#define NEW_SIMD_CODE
-#include "include/constants.h"
-#include "include/kernel_vendor.h"
+#include "inc_hash_constants.h"
+#include "inc_vendor.cl"
#define DGST_R0 3
#define DGST_R1 4
#define DGST_R2 2
#define DGST_R3 1
-#include "include/kernel_functions.c"
-#include "OpenCL/types_ocl.c"
-#include "OpenCL/common.c"
-#include "include/rp_kernel.h"
-#include "OpenCL/rp.c"
-#include "OpenCL/simd.c"
+#include "inc_hash_functions.cl"
+#include "inc_types.cl"
+#include "inc_common.cl"
+#include "inc_rp.h"
+#include "inc_rp.cl"
+#include "inc_simd.cl"
-static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
+void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = 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;
const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
u32 salt_buf0[4];
+ u32 salt_buf1[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];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
u32 domain_buf0[4];
+ u32 domain_buf1[4];
domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
-
- u32 domain_buf1[4];
-
domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ /**
+ * salt
+ */
+
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];
-
- 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] = w3[2];
w0_t[0] |= out_len & 0xff;
- /**
- * salt
- */
-
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
+
+ s0[0] = domain_buf0[0];
+ s0[1] = domain_buf0[1];
+ s0[2] = domain_buf0[2];
+ s0[3] = domain_buf0[3];
+ s1[0] = domain_buf1[0];
+ s1[1] = domain_buf1[1];
+ s1[2] = domain_buf1[2];
+ s1[3] = domain_buf1[3];
+ s2[0] = 0;
+ s2[1] = 0;
+ s2[2] = 0;
+ s2[3] = 0;
+ s3[0] = 0;
+ s3[1] = 0;
+ s3[2] = 0;
+ s3[3] = 0;
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, 1 + out_len);
+
+ 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];
+
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;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, 1 + out_len + domain_len + 1);
- u32x d0[4];
-
- d0[0] = domain_buf0[0];
- d0[1] = domain_buf0[1];
- d0[2] = domain_buf0[2];
- d0[3] = domain_buf0[3];
-
- u32x d1[4];
-
- d1[0] = domain_buf1[0];
- d1[1] = domain_buf1[1];
- d1[2] = domain_buf1[2];
- d1[3] = 0;
-
- u32x d2[4];
-
- d2[0] = 0;
- d2[1] = 0;
- d2[2] = 0;
- d2[3] = 0;
-
- u32x d3[4];
-
- d3[0] = 0;
- d3[1] = 0;
- d3[2] = 0;
- d3[3] = 0;
-
- switch_buffer_by_offset_le_VV (d0, d1, d2, d3, 1 + out_len);
+ 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];
/**
* sha1
*/
- u32x w0_t2[4];
-
- w0_t2[0] = swap32 (w0_t[0] | d0[0] | s0[0]);
- w0_t2[1] = swap32 (w0_t[1] | d0[1] | s0[1]);
- w0_t2[2] = swap32 (w0_t[2] | d0[2] | s0[2]);
- w0_t2[3] = swap32 (w0_t[3] | d0[3] | s0[3]);
-
- u32x w1_t2[4];
-
- w1_t2[0] = swap32 (w1_t[0] | d1[0] | s1[0]);
- w1_t2[1] = swap32 (w1_t[1] | d1[1] | s1[1]);
- w1_t2[2] = swap32 (w1_t[2] | d1[2] | s1[2]);
- w1_t2[3] = swap32 (w1_t[3] | d1[3] | s1[3]);
-
- u32x w2_t2[4];
-
- w2_t2[0] = swap32 (w2_t[0] | d2[0] | s2[0]);
- w2_t2[1] = swap32 (w2_t[1] | d2[1] | s2[1]);
