arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-22 08:08:10 +00:00
Code Issues Releases Wiki Activity

Add HMAC functionality to inc_hash_md4.cl and make DCC2 use it, resulting in support for longer domain names

Browse Source
This commit is contained in:
jsteube 2017-06-19 16:42:21 +02:00
parent 4174f06008
commit 2c92465036
3 changed files with 1205 additions and 705 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

438
OpenCL/inc_hash_md4.cl
View File

@ -4,20 +4,20 @@
// input buf need to be in algorithm native byte order (md5 = LE, sha1 = BE, etc)
// input buf need to be 64 byte aligned when usin md5_update()
typedef struct md4_ctx_scalar
typedef struct md4_ctx
{
u32 h[4];
u32 h[4];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int len;
int len;
} md4_ctx_scalar_t;
} md4_ctx_t;
void md4_transform_scalar (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
void md4_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
{
u32 a = digest[0];
u32 b = digest[1];
@ -81,59 +81,59 @@ void md4_transform_scalar (const u32 w0[4], const u32 w1[4], const u32 w2[4], co
digest[3] += d;
}
void md4_init_scalar (md4_ctx_scalar_t *md4_ctx)
void md4_init (md4_ctx_t *ctx)
{
md4_ctx->h[0] = MD4M_A;
md4_ctx->h[1] = MD4M_B;
md4_ctx->h[2] = MD4M_C;
md4_ctx->h[3] = MD4M_D;
ctx->h[0] = MD4M_A;
ctx->h[1] = MD4M_B;
ctx->h[2] = MD4M_C;
ctx->h[3] = MD4M_D;
md4_ctx->w0[0] = 0;
md4_ctx->w0[1] = 0;
md4_ctx->w0[2] = 0;
md4_ctx->w0[3] = 0;
md4_ctx->w1[0] = 0;
md4_ctx->w1[1] = 0;
md4_ctx->w1[2] = 0;
md4_ctx->w1[3] = 0;
md4_ctx->w2[0] = 0;
md4_ctx->w2[1] = 0;
md4_ctx->w2[2] = 0;
md4_ctx->w2[3] = 0;
md4_ctx->w3[0] = 0;
md4_ctx->w3[1] = 0;
md4_ctx->w3[2] = 0;
md4_ctx->w3[3] = 0;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
md4_ctx->len = 0;
ctx->len = 0;
}
void md4_update_scalar_64 (md4_ctx_scalar_t *md4_ctx, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const int len)
void md4_update_64 (md4_ctx_t *ctx, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const int len)
{
const int pos = md4_ctx->len & 0x3f;
const int pos = ctx->len & 63;
md4_ctx->len += len;
ctx->len += len;
if ((pos + len) < 64)
{
switch_buffer_by_offset_le_S (w0, w1, w2, w3, pos);
md4_ctx->w0[0] |= w0[0];
md4_ctx->w0[1] |= w0[1];
md4_ctx->w0[2] |= w0[2];
md4_ctx->w0[3] |= w0[3];
md4_ctx->w1[0] |= w1[0];
md4_ctx->w1[1] |= w1[1];
md4_ctx->w1[2] |= w1[2];
md4_ctx->w1[3] |= w1[3];
md4_ctx->w2[0] |= w2[0];
md4_ctx->w2[1] |= w2[1];
md4_ctx->w2[2] |= w2[2];
md4_ctx->w2[3] |= w2[3];
md4_ctx->w3[0] |= w3[0];
md4_ctx->w3[1] |= w3[1];
md4_ctx->w3[2] |= w3[2];
md4_ctx->w3[3] |= w3[3];
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
}
else
{
@ -144,45 +144,45 @@ void md4_update_scalar_64 (md4_ctx_scalar_t *md4_ctx, u32 w0[4], u32 w1[4], u32
switch_buffer_by_offset_carry_le_S (w0, w1, w2, w3, c0, c1, c2, c3, pos);
md4_ctx->w0[0] |= w0[0];
md4_ctx->w0[1] |= w0[1];
md4_ctx->w0[2] |= w0[2];
md4_ctx->w0[3] |= w0[3];
md4_ctx->w1[0] |= w1[0];
md4_ctx->w1[1] |= w1[1];
md4_ctx->w1[2] |= w1[2];
md4_ctx->w1[3] |= w1[3];
md4_ctx->w2[0] |= w2[0];
md4_ctx->w2[1] |= w2[1];
md4_ctx->w2[2] |= w2[2];
md4_ctx->w2[3] |= w2[3];
md4_ctx->w3[0] |= w3[0];
md4_ctx->w3[1] |= w3[1];
md4_ctx->w3[2] |= w3[2];
md4_ctx->w3[3] |= w3[3];
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
md4_transform_scalar (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
md4_ctx->w0[0] = c0[0];
md4_ctx->w0[1] = c0[1];
md4_ctx->w0[2] = c0[2];
md4_ctx->w0[3] = c0[3];
md4_ctx->w1[0] = c1[0];
md4_ctx->w1[1] = c1[1];
md4_ctx->w1[2] = c1[2];
md4_ctx->w1[3] = c1[3];
md4_ctx->w2[0] = c2[0];
md4_ctx->w2[1] = c2[1];
md4_ctx->w2[2] = c2[2];
md4_ctx->w2[3] = c2[3];
md4_ctx->w3[0] = c3[0];
md4_ctx->w3[1] = c3[1];
md4_ctx->w3[2] = c3[2];
md4_ctx->w3[3] = c3[3];
ctx->w0[0] = c0[0];
ctx->w0[1] = c0[1];
ctx->w0[2] = c0[2];
ctx->w0[3] = c0[3];
ctx->w1[0] = c1[0];
ctx->w1[1] = c1[1];
ctx->w1[2] = c1[2];
ctx->w1[3] = c1[3];
ctx->w2[0] = c2[0];
ctx->w2[1] = c2[1];
ctx->w2[2] = c2[2];
ctx->w2[3] = c2[3];
ctx->w3[0] = c3[0];
ctx->w3[1] = c3[1];
ctx->w3[2] = c3[2];
ctx->w3[3] = c3[3];
}
}
void md4_update_scalar (md4_ctx_scalar_t *md4_ctx, const u32 *w, const int len)
void md4_update (md4_ctx_t *ctx, const u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
@ -211,7 +211,7 @@ void md4_update_scalar (md4_ctx_scalar_t *md4_ctx, const u32 *w, const int len)
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, 64);
md4_update_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
@ -231,10 +231,10 @@ void md4_update_scalar (md4_ctx_scalar_t *md4_ctx, const u32 *w, const int len)
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, len & 0x3f);
md4_update_64 (ctx, w0, w1, w2, w3, len & 63);
}
