arno
/
hashcat
mirror of https://github.com/hashcat/hashcat.git
1
0
Fork 0
Code Issues Releases Wiki Activity

Mix in pure kernel functions in various optimized kernels

Browse Source
pull/1315/head
jsteube 7 years ago
parent 4b2f3011e9
commit 560a786ea7
26 changed files with 459 additions and 3097 deletions
Show Stats Download Patch File Download Diff File

107
OpenCL/m00050_a0-optimized.cl
Unescape View File

@ -13,104 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -136,7 +39,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -160,7 +63,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -170,7 +73,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -194,7 +97,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00050_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

107
OpenCL/m00050_a1-optimized.cl
Unescape View File

@ -11,104 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -134,7 +37,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -158,7 +61,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -168,7 +71,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -192,7 +95,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00050_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

107
OpenCL/m00050_a3-optimized.cl
Unescape View File

@ -11,104 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -134,7 +37,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -158,7 +61,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -168,7 +71,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -192,7 +95,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
void m00050m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

107
OpenCL/m00060_a0-optimized.cl
Unescape View File

@ -13,104 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -136,7 +39,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -160,7 +63,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -170,7 +73,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -194,7 +97,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00060_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

107
OpenCL/m00060_a1-optimized.cl
Unescape View File

@ -11,104 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -134,7 +37,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -158,7 +61,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -168,7 +71,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -192,7 +95,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00060_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

107
OpenCL/m00060_a3-optimized.cl
Unescape View File

@ -11,104 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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];
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
#include "inc_hash_md5.cl"
void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
{
@ -134,7 +37,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
ipad[2] = MD5M_C;
ipad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, ipad);
md5_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -158,7 +61,7 @@ void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
opad[2] = MD5M_C;
opad[3] = MD5M_D;
md5_transform (w0, w1, w2, w3, opad);
md5_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
@ -168,7 +71,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = ipad[2];
digest[3] = ipad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -192,7 +95,7 @@ void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4],
digest[2] = opad[2];
digest[3] = opad[3];
md5_transform (w0, w1, w2, w3, digest);
md5_transform_vector (w0, w1, w2, w3, digest);
}
void m00060m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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 *esal_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)

137
OpenCL/m00150_a0-optimized.cl
Unescape View File

@ -13,134 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -167,7 +40,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -192,7 +65,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -203,7 +76,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -228,7 +101,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00150_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

137
OpenCL/m00150_a1-optimized.cl
Unescape View File

@ -11,134 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -165,7 +38,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -190,7 +63,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -201,7 +74,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -226,7 +99,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00150_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

137
OpenCL/m00150_a3-optimized.cl
Unescape View File

@ -11,134 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -165,7 +38,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -190,7 +63,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -201,7 +74,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -226,7 +99,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
void m00150m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

137
OpenCL/m00160_a0-optimized.cl
Unescape View File

@ -13,134 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -167,7 +40,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -192,7 +65,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -203,7 +76,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -228,7 +101,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00160_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

137
OpenCL/m00160_a1-optimized.cl
Unescape View File

@ -11,134 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -165,7 +38,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -190,7 +63,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -201,7 +74,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -226,7 +99,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m00160_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

137
OpenCL/m00160_a3-optimized.cl
Unescape View File

@ -11,134 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
{
u32x A = digest[0];
u32x B = digest[1];
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;
}
#include "inc_hash_sha1.cl"
void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
{
@ -165,7 +38,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
ipad[3] = SHA1M_D;
ipad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, ipad);
sha1_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -190,7 +63,7 @@ void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
opad[3] = SHA1M_D;
opad[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, opad);
sha1_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
@ -201,7 +74,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = ipad[3];
digest[4] = ipad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -226,7 +99,7 @@ void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5]
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
void m00160m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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 *esal_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)

