arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2025-05-28 11:48:49 +00:00
Code Issues Releases Wiki Activity

Merge pull request #1 from hashcat/master

Browse Source 
update
This commit is contained in:
Sein Coray 2020-03-31 20:23:37 +02:00 committed by GitHub
commit f8ef6f67bf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
415 changed files with 46927 additions and 12604 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

69
OpenCL/amp_a1.cl
View File

@ -34,11 +34,70 @@ KERNEL_FQ void amp (GLOBAL_AS pw_t *pws, GLOBAL_AS pw_t *pws_amp, GLOBAL_AS cons
switch_buffer_by_offset_1x64_le_S (pw.i, comb_len);
}
#pragma unroll
for (int i = 0; i < 64; i++)
{
pw.i[i] |= comb.i[i];
}
pw.i[ 0] |= comb.i[ 0];
pw.i[ 1] |= comb.i[ 1];
pw.i[ 2] |= comb.i[ 2];
pw.i[ 3] |= comb.i[ 3];
pw.i[ 4] |= comb.i[ 4];
pw.i[ 5] |= comb.i[ 5];
pw.i[ 6] |= comb.i[ 6];
pw.i[ 7] |= comb.i[ 7];
pw.i[ 8] |= comb.i[ 8];
pw.i[ 9] |= comb.i[ 9];
pw.i[10] |= comb.i[10];
pw.i[11] |= comb.i[11];
pw.i[12] |= comb.i[12];
pw.i[13] |= comb.i[13];
pw.i[14] |= comb.i[14];
pw.i[15] |= comb.i[15];
pw.i[16] |= comb.i[16];
pw.i[17] |= comb.i[17];
pw.i[18] |= comb.i[18];
pw.i[19] |= comb.i[19];
pw.i[20] |= comb.i[20];
pw.i[21] |= comb.i[21];
pw.i[22] |= comb.i[22];
pw.i[23] |= comb.i[23];
pw.i[24] |= comb.i[24];
pw.i[25] |= comb.i[25];
pw.i[26] |= comb.i[26];
pw.i[27] |= comb.i[27];
pw.i[28] |= comb.i[28];
pw.i[29] |= comb.i[29];
pw.i[30] |= comb.i[30];
pw.i[31] |= comb.i[31];
pw.i[32] |= comb.i[32];
pw.i[33] |= comb.i[33];
pw.i[34] |= comb.i[34];
pw.i[35] |= comb.i[35];
pw.i[36] |= comb.i[36];
pw.i[37] |= comb.i[37];
pw.i[38] |= comb.i[38];
pw.i[39] |= comb.i[39];
pw.i[40] |= comb.i[40];
pw.i[41] |= comb.i[41];
pw.i[42] |= comb.i[42];
pw.i[43] |= comb.i[43];
pw.i[44] |= comb.i[44];
pw.i[45] |= comb.i[45];
pw.i[46] |= comb.i[46];
pw.i[47] |= comb.i[47];
pw.i[48] |= comb.i[48];
pw.i[49] |= comb.i[49];
pw.i[50] |= comb.i[50];
pw.i[51] |= comb.i[51];
pw.i[52] |= comb.i[52];
pw.i[53] |= comb.i[53];
pw.i[54] |= comb.i[54];
pw.i[55] |= comb.i[55];
pw.i[56] |= comb.i[56];
pw.i[57] |= comb.i[57];
pw.i[58] |= comb.i[58];
pw.i[59] |= comb.i[59];
pw.i[60] |= comb.i[60];
pw.i[61] |= comb.i[61];
pw.i[62] |= comb.i[62];
pw.i[63] |= comb.i[63];
pw.pw_len = pw_len + comb_len;

836
OpenCL/inc_common.cl
View File

File diff suppressed because it is too large Load Diff

68
OpenCL/inc_common.h
View File

@ -110,27 +110,29 @@
*/
#ifdef IS_CUDA
#define KERN_ATTR_BASIC() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, void)
#define KERN_ATTR_BITSLICE() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bs_word_t *g_words_buf_s, void, void, void)
#define KERN_ATTR_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, e)
#define KERN_ATTR_RULES() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, void)
#define KERN_ATTR_RULES_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, e)
#define KERN_ATTR_TMPS(t) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, void, void)
#define KERN_ATTR_TMPS_ESALT(t,e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, void, e)
#define KERN_ATTR_TMPS_HOOKS(t,h) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, h, void)
#define KERN_ATTR_VECTOR() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const u32x *g_words_buf_r, void, void, void)
#define KERN_ATTR_VECTOR_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const u32x *g_words_buf_r, void, void, e)
#define KERN_ATTR_BASIC() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, void)
#define KERN_ATTR_BITSLICE() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bs_word_t *g_words_buf_s, void, void, void)
#define KERN_ATTR_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, e)
#define KERN_ATTR_RULES() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, void)
#define KERN_ATTR_RULES_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, void, void, e)
#define KERN_ATTR_TMPS(t) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, void, void)
#define KERN_ATTR_TMPS_ESALT(t,e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, void, e)
#define KERN_ATTR_TMPS_HOOKS(t,h) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, h, void)
#define KERN_ATTR_TMPS_HOOKS_ESALT(t,h,e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const bf_t *g_bfs_buf, t, h, e)
#define KERN_ATTR_VECTOR() KERN_ATTR (GLOBAL_AS, GLOBAL_AS const u32x *g_words_buf_r, void, void, void)
#define KERN_ATTR_VECTOR_ESALT(e) KERN_ATTR (GLOBAL_AS, GLOBAL_AS const u32x *g_words_buf_r, void, void, e)
#else
#define KERN_ATTR_BASIC() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, void, void, void)
#define KERN_ATTR_BITSLICE() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bs_word_t *words_buf_s, void, void, void)
#define KERN_ATTR_ESALT(e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, void, void, e)
#define KERN_ATTR_RULES() KERN_ATTR (CONSTANT_AS, GLOBAL_AS const bf_t *bfs_buf, void, void, void)
#define KERN_ATTR_RULES_ESALT(e) KERN_ATTR (CONSTANT_AS, GLOBAL_AS const bf_t *bfs_buf, void, void, e)
#define KERN_ATTR_TMPS(t) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, void, void)
#define KERN_ATTR_TMPS_ESALT(t,e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, void, e)
#define KERN_ATTR_TMPS_HOOKS(t,h) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, h, void)
#define KERN_ATTR_VECTOR() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const u32x *words_buf_r, void, void, void)
#define KERN_ATTR_VECTOR_ESALT(e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const u32x *words_buf_r, void, void, e)
#define KERN_ATTR_BASIC() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, void, void, void)
#define KERN_ATTR_BITSLICE() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bs_word_t *words_buf_s, void, void, void)
#define KERN_ATTR_ESALT(e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, void, void, e)
#define KERN_ATTR_RULES() KERN_ATTR (CONSTANT_AS, GLOBAL_AS const bf_t *bfs_buf, void, void, void)
#define KERN_ATTR_RULES_ESALT(e) KERN_ATTR (CONSTANT_AS, GLOBAL_AS const bf_t *bfs_buf, void, void, e)
#define KERN_ATTR_TMPS(t) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, void, void)
#define KERN_ATTR_TMPS_ESALT(t,e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, void, e)
#define KERN_ATTR_TMPS_HOOKS(t,h) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, h, void)
#define KERN_ATTR_TMPS_HOOKS_ESALT(t,h,e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const bf_t *bfs_buf, t, h, e)
#define KERN_ATTR_VECTOR() KERN_ATTR (GLOBAL_AS, CONSTANT_AS const u32x *words_buf_r, void, void, void)
#define KERN_ATTR_VECTOR_ESALT(e) KERN_ATTR (GLOBAL_AS, CONSTANT_AS const u32x *words_buf_r, void, void, e)
#endif
// union based packing
@ -140,6 +142,24 @@ DECLSPEC u8 v8b_from_v32_S (const u32 v32);
DECLSPEC u8 v8c_from_v32_S (const u32 v32);
DECLSPEC u8 v8d_from_v32_S (const u32 v32);
DECLSPEC u8 v8a_from_v64_S (const u64 v64);
DECLSPEC u8 v8b_from_v64_S (const u64 v64);
DECLSPEC u8 v8c_from_v64_S (const u64 v64);
DECLSPEC u8 v8d_from_v64_S (const u64 v64);
DECLSPEC u8 v8e_from_v64_S (const u64 v64);
DECLSPEC u8 v8f_from_v64_S (const u64 v64);
DECLSPEC u8 v8g_from_v64_S (const u64 v64);
DECLSPEC u8 v8h_from_v64_S (const u64 v64);
DECLSPEC u8x v8a_from_v64 (const u64x v64);
DECLSPEC u8x v8b_from_v64 (const u64x v64);
DECLSPEC u8x v8c_from_v64 (const u64x v64);
DECLSPEC u8x v8d_from_v64 (const u64x v64);
DECLSPEC u8x v8e_from_v64 (const u64x v64);
DECLSPEC u8x v8f_from_v64 (const u64x v64);
DECLSPEC u8x v8g_from_v64 (const u64x v64);
DECLSPEC u8x v8h_from_v64 (const u64x v64);
DECLSPEC u16 v16a_from_v32_S (const u32 v32);
DECLSPEC u16 v16b_from_v32_S (const u32 v32);
@ -151,6 +171,11 @@ DECLSPEC u64 v64_from_v32ab_S (const u32 v32a, const u32 v32b);
// inline asm packing
DECLSPEC u32x unpack_v8a_from_v32 (const u32x v32);
DECLSPEC u32x unpack_v8b_from_v32 (const u32x v32);
DECLSPEC u32x unpack_v8c_from_v32 (const u32x v32);
DECLSPEC u32x unpack_v8d_from_v32 (const u32x v32);
DECLSPEC u32 unpack_v8a_from_v32_S (const u32 v32);
DECLSPEC u32 unpack_v8b_from_v32_S (const u32 v32);
DECLSPEC u32 unpack_v8c_from_v32_S (const u32 v32);
@ -217,7 +242,7 @@ DECLSPEC int is_valid_hex_32 (const u32 v);
DECLSPEC int is_valid_base58_8 (const u8 v);
DECLSPEC int is_valid_base58_32 (const u32 v);
DECLSPEC int hc_find_keyboard_layout_map (const u32 search, const int search_len, LOCAL_AS keyboard_layout_mapping_t *s_keyboard_layout_mapping_buf, const int keyboard_layout_mapping_cnt);
DECLSPEC int hc_execute_keyboard_layout_mapping (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], const int pw_len, LOCAL_AS keyboard_layout_mapping_t *s_keyboard_layout_mapping_buf, const int keyboard_layout_mapping_cnt);
DECLSPEC int hc_execute_keyboard_layout_mapping (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const int pw_len, LOCAL_AS keyboard_layout_mapping_t *s_keyboard_layout_mapping_buf, const int keyboard_layout_mapping_cnt);
DECLSPEC void make_utf16be (const u32x *in, u32x *out1, u32x *out2);
DECLSPEC void make_utf16beN (const u32x *in, u32x *out1, u32x *out2);
DECLSPEC void make_utf16le (const u32x *in, u32x *out1, u32x *out2);
@ -275,6 +300,5 @@ DECLSPEC void append_0x01_4x4_VV (u32x *w0, u32x *w1, u32x *w2, u32x *w3, const
DECLSPEC void append_0x06_2x4_VV (u32x *w0, u32x *w1, const u32x offset);
DECLSPEC void append_0x80_2x4_VV (u32x *w0, u32x *w1, const u32x offset);
DECLSPEC void append_0x80_4x4_VV (u32x *w0, u32x *w1, u32x *w2, u32x *w3, const u32x offset);
DECLSPEC void gpu_decompress_entry (GLOBAL_AS pw_idx_t *pws_idx, GLOBAL_AS u32 *pws_comp, pw_t *pw, const u64 gid);
#endif

2064
OpenCL/inc_ecc_secp256k1.cl Normal file
View File

File diff suppressed because it is too large Load Diff

41
OpenCL/inc_ecc_secp256k1.h Normal file
View File

@ -0,0 +1,41 @@
/**
* Author......: See docs/credits.txt
* License.....: MIT
*/
#ifndef _INC_ECC_SECP256K1_H
#define _INC_ECC_SECP256K1_H
// y^2 = x^3 + ax + b with a = 0 and b = 7 => y^2 = x^3 + 7:
#define SECP256K1_B 7
#define SECP256K1_P0 0xfffffc2f
#define SECP256K1_P1 0xfffffffe
#define SECP256K1_P2 0xffffffff
#define SECP256K1_P3 0xffffffff
#define SECP256K1_P4 0xffffffff
#define SECP256K1_P5 0xffffffff
#define SECP256K1_P6 0xffffffff
#define SECP256K1_P7 0xffffffff
#define SECP256K1_N0 0xd0364141
#define SECP256K1_N1 0xbfd25e8c
#define SECP256K1_N2 0xaf48a03b
#define SECP256K1_N3 0xbaaedce6
#define SECP256K1_N4 0xfffffffe
#define SECP256K1_N5 0xffffffff
#define SECP256K1_N6 0xffffffff
#define SECP256K1_N7 0xffffffff
typedef struct secp256k1
{
u32 xy[96]; // pre-computed points: (x1,y1,-y1),(x3,y3,-y3),(x5,y5,-y5),(x7,y7,-y7)
} secp256k1_t;
DECLSPEC u32 parse_public (secp256k1_t *r, const u32 *k);
DECLSPEC void point_mul (u32 *r, const u32 *k, GLOBAL_AS const secp256k1_t *tmps);
#endif // _INC_ECC_SECP256K1_H

16
OpenCL/inc_hash_md4.cl
View File

@ -799,9 +799,7 @@ DECLSPEC void md4_hmac_init_64 (md4_hmac_ctx_t *ctx, const u32 *w0, const u32 *w
md4_init (&ctx->ipad);
md4_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
md4_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -824,9 +822,7 @@ DECLSPEC void md4_hmac_init_64 (md4_hmac_ctx_t *ctx, const u32 *w0, const u32 *w
md4_init (&ctx->opad);
md4_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
md4_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void md4_hmac_init (md4_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1686,9 +1682,7 @@ DECLSPEC void md4_hmac_init_vector_64 (md4_hmac_ctx_vector_t *ctx, const u32x *w
md4_init_vector (&ctx->ipad);
md4_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
md4_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1711,9 +1705,7 @@ DECLSPEC void md4_hmac_init_vector_64 (md4_hmac_ctx_vector_t *ctx, const u32x *w
md4_init_vector (&ctx->opad);
md4_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
md4_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void md4_hmac_init_vector (md4_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

18
OpenCL/inc_hash_md4.h
View File

@ -10,26 +10,14 @@
#define MD4_G_S(x,y,z) (((x) & (y)) | ((x) & (z)) | ((y) & (z)))
#define MD4_H_S(x,y,z) ((x) ^ (y) ^ (z))
#ifdef IS_NV
#define MD4_F(x,y,z) (((x) & (y)) | ((~(x)) & (z)))
#define MD4_G(x,y,z) (((x) & (y)) | ((x) & (z)) | ((y) & (z)))
#define MD4_H(x,y,z) ((x) ^ (y) ^ (z))
#ifdef USE_BITSELECT
#define MD4_Fo(x,y,z) (bitselect ((z), (y), (x)))
#define MD4_Go(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_AMD
#define MD4_F(x,y,z) (((x) & (y)) | ((~(x)) & (z)))
#define MD4_G(x,y,z) (((x) & (y)) | ((x) & (z)) | ((y) & (z)))
#define MD4_H(x,y,z) ((x) ^ (y) ^ (z))
#define MD4_Fo(x,y,z) (bitselect ((z), (y), (x)))
#define MD4_Go(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_GENERIC
#define MD4_F(x,y,z) (((x) & (y)) | ((~(x)) & (z)))
#define MD4_G(x,y,z) (((x) & (y)) | ((x) & (z)) | ((y) & (z)))
#define MD4_H(x,y,z) ((x) ^ (y) ^ (z))
#else
#define MD4_Fo(x,y,z) (MD4_F((x), (y), (z)))
#define MD4_Go(x,y,z) (MD4_G((x), (y), (z)))
#endif

16
OpenCL/inc_hash_md5.cl
View File

@ -835,9 +835,7 @@ DECLSPEC void md5_hmac_init_64 (md5_hmac_ctx_t *ctx, const u32 *w0, const u32 *w
md5_init (&ctx->ipad);
md5_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
md5_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -860,9 +858,7 @@ DECLSPEC void md5_hmac_init_64 (md5_hmac_ctx_t *ctx, const u32 *w0, const u32 *w
md5_init (&ctx->opad);
md5_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
md5_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void md5_hmac_init (md5_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1758,9 +1754,7 @@ DECLSPEC void md5_hmac_init_vector_64 (md5_hmac_ctx_vector_t *ctx, const u32x *w
md5_init_vector (&ctx->ipad);
md5_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
md5_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1783,9 +1777,7 @@ DECLSPEC void md5_hmac_init_vector_64 (md5_hmac_ctx_vector_t *ctx, const u32x *w
md5_init_vector (&ctx->opad);
md5_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
md5_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void md5_hmac_init_vector (md5_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

22
OpenCL/inc_hash_md5.h
View File

@ -11,35 +11,17 @@
#define MD5_H_S(x,y,z) ((x) ^ (y) ^ (z))
#define MD5_I_S(x,y,z) ((y) ^ ((x) | ~(z)))
#ifdef IS_NV
#define MD5_F(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define MD5_G(x,y,z) ((y) ^ ((z) & ((x) ^ (y))))
#define MD5_H(x,y,z) ((x) ^ (y) ^ (z))
#define MD5_H1(x,y,z) ((t = (x) ^ (y)) ^ (z))
#define MD5_H2(x,y,z) ((x) ^ t)
#define MD5_I(x,y,z) ((y) ^ ((x) | ~(z)))
#define MD5_Fo(x,y,z) (MD5_F((x), (y), (z)))
#define MD5_Go(x,y,z) (MD5_G((x), (y), (z)))
#endif
#ifdef IS_AMD
#define MD5_F(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define MD5_G(x,y,z) ((y) ^ ((z) & ((x) ^ (y))))
#define MD5_H(x,y,z) ((x) ^ (y) ^ (z))
#define MD5_H1(x,y,z) ((t = (x) ^ (y)) ^ (z))
#define MD5_H2(x,y,z) ((x) ^ t)
#define MD5_I(x,y,z) ((y) ^ ((x) | ~(z)))
#ifdef USE_BITSELECT
#define MD5_Fo(x,y,z) (bitselect ((z), (y), (x)))
#define MD5_Go(x,y,z) (bitselect ((y), (x), (z)))
#endif
#ifdef IS_GENERIC
#define MD5_F(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define MD5_G(x,y,z) ((y) ^ ((z) & ((x) ^ (y))))
#define MD5_H(x,y,z) ((x) ^ (y) ^ (z))
#define MD5_H1(x,y,z) ((t = (x) ^ (y)) ^ (z))
#define MD5_H2(x,y,z) ((x) ^ t)
#define MD5_I(x,y,z) ((y) ^ ((x) | ~(z)))
#else
#define MD5_Fo(x,y,z) (MD5_F((x), (y), (z)))
#define MD5_Go(x,y,z) (MD5_G((x), (y), (z)))
#endif

16
OpenCL/inc_hash_ripemd160.cl
View File

@ -933,9 +933,7 @@ DECLSPEC void ripemd160_hmac_init_64 (ripemd160_hmac_ctx_t *ctx, const u32 *w0,
ripemd160_init (&ctx->ipad);
ripemd160_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
ripemd160_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -958,9 +956,7 @@ DECLSPEC void ripemd160_hmac_init_64 (ripemd160_hmac_ctx_t *ctx, const u32 *w0,
ripemd160_init (&ctx->opad);
ripemd160_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
ripemd160_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void ripemd160_hmac_init (ripemd160_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1955,9 +1951,7 @@ DECLSPEC void ripemd160_hmac_init_vector_64 (ripemd160_hmac_ctx_vector_t *ctx, c
ripemd160_init_vector (&ctx->ipad);
ripemd160_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
ripemd160_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1980,9 +1974,7 @@ DECLSPEC void ripemd160_hmac_init_vector_64 (ripemd160_hmac_ctx_vector_t *ctx, c
ripemd160_init_vector (&ctx->opad);
ripemd160_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
ripemd160_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void ripemd160_hmac_init_vector (ripemd160_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

22
OpenCL/inc_hash_ripemd160.h
View File

@ -6,32 +6,16 @@
#ifndef _INC_HASH_RIPEMD160_H
#define _INC_HASH_RIPEMD160_H
#ifdef IS_NV
#define RIPEMD160_F(x,y,z) ((x) ^ (y) ^ (z))
#define RIPEMD160_G(x,y,z) ((z) ^ ((x) & ((y) ^ (z)))) /* x ? y : z */
#define RIPEMD160_H(x,y,z) (((x) | ~(y)) ^ (z))
#define RIPEMD160_I(x,y,z) ((y) ^ ((z) & ((x) ^ (y)))) /* z ? x : y */
#define RIPEMD160_J(x,y,z) ((x) ^ ((y) | ~(z)))
#ifdef USE_BITSELECT
#define RIPEMD160_Go(x,y,z) (bitselect ((z), (y), (x)))
#define RIPEMD160_Io(x,y,z) (bitselect ((y), (x), (z)))
#endif
#ifdef IS_AMD
#define RIPEMD160_F(x,y,z) ((x) ^ (y) ^ (z))
#define RIPEMD160_G(x,y,z) ((z) ^ ((x) & ((y) ^ (z)))) /* x ? y : z */
#define RIPEMD160_H(x,y,z) (((x) | ~(y)) ^ (z))
#define RIPEMD160_I(x,y,z) ((y) ^ ((z) & ((x) ^ (y)))) /* z ? x : y */
#define RIPEMD160_J(x,y,z) ((x) ^ ((y) | ~(z)))
#define RIPEMD160_Go(x,y,z) (bitselect ((z), (y), (x)))
#define RIPEMD160_Io(x,y,z) (bitselect ((y), (x), (z)))
#endif
#ifdef IS_GENERIC
#define RIPEMD160_F(x,y,z) ((x) ^ (y) ^ (z))
#define RIPEMD160_G(x,y,z) ((z) ^ ((x) & ((y) ^ (z)))) /* x ? y : z */
#define RIPEMD160_H(x,y,z) (((x) | ~(y)) ^ (z))
#define RIPEMD160_I(x,y,z) ((y) ^ ((z) & ((x) ^ (y)))) /* z ? x : y */
#define RIPEMD160_J(x,y,z) ((x) ^ ((y) | ~(z)))
#else
#define RIPEMD160_Go(x,y,z) (RIPEMD160_G ((x), (y), (z)))
#define RIPEMD160_Io(x,y,z) (RIPEMD160_I ((x), (y), (z)))
#endif

