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

Rewrite MT[X][256] constants to MTX[256] constants in whirlpool hash

Browse Source
pull/2310/head
Jens Steube 4 years ago
parent 193b4a38c7
commit 5f57ab35b6
14 changed files with 1379 additions and 1054 deletions
Show Stats Download Patch File Download Diff File

1267
OpenCL/inc_hash_whirlpool.cl
View File

File diff suppressed because it is too large Load Diff

66
OpenCL/inc_hash_whirlpool.h
Unescape View File

@ -6,27 +6,25 @@
#ifndef _INC_HASH_WHIRLPOOL_H
#define _INC_HASH_WHIRLPOOL_H
#define R 10
#if VECT_SIZE == 1
#define BOX(S,n,i) (S)[(n)][(i)]
#define BOX64(S,n,i) (S)[(n)][(i)]
#define BOX(S,i) (S)[(i)]
#define BOX64(S,i) (S)[(i)]
#elif VECT_SIZE == 2
#define BOX(S,n,i) make_u32x ((S)[(n)][(i).s0], (S)[(n)][(i).s1])
#define BOX64(S,n,i) make_u64x ((S)[(n)][(i).s0], (S)[(n)][(i).s1])
#define BOX(S,i) make_u32x ((S)[(i).s0], (S)[(i).s1])
#define BOX64(S,i) make_u64x ((S)[(i).s0], (S)[(i).s1])
#elif VECT_SIZE == 4
#define BOX(S,n,i) make_u32x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3])
#define BOX64(S,n,i) make_u64x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3])
#define BOX(S,i) make_u32x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3])
#define BOX64(S,i) make_u64x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3])
#elif VECT_SIZE == 8
#define BOX(S,n,i) make_u32x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3], (S)[(n)][(i).s4], (S)[(n)][(i).s5], (S)[(n)][(i).s6], (S)[(n)][(i).s7])
#define BOX64(S,n,i) make_u64x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3], (S)[(n)][(i).s4], (S)[(n)][(i).s5], (S)[(n)][(i).s6], (S)[(n)][(i).s7])
#define BOX(S,i) make_u32x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7])
#define BOX64(S,i) make_u64x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7])
#elif VECT_SIZE == 16
#define BOX(S,n,i) make_u32x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3], (S)[(n)][(i).s4], (S)[(n)][(i).s5], (S)[(n)][(i).s6], (S)[(n)][(i).s7], (S)[(n)][(i).s8], (S)[(n)][(i).s9], (S)[(n)][(i).sa], (S)[(n)][(i).sb], (S)[(n)][(i).sc], (S)[(n)][(i).sd], (S)[(n)][(i).se], (S)[(n)][(i).sf])
#define BOX64(S,n,i) make_u64x ((S)[(n)][(i).s0], (S)[(n)][(i).s1], (S)[(n)][(i).s2], (S)[(n)][(i).s3], (S)[(n)][(i).s4], (S)[(n)][(i).s5], (S)[(n)][(i).s6], (S)[(n)][(i).s7], (S)[(n)][(i).s8], (S)[(n)][(i).s9], (S)[(n)][(i).sa], (S)[(n)][(i).sb], (S)[(n)][(i).sc], (S)[(n)][(i).sd], (S)[(n)][(i).se], (S)[(n)][(i).sf])
#define BOX(S,i) make_u32x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7], (S)[(i).s8], (S)[(i).s9], (S)[(i).sa], (S)[(i).sb], (S)[(i).sc], (S)[(i).sd], (S)[(i).se], (S)[(i).sf])
#define BOX64(S,i) make_u64x ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7], (S)[(i).s8], (S)[(i).s9], (S)[(i).sa], (S)[(i).sb], (S)[(i).sc], (S)[(i).sd], (S)[(i).se], (S)[(i).sf])
#endif
#define BOX_S(S,n,i) (S)[(n)][(i)]
#define BOX64_S(S,n,i) (S)[(n)][(i)]
#define BOX_S(S,i) (S)[(i)]
#define BOX64_S(S,i) (S)[(i)]
typedef struct whirlpool_ctx
{
@ -39,7 +37,14 @@ typedef struct whirlpool_ctx
int len;
SHM_TYPE u64 (*s_MT)[256];
SHM_TYPE u64 *s_MT0;
SHM_TYPE u64 *s_MT1;
SHM_TYPE u64 *s_MT2;
SHM_TYPE u64 *s_MT3;
SHM_TYPE u64 *s_MT4;
SHM_TYPE u64 *s_MT5;
SHM_TYPE u64 *s_MT6;
SHM_TYPE u64 *s_MT7;
} whirlpool_ctx_t;
@ -61,7 +66,14 @@ typedef struct whirlpool_ctx_vector
int len;
SHM_TYPE u64 (*s_MT)[256];
SHM_TYPE u64 *s_MT0;
SHM_TYPE u64 *s_MT1;
SHM_TYPE u64 *s_MT2;
SHM_TYPE u64 *s_MT3;
SHM_TYPE u64 *s_MT4;
SHM_TYPE u64 *s_MT5;
SHM_TYPE u64 *s_MT6;
SHM_TYPE u64 *s_MT7;
} whirlpool_ctx_vector_t;
@ -72,8 +84,8 @@ typedef struct whirlpool_hmac_ctx_vector
} whirlpool_hmac_ctx_vector_t;
DECLSPEC void whirlpool_transform (const u32 *w0, const u32 *w1, const u32 *w2, const u32 *w3, u32 *digest, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_init (whirlpool_ctx_t *ctx, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_transform (const u32 *w0, const u32 *w1, const u32 *w2, const u32 *w3, u32 *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_init (whirlpool_ctx_t *ctx, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_update_64 (whirlpool_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const int len);