201
OpenCL/m00400-optimized.cl
Unescape View File

@ -11,206 +11,11 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#define COMPARE_S "inc_comp_single.cl"
#define COMPARE_M "inc_comp_multi.cl"
void md5_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];
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 = 0;
MD5_STEP_S (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP_S (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP_S (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP_S (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP_S (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP_S (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP_S (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP_S (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP_S (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP_S (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP_S (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP_S (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP_S (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP_S (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP_S (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP_S (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP_S (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP_S (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP_S (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP_S (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP_S (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP_S (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP_S (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP_S (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP_S (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP_S (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP_S (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP_S (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP_S (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP_S (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP_S (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP_S (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP_S (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP_S (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP_S (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP_S (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP_S (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP_S (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP_S (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP_S (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP_S (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP_S (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP_S (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP_S (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP_S (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP_S (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP_S (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP_S (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP_S (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP_S (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP_S (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP_S (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP_S (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP_S (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP_S (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP_S (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP_S (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP_S (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP_S (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP_S (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP_S (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP_S (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP_S (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP_S (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
void md5_transform_V (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
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 = 0;
MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
}
__kernel void m00400_init (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global phpass_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)
{
/**
@ -300,7 +105,7 @@ __kernel void m00400_init (__global pw_t *pws, __global const kernel_rule_t *rul
digest[2] = MD5M_C;
digest[3] = MD5M_D;
md5_transform_S (block0, block1, block2, block3, digest);
md5_transform (block0, block1, block2, block3, digest);
tmps[gid].digest_buf[0] = digest[0];
tmps[gid].digest_buf[1] = digest[1];
@ -390,7 +195,7 @@ __kernel void m00400_loop (__global pw_t *pws, __global const kernel_rule_t *rul
digest[2] = MD5M_C;
digest[3] = MD5M_D;
md5_transform_V (block0, block1, block2, block3, digest);
md5_transform_vector (block0, block1, block2, block3, digest);
}
unpackv (tmps, digest_buf, gid, 0, digest[0]);

117
OpenCL/m01450_a0-optimized.cl
Unescape View File

@ -13,114 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -150,7 +43,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -178,7 +71,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -192,7 +85,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -220,7 +113,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m01450_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

117
OpenCL/m01450_a1-optimized.cl
Unescape View File

@ -11,114 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -148,7 +41,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -176,7 +69,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -190,7 +83,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -218,7 +111,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m01450_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

117
OpenCL/m01450_a3-optimized.cl
Unescape View File

@ -11,114 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -148,7 +41,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -176,7 +69,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -190,7 +83,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -218,7 +111,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
void m01450m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

117
OpenCL/m01460_a0-optimized.cl
Unescape View File

@ -13,114 +13,7 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -150,7 +43,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -178,7 +71,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -192,7 +85,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -220,7 +113,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m01460_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

117
OpenCL/m01460_a1-optimized.cl
Unescape View File

@ -11,114 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -148,7 +41,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -176,7 +69,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -190,7 +83,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -218,7 +111,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
__kernel void m01460_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

117
OpenCL/m01460_a3-optimized.cl
Unescape View File

@ -11,114 +11,7 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
__constant u32a k_sha256[64] =
{
SHA256C00, SHA256C01, SHA256C02, SHA256C03,
SHA256C04, SHA256C05, SHA256C06, SHA256C07,
SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
SHA256C10, SHA256C11, SHA256C12, SHA256C13,
SHA256C14, SHA256C15, SHA256C16, SHA256C17,
SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
SHA256C20, SHA256C21, SHA256C22, SHA256C23,
SHA256C24, SHA256C25, SHA256C26, SHA256C27,
SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
SHA256C30, SHA256C31, SHA256C32, SHA256C33,
SHA256C34, SHA256C35, SHA256C36, SHA256C37,
SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
};
void sha256_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[8])
{
u32x a = digest[0];
u32x b = digest[1];
u32x c = digest[2];
u32x d = digest[3];
u32x e = digest[4];
u32x f = digest[5];
u32x g = digest[6];
u32x h = digest[7];
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];
#define ROUND_EXPAND() \
{ \
w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 64; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
}
#include "inc_hash_sha256.cl"
void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8])
{
@ -148,7 +41,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
ipad[6] = SHA256M_G;
ipad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, ipad);
sha256_transform_vector (w0, w1, w2, w3, ipad);
w0[0] = w0[0] ^ 0x6a6a6a6a;
w0[1] = w0[1] ^ 0x6a6a6a6a;
@ -176,7 +69,7 @@ void hmac_sha256_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
opad[6] = SHA256M_G;
opad[7] = SHA256M_H;
sha256_transform (w0, w1, w2, w3, opad);
sha256_transform_vector (w0, w1, w2, w3, opad);
}
void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[8], u32x opad[8], u32x digest[8])
@ -190,7 +83,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
w0[0] = digest[0];
w0[1] = digest[1];
@ -218,7 +111,7 @@ void hmac_sha256_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha256_transform (w0, w1, w2, w3, digest);
sha256_transform_vector (w0, w1, w2, w3, digest);
}
void m01460m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01750_a0-optimized.cl
Unescape View File