384
OpenCL/inc_hash_sha1.cl
View File

@ -22,6 +22,8 @@ DECLSPEC void sha1_transform (const u32 *w0, const u32 *w1, const u32 *w2, const
u32 d = digest[3];
u32 e = digest[4];
#ifdef IS_CPU
u32 w0_t = w0[0];
u32 w1_t = w0[1];
u32 w2_t = w0[2];
@ -136,6 +138,187 @@ DECLSPEC void sha1_transform (const u32 *w0, const u32 *w1, const u32 *w2, const
#undef K
#else
u32 w00_t = w0[0];
u32 w01_t = w0[1];
u32 w02_t = w0[2];
u32 w03_t = w0[3];
u32 w04_t = w1[0];
u32 w05_t = w1[1];
u32 w06_t = w1[2];
u32 w07_t = w1[3];
u32 w08_t = w2[0];
u32 w09_t = w2[1];
u32 w0a_t = w2[2];
u32 w0b_t = w2[3];
u32 w0c_t = w3[0];
u32 w0d_t = w3[1];
u32 w0e_t = w3[2];
u32 w0f_t = w3[3];
u32 w10_t;
u32 w11_t;
u32 w12_t;
u32 w13_t;
u32 w14_t;
u32 w15_t;
u32 w16_t;
u32 w17_t;
u32 w18_t;
u32 w19_t;
u32 w1a_t;
u32 w1b_t;
u32 w1c_t;
u32 w1d_t;
u32 w1e_t;
u32 w1f_t;
u32 w20_t;
u32 w21_t;
u32 w22_t;
u32 w23_t;
u32 w24_t;
u32 w25_t;
u32 w26_t;
u32 w27_t;
u32 w28_t;
u32 w29_t;
u32 w2a_t;
u32 w2b_t;
u32 w2c_t;
u32 w2d_t;
u32 w2e_t;
u32 w2f_t;
u32 w30_t;
u32 w31_t;
u32 w32_t;
u32 w33_t;
u32 w34_t;
u32 w35_t;
u32 w36_t;
u32 w37_t;
u32 w38_t;
u32 w39_t;
u32 w3a_t;
u32 w3b_t;
u32 w3c_t;
u32 w3d_t;
u32 w3e_t;
u32 w3f_t;
u32 w40_t;
u32 w41_t;
u32 w42_t;
u32 w43_t;
u32 w44_t;
u32 w45_t;
u32 w46_t;
u32 w47_t;
u32 w48_t;
u32 w49_t;
u32 w4a_t;
u32 w4b_t;
u32 w4c_t;
u32 w4d_t;
u32 w4e_t;
u32 w4f_t;
#define K SHA1C00
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w00_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w01_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w02_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w03_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w04_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w05_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w06_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w07_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w08_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w09_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w0a_t);
SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w0b_t);
SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w0c_t);
SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w0d_t);
SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w0e_t);
SHA1_STEP_S (SHA1_F0o, a, b, c, d, e, w0f_t);
w10_t = hc_rotl32_S ((w0d_t ^ w08_t ^ w02_t ^ w00_t), 1u); SHA1_STEP_S (SHA1_F0o, e, a, b, c, d, w10_t);
w11_t = hc_rotl32_S ((w0e_t ^ w09_t ^ w03_t ^ w01_t), 1u); SHA1_STEP_S (SHA1_F0o, d, e, a, b, c, w11_t);
w12_t = hc_rotl32_S ((w0f_t ^ w0a_t ^ w04_t ^ w02_t), 1u); SHA1_STEP_S (SHA1_F0o, c, d, e, a, b, w12_t);
w13_t = hc_rotl32_S ((w10_t ^ w0b_t ^ w05_t ^ w03_t), 1u); SHA1_STEP_S (SHA1_F0o, b, c, d, e, a, w13_t);
#undef K
#define K SHA1C01
w14_t = hc_rotl32_S ((w11_t ^ w0c_t ^ w06_t ^ w04_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w14_t);
w15_t = hc_rotl32_S ((w12_t ^ w0d_t ^ w07_t ^ w05_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w15_t);
w16_t = hc_rotl32_S ((w13_t ^ w0e_t ^ w08_t ^ w06_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w16_t);
w17_t = hc_rotl32_S ((w14_t ^ w0f_t ^ w09_t ^ w07_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w17_t);
w18_t = hc_rotl32_S ((w15_t ^ w10_t ^ w0a_t ^ w08_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w18_t);
w19_t = hc_rotl32_S ((w16_t ^ w11_t ^ w0b_t ^ w09_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w19_t);
w1a_t = hc_rotl32_S ((w17_t ^ w12_t ^ w0c_t ^ w0a_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w1a_t);
w1b_t = hc_rotl32_S ((w18_t ^ w13_t ^ w0d_t ^ w0b_t), 1u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w1b_t);
w1c_t = hc_rotl32_S ((w19_t ^ w14_t ^ w0e_t ^ w0c_t), 1u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w1c_t);
w1d_t = hc_rotl32_S ((w1a_t ^ w15_t ^ w0f_t ^ w0d_t), 1u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w1d_t);
w1e_t = hc_rotl32_S ((w1b_t ^ w16_t ^ w10_t ^ w0e_t), 1u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w1e_t);
w1f_t = hc_rotl32_S ((w1c_t ^ w17_t ^ w11_t ^ w0f_t), 1u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w1f_t);
w20_t = hc_rotl32_S ((w1a_t ^ w10_t ^ w04_t ^ w00_t), 2u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w20_t);
w21_t = hc_rotl32_S ((w1b_t ^ w11_t ^ w05_t ^ w01_t), 2u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w21_t);
w22_t = hc_rotl32_S ((w1c_t ^ w12_t ^ w06_t ^ w02_t), 2u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w22_t);
w23_t = hc_rotl32_S ((w1d_t ^ w13_t ^ w07_t ^ w03_t), 2u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w23_t);
w24_t = hc_rotl32_S ((w1e_t ^ w14_t ^ w08_t ^ w04_t), 2u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w24_t);
w25_t = hc_rotl32_S ((w1f_t ^ w15_t ^ w09_t ^ w05_t), 2u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w25_t);
w26_t = hc_rotl32_S ((w20_t ^ w16_t ^ w0a_t ^ w06_t), 2u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w26_t);
w27_t = hc_rotl32_S ((w21_t ^ w17_t ^ w0b_t ^ w07_t), 2u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w27_t);
#undef K
#define K SHA1C02
w28_t = hc_rotl32_S ((w22_t ^ w18_t ^ w0c_t ^ w08_t), 2u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w28_t);
w29_t = hc_rotl32_S ((w23_t ^ w19_t ^ w0d_t ^ w09_t), 2u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w29_t);
w2a_t = hc_rotl32_S ((w24_t ^ w1a_t ^ w0e_t ^ w0a_t), 2u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w2a_t);
w2b_t = hc_rotl32_S ((w25_t ^ w1b_t ^ w0f_t ^ w0b_t), 2u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w2b_t);
w2c_t = hc_rotl32_S ((w26_t ^ w1c_t ^ w10_t ^ w0c_t), 2u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w2c_t);
w2d_t = hc_rotl32_S ((w27_t ^ w1d_t ^ w11_t ^ w0d_t), 2u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w2d_t);
w2e_t = hc_rotl32_S ((w28_t ^ w1e_t ^ w12_t ^ w0e_t), 2u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w2e_t);
w2f_t = hc_rotl32_S ((w29_t ^ w1f_t ^ w13_t ^ w0f_t), 2u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w2f_t);
w30_t = hc_rotl32_S ((w2a_t ^ w20_t ^ w14_t ^ w10_t), 2u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w30_t);
w31_t = hc_rotl32_S ((w2b_t ^ w21_t ^ w15_t ^ w11_t), 2u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w31_t);
w32_t = hc_rotl32_S ((w2c_t ^ w22_t ^ w16_t ^ w12_t), 2u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w32_t);
w33_t = hc_rotl32_S ((w2d_t ^ w23_t ^ w17_t ^ w13_t), 2u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w33_t);
w34_t = hc_rotl32_S ((w2e_t ^ w24_t ^ w18_t ^ w14_t), 2u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w34_t);
w35_t = hc_rotl32_S ((w2f_t ^ w25_t ^ w19_t ^ w15_t), 2u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w35_t);
w36_t = hc_rotl32_S ((w30_t ^ w26_t ^ w1a_t ^ w16_t), 2u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w36_t);
w37_t = hc_rotl32_S ((w31_t ^ w27_t ^ w1b_t ^ w17_t), 2u); SHA1_STEP_S (SHA1_F2o, a, b, c, d, e, w37_t);
w38_t = hc_rotl32_S ((w32_t ^ w28_t ^ w1c_t ^ w18_t), 2u); SHA1_STEP_S (SHA1_F2o, e, a, b, c, d, w38_t);
w39_t = hc_rotl32_S ((w33_t ^ w29_t ^ w1d_t ^ w19_t), 2u); SHA1_STEP_S (SHA1_F2o, d, e, a, b, c, w39_t);
w3a_t = hc_rotl32_S ((w34_t ^ w2a_t ^ w1e_t ^ w1a_t), 2u); SHA1_STEP_S (SHA1_F2o, c, d, e, a, b, w3a_t);
w3b_t = hc_rotl32_S ((w35_t ^ w2b_t ^ w1f_t ^ w1b_t), 2u); SHA1_STEP_S (SHA1_F2o, b, c, d, e, a, w3b_t);
#undef K
#define K SHA1C03
w3c_t = hc_rotl32_S ((w36_t ^ w2c_t ^ w20_t ^ w1c_t), 2u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w3c_t);
w3d_t = hc_rotl32_S ((w37_t ^ w2d_t ^ w21_t ^ w1d_t), 2u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w3d_t);
w3e_t = hc_rotl32_S ((w38_t ^ w2e_t ^ w22_t ^ w1e_t), 2u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w3e_t);
w3f_t = hc_rotl32_S ((w39_t ^ w2f_t ^ w23_t ^ w1f_t), 2u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w3f_t);
w40_t = hc_rotl32_S ((w34_t ^ w20_t ^ w08_t ^ w00_t), 4u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w40_t);
w41_t = hc_rotl32_S ((w35_t ^ w21_t ^ w09_t ^ w01_t), 4u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w41_t);
w42_t = hc_rotl32_S ((w36_t ^ w22_t ^ w0a_t ^ w02_t), 4u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w42_t);
w43_t = hc_rotl32_S ((w37_t ^ w23_t ^ w0b_t ^ w03_t), 4u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w43_t);
w44_t = hc_rotl32_S ((w38_t ^ w24_t ^ w0c_t ^ w04_t), 4u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w44_t);
w45_t = hc_rotl32_S ((w39_t ^ w25_t ^ w0d_t ^ w05_t), 4u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w45_t);
w46_t = hc_rotl32_S ((w3a_t ^ w26_t ^ w0e_t ^ w06_t), 4u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w46_t);
w47_t = hc_rotl32_S ((w3b_t ^ w27_t ^ w0f_t ^ w07_t), 4u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w47_t);
w48_t = hc_rotl32_S ((w3c_t ^ w28_t ^ w10_t ^ w08_t), 4u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w48_t);
w49_t = hc_rotl32_S ((w3d_t ^ w29_t ^ w11_t ^ w09_t), 4u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w49_t);
w4a_t = hc_rotl32_S ((w3e_t ^ w2a_t ^ w12_t ^ w0a_t), 4u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w4a_t);
w4b_t = hc_rotl32_S ((w3f_t ^ w2b_t ^ w13_t ^ w0b_t), 4u); SHA1_STEP_S (SHA1_F1, a, b, c, d, e, w4b_t);
w4c_t = hc_rotl32_S ((w40_t ^ w2c_t ^ w14_t ^ w0c_t), 4u); SHA1_STEP_S (SHA1_F1, e, a, b, c, d, w4c_t);
w4d_t = hc_rotl32_S ((w41_t ^ w2d_t ^ w15_t ^ w0d_t), 4u); SHA1_STEP_S (SHA1_F1, d, e, a, b, c, w4d_t);
w4e_t = hc_rotl32_S ((w42_t ^ w2e_t ^ w16_t ^ w0e_t), 4u); SHA1_STEP_S (SHA1_F1, c, d, e, a, b, w4e_t);
w4f_t = hc_rotl32_S ((w43_t ^ w2f_t ^ w17_t ^ w0f_t), 4u); SHA1_STEP_S (SHA1_F1, b, c, d, e, a, w4f_t);
#undef K
#endif
digest[0] += a;
digest[1] += b;
digest[2] += c;
@ -1101,9 +1284,7 @@ DECLSPEC void sha1_hmac_init_64 (sha1_hmac_ctx_t *ctx, const u32 *w0, const u32
sha1_init (&ctx->ipad);
sha1_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha1_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1126,9 +1307,7 @@ DECLSPEC void sha1_hmac_init_64 (sha1_hmac_ctx_t *ctx, const u32 *w0, const u32
sha1_init (&ctx->opad);
sha1_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha1_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha1_hmac_init (sha1_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1440,6 +1619,9 @@ DECLSPEC void sha1_transform_vector (const u32x *w0, const u32x *w1, const u32x
u32x d = digest[3];
u32x e = digest[4];
#ifdef IS_CPU
u32x w0_t = w0[0];
u32x w1_t = w0[1];
u32x w2_t = w0[2];
@ -1554,6 +1736,188 @@ DECLSPEC void sha1_transform_vector (const u32x *w0, const u32x *w1, const u32x
#undef K
#else
u32x w00_t = w0[0];
u32x w01_t = w0[1];
u32x w02_t = w0[2];
u32x w03_t = w0[3];
u32x w04_t = w1[0];
u32x w05_t = w1[1];
u32x w06_t = w1[2];
u32x w07_t = w1[3];
u32x w08_t = w2[0];
u32x w09_t = w2[1];
u32x w0a_t = w2[2];
u32x w0b_t = w2[3];
u32x w0c_t = w3[0];
u32x w0d_t = w3[1];
u32x w0e_t = w3[2];
u32x w0f_t = w3[3];
u32x w10_t;
u32x w11_t;
u32x w12_t;
u32x w13_t;
u32x w14_t;
u32x w15_t;
u32x w16_t;
u32x w17_t;
u32x w18_t;
u32x w19_t;
u32x w1a_t;
u32x w1b_t;
u32x w1c_t;
u32x w1d_t;
u32x w1e_t;
u32x w1f_t;
u32x w20_t;
u32x w21_t;
u32x w22_t;
u32x w23_t;
u32x w24_t;
u32x w25_t;
u32x w26_t;
u32x w27_t;
u32x w28_t;
u32x w29_t;
u32x w2a_t;
u32x w2b_t;
u32x w2c_t;
u32x w2d_t;
u32x w2e_t;
u32x w2f_t;
u32x w30_t;
u32x w31_t;
u32x w32_t;
u32x w33_t;
u32x w34_t;
u32x w35_t;
u32x w36_t;
u32x w37_t;
u32x w38_t;
u32x w39_t;
u32x w3a_t;
u32x w3b_t;
u32x w3c_t;
u32x w3d_t;
u32x w3e_t;
u32x w3f_t;
u32x w40_t;
u32x w41_t;
u32x w42_t;
u32x w43_t;
u32x w44_t;
u32x w45_t;
u32x w46_t;
u32x w47_t;
u32x w48_t;
u32x w49_t;
u32x w4a_t;
u32x w4b_t;
u32x w4c_t;
u32x w4d_t;
u32x w4e_t;
u32x w4f_t;
#define K SHA1C00
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w00_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w01_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w02_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w03_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w04_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w05_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w06_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w07_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w08_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w09_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0a_t);
SHA1_STEP (SHA1_F0o, e, a, b, c, d, w0b_t);
SHA1_STEP (SHA1_F0o, d, e, a, b, c, w0c_t);
SHA1_STEP (SHA1_F0o, c, d, e, a, b, w0d_t);
SHA1_STEP (SHA1_F0o, b, c, d, e, a, w0e_t);
SHA1_STEP (SHA1_F0o, a, b, c, d, e, w0f_t);
w10_t = hc_rotl32 ((w0d_t ^ w08_t ^ w02_t ^ w00_t), 1u); SHA1_STEP (SHA1_F0o, e, a, b, c, d, w10_t);
w11_t = hc_rotl32 ((w0e_t ^ w09_t ^ w03_t ^ w01_t), 1u); SHA1_STEP (SHA1_F0o, d, e, a, b, c, w11_t);
w12_t = hc_rotl32 ((w0f_t ^ w0a_t ^ w04_t ^ w02_t), 1u); SHA1_STEP (SHA1_F0o, c, d, e, a, b, w12_t);
w13_t = hc_rotl32 ((w10_t ^ w0b_t ^ w05_t ^ w03_t), 1u); SHA1_STEP (SHA1_F0o, b, c, d, e, a, w13_t);
#undef K
#define K SHA1C01
w14_t = hc_rotl32 ((w11_t ^ w0c_t ^ w06_t ^ w04_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w14_t);
w15_t = hc_rotl32 ((w12_t ^ w0d_t ^ w07_t ^ w05_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w15_t);
w16_t = hc_rotl32 ((w13_t ^ w0e_t ^ w08_t ^ w06_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w16_t);
w17_t = hc_rotl32 ((w14_t ^ w0f_t ^ w09_t ^ w07_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w17_t);
w18_t = hc_rotl32 ((w15_t ^ w10_t ^ w0a_t ^ w08_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w18_t);
w19_t = hc_rotl32 ((w16_t ^ w11_t ^ w0b_t ^ w09_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w19_t);
w1a_t = hc_rotl32 ((w17_t ^ w12_t ^ w0c_t ^ w0a_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1a_t);
w1b_t = hc_rotl32 ((w18_t ^ w13_t ^ w0d_t ^ w0b_t), 1u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w1b_t);
w1c_t = hc_rotl32 ((w19_t ^ w14_t ^ w0e_t ^ w0c_t), 1u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w1c_t);
w1d_t = hc_rotl32 ((w1a_t ^ w15_t ^ w0f_t ^ w0d_t), 1u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w1d_t);
w1e_t = hc_rotl32 ((w1b_t ^ w16_t ^ w10_t ^ w0e_t), 1u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w1e_t);
w1f_t = hc_rotl32 ((w1c_t ^ w17_t ^ w11_t ^ w0f_t), 1u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w1f_t);
w20_t = hc_rotl32 ((w1a_t ^ w10_t ^ w04_t ^ w00_t), 2u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w20_t);
w21_t = hc_rotl32 ((w1b_t ^ w11_t ^ w05_t ^ w01_t), 2u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w21_t);
w22_t = hc_rotl32 ((w1c_t ^ w12_t ^ w06_t ^ w02_t), 2u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w22_t);
w23_t = hc_rotl32 ((w1d_t ^ w13_t ^ w07_t ^ w03_t), 2u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w23_t);
w24_t = hc_rotl32 ((w1e_t ^ w14_t ^ w08_t ^ w04_t), 2u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w24_t);
w25_t = hc_rotl32 ((w1f_t ^ w15_t ^ w09_t ^ w05_t), 2u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w25_t);
w26_t = hc_rotl32 ((w20_t ^ w16_t ^ w0a_t ^ w06_t), 2u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w26_t);
w27_t = hc_rotl32 ((w21_t ^ w17_t ^ w0b_t ^ w07_t), 2u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w27_t);
#undef K
#define K SHA1C02
w28_t = hc_rotl32 ((w22_t ^ w18_t ^ w0c_t ^ w08_t), 2u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w28_t);
w29_t = hc_rotl32 ((w23_t ^ w19_t ^ w0d_t ^ w09_t), 2u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w29_t);
w2a_t = hc_rotl32 ((w24_t ^ w1a_t ^ w0e_t ^ w0a_t), 2u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2a_t);
w2b_t = hc_rotl32 ((w25_t ^ w1b_t ^ w0f_t ^ w0b_t), 2u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w2b_t);
w2c_t = hc_rotl32 ((w26_t ^ w1c_t ^ w10_t ^ w0c_t), 2u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w2c_t);
w2d_t = hc_rotl32 ((w27_t ^ w1d_t ^ w11_t ^ w0d_t), 2u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w2d_t);
w2e_t = hc_rotl32 ((w28_t ^ w1e_t ^ w12_t ^ w0e_t), 2u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w2e_t);
w2f_t = hc_rotl32 ((w29_t ^ w1f_t ^ w13_t ^ w0f_t), 2u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w2f_t);
w30_t = hc_rotl32 ((w2a_t ^ w20_t ^ w14_t ^ w10_t), 2u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w30_t);
w31_t = hc_rotl32 ((w2b_t ^ w21_t ^ w15_t ^ w11_t), 2u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w31_t);
w32_t = hc_rotl32 ((w2c_t ^ w22_t ^ w16_t ^ w12_t), 2u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w32_t);
w33_t = hc_rotl32 ((w2d_t ^ w23_t ^ w17_t ^ w13_t), 2u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w33_t);
w34_t = hc_rotl32 ((w2e_t ^ w24_t ^ w18_t ^ w14_t), 2u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w34_t);
w35_t = hc_rotl32 ((w2f_t ^ w25_t ^ w19_t ^ w15_t), 2u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w35_t);
w36_t = hc_rotl32 ((w30_t ^ w26_t ^ w1a_t ^ w16_t), 2u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w36_t);
w37_t = hc_rotl32 ((w31_t ^ w27_t ^ w1b_t ^ w17_t), 2u); SHA1_STEP (SHA1_F2o, a, b, c, d, e, w37_t);
w38_t = hc_rotl32 ((w32_t ^ w28_t ^ w1c_t ^ w18_t), 2u); SHA1_STEP (SHA1_F2o, e, a, b, c, d, w38_t);
w39_t = hc_rotl32 ((w33_t ^ w29_t ^ w1d_t ^ w19_t), 2u); SHA1_STEP (SHA1_F2o, d, e, a, b, c, w39_t);
w3a_t = hc_rotl32 ((w34_t ^ w2a_t ^ w1e_t ^ w1a_t), 2u); SHA1_STEP (SHA1_F2o, c, d, e, a, b, w3a_t);
w3b_t = hc_rotl32 ((w35_t ^ w2b_t ^ w1f_t ^ w1b_t), 2u); SHA1_STEP (SHA1_F2o, b, c, d, e, a, w3b_t);
#undef K
#define K SHA1C03
w3c_t = hc_rotl32 ((w36_t ^ w2c_t ^ w20_t ^ w1c_t), 2u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w3c_t);
w3d_t = hc_rotl32 ((w37_t ^ w2d_t ^ w21_t ^ w1d_t), 2u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w3d_t);
w3e_t = hc_rotl32 ((w38_t ^ w2e_t ^ w22_t ^ w1e_t), 2u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w3e_t);
w3f_t = hc_rotl32 ((w39_t ^ w2f_t ^ w23_t ^ w1f_t), 2u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w3f_t);
w40_t = hc_rotl32 ((w34_t ^ w20_t ^ w08_t ^ w00_t), 4u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w40_t);
w41_t = hc_rotl32 ((w35_t ^ w21_t ^ w09_t ^ w01_t), 4u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w41_t);
w42_t = hc_rotl32 ((w36_t ^ w22_t ^ w0a_t ^ w02_t), 4u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w42_t);
w43_t = hc_rotl32 ((w37_t ^ w23_t ^ w0b_t ^ w03_t), 4u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w43_t);
w44_t = hc_rotl32 ((w38_t ^ w24_t ^ w0c_t ^ w04_t), 4u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w44_t);
w45_t = hc_rotl32 ((w39_t ^ w25_t ^ w0d_t ^ w05_t), 4u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w45_t);
w46_t = hc_rotl32 ((w3a_t ^ w26_t ^ w0e_t ^ w06_t), 4u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w46_t);
w47_t = hc_rotl32 ((w3b_t ^ w27_t ^ w0f_t ^ w07_t), 4u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w47_t);
w48_t = hc_rotl32 ((w3c_t ^ w28_t ^ w10_t ^ w08_t), 4u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w48_t);
w49_t = hc_rotl32 ((w3d_t ^ w29_t ^ w11_t ^ w09_t), 4u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w49_t);
w4a_t = hc_rotl32 ((w3e_t ^ w2a_t ^ w12_t ^ w0a_t), 4u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w4a_t);
w4b_t = hc_rotl32 ((w3f_t ^ w2b_t ^ w13_t ^ w0b_t), 4u); SHA1_STEP (SHA1_F1, a, b, c, d, e, w4b_t);
w4c_t = hc_rotl32 ((w40_t ^ w2c_t ^ w14_t ^ w0c_t), 4u); SHA1_STEP (SHA1_F1, e, a, b, c, d, w4c_t);
w4d_t = hc_rotl32 ((w41_t ^ w2d_t ^ w15_t ^ w0d_t), 4u); SHA1_STEP (SHA1_F1, d, e, a, b, c, w4d_t);
w4e_t = hc_rotl32 ((w42_t ^ w2e_t ^ w16_t ^ w0e_t), 4u); SHA1_STEP (SHA1_F1, c, d, e, a, b, w4e_t);
w4f_t = hc_rotl32 ((w43_t ^ w2f_t ^ w17_t ^ w0f_t), 4u); SHA1_STEP (SHA1_F1, b, c, d, e, a, w4f_t);
#undef K
#endif
digest[0] += a;
digest[1] += b;
digest[2] += c;
@ -2139,9 +2503,7 @@ DECLSPEC void sha1_hmac_init_vector_64 (sha1_hmac_ctx_vector_t *ctx, const u32x
sha1_init_vector (&ctx->ipad);
sha1_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha1_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -2164,9 +2526,7 @@ DECLSPEC void sha1_hmac_init_vector_64 (sha1_hmac_ctx_vector_t *ctx, const u32x
sha1_init_vector (&ctx->opad);
sha1_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha1_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha1_hmac_init_vector (sha1_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

18
OpenCL/inc_hash_sha1.h
View File

@ -6,26 +6,14 @@
#ifndef _INC_HASH_SHA1_H
#define _INC_HASH_SHA1_H
#ifdef IS_NV
#define SHA1_F0(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA1_F1(x,y,z) ((x) ^ (y) ^ (z))
#define SHA1_F2(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#ifdef USE_BITSELECT
#define SHA1_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA1_F2o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_AMD
#define SHA1_F0(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA1_F1(x,y,z) ((x) ^ (y) ^ (z))
#define SHA1_F2(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA1_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA1_F2o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_GENERIC
#define SHA1_F0(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA1_F1(x,y,z) ((x) ^ (y) ^ (z))
#define SHA1_F2(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#else
#define SHA1_F0o(x,y,z) (SHA1_F0 ((x), (y), (z)))
#define SHA1_F2o(x,y,z) (SHA1_F2 ((x), (y), (z)))
#endif

16
OpenCL/inc_hash_sha224.cl
View File

@ -850,9 +850,7 @@ DECLSPEC void sha224_hmac_init_64 (sha224_hmac_ctx_t *ctx, const u32 *w0, const
sha224_init (&ctx->ipad);
sha224_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha224_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -875,9 +873,7 @@ DECLSPEC void sha224_hmac_init_64 (sha224_hmac_ctx_t *ctx, const u32 *w0, const
sha224_init (&ctx->opad);
sha224_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha224_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha224_hmac_init (sha224_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1814,9 +1810,7 @@ DECLSPEC void sha224_hmac_init_vector_64 (sha224_hmac_ctx_vector_t *ctx, const u
sha224_init_vector (&ctx->ipad);
sha224_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha224_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1839,9 +1833,7 @@ DECLSPEC void sha224_hmac_init_vector_64 (sha224_hmac_ctx_vector_t *ctx, const u
sha224_init_vector (&ctx->opad);
sha224_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha224_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha224_hmac_init_vector (sha224_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

16
OpenCL/inc_hash_sha224.h
View File

@ -18,23 +18,13 @@
#define SHA224_S2(x) (hc_rotl32 ((x), 30u) ^ hc_rotl32 ((x), 19u) ^ hc_rotl32 ((x), 10u))
#define SHA224_S3(x) (hc_rotl32 ((x), 26u) ^ hc_rotl32 ((x), 21u) ^ hc_rotl32 ((x), 7u))
#ifdef IS_NV
#define SHA224_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA224_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#ifdef USE_BITSELECT
#define SHA224_F0o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#define SHA224_F1o(x,y,z) (bitselect ((z), (y), (x)))
#endif
#ifdef IS_AMD
#define SHA224_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA224_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA224_F0o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#define SHA224_F1o(x,y,z) (bitselect ((z), (y), (x)))
#endif
#ifdef IS_GENERIC
#define SHA224_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA224_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#else
#define SHA224_F0o(x,y,z) (SHA224_F0 ((x), (y), (z)))
#define SHA224_F1o(x,y,z) (SHA224_F1 ((x), (y), (z)))
#endif