DECLSPEC void whirlpool_update (whirlpool_ctx_t *ctx, const u32 *w, const int len);
DECLSPEC void whirlpool_update_swap (whirlpool_ctx_t *ctx, const u32 *w, const int len);
@ -84,11 +96,11 @@ DECLSPEC void whirlpool_update_global_swap (whirlpool_ctx_t *ctx, GLOBAL_AS cons
DECLSPEC void whirlpool_update_global_utf16le (whirlpool_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len);
DECLSPEC void whirlpool_update_global_utf16le_swap (whirlpool_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len);
DECLSPEC void whirlpool_final (whirlpool_ctx_t *ctx);
DECLSPEC void whirlpool_hmac_init_64 (whirlpool_hmac_ctx_t *ctx, const u32 *w0, const u32 *w1, const u32 *w2, const u32 *w3, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_swap (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_global (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_global_swap (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_64 (whirlpool_hmac_ctx_t *ctx, const u32 *w0, const u32 *w1, const u32 *w2, const u32 *w3, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_init (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_init_swap (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_init_global (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_init_global_swap (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_update_64 (whirlpool_hmac_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const int len);
DECLSPEC void whirlpool_hmac_update (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len);
DECLSPEC void whirlpool_hmac_update_swap (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len);
@ -99,8 +111,8 @@ DECLSPEC void whirlpool_hmac_update_global_swap (whirlpool_hmac_ctx_t *ctx, GLOB
DECLSPEC void whirlpool_hmac_update_global_utf16le (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len);
DECLSPEC void whirlpool_hmac_update_global_utf16le_swap (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len);
DECLSPEC void whirlpool_hmac_final (whirlpool_hmac_ctx_t *ctx);
DECLSPEC void whirlpool_transform_vector (const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, u32x *digest, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_init_vector (whirlpool_ctx_vector_t *ctx, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_transform_vector (const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_init_vector (whirlpool_ctx_vector_t *ctx, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_init_vector_from_scalar (whirlpool_ctx_vector_t *ctx, whirlpool_ctx_t *ctx0);
DECLSPEC void whirlpool_update_vector_64 (whirlpool_ctx_vector_t *ctx, u32x *w0, u32x *w1, u32x *w2, u32x *w3, const int len);
DECLSPEC void whirlpool_update_vector (whirlpool_ctx_vector_t *ctx, const u32x *w, const int len);
@ -108,8 +120,8 @@ DECLSPEC void whirlpool_update_vector_swap (whirlpool_ctx_vector_t *ctx, const u
DECLSPEC void whirlpool_update_vector_utf16le (whirlpool_ctx_vector_t *ctx, const u32x *w, const int len);
DECLSPEC void whirlpool_update_vector_utf16le_swap (whirlpool_ctx_vector_t *ctx, const u32x *w, const int len);
DECLSPEC void whirlpool_final_vector (whirlpool_ctx_vector_t *ctx);
DECLSPEC void whirlpool_hmac_init_vector_64 (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_vector (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w, const int len, SHM_TYPE u64 (*s_MT)[256]);
DECLSPEC void whirlpool_hmac_init_vector_64 (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_init_vector (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w, const int len, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7);
DECLSPEC void whirlpool_hmac_update_vector_64 (whirlpool_hmac_ctx_vector_t *ctx, u32x *w0, u32x *w1, u32x *w2, u32x *w3, const int len);
DECLSPEC void whirlpool_hmac_update_vector (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w, const int len);
DECLSPEC void whirlpool_hmac_final_vector (whirlpool_hmac_ctx_vector_t *ctx);