@ -13,117 +13,53 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -159,7 +95,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@ -187,7 +123,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -223,7 +159,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -251,7 +187,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
__kernel void m01750_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01750_a1-optimized.cl
Unescape View File

@ -11,117 +11,53 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -157,7 +93,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@ -185,7 +121,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -221,7 +157,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -249,7 +185,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
__kernel void m01750_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01750_a3-optimized.cl
Unescape View File

@ -11,117 +11,53 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -157,7 +93,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ 0x5c5c5c5c5c5c5c5c;
@ -185,7 +121,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -221,7 +157,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -249,7 +185,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
void m01750m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01760_a0-optimized.cl
Unescape View File

@ -13,117 +13,53 @@
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -159,7 +95,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@ -187,7 +123,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -223,7 +159,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -251,7 +187,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
__kernel void m01760_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01760_a1-optimized.cl
Unescape View File

@ -11,117 +11,53 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -157,7 +93,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@ -185,7 +121,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -221,7 +157,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -249,7 +185,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
__kernel void m01760_m04 (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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)

162
OpenCL/m01760_a3-optimized.cl
Unescape View File

@ -11,117 +11,53 @@
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_sha512.cl"
__constant u64a k_sha512[80] =
void sha512_transform_transport_vector (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
{
u64x w0_t = w0[0];
u64x w1_t = w0[1];
u64x w2_t = w0[2];
u64x w3_t = w0[3];
u64x w4_t = w1[0];
u64x w5_t = w1[1];
u64x w6_t = w1[2];
u64x w7_t = w1[3];
u64x w8_t = w2[0];
u64x w9_t = w2[1];
u64x wa_t = w2[2];
u64x wb_t = w2[3];
u64x wc_t = w3[0];
u64x wd_t = w3[1];
u64x we_t = w3[2];
u64x wf_t = w3[3];
u64x a = digest[0];
u64x b = digest[1];
u64x c = digest[2];
u64x d = digest[3];
u64x e = digest[4];
u64x f = digest[5];
u64x g = digest[6];
u64x h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32x t0[4];
u32x t1[4];
u32x t2[4];
u32x t3[4];
u32x t4[4];
u32x t5[4];
u32x t6[4];
u32x t7[4];
t0[0] = h32_from_64 (w0[0]);
t0[1] = l32_from_64 (w0[0]);
t0[2] = h32_from_64 (w0[1]);
t0[3] = l32_from_64 (w0[1]);
t1[0] = h32_from_64 (w0[2]);
t1[1] = l32_from_64 (w0[2]);
t1[2] = h32_from_64 (w0[3]);
t1[3] = l32_from_64 (w0[3]);
t2[0] = h32_from_64 (w1[0]);
t2[1] = l32_from_64 (w1[0]);
t2[2] = h32_from_64 (w1[1]);
t2[3] = l32_from_64 (w1[1]);
t3[0] = h32_from_64 (w1[2]);
t3[1] = l32_from_64 (w1[2]);
t3[2] = h32_from_64 (w1[3]);
t3[3] = l32_from_64 (w1[3]);
t4[0] = h32_from_64 (w2[0]);
t4[1] = l32_from_64 (w2[0]);
t4[2] = h32_from_64 (w2[1]);
t4[3] = l32_from_64 (w2[1]);
t5[0] = h32_from_64 (w2[2]);
t5[1] = l32_from_64 (w2[2]);
t5[2] = h32_from_64 (w2[3]);
t5[3] = l32_from_64 (w2[3]);
t6[0] = h32_from_64 (w3[0]);
t6[1] = l32_from_64 (w3[0]);
t6[2] = h32_from_64 (w3[1]);
t6[3] = l32_from_64 (w3[1]);
t7[0] = h32_from_64 (w3[2]);
t7[1] = l32_from_64 (w3[2]);
t7[2] = h32_from_64 (w3[3]);
t7[3] = l32_from_64 (w3[3]);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
@ -157,7 +93,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
ipad[6] = SHA512M_G;
ipad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, ipad);
w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
@ -185,7 +121,7 @@ void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
opad[6] = SHA512M_G;
opad[7] = SHA512M_H;
sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, opad);
}
void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
@ -221,7 +157,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = ipad[6];
digest[7] = ipad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
@ -249,7 +185,7 @@ void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[
digest[6] = opad[6];
digest[7] = opad[7];
sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha512_transform_transport_vector (w0_t, w1_t, w2_t, w3_t, digest);
}
void m01760m (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const u32 pw_len, __global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global void *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 *esal_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)