void md4_update_scalar_global (md4_ctx_scalar_t *md4_ctx, const __global u32 *w, const int len)
void md4_update_global (md4_ctx_t *ctx, const __global u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
@ -263,7 +263,7 @@ void md4_update_scalar_global (md4_ctx_scalar_t *md4_ctx, const __global u32 *w,
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, 64);
md4_update_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
@ -283,10 +283,10 @@ void md4_update_scalar_global (md4_ctx_scalar_t *md4_ctx, const __global u32 *w,
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, len & 0x3f);
md4_update_64 (ctx, w0, w1, w2, w3, len & 63);
}
void md4_update_scalar_global_utf16le (md4_ctx_scalar_t *md4_ctx, const __global u32 *w, const int len)
void md4_update_global_utf16le (md4_ctx_t *ctx, const __global u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
@ -310,7 +310,7 @@ void md4_update_scalar_global_utf16le (md4_ctx_scalar_t *md4_ctx, const __global
make_utf16le_S (w1, w2, w3);
make_utf16le_S (w0, w0, w1);
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, 64);
md4_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
}
w0[0] = w[i + 0];
@ -325,45 +325,46 @@ void md4_update_scalar_global_utf16le (md4_ctx_scalar_t *md4_ctx, const __global
make_utf16le_S (w1, w2, w3);
make_utf16le_S (w0, w0, w1);
md4_update_scalar_64 (md4_ctx, w0, w1, w2, w3, (len * 2) & 0x3f);
md4_update_64 (ctx, w0, w1, w2, w3, (len & 31) * 2);
}
void md4_final_scalar (md4_ctx_scalar_t *md4_ctx)
void md4_final (md4_ctx_t *ctx)
{
const int pos = md4_ctx->len & 0x3f;
const int pos = ctx->len & 63;
append_0x80_4x4_S (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, pos);
append_0x80_4x4_S (ctx->w0, ctx->w1, ctx->w2, ctx->w3, pos);
if (pos >= 56)
{
md4_transform_scalar (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
md4_ctx->w0[0] = 0;
md4_ctx->w0[1] = 0;
md4_ctx->w0[2] = 0;
md4_ctx->w0[3] = 0;
md4_ctx->w1[0] = 0;
md4_ctx->w1[1] = 0;
md4_ctx->w1[2] = 0;
md4_ctx->w1[3] = 0;
md4_ctx->w2[0] = 0;
md4_ctx->w2[1] = 0;
md4_ctx->w2[2] = 0;
md4_ctx->w2[3] = 0;
md4_ctx->w3[0] = 0;
md4_ctx->w3[1] = 0;
md4_ctx->w3[2] = 0;
md4_ctx->w3[3] = 0;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
}
md4_ctx->w3[2] = md4_ctx->len * 8;
ctx->w3[2] = ctx->len * 8;
ctx->w3[3] = 0;
md4_transform_scalar (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
}
void md4_optimize_max_length_scalar (md4_ctx_scalar_t *md4_ctx, const int bits)
void md4_optimize_max_length (md4_ctx_t *ctx, const int bits)
{
md4_ctx->len &= (1 << bits) - 1;
ctx->len &= (1 << bits) - 1;
}
// while input buf can be a vector datatype, the length of the different elements can not
@ -445,59 +446,59 @@ void md4_transform_vector (const u32x w0[4], const u32x w1[4], const u32x w2[4],
digest[3] += d;
}
void md4_init_vector (md4_ctx_vector_t *md4_ctx)
void md4_init_vector (md4_ctx_vector_t *ctx)
{
md4_ctx->h[0] = MD4M_A;
md4_ctx->h[1] = MD4M_B;
md4_ctx->h[2] = MD4M_C;
md4_ctx->h[3] = MD4M_D;
ctx->h[0] = MD4M_A;
ctx->h[1] = MD4M_B;
ctx->h[2] = MD4M_C;
ctx->h[3] = MD4M_D;
md4_ctx->w0[0] = 0;
md4_ctx->w0[1] = 0;
md4_ctx->w0[2] = 0;
md4_ctx->w0[3] = 0;
md4_ctx->w1[0] = 0;
md4_ctx->w1[1] = 0;
md4_ctx->w1[2] = 0;
md4_ctx->w1[3] = 0;
md4_ctx->w2[0] = 0;
md4_ctx->w2[1] = 0;
md4_ctx->w2[2] = 0;
md4_ctx->w2[3] = 0;
md4_ctx->w3[0] = 0;
md4_ctx->w3[1] = 0;
md4_ctx->w3[2] = 0;
md4_ctx->w3[3] = 0;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
md4_ctx->len = 0;
ctx->len = 0;
}
void md4_update_vector_64 (md4_ctx_vector_t *md4_ctx, u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const int len)
void md4_update_vector_64 (md4_ctx_vector_t *ctx, u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const int len)
{
const int pos = md4_ctx->len & 0x3f;
const int pos = ctx->len & 63;
md4_ctx->len += len;
ctx->len += len;
if ((pos + len) < 64)
{
switch_buffer_by_offset_le (w0, w1, w2, w3, pos);
md4_ctx->w0[0] |= w0[0];
md4_ctx->w0[1] |= w0[1];
md4_ctx->w0[2] |= w0[2];
md4_ctx->w0[3] |= w0[3];
md4_ctx->w1[0] |= w1[0];
md4_ctx->w1[1] |= w1[1];
md4_ctx->w1[2] |= w1[2];
md4_ctx->w1[3] |= w1[3];
md4_ctx->w2[0] |= w2[0];
md4_ctx->w2[1] |= w2[1];
md4_ctx->w2[2] |= w2[2];
md4_ctx->w2[3] |= w2[3];
md4_ctx->w3[0] |= w3[0];
md4_ctx->w3[1] |= w3[1];
md4_ctx->w3[2] |= w3[2];
md4_ctx->w3[3] |= w3[3];
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
}
else
{
@ -508,45 +509,45 @@ void md4_update_vector_64 (md4_ctx_vector_t *md4_ctx, u32x w0[4], u32x w1[4], u3
switch_buffer_by_offset_carry_le (w0, w1, w2, w3, c0, c1, c2, c3, pos);
md4_ctx->w0[0] |= w0[0];
md4_ctx->w0[1] |= w0[1];
md4_ctx->w0[2] |= w0[2];
md4_ctx->w0[3] |= w0[3];
md4_ctx->w1[0] |= w1[0];
md4_ctx->w1[1] |= w1[1];
md4_ctx->w1[2] |= w1[2];
md4_ctx->w1[3] |= w1[3];