76
OpenCL/m06100_a0-optimized.cl
Unescape View File

@ -16,9 +16,9 @@
#include "inc_hash_whirlpool.cl"
#endif
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT);
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_m04 (KERN_ATTR_RULES ())
@ -37,25 +37,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -136,7 +150,7 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_RULES ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_M_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}
@ -166,25 +180,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -277,7 +305,7 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_RULES ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_S_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}

72
OpenCL/m06100_a0-pure.cl
Unescape View File

@ -32,25 +32,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -74,7 +88,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_RULES ())
whirlpool_ctx_t ctx;
whirlpool_init (&ctx, s_MT);
whirlpool_init (&ctx, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_swap (&ctx, tmp.i, tmp.pw_len);
@ -105,25 +119,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -159,7 +187,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_RULES ())
whirlpool_ctx_t ctx;
whirlpool_init (&ctx, s_MT);
whirlpool_init (&ctx, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_swap (&ctx, tmp.i, tmp.pw_len);

76
OpenCL/m06100_a1-optimized.cl
Unescape View File

@ -14,9 +14,9 @@
#include "inc_hash_whirlpool.cl"
#endif
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT);
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
@ -35,25 +35,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -192,7 +206,7 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_M_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}
@ -222,25 +236,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -391,7 +419,7 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_S_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}

72
OpenCL/m06100_a1-pure.cl
Unescape View File

@ -30,25 +30,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -60,7 +74,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_BASIC ())
whirlpool_ctx_t ctx0;
whirlpool_init (&ctx0, s_MT);
whirlpool_init (&ctx0, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len);
@ -101,25 +115,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -143,7 +171,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_BASIC ())
whirlpool_ctx_t ctx0;
whirlpool_init (&ctx0, s_MT);
whirlpool_init (&ctx0, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_global_swap (&ctx0, pws[gid].i, pws[gid].pw_len);

156
OpenCL/m06100_a3-optimized.cl
Unescape View File

@ -14,12 +14,12 @@
#include "inc_hash_whirlpool.cl"
#endif
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void whirlpool_transform_transport_vector (const u32x *w, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT);
whirlpool_transform_vector (w + 0, w + 4, w + 8, w + 12, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
DECLSPEC void m06100m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void m06100m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
/**
* modifier
@ -82,13 +82,13 @@ DECLSPEC void m06100m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_M_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}
}
DECLSPEC void m06100s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void m06100s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC (), SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
/**
* modifier
@ -163,7 +163,7 @@ DECLSPEC void m06100s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_MT);
whirlpool_transform_transport_vector (w, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
COMPARE_S_SIMD (dgst[0], dgst[1], dgst[2], dgst[3]);
}
@ -185,25 +185,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -247,7 +261,7 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
* main
*/
m06100m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT);
m06100m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_m08 (KERN_ATTR_BASIC ())
@ -266,25 +280,39 @@ KERNEL_FQ void m06100_m08 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -328,7 +356,7 @@ KERNEL_FQ void m06100_m08 (KERN_ATTR_BASIC ())
* main
*/
m06100m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT);
m06100m (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_m16 (KERN_ATTR_BASIC ())
@ -351,25 +379,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -413,7 +455,7 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
* main
*/
m06100s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT);
m06100s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_s08 (KERN_ATTR_BASIC ())
@ -432,25 +474,39 @@ KERNEL_FQ void m06100_s08 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -494,7 +550,7 @@ KERNEL_FQ void m06100_s08 (KERN_ATTR_BASIC ())
* main
*/
m06100s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT);
m06100s (w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, il_cnt, digests_cnt, digests_offset, combs_mode, gid_max, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06100_s16 (KERN_ATTR_BASIC ())