217
OpenCL/m01800-optimized.cl
Unescape View File

@ -8,19 +8,11 @@
#include "inc_hash_functions.cl"
#include "inc_types.cl"
#include "inc_common.cl"
#include "inc_hash_sha512.cl"
#define COMPARE_S "inc_comp_single.cl"
#define COMPARE_M "inc_comp_multi.cl"
// Buggy drivers...
#ifdef IS_AMD
#define STATE_DECL volatile
//#define STATE_DECL
#else
#define STATE_DECL
#endif
#define PUTCHAR64_BE(a,p,c) ((u8 *)(a))[(p) ^ 7] = (u8) (c)
#define GETCHAR64_BE(a,p) ((u8 *)(a))[(p) ^ 7]
@ -30,121 +22,56 @@ typedef struct
u64 buf[16];
int len;
} sha512_ctx_t;
} orig_sha512_ctx_t;
__constant u64a k_sha512[80] =
{
SHA512C00, SHA512C01, SHA512C02, SHA512C03,
SHA512C04, SHA512C05, SHA512C06, SHA512C07,
SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
SHA512C10, SHA512C11, SHA512C12, SHA512C13,
SHA512C14, SHA512C15, SHA512C16, SHA512C17,
SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
SHA512C20, SHA512C21, SHA512C22, SHA512C23,
SHA512C24, SHA512C25, SHA512C26, SHA512C27,
SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
SHA512C30, SHA512C31, SHA512C32, SHA512C33,
SHA512C34, SHA512C35, SHA512C36, SHA512C37,
SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
SHA512C40, SHA512C41, SHA512C42, SHA512C43,
SHA512C44, SHA512C45, SHA512C46, SHA512C47,
SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
};
void sha512_transform (const u64 *w, u64 *digest)
void sha512_transform_transport (const u64 *w, u64 *digest)
{
u64 w0_t = w[ 0];
u64 w1_t = w[ 1];
u64 w2_t = w[ 2];
u64 w3_t = w[ 3];
u64 w4_t = w[ 4];
u64 w5_t = w[ 5];
u64 w6_t = w[ 6];
u64 w7_t = w[ 7];
u64 w8_t = w[ 8];
u64 w9_t = w[ 9];
u64 wa_t = w[10];
u64 wb_t = w[11];
u64 wc_t = w[12];
u64 wd_t = w[13];
u64 we_t = w[14];
u64 wf_t = w[15];
STATE_DECL u64 a = digest[0];
STATE_DECL u64 b = digest[1];
STATE_DECL u64 c = digest[2];
STATE_DECL u64 d = digest[3];
STATE_DECL u64 e = digest[4];
STATE_DECL u64 f = digest[5];
STATE_DECL u64 g = digest[6];
STATE_DECL u64 h = digest[7];
#define ROUND_EXPAND() \
{ \
w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
}
#define ROUND_STEP(i) \
{ \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
}
ROUND_STEP (0);
#ifdef _unroll
#pragma unroll
#endif
for (int i = 16; i < 80; i += 16)
{
ROUND_EXPAND (); ROUND_STEP (i);
}
digest[0] += a;
digest[1] += b;
digest[2] += c;
digest[3] += d;
digest[4] += e;
digest[5] += f;
digest[6] += g;
digest[7] += h;
u32 t0[4];
u32 t1[4];
u32 t2[4];
u32 t3[4];
u32 t4[4];
u32 t5[4];
u32 t6[4];
u32 t7[4];
t0[0] = h32_from_64_S (w[ 0]);
t0[1] = l32_from_64_S (w[ 0]);
t0[2] = h32_from_64_S (w[ 1]);
t0[3] = l32_from_64_S (w[ 1]);