md4_ctx->w2[0] |= w2[0];
md4_ctx->w2[1] |= w2[1];
md4_ctx->w2[2] |= w2[2];
md4_ctx->w2[3] |= w2[3];
md4_ctx->w3[0] |= w3[0];
md4_ctx->w3[1] |= w3[1];
md4_ctx->w3[2] |= w3[2];
md4_ctx->w3[3] |= w3[3];
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
md4_transform_vector (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
md4_ctx->w0[0] = c0[0];
md4_ctx->w0[1] = c0[1];
md4_ctx->w0[2] = c0[2];
md4_ctx->w0[3] = c0[3];
md4_ctx->w1[0] = c1[0];
md4_ctx->w1[1] = c1[1];
md4_ctx->w1[2] = c1[2];
md4_ctx->w1[3] = c1[3];
md4_ctx->w2[0] = c2[0];
md4_ctx->w2[1] = c2[1];
md4_ctx->w2[2] = c2[2];
md4_ctx->w2[3] = c2[3];
md4_ctx->w3[0] = c3[0];
md4_ctx->w3[1] = c3[1];
md4_ctx->w3[2] = c3[2];
md4_ctx->w3[3] = c3[3];
ctx->w0[0] = c0[0];
ctx->w0[1] = c0[1];
ctx->w0[2] = c0[2];
ctx->w0[3] = c0[3];
ctx->w1[0] = c1[0];
ctx->w1[1] = c1[1];
ctx->w1[2] = c1[2];
ctx->w1[3] = c1[3];
ctx->w2[0] = c2[0];
ctx->w2[1] = c2[1];
ctx->w2[2] = c2[2];
ctx->w2[3] = c2[3];
ctx->w3[0] = c3[0];
ctx->w3[1] = c3[1];
ctx->w3[2] = c3[2];
ctx->w3[3] = c3[3];
}
}
void md4_update_vector (md4_ctx_vector_t *md4_ctx, const u32x *w, const int len)
void md4_update_vector (md4_ctx_vector_t *ctx, const u32x *w, const int len)
{
u32x w0[4];
u32x w1[4];
@ -575,7 +576,7 @@ void md4_update_vector (md4_ctx_vector_t *md4_ctx, const u32x *w, const int len)
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_vector_64 (md4_ctx, w0, w1, w2, w3, 64);
md4_update_vector_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
@ -595,45 +596,44 @@ void md4_update_vector (md4_ctx_vector_t *md4_ctx, const u32x *w, const int len)
w3[2] = w[i + 14];
w3[3] = w[i + 15];
md4_update_vector_64 (md4_ctx, w0, w1, w2, w3, len & 0x3f);
md4_update_vector_64 (ctx, w0, w1, w2, w3, len & 63);
}
void md4_final_vector (md4_ctx_vector_t *md4_ctx)
void md4_final_vector (md4_ctx_vector_t *ctx)
{
const int pos = md4_ctx->len & 0x3f;
const int pos = ctx->len & 63;
append_0x80_4x4 (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, pos);
append_0x80_4x4 (ctx->w0, ctx->w1, ctx->w2, ctx->w3, pos);
if (pos >= 56)
{
md4_transform_vector (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
md4_ctx->w0[0] = 0;
md4_ctx->w0[1] = 0;
md4_ctx->w0[2] = 0;
md4_ctx->w0[3] = 0;
md4_ctx->w1[0] = 0;
md4_ctx->w1[1] = 0;
md4_ctx->w1[2] = 0;
md4_ctx->w1[3] = 0;
md4_ctx->w2[0] = 0;
md4_ctx->w2[1] = 0;
md4_ctx->w2[2] = 0;
md4_ctx->w2[3] = 0;
md4_ctx->w3[0] = 0;
md4_ctx->w3[1] = 0;
md4_ctx->w3[2] = 0;
md4_ctx->w3[3] = 0;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
}
md4_ctx->w3[2] = md4_ctx->len * 8;
ctx->w3[2] = ctx->len * 8;
ctx->w3[3] = 0;
md4_transform_vector (md4_ctx->w0, md4_ctx->w1, md4_ctx->w2, md4_ctx->w3, md4_ctx->h);
md4_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
}
void md4_optimize_max_length_vector (md4_ctx_vector_t *md4_ctx, const int bits)
void md4_optimize_max_length_vector (md4_ctx_vector_t *ctx, const int bits)
{
md4_ctx->len &= (1 << bits) - 1;
ctx->len &= (1 << bits) - 1;
}

935
OpenCL/inc_hash_sha1.cl Normal file
View File

@ -0,0 +1,935 @@
// important notes on this:
// input buf unused bytes needs to be set to zero
// input buf need to be in algorithm native byte order (md5 = LE, sha1 = BE, etc)
// input buf need to be 64 byte aligned when usin md5_update()
typedef struct sha1_ctx
{
u32 h[5];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int len;
} sha1_ctx_t;
void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 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];
#undef K
#define K SHA1C00
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w0_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w1_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w2_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w3_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w4_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w5_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w6_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w7_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w8_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w9_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, wa_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, wb_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, wc_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, wd_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, we_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w3_t);
#undef K
#define K SHA1C01
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w7_t);
#undef K
#define K SHA1C02
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, wb_t);
#undef K
#define K SHA1C03
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, wf_t);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