16
OpenCL/inc_hash_sha256.cl
View File

@ -850,9 +850,7 @@ DECLSPEC void sha256_hmac_init_64 (sha256_hmac_ctx_t *ctx, const u32 *w0, const
sha256_init (&ctx->ipad);
sha256_transform (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha256_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -875,9 +873,7 @@ DECLSPEC void sha256_hmac_init_64 (sha256_hmac_ctx_t *ctx, const u32 *w0, const
sha256_init (&ctx->opad);
sha256_transform (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha256_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha256_hmac_init (sha256_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -1814,9 +1810,7 @@ DECLSPEC void sha256_hmac_init_vector_64 (sha256_hmac_ctx_vector_t *ctx, const u
sha256_init_vector (&ctx->ipad);
sha256_transform_vector (t0, t1, t2, t3, ctx->ipad.h);
ctx->ipad.len = 64;
sha256_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1839,9 +1833,7 @@ DECLSPEC void sha256_hmac_init_vector_64 (sha256_hmac_ctx_vector_t *ctx, const u
sha256_init_vector (&ctx->opad);
sha256_transform_vector (t0, t1, t2, t3, ctx->opad.h);
ctx->opad.len = 64;
sha256_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void sha256_hmac_init_vector (sha256_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

16
OpenCL/inc_hash_sha256.h
View File

@ -18,23 +18,13 @@
#define SHA256_S2(x) (hc_rotl32 ((x), 30u) ^ hc_rotl32 ((x), 19u) ^ hc_rotl32 ((x), 10u))
#define SHA256_S3(x) (hc_rotl32 ((x), 26u) ^ hc_rotl32 ((x), 21u) ^ hc_rotl32 ((x), 7u))
#ifdef IS_NV
#define SHA256_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA256_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#ifdef USE_BITSELECT
#define SHA256_F0o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#define SHA256_F1o(x,y,z) (bitselect ((z), (y), (x)))
#endif
#ifdef IS_AMD
#define SHA256_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA256_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA256_F0o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#define SHA256_F1o(x,y,z) (bitselect ((z), (y), (x)))
#endif
#ifdef IS_GENERIC
#define SHA256_F0(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#define SHA256_F1(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#else
#define SHA256_F0o(x,y,z) (SHA256_F0 ((x), (y), (z)))
#define SHA256_F1o(x,y,z) (SHA256_F1 ((x), (y), (z)))
#endif

20
OpenCL/inc_hash_sha384.cl
View File

@ -219,7 +219,7 @@ DECLSPEC void sha384_update_128 (sha384_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u
if (len == 128)
{
sha384_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
sha384_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
ctx->w0[0] = 0;
ctx->w0[1] = 0;
@ -1358,9 +1358,7 @@ DECLSPEC void sha384_hmac_init_128 (sha384_hmac_ctx_t *ctx, const u32 *w0, const
sha384_init (&ctx->ipad);
sha384_transform (t0, t1, t2, t3, t4, t5, t6, t7, ctx->ipad.h);
ctx->ipad.len = 128;
sha384_update_128 (&ctx->ipad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
// opad
@ -1399,9 +1397,7 @@ DECLSPEC void sha384_hmac_init_128 (sha384_hmac_ctx_t *ctx, const u32 *w0, const
sha384_init (&ctx->opad);
sha384_transform (t0, t1, t2, t3, t4, t5, t6, t7, ctx->opad.h);
ctx->opad.len = 128;
sha384_update_128 (&ctx->opad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
}
DECLSPEC void sha384_hmac_init (sha384_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -2093,7 +2089,7 @@ DECLSPEC void sha384_update_vector_128 (sha384_ctx_vector_t *ctx, u32x *w0, u32x
if (len == 128)
{
sha384_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
sha384_transform_vector (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
ctx->w0[0] = 0;
ctx->w0[1] = 0;
@ -2858,9 +2854,7 @@ DECLSPEC void sha384_hmac_init_vector_128 (sha384_hmac_ctx_vector_t *ctx, const
sha384_init_vector (&ctx->ipad);
sha384_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, ctx->ipad.h);
ctx->ipad.len = 128;
sha384_update_vector_128 (&ctx->ipad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
// opad
@ -2899,9 +2893,7 @@ DECLSPEC void sha384_hmac_init_vector_128 (sha384_hmac_ctx_vector_t *ctx, const
sha384_init_vector (&ctx->opad);
sha384_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, ctx->opad.h);
ctx->opad.len = 128;
sha384_update_vector_128 (&ctx->opad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
}
DECLSPEC void sha384_hmac_init_vector (sha384_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

11
OpenCL/inc_hash_sha384.h
View File

@ -21,17 +21,10 @@
#define SHA384_F0(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA384_F1(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#ifdef IS_NV
#ifdef USE_BITSELECT
#define SHA384_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA384_F1o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_AMD
#define SHA384_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA384_F1o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_GENERIC
#else
#define SHA384_F0o(x,y,z) (SHA384_F0 ((x), (y), (z)))
#define SHA384_F1o(x,y,z) (SHA384_F1 ((x), (y), (z)))
#endif

16
OpenCL/inc_hash_sha512.cl
View File

@ -1358,9 +1358,7 @@ DECLSPEC void sha512_hmac_init_128 (sha512_hmac_ctx_t *ctx, const u32 *w0, const
sha512_init (&ctx->ipad);
sha512_transform (t0, t1, t2, t3, t4, t5, t6, t7, ctx->ipad.h);
ctx->ipad.len = 128;
sha512_update_128 (&ctx->ipad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
// opad
@ -1399,9 +1397,7 @@ DECLSPEC void sha512_hmac_init_128 (sha512_hmac_ctx_t *ctx, const u32 *w0, const
sha512_init (&ctx->opad);
sha512_transform (t0, t1, t2, t3, t4, t5, t6, t7, ctx->opad.h);
ctx->opad.len = 128;
sha512_update_128 (&ctx->opad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
}
DECLSPEC void sha512_hmac_init (sha512_hmac_ctx_t *ctx, const u32 *w, const int len)
@ -2975,9 +2971,7 @@ DECLSPEC void sha512_hmac_init_vector_128 (sha512_hmac_ctx_vector_t *ctx, const
sha512_init_vector (&ctx->ipad);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, ctx->ipad.h);
ctx->ipad.len = 128;
sha512_update_vector_128 (&ctx->ipad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
// opad
@ -3016,9 +3010,7 @@ DECLSPEC void sha512_hmac_init_vector_128 (sha512_hmac_ctx_vector_t *ctx, const
sha512_init_vector (&ctx->opad);
sha512_transform_vector (t0, t1, t2, t3, t4, t5, t6, t7, ctx->opad.h);
ctx->opad.len = 128;
sha512_update_vector_128 (&ctx->opad, t0, t1, t2, t3, t4, t5, t6, t7, 128);
}
DECLSPEC void sha512_hmac_init_vector (sha512_hmac_ctx_vector_t *ctx, const u32x *w, const int len)

11
OpenCL/inc_hash_sha512.h
View File

@ -21,17 +21,10 @@
#define SHA512_F0(x,y,z) ((z) ^ ((x) & ((y) ^ (z))))
#define SHA512_F1(x,y,z) (((x) & (y)) | ((z) & ((x) ^ (y))))
#ifdef IS_NV
#ifdef USE_BITSELECT
#define SHA512_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA512_F1o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_AMD
#define SHA512_F0o(x,y,z) (bitselect ((z), (y), (x)))
#define SHA512_F1o(x,y,z) (bitselect ((x), (y), ((x) ^ (z))))
#endif
#ifdef IS_GENERIC
#else
#define SHA512_F0o(x,y,z) (SHA512_F0 ((x), (y), (z)))
#define SHA512_F1o(x,y,z) (SHA512_F1 ((x), (y), (z)))
#endif

16
OpenCL/inc_hash_streebog256.cl
View File

@ -1106,9 +1106,7 @@ DECLSPEC void streebog256_hmac_init_64 (streebog256_hmac_ctx_t *ctx, const u32 *
streebog256_init (&ctx->ipad, s_sbob_sl64);
streebog256_transform (&ctx->ipad, t0, t1, t2, t3);
ctx->ipad.len = 64;
streebog256_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1131,9 +1129,7 @@ DECLSPEC void streebog256_hmac_init_64 (streebog256_hmac_ctx_t *ctx, const u32 *
streebog256_init (&ctx->opad, s_sbob_sl64);
streebog256_transform (&ctx->opad, t0, t1, t2, t3);
ctx->opad.len = 64;
streebog256_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void streebog256_hmac_init (streebog256_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64a (*s_sbob_sl64)[256])
@ -1752,9 +1748,7 @@ DECLSPEC void streebog256_hmac_init_vector_64 (streebog256_hmac_ctx_vector_t *ct
streebog256_init_vector (&ctx->ipad, s_sbob_sl64);
streebog256_transform_vector (&ctx->ipad, t0, t1, t2, t3);
ctx->ipad.len = 64;
streebog256_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1777,9 +1771,7 @@ DECLSPEC void streebog256_hmac_init_vector_64 (streebog256_hmac_ctx_vector_t *ct
streebog256_init_vector (&ctx->opad, s_sbob_sl64);
streebog256_transform_vector (&ctx->opad, t0, t1, t2, t3);
ctx->opad.len = 64;
streebog256_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void streebog256_hmac_init_vector (streebog256_hmac_ctx_vector_t *ctx, const u32x *w, const int len, SHM_TYPE u64a (*s_sbob_sl64)[256])

22
OpenCL/inc_hash_streebog512.cl
View File

@ -682,7 +682,7 @@ DECLSPEC void streebog512_g (u64 *h, const u64 *n, const u64 *m, SHM_TYPE u64a (
for (int i = 0; i < 8; i++)
{
k[i] = SBOG_LPSti64;
k[i] = SBOG_LPSti64_S;
}
#ifdef _unroll
@ -708,7 +708,7 @@ DECLSPEC void streebog512_g (u64 *h, const u64 *n, const u64 *m, SHM_TYPE u64a (
#endif
for (int i = 0; i < 8; i++)
{
s[i] = SBOG_LPSti64;
s[i] = SBOG_LPSti64_S;
}
for (int i = 0; i < 8; i++)
@ -721,7 +721,7 @@ DECLSPEC void streebog512_g (u64 *h, const u64 *n, const u64 *m, SHM_TYPE u64a (
#endif
for (int i = 0; i < 8; i++)
{
k[i] = SBOG_LPSti64;
k[i] = SBOG_LPSti64_S;
}
}
@ -1106,9 +1106,7 @@ DECLSPEC void streebog512_hmac_init_64 (streebog512_hmac_ctx_t *ctx, const u32 *
streebog512_init (&ctx->ipad, s_sbob_sl64);
streebog512_transform (&ctx->ipad, t0, t1, t2, t3);
ctx->ipad.len = 64;
streebog512_update_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1131,9 +1129,7 @@ DECLSPEC void streebog512_hmac_init_64 (streebog512_hmac_ctx_t *ctx, const u32 *
streebog512_init (&ctx->opad, s_sbob_sl64);
streebog512_transform (&ctx->opad, t0, t1, t2, t3);
ctx->opad.len = 64;
streebog512_update_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void streebog512_hmac_init (streebog512_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u64a (*s_sbob_sl64)[256])
@ -1771,9 +1767,7 @@ DECLSPEC void streebog512_hmac_init_vector_64 (streebog512_hmac_ctx_vector_t *ct
streebog512_init_vector (&ctx->ipad, s_sbob_sl64);
streebog512_transform_vector (&ctx->ipad, t0, t1, t2, t3);
ctx->ipad.len = 64;
streebog512_update_vector_64 (&ctx->ipad, t0, t1, t2, t3, 64);
// opad
@ -1796,9 +1790,7 @@ DECLSPEC void streebog512_hmac_init_vector_64 (streebog512_hmac_ctx_vector_t *ct
streebog512_init_vector (&ctx->opad, s_sbob_sl64);
streebog512_transform_vector (&ctx->opad, t0, t1, t2, t3);
ctx->opad.len = 64;
streebog512_update_vector_64 (&ctx->opad, t0, t1, t2, t3, 64);
}
DECLSPEC void streebog512_hmac_init_vector (streebog512_hmac_ctx_vector_t *ctx, const u32x *w, const int len, SHM_TYPE u64a (*s_sbob_sl64)[256])

20
OpenCL/inc_hash_streebog512.h
View File

@ -6,6 +6,8 @@
#ifndef _INC_HASH_STREEBOG512_H
#define _INC_HASH_STREEBOG512_H
#define BOX_S(S,n,i) ((S)[(n)][(i)])
#if VECT_SIZE == 1
#define BOX(S,n,i) ((S)[(n)][(i)])
@ -17,15 +19,25 @@
#elif VECT_SIZE == 8
#define BOX(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).s4], (S)[(n)][(i).s5], (S)[(n)][(i).s6], (S)[(n)][(i).s7])
#elif VECT_SIZE == 16
#define BOX(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])
(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])
#endif
#define SBOG_LPSti64_S \
BOX_S (s_sbob_sl64, 0, ((t[0] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 1, ((t[1] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 2, ((t[2] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 3, ((t[3] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 4, ((t[4] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 5, ((t[5] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 6, ((t[6] >> (i * 8)) & 0xff)) ^ \
BOX_S (s_sbob_sl64, 7, ((t[7] >> (i * 8)) & 0xff))
#define SBOG_LPSti64 \
BOX (s_sbob_sl64, 0, ((t[0] >> (i * 8)) & 0xff)) ^ \
BOX (s_sbob_sl64, 1, ((t[1] >> (i * 8)) & 0xff)) ^ \

2370
OpenCL/inc_hash_whirlpool.cl
View File

File diff suppressed because it is too large Load Diff

62
OpenCL/inc_hash_whirlpool.h
View File

@ -6,21 +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 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 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 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 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 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 BOX_S(S,i) (S)[(i)]
#define BOX64_S(S,i) (S)[(i)]
typedef struct whirlpool_ctx
{
@ -33,8 +37,14 @@ typedef struct whirlpool_ctx
int len;
SHM_TYPE u32 (*s_Ch)[256];
SHM_TYPE u32 (*s_Cl)[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;
@ -56,8 +66,14 @@ typedef struct whirlpool_ctx_vector
int len;
SHM_TYPE u32 (*s_Ch)[256];
SHM_TYPE u32 (*s_Cl)[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;
@ -68,8 +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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_init (whirlpool_ctx_t *ctx, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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);
@ -80,11 +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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_hmac_init (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_hmac_init_swap (whirlpool_hmac_ctx_t *ctx, const u32 *w, const int len, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_hmac_init_global (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_hmac_init_global_swap (whirlpool_hmac_ctx_t *ctx, GLOBAL_AS const u32 *w, const int len, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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);
@ -95,8 +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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_init_vector (whirlpool_ctx_vector_t *ctx, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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);
@ -104,8 +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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[256]);
DECLSPEC void whirlpool_hmac_init_vector (whirlpool_hmac_ctx_vector_t *ctx, const u32x *w, const int len, SHM_TYPE u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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);

13
OpenCL/inc_rp.cl
View File

@ -298,9 +298,17 @@ DECLSPEC int mangle_dupeword_times (MAYBE_UNUSED const u8 p0, MAYBE_UNUSED const
if (out_len >= RP_PASSWORD_SIZE) return (len);
u8 *out = buf + len;
int out_pos = len;
for (int t = 0; t < p0; t++) for (int i = 0; i < len; i++) *out++ = *buf++;
for (int t = 0; t < p0; t++)
{
for (int i = 0; i < len; i++)
{
buf[out_pos] = buf[i];
out_pos++;
}
}
return (out_len);
}
@ -769,6 +777,7 @@ DECLSPEC int apply_rules (CONSTANT_AS const u32 *cmds, u32 *buf, const int in_le
const u8 p0 = (cmd >> 8) & 0xff;
const u8 p1 = (cmd >> 16) & 0xff;
// we need to guarantee input length < 256 otherwise functions like rule_op_mangle_switch_last() and others will read out of boundary
out_len = apply_rule (name, p0, p1, buf, out_len);
}

1
OpenCL/inc_rp_optimized.cl
View File

@ -2349,6 +2349,7 @@ DECLSPEC u32 apply_rules_optimized (CONSTANT_AS const u32 *cmds, u32 *buf0, u32
const u32 p0 = (cmd >> 8) & 0xff;
const u32 p1 = (cmd >> 16) & 0xff;
// we need to guarantee input length < 32 otherwise functions like rule_op_mangle_switch_last() and others will read out of boundary
out_len = apply_rule_optimized (name, p0, p1, buf0, buf1, out_len);
}

48
OpenCL/inc_vendor.h
View File

@ -88,6 +88,10 @@
#define IS_GENERIC
#endif
#if defined IS_AMD && HAS_VPERM == 1
#define IS_ROCM
#endif
#define LOCAL_MEM_TYPE_LOCAL 1
#define LOCAL_MEM_TYPE_GLOBAL 2
@ -95,6 +99,11 @@
#define REAL_SHM
#endif
// So far, only used by -m 22100 and only affects NVIDIA on OpenCL. CUDA seems to work fine.
#ifdef FORCE_DISABLE_SHM
#undef REAL_SHM
#endif
#ifdef REAL_SHM
#define SHM_TYPE LOCAL_AS
#else
@ -106,17 +115,11 @@
* fast but pure kernels on rocm is a good example
*/
#if defined IS_CPU
#define DECLSPEC inline
#elif defined IS_GPU
#if defined IS_AMD
#if defined IS_AMD && defined IS_GPU
#define DECLSPEC inline static
#else
#define DECLSPEC
#endif
#else
#define DECLSPEC
#endif
/**
* AMD specific
@ -131,17 +134,30 @@
#endif
#endif
/**
* Unrolling is generally enabled, for all device types and hash modes
* There's a few exception when it's better not to unroll
* Some algorithms run into too much register pressure due to loop unrolling
*/
// Whitelist some OpenCL specific functions
// This could create more stable kernels on systems with bad OpenCL drivers
// generic vendors: those algos have shown that they produce better results on both amd and nv when not unrolled
// so we can assume they will produce better results on other vendors as well
#ifdef IS_CUDA
#define USE_BITSELECT
#define USE_ROTATE
#endif
#ifdef NO_UNROLL
#undef _unroll
#ifdef IS_ROCM
#define USE_BITSELECT
#define USE_ROTATE
#endif
#ifdef IS_INTEL_SDK
#ifdef IS_CPU
//#define USE_BITSELECT
//#define USE_ROTATE
#endif
#endif
#ifdef IS_OPENCL
//#define USE_BITSELECT
//#define USE_ROTATE
//#define USE_SWIZZLE
#endif
#endif

41
OpenCL/inc_zip_inflate.cl
View File

@ -209,6 +209,15 @@ DECLSPEC void *memset(u8 *s, int c, u32 len){
#define TINFL_MEMSET(p, c, l) memset(p, c, (u32)l)
#define MZ_CLEAR_OBJ(obj) memset(&(obj), 0, sizeof(obj))
// hashcat-patched/hashcat-specific:
#ifdef CRC32_IN_INFLATE
#define M_DICT_SIZE 1
#define MAYBE_GLOBAL GLOBAL_AS
#else
#define M_DICT_SIZE TINFL_LZ_DICT_SIZE
#define MAYBE_GLOBAL
#endif
#define TINFL_CR_FINISH }
#define TINFL_CR_BEGIN \
switch (r->m_state) \
@ -411,14 +420,16 @@ typedef struct
tinfl_decompressor m_decomp;
mz_uint m_dict_ofs, m_dict_avail, m_first_call, m_has_flushed;
int m_window_bits;
mz_uint8 m_dict[1]; // hashcat-patched: we do not need m_dict because we have our own output buffer
// hashcat-patched: we do not need m_dict in case of CRC32 checksums,
// because we have our own output buffer:
mz_uint8 m_dict[M_DICT_SIZE];
tinfl_status m_last_status;
} inflate_state;
typedef struct mz_stream_s
{
GLOBAL_AS const unsigned char *next_in; /* pointer to next byte to read */
MAYBE_GLOBAL const unsigned char *next_in; /* pointer to next byte to read */
unsigned int avail_in; /* number of bytes available at next_in */
mz_ulong total_in; /* total number of bytes consumed so far */
@ -457,9 +468,10 @@ DECLSPEC int mz_inflateEnd(mz_streamp pStream);
DECLSPEC int mz_inflateInit2(mz_streamp pStream, int window_bits, inflate_state*);
// hashcat-patched/hashcat-specific:
DECLSPEC const mz_uint8 pIn_xor_byte (const mz_uint8 c, mz_streamp pStream)
{
#ifdef CRC32_IN_INFLATE
mz_uint8 r = c;
u32 key3;
@ -469,18 +481,21 @@ DECLSPEC const mz_uint8 pIn_xor_byte (const mz_uint8 c, mz_streamp pStream)
update_key012 (pStream->key0, pStream->key1, pStream->key2, plain, pStream->crc32tab);
return (mz_uint8) plain;
#else
return c;
#endif
}
DECLSPEC void memcpy_g(void *dest, GLOBAL_AS const void *src, size_t n, mz_streamp pStream){
GLOBAL_AS char *csrc = (GLOBAL_AS char *)src;
DECLSPEC void memcpy_g(void *dest, MAYBE_GLOBAL const void *src, size_t n, mz_streamp pStream){
MAYBE_GLOBAL char *csrc = (MAYBE_GLOBAL char *)src;
char *cdest = (char *)dest;
for (int i=0; i<n; i++){
cdest[i] = pIn_xor_byte (csrc[i], pStream);
}
}
DECLSPEC tinfl_status tinfl_decompress(tinfl_decompressor *r, GLOBAL_AS const mz_uint8 *pIn_buf_next, size_t *pIn_buf_size, mz_uint8 *pOut_buf_start, mz_uint8 *pOut_buf_next, size_t *pOut_buf_size, const mz_uint32 decomp_flags, mz_streamp pStream)
DECLSPEC tinfl_status tinfl_decompress(tinfl_decompressor *r, MAYBE_GLOBAL const mz_uint8 *pIn_buf_next, size_t *pIn_buf_size, mz_uint8 *pOut_buf_start, mz_uint8 *pOut_buf_next, size_t *pOut_buf_size, const mz_uint32 decomp_flags, mz_streamp pStream)
{
const int s_length_base[31] = { 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0 };
@ -493,8 +508,8 @@ DECLSPEC tinfl_status tinfl_decompress(tinfl_decompressor *r, GLOBAL_AS const mz
tinfl_status status = TINFL_STATUS_FAILED;
mz_uint32 num_bits, dist, counter, num_extra;
tinfl_bit_buf_t bit_buf;
GLOBAL_AS const mz_uint8 *pIn_buf_cur = pIn_buf_next;
GLOBAL_AS const mz_uint8 *pIn_buf_end = pIn_buf_next + *pIn_buf_size;
MAYBE_GLOBAL const mz_uint8 *pIn_buf_cur = pIn_buf_next;
MAYBE_GLOBAL const mz_uint8 *pIn_buf_end = pIn_buf_next + *pIn_buf_size;
mz_uint8 *pOut_buf_cur = pOut_buf_next, *const pOut_buf_end = pOut_buf_next + *pOut_buf_size;
size_t out_buf_size_mask = (decomp_flags & TINFL_FLAG_USING_NON_WRAPPING_OUTPUT_BUF) ? (size_t)-1 : ((pOut_buf_next - pOut_buf_start) + *pOut_buf_size) - 1, dist_from_out_buf_start;
@ -851,7 +866,7 @@ DECLSPEC tinfl_status tinfl_decompress(tinfl_decompressor *r, GLOBAL_AS const mz
pOut_buf_cur[2] = pSrc[2];
pOut_buf_cur += 3;
pSrc += 3;
counter -= 3;
counter -= 3;
}
if (counter > 0)
{
@ -1008,10 +1023,12 @@ DECLSPEC int mz_inflate(mz_streamp pStream, int flush)
out_bytes = pStream->avail_out;
status = tinfl_decompress(&pState->m_decomp, pStream->next_in, &in_bytes, pStream->next_out, pStream->next_out, &out_bytes, decomp_flags, pStream);
#ifdef CRC32_IN_INFLATE
for (int i = 0; i < out_bytes; i++)
{
pStream->crc32 = CRC32 (pStream->crc32, pStream->next_out[i], pStream->crc32tab);
}
#endif
pState->m_last_status = status;
pStream->next_in += (mz_uint)in_bytes;
@ -1040,10 +1057,12 @@ DECLSPEC int mz_inflate(mz_streamp pStream, int flush)
n = MZ_MIN(pState->m_dict_avail, pStream->avail_out);
memcpy(pStream->next_out, pState->m_dict + pState->m_dict_ofs, n);
#ifdef CRC32_IN_INFLATE
for (int i = 0; i < n; i++)
{
pStream->crc32 = CRC32 (pStream->crc32, pStream->next_out[i], pStream->crc32tab);
}
#endif
//pStream->next_out += n;
//pStream->avail_out -= n;
@ -1072,10 +1091,12 @@ DECLSPEC int mz_inflate(mz_streamp pStream, int flush)
n = MZ_MIN(pState->m_dict_avail, pStream->avail_out);
memcpy(pStream->next_out, pState->m_dict + pState->m_dict_ofs, n);
#ifdef CRC32_IN_INFLATE
for (int i = 0; i < n; i++)
{
pStream->crc32 = CRC32 (pStream->crc32, pStream->next_out[i], pStream->crc32tab);
}
#endif
//pStream->next_out += n;
//pStream->avail_out -= n;
@ -1158,10 +1179,12 @@ DECLSPEC int hc_inflate (mz_streamp pStream)
tinfl_status status = tinfl_decompress (&pState->m_decomp, pStream->next_in, &in_bytes, pStream->next_out, pStream->next_out + pStream->total_out, &out_bytes, decomp_flags, pStream);
#ifdef CRC32_IN_INFLATE
for (int i = 0; i < out_bytes; i++)
{
pStream->crc32 = CRC32 (pStream->crc32, pStream->next_out[pStream->total_out + i], pStream->crc32tab);
}
#endif
pStream->next_in += (mz_uint) in_bytes;
pStream->avail_in -= (mz_uint) in_bytes;