- w2_t2[2] = swap32 (w2_t[2] | d2[2] | s2[2]);
- w2_t2[3] = swap32 (w2_t[3] | d2[3] | s2[3]);
-
- u32x w3_t2[4];
-
- w3_t2[0] = swap32 (w3_t[0] | d3[0] | s3[0]);
- w3_t2[1] = swap32 (w3_t[1] | d3[1] | s3[1]);
- w3_t2[2] = 0;
- w3_t2[3] = (1 + out_len + domain_len + 1 + salt_len) * 8;
+ 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] = 0;
+ w3_t[3] = (1 + out_len + domain_len + 1 + salt_len) * 8;
u32x digest[5];
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
- sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
+ sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
// iterations
for (u32 i = 0; i < salt_iter; i++)
{
- u32x w0_t3[4];
-
- w0_t3[0] = digest[0];
- w0_t3[1] = digest[1];
- w0_t3[2] = digest[2];
- w0_t3[3] = digest[3];
-
- u32x w1_t3[4];
-
- w1_t3[0] = digest[4];
- w1_t3[1] = swap32 (salt_buf0[0]);
- w1_t3[2] = swap32 (salt_buf0[1]);
- w1_t3[3] = swap32 (salt_buf0[2]);
-
- u32x w2_t3[4];
-
- w2_t3[0] = swap32 (salt_buf0[3]);
- w2_t3[1] = swap32 (salt_buf1[0]);
- w2_t3[2] = swap32 (salt_buf1[1]);
- w2_t3[3] = swap32 (salt_buf1[2]);
-
- u32x w3_t3[4];
-
- w3_t3[0] = swap32 (salt_buf1[3]);
- w3_t3[1] = 0;
- w3_t3[2] = 0;
- w3_t3[3] = (20 + salt_len) * 8;
+ w0_t[0] = digest[0];
+ w0_t[1] = digest[1];
+ w0_t[2] = digest[2];
+ w0_t[3] = digest[3];
+ w1_t[0] = digest[4];
+ w1_t[1] = swap32 (salt_buf0[0]);
+ w1_t[2] = swap32 (salt_buf0[1]);
+ w1_t[3] = swap32 (salt_buf0[2]);
+ w2_t[0] = swap32 (salt_buf0[3]);
+ w2_t[1] = swap32 (salt_buf1[0]);
+ w2_t[2] = swap32 (salt_buf1[1]);
+ w2_t[3] = swap32 (salt_buf1[2]);
+ w3_t[0] = swap32 (salt_buf1[3]);
+ w3_t[1] = 0;
+ w3_t[2] = 0;
+ w3_t[3] = (20 + salt_len) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
- sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
+ sha1_transform (w0_t, w1_t, w2_t, w3_t, 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;
const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
u32 salt_buf0[4];
+ u32 salt_buf1[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];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
u32 domain_buf0[4];
+ u32 domain_buf1[4];
domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
-
- u32 domain_buf1[4];
-
domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
+ /**
+ * salt
+ */
+
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];
-
- 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] = w3[2];
w0_t[0] |= out_len & 0xff;
- /**
- * salt
- */
-
u32x s0[4];
+ u32x s1[4];
+ u32x s2[4];
+ u32x s3[4];
+
+ s0[0] = domain_buf0[0];
+ s0[1] = domain_buf0[1];
+ s0[2] = domain_buf0[2];
+ s0[3] = domain_buf0[3];
+ s1[0] = domain_buf1[0];
+ s1[1] = domain_buf1[1];
+ s1[2] = domain_buf1[2];
+ s1[3] = domain_buf1[3];
+ s2[0] = 0;
+ s2[1] = 0;
+ s2[2] = 0;
+ s2[3] = 0;
+ s3[0] = 0;
+ s3[1] = 0;
+ s3[2] = 0;
+ s3[3] = 0;
+
+ switch_buffer_by_offset_le_VV (s0, s1, s2, s3, 1 + out_len);
+
+ 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];
+