}
void sha1_init (sha1_ctx_t *ctx)
{
ctx->h[0] = SHA1M_A;
ctx->h[1] = SHA1M_B;
ctx->h[2] = SHA1M_C;
ctx->h[3] = SHA1M_D;
ctx->h[4] = SHA1M_E;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
ctx->len = 0;
}
void sha1_update_64 (sha1_ctx_t *ctx, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const int len)
{
const int pos = ctx->len & 63;
ctx->len += len;
if ((pos + len) < 64)
{
switch_buffer_by_offset_be_S (w0, w1, w2, w3, pos);
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
}
else
{
u32 c0[4] = { 0 };
u32 c1[4] = { 0 };
u32 c2[4] = { 0 };
u32 c3[4] = { 0 };
switch_buffer_by_offset_carry_be_S (w0, w1, w2, w3, c0, c1, c2, c3, pos);
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
sha1_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
ctx->w0[0] = c0[0];
ctx->w0[1] = c0[1];
ctx->w0[2] = c0[2];
ctx->w0[3] = c0[3];
ctx->w1[0] = c1[0];
ctx->w1[1] = c1[1];
ctx->w1[2] = c1[2];
ctx->w1[3] = c1[3];
ctx->w2[0] = c2[0];
ctx->w2[1] = c2[1];
ctx->w2[2] = c2[2];
ctx->w2[3] = c2[3];
ctx->w3[0] = c3[0];
ctx->w3[1] = c3[1];
ctx->w3[2] = c3[2];
ctx->w3[3] = c3[3];
}
}
void sha1_update (sha1_ctx_t *ctx, const u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int i;
int j;
for (i = 0, j = 0; i < len - 64; i += 64, j += 16)
{
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_64 (ctx, w0, w1, w2, w3, len & 63);
}
void sha1_update_global (sha1_ctx_t *ctx, const __global u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int i;
int j;
for (i = 0, j = 0; i < len - 64; i += 64, j += 16)
{
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_64 (ctx, w0, w1, w2, w3, len & 63);
}
void sha1_update_global_swap (sha1_ctx_t *ctx, const __global u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int i;
int j;
for (i = 0, j = 0; i < len - 64; i += 64, j += 16)
{
w0[0] = swap32_S (w[i + 0]);
w0[1] = swap32_S (w[i + 1]);
w0[2] = swap32_S (w[i + 2]);
w0[3] = swap32_S (w[i + 3]);
w1[0] = swap32_S (w[i + 4]);
w1[1] = swap32_S (w[i + 5]);
w1[2] = swap32_S (w[i + 6]);
w1[3] = swap32_S (w[i + 7]);
w2[0] = swap32_S (w[i + 8]);
w2[1] = swap32_S (w[i + 9]);
w2[2] = swap32_S (w[i + 10]);
w2[3] = swap32_S (w[i + 11]);
w3[0] = swap32_S (w[i + 12]);
w3[1] = swap32_S (w[i + 13]);
w3[2] = swap32_S (w[i + 14]);
w3[3] = swap32_S (w[i + 15]);
sha1_update_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = swap32_S (w[i + 0]);
w0[1] = swap32_S (w[i + 1]);
w0[2] = swap32_S (w[i + 2]);
w0[3] = swap32_S (w[i + 3]);
w1[0] = swap32_S (w[i + 4]);
w1[1] = swap32_S (w[i + 5]);
w1[2] = swap32_S (w[i + 6]);
w1[3] = swap32_S (w[i + 7]);
w2[0] = swap32_S (w[i + 8]);
w2[1] = swap32_S (w[i + 9]);
w2[2] = swap32_S (w[i + 10]);
w2[3] = swap32_S (w[i + 11]);
w3[0] = swap32_S (w[i + 12]);
w3[1] = swap32_S (w[i + 13]);
w3[2] = swap32_S (w[i + 14]);
w3[3] = swap32_S (w[i + 15]);
sha1_update_64 (ctx, w0, w1, w2, w3, len & 63);
}
void sha1_update_global_utf16le (sha1_ctx_t *ctx, const __global u32 *w, const int len)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int i;
int j;
for (i = 0, j = 0; i < len - 32; i += 32, j += 8)
{
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
make_utf16le_S (w1, w2, w3);
make_utf16le_S (w0, w0, w1);
sha1_update_64 (ctx, w0, w1, w2, w3, 32 * 2);
}
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
make_utf16le_S (w1, w2, w3);
make_utf16le_S (w0, w0, w1);
sha1_update_64 (ctx, w0, w1, w2, w3, (len & 31) * 2);
}
void sha1_final (sha1_ctx_t *ctx)
{
int pos = ctx->len & 63;
append_0x80_4x4_S (ctx->w0, ctx->w1, ctx->w2, ctx->w3, pos ^ 3);
if (pos >= 56)
{
sha1_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
}
ctx->w3[2] = 0;
ctx->w3[3] = ctx->len * 8;
sha1_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
}
void sha1_optimize_max_length (sha1_ctx_t *ctx, const int bits)
{
ctx->len &= (1 << bits) - 1;
}
// sha1_hmac
typedef struct sha1_hmac_ctx
{
sha1_ctx_t ipad;
sha1_ctx_t opad;
} sha1_hmac_ctx_t;
void sha1_hmac_init (sha1_hmac_ctx_t *ctx, const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4])
{
u32 t0[4];
u32 t1[4];
u32 t2[4];
u32 t3[4];
// ipad
t0[0] = w0[0] ^ 0x36363636;
t0[1] = w0[1] ^ 0x36363636;
t0[2] = w0[2] ^ 0x36363636;
t0[3] = w0[3] ^ 0x36363636;
t1[0] = w1[0] ^ 0x36363636;
t1[1] = w1[1] ^ 0x36363636;
t1[2] = w1[2] ^ 0x36363636;
t1[3] = w1[3] ^ 0x36363636;
t2[0] = w2[0] ^ 0x36363636;
t2[1] = w2[1] ^ 0x36363636;
t2[2] = w2[2] ^ 0x36363636;
t2[3] = w2[3] ^ 0x36363636;
t3[0] = w3[0] ^ 0x36363636;
t3[1] = w3[1] ^ 0x36363636;
t3[2] = w3[2] ^ 0x36363636;
t3[3] = w3[3] ^ 0x36363636;
sha1_init (&ctx->ipad);
sha1_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
t0[0] = w0[0] ^ 0x5c5c5c5c;
t0[1] = w0[1] ^ 0x5c5c5c5c;
t0[2] = w0[2] ^ 0x5c5c5c5c;
t0[3] = w0[3] ^ 0x5c5c5c5c;