40
OpenCL/m00200_a0-optimized.cl
View File

@ -99,10 +99,10 @@ KERNEL_FQ void m00200_m04 (KERN_ATTR_RULES ())
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
ROUND (unpack_v8d_from_v32 (wj));
}
const u32x wj = w_t[j];
@ -111,18 +111,18 @@ KERNEL_FQ void m00200_m04 (KERN_ATTR_RULES ())
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
}
a &= 0x7fffffff;
@ -237,10 +237,10 @@ KERNEL_FQ void m00200_s04 (KERN_ATTR_RULES ())
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
ROUND (unpack_v8d_from_v32 (wj));
}
const u32x wj = w_t[j];
@ -249,18 +249,18 @@ KERNEL_FQ void m00200_s04 (KERN_ATTR_RULES ())
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
}
a &= 0x7fffffff;

42
OpenCL/m00200_a1-optimized.cl
View File

@ -142,6 +142,8 @@ KERNEL_FQ void m00200_m04 (KERN_ATTR_BASIC ())
u32x a = MYSQL323_A;
u32x b = MYSQL323_B;
u32x c = 0;
u32x d = 0;
u32x add = 7;
@ -159,10 +161,10 @@ KERNEL_FQ void m00200_m04 (KERN_ATTR_BASIC ())
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
ROUND (unpack_v8d_from_v32 (wj));
}
const u32x wj = w_t[j];
@ -171,18 +173,18 @@ KERNEL_FQ void m00200_m04 (KERN_ATTR_BASIC ())
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
}
a &= 0x7fffffff;
@ -361,10 +363,10 @@ KERNEL_FQ void m00200_s04 (KERN_ATTR_BASIC ())
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
ROUND (unpack_v8d_from_v32 (wj));
}
const u32x wj = w_t[j];
@ -373,18 +375,18 @@ KERNEL_FQ void m00200_s04 (KERN_ATTR_BASIC ())
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
ROUND (unpack_v8c_from_v32 (wj));
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
ROUND (unpack_v8b_from_v32 (wj));
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
ROUND (unpack_v8a_from_v32 (wj));
}
a &= 0x7fffffff;

272
OpenCL/m00200_a3-optimized.cl
View File

@ -42,10 +42,10 @@
{ \
const u32 wj = w[j]; \
\
ROUND ((wj >> 0) & 0xff); \
ROUND ((wj >> 8) & 0xff); \
ROUND ((wj >> 16) & 0xff); \
ROUND ((wj >> 24) & 0xff); \
ROUND (unpack_v8a_from_v32 (wj)); \
ROUND (unpack_v8b_from_v32 (wj)); \
ROUND (unpack_v8c_from_v32 (wj)); \
ROUND (unpack_v8d_from_v32 (wj)); \
} \
\
const u32 wj = w[j]; \
@ -54,18 +54,18 @@
\
if (left == 3) \
{ \
ROUND ((wj >> 0) & 0xff); \
ROUND ((wj >> 8) & 0xff); \
ROUND ((wj >> 16) & 0xff); \
ROUND (unpack_v8a_from_v32 (wj)); \
ROUND (unpack_v8b_from_v32 (wj)); \
ROUND (unpack_v8c_from_v32 (wj)); \
} \
else if (left == 2) \
{ \
ROUND ((wj >> 0) & 0xff); \
ROUND ((wj >> 8) & 0xff); \
ROUND (unpack_v8a_from_v32 (wj)); \
ROUND (unpack_v8b_from_v32 (wj)); \
} \
else if (left == 1) \
{ \
ROUND ((wj >> 0) & 0xff); \
ROUND (unpack_v8a_from_v32 (wj)); \
}
#define CODE_POST_M \
@ -99,141 +99,123 @@ DECLSPEC void m00200m (u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* digest
*/
const u32 search[4] =
{
digests_buf[digests_offset].digest_buf[DGST_R0],
digests_buf[digests_offset].digest_buf[DGST_R1],
0,
0
};
/**
* loop
*/
u32 w0l = w[0];
CODE_PRE;
switch (pw_len)
{
case 1:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0));
break;
case 2:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0));
break;
case 3:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0));
break;
case 4:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
break;
case 5:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1]));
break;
case 6:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1]));
break;
case 7:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1]));
break;
case 8:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
break;
case 9:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2]));
break;
case 10:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2]));
break;
case 11:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2]));
break;
case 12:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
break;
case 13:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3]));
break;
case 14:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3]));
break;
case 15:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff); ROUND ((w[3] >> 16) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3])); ROUND (unpack_v8c_from_v32 (w[3]));
break;
case 16:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff); ROUND ((w[3] >> 16) & 0xff); ROUND ((w[3] >> 24) & 0xff);
CODE_POST_M;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3])); ROUND (unpack_v8c_from_v32 (w[3])); ROUND (unpack_v8d_from_v32 (w[3]));
break;
default:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
CODE_LOOP (pw_len - 4);
CODE_POST_M;
break;
}
CODE_POST_M;
}
DECLSPEC void m00200s (u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
@ -263,135 +245,105 @@ DECLSPEC void m00200s (u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
u32 w0l = w[0];
CODE_PRE;
switch (pw_len)
{
case 1:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0));
break;
case 2:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0));
break;
case 3:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0));
break;
case 4:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
break;
case 5:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1]));
break;
case 6:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1]));
break;
case 7:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1]));
break;
case 8:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
break;
case 9:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2]));
break;
case 10:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2]));
break;
case 11:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2]));
break;
case 12:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
break;
case 13:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3]));
break;
case 14:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3]));
break;
case 15:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff); ROUND ((w[3] >> 16) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3])); ROUND (unpack_v8c_from_v32 (w[3]));
break;
case 16:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND ((w[1] >> 0) & 0xff); ROUND ((w[1] >> 8) & 0xff); ROUND ((w[1] >> 16) & 0xff); ROUND ((w[1] >> 24) & 0xff);
ROUND ((w[2] >> 0) & 0xff); ROUND ((w[2] >> 8) & 0xff); ROUND ((w[2] >> 16) & 0xff); ROUND ((w[2] >> 24) & 0xff);
ROUND ((w[3] >> 0) & 0xff); ROUND ((w[3] >> 8) & 0xff); ROUND ((w[3] >> 16) & 0xff); ROUND ((w[3] >> 24) & 0xff);
CODE_POST_S;
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
ROUND (unpack_v8a_from_v32 (w[1])); ROUND (unpack_v8b_from_v32 (w[1])); ROUND (unpack_v8c_from_v32 (w[1])); ROUND (unpack_v8d_from_v32 (w[1]));
ROUND (unpack_v8a_from_v32 (w[2])); ROUND (unpack_v8b_from_v32 (w[2])); ROUND (unpack_v8c_from_v32 (w[2])); ROUND (unpack_v8d_from_v32 (w[2]));
ROUND (unpack_v8a_from_v32 (w[3])); ROUND (unpack_v8b_from_v32 (w[3])); ROUND (unpack_v8c_from_v32 (w[3])); ROUND (unpack_v8d_from_v32 (w[3]));
break;
default:
CODE_PRE;
ROUND ((w0 >> 0) & 0xff); ROUND ((w0 >> 8) & 0xff); ROUND ((w0 >> 16) & 0xff); ROUND ((w0 >> 24) & 0xff);
ROUND (unpack_v8a_from_v32 ( w0)); ROUND (unpack_v8b_from_v32 ( w0)); ROUND (unpack_v8c_from_v32 ( w0)); ROUND (unpack_v8d_from_v32 ( w0));
CODE_LOOP (pw_len - 4);
CODE_POST_S;
break;
}
CODE_POST_S;
}
KERNEL_FQ void m00200_m04 (KERN_ATTR_VECTOR ())

80
OpenCL/m00600_a0-optimized.cl
View File

@ -28,28 +28,28 @@ typedef struct blake2
#define BLAKE2B_FINAL 1
#define BLAKE2B_UPDATE 0
#define BLAKE2B_G(r,i,a,b,c,d) \
do { \
a = a + b + m[blake2b_sigma[r][2*i+0]]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[blake2b_sigma[r][2*i+1]]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
#define BLAKE2B_G(k0,k1,a,b,c,d) \
do { \
a = a + b + m[(k0)]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[(k1)]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
} while(0)
#define BLAKE2B_ROUND(r) \
do { \
BLAKE2B_G (r,0,v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G (r,1,v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G (r,2,v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G (r,3,v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G (r,4,v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G (r,5,v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G (r,6,v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G (r,7,v[ 3],v[ 4],v[ 9],v[14]); \
#define BLAKE2B_ROUND(c0,c1,c2,c3,c4,c5,c6,c7,c8,c9,ca,cb,cc,cd,ce,cf) \
do { \
BLAKE2B_G ((c0),(c1),v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G ((c2),(c3),v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G ((c4),(c5),v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G ((c6),(c7),v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G ((c8),(c9),v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G ((ca),(cb),v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G ((cc),(cd),v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G ((ce),(cf),v[ 3],v[ 4],v[ 9],v[14]); \
} while(0)
DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, const u32x out_len, const u8 isFinal)
@ -93,34 +93,18 @@ DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, co
v[14] = BLAKE2B_IV_06 ^ f[0];
v[15] = BLAKE2B_IV_07 ^ f[1];
const int blake2b_sigma[12][16] =
{
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 } ,
{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 } ,
{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 } ,
{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 } ,
{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 } ,
{ 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 } ,
{ 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 } ,
{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 } ,
{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0 } ,
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }
};
BLAKE2B_ROUND ( 0);
BLAKE2B_ROUND ( 1);
BLAKE2B_ROUND ( 2);
BLAKE2B_ROUND ( 3);
BLAKE2B_ROUND ( 4);
BLAKE2B_ROUND ( 5);
BLAKE2B_ROUND ( 6);
BLAKE2B_ROUND ( 7);
BLAKE2B_ROUND ( 8);
BLAKE2B_ROUND ( 9);
BLAKE2B_ROUND (10);
BLAKE2B_ROUND (11);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
BLAKE2B_ROUND (11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4);
BLAKE2B_ROUND ( 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8);
BLAKE2B_ROUND ( 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13);
BLAKE2B_ROUND ( 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9);
BLAKE2B_ROUND (12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11);
BLAKE2B_ROUND (13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10);
BLAKE2B_ROUND ( 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5);
BLAKE2B_ROUND (10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
h[0] = h[0] ^ v[0] ^ v[ 8];
h[1] = h[1] ^ v[1] ^ v[ 9];

80
OpenCL/m00600_a1-optimized.cl
View File

@ -26,28 +26,28 @@ typedef struct blake2
#define BLAKE2B_FINAL 1
#define BLAKE2B_UPDATE 0
#define BLAKE2B_G(r,i,a,b,c,d) \
do { \
a = a + b + m[blake2b_sigma[r][2*i+0]]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[blake2b_sigma[r][2*i+1]]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
#define BLAKE2B_G(k0,k1,a,b,c,d) \
do { \
a = a + b + m[(k0)]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[(k1)]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
} while(0)
#define BLAKE2B_ROUND(r) \
do { \
BLAKE2B_G (r,0,v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G (r,1,v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G (r,2,v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G (r,3,v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G (r,4,v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G (r,5,v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G (r,6,v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G (r,7,v[ 3],v[ 4],v[ 9],v[14]); \
#define BLAKE2B_ROUND(c0,c1,c2,c3,c4,c5,c6,c7,c8,c9,ca,cb,cc,cd,ce,cf) \
do { \
BLAKE2B_G ((c0),(c1),v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G ((c2),(c3),v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G ((c4),(c5),v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G ((c6),(c7),v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G ((c8),(c9),v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G ((ca),(cb),v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G ((cc),(cd),v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G ((ce),(cf),v[ 3],v[ 4],v[ 9],v[14]); \
} while(0)
DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, const u32x out_len, const u8 isFinal)
@ -91,34 +91,18 @@ DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, co
v[14] = BLAKE2B_IV_06 ^ f[0];
v[15] = BLAKE2B_IV_07 ^ f[1];
const int blake2b_sigma[12][16] =
{
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 } ,
{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 } ,
{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 } ,
{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 } ,
{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 } ,
{ 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 } ,
{ 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 } ,
{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 } ,
{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0 } ,
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }
};
BLAKE2B_ROUND ( 0);
BLAKE2B_ROUND ( 1);
BLAKE2B_ROUND ( 2);
BLAKE2B_ROUND ( 3);
BLAKE2B_ROUND ( 4);
BLAKE2B_ROUND ( 5);
BLAKE2B_ROUND ( 6);
BLAKE2B_ROUND ( 7);
BLAKE2B_ROUND ( 8);
BLAKE2B_ROUND ( 9);
BLAKE2B_ROUND (10);
BLAKE2B_ROUND (11);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
BLAKE2B_ROUND (11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4);
BLAKE2B_ROUND ( 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8);
BLAKE2B_ROUND ( 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13);
BLAKE2B_ROUND ( 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9);
BLAKE2B_ROUND (12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11);
BLAKE2B_ROUND (13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10);
BLAKE2B_ROUND ( 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5);
BLAKE2B_ROUND (10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
h[0] = h[0] ^ v[0] ^ v[ 8];
h[1] = h[1] ^ v[1] ^ v[ 9];

80
OpenCL/m00600_a3-optimized.cl
View File

@ -26,28 +26,28 @@ typedef struct blake2
#define BLAKE2B_FINAL 1
#define BLAKE2B_UPDATE 0
#define BLAKE2B_G(r,i,a,b,c,d) \
do { \
a = a + b + m[blake2b_sigma[r][2*i+0]]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[blake2b_sigma[r][2*i+1]]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
#define BLAKE2B_G(k0,k1,a,b,c,d) \
do { \
a = a + b + m[(k0)]; \
d = hc_rotr64 (d ^ a, 32); \
c = c + d; \
b = hc_rotr64 (b ^ c, 24); \
a = a + b + m[(k1)]; \
d = hc_rotr64 (d ^ a, 16); \
c = c + d; \
b = hc_rotr64 (b ^ c, 63); \
} while(0)
#define BLAKE2B_ROUND(r) \
do { \
BLAKE2B_G (r,0,v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G (r,1,v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G (r,2,v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G (r,3,v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G (r,4,v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G (r,5,v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G (r,6,v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G (r,7,v[ 3],v[ 4],v[ 9],v[14]); \
#define BLAKE2B_ROUND(c0,c1,c2,c3,c4,c5,c6,c7,c8,c9,ca,cb,cc,cd,ce,cf) \
do { \
BLAKE2B_G ((c0),(c1),v[ 0],v[ 4],v[ 8],v[12]); \
BLAKE2B_G ((c2),(c3),v[ 1],v[ 5],v[ 9],v[13]); \
BLAKE2B_G ((c4),(c5),v[ 2],v[ 6],v[10],v[14]); \
BLAKE2B_G ((c6),(c7),v[ 3],v[ 7],v[11],v[15]); \
BLAKE2B_G ((c8),(c9),v[ 0],v[ 5],v[10],v[15]); \
BLAKE2B_G ((ca),(cb),v[ 1],v[ 6],v[11],v[12]); \
BLAKE2B_G ((cc),(cd),v[ 2],v[ 7],v[ 8],v[13]); \
BLAKE2B_G ((ce),(cf),v[ 3],v[ 4],v[ 9],v[14]); \
} while(0)
DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, const u32x *w0, const u32x *w1, const u32x *w2, const u32x *w3, const u32x out_len, const u8 isFinal)
@ -91,34 +91,18 @@ DECLSPEC void blake2b_transform (u64x *h, u64x *t, u64x *f, u64x *m, u64x *v, co
v[14] = BLAKE2B_IV_06 ^ f[0];
v[15] = BLAKE2B_IV_07 ^ f[1];
const int blake2b_sigma[12][16] =
{
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 } ,
{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 } ,
{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 } ,
{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 } ,
{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 } ,
{ 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 } ,
{ 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 } ,
{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 } ,
{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0 } ,
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 } ,
{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 }
};
BLAKE2B_ROUND ( 0);
BLAKE2B_ROUND ( 1);
BLAKE2B_ROUND ( 2);
BLAKE2B_ROUND ( 3);
BLAKE2B_ROUND ( 4);
BLAKE2B_ROUND ( 5);
BLAKE2B_ROUND ( 6);
BLAKE2B_ROUND ( 7);
BLAKE2B_ROUND ( 8);
BLAKE2B_ROUND ( 9);
BLAKE2B_ROUND (10);
BLAKE2B_ROUND (11);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
BLAKE2B_ROUND (11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4);
BLAKE2B_ROUND ( 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8);
BLAKE2B_ROUND ( 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13);
BLAKE2B_ROUND ( 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9);
BLAKE2B_ROUND (12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11);
BLAKE2B_ROUND (13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10);
BLAKE2B_ROUND ( 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5);
BLAKE2B_ROUND (10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13 , 0);
BLAKE2B_ROUND ( 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
BLAKE2B_ROUND (14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3);
h[0] = h[0] ^ v[0] ^ v[ 8];
h[1] = h[1] ^ v[1] ^ v[ 9];

1109
OpenCL/m01500_a3-pure.cl
View File

File diff suppressed because it is too large Load Diff

1143
OpenCL/m03000_a3-pure.cl
View File

File diff suppressed because it is too large Load Diff

25
OpenCL/m03200-pure.cl
View File

@ -356,6 +356,10 @@ CONSTANT_VK u32a c_pbox[18] =
L ^= P[17]; \
}
#ifdef DYNAMIC_LOCAL
extern __shared__ u32 lm[];
#endif
DECLSPEC void expand_key (u32 *E, u32 *W, const int len)
{
u8 *E_ptr = (u8 *) E;
@ -456,6 +460,12 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_init (KERN_ATTR_TMPS
P[i] = c_pbox[i];
}
#ifdef DYNAMIC_LOCAL
LOCAL_AS u32 *S0 = lm + (lid * 1024) + 0;
LOCAL_AS u32 *S1 = lm + (lid * 1024) + 256;
LOCAL_AS u32 *S2 = lm + (lid * 1024) + 512;
LOCAL_AS u32 *S3 = lm + (lid * 1024) + 768;
#else
LOCAL_VK u32 S0_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S1_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S2_all[FIXED_LOCAL_SIZE][256];
@ -465,6 +475,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_init (KERN_ATTR_TMPS
LOCAL_AS u32 *S1 = S1_all[lid];
LOCAL_AS u32 *S2 = S2_all[lid];
LOCAL_AS u32 *S3 = S3_all[lid];
#endif
for (u32 i = 0; i < 256; i++)
{
@ -614,6 +625,12 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_loop (KERN_ATTR_TMPS
P[i] = tmps[gid].P[i];
}
#ifdef DYNAMIC_LOCAL
LOCAL_AS u32 *S0 = lm + (lid * 1024) + 0;
LOCAL_AS u32 *S1 = lm + (lid * 1024) + 256;
LOCAL_AS u32 *S2 = lm + (lid * 1024) + 512;
LOCAL_AS u32 *S3 = lm + (lid * 1024) + 768;
#else
LOCAL_VK u32 S0_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S1_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S2_all[FIXED_LOCAL_SIZE][256];
@ -623,6 +640,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_loop (KERN_ATTR_TMPS
LOCAL_AS u32 *S1 = S1_all[lid];
LOCAL_AS u32 *S2 = S2_all[lid];
LOCAL_AS u32 *S3 = S3_all[lid];
#endif
for (u32 i = 0; i < 256; i++)
{
@ -799,6 +817,12 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_comp (KERN_ATTR_TMPS
P[i] = tmps[gid].P[i];
}
#ifdef DYNAMIC_LOCAL
LOCAL_AS u32 *S0 = lm + (lid * 1024) + 0;
LOCAL_AS u32 *S1 = lm + (lid * 1024) + 256;
LOCAL_AS u32 *S2 = lm + (lid * 1024) + 512;
LOCAL_AS u32 *S3 = lm + (lid * 1024) + 768;
#else
LOCAL_VK u32 S0_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S1_all[FIXED_LOCAL_SIZE][256];
LOCAL_VK u32 S2_all[FIXED_LOCAL_SIZE][256];
@ -808,6 +832,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m03200_comp (KERN_ATTR_TMPS
LOCAL_AS u32 *S1 = S1_all[lid];
LOCAL_AS u32 *S2 = S2_all[lid];
LOCAL_AS u32 *S3 = S3_all[lid];
#endif
for (u32 i = 0; i < 256; i++)
{

22
OpenCL/m05500_a0-optimized.cl
View File

@ -356,7 +356,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32x r = data[0];
u32x l = data[1];
@ -398,7 +398,7 @@ DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCA
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32x tt;
@ -516,6 +516,8 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_RULES ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -542,6 +544,13 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_RULES ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -729,6 +738,8 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_RULES ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -755,6 +766,13 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_RULES ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

22
OpenCL/m05500_a0-pure.cl
View File

@ -356,7 +356,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32 r = data[0];
u32 l = data[1];
@ -398,7 +398,7 @@ DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, LOCAL_AS
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32 c, u32 d, u32 *Kc, u32 *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32 c, u32 d, u32 *Kc, u32 *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32 tt;
@ -516,6 +516,8 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_RULES ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -542,6 +544,13 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_RULES ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -640,6 +649,8 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_RULES ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -666,6 +677,13 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_RULES ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

22
OpenCL/m05500_a1-optimized.cl
View File

@ -354,7 +354,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32x r = data[0];
u32x l = data[1];
@ -396,7 +396,7 @@ DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCA
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32x tt;
@ -514,6 +514,8 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_BASIC ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -540,6 +542,13 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_BASIC ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -780,6 +789,8 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_BASIC ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -806,6 +817,13 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_BASIC ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

22
OpenCL/m05500_a1-pure.cl
View File

@ -354,7 +354,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32 r = data[0];
u32 l = data[1];
@ -396,7 +396,7 @@ DECLSPEC void _des_crypt_encrypt (u32 *iv, u32 *data, u32 *Kc, u32 *Kd, LOCAL_AS
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32 c, u32 d, u32 *Kc, u32 *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32 c, u32 d, u32 *Kc, u32 *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32 tt;
@ -514,6 +514,8 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_BASIC ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -540,6 +542,13 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_BASIC ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -636,6 +645,8 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_BASIC ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -662,6 +673,13 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_BASIC ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

62
OpenCL/m05500_a3-optimized.cl
View File

@ -354,7 +354,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32x r = data[0];
u32x l = data[1];
@ -396,7 +396,7 @@ DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCA
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32x tt;
@ -500,7 +500,7 @@ DECLSPEC void transform_netntlmv1_key (const u32x w0, const u32x w1, u32x *out)
| ((k[7] & 0xff) << 24);
}
DECLSPEC void m05500m (LOCAL_AS u32 (*s_SPtrans)[64], LOCAL_AS u32 (*s_skb)[64], u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
DECLSPEC void m05500m (SHM_TYPE u32 (*s_SPtrans)[64], SHM_TYPE u32 (*s_skb)[64], u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
{
/**
* modifier
@ -657,7 +657,7 @@ DECLSPEC void m05500m (LOCAL_AS u32 (*s_SPtrans)[64], LOCAL_AS u32 (*s_skb)[64],
}
}
DECLSPEC void m05500s (LOCAL_AS u32 (*s_SPtrans)[64], LOCAL_AS u32 (*s_skb)[64], u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
DECLSPEC void m05500s (SHM_TYPE u32 (*s_SPtrans)[64], SHM_TYPE u32 (*s_skb)[64], u32 *w, const u32 pw_len, KERN_ATTR_VECTOR ())
{
/**
* modifier
@ -847,6 +847,8 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -873,6 +875,13 @@ KERNEL_FQ void m05500_m04 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -921,6 +930,8 @@ KERNEL_FQ void m05500_m08 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -947,6 +958,13 @@ KERNEL_FQ void m05500_m08 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -995,6 +1013,8 @@ KERNEL_FQ void m05500_m16 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -1021,6 +1041,13 @@ KERNEL_FQ void m05500_m16 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -1069,6 +1096,8 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -1095,6 +1124,13 @@ KERNEL_FQ void m05500_s04 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -1143,6 +1179,8 @@ KERNEL_FQ void m05500_s08 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -1169,6 +1207,13 @@ KERNEL_FQ void m05500_s08 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -1217,6 +1262,8 @@ KERNEL_FQ void m05500_s16 (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -1243,6 +1290,13 @@ KERNEL_FQ void m05500_s16 (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

22
OpenCL/m05500_a3-pure.cl
View File

@ -354,7 +354,7 @@ CONSTANT_VK u32a c_skb[8][64] =
#define BOX(i,n,S) 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])
#endif
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_SPtrans)[64])
DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_SPtrans)[64])
{
u32x r = data[0];
u32x l = data[1];
@ -396,7 +396,7 @@ DECLSPEC void _des_crypt_encrypt (u32x *iv, u32x *data, u32x *Kc, u32x *Kd, LOCA
iv[1] = r;
}
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, LOCAL_AS u32 (*s_skb)[64])
DECLSPEC void _des_crypt_keysetup (u32x c, u32x d, u32x *Kc, u32x *Kd, SHM_TYPE u32 (*s_skb)[64])
{
u32x tt;
@ -514,6 +514,8 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -540,6 +542,13 @@ KERNEL_FQ void m05500_mxx (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**
@ -649,6 +658,8 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_VECTOR ())
* sbox, kbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_SPtrans[8][64];
LOCAL_VK u32 s_skb[8][64];
@ -675,6 +686,13 @@ KERNEL_FQ void m05500_sxx (KERN_ATTR_VECTOR ())
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_SPtrans)[64] = c_SPtrans;
CONSTANT_AS u32a (*s_skb)[64] = c_skb;
#endif
if (gid >= gid_max) return;
/**

98
OpenCL/m06100_a0-optimized.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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 ())
@ -37,36 +37,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -147,7 +150,7 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_RULES ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_Ch, s_Cl);
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]);
}
@ -177,36 +180,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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 +305,7 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_RULES ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_Ch, s_Cl);
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]);
}

94
OpenCL/m06100_a0-pure.cl
View File

@ -32,36 +32,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -85,7 +88,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_RULES ())
whirlpool_ctx_t ctx;
whirlpool_init (&ctx, s_Ch, s_Cl);
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);
@ -116,36 +119,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_RULES ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -181,7 +187,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_RULES ())
whirlpool_ctx_t ctx;
whirlpool_init (&ctx, s_Ch, s_Cl);
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);

