_t = swap32 (w1[0]);
- u32 w5_t = swap32 (w1[1]);
- u32 w6_t = swap32 (w1[2]);
- u32 w7_t = swap32 (w1[3]);
- u32 w8_t = swap32 (w2[0]);
- u32 w9_t = swap32 (w2[1]);
- u32 wa_t = swap32 (w2[2]);
- u32 wb_t = swap32 (w2[3]);
- u32 wc_t = swap32 (w3[0]);
- u32 wd_t = swap32 (w3[1]);
- u32 we_t = 0;
- u32 wf_t = pw_len * 8;
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
d += SHA1M_D;
c += SHA1M_C;
- {
- const u32 r0 = a;
- const u32 r1 = e;
- const u32 r2 = d;
- const u32 r3 = c;
-
- #include COMPARE_M
- }
+ COMPARE_M_SIMD (a, e, d, c);
a &= 0x00000fff;
- {
- const u32 r0 = a;
- const u32 r1 = e;
- const u32 r2 = d;
- const u32 r3 = c;
-
- #include COMPARE_M
- }
+ COMPARE_M_SIMD (a, e, d, c);
}
}
if (gid >= gid_max) return;
- u32 wordl0[4];
-
- wordl0[0] = pws[gid].i[ 0];
- wordl0[1] = pws[gid].i[ 1];
- wordl0[2] = pws[gid].i[ 2];
- wordl0[3] = pws[gid].i[ 3];
-
- u32 wordl1[4];
-
- wordl1[0] = pws[gid].i[ 4];
- wordl1[1] = pws[gid].i[ 5];
- wordl1[2] = pws[gid].i[ 6];
- wordl1[3] = pws[gid].i[ 7];
-
- u32 wordl2[4];
+ u32 pw_buf0[4];
+ u32 pw_buf1[4];
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
-
- u32 wordl3[4];
-
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ 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;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- append_0x80_2x4 (wordl0, wordl1, pw_l_len);
-
- switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* digest
*/
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
-
- u32 wordr0[4];
-
- wordr0[0] = combs_buf[il_pos].i[0];
- wordr0[1] = combs_buf[il_pos].i[1];
- wordr0[2] = combs_buf[il_pos].i[2];
- wordr0[3] = combs_buf[il_pos].i[3];
-
- u32 wordr1[4];
-
- wordr1[0] = combs_buf[il_pos].i[4];
- wordr1[1] = combs_buf[il_pos].i[5];
- wordr1[2] = combs_buf[il_pos].i[6];
- wordr1[3] = combs_buf[il_pos].i[7];
-
- u32 wordr2[4];
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- wordr2[0] = 0;
- wordr2[1] = 0;
- wordr2[2] = 0;
- wordr2[3] = 0;
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr3[4];
+ /**
+ * concat password candidate
+ */
- wordr3[0] = 0;
- wordr3[1] = 0;
- wordr3[2] = 0;
- wordr3[3] = 0;
+ 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)
{
- append_0x80_2x4 (wordr0, wordr1, pw_r_len);
-
- switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ 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);
}
- u32 w0[4];
+ 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];
-
- u32 w1[4];
-
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
-
- u32 w2[4];
-
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
-
- u32 w3[4];
-
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
- w3[2] = 0;
- w3[3] = 0;
+ w3[2] = wordl3[2] | wordr3[2];
+ w3[3] = wordl3[3] | wordr3[3];
/**
* sha1
*/
- u32 w0_t = swap32 (w0[0]);
- u32 w1_t = swap32 (w0[1]);
- u32 w2_t = swap32 (w0[2]);
- u32 w3_t = swap32 (w0[3]);
- u32 w4_t = swap32 (w1[0]);
- u32 w5_t = swap32 (w1[1]);
- u32 w6_t = swap32 (w1[2]);
- u32 w7_t = swap32 (w1[3]);
- u32 w8_t = swap32 (w2[0]);
- u32 w9_t = swap32 (w2[1]);
- u32 wa_t = swap32 (w2[2]);
- u32 wb_t = swap32 (w2[3]);
- u32 wc_t = swap32 (w3[0]);
- u32 wd_t = swap32 (w3[1]);
- u32 we_t = 0;
- u32 wf_t = pw_len * 8;
-
- u32 a = SHA1M_A;
- u32 b = SHA1M_B;
- u32 c = SHA1M_C;
- u32 d = SHA1M_D;
- u32 e = SHA1M_E;
+ u32x w0_t = swap32 (w0[0]);
+ u32x w1_t = swap32 (w0[1]);
+ u32x w2_t = swap32 (w0[2]);
+ u32x w3_t = swap32 (w0[3]);
+ u32x w4_t = swap32 (w1[0]);
+ u32x w5_t = swap32 (w1[1]);
+ u32x w6_t = swap32 (w1[2]);
+ u32x w7_t = swap32 (w1[3]);
+ u32x w8_t = swap32 (w2[0]);
+ u32x w9_t = swap32 (w2[1]);
+ u32x wa_t = swap32 (w2[2]);
+ u32x wb_t = swap32 (w2[3]);
+ u32x wc_t = swap32 (w3[0]);
+ u32x wd_t = swap32 (w3[1]);
+ u32x we_t = 0;
+ u32x wf_t = pw_len * 8;
+
+ u32x a = SHA1M_A;
+ u32x b = SHA1M_B;
+ u32x c = SHA1M_C;
+ u32x d = SHA1M_D;
+ u32x e = SHA1M_E;
#undef K
#define K SHA1C00
d += SHA1M_D;
c += SHA1M_C;
- {
- const u32 r0 = a;
- const u32 r1 = e;
- const u32 r2 = d;
- const u32 r3 = c;
-
- #include COMPARE_S
- }
+ COMPARE_S_SIMD (a, e, d, c);
a &= 0x00000fff;
- {
- const u32 r0 = a;
- const u32 r1 = e;
- const u32 r2 = d;
- const u32 r3 = c;
-
- #include COMPARE_S
- }
+ COMPARE_S_SIMD (a, e, d, c);
}
}
digests_buf[digests_offset].digest_buf[DGST_R3]
};
- /**
+ /**
* loop
*/
projects / hashcat.git / commitdiff