t1[0] = w1[0] ^ 0x5c5c5c5c;
t1[1] = w1[1] ^ 0x5c5c5c5c;
t1[2] = w1[2] ^ 0x5c5c5c5c;
t1[3] = w1[3] ^ 0x5c5c5c5c;
t2[0] = w2[0] ^ 0x5c5c5c5c;
t2[1] = w2[1] ^ 0x5c5c5c5c;
t2[2] = w2[2] ^ 0x5c5c5c5c;
t2[3] = w2[3] ^ 0x5c5c5c5c;
t3[0] = w3[0] ^ 0x5c5c5c5c;
t3[1] = w3[1] ^ 0x5c5c5c5c;
t3[2] = w3[2] ^ 0x5c5c5c5c;
t3[3] = w3[3] ^ 0x5c5c5c5c;
sha1_init (&ctx->opad);
sha1_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
void sha1_hmac_update_64 (sha1_hmac_ctx_t *ctx, u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const int len)
{
sha1_update_64 (&ctx->ipad, w0, w1, w2, w3, len);
}
void sha1_hmac_update (sha1_hmac_ctx_t *ctx, const u32 *w, const int len)
{
sha1_update (&ctx->ipad, w, len);
}
void sha1_hmac_update_global (sha1_hmac_ctx_t *ctx, const __global u32 *w, const int len)
{
sha1_update_global (&ctx->ipad, w, len);
}
void sha1_hmac_update_global_swap (sha1_hmac_ctx_t *ctx, const __global u32 *w, const int len)
{
sha1_update_global_swap (&ctx->ipad, w, len);
}
void sha1_hmac_final (sha1_hmac_ctx_t *ctx)
{
sha1_final (&ctx->ipad);
u32 t0[4];
u32 t1[4];
u32 t2[4];
u32 t3[4];
t0[0] = ctx->ipad.h[0];
t0[1] = ctx->ipad.h[1];
t0[2] = ctx->ipad.h[2];
t0[3] = ctx->ipad.h[3];
t1[0] = ctx->ipad.h[4];
t1[1] = 0;
t1[2] = 0;
t1[3] = 0;
t2[0] = 0;
t2[1] = 0;
t2[2] = 0;
t2[3] = 0;
t3[0] = 0;
t3[1] = 0;
t3[2] = 0;
t3[3] = 0;
sha1_update_64 (&ctx->opad, t0, t1, t2, t3, 20);
sha1_final (&ctx->opad);
}
// while input buf can be a vector datatype, the length of the different elements can not
typedef struct sha1_ctx_vector
{
u32x h[5];
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
int len;
} sha1_ctx_vector_t;
void sha1_transform_vector (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];
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
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
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
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
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);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
}
void sha1_init_vector (sha1_ctx_vector_t *ctx)
{
ctx->h[0] = SHA1M_A;
ctx->h[1] = SHA1M_B;
ctx->h[2] = SHA1M_C;
ctx->h[3] = SHA1M_D;
ctx->h[4] = SHA1M_E;
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
ctx->len = 0;
}
void sha1_update_vector_64 (sha1_ctx_vector_t *ctx, u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const int len)
{
const int pos = ctx->len & 63;
ctx->len += len;
if ((pos + len) < 64)
{
switch_buffer_by_offset_be (w0, w1, w2, w3, pos);
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
}
else
{
u32x c0[4] = { 0 };
u32x c1[4] = { 0 };
u32x c2[4] = { 0 };
u32x c3[4] = { 0 };
switch_buffer_by_offset_carry_be (w0, w1, w2, w3, c0, c1, c2, c3, pos);
ctx->w0[0] |= w0[0];
ctx->w0[1] |= w0[1];
ctx->w0[2] |= w0[2];
ctx->w0[3] |= w0[3];
ctx->w1[0] |= w1[0];
ctx->w1[1] |= w1[1];
ctx->w1[2] |= w1[2];
ctx->w1[3] |= w1[3];
ctx->w2[0] |= w2[0];
ctx->w2[1] |= w2[1];
ctx->w2[2] |= w2[2];
ctx->w2[3] |= w2[3];
ctx->w3[0] |= w3[0];
ctx->w3[1] |= w3[1];
ctx->w3[2] |= w3[2];
ctx->w3[3] |= w3[3];
sha1_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
ctx->w0[0] = c0[0];
ctx->w0[1] = c0[1];
ctx->w0[2] = c0[2];
ctx->w0[3] = c0[3];
ctx->w1[0] = c1[0];
ctx->w1[1] = c1[1];
ctx->w1[2] = c1[2];
ctx->w1[3] = c1[3];
ctx->w2[0] = c2[0];
ctx->w2[1] = c2[1];
ctx->w2[2] = c2[2];
ctx->w2[3] = c2[3];
ctx->w3[0] = c3[0];
ctx->w3[1] = c3[1];
ctx->w3[2] = c3[2];
ctx->w3[3] = c3[3];
}
}
void sha1_update_vector (sha1_ctx_vector_t *ctx, const u32x *w, const int len)
{
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
int i;
int j;
for (i = 0, j = 0; i < len - 64; i += 64, j += 16)
{
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_vector_64 (ctx, w0, w1, w2, w3, 64);
}
w0[0] = w[i + 0];
w0[1] = w[i + 1];
w0[2] = w[i + 2];
w0[3] = w[i + 3];
w1[0] = w[i + 4];
w1[1] = w[i + 5];
w1[2] = w[i + 6];
w1[3] = w[i + 7];
w2[0] = w[i + 8];
w2[1] = w[i + 9];
w2[2] = w[i + 10];
w2[3] = w[i + 11];
w3[0] = w[i + 12];
w3[1] = w[i + 13];
w3[2] = w[i + 14];
w3[3] = w[i + 15];
sha1_update_vector_64 (ctx, w0, w1, w2, w3, len & 63);
}
void sha1_final_vector (sha1_ctx_vector_t *ctx)
{
int pos = ctx->len & 63;
append_0x80_4x4 (ctx->w0, ctx->w1, ctx->w2, ctx->w3, pos ^ 3);
if (pos >= 56)
{
sha1_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
ctx->w0[0] = 0;
ctx->w0[1] = 0;
ctx->w0[2] = 0;
ctx->w0[3] = 0;
ctx->w1[0] = 0;
ctx->w1[1] = 0;
ctx->w1[2] = 0;
ctx->w1[3] = 0;
ctx->w2[0] = 0;
ctx->w2[1] = 0;
ctx->w2[2] = 0;
ctx->w2[3] = 0;
ctx->w3[0] = 0;
ctx->w3[1] = 0;
ctx->w3[2] = 0;
ctx->w3[3] = 0;
}
ctx->w3[2] = 0;
ctx->w3[3] = ctx->len * 8;
sha1_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
}
void sha1_optimize_max_length_vector (sha1_ctx_vector_t *ctx, const int bits)
