arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2025-01-11 00:01:16 +00:00
Code Issues Releases Wiki Activity

Optimize some -m 780x code in kernels

Browse Source
This commit is contained in:
Jens Steube 2020-02-20 11:01:56 +01:00
parent 189bbb2661
commit 1449e239c2
6 changed files with 396 additions and 348 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

124
OpenCL/m07800_a0-optimized.cl
View File

@ -191,33 +191,33 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -278,17 +278,21 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }
@ -451,33 +455,33 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -538,17 +542,21 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }

124
OpenCL/m07800_a1-optimized.cl
View File

@ -249,33 +249,33 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -336,17 +336,21 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }
@ -569,33 +573,33 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -656,17 +660,21 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }

124
OpenCL/m07800_a3-optimized.cl
View File

@ -165,33 +165,33 @@ DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -252,17 +252,21 @@ DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }
@ -393,33 +397,33 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -480,17 +484,21 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]); COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
} }

124
OpenCL/m07801_a0-optimized.cl
View File

@ -191,33 +191,33 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -278,17 +278,21 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }
@ -451,33 +455,33 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -538,17 +542,21 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }

124
OpenCL/m07801_a1-optimized.cl
View File

@ -249,33 +249,33 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -336,17 +336,21 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }
@ -569,33 +573,33 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -656,17 +660,21 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }

124
OpenCL/m07801_a3-optimized.cl
View File

@ -165,33 +165,33 @@ DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -252,17 +252,21 @@ DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }
@ -393,33 +397,33 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D; digest[3] = SHA1M_D;
digest[4] = SHA1M_E; digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest); sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range // prepare magic array range
u32 lengthMagicArray = 0x20; u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0; u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6; lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6; lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6; lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6; lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8; offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8; offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8; offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8; offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final // final
@ -480,17 +484,21 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate // calculate
int left; if (final_len >= 56)
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
{ {
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest); final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
} }
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8; sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]); COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
} }