98
OpenCL/m06100_a1-optimized.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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 ())
@ -35,36 +35,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -203,7 +206,7 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_Ch, s_Cl);
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]);
}
@ -233,36 +236,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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 +419,7 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
dgst[14] = 0;
dgst[15] = 0;
whirlpool_transform_transport_vector (w, dgst, s_Ch, s_Cl);
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]);
}

94
OpenCL/m06100_a1-pure.cl
View File

@ -30,36 +30,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -71,7 +74,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_BASIC ())
whirlpool_ctx_t ctx0;
whirlpool_init (&ctx0, s_Ch, s_Cl);
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);
@ -112,36 +115,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -165,7 +171,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_BASIC ())
whirlpool_ctx_t ctx0;
whirlpool_init (&ctx0, s_Ch, s_Cl);
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);

200
OpenCL/m06100_a3-optimized.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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 u32 (*s_Cl)[256], SHM_TYPE u32 (*s_Ch)[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_Ch, s_Cl);
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 u32 (*s_Cl)[256], SHM_TYPE u32 (*s_Ch)[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_Ch, s_Cl);
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]);
}
@ -185,36 +185,39 @@ KERNEL_FQ void m06100_m04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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 +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_Cl, s_Ch);
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 ())
@ -277,36 +280,39 @@ KERNEL_FQ void m06100_m08 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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 +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_Cl, s_Ch);
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 ())
@ -373,36 +379,39 @@ KERNEL_FQ void m06100_s04 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -446,7 +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_Cl, s_Ch);
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 ())
@ -465,36 +474,39 @@ KERNEL_FQ void m06100_s08 (KERN_ATTR_BASIC ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -538,7 +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_Cl, s_Ch);
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 ())

94
OpenCL/m06100_a3-pure.cl
View File

@ -30,36 +30,39 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_VECTOR ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -94,7 +97,7 @@ KERNEL_FQ void m06100_mxx (KERN_ATTR_VECTOR ())
whirlpool_ctx_vector_t ctx;
whirlpool_init_vector (&ctx, s_Ch, s_Cl);
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);
@ -125,36 +128,39 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_VECTOR ())
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -201,7 +207,7 @@ KERNEL_FQ void m06100_sxx (KERN_ATTR_VECTOR ())
whirlpool_ctx_vector_t ctx;
whirlpool_init_vector (&ctx, s_Ch, s_Cl);
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);

143
OpenCL/m06231-pure.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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))
@ -168,36 +168,39 @@ KERNEL_FQ void m06231_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -269,7 +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_Ch, s_Cl);
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];
@ -380,36 +383,39 @@ KERNEL_FQ void m06231_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -515,7 +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_Ch, s_Cl);
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];
@ -628,45 +634,6 @@ KERNEL_FQ void m06231_comp (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
#endif
if (gid >= gid_max) return;
u32 ukey1[8];

143
OpenCL/m06232-pure.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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))
@ -168,36 +168,39 @@ KERNEL_FQ void m06232_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -269,7 +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_Ch, s_Cl);
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];
@ -380,36 +383,39 @@ KERNEL_FQ void m06232_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -515,7 +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_Ch, s_Cl);
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];
@ -628,45 +634,6 @@ KERNEL_FQ void m06232_comp (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
#endif
if (gid >= gid_max) return;
u32 ukey1[8];

143
OpenCL/m06233-pure.cl
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 u32 (*s_Ch)[256], SHM_TYPE u32 (*s_Cl)[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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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_Ch, s_Cl);
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))
@ -168,36 +168,39 @@ KERNEL_FQ void m06233_init (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -269,7 +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_Ch, s_Cl);
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];
@ -380,36 +383,39 @@ KERNEL_FQ void m06233_loop (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[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_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[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 u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
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
@ -515,7 +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_Ch, s_Cl);
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];
@ -628,45 +634,6 @@ KERNEL_FQ void m06233_comp (KERN_ATTR_TMPS_ESALT (tc_tmp_t, tc_t))
#endif
/**
* Whirlpool shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_Ch[8][256];
LOCAL_VK u32 s_Cl[8][256];
for (u32 i = lid; i < 256; i += lsz)
{
s_Ch[0][i] = Ch[0][i];
s_Ch[1][i] = Ch[1][i];
s_Ch[2][i] = Ch[2][i];
s_Ch[3][i] = Ch[3][i];
s_Ch[4][i] = Ch[4][i];
s_Ch[5][i] = Ch[5][i];
s_Ch[6][i] = Ch[6][i];
s_Ch[7][i] = Ch[7][i];
s_Cl[0][i] = Cl[0][i];
s_Cl[1][i] = Cl[1][i];
s_Cl[2][i] = Cl[2][i];
s_Cl[3][i] = Cl[3][i];
s_Cl[4][i] = Cl[4][i];
s_Cl[5][i] = Cl[5][i];
s_Cl[6][i] = Cl[6][i];
s_Cl[7][i] = Cl[7][i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_Ch)[256] = Ch;
CONSTANT_AS u32a (*s_Cl)[256] = Cl;
#endif
if (gid >= gid_max) return;
u32 ukey1[8];

302
OpenCL/m06500-pure.cl
View File

@ -119,74 +119,69 @@ KERNEL_FQ void m06500_init (KERN_ATTR_TMPS (sha512aix_tmp_t))
sha512_hmac_update_global_swap (&sha512_hmac_ctx, salt_bufs[salt_pos].salt_buf, salt_bufs[salt_pos].salt_len);
for (u32 i = 0, j = 1; i < 8; i += 8, j += 1)
{
sha512_hmac_ctx_t sha512_hmac_ctx2 = sha512_hmac_ctx;
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
u32 w4[4];
u32 w5[4];
u32 w6[4];
u32 w7[4];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
u32 w4[4];
u32 w5[4];
u32 w6[4];
u32 w7[4];
w0[0] = 1;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
w4[0] = 0;
w4[1] = 0;
w4[2] = 0;
w4[3] = 0;
w5[0] = 0;
w5[1] = 0;
w5[2] = 0;
w5[3] = 0;
w6[0] = 0;
w6[1] = 0;
w6[2] = 0;
w6[3] = 0;
w7[0] = 0;
w7[1] = 0;
w7[2] = 0;
w7[3] = 0;
w0[0] = j;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
w4[0] = 0;
w4[1] = 0;
w4[2] = 0;
w4[3] = 0;
w5[0] = 0;
w5[1] = 0;
w5[2] = 0;
w5[3] = 0;
w6[0] = 0;
w6[1] = 0;
w6[2] = 0;
w6[3] = 0;
w7[0] = 0;
w7[1] = 0;
w7[2] = 0;
w7[3] = 0;
sha512_hmac_update_128 (&sha512_hmac_ctx, w0, w1, w2, w3, w4, w5, w6, w7, 4);
sha512_hmac_update_128 (&sha512_hmac_ctx2, w0, w1, w2, w3, w4, w5, w6, w7, 4);
sha512_hmac_final (&sha512_hmac_ctx);
sha512_hmac_final (&sha512_hmac_ctx2);
tmps[gid].dgst[0] = sha512_hmac_ctx.opad.h[0];
tmps[gid].dgst[1] = sha512_hmac_ctx.opad.h[1];
tmps[gid].dgst[2] = sha512_hmac_ctx.opad.h[2];
tmps[gid].dgst[3] = sha512_hmac_ctx.opad.h[3];
tmps[gid].dgst[4] = sha512_hmac_ctx.opad.h[4];
tmps[gid].dgst[5] = sha512_hmac_ctx.opad.h[5];
tmps[gid].dgst[6] = sha512_hmac_ctx.opad.h[6];
tmps[gid].dgst[7] = sha512_hmac_ctx.opad.h[7];
tmps[gid].dgst[i + 0] = sha512_hmac_ctx2.opad.h[0];
tmps[gid].dgst[i + 1] = sha512_hmac_ctx2.opad.h[1];
tmps[gid].dgst[i + 2] = sha512_hmac_ctx2.opad.h[2];
tmps[gid].dgst[i + 3] = sha512_hmac_ctx2.opad.h[3];
tmps[gid].dgst[i + 4] = sha512_hmac_ctx2.opad.h[4];
tmps[gid].dgst[i + 5] = sha512_hmac_ctx2.opad.h[5];
tmps[gid].dgst[i + 6] = sha512_hmac_ctx2.opad.h[6];
tmps[gid].dgst[i + 7] = sha512_hmac_ctx2.opad.h[7];
tmps[gid].out[i + 0] = tmps[gid].dgst[i + 0];
tmps[gid].out[i + 1] = tmps[gid].dgst[i + 1];
tmps[gid].out[i + 2] = tmps[gid].dgst[i + 2];
tmps[gid].out[i + 3] = tmps[gid].dgst[i + 3];
tmps[gid].out[i + 4] = tmps[gid].dgst[i + 4];
tmps[gid].out[i + 5] = tmps[gid].dgst[i + 5];
tmps[gid].out[i + 6] = tmps[gid].dgst[i + 6];
tmps[gid].out[i + 7] = tmps[gid].dgst[i + 7];
}
tmps[gid].out[0] = sha512_hmac_ctx.opad.h[0];
tmps[gid].out[1] = sha512_hmac_ctx.opad.h[1];
tmps[gid].out[2] = sha512_hmac_ctx.opad.h[2];
tmps[gid].out[3] = sha512_hmac_ctx.opad.h[3];
tmps[gid].out[4] = sha512_hmac_ctx.opad.h[4];
tmps[gid].out[5] = sha512_hmac_ctx.opad.h[5];
tmps[gid].out[6] = sha512_hmac_ctx.opad.h[6];
tmps[gid].out[7] = sha512_hmac_ctx.opad.h[7];
}
KERNEL_FQ void m06500_loop (KERN_ATTR_TMPS (sha512aix_tmp_t))
@ -216,103 +211,100 @@ KERNEL_FQ void m06500_loop (KERN_ATTR_TMPS (sha512aix_tmp_t))
opad[6] = pack64v (tmps, opad, gid, 6);
opad[7] = pack64v (tmps, opad, gid, 7);
for (u32 i = 0; i < 8; i += 8)
u64x dgst[8];
u64x out[8];
dgst[0] = pack64v (tmps, dgst, gid, 0);
dgst[1] = pack64v (tmps, dgst, gid, 1);
dgst[2] = pack64v (tmps, dgst, gid, 2);
dgst[3] = pack64v (tmps, dgst, gid, 3);
dgst[4] = pack64v (tmps, dgst, gid, 4);
dgst[5] = pack64v (tmps, dgst, gid, 5);
dgst[6] = pack64v (tmps, dgst, gid, 6);
dgst[7] = pack64v (tmps, dgst, gid, 7);
out[0] = pack64v (tmps, out, gid, 0);
out[1] = pack64v (tmps, out, gid, 1);
out[2] = pack64v (tmps, out, gid, 2);
out[3] = pack64v (tmps, out, gid, 3);
out[4] = pack64v (tmps, out, gid, 4);
out[5] = pack64v (tmps, out, gid, 5);
out[6] = pack64v (tmps, out, gid, 6);
out[7] = pack64v (tmps, out, gid, 7);
for (u32 j = 0; j < loop_cnt; j++)
{
u64x dgst[8];
u64x out[8];
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
u32x w4[4];
u32x w5[4];
u32x w6[4];
u32x w7[4];
dgst[0] = pack64v (tmps, dgst, gid, i + 0);
dgst[1] = pack64v (tmps, dgst, gid, i + 1);
dgst[2] = pack64v (tmps, dgst, gid, i + 2);
dgst[3] = pack64v (tmps, dgst, gid, i + 3);
dgst[4] = pack64v (tmps, dgst, gid, i + 4);
dgst[5] = pack64v (tmps, dgst, gid, i + 5);
dgst[6] = pack64v (tmps, dgst, gid, i + 6);
dgst[7] = pack64v (tmps, dgst, gid, i + 7);
w0[0] = h32_from_64 (dgst[0]);
w0[1] = l32_from_64 (dgst[0]);
w0[2] = h32_from_64 (dgst[1]);
w0[3] = l32_from_64 (dgst[1]);
w1[0] = h32_from_64 (dgst[2]);
w1[1] = l32_from_64 (dgst[2]);
w1[2] = h32_from_64 (dgst[3]);
w1[3] = l32_from_64 (dgst[3]);
w2[0] = h32_from_64 (dgst[4]);
w2[1] = l32_from_64 (dgst[4]);
w2[2] = h32_from_64 (dgst[5]);
w2[3] = l32_from_64 (dgst[5]);
w3[0] = h32_from_64 (dgst[6]);
w3[1] = l32_from_64 (dgst[6]);
w3[2] = h32_from_64 (dgst[7]);
w3[3] = l32_from_64 (dgst[7]);
w4[0] = 0x80000000;
w4[1] = 0;
w4[2] = 0;
w4[3] = 0;
w5[0] = 0;
w5[1] = 0;
w5[2] = 0;
w5[3] = 0;
w6[0] = 0;
w6[1] = 0;
w6[2] = 0;
w6[3] = 0;
w7[0] = 0;
w7[1] = 0;
w7[2] = 0;
w7[3] = (128 + 64) * 8;
out[0] = pack64v (tmps, out, gid, i + 0);
out[1] = pack64v (tmps, out, gid, i + 1);
out[2] = pack64v (tmps, out, gid, i + 2);
out[3] = pack64v (tmps, out, gid, i + 3);
out[4] = pack64v (tmps, out, gid, i + 4);
out[5] = pack64v (tmps, out, gid, i + 5);
out[6] = pack64v (tmps, out, gid, i + 6);
out[7] = pack64v (tmps, out, gid, i + 7);
hmac_sha512_run_V (w0, w1, w2, w3, w4, w5, w6, w7, ipad, opad, dgst);
for (u32 j = 0; j < loop_cnt; j++)
{
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
u32x w4[4];
u32x w5[4];
u32x w6[4];
u32x w7[4];
w0[0] = h32_from_64 (dgst[0]);
w0[1] = l32_from_64 (dgst[0]);
w0[2] = h32_from_64 (dgst[1]);
w0[3] = l32_from_64 (dgst[1]);
w1[0] = h32_from_64 (dgst[2]);
w1[1] = l32_from_64 (dgst[2]);
w1[2] = h32_from_64 (dgst[3]);
w1[3] = l32_from_64 (dgst[3]);
w2[0] = h32_from_64 (dgst[4]);
w2[1] = l32_from_64 (dgst[4]);
w2[2] = h32_from_64 (dgst[5]);
w2[3] = l32_from_64 (dgst[5]);
w3[0] = h32_from_64 (dgst[6]);
w3[1] = l32_from_64 (dgst[6]);
w3[2] = h32_from_64 (dgst[7]);
w3[3] = l32_from_64 (dgst[7]);
w4[0] = 0x80000000;
w4[1] = 0;
w4[2] = 0;
w4[3] = 0;
w5[0] = 0;
w5[1] = 0;
w5[2] = 0;
w5[3] = 0;
w6[0] = 0;
w6[1] = 0;
w6[2] = 0;
w6[3] = 0;
w7[0] = 0;
w7[1] = 0;
w7[2] = 0;
w7[3] = (128 + 64) * 8;
hmac_sha512_run_V (w0, w1, w2, w3, w4, w5, w6, w7, ipad, opad, dgst);
out[0] ^= dgst[0];
out[1] ^= dgst[1];
out[2] ^= dgst[2];
out[3] ^= dgst[3];
out[4] ^= dgst[4];
out[5] ^= dgst[5];
out[6] ^= dgst[6];
out[7] ^= dgst[7];
}
unpack64v (tmps, dgst, gid, i + 0, dgst[0]);
unpack64v (tmps, dgst, gid, i + 1, dgst[1]);
unpack64v (tmps, dgst, gid, i + 2, dgst[2]);
unpack64v (tmps, dgst, gid, i + 3, dgst[3]);
unpack64v (tmps, dgst, gid, i + 4, dgst[4]);
unpack64v (tmps, dgst, gid, i + 5, dgst[5]);
unpack64v (tmps, dgst, gid, i + 6, dgst[6]);
unpack64v (tmps, dgst, gid, i + 7, dgst[7]);
unpack64v (tmps, out, gid, i + 0, out[0]);
unpack64v (tmps, out, gid, i + 1, out[1]);
unpack64v (tmps, out, gid, i + 2, out[2]);
unpack64v (tmps, out, gid, i + 3, out[3]);
unpack64v (tmps, out, gid, i + 4, out[4]);
unpack64v (tmps, out, gid, i + 5, out[5]);
unpack64v (tmps, out, gid, i + 6, out[6]);
unpack64v (tmps, out, gid, i + 7, out[7]);
out[0] ^= dgst[0];
out[1] ^= dgst[1];
out[2] ^= dgst[2];
out[3] ^= dgst[3];
out[4] ^= dgst[4];
out[5] ^= dgst[5];
out[6] ^= dgst[6];
out[7] ^= dgst[7];
}
unpack64v (tmps, dgst, gid, 0, dgst[0]);
unpack64v (tmps, dgst, gid, 1, dgst[1]);
unpack64v (tmps, dgst, gid, 2, dgst[2]);
unpack64v (tmps, dgst, gid, 3, dgst[3]);
unpack64v (tmps, dgst, gid, 4, dgst[4]);
unpack64v (tmps, dgst, gid, 5, dgst[5]);
unpack64v (tmps, dgst, gid, 6, dgst[6]);
unpack64v (tmps, dgst, gid, 7, dgst[7]);
unpack64v (tmps, out, gid, 0, out[0]);
unpack64v (tmps, out, gid, 1, out[1]);
unpack64v (tmps, out, gid, 2, out[2]);
unpack64v (tmps, out, gid, 3, out[3]);
unpack64v (tmps, out, gid, 4, out[4]);
unpack64v (tmps, out, gid, 5, out[5]);
unpack64v (tmps, out, gid, 6, out[6]);
unpack64v (tmps, out, gid, 7, out[7]);
}
KERNEL_FQ void m06500_comp (KERN_ATTR_TMPS (sha512aix_tmp_t))

100
OpenCL/m07000_a3-optimized.cl
View File

@ -32,9 +32,9 @@ DECLSPEC void m07000m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
u32 salt_buf2[4];
u32 salt_buf3[4];
salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf0[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf0[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf0[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf0[3] = 0;
salt_buf1[0] = 0;
salt_buf1[1] = 0;
@ -56,12 +56,12 @@ DECLSPEC void m07000m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
u32 magic_buf2[4];
u32 magic_buf3[4];
magic_buf0[0] = FORTIGATE_A;
magic_buf0[1] = FORTIGATE_B;
magic_buf0[2] = FORTIGATE_C;
magic_buf0[3] = FORTIGATE_D;
magic_buf1[0] = FORTIGATE_E;
magic_buf1[1] = FORTIGATE_F;
magic_buf0[0] = hc_swap32_S (FORTIGATE_A);
magic_buf0[1] = hc_swap32_S (FORTIGATE_B);
magic_buf0[2] = hc_swap32_S (FORTIGATE_C);
magic_buf0[3] = hc_swap32_S (FORTIGATE_D);
magic_buf1[0] = hc_swap32_S (FORTIGATE_E);
magic_buf1[1] = hc_swap32_S (FORTIGATE_F);
magic_buf1[2] = 0;
magic_buf1[3] = 0;
magic_buf2[0] = 0;
@ -77,7 +77,7 @@ DECLSPEC void m07000m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u32 salt_pw_len = salt_len + pw_len;
switch_buffer_by_offset_le_S (magic_buf0, magic_buf1, magic_buf2, magic_buf3, salt_pw_len);
switch_buffer_by_offset_be_S (magic_buf0, magic_buf1, magic_buf2, magic_buf3, salt_pw_len);
salt_buf0[0] |= magic_buf0[0];
salt_buf0[1] |= magic_buf0[1];
@ -98,7 +98,7 @@ DECLSPEC void m07000m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u32 final_len = salt_len + pw_len + magic_len;
append_0x80_4x4_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, final_len);
append_0x80_4x4_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, final_len ^ 3);
/**
* loop
@ -158,20 +158,20 @@ DECLSPEC void m07000m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
* sha1
*/
u32x w0_t = hc_swap32 (t0[0]);
u32x w1_t = hc_swap32 (t0[1]);
u32x w2_t = hc_swap32 (t0[2]);
u32x w3_t = hc_swap32 (t0[3]);
u32x w4_t = hc_swap32 (t1[0]);
u32x w5_t = hc_swap32 (t1[1]);
u32x w6_t = hc_swap32 (t1[2]);
u32x w7_t = hc_swap32 (t1[3]);
u32x w8_t = hc_swap32 (t2[0]);
u32x w9_t = hc_swap32 (t2[1]);
u32x wa_t = hc_swap32 (t2[2]);
u32x wb_t = hc_swap32 (t2[3]);
u32x wc_t = hc_swap32 (t3[0]);
u32x wd_t = hc_swap32 (t3[1]);
u32x w0_t = t0[0];
u32x w1_t = t0[1];
u32x w2_t = t0[2];
u32x w3_t = t0[3];
u32x w4_t = t1[0];
u32x w5_t = t1[1];
u32x w6_t = t1[2];
u32x w7_t = t1[3];
u32x w8_t = t2[0];
u32x w9_t = t2[1];
u32x wa_t = t2[2];
u32x wb_t = t2[3];
u32x wc_t = t3[0];
u32x wd_t = t3[1];
u32x we_t = 0;
u32x wf_t = final_len * 8;
@ -299,9 +299,9 @@ DECLSPEC void m07000s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
u32 salt_buf2[4];
u32 salt_buf3[4];
salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf0[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf0[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf0[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf0[3] = 0;
salt_buf1[0] = 0;
salt_buf1[1] = 0;
@ -323,12 +323,12 @@ DECLSPEC void m07000s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
u32 magic_buf2[4];
u32 magic_buf3[4];
magic_buf0[0] = FORTIGATE_A;
magic_buf0[1] = FORTIGATE_B;
magic_buf0[2] = FORTIGATE_C;
magic_buf0[3] = FORTIGATE_D;
magic_buf1[0] = FORTIGATE_E;
magic_buf1[1] = FORTIGATE_F;
magic_buf0[0] = hc_swap32_S (FORTIGATE_A);
magic_buf0[1] = hc_swap32_S (FORTIGATE_B);
magic_buf0[2] = hc_swap32_S (FORTIGATE_C);
magic_buf0[3] = hc_swap32_S (FORTIGATE_D);
magic_buf1[0] = hc_swap32_S (FORTIGATE_E);
magic_buf1[1] = hc_swap32_S (FORTIGATE_F);
magic_buf1[2] = 0;
magic_buf1[3] = 0;
magic_buf2[0] = 0;
@ -344,7 +344,7 @@ DECLSPEC void m07000s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u32 salt_pw_len = salt_len + pw_len;
switch_buffer_by_offset_le_S (magic_buf0, magic_buf1, magic_buf2, magic_buf3, salt_pw_len);
switch_buffer_by_offset_be_S (magic_buf0, magic_buf1, magic_buf2, magic_buf3, salt_pw_len);
salt_buf0[0] |= magic_buf0[0];
salt_buf0[1] |= magic_buf0[1];
@ -365,7 +365,7 @@ DECLSPEC void m07000s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u32 final_len = salt_len + pw_len + magic_len;
append_0x80_4x4_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, final_len);
append_0x80_4x4_S (salt_buf0, salt_buf1, salt_buf2, salt_buf3, final_len ^ 3);
/**
* digest
@ -443,20 +443,20 @@ DECLSPEC void m07000s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
* sha1
*/
u32x w0_t = hc_swap32 (t0[0]);
u32x w1_t = hc_swap32 (t0[1]);
u32x w2_t = hc_swap32 (t0[2]);
u32x w3_t = hc_swap32 (t0[3]);
u32x w4_t = hc_swap32 (t1[0]);
u32x w5_t = hc_swap32 (t1[1]);
u32x w6_t = hc_swap32 (t1[2]);
u32x w7_t = hc_swap32 (t1[3]);
u32x w8_t = hc_swap32 (t2[0]);
u32x w9_t = hc_swap32 (t2[1]);
u32x wa_t = hc_swap32 (t2[2]);
u32x wb_t = hc_swap32 (t2[3]);
u32x wc_t = hc_swap32 (t3[0]);
u32x wd_t = hc_swap32 (t3[1]);
u32x w0_t = t0[0];
u32x w1_t = t0[1];
u32x w2_t = t0[2];
u32x w3_t = t0[3];
u32x w4_t = t1[0];
u32x w5_t = t1[1];
u32x w6_t = t1[2];
u32x w7_t = t1[3];
u32x w8_t = t2[0];
u32x w9_t = t2[1];
u32x wa_t = t2[2];
u32x wb_t = t2[3];
u32x wc_t = t3[0];
u32x wd_t = t3[1];
u32x we_t = 0;
u32x wf_t = final_len * 8;

4
OpenCL/m07000_a3-pure.cl
View File

@ -62,7 +62,7 @@ KERNEL_FQ void m07000_mxx (KERN_ATTR_VECTOR ())
sha1_init_vector_from_scalar (&ctx, &ctx0);
sha1_update_vector_swap (&ctx, w, pw_len);
sha1_update_vector (&ctx, w, pw_len);
/**
* pepper
@ -163,7 +163,7 @@ KERNEL_FQ void m07000_sxx (KERN_ATTR_VECTOR ())
sha1_init_vector_from_scalar (&ctx, &ctx0);
sha1_update_vector_swap (&ctx, w, pw_len);
sha1_update_vector (&ctx, w, pw_len);
/**
* pepper