{
ctx->len &= (1 << bits) - 1;
}

537
OpenCL/m02100.cl
View File

@ -12,419 +12,11 @@
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md4.cl"
#include "inc_hash_sha1.cl"
#define COMPARE_S "inc_comp_single.cl"
#define COMPARE_M "inc_comp_multi.cl"
void md4_transform_S (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
{
u32 a = digest[0];
u32 b = digest[1];
u32 c = digest[2];
u32 d = digest[3];
MD4_STEP_S (MD4_Fo, a, b, c, d, w0[0], MD4C00, MD4S00);
MD4_STEP_S (MD4_Fo, d, a, b, c, w0[1], MD4C00, MD4S01);
MD4_STEP_S (MD4_Fo, c, d, a, b, w0[2], MD4C00, MD4S02);
MD4_STEP_S (MD4_Fo, b, c, d, a, w0[3], MD4C00, MD4S03);
MD4_STEP_S (MD4_Fo, a, b, c, d, w1[0], MD4C00, MD4S00);
MD4_STEP_S (MD4_Fo, d, a, b, c, w1[1], MD4C00, MD4S01);
MD4_STEP_S (MD4_Fo, c, d, a, b, w1[2], MD4C00, MD4S02);
MD4_STEP_S (MD4_Fo, b, c, d, a, w1[3], MD4C00, MD4S03);
MD4_STEP_S (MD4_Fo, a, b, c, d, w2[0], MD4C00, MD4S00);
MD4_STEP_S (MD4_Fo, d, a, b, c, w2[1], MD4C00, MD4S01);
MD4_STEP_S (MD4_Fo, c, d, a, b, w2[2], MD4C00, MD4S02);
MD4_STEP_S (MD4_Fo, b, c, d, a, w2[3], MD4C00, MD4S03);
MD4_STEP_S (MD4_Fo, a, b, c, d, w3[0], MD4C00, MD4S00);
MD4_STEP_S (MD4_Fo, d, a, b, c, w3[1], MD4C00, MD4S01);
MD4_STEP_S (MD4_Fo, c, d, a, b, w3[2], MD4C00, MD4S02);
MD4_STEP_S (MD4_Fo, b, c, d, a, w3[3], MD4C00, MD4S03);
MD4_STEP_S (MD4_Go, a, b, c, d, w0[0], MD4C01, MD4S10);
MD4_STEP_S (MD4_Go, d, a, b, c, w1[0], MD4C01, MD4S11);
MD4_STEP_S (MD4_Go, c, d, a, b, w2[0], MD4C01, MD4S12);
MD4_STEP_S (MD4_Go, b, c, d, a, w3[0], MD4C01, MD4S13);
MD4_STEP_S (MD4_Go, a, b, c, d, w0[1], MD4C01, MD4S10);
MD4_STEP_S (MD4_Go, d, a, b, c, w1[1], MD4C01, MD4S11);
MD4_STEP_S (MD4_Go, c, d, a, b, w2[1], MD4C01, MD4S12);
MD4_STEP_S (MD4_Go, b, c, d, a, w3[1], MD4C01, MD4S13);
MD4_STEP_S (MD4_Go, a, b, c, d, w0[2], MD4C01, MD4S10);
MD4_STEP_S (MD4_Go, d, a, b, c, w1[2], MD4C01, MD4S11);
MD4_STEP_S (MD4_Go, c, d, a, b, w2[2], MD4C01, MD4S12);
MD4_STEP_S (MD4_Go, b, c, d, a, w3[2], MD4C01, MD4S13);
MD4_STEP_S (MD4_Go, a, b, c, d, w0[3], MD4C01, MD4S10);
MD4_STEP_S (MD4_Go, d, a, b, c, w1[3], MD4C01, MD4S11);
MD4_STEP_S (MD4_Go, c, d, a, b, w2[3], MD4C01, MD4S12);
MD4_STEP_S (MD4_Go, b, c, d, a, w3[3], MD4C01, MD4S13);
MD4_STEP_S (MD4_H , a, b, c, d, w0[0], MD4C02, MD4S20);
MD4_STEP_S (MD4_H , d, a, b, c, w2[0], MD4C02, MD4S21);
MD4_STEP_S (MD4_H , c, d, a, b, w1[0], MD4C02, MD4S22);
MD4_STEP_S (MD4_H , b, c, d, a, w3[0], MD4C02, MD4S23);
MD4_STEP_S (MD4_H , a, b, c, d, w0[2], MD4C02, MD4S20);
MD4_STEP_S (MD4_H , d, a, b, c, w2[2], MD4C02, MD4S21);
MD4_STEP_S (MD4_H , c, d, a, b, w1[2], MD4C02, MD4S22);
MD4_STEP_S (MD4_H , b, c, d, a, w3[2], MD4C02, MD4S23);
MD4_STEP_S (MD4_H , a, b, c, d, w0[1], MD4C02, MD4S20);
MD4_STEP_S (MD4_H , d, a, b, c, w2[1], MD4C02, MD4S21);
MD4_STEP_S (MD4_H , c, d, a, b, w1[1], MD4C02, MD4S22);
MD4_STEP_S (MD4_H , b, c, d, a, w3[1], MD4C02, MD4S23);
MD4_STEP_S (MD4_H , a, b, c, d, w0[3], MD4C02, MD4S20);
MD4_STEP_S (MD4_H , d, a, b, c, w2[3], MD4C02, MD4S21);
MD4_STEP_S (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22);
MD4_STEP_S (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
void sha1_transform_S (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 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];
#undef K
#define K SHA1C00
SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, w0_t);
SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w1_t);
SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w2_t);
SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w3_t);
SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w4_t);
SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, w5_t);
SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w6_t);
SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w7_t);
SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w8_t);
SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w9_t);
SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, wa_t);
SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, wb_t);
SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, wc_t);
SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, wd_t);
SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, we_t);
SHA1_STEP_S (SHA1_F0o, A, B, C, D, E, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F0o, E, A, B, C, D, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F0o, D, E, A, B, C, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F0o, C, D, E, A, B, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F0o, B, C, D, E, A, w3_t);
#undef K
#define K SHA1C01
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w7_t);
#undef K
#define K SHA1C02