72
OpenCL/m06100_a3-pure.cl
Unescape View File

@ -30,25 +30,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_VECTOR ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -83,7 +97,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_VECTOR ())
whirlpool_ctx_vector_t ctx;
whirlpool_init_vector (&ctx, s_MT);
whirlpool_init_vector (&ctx, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_vector (&ctx, w, pw_len);
@ -114,25 +128,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_VECTOR ())
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -179,7 +207,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_VECTOR ())
whirlpool_ctx_vector_t ctx;
whirlpool_init_vector (&ctx, s_MT);
whirlpool_init_vector (&ctx, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
whirlpool_update_vector (&ctx, w, pw_len);

82
OpenCL/m06231-pure.cl
Unescape View File

@ -45,7 +45,7 @@ typedef struct tc_tmp
} tc_tmp_t;
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -64,7 +64,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -83,7 +83,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -119,7 +119,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -138,7 +138,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06231_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
@ -168,25 +168,39 @@ KERNEL_FQ void m06231_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -258,7 +272,7 @@ KERNEL_FQ void m06231_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -369,25 +383,39 @@ KERNEL_FQ void m06231_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -493,7 +521,7 @@ KERNEL_FQ void m06231_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];

82
OpenCL/m06232-pure.cl
Unescape View File

@ -45,7 +45,7 @@ typedef struct tc_tmp
} tc_tmp_t;
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -64,7 +64,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -83,7 +83,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -119,7 +119,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -138,7 +138,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06232_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
@ -168,25 +168,39 @@ KERNEL_FQ void m06232_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -258,7 +272,7 @@ KERNEL_FQ void m06232_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -369,25 +383,39 @@ KERNEL_FQ void m06232_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -493,7 +521,7 @@ KERNEL_FQ void m06232_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];

82
OpenCL/m06233-pure.cl
Unescape View File

@ -45,7 +45,7 @@ typedef struct tc_tmp
} tc_tmp_t;
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -64,7 +64,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -83,7 +83,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -119,7 +119,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -138,7 +138,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m06233_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
@ -168,25 +168,39 @@ KERNEL_FQ void m06233_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -258,7 +272,7 @@ KERNEL_FQ void m06233_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -369,25 +383,39 @@ KERNEL_FQ void m06233_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -493,7 +521,7 @@ KERNEL_FQ void m06233_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];

110
OpenCL/m13731-pure.cl
Unescape View File

@ -86,7 +86,7 @@ DECLSPEC int check_header_0512 (GLOBAL_AS const vc_t *esalt_bufs, GLOBAL_AS u32
return -1;
}
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -105,7 +105,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -124,7 +124,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -160,7 +160,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -179,7 +179,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m13731_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
@ -209,25 +209,39 @@ KERNEL_FQ void m13731_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -299,7 +313,7 @@ KERNEL_FQ void m13731_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -461,25 +475,39 @@ KERNEL_FQ void m13731_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -616,7 +644,7 @@ KERNEL_FQ void m13731_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];
@ -755,34 +783,6 @@ KERNEL_FQ void m13731_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
#endif
if (gid >= gid_max) return;
if (tmps[gid].pim)

110
OpenCL/m13732-pure.cl
Unescape View File

@ -137,7 +137,7 @@ DECLSPEC int check_header_1024 (GLOBAL_AS const vc_t *esalt_bufs, GLOBAL_AS u32
return -1;
}
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -156,7 +156,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -175,7 +175,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -211,7 +211,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -230,7 +230,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m13732_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
@ -260,25 +260,39 @@ KERNEL_FQ void m13732_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -350,7 +364,7 @@ KERNEL_FQ void m13732_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -512,25 +526,39 @@ KERNEL_FQ void m13732_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -667,7 +695,7 @@ KERNEL_FQ void m13732_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];
@ -807,34 +835,6 @@ KERNEL_FQ void m13732_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
#endif
if (gid >= gid_max) return;
if (tmps[gid].pim)