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;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, 1 + out_len + domain_len + 1);
- u32x d0[4];
-
- d0[0] = domain_buf0[0];
- d0[1] = domain_buf0[1];
- d0[2] = domain_buf0[2];
- d0[3] = domain_buf0[3];
-
- u32x d1[4];
-
- d1[0] = domain_buf1[0];
- d1[1] = domain_buf1[1];
- d1[2] = domain_buf1[2];
- d1[3] = 0;
-
- u32x d2[4];
-
- d2[0] = 0;
- d2[1] = 0;
- d2[2] = 0;
- d2[3] = 0;
-
- u32x d3[4];
-
- d3[0] = 0;
- d3[1] = 0;
- d3[2] = 0;
- d3[3] = 0;
-
- switch_buffer_by_offset_le_VV (d0, d1, d2, d3, 1 + out_len);
+ 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];
/**
* sha1
*/
- u32x w0_t2[4];
-
- w0_t2[0] = swap32 (w0_t[0] | d0[0] | s0[0]);
- w0_t2[1] = swap32 (w0_t[1] | d0[1] | s0[1]);
- w0_t2[2] = swap32 (w0_t[2] | d0[2] | s0[2]);
- w0_t2[3] = swap32 (w0_t[3] | d0[3] | s0[3]);
-
- u32x w1_t2[4];
-
- w1_t2[0] = swap32 (w1_t[0] | d1[0] | s1[0]);
- w1_t2[1] = swap32 (w1_t[1] | d1[1] | s1[1]);
- w1_t2[2] = swap32 (w1_t[2] | d1[2] | s1[2]);
- w1_t2[3] = swap32 (w1_t[3] | d1[3] | s1[3]);
-
- u32x w2_t2[4];
-
- w2_t2[0] = swap32 (w2_t[0] | d2[0] | s2[0]);
- w2_t2[1] = swap32 (w2_t[1] | d2[1] | s2[1]);
- w2_t2[2] = swap32 (w2_t[2] | d2[2] | s2[2]);
- w2_t2[3] = swap32 (w2_t[3] | d2[3] | s2[3]);
-
- u32x w3_t2[4];
-
- w3_t2[0] = swap32 (w3_t[0] | d3[0] | s3[0]);
- w3_t2[1] = swap32 (w3_t[1] | d3[1] | s3[1]);
- w3_t2[2] = 0;
- w3_t2[3] = (1 + out_len + domain_len + 1 + salt_len) * 8;
+ 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] = 0;
+ w3_t[3] = (1 + out_len + domain_len + 1 + salt_len) * 8;
u32x digest[5];
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
- sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
+ sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
// iterations
for (u32 i = 0; i < salt_iter; i++)
{
- u32x w0_t3[4];
-
- w0_t3[0] = digest[0];
- w0_t3[1] = digest[1];
- w0_t3[2] = digest[2];
- w0_t3[3] = digest[3];
-
- u32x w1_t3[4];
-
- w1_t3[0] = digest[4];
- w1_t3[1] = swap32 (salt_buf0[0]);
- w1_t3[2] = swap32 (salt_buf0[1]);
- w1_t3[3] = swap32 (salt_buf0[2]);
-
- u32x w2_t3[4];
-
- w2_t3[0] = swap32 (salt_buf0[3]);
- w2_t3[1] = swap32 (salt_buf1[0]);
- w2_t3[2] = swap32 (salt_buf1[1]);
- w2_t3[3] = swap32 (salt_buf1[2]);
-
- u32x w3_t3[4];
-
- w3_t3[0] = swap32 (salt_buf1[3]);
- w3_t3[1] = 0;
- w3_t3[2] = 0;
- w3_t3[3] = (20 + salt_len) * 8;
+ w0_t[0] = digest[0];
+ w0_t[1] = digest[1];
+ w0_t[2] = digest[2];
+ w0_t[3] = digest[3];
+ w1_t[0] = digest[4];
+ w1_t[1] = swap32 (salt_buf0[0]);
+ w1_t[2] = swap32 (salt_buf0[1]);
+ w1_t[3] = swap32 (salt_buf0[2]);
+ w2_t[0] = swap32 (salt_buf0[3]);
+ w2_t[1] = swap32 (salt_buf1[0]);
+ w2_t[2] = swap32 (salt_buf1[1]);
+ w2_t[3] = swap32 (salt_buf1[2]);
+ w3_t[0] = swap32 (salt_buf1[3]);
+ w3_t[1] = 0;
+ w3_t[2] = 0;
+ w3_t[3] = (20 + salt_len) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
- sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
+ sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
}
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);