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F2o, A, B, C, D, E, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F2o, E, A, B, C, D, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F2o, D, E, A, B, C, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F2o, C, D, E, A, B, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F2o, B, C, D, E, A, wb_t);
#undef K
#define K SHA1C03
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, wf_t);
w0_t = rotl32_S ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w0_t);
w1_t = rotl32_S ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w1_t);
w2_t = rotl32_S ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w2_t);
w3_t = rotl32_S ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w3_t);
w4_t = rotl32_S ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w4_t);
w5_t = rotl32_S ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, w5_t);
w6_t = rotl32_S ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, w6_t);
w7_t = rotl32_S ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, w7_t);
w8_t = rotl32_S ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, w8_t);
w9_t = rotl32_S ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, w9_t);
wa_t = rotl32_S ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wa_t);
wb_t = rotl32_S ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP_S (SHA1_F1, A, B, C, D, E, wb_t);
wc_t = rotl32_S ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP_S (SHA1_F1, E, A, B, C, D, wc_t);
wd_t = rotl32_S ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP_S (SHA1_F1, D, E, A, B, C, wd_t);
we_t = rotl32_S ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP_S (SHA1_F1, C, D, E, A, B, we_t);
wf_t = rotl32_S ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP_S (SHA1_F1, B, C, D, E, A, wf_t);
digest[0] += A;
digest[1] += B;
digest[2] += C;
digest[3] += D;
digest[4] += E;
}
void hmac_sha1_pad_S (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
{
w0[0] = w0[0] ^ 0x36363636;
w0[1] = w0[1] ^ 0x36363636;
w0[2] = w0[2] ^ 0x36363636;
w0[3] = w0[3] ^ 0x36363636;
w1[0] = w1[0] ^ 0x36363636;
w1[1] = w1[1] ^ 0x36363636;
w1[2] = w1[2] ^ 0x36363636;
w1[3] = w1[3] ^ 0x36363636;
w2[0] = w2[0] ^ 0x36363636;
w2[1] = w2[1] ^ 0x36363636;
w2[2] = w2[2] ^ 0x36363636;
w2[3] = w2[3] ^ 0x36363636;
w3[0] = w3[0] ^ 0x36363636;
w3[1] = w3[1] ^ 0x36363636;
w3[2] = w3[2] ^ 0x36363636;
w3[3] = w3[3] ^ 0x36363636;
ipad[0] = SHA1M_A;
ipad[1] = SHA1M_B;
ipad[2] = SHA1M_C;
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform_S (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
w0[2] = w0[2] ^ 0x6a6a6a6a;
w0[3] = w0[3] ^ 0x6a6a6a6a;
w1[0] = w1[0] ^ 0x6a6a6a6a;
w1[1] = w1[1] ^ 0x6a6a6a6a;
w1[2] = w1[2] ^ 0x6a6a6a6a;
w1[3] = w1[3] ^ 0x6a6a6a6a;
w2[0] = w2[0] ^ 0x6a6a6a6a;
w2[1] = w2[1] ^ 0x6a6a6a6a;
w2[2] = w2[2] ^ 0x6a6a6a6a;
w2[3] = w2[3] ^ 0x6a6a6a6a;
w3[0] = w3[0] ^ 0x6a6a6a6a;
w3[1] = w3[1] ^ 0x6a6a6a6a;
w3[2] = w3[2] ^ 0x6a6a6a6a;
w3[3] = w3[3] ^ 0x6a6a6a6a;
opad[0] = SHA1M_A;
opad[1] = SHA1M_B;
opad[2] = SHA1M_C;
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform_S (w0, w1, w2, w3, opad);
}
void hmac_sha1_run_S (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5])
{
digest[0] = ipad[0];
digest[1] = ipad[1];
digest[2] = ipad[2];
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform_S (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w1[1] = 0x80000000;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (64 + 20) * 8;
digest[0] = opad[0];
digest[1] = opad[1];
digest[2] = opad[2];
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform_S (w0, w1, w2, w3, digest);
}
void sha1_transform_V (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];
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
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
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
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
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);
digest[0] += A;
digest[1] += B;
digest[2] += C;
digest[3] += D;
digest[4] += E;
}
void hmac_sha1_run_V (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];
@ -433,7 +25,7 @@ void hmac_sha1_run_V (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform_V (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -458,7 +50,7 @@ void hmac_sha1_run_V (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform_V (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m02100_init (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
@ -471,21 +63,19 @@ __kernel void m02100_init (__global pw_t *pws, __global const kernel_rule_t *rul
if (gid >= gid_max) return;
md4_ctx_scalar_t md4_ctx1;
// dcc1
md4_init_scalar (&md4_ctx1);