110
OpenCL/m13733-pure.cl
Unescape View File

@ -202,7 +202,7 @@ DECLSPEC int check_header_1536 (GLOBAL_AS const vc_t *esalt_bufs, GLOBAL_AS u32
return -1;
}
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 (*s_MT)[256])
DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest, SHM_TYPE u64 *s_MT0, SHM_TYPE u64 *s_MT1, SHM_TYPE u64 *s_MT2, SHM_TYPE u64 *s_MT3, SHM_TYPE u64 *s_MT4, SHM_TYPE u64 *s_MT5, SHM_TYPE u64 *s_MT6, SHM_TYPE u64 *s_MT7)
{
digest[ 0] = ipad[ 0];
digest[ 1] = ipad[ 1];
@ -221,7 +221,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = ipad[14];
digest[15] = ipad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -240,7 +240,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = digest[ 0];
w0[1] = digest[ 1];
@ -276,7 +276,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
digest[14] = opad[14];
digest[15] = opad[15];
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
w0[0] = 0x80000000;
w0[1] = 0;
@ -295,7 +295,7 @@ DECLSPEC void hmac_whirlpool_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x
w3[2] = 0;
w3[3] = (64 + 64) * 8;
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT);
whirlpool_transform_vector (w0, w1, w2, w3, digest, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
}
KERNEL_FQ void m13733_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
@ -325,25 +325,39 @@ KERNEL_FQ void m13733_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -415,7 +429,7 @@ KERNEL_FQ void m13733_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
whirlpool_hmac_ctx_t whirlpool_hmac_ctx;
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT);
whirlpool_hmac_init_64 (&whirlpool_hmac_ctx, w0, w1, w2, w3, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
tmps[gid].ipad[ 0] = whirlpool_hmac_ctx.ipad.h[ 0];
tmps[gid].ipad[ 1] = whirlpool_hmac_ctx.ipad.h[ 1];
@ -577,25 +591,39 @@ KERNEL_FQ void m13733_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
LOCAL_VK u64 s_MT0[256];
LOCAL_VK u64 s_MT1[256];
LOCAL_VK u64 s_MT2[256];
LOCAL_VK u64 s_MT3[256];
LOCAL_VK u64 s_MT4[256];
LOCAL_VK u64 s_MT5[256];
LOCAL_VK u64 s_MT6[256];
LOCAL_VK u64 s_MT7[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
s_MT0[i] = MT0[i];
s_MT1[i] = MT1[i];
s_MT2[i] = MT2[i];
s_MT3[i] = MT3[i];
s_MT4[i] = MT4[i];
s_MT5[i] = MT5[i];
s_MT6[i] = MT6[i];
s_MT7[i] = MT7[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
CONSTANT_AS u64a *s_MT0 = MT0;
CONSTANT_AS u64a *s_MT1 = MT1;
CONSTANT_AS u64a *s_MT2 = MT2;
CONSTANT_AS u64a *s_MT3 = MT3;
CONSTANT_AS u64a *s_MT4 = MT4;
CONSTANT_AS u64a *s_MT5 = MT5;
CONSTANT_AS u64a *s_MT6 = MT6;
CONSTANT_AS u64a *s_MT7 = MT7;
#endif
@ -732,7 +760,7 @@ KERNEL_FQ void m13733_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
w3[2] = dgst[14];
w3[3] = dgst[15];
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT);
hmac_whirlpool_run_V (w0, w1, w2, w3, ipad, opad, dgst, s_MT0, s_MT1, s_MT2, s_MT3, s_MT4, s_MT5, s_MT6, s_MT7);
out[ 0] ^= dgst[ 0];
out[ 1] ^= dgst[ 1];
@ -873,34 +901,6 @@ KERNEL_FQ void m13733_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u64 s_MT[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_MT[0][i] = MT[0][i];
s_MT[1][i] = MT[1][i];
s_MT[2][i] = MT[2][i];
s_MT[3][i] = MT[3][i];
s_MT[4][i] = MT[4][i];
s_MT[5][i] = MT[5][i];
s_MT[6][i] = MT[6][i];
s_MT[7][i] = MT[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u64a (*s_MT)[256] = MT;
#endif
if (gid >= gid_max) return;
if (tmps[gid].pim)

Write Preview
Loading…
Cancel
Save