t1[0] = h32_from_64_S (w[ 2]);
t1[1] = l32_from_64_S (w[ 2]);
t1[2] = h32_from_64_S (w[ 3]);
t1[3] = l32_from_64_S (w[ 3]);
t2[0] = h32_from_64_S (w[ 4]);
t2[1] = l32_from_64_S (w[ 4]);
t2[2] = h32_from_64_S (w[ 5]);
t2[3] = l32_from_64_S (w[ 5]);
t3[0] = h32_from_64_S (w[ 6]);
t3[1] = l32_from_64_S (w[ 6]);
t3[2] = h32_from_64_S (w[ 7]);
t3[3] = l32_from_64_S (w[ 7]);
t4[0] = h32_from_64_S (w[ 8]);
t4[1] = l32_from_64_S (w[ 8]);
t4[2] = h32_from_64_S (w[ 9]);
t4[3] = l32_from_64_S (w[ 9]);
t5[0] = h32_from_64_S (w[10]);
t5[1] = l32_from_64_S (w[10]);
t5[2] = h32_from_64_S (w[11]);
t5[3] = l32_from_64_S (w[11]);
t6[0] = h32_from_64_S (w[12]);
t6[1] = l32_from_64_S (w[12]);
t6[2] = h32_from_64_S (w[13]);
t6[3] = l32_from_64_S (w[13]);
t7[0] = h32_from_64_S (w[14]);
t7[1] = l32_from_64_S (w[14]);
t7[2] = h32_from_64_S (w[15]);
t7[3] = l32_from_64_S (w[15]);
sha512_transform (t0, t1, t2, t3, t4, t5, t6, t7, digest);
}
void sha512_init (sha512_ctx_t *sha512_ctx)
void orig_sha512_init (orig_sha512_ctx_t *sha512_ctx)
{
sha512_ctx->state[0] = SHA512M_A;
sha512_ctx->state[1] = SHA512M_B;
@ -158,7 +85,7 @@ void sha512_init (sha512_ctx_t *sha512_ctx)
sha512_ctx->len = 0;
}
void sha512_update (sha512_ctx_t *sha512_ctx, const u64 *buf, int len)
void orig_sha512_update (orig_sha512_ctx_t *sha512_ctx, const u64 *buf, int len)
{
int pos = sha512_ctx->len & 0x7f;
@ -181,7 +108,7 @@ void sha512_update (sha512_ctx_t *sha512_ctx, const u64 *buf, int len)
PUTCHAR64_BE (sha512_ctx->buf, pos++, GETCHAR64_BE (buf, i));
}
sha512_transform (sha512_ctx->buf, sha512_ctx->state);
sha512_transform_transport (sha512_ctx->buf, sha512_ctx->state);
len -= cnt;
@ -191,7 +118,7 @@ void sha512_update (sha512_ctx_t *sha512_ctx, const u64 *buf, int len)
}
}
void sha512_final (sha512_ctx_t *sha512_ctx)
void orig_sha512_final (orig_sha512_ctx_t *sha512_ctx)
{
int pos = sha512_ctx->len & 0x7f;
@ -204,7 +131,7 @@ void sha512_final (sha512_ctx_t *sha512_ctx)
if (pos >= 112)
{
sha512_transform (sha512_ctx->buf, sha512_ctx->state);
sha512_transform_transport (sha512_ctx->buf, sha512_ctx->state);
sha512_ctx->buf[ 0] = 0;
sha512_ctx->buf[ 1] = 0;
@ -226,7 +153,7 @@ void sha512_final (sha512_ctx_t *sha512_ctx)
sha512_ctx->buf[15] = sha512_ctx->len * 8;
sha512_transform (sha512_ctx->buf, sha512_ctx->state);
sha512_transform_transport (sha512_ctx->buf, sha512_ctx->state);
}
__kernel void m01800_init (__global pw_t *pws, __global const kernel_rule_t *rules_buf, __global const pw_t *combs_buf, __global const bf_t *bfs_buf, __global sha512crypt_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)
@ -279,15 +206,15 @@ __kernel void m01800_init (__global pw_t *pws, __global const kernel_rule_t *rul