md4_ctx_t md4_ctx1;
md4_update_scalar_global_utf16le (&md4_ctx1, pws[gid].i, pws[gid].pw_len);
md4_init (&md4_ctx1);
md4_final_scalar (&md4_ctx1);
md4_update_global_utf16le (&md4_ctx1, pws[gid].i, pws[gid].pw_len);
/**
* salt
*/
md4_final (&md4_ctx1);
md4_ctx_scalar_t md4_ctx2;
md4_ctx_t md4_ctx2;
md4_init_scalar (&md4_ctx2);
md4_init (&md4_ctx2);
md4_ctx2.w0[0] = md4_ctx1.h[0];
md4_ctx2.w0[1] = md4_ctx1.h[1];
@ -494,18 +84,16 @@ __kernel void m02100_init (__global pw_t *pws, __global const kernel_rule_t *rul
md4_ctx2.len = 16;
md4_update_scalar_global (&md4_ctx2, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len);
md4_update_global (&md4_ctx2, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len);
md4_final_scalar (&md4_ctx2);
md4_final (&md4_ctx2);
md4_ctx2.h[0] = swap32_S (md4_ctx2.h[0]);
md4_ctx2.h[1] = swap32_S (md4_ctx2.h[1]);
md4_ctx2.h[2] = swap32_S (md4_ctx2.h[2]);
md4_ctx2.h[3] = swap32_S (md4_ctx2.h[3]);
/**
* pads
*/
// dcc2
u32 w0[4];
u32 w1[4];
@ -529,78 +117,55 @@ __kernel void m02100_init (__global pw_t *pws, __global const kernel_rule_t *rul
w3[2] = 0;
w3[3] = 0;
u32 ipad[5];
u32 opad[5];
sha1_hmac_ctx_t sha1_hmac_ctx;
hmac_sha1_pad_S (w0, w1, w2, w3, ipad, opad);
sha1_hmac_init (&sha1_hmac_ctx, w0, w1, w2, w3);
tmps[gid].ipad[0] = ipad[0];
tmps[gid].ipad[1] = ipad[1];
tmps[gid].ipad[2] = ipad[2];
tmps[gid].ipad[3] = ipad[3];
tmps[gid].ipad[4] = ipad[4];
tmps[gid].ipad[0] = sha1_hmac_ctx.ipad.h[0];
tmps[gid].ipad[1] = sha1_hmac_ctx.ipad.h[1];
tmps[gid].ipad[2] = sha1_hmac_ctx.ipad.h[2];
tmps[gid].ipad[3] = sha1_hmac_ctx.ipad.h[3];
tmps[gid].ipad[4] = sha1_hmac_ctx.ipad.h[4];
tmps[gid].opad[0] = opad[0];
tmps[gid].opad[1] = opad[1];
tmps[gid].opad[2] = opad[2];
tmps[gid].opad[3] = opad[3];
tmps[gid].opad[4] = opad[4];
tmps[gid].opad[0] = sha1_hmac_ctx.opad.h[0];
tmps[gid].opad[1] = sha1_hmac_ctx.opad.h[1];
tmps[gid].opad[2] = sha1_hmac_ctx.opad.h[2];
tmps[gid].opad[3] = sha1_hmac_ctx.opad.h[3];
tmps[gid].opad[4] = sha1_hmac_ctx.opad.h[4];
/**
* hmac1
*/
sha1_hmac_update_global_swap (&sha1_hmac_ctx, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
w0[0] = salt_bufs[salt_pos].salt_buf[ 0];
w0[1] = salt_bufs[salt_pos].salt_buf[ 1];
w0[2] = salt_bufs[salt_pos].salt_buf[ 2];
w0[3] = salt_bufs[salt_pos].salt_buf[ 3];
w1[0] = salt_bufs[salt_pos].salt_buf[ 4];
w1[1] = salt_bufs[salt_pos].salt_buf[ 5];
w1[2] = salt_bufs[salt_pos].salt_buf[ 6];
w1[3] = salt_bufs[salt_pos].salt_buf[ 7];
w2[0] = salt_bufs[salt_pos].salt_buf[ 8];
w2[1] = salt_bufs[salt_pos].salt_buf[ 9];
w2[2] = salt_bufs[salt_pos].salt_buf[10];
w2[3] = salt_bufs[salt_pos].salt_buf[11];
w3[0] = salt_bufs[salt_pos].salt_buf[12];
w0[0] = 1;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (64 + salt_len + 4) * 8;
w3[3] = 0;
append_0x01_4x4_S (w0, w1, w2, w3, salt_len + 3);
append_0x80_4x4_S (w0, w1, w2, w3, salt_len + 4);
sha1_hmac_update_64 (&sha1_hmac_ctx, w0, w1, w2, w3, 4);
w0[0] = swap32_S (w0[0]);
w0[1] = swap32_S (w0[1]);
w0[2] = swap32_S (w0[2]);
w0[3] = swap32_S (w0[3]);
w1[0] = swap32_S (w1[0]);
w1[1] = swap32_S (w1[1]);
w1[2] = swap32_S (w1[2]);
w1[3] = swap32_S (w1[3]);
w2[0] = swap32_S (w2[0]);
w2[1] = swap32_S (w2[1]);
w2[2] = swap32_S (w2[2]);
w2[3] = swap32_S (w2[3]);
w3[0] = swap32_S (w3[0]);
w3[1] = swap32_S (w3[1]);
sha1_hmac_final (&sha1_hmac_ctx);
u32 digest[5];
tmps[gid].dgst[0] = sha1_hmac_ctx.opad.h[0];
tmps[gid].dgst[1] = sha1_hmac_ctx.opad.h[1];
tmps[gid].dgst[2] = sha1_hmac_ctx.opad.h[2];
tmps[gid].dgst[3] = sha1_hmac_ctx.opad.h[3];
tmps[gid].dgst[4] = sha1_hmac_ctx.opad.h[4];
hmac_sha1_run_S (w0, w1, w2, w3, ipad, opad, digest);
tmps[gid].dgst[0] = digest[0];
tmps[gid].dgst[1] = digest[1];
tmps[gid].dgst[2] = digest[2];
tmps[gid].dgst[3] = digest[3];
tmps[gid].dgst[4] = digest[4];
tmps[gid].out[0] = digest[0];
tmps[gid].out[1] = digest[1];
tmps[gid].out[2] = digest[2];
tmps[gid].out[3] = digest[3];
tmps[gid].out[0] = tmps[gid].dgst[0];
tmps[gid].out[1] = tmps[gid].dgst[1];
tmps[gid].out[2] = tmps[gid].dgst[2];
tmps[gid].out[3] = tmps[gid].dgst[3];
}
__kernel void m02100_loop (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const comb_t *combs_buf, __global const bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global const u32 *bitmaps_buf_s1_a, __global const u32 *bitmaps_buf_s1_b, __global const u32 *bitmaps_buf_s1_c, __global const u32 *bitmaps_buf_s1_d, __global const u32 *bitmaps_buf_s2_a, __global const u32 *bitmaps_buf_s2_b, __global const u32 *bitmaps_buf_s2_c, __global const u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global const digest_t *digests_buf, __global u32 *hashes_shown, __global const salt_t *salt_bufs, __global const void *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)