146
OpenCL/m07300_a0-optimized.cl
View File

@ -118,13 +118,22 @@ KERNEL_FQ void m07300_m04 (KERN_ATTR_RULES_ESALT (rakp_t))
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* s_msg
*/
const u64 gid = get_global_id (0);
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
@ -186,40 +195,40 @@ KERNEL_FQ void m07300_m04 (KERN_ATTR_RULES_ESALT (rakp_t))
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = s_esalt_buf[esalt_off + 14];
w3[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0, w1, w2, w3, ipad);
}
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = 0;
w3[3] = (64 + esalt_size) * 8;
@ -245,13 +254,22 @@ KERNEL_FQ void m07300_s04 (KERN_ATTR_RULES_ESALT (rakp_t))
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* s_msg
*/
const u64 gid = get_global_id (0);
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
@ -325,40 +343,40 @@ KERNEL_FQ void m07300_s04 (KERN_ATTR_RULES_ESALT (rakp_t))
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = s_esalt_buf[esalt_off + 14];
w3[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0, w1, w2, w3, ipad);
}
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = 0;
w3[3] = (64 + esalt_size) * 8;

146
OpenCL/m07300_a1-optimized.cl
View File

@ -116,13 +116,22 @@ KERNEL_FQ void m07300_m04 (KERN_ATTR_ESALT (rakp_t))
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* s_msg
*/
const u64 gid = get_global_id (0);
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
@ -252,40 +261,40 @@ KERNEL_FQ void m07300_m04 (KERN_ATTR_ESALT (rakp_t))
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = s_esalt_buf[esalt_off + 14];
w3[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0, w1, w2, w3, ipad);
}
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = 0;
w3[3] = (64 + esalt_size) * 8;
@ -311,13 +320,22 @@ KERNEL_FQ void m07300_s04 (KERN_ATTR_ESALT (rakp_t))
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* s_msg
*/
const u64 gid = get_global_id (0);
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
@ -459,40 +477,40 @@ KERNEL_FQ void m07300_s04 (KERN_ATTR_ESALT (rakp_t))
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = s_esalt_buf[esalt_off + 14];
w3[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0, w1, w2, w3, ipad);
}
w0[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0[0] = s_esalt_buf[esalt_off + 0];
w0[1] = s_esalt_buf[esalt_off + 1];
w0[2] = s_esalt_buf[esalt_off + 2];
w0[3] = s_esalt_buf[esalt_off + 3];
w1[0] = s_esalt_buf[esalt_off + 4];
w1[1] = s_esalt_buf[esalt_off + 5];
w1[2] = s_esalt_buf[esalt_off + 6];
w1[3] = s_esalt_buf[esalt_off + 7];
w2[0] = s_esalt_buf[esalt_off + 8];
w2[1] = s_esalt_buf[esalt_off + 9];
w2[2] = s_esalt_buf[esalt_off + 10];
w2[3] = s_esalt_buf[esalt_off + 11];
w3[0] = s_esalt_buf[esalt_off + 12];
w3[1] = s_esalt_buf[esalt_off + 13];
w3[2] = 0;
w3[3] = (64 + esalt_size) * 8;

262
OpenCL/m07300_a3-optimized.cl
View File

@ -110,7 +110,7 @@ DECLSPEC void hmac_sha1_run (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad,
sha1_transform_vector (w0, w1, w2, w3, digest);
}
DECLSPEC void m07300m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (rakp_t))
DECLSPEC void m07300m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (rakp_t), LOCAL_AS u32 *s_esalt_buf)
{
/**
* modifier
@ -175,40 +175,40 @@ DECLSPEC void m07300m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0_t[0] = s_esalt_buf[esalt_off + 0];
w0_t[1] = s_esalt_buf[esalt_off + 1];
w0_t[2] = s_esalt_buf[esalt_off + 2];
w0_t[3] = s_esalt_buf[esalt_off + 3];
w1_t[0] = s_esalt_buf[esalt_off + 4];
w1_t[1] = s_esalt_buf[esalt_off + 5];
w1_t[2] = s_esalt_buf[esalt_off + 6];
w1_t[3] = s_esalt_buf[esalt_off + 7];
w2_t[0] = s_esalt_buf[esalt_off + 8];
w2_t[1] = s_esalt_buf[esalt_off + 9];
w2_t[2] = s_esalt_buf[esalt_off + 10];
w2_t[3] = s_esalt_buf[esalt_off + 11];
w3_t[0] = s_esalt_buf[esalt_off + 12];
w3_t[1] = s_esalt_buf[esalt_off + 13];
w3_t[2] = s_esalt_buf[esalt_off + 14];
w3_t[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, ipad);
}
w0_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0_t[0] = s_esalt_buf[esalt_off + 0];
w0_t[1] = s_esalt_buf[esalt_off + 1];
w0_t[2] = s_esalt_buf[esalt_off + 2];
w0_t[3] = s_esalt_buf[esalt_off + 3];
w1_t[0] = s_esalt_buf[esalt_off + 4];
w1_t[1] = s_esalt_buf[esalt_off + 5];
w1_t[2] = s_esalt_buf[esalt_off + 6];
w1_t[3] = s_esalt_buf[esalt_off + 7];
w2_t[0] = s_esalt_buf[esalt_off + 8];
w2_t[1] = s_esalt_buf[esalt_off + 9];
w2_t[2] = s_esalt_buf[esalt_off + 10];
w2_t[3] = s_esalt_buf[esalt_off + 11];
w3_t[0] = s_esalt_buf[esalt_off + 12];
w3_t[1] = s_esalt_buf[esalt_off + 13];
w3_t[2] = 0;
w3_t[3] = (64 + esalt_size) * 8;
@ -220,7 +220,7 @@ DECLSPEC void m07300m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
}
}
DECLSPEC void m07300s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (rakp_t))
DECLSPEC void m07300s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (rakp_t), LOCAL_AS u32 *s_esalt_buf)
{
/**
* modifier
@ -297,40 +297,40 @@ DECLSPEC void m07300s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
{
w0_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w3_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 14];
w3_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 15];
w0_t[0] = s_esalt_buf[esalt_off + 0];
w0_t[1] = s_esalt_buf[esalt_off + 1];
w0_t[2] = s_esalt_buf[esalt_off + 2];
w0_t[3] = s_esalt_buf[esalt_off + 3];
w1_t[0] = s_esalt_buf[esalt_off + 4];
w1_t[1] = s_esalt_buf[esalt_off + 5];
w1_t[2] = s_esalt_buf[esalt_off + 6];
w1_t[3] = s_esalt_buf[esalt_off + 7];
w2_t[0] = s_esalt_buf[esalt_off + 8];
w2_t[1] = s_esalt_buf[esalt_off + 9];
w2_t[2] = s_esalt_buf[esalt_off + 10];
w2_t[3] = s_esalt_buf[esalt_off + 11];
w3_t[0] = s_esalt_buf[esalt_off + 12];
w3_t[1] = s_esalt_buf[esalt_off + 13];
w3_t[2] = s_esalt_buf[esalt_off + 14];
w3_t[3] = s_esalt_buf[esalt_off + 15];
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, ipad);
}
w0_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 0];
w0_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 1];
w0_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 2];
w0_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 3];
w1_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 4];
w1_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 5];
w1_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 6];
w1_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 7];
w2_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 8];
w2_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 9];
w2_t[2] = esalt_bufs[digests_offset].salt_buf[esalt_off + 10];
w2_t[3] = esalt_bufs[digests_offset].salt_buf[esalt_off + 11];
w3_t[0] = esalt_bufs[digests_offset].salt_buf[esalt_off + 12];
w3_t[1] = esalt_bufs[digests_offset].salt_buf[esalt_off + 13];
w0_t[0] = s_esalt_buf[esalt_off + 0];
w0_t[1] = s_esalt_buf[esalt_off + 1];
w0_t[2] = s_esalt_buf[esalt_off + 2];
w0_t[3] = s_esalt_buf[esalt_off + 3];
w1_t[0] = s_esalt_buf[esalt_off + 4];
w1_t[1] = s_esalt_buf[esalt_off + 5];
w1_t[2] = s_esalt_buf[esalt_off + 6];
w1_t[3] = s_esalt_buf[esalt_off + 7];
w2_t[0] = s_esalt_buf[esalt_off + 8];
w2_t[1] = s_esalt_buf[esalt_off + 9];
w2_t[2] = s_esalt_buf[esalt_off + 10];
w2_t[3] = s_esalt_buf[esalt_off + 11];
w3_t[0] = s_esalt_buf[esalt_off + 12];
w3_t[1] = s_esalt_buf[esalt_off + 13];
w3_t[2] = 0;
w3_t[3] = (64 + esalt_size) * 8;
@ -345,13 +345,32 @@ DECLSPEC void m07300s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
KERNEL_FQ void m07300_m04 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -386,19 +405,38 @@ KERNEL_FQ void m07300_m04 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300m (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);
m07300m (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_esalt_buf);
}
KERNEL_FQ void m07300_m08 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -433,19 +471,38 @@ KERNEL_FQ void m07300_m08 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300m (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);
m07300m (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_esalt_buf);
}
KERNEL_FQ void m07300_m16 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -480,19 +537,38 @@ KERNEL_FQ void m07300_m16 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300m (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);
m07300m (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_esalt_buf);
}
KERNEL_FQ void m07300_s04 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -527,19 +603,38 @@ KERNEL_FQ void m07300_s04 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300s (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);
m07300s (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_esalt_buf);
}
KERNEL_FQ void m07300_s08 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -574,19 +669,38 @@ KERNEL_FQ void m07300_s08 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300s (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);
m07300s (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_esalt_buf);
}
KERNEL_FQ void m07300_s16 (KERN_ATTR_ESALT (rakp_t))
{
/**
* base
* modifier
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* s_msg
*/
LOCAL_VK u32 s_esalt_buf[128];
for (u32 i = lid; i < 128; i += lsz)
{
s_esalt_buf[i] = esalt_bufs[digests_offset].salt_buf[i];
}
SYNC_THREADS ();
if (gid >= gid_max) return;
/**
* base
*/
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -621,5 +735,5 @@ KERNEL_FQ void m07300_s16 (KERN_ATTR_ESALT (rakp_t))
* main
*/
m07300s (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);
m07300s (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_esalt_buf);
}

1143
OpenCL/m07400-optimized.cl
View File

File diff suppressed because it is too large Load Diff

596
OpenCL/m07800_a0-optimized.cl
View File

@ -17,30 +17,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -56,20 +116,52 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
@ -90,14 +182,14 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -114,6 +206,15 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
/**
* SAP
*/
@ -140,7 +241,7 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, out_len);
const u32x pw_salt_len = out_len + salt_len;
@ -150,20 +251,20 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -191,33 +292,33 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -227,49 +328,61 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, out_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = out_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -278,17 +391,21 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_RULES ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
@ -304,20 +421,52 @@ KERNEL_FQ void m07800_m16 (KERN_ATTR_RULES ())
KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
@ -338,14 +487,14 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -374,6 +523,15 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
/**
* SAP
*/
@ -400,7 +558,7 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, out_len);
const u32x pw_salt_len = out_len + salt_len;
@ -410,20 +568,20 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -451,33 +609,33 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -487,49 +645,61 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, out_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = out_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -538,17 +708,21 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_RULES ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}

604
OpenCL/m07800_a1-optimized.cl
View File

@ -15,30 +15,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -54,17 +114,45 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -88,14 +176,14 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -172,6 +260,23 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
w2[0] = hc_swap32_S (w2[0]);
w2[1] = hc_swap32_S (w2[1]);
w2[2] = hc_swap32_S (w2[2]);
w2[3] = hc_swap32_S (w2[3]);
w3[0] = hc_swap32_S (w3[0]);
w3[1] = hc_swap32_S (w3[1]);
w3[2] = hc_swap32_S (w3[2]);
w3[3] = hc_swap32_S (w3[3]);
/**
* SAP
*/
@ -198,7 +303,7 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, pw_len);
const u32x pw_salt_len = pw_len + salt_len;
@ -208,20 +313,20 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -249,33 +354,33 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -285,49 +390,61 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = hc_swap32_S (w2[0]);
final[ 9] = hc_swap32_S (w2[1]);
final[10] = hc_swap32_S (w2[2]);
final[11] = hc_swap32_S (w2[3]);
final[12] = hc_swap32_S (w3[0]);
final[13] = hc_swap32_S (w3[1]);
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -336,17 +453,21 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
@ -362,17 +483,45 @@ KERNEL_FQ void m07800_m16 (KERN_ATTR_BASIC ())
KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -396,14 +545,14 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -492,6 +641,23 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
w2[0] = hc_swap32_S (w2[0]);
w2[1] = hc_swap32_S (w2[1]);
w2[2] = hc_swap32_S (w2[2]);
w2[3] = hc_swap32_S (w2[3]);
w3[0] = hc_swap32_S (w3[0]);
w3[1] = hc_swap32_S (w3[1]);
w3[2] = hc_swap32_S (w3[2]);
w3[3] = hc_swap32_S (w3[3]);
/**
* SAP
*/
@ -518,7 +684,7 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, pw_len);
const u32x pw_salt_len = pw_len + salt_len;
@ -528,20 +694,20 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -569,33 +735,33 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -605,49 +771,61 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = hc_swap32_S (w2[0]);
final[ 9] = hc_swap32_S (w2[1]);
final[10] = hc_swap32_S (w2[2]);
final[11] = hc_swap32_S (w2[3]);
final[12] = hc_swap32_S (w3[0]);
final[13] = hc_swap32_S (w3[1]);
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -656,17 +834,21 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}

550
OpenCL/m07800_a3-optimized.cl
View File

@ -15,30 +15,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -52,7 +112,7 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
a[d + 1] = l32_from_64_S (tmp);
}
DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
DECLSPEC void m07800m (SHM_TYPE u32a (*s_theMagicArray)[16], u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
{
/**
* modifier
@ -165,33 +225,33 @@ DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -201,47 +261,59 @@ DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = w0[0];
final[ 1] = w0[1];
final[ 2] = w0[2];
final[ 3] = w0[3];
final[ 4] = w1[0];
final[ 5] = w1[1];
final[ 6] = w1[2];
final[ 7] = w1[3];
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
@ -252,23 +324,27 @@ DECLSPEC void m07800m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
}
DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
DECLSPEC void m07800s (SHM_TYPE u32a (*s_theMagicArray)[16], u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
{
/**
* modifier
@ -393,33 +469,33 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -429,47 +505,59 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = w0[0];
final[ 1] = w0[1];
final[ 2] = w0[2];
final[ 3] = w0[3];
final[ 4] = w1[0];
final[ 5] = w1[1];
final[ 6] = w1[2];
final[ 7] = w1[3];
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
@ -480,17 +568,21 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
}
@ -498,11 +590,45 @@ DECLSPEC void m07800s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -510,8 +636,6 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -546,16 +670,50 @@ KERNEL_FQ void m07800_m04 (KERN_ATTR_BASIC ())
* main
*/
m07800m (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);
m07800m (s_theMagicArray, 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);
}
KERNEL_FQ void m07800_m08 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -563,8 +721,6 @@ KERNEL_FQ void m07800_m08 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -599,7 +755,7 @@ KERNEL_FQ void m07800_m08 (KERN_ATTR_BASIC ())
* main
*/
m07800m (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);
m07800m (s_theMagicArray, 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);
}
KERNEL_FQ void m07800_m16 (KERN_ATTR_BASIC ())
@ -608,11 +764,45 @@ KERNEL_FQ void m07800_m16 (KERN_ATTR_BASIC ())
KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -620,8 +810,6 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -656,16 +844,50 @@ KERNEL_FQ void m07800_s04 (KERN_ATTR_BASIC ())
* main
*/
m07800s (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);
m07800s (s_theMagicArray, 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);
}
KERNEL_FQ void m07800_s08 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -673,8 +895,6 @@ KERNEL_FQ void m07800_s08 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -709,7 +929,7 @@ KERNEL_FQ void m07800_s08 (KERN_ATTR_BASIC ())
* main
*/
m07800s (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);
m07800s (s_theMagicArray, 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);
}
KERNEL_FQ void m07800_s16 (KERN_ATTR_BASIC ())

596
OpenCL/m07801_a0-optimized.cl
View File

@ -17,30 +17,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -56,20 +116,52 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
@ -90,14 +182,14 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -114,6 +206,15 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
/**
* SAP
*/
@ -140,7 +241,7 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, out_len);
const u32x pw_salt_len = out_len + salt_len;
@ -150,20 +251,20 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -191,33 +292,33 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -227,49 +328,61 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, out_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = out_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -278,17 +391,21 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_RULES ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}
@ -304,20 +421,52 @@ KERNEL_FQ void m07801_m16 (KERN_ATTR_RULES ())
KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf0[4];
u32 pw_buf1[4];
@ -338,14 +487,14 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -374,6 +523,15 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
const u32x out_len = apply_rules_vect_optimized (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
/**
* SAP
*/
@ -400,7 +558,7 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, out_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, out_len);
const u32x pw_salt_len = out_len + salt_len;
@ -410,20 +568,20 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -451,33 +609,33 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -487,49 +645,61 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, out_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = out_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -538,17 +708,21 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_RULES ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}

604
OpenCL/m07801_a1-optimized.cl
View File

@ -15,30 +15,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -54,17 +114,45 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -88,14 +176,14 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -172,6 +260,23 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
w2[0] = hc_swap32_S (w2[0]);
w2[1] = hc_swap32_S (w2[1]);
w2[2] = hc_swap32_S (w2[2]);
w2[3] = hc_swap32_S (w2[3]);
w3[0] = hc_swap32_S (w3[0]);
w3[1] = hc_swap32_S (w3[1]);
w3[2] = hc_swap32_S (w3[2]);
w3[3] = hc_swap32_S (w3[3]);
/**
* SAP
*/
@ -198,7 +303,7 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, pw_len);
const u32x pw_salt_len = pw_len + salt_len;
@ -208,20 +313,20 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -249,33 +354,33 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -285,49 +390,61 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = hc_swap32_S (w2[0]);
final[ 9] = hc_swap32_S (w2[1]);
final[10] = hc_swap32_S (w2[2]);
final[11] = hc_swap32_S (w2[3]);
final[12] = hc_swap32_S (w3[0]);
final[13] = hc_swap32_S (w3[1]);
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -336,17 +453,21 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}
@ -362,17 +483,45 @@ KERNEL_FQ void m07801_m16 (KERN_ATTR_BASIC ())
KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
{
/**
* modifier
*/
const u64 lid = get_local_id (0);
/**
* base
*/
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -396,14 +545,14 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
u32 salt_buf[8];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
salt_buf[0] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[0]);
salt_buf[1] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[1]);
salt_buf[2] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[2]);
salt_buf[3] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[3]);
salt_buf[4] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[4]);
salt_buf[5] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[5]);
salt_buf[6] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[6]);
salt_buf[7] = hc_swap32_S (salt_bufs[salt_pos].salt_buf[7]);
const u32 salt_len = salt_bufs[salt_pos].salt_len;
@ -492,6 +641,23 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
w3[2] = wordl3[2] | wordr3[2];
w3[3] = wordl3[3] | wordr3[3];
w0[0] = hc_swap32_S (w0[0]);
w0[1] = hc_swap32_S (w0[1]);
w0[2] = hc_swap32_S (w0[2]);
w0[3] = hc_swap32_S (w0[3]);
w1[0] = hc_swap32_S (w1[0]);
w1[1] = hc_swap32_S (w1[1]);
w1[2] = hc_swap32_S (w1[2]);
w1[3] = hc_swap32_S (w1[3]);
w2[0] = hc_swap32_S (w2[0]);
w2[1] = hc_swap32_S (w2[1]);
w2[2] = hc_swap32_S (w2[2]);
w2[3] = hc_swap32_S (w2[3]);
w3[0] = hc_swap32_S (w3[0]);
w3[1] = hc_swap32_S (w3[1]);
w3[2] = hc_swap32_S (w3[2]);
w3[3] = hc_swap32_S (w3[3]);
/**
* SAP
*/
@ -518,7 +684,7 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
s3[2] = 0;
s3[3] = 0;
switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
switch_buffer_by_offset_be_S (s0, s1, s2, s3, pw_len);
const u32x pw_salt_len = pw_len + salt_len;
@ -528,20 +694,20 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
u32 final[32];
final[ 0] = hc_swap32_S (w0[0] | s0[0]);
final[ 1] = hc_swap32_S (w0[1] | s0[1]);
final[ 2] = hc_swap32_S (w0[2] | s0[2]);
final[ 3] = hc_swap32_S (w0[3] | s0[3]);
final[ 4] = hc_swap32_S (w1[0] | s1[0]);
final[ 5] = hc_swap32_S (w1[1] | s1[1]);
final[ 6] = hc_swap32_S (w1[2] | s1[2]);
final[ 7] = hc_swap32_S (w1[3] | s1[3]);
final[ 8] = hc_swap32_S (w2[0] | s2[0]);
final[ 9] = hc_swap32_S (w2[1] | s2[1]);
final[10] = hc_swap32_S (w2[2] | s2[2]);
final[11] = hc_swap32_S (w2[3] | s2[3]);
final[12] = hc_swap32_S (w3[0] | s3[0]);
final[13] = hc_swap32_S (w3[1] | s3[1]);
final[ 0] = w0[0] | s0[0];
final[ 1] = w0[1] | s0[1];
final[ 2] = w0[2] | s0[2];
final[ 3] = w0[3] | s0[3];
final[ 4] = w1[0] | s1[0];
final[ 5] = w1[1] | s1[1];
final[ 6] = w1[2] | s1[2];
final[ 7] = w1[3] | s1[3];
final[ 8] = w2[0] | s2[0];
final[ 9] = w2[1] | s2[1];
final[10] = w2[2] | s2[2];
final[11] = w2[3] | s2[3];
final[12] = w3[0] | s3[0];
final[13] = w3[1] | s3[1];
final[14] = 0;
final[15] = pw_salt_len * 8;
final[16] = 0;
@ -569,33 +735,33 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -605,49 +771,61 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = hc_swap32_S (w0[0]);
final[ 1] = hc_swap32_S (w0[1]);
final[ 2] = hc_swap32_S (w0[2]);
final[ 3] = hc_swap32_S (w0[3]);
final[ 4] = hc_swap32_S (w1[0]);
final[ 5] = hc_swap32_S (w1[1]);
final[ 6] = hc_swap32_S (w1[2]);
final[ 7] = hc_swap32_S (w1[3]);
final[ 8] = hc_swap32_S (w2[0]);
final[ 9] = hc_swap32_S (w2[1]);
final[10] = hc_swap32_S (w2[2]);
final[11] = hc_swap32_S (w2[3]);
final[12] = hc_swap32_S (w3[0]);
final[13] = hc_swap32_S (w3[1]);
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = hc_swap32_S (salt_buf[i / 4]); // attention, int[] not char[]
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
SETSHIFTEDINT (final, final_len + i, tmp);
}
@ -656,17 +834,21 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}