* begin
*/
sha512_ctx_t sha512_ctx;
orig_sha512_ctx_t sha512_ctx;
sha512_init (&sha512_ctx);
orig_sha512_init (&sha512_ctx);
sha512_update (&sha512_ctx, pw, pw_len);
sha512_update (&sha512_ctx, salt, salt_len);
sha512_update (&sha512_ctx, pw, pw_len);
orig_sha512_update (&sha512_ctx, pw, pw_len);
orig_sha512_update (&sha512_ctx, salt, salt_len);
orig_sha512_update (&sha512_ctx, pw, pw_len);
sha512_final (&sha512_ctx);
orig_sha512_final (&sha512_ctx);
u64 tmp[8];
@ -300,25 +227,25 @@ __kernel void m01800_init (__global pw_t *pws, __global const kernel_rule_t *rul
tmp[6] = sha512_ctx.state[6];
tmp[7] = sha512_ctx.state[7];
sha512_init (&sha512_ctx);
orig_sha512_init (&sha512_ctx);
sha512_update (&sha512_ctx, pw, pw_len);
sha512_update (&sha512_ctx, salt, salt_len);
sha512_update (&sha512_ctx, tmp, pw_len);
orig_sha512_update (&sha512_ctx, pw, pw_len);
orig_sha512_update (&sha512_ctx, salt, salt_len);
orig_sha512_update (&sha512_ctx, tmp, pw_len);
for (u32 j = pw_len; j; j >>= 1)
{
if (j & 1)
{
sha512_update (&sha512_ctx, tmp, 64);
orig_sha512_update (&sha512_ctx, tmp, 64);
}
else
{
sha512_update (&sha512_ctx, pw, pw_len);
orig_sha512_update (&sha512_ctx, pw, pw_len);
}
}
sha512_final (&sha512_ctx);
orig_sha512_final (&sha512_ctx);
tmps[gid].l_alt_result[0] = sha512_ctx.state[0];
tmps[gid].l_alt_result[1] = sha512_ctx.state[1];
@ -331,28 +258,28 @@ __kernel void m01800_init (__global pw_t *pws, __global const kernel_rule_t *rul
// p_bytes
sha512_init (&sha512_ctx);
orig_sha512_init (&sha512_ctx);
for (u32 j = 0; j < pw_len; j++)
{
sha512_update (&sha512_ctx, pw, pw_len);
orig_sha512_update (&sha512_ctx, pw, pw_len);
}
sha512_final (&sha512_ctx);
orig_sha512_final (&sha512_ctx);
tmps[gid].l_p_bytes[0] = sha512_ctx.state[0];
tmps[gid].l_p_bytes[1] = sha512_ctx.state[1];
// s_bytes
sha512_init (&sha512_ctx);
orig_sha512_init (&sha512_ctx);
for (u32 j = 0; j < 16 + ((tmps[gid].l_alt_result[0] >> 56) & 0xff); j++)
{
sha512_update (&sha512_ctx, salt, salt_len);
orig_sha512_update (&sha512_ctx, salt, salt_len);
}
sha512_final (&sha512_ctx);
orig_sha512_final (&sha512_ctx);
tmps[gid].l_s_bytes[0] = sha512_ctx.state[0];
tmps[gid].l_s_bytes[1] = sha512_ctx.state[1];
@ -522,7 +449,7 @@ __kernel void m01800_loop (__global pw_t *pws, __global const kernel_rule_t *rul
l_alt_result[6] = SHA512M_G;
l_alt_result[7] = SHA512M_H;
sha512_transform (block, l_alt_result);
sha512_transform_transport (block, l_alt_result);
}
tmps[gid].l_alt_result[0] = l_alt_result[0];

Write Preview
Loading…
Cancel
Save