550
OpenCL/m07801_a3-optimized.cl
View File

@ -15,30 +15,90 @@
#include "inc_hash_sha1.cl"
#endif
CONSTANT_VK u32a theMagicArray[64] =
CONSTANT_VK u32a theMagicArray[80][16] =
{
0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f,
0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194,
0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3,
0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3,
0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f, 0x37af944c, 0x29085282, 0xb23b4e37, 0x9f170791, 0x113bfdcd,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,
{ 0x91ac5114, 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194 },
{ 0xac51149f, 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f },
{ 0x51149f67, 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b },
{ 0x149f6754, 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1 },
{ 0x9f675443, 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b },
{ 0x67544324, 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06 },
{ 0x544324e7, 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605 },
{ 0x4324e73b, 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d },
{ 0x24e73be0, 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03 },
{ 0xe73be028, 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d },
{ 0x3be02874, 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e },
{ 0xe028747b, 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13 },
{ 0x28747bc2, 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a },
{ 0x747bc286, 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e },
{ 0x7bc28633, 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a },
{ 0xc2863313, 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a },
{ 0x863313eb, 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8 },
{ 0x3313eb5a, 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9 },
{ 0x13eb5a4f, 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c },
{ 0xeb5a4fcb, 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14 },
{ 0x5a4fcb5c, 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417 },
{ 0x4fcb5c08, 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758 },
{ 0xcb5c080a, 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7 },
{ 0x5c080a73, 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a },
{ 0x080a7337, 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6 },
{ 0x0a73370e, 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1 },
{ 0x73370e5d, 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199 },
{ 0x370e5d1c, 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963 },
{ 0x0e5d1c2f, 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a },
{ 0x5d1c2f33, 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7 },
{ 0x1c2f338f, 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd },
{ 0x2f338fe6, 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70 },
{ 0x338fe6e5, 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3 },
{ 0x8fe6e5f8, 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6 },
{ 0xe6e5f89b, 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e },
{ 0xe5f89bae, 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74 },
{ 0xf89baedd, 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413 },
{ 0x9baedd16, 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303 },
{ 0xaedd16f2, 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9 },
{ 0xdd16f24b, 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b },
{ 0x16f24b8d, 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04 },
{ 0xf24b8d2c, 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426 },
{ 0x4b8d2ce1, 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698 },
{ 0x8d2ce1d4, 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7 },
{ 0x2ce1d4dc, 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726 },
{ 0xe1d4dcb0, 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a },
{ 0xd4dcb0cb, 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92 },
{ 0xdcb0cbdf, 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293 },
{ 0xb0cbdf9d, 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325 },
{ 0xcbdf9dd4, 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0 },
{ 0xdf9dd470, 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2 },
{ 0x9dd4706d, 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d },
{ 0xd4706d17, 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23 },
{ 0x706d17f9, 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed },
{ 0x6d17f94d, 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63 },
{ 0x17f94d42, 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379 },
{ 0xf94d423f, 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d },
{ 0x4d423f9b, 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13 },
{ 0x423f9b1b, 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332 },
{ 0x3f9b1b11, 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa },
{ 0x9b1b1194, 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c },
{ 0x1b11949f, 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35 },
{ 0x11949f5b, 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502 },
{ 0x949f5bc1, 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a },
{ 0x9f5bc19b, 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3 },
{ 0x5bc19b06, 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3 },
{ 0xc19b0605, 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd },
{ 0x9b06059d, 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e },
{ 0x06059d03, 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a },
{ 0x059d039d, 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24 },
{ 0x9d039d5e, 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf },
{ 0x039d5e13, 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51 },
{ 0x9d5e138a, 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3 },
{ 0x5e138a1e, 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c },
{ 0x138a1e9a, 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd },
{ 0x8a1e9a6a, 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55 },
{ 0x1e9a6ae8, 0xd97c1417, 0x58c72af6, 0xa199630a, 0xd7fd70c3, 0xf65e7413, 0x03c90b04, 0x2698f726, 0x8a929325, 0xb0a20d23, 0xed63796d, 0x1332fa3c, 0x35029aa3, 0xb3dd8e0a, 0x24bf51c3, 0x7ccd559f },
{ 0x9a6ae8d9, 0x7c141758, 0xc72af6a1, 0x99630ad7, 0xfd70c3f6, 0x5e741303, 0xc90b0426, 0x98f7268a, 0x929325b0, 0xa20d23ed, 0x63796d13, 0x32fa3c35, 0x029aa3b3, 0xdd8e0a24, 0xbf51c37c, 0xcd559f37 },
{ 0x6ae8d97c, 0x141758c7, 0x2af6a199, 0x630ad7fd, 0x70c3f65e, 0x741303c9, 0x0b042698, 0xf7268a92, 0x9325b0a2, 0x0d23ed63, 0x796d1332, 0xfa3c3502, 0x9aa3b3dd, 0x8e0a24bf, 0x51c37ccd, 0x559f37af },
{ 0xe8d97c14, 0x1758c72a, 0xf6a19963, 0x0ad7fd70, 0xc3f65e74, 0x1303c90b, 0x042698f7, 0x268a9293, 0x25b0a20d, 0x23ed6379, 0x6d1332fa, 0x3c35029a, 0xa3b3dd8e, 0x0a24bf51, 0xc37ccd55, 0x9f37af94 },
};
DECLSPEC u32 GETSHIFTEDINT_CONST (CONSTANT_AS u32a *a, const int n)
{
const int d = n / 4;
const int m = n & 3;
u64 tmp = hl32_to_64_S (a[d + 0], a[d + 1]);
tmp <<= m * 8;
return h32_from_64_S (tmp);
}
DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
{
const int d = n / 4;
@ -52,7 +112,7 @@ DECLSPEC void SETSHIFTEDINT (u32 *a, const int n, const u32 v)
a[d + 1] = l32_from_64_S (tmp);
}
DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
DECLSPEC void m07801m (SHM_TYPE u32a (*s_theMagicArray)[16], u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
{
/**
* modifier
@ -165,33 +225,33 @@ DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -201,47 +261,59 @@ DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = w0[0];
final[ 1] = w0[1];
final[ 2] = w0[2];
final[ 3] = w0[3];
final[ 4] = w1[0];
final[ 5] = w1[1];
final[ 6] = w1[2];
final[ 7] = w1[3];
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
@ -252,23 +324,27 @@ DECLSPEC void m07801m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_M_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}
}
DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
DECLSPEC void m07801s (SHM_TYPE u32a (*s_theMagicArray)[16], u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_BASIC ())
{
/**
* modifier
@ -393,33 +469,33 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
// prepare magic array range
u32 lengthMagicArray = 0x20;
u32 offsetMagicArray = 0;
lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
lengthMagicArray += unpack_v8d_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[0]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8b_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8a_from_v32_S (digest[1]) % 6;
lengthMagicArray += unpack_v8d_from_v32_S (digest[2]) % 6;
lengthMagicArray += unpack_v8c_from_v32_S (digest[2]) % 6;
offsetMagicArray += unpack_v8b_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[2]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[3]) & 7;
offsetMagicArray += unpack_v8d_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8c_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8b_from_v32_S (digest[4]) & 7;
offsetMagicArray += unpack_v8a_from_v32_S (digest[4]) & 7;
// final
@ -429,47 +505,59 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
final[ 0] = w0[0];
final[ 1] = w0[1];
final[ 2] = w0[2];
final[ 3] = w0[3];
final[ 4] = w1[0];
final[ 5] = w1[1];
final[ 6] = w1[2];
final[ 7] = w1[3];
final[ 8] = 0;
final[ 9] = 0;
final[10] = 0;
final[11] = 0;
final[12] = 0;
final[13] = 0;
final[14] = 0;
final[15] = 0;
u32 final_len = pw_len;
u32 i;
// append MagicArray
for (i = 0; i < lengthMagicArray - 4; i += 4)
{
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i);
final[ 0] = s_theMagicArray[offsetMagicArray][ 0];
final[ 1] = s_theMagicArray[offsetMagicArray][ 1];
final[ 2] = s_theMagicArray[offsetMagicArray][ 2];
final[ 3] = s_theMagicArray[offsetMagicArray][ 3];
final[ 4] = s_theMagicArray[offsetMagicArray][ 4];
final[ 5] = s_theMagicArray[offsetMagicArray][ 5];
final[ 6] = s_theMagicArray[offsetMagicArray][ 6];
final[ 7] = s_theMagicArray[offsetMagicArray][ 7];
final[ 8] = s_theMagicArray[offsetMagicArray][ 8];
final[ 9] = s_theMagicArray[offsetMagicArray][ 9];
final[10] = s_theMagicArray[offsetMagicArray][10];
final[11] = s_theMagicArray[offsetMagicArray][11];
final[12] = s_theMagicArray[offsetMagicArray][12];
final[13] = s_theMagicArray[offsetMagicArray][13];
final[14] = s_theMagicArray[offsetMagicArray][14];
final[15] = s_theMagicArray[offsetMagicArray][15];
final[16] = 0;
final[17] = 0;
final[18] = 0;
final[19] = 0;
final[20] = 0;
final[21] = 0;
final[22] = 0;
final[23] = 0;
final[24] = 0;
final[25] = 0;
final[26] = 0;
final[27] = 0;
final[28] = 0;
final[29] = 0;
final[30] = 0;
final[31] = 0;
SETSHIFTEDINT (final, final_len + i, tmp);
}
truncate_block_16x4_be_S (final + 0, final + 4, final + 8, final + 12, lengthMagicArray);
const u32 mask = 0xffffffff << (((4 - (lengthMagicArray - i)) & 3) * 8);
switch_buffer_by_offset_8x4_be_S (final + 0, final + 4, final + 8, final + 12, final + 16, final + 20, final + 24, final + 28, pw_len);
const u32 tmp = GETSHIFTEDINT_CONST (theMagicArray, offsetMagicArray + i) & mask;
final[0] |= w0[0];
final[1] |= w0[1];
final[2] |= w0[2];
final[3] |= w0[3];
final[4] |= w1[0];
final[5] |= w1[1];
final[6] |= w1[2];
final[7] |= w1[3];
SETSHIFTEDINT (final, final_len + i, tmp);
final_len += lengthMagicArray;
u32 final_len = pw_len + lengthMagicArray;
// append Salt
for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
for (int i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
{
const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
@ -480,17 +568,21 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
// calculate
int left;
int off;
for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
if (final_len >= 56)
{
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
final[30] = 0;
final[31] = final_len * 8;
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
sha1_transform (final + 16, final + 20, final + 24, final + 28, digest);
}
else
{
final[14] = 0;
final[15] = final_len * 8;
final[off + 15] = final_len * 8;
sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
sha1_transform (final + 0, final + 4, final + 8, final + 12, digest);
}
COMPARE_S_SIMD (0, 0, digest[2] & 0xffff0000, digest[1]);
}
@ -498,11 +590,45 @@ DECLSPEC void m07801s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -510,8 +636,6 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -546,16 +670,50 @@ KERNEL_FQ void m07801_m04 (KERN_ATTR_BASIC ())
* main
*/
m07801m (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);
m07801m (s_theMagicArray, 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);
}
KERNEL_FQ void m07801_m08 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -563,8 +721,6 @@ KERNEL_FQ void m07801_m08 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -599,7 +755,7 @@ KERNEL_FQ void m07801_m08 (KERN_ATTR_BASIC ())
* main
*/
m07801m (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);
m07801m (s_theMagicArray, 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);
}
KERNEL_FQ void m07801_m16 (KERN_ATTR_BASIC ())
@ -608,11 +764,45 @@ KERNEL_FQ void m07801_m16 (KERN_ATTR_BASIC ())
KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -620,8 +810,6 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -656,16 +844,50 @@ KERNEL_FQ void m07801_s04 (KERN_ATTR_BASIC ())
* main
*/
m07801s (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);
m07801s (s_theMagicArray, 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);
}
KERNEL_FQ void m07801_s08 (KERN_ATTR_BASIC ())
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* base
* aes shared
*/
const u64 gid = get_global_id (0);
#ifdef REAL_SHM
LOCAL_VK u32 s_theMagicArray[80][16];
for (u32 i = lid; i < 80; i += lsz)
{
s_theMagicArray[i][ 0] = theMagicArray[i][ 0];
s_theMagicArray[i][ 1] = theMagicArray[i][ 1];
s_theMagicArray[i][ 2] = theMagicArray[i][ 2];
s_theMagicArray[i][ 3] = theMagicArray[i][ 3];
s_theMagicArray[i][ 4] = theMagicArray[i][ 4];
s_theMagicArray[i][ 5] = theMagicArray[i][ 5];
s_theMagicArray[i][ 6] = theMagicArray[i][ 6];
s_theMagicArray[i][ 7] = theMagicArray[i][ 7];
s_theMagicArray[i][ 8] = theMagicArray[i][ 8];
s_theMagicArray[i][ 9] = theMagicArray[i][ 9];
s_theMagicArray[i][10] = theMagicArray[i][10];
s_theMagicArray[i][11] = theMagicArray[i][11];
s_theMagicArray[i][12] = theMagicArray[i][12];
s_theMagicArray[i][13] = theMagicArray[i][13];
s_theMagicArray[i][14] = theMagicArray[i][14];
s_theMagicArray[i][15] = theMagicArray[i][15];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a (*s_theMagicArray)[16] = theMagicArray;
#endif
if (gid >= gid_max) return;
@ -673,8 +895,6 @@ KERNEL_FQ void m07801_s08 (KERN_ATTR_BASIC ())
* modifier
*/
//const u64 lid = get_local_id (0);
u32 w0[4];
w0[0] = pws[gid].i[ 0];
@ -709,7 +929,7 @@ KERNEL_FQ void m07801_s08 (KERN_ATTR_BASIC ())
* main
*/
m07801s (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);
m07801s (s_theMagicArray, 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);
}
KERNEL_FQ void m07801_s16 (KERN_ATTR_BASIC ())

16
OpenCL/m07900-pure.cl
View File

@ -66,18 +66,6 @@ KERNEL_FQ void m07900_loop (KERN_ATTR_TMPS (drupal7_tmp_t))
const u32 pw_len = pws[gid].pw_len;
u32 w[64] = { 0 };
for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1)
{
w[idx] = pws[gid].i[idx];
}
for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1)
{
w[idx] = hc_swap32_S (w[idx]);
}
/**
* load
*/
@ -120,7 +108,7 @@ KERNEL_FQ void m07900_loop (KERN_ATTR_TMPS (drupal7_tmp_t))
sha512_ctx.len = 64;
sha512_update (&sha512_ctx, w, pw_len);
sha512_update_global_swap (&sha512_ctx, pws[gid].i, pw_len);
sha512_final (&sha512_ctx);
@ -158,7 +146,7 @@ KERNEL_FQ void m07900_loop (KERN_ATTR_TMPS (drupal7_tmp_t))
sha512_ctx.len = 64;
sha512_update (&sha512_ctx, w, pw_len);
sha512_update_global_swap (&sha512_ctx, pws[gid].i, pw_len);
sha512_final (&sha512_ctx);

45
OpenCL/m08600_a0-pure.cl
View File

@ -16,7 +16,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -50,6 +52,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#if VECT_SIZE == 1
@ -244,9 +279,9 @@ KERNEL_FQ void m08600_mxx (KERN_ATTR_RULES ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -308,9 +343,9 @@ KERNEL_FQ void m08600_sxx (KERN_ATTR_RULES ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

45
OpenCL/m08600_a1-pure.cl
View File

@ -14,7 +14,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -48,6 +50,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#if VECT_SIZE == 1
@ -242,9 +277,9 @@ KERNEL_FQ void m08600_mxx (KERN_ATTR_BASIC ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -366,9 +401,9 @@ KERNEL_FQ void m08600_sxx (KERN_ATTR_BASIC ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

45
OpenCL/m08600_a3-pure.cl
View File

@ -13,7 +13,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -47,6 +49,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#if VECT_SIZE == 1
@ -347,9 +382,9 @@ KERNEL_FQ void m08600_mxx (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -404,9 +439,9 @@ KERNEL_FQ void m08600_sxx (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

55
OpenCL/m08700_a0-optimized.cl
View File

@ -16,7 +16,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -50,6 +52,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#if VECT_SIZE == 1
@ -89,17 +124,17 @@ DECLSPEC void lotus_mix (u32x *in, LOCAL_AS u32 *s_lotus_magic_table)
u32x tmp_in = in[j];
u32x tmp_out = 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
p = p + s--; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = p + s--; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = p + s--; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = p + s--; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
in[j] = tmp_out;
}
}
}
DECLSPEC void lotus_transform_password (u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
DECLSPEC void lotus_transform_password (const u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
{
u32x t = out[3] >> 24;
@ -281,9 +316,9 @@ KERNEL_FQ void m08700_m04 (KERN_ATTR_RULES ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -468,9 +503,9 @@ KERNEL_FQ void m08700_s04 (KERN_ATTR_RULES ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

55
OpenCL/m08700_a1-optimized.cl
View File

@ -14,7 +14,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -48,6 +50,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#if VECT_SIZE == 1
@ -87,17 +122,17 @@ DECLSPEC void lotus_mix (u32x *in, LOCAL_AS u32 *s_lotus_magic_table)
u32x tmp_in = in[j];
u32x tmp_out = 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
p = p + s--; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = p + s--; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = p + s--; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = p + s--; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
in[j] = tmp_out;
}
}
}
DECLSPEC void lotus_transform_password (u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
DECLSPEC void lotus_transform_password (const u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
{
u32x t = out[3] >> 24;
@ -279,9 +314,9 @@ KERNEL_FQ void m08700_m04 (KERN_ATTR_BASIC ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -526,9 +561,9 @@ KERNEL_FQ void m08700_s04 (KERN_ATTR_BASIC ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

71
OpenCL/m08700_a3-optimized.cl
View File

@ -13,7 +13,9 @@
#include "inc_simd.cl"
#endif
CONSTANT_VK u32a lotus_magic_table[256] =
// we just double this buffer so we can safe the & 0xff ;)
CONSTANT_VK u32a lotus_magic_table[512] =
{
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
@ -47,6 +49,39 @@ CONSTANT_VK u32a lotus_magic_table[256] =
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
};
#define BOX(S,i) (S)[(i)]
@ -88,17 +123,17 @@ DECLSPEC void lotus_mix (u32x *in, LOCAL_AS u32 *s_lotus_magic_table)
u32x tmp_in = in[j];
u32x tmp_out = 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
p = p + s--; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
p = p + s--; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
p = p + s--; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
p = p + s--; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
in[j] = tmp_out;
}
}
}
DECLSPEC void lotus_transform_password (u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
DECLSPEC void lotus_transform_password (const u32x *in, u32x *out, LOCAL_AS u32 *s_lotus_magic_table)
{
u32x t = out[3] >> 24;
@ -558,9 +593,9 @@ KERNEL_FQ void m08700_m04 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -626,9 +661,9 @@ KERNEL_FQ void m08700_m08 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -694,9 +729,9 @@ KERNEL_FQ void m08700_m16 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -762,9 +797,9 @@ KERNEL_FQ void m08700_s04 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -830,9 +865,9 @@ KERNEL_FQ void m08700_s08 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}
@ -898,9 +933,9 @@ KERNEL_FQ void m08700_s16 (KERN_ATTR_VECTOR ())
* sbox
*/
LOCAL_VK u32 s_lotus_magic_table[256];
LOCAL_VK u32 s_lotus_magic_table[512];
for (u32 i = lid; i < 256; i += lsz)
for (u32 i = lid; i < 512; i += lsz)
{
s_lotus_magic_table[i] = lotus_magic_table[i];
}

62
OpenCL/m09100-pure.cl
View File

@ -27,6 +27,42 @@ typedef struct lotus8_tmp
} lotus8_tmp_t;
CONSTANT_VK u32a bin2asc[256] =
{
0x00003030, 0x00003130, 0x00003230, 0x00003330, 0x00003430, 0x00003530, 0x00003630, 0x00003730,
0x00003830, 0x00003930, 0x00004130, 0x00004230, 0x00004330, 0x00004430, 0x00004530, 0x00004630,
0x00003031, 0x00003131, 0x00003231, 0x00003331, 0x00003431, 0x00003531, 0x00003631, 0x00003731,
0x00003831, 0x00003931, 0x00004131, 0x00004231, 0x00004331, 0x00004431, 0x00004531, 0x00004631,
0x00003032, 0x00003132, 0x00003232, 0x00003332, 0x00003432, 0x00003532, 0x00003632, 0x00003732,
0x00003832, 0x00003932, 0x00004132, 0x00004232, 0x00004332, 0x00004432, 0x00004532, 0x00004632,
0x00003033, 0x00003133, 0x00003233, 0x00003333, 0x00003433, 0x00003533, 0x00003633, 0x00003733,
0x00003833, 0x00003933, 0x00004133, 0x00004233, 0x00004333, 0x00004433, 0x00004533, 0x00004633,
0x00003034, 0x00003134, 0x00003234, 0x00003334, 0x00003434, 0x00003534, 0x00003634, 0x00003734,
0x00003834, 0x00003934, 0x00004134, 0x00004234, 0x00004334, 0x00004434, 0x00004534, 0x00004634,
0x00003035, 0x00003135, 0x00003235, 0x00003335, 0x00003435, 0x00003535, 0x00003635, 0x00003735,
0x00003835, 0x00003935, 0x00004135, 0x00004235, 0x00004335, 0x00004435, 0x00004535, 0x00004635,
0x00003036, 0x00003136, 0x00003236, 0x00003336, 0x00003436, 0x00003536, 0x00003636, 0x00003736,
0x00003836, 0x00003936, 0x00004136, 0x00004236, 0x00004336, 0x00004436, 0x00004536, 0x00004636,
0x00003037, 0x00003137, 0x00003237, 0x00003337, 0x00003437, 0x00003537, 0x00003637, 0x00003737,
0x00003837, 0x00003937, 0x00004137, 0x00004237, 0x00004337, 0x00004437, 0x00004537, 0x00004637,
0x00003038, 0x00003138, 0x00003238, 0x00003338, 0x00003438, 0x00003538, 0x00003638, 0x00003738,
0x00003838, 0x00003938, 0x00004138, 0x00004238, 0x00004338, 0x00004438, 0x00004538, 0x00004638,
0x00003039, 0x00003139, 0x00003239, 0x00003339, 0x00003439, 0x00003539, 0x00003639, 0x00003739,
0x00003839, 0x00003939, 0x00004139, 0x00004239, 0x00004339, 0x00004439, 0x00004539, 0x00004639,
0x00003041, 0x00003141, 0x00003241, 0x00003341, 0x00003441, 0x00003541, 0x00003641, 0x00003741,
0x00003841, 0x00003941, 0x00004141, 0x00004241, 0x00004341, 0x00004441, 0x00004541, 0x00004641,
0x00003042, 0x00003142, 0x00003242, 0x00003342, 0x00003442, 0x00003542, 0x00003642, 0x00003742,
0x00003842, 0x00003942, 0x00004142, 0x00004242, 0x00004342, 0x00004442, 0x00004542, 0x00004642,
0x00003043, 0x00003143, 0x00003243, 0x00003343, 0x00003443, 0x00003543, 0x00003643, 0x00003743,
0x00003843, 0x00003943, 0x00004143, 0x00004243, 0x00004343, 0x00004443, 0x00004543, 0x00004643,
0x00003044, 0x00003144, 0x00003244, 0x00003344, 0x00003444, 0x00003544, 0x00003644, 0x00003744,
0x00003844, 0x00003944, 0x00004144, 0x00004244, 0x00004344, 0x00004444, 0x00004544, 0x00004644,
0x00003045, 0x00003145, 0x00003245, 0x00003345, 0x00003445, 0x00003545, 0x00003645, 0x00003745,
0x00003845, 0x00003945, 0x00004145, 0x00004245, 0x00004345, 0x00004445, 0x00004545, 0x00004645,
0x00003046, 0x00003146, 0x00003246, 0x00003346, 0x00003446, 0x00003546, 0x00003646, 0x00003746,
0x00003846, 0x00003946, 0x00004146, 0x00004246, 0x00004346, 0x00004446, 0x00004546, 0x00004646,
};
CONSTANT_VK u32a lotus64_table[64] =
{
'0', '1', '2', '3', '4', '5', '6', '7',
@ -79,7 +115,7 @@ CONSTANT_VK u32a lotus_magic_table[256] =
#define BOX1(S,i) (S)[(i)]
DECLSPEC void lotus_mix (u32 *in, LOCAL_AS const u32 *s_lotus_magic_table)
DECLSPEC void lotus_mix (u32 *in, SHM_TYPE const u32 *s_lotus_magic_table)
{
u8 p = 0;
@ -102,7 +138,7 @@ DECLSPEC void lotus_mix (u32 *in, LOCAL_AS const u32 *s_lotus_magic_table)
}
}
DECLSPEC void lotus_transform_password (const u32 *in, u32 *out, LOCAL_AS const u32 *s_lotus_magic_table)
DECLSPEC void lotus_transform_password (const u32 *in, u32 *out, SHM_TYPE const u32 *s_lotus_magic_table)
{
u8 t = (u8) (out[3] >> 24);
@ -199,7 +235,7 @@ DECLSPEC void pad (u32 *w, const u32 len)
}
}
DECLSPEC void mdtransform_norecalc (u32 *state, const u32 *block, LOCAL_AS const u32 *s_lotus_magic_table)
DECLSPEC void mdtransform_norecalc (u32 *state, const u32 *block, SHM_TYPE const u32 *s_lotus_magic_table)
{
u32 x[12];
@ -224,14 +260,14 @@ DECLSPEC void mdtransform_norecalc (u32 *state, const u32 *block, LOCAL_AS const
state[3] = x[3];
}
DECLSPEC void mdtransform (u32 *state, u32 *checksum, const u32 *block, LOCAL_AS const u32 *s_lotus_magic_table)
DECLSPEC void mdtransform (u32 *state, u32 *checksum, const u32 *block, SHM_TYPE const u32 *s_lotus_magic_table)
{
mdtransform_norecalc (state, block, s_lotus_magic_table);
lotus_transform_password (block, checksum, s_lotus_magic_table);
}
DECLSPEC void domino_big_md (const u32 *saved_key, const u32 size, u32 *state, LOCAL_AS const u32 *s_lotus_magic_table)
DECLSPEC void domino_big_md (const u32 *saved_key, const u32 size, u32 *state, SHM_TYPE const u32 *s_lotus_magic_table)
{
u32 checksum[4];
@ -402,6 +438,8 @@ KERNEL_FQ void m09100_init (KERN_ATTR_TMPS (lotus8_tmp_t))
* sbox
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_lotus_magic_table[256];
for (u32 i = lid; i < 256; i += lsz)
@ -413,15 +451,19 @@ KERNEL_FQ void m09100_init (KERN_ATTR_TMPS (lotus8_tmp_t))
for (u32 i = lid; i < 256; i += lsz)
{
const u32 i0 = (i >> 0) & 15;
const u32 i1 = (i >> 4) & 15;
l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'A' - 10 + i0) << 8
| ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
l_bin2asc[i] = bin2asc[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_lotus_magic_table = lotus_magic_table;
CONSTANT_AS u32a *l_bin2asc = bin2asc;
#endif
if (gid >= gid_max) return;
/**

11
OpenCL/m09500-pure.cl
View File

@ -271,12 +271,15 @@ KERNEL_FQ void m09500_comp (KERN_ATTR_TMPS_ESALT (office2010_tmp_t, office2010_t
AES128_set_decrypt_key (ks, ukey, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3);
const u32 digest_cur = digests_offset + loop_pos;
u32 data[4];
data[0] = esalt_bufs[digests_offset].encryptedVerifier[0];
data[1] = esalt_bufs[digests_offset].encryptedVerifier[1];
data[2] = esalt_bufs[digests_offset].encryptedVerifier[2];
data[3] = esalt_bufs[digests_offset].encryptedVerifier[3];
data[0] = esalt_bufs[digest_cur].encryptedVerifier[0];
data[1] = esalt_bufs[digest_cur].encryptedVerifier[1];
data[2] = esalt_bufs[digest_cur].encryptedVerifier[2];
data[3] = esalt_bufs[digest_cur].encryptedVerifier[3];
u32 out[4];

231
OpenCL/m09800_a0-optimized.cl
View File

@ -17,11 +17,15 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
@ -252,21 +256,21 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -279,6 +283,8 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -337,7 +343,93 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
int digest_pos = find_hash (out, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
@ -464,21 +556,21 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -491,6 +583,8 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -549,7 +643,92 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
if (out[0] != search[0]) continue;
if (out[1] != search[1]) continue;
if (out[2] != search[2]) continue;
if (out[3] != search[3]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}

231
OpenCL/m09800_a1-optimized.cl
View File

@ -15,11 +15,15 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
@ -300,21 +304,21 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -327,6 +331,8 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -385,7 +391,93 @@ KERNEL_FQ void m09800_m04 (KERN_ATTR_ESALT (oldoffice34_t))
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
int digest_pos = find_hash (out, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
@ -562,21 +654,21 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -589,6 +681,8 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -647,7 +741,92 @@ KERNEL_FQ void m09800_s04 (KERN_ATTR_ESALT (oldoffice34_t))
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
if (out[0] != search[0]) continue;
if (out[1] != search[1]) continue;
if (out[2] != search[2]) continue;
if (out[3] != search[3]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}

231
OpenCL/m09800_a3-optimized.cl
View File

@ -12,11 +12,15 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
@ -223,21 +227,21 @@ DECLSPEC void m09800m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
w3_t[2] = 0;
w3_t[3] = pw_salt_len * 8;
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, pass_hash);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
w0_t[2] = digest[2];
w0_t[3] = digest[3];
w1_t[0] = digest[4];
w0_t[0] = pass_hash[0];
w0_t[1] = pass_hash[1];
w0_t[2] = pass_hash[2];
w0_t[3] = pass_hash[3];
w1_t[0] = pass_hash[4];
w1_t[1] = 0;
w1_t[2] = 0x80000000;
w1_t[3] = 0;
@ -250,6 +254,8 @@ DECLSPEC void m09800m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
w3_t[2] = 0;
w3_t[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -308,7 +314,93 @@ DECLSPEC void m09800m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
int digest_pos = find_hash (out, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
@ -403,21 +495,21 @@ DECLSPEC void m09800s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
w3_t[2] = 0;
w3_t[3] = pw_salt_len * 8;
u32 digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, pass_hash);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
w0_t[2] = digest[2];
w0_t[3] = digest[3];
w1_t[0] = digest[4];
w0_t[0] = pass_hash[0];
w0_t[1] = pass_hash[1];
w0_t[2] = pass_hash[2];
w0_t[3] = pass_hash[3];
w1_t[0] = pass_hash[4];
w1_t[1] = 0;
w1_t[2] = 0x80000000;
w1_t[3] = 0;
@ -430,6 +522,8 @@ DECLSPEC void m09800s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
w3_t[2] = 0;
w3_t[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
@ -488,7 +582,92 @@ DECLSPEC void m09800s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
rc4_next_16 (rc4_key, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
// initial compare
if (out[0] != search[0]) continue;
if (out[1] != search[1]) continue;
if (out[2] != search[2]) continue;
if (out[3] != search[3]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}

2
OpenCL/m09810_a0-optimized.cl
View File

@ -22,6 +22,8 @@ typedef struct oldoffice34
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;

2
OpenCL/m09810_a1-optimized.cl
View File

@ -20,6 +20,8 @@ typedef struct oldoffice34
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;

2
OpenCL/m09810_a3-optimized.cl
View File

@ -20,6 +20,8 @@ typedef struct oldoffice34
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;

379
OpenCL/m09820_a0-optimized.cl
View File

@ -3,7 +3,8 @@
* License.....: MIT
*/
#define NEW_SIMD_CODE
//too much register pressure
//#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -16,15 +17,142 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
KERNEL_FQ void m09820_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
{
/**
@ -55,6 +183,14 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
/**
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -107,21 +243,21 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -134,20 +270,110 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_M_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
int digest_pos = find_hash (digest, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
@ -189,6 +415,14 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
/**
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -253,21 +487,21 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -280,20 +514,107 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_RULES_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_S_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
if (digest[0] != search[0]) continue;
if (digest[1] != search[1]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}

379
OpenCL/m09820_a1-optimized.cl
View File

@ -3,7 +3,8 @@
* License.....: MIT
*/
#define NEW_SIMD_CODE
//too much register pressure
//#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -14,15 +15,142 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
KERNEL_FQ void m09820_m04 (KERN_ATTR_ESALT (oldoffice34_t))
{
/**
@ -53,6 +181,14 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -155,21 +291,21 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -182,20 +318,110 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_M_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
int digest_pos = find_hash (digest, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
@ -237,6 +463,14 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -351,21 +585,21 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_ESALT (oldoffice34_t))
w0[1] = salt_buf[1];
w0[0] = salt_buf[0];
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, pass_hash);
w0[0] = digest[0];
w0[1] = digest[1];
w0[2] = digest[2];
w0[3] = digest[3];
w1[0] = digest[4];
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0;
w1[2] = 0x80000000;
w1[3] = 0;
@ -378,20 +612,107 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_ESALT (oldoffice34_t))
w3[2] = 0;
w3[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0, w1, w2, w3, digest);
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_S_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
if (digest[0] != search[0]) continue;
if (digest[1] != search[1]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}

427
OpenCL/m09820_a3-optimized.cl
View File

@ -3,7 +3,8 @@
* License.....: MIT
*/
#define NEW_SIMD_CODE
//too much register pressure
//#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -14,16 +15,143 @@
#include "inc_hash_sha1.cl"
#endif
#define MIN_NULL_BYTES 10
typedef struct oldoffice34
{
u32 version;
u32 encryptedVerifier[4];
u32 encryptedVerifierHash[5];
u32 secondBlockData[8];
u32 secondBlockLen;
u32 rc4key[2];
} oldoffice34_t;
DECLSPEC void m09820m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice34_t))
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
DECLSPEC void m09820m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice34_t))
{
/**
* modifier
@ -32,6 +160,12 @@ DECLSPEC void m09820m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -81,21 +215,21 @@ DECLSPEC void m09820m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
w3_t[2] = 0;
w3_t[3] = (pw_len + 16) * 8;
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, pass_hash);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
w0_t[2] = digest[2];
w0_t[3] = digest[3];
w1_t[0] = digest[4];
w0_t[0] = pass_hash[0];
w0_t[1] = pass_hash[1];
w0_t[2] = pass_hash[2];
w0_t[3] = pass_hash[3];
w1_t[0] = pass_hash[4];
w1_t[1] = 0;
w1_t[2] = 0x80000000;
w1_t[3] = 0;
@ -108,24 +242,114 @@ DECLSPEC void m09820m (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
w3_t[2] = 0;
w3_t[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_M_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
int digest_pos = find_hash (digest, digests_cnt, &digests_buf[digests_offset]);
if (digest_pos == -1) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
const u32 final_hash_pos = digests_offset + digest_pos;
if (atomic_inc (&hashes_shown[final_hash_pos]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, digest_pos, final_hash_pos, gid, il_pos, 0, 0);
}
}
}
DECLSPEC void m09820s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice34_t))
DECLSPEC void m09820s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice34_t))
{
/**
* modifier
@ -134,6 +358,12 @@ DECLSPEC void m09820s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -195,21 +425,21 @@ DECLSPEC void m09820s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
w3_t[2] = 0;
w3_t[3] = (pw_len + 16) * 8;
u32x digest[5];
u32 pass_hash[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
pass_hash[0] = SHA1M_A;
pass_hash[1] = SHA1M_B;
pass_hash[2] = SHA1M_C;
pass_hash[3] = SHA1M_D;
pass_hash[4] = SHA1M_E;
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, pass_hash);
w0_t[0] = digest[0];
w0_t[1] = digest[1];
w0_t[2] = digest[2];
w0_t[3] = digest[3];
w1_t[0] = digest[4];
w0_t[0] = pass_hash[0];
w0_t[1] = pass_hash[1];
w0_t[2] = pass_hash[2];
w0_t[3] = pass_hash[3];
w1_t[0] = pass_hash[4];
w1_t[1] = 0;
w1_t[2] = 0x80000000;
w1_t[3] = 0;
@ -222,20 +452,107 @@ DECLSPEC void m09820s (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KER
w3_t[2] = 0;
w3_t[3] = (20 + 4) * 8;
u32 digest[5];
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform_vector (w0_t, w1_t, w2_t, w3_t, digest);
sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
digest[0] = hc_swap32 (digest[0]);
digest[1] = hc_swap32 (digest[1]) & 0xff;
digest[2] = 0;
digest[3] = 0;
COMPARE_S_SIMD (digest[0], digest[1], digest[2], digest[3]);
// initial compare
if (digest[0] != search[0]) continue;
if (digest[1] != search[1]) continue;
if (esalt_bufs[digests_offset].secondBlockLen != 0)
{
w0[0] = pass_hash[0];
w0[1] = pass_hash[1];
w0[2] = pass_hash[2];
w0[3] = pass_hash[3];
w1[0] = pass_hash[4];
w1[1] = 0x01000000;
w1[2] = 0x80000000;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (20 + 4) * 8;
digest[0] = SHA1M_A;
digest[1] = SHA1M_B;
digest[2] = SHA1M_C;
digest[3] = SHA1M_D;
digest[4] = SHA1M_E;
sha1_transform (w0, w1, w2, w3, digest);
digest[0] = hc_swap32_S (digest[0]);
digest[1] = hc_swap32_S (digest[1]);
digest[2] = 0;
digest[3] = 0;
digest[1] &= 0xff; // only 40-bit key
// second block decrypt:
rc4_init_16 (rc4_key, digest);
u32 secondBlockData[4];
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[0];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[1];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[2];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[3];
u32 out[4];
u32 j = rc4_next_16 (rc4_key, 0, 0, secondBlockData, out);
int null_bytes = 0;
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
secondBlockData[0] = esalt_bufs[digests_offset].secondBlockData[4];
secondBlockData[1] = esalt_bufs[digests_offset].secondBlockData[5];
secondBlockData[2] = esalt_bufs[digests_offset].secondBlockData[6];
secondBlockData[3] = esalt_bufs[digests_offset].secondBlockData[7];
rc4_next_16 (rc4_key, 16, j, secondBlockData, out);
for (int k = 0; k < 4; k++)
{
if ((out[k] & 0x000000ff) == 0) null_bytes++;
if ((out[k] & 0x0000ff00) == 0) null_bytes++;
if ((out[k] & 0x00ff0000) == 0) null_bytes++;
if ((out[k] & 0xff000000) == 0) null_bytes++;
}
if (null_bytes < MIN_NULL_BYTES) continue;
}
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, il_pos, 0, 0);
}
}
}
@ -279,7 +596,13 @@ KERNEL_FQ void m09820_m04 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820m (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820m (rc4_keys, 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);
}
KERNEL_FQ void m09820_m08 (KERN_ATTR_ESALT (oldoffice34_t))
@ -322,7 +645,13 @@ KERNEL_FQ void m09820_m08 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820m (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820m (rc4_keys, 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);
}
KERNEL_FQ void m09820_m16 (KERN_ATTR_ESALT (oldoffice34_t))
@ -365,7 +694,13 @@ KERNEL_FQ void m09820_m16 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820m (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820m (rc4_keys, 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);
}
KERNEL_FQ void m09820_s04 (KERN_ATTR_ESALT (oldoffice34_t))
@ -408,7 +743,13 @@ KERNEL_FQ void m09820_s04 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820s (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820s (rc4_keys, 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);
}
KERNEL_FQ void m09820_s08 (KERN_ATTR_ESALT (oldoffice34_t))
@ -451,7 +792,13 @@ KERNEL_FQ void m09820_s08 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820s (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820s (rc4_keys, 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);
}
KERNEL_FQ void m09820_s16 (KERN_ATTR_ESALT (oldoffice34_t))
@ -494,5 +841,11 @@ KERNEL_FQ void m09820_s16 (KERN_ATTR_ESALT (oldoffice34_t))
const u32 pw_len = pws[gid].pw_len & 63;
m09820s (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);
/**
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
m09820s (rc4_keys, 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);
}

68
OpenCL/m10700-pure.cl
View File

@ -257,8 +257,10 @@ DECLSPEC void sha256_final_aes (sha256_ctx_t *ctx, const u32 *aes_ks, u32 *aes_i
sha256_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->h);
}
DECLSPEC void sha384_update_aes_128 (sha384_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *w4, u32 *w5, u32 *w6, u32 *w7, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC u32 sha384_update_aes_128 (sha384_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *w4, u32 *w5, u32 *w6, u32 *w7, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{
u32 ex = 0;
MAYBE_VOLATILE const int pos = ctx->len & 127;
ctx->len += len;
@ -355,6 +357,8 @@ DECLSPEC void sha384_update_aes_128 (sha384_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w6, ctx->w6, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w7, ctx->w7, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = ctx->w7[3] & 0xff;
sha384_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
ctx->w0[0] = c0[0];
@ -390,6 +394,8 @@ DECLSPEC void sha384_update_aes_128 (sha384_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w
ctx->w7[2] = c7[2];
ctx->w7[3] = c7[3];
}
return ex;
}
DECLSPEC void sha384_update_aes (sha384_ctx_t *ctx, const u32 *w, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
@ -480,14 +486,21 @@ DECLSPEC void sha384_update_aes (sha384_ctx_t *ctx, const u32 *w, const int len,
sha384_update_aes_128 (ctx, w0, w1, w2, w3, w4, w5, w6, w7, len - pos1, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
}
DECLSPEC void sha384_final_aes (sha384_ctx_t *ctx, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC u32 sha384_final_aes (sha384_ctx_t *ctx, const u32 ex, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{
u32 ret = ex;
int pos = ctx->len & 127;
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w0, ctx->w0, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w1, ctx->w1, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w2, ctx->w2, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w3, ctx->w3, s_te0, s_te1, s_te2, s_te3, s_te4);
if (pos)
{
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w0, ctx->w0, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w1, ctx->w1, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w2, ctx->w2, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w3, ctx->w3, s_te0, s_te1, s_te2, s_te3, s_te4);
ret = ctx->w3[3] & 0xff;
}
append_0x80_8x4_S (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, pos ^ 3);
@ -533,10 +546,14 @@ DECLSPEC void sha384_final_aes (sha384_ctx_t *ctx, const u32 *aes_ks, u32 *aes_i
ctx->w7[3] = ctx->len * 8;
sha384_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
return ret;
}
DECLSPEC void sha512_update_aes_128 (sha512_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *w4, u32 *w5, u32 *w6, u32 *w7, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC u32 sha512_update_aes_128 (sha512_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *w4, u32 *w5, u32 *w6, u32 *w7, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{
u32 ex = 0;
MAYBE_VOLATILE const int pos = ctx->len & 127;
ctx->len += len;
@ -633,6 +650,8 @@ DECLSPEC void sha512_update_aes_128 (sha512_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w6, ctx->w6, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w7, ctx->w7, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = ctx->w7[3] & 0xff;
sha512_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
ctx->w0[0] = c0[0];
@ -668,6 +687,8 @@ DECLSPEC void sha512_update_aes_128 (sha512_ctx_t *ctx, u32 *w0, u32 *w1, u32 *w
ctx->w7[2] = c7[2];
ctx->w7[3] = c7[3];
}
return ex;
}
DECLSPEC void sha512_update_aes (sha512_ctx_t *ctx, const u32 *w, const int len, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
@ -758,14 +779,21 @@ DECLSPEC void sha512_update_aes (sha512_ctx_t *ctx, const u32 *w, const int len,
sha512_update_aes_128 (ctx, w0, w1, w2, w3, w4, w5, w6, w7, len - pos1, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
}
DECLSPEC void sha512_final_aes (sha512_ctx_t *ctx, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
DECLSPEC u32 sha512_final_aes (sha512_ctx_t *ctx, const u32 ex, const u32 *aes_ks, u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
{
u32 ret = ex;
int pos = ctx->len & 127;
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w0, ctx->w0, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w1, ctx->w1, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w2, ctx->w2, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w3, ctx->w3, s_te0, s_te1, s_te2, s_te3, s_te4);
if (pos)
{
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w0, ctx->w0, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w1, ctx->w1, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w2, ctx->w2, s_te0, s_te1, s_te2, s_te3, s_te4);
aes128_encrypt_cbc (aes_ks, aes_iv, ctx->w3, ctx->w3, s_te0, s_te1, s_te2, s_te3, s_te4);
ret = ctx->w3[3] & 0xff;
}
append_0x80_8x4_S (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, pos ^ 3);
@ -811,6 +839,8 @@ DECLSPEC void sha512_final_aes (sha512_ctx_t *ctx, const u32 *aes_ks, u32 *aes_i
ctx->w7[3] = ctx->len * 8;
sha512_transform (ctx->w0, ctx->w1, ctx->w2, ctx->w3, ctx->w4, ctx->w5, ctx->w6, ctx->w7, ctx->h);
return ret;
}
DECLSPEC int find_sum (const u32 *w, const u32 pw_len, u32 *bb, const u32 *aes_ks, const u32 *aes_iv, SHM_TYPE u32 *s_te0, SHM_TYPE u32 *s_te1, SHM_TYPE u32 *s_te2, SHM_TYPE u32 *s_te3, SHM_TYPE u32 *s_te4)
@ -1040,12 +1070,10 @@ DECLSPEC u32 do_round (const u32 *w, const u32 pw_len, pdf17l8_tmp_t *tmp, SHM_T
w7[2] = 0;
w7[3] = 0;
sha384_update_aes_128 (&ctx384, w0, w1, w2, w3, w4, w5, w6, w7, tmp->dgst_len, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = sha384_update_aes_128 (&ctx384, w0, w1, w2, w3, w4, w5, w6, w7, tmp->dgst_len, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
}
sha384_final_aes (&ctx384, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = ctx384.w3[3] & 0xff;
ex = sha384_final_aes (&ctx384, ex, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
tmp->dgst32[ 0] = h32_from_64_S (ctx384.h[0]);
tmp->dgst32[ 1] = l32_from_64_S (ctx384.h[0]);
@ -1109,12 +1137,10 @@ DECLSPEC u32 do_round (const u32 *w, const u32 pw_len, pdf17l8_tmp_t *tmp, SHM_T
w7[2] = 0;
w7[3] = 0;
sha512_update_aes_128 (&ctx512, w0, w1, w2, w3, w4, w5, w6, w7, tmp->dgst_len, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = sha512_update_aes_128 (&ctx512, w0, w1, w2, w3, w4, w5, w6, w7, tmp->dgst_len, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
}
sha512_final_aes (&ctx512, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
ex = ctx512.w3[3] & 0xff;
ex = sha512_final_aes (&ctx512, ex, aes_ks, aes_iv, s_te0, s_te1, s_te2, s_te3, s_te4);
tmp->dgst32[ 0] = h32_from_64_S (ctx512.h[0]);
tmp->dgst32[ 1] = l32_from_64_S (ctx512.h[0]);
@ -1230,6 +1256,8 @@ KERNEL_FQ void m10700_loop (KERN_ATTR_TMPS_ESALT (pdf17l8_tmp_t, pdf_t))
const u32 pw_len = pws[gid].pw_len;
if (pw_len == 0) return;
u32 w[64] = { 0 };
for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1)

70
OpenCL/m11300-pure.cl
View File

@ -26,6 +26,9 @@ typedef struct bitcoin_wallet
u32 cry_master_buf[64];
u32 cry_master_len;
u32 cry_salt_buf[16];
u32 cry_salt_len;
} bitcoin_wallet_t;
DECLSPEC void hmac_sha512_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *w4, u32x *w5, u32x *w6, u32x *w7, u64x *ipad, u64x *opad, u64x *digest)
@ -293,47 +296,56 @@ KERNEL_FQ void m11300_comp (KERN_ATTR_TMPS_ESALT (bitcoin_wallet_tmp_t, bitcoin_
key[6] = h32_from_64_S (dgst[3]);
key[7] = l32_from_64_S (dgst[3]);
u32 iv[4];
iv[0] = h32_from_64_S (dgst[4]);
iv[1] = l32_from_64_S (dgst[4]);
iv[2] = h32_from_64_S (dgst[5]);
iv[3] = l32_from_64_S (dgst[5]);
#define KEYLEN 60
u32 ks[KEYLEN];
AES256_set_decrypt_key (ks, key, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3);
u32 i = esalt_bufs[digests_offset].cry_master_len - 32;
u32 iv[4];
iv[0] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 0]);
iv[1] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 1]);
iv[2] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 2]);
iv[3] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 3]);
i += 16;
u32 data[4];
data[0] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 0]);
data[1] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 1]);
data[2] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 2]);
data[3] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 3]);
u32 out[4];
for (u32 i = 0; i < esalt_bufs[digests_offset].cry_master_len; i += 16)
AES256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4);
out[0] ^= iv[0];
out[1] ^= iv[1];
out[2] ^= iv[2];
out[3] ^= iv[3];
u32 pad = 0;
if (esalt_bufs[digests_offset].cry_salt_len != 18)
{
u32 data[4];
/* most wallets */
pad = 0x10101010;
data[0] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 0]);
data[1] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 1]);
data[2] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 2]);
data[3] = hc_swap32_S (esalt_bufs[digests_offset].cry_master_buf[(i / 4) + 3]);
AES256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4);
out[0] ^= iv[0];
out[1] ^= iv[1];
out[2] ^= iv[2];
out[3] ^= iv[3];
iv[0] = data[0];
iv[1] = data[1];
iv[2] = data[2];
iv[3] = data[3];
if (out[0] != pad) return;
if (out[1] != pad) return;
}
else
{
/* Nexus legacy wallet */
pad = 0x08080808;
}
if ((out[0] == 0x10101010)
&& (out[1] == 0x10101010)
&& (out[2] == 0x10101010)
&& (out[3] == 0x10101010))
if (out[2] == pad && out[3] == pad)
{
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{

255
OpenCL/m11600-optimized.cl Normal file
View File

@ -0,0 +1,255 @@
/**
* Author......: See docs/credits.txt
* License.....: MIT
*/
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
#include "inc_types.h"
#include "inc_platform.cl"
#include "inc_common.cl"
#include "inc_hash_sha256.cl"
#endif
typedef struct seven_zip_tmp
{
u32 h[8];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
int len;
} seven_zip_tmp_t;
typedef struct
{
u32 ukey[8];
u32 hook_success;
} seven_zip_hook_t;
#define PUTCHAR(a,p,c) ((u8 *)(a))[(p)] = (u8) (c)
#define GETCHAR(a,p) ((u8 *)(a))[(p)]
#define PUTCHAR_BE(a,p,c) ((u8 *)(a))[(p) ^ 3] = (u8) (c)
#define GETCHAR_BE(a,p) ((u8 *)(a))[(p) ^ 3]
#define MIN(a,b) (((a) < (b)) ? (a) : (b))
KERNEL_FQ void m11600_init (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_hook_t))
{
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
tmps[gid].h[0] = SHA256M_A;
tmps[gid].h[1] = SHA256M_B;
tmps[gid].h[2] = SHA256M_C;
tmps[gid].h[3] = SHA256M_D;
tmps[gid].h[4] = SHA256M_E;
tmps[gid].h[5] = SHA256M_F;
tmps[gid].h[6] = SHA256M_G;
tmps[gid].h[7] = SHA256M_H;
tmps[gid].len = 0;
}
KERNEL_FQ void m11600_loop (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_hook_t))
{
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf[5];
pw_buf[0] = pws[gid].i[0];
pw_buf[1] = pws[gid].i[1];
pw_buf[2] = pws[gid].i[2];
pw_buf[3] = pws[gid].i[3];
pw_buf[4] = pws[gid].i[4];
const u32 pw_len = MIN (pws[gid].pw_len, 20);
// this is large enough to hold all possible w[] arrays for 64 iterations
#define LARGEBLOCK_ELEMS ((40 + 8) * 16)
u32 largeblock[LARGEBLOCK_ELEMS];
u8 *ptr = (u8 *) largeblock;
for (u32 i = 0; i < LARGEBLOCK_ELEMS; i++) largeblock[i] = 0;
u32 loop_pos_pos = loop_pos;
for (u32 i = 0, p = 0; i < 64; i++)
{
for (u32 j = 0; j < pw_len; j++, p += 2)
{
PUTCHAR_BE (largeblock, p, GETCHAR (pw_buf, j));
}
const u8 byte2 = unpack_v8c_from_v32_S (loop_pos_pos);
const u8 byte3 = unpack_v8d_from_v32_S (loop_pos_pos);
PUTCHAR_BE (largeblock, p + 2, byte2);
PUTCHAR_BE (largeblock, p + 3, byte3);
loop_pos_pos++;
p += 8;
}
u32 h[8];
h[0] = tmps[gid].h[0];
h[1] = tmps[gid].h[1];
h[2] = tmps[gid].h[2];
h[3] = tmps[gid].h[3];
h[4] = tmps[gid].h[4];
h[5] = tmps[gid].h[5];
h[6] = tmps[gid].h[6];
h[7] = tmps[gid].h[7];
const int iter64 = (pw_len * 2) + 8;
loop_pos_pos = loop_pos;
for (u32 i = 0; i < loop_cnt; i += 64)
{
// iteration set
for (u32 i = 0, p = pw_len * 2; i < 64; i++, p += iter64)
{
const u8 byte0 = unpack_v8a_from_v32_S (loop_pos_pos);
const u8 byte1 = unpack_v8b_from_v32_S (loop_pos_pos);
PUTCHAR_BE (largeblock, p + 0, byte0);
PUTCHAR_BE (largeblock, p + 1, byte1);
loop_pos_pos++;
}
// full 64 byte buffer
for (int j = 0, j16 = 0; j < iter64; j++, j16 += 16)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
w0[0] = largeblock[j16 + 0];
w0[1] = largeblock[j16 + 1];
w0[2] = largeblock[j16 + 2];
w0[3] = largeblock[j16 + 3];
w1[0] = largeblock[j16 + 4];
w1[1] = largeblock[j16 + 5];
w1[2] = largeblock[j16 + 6];
w1[3] = largeblock[j16 + 7];
w2[0] = largeblock[j16 + 8];
w2[1] = largeblock[j16 + 9];
w2[2] = largeblock[j16 + 10];
w2[3] = largeblock[j16 + 11];
w3[0] = largeblock[j16 + 12];
w3[1] = largeblock[j16 + 13];
w3[2] = largeblock[j16 + 14];
w3[3] = largeblock[j16 + 15];
sha256_transform (w0, w1, w2, w3, h);
}
}
tmps[gid].len += loop_cnt * iter64;
tmps[gid].h[0] = h[0];
tmps[gid].h[1] = h[1];
tmps[gid].h[2] = h[2];
tmps[gid].h[3] = h[3];
tmps[gid].h[4] = h[4];
tmps[gid].h[5] = h[5];
tmps[gid].h[6] = h[6];
tmps[gid].h[7] = h[7];
}
KERNEL_FQ void m11600_hook23 (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_hook_t))
{
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
/**
* context load
*/
u32 h[8];
h[0] = tmps[gid].h[0];
h[1] = tmps[gid].h[1];
h[2] = tmps[gid].h[2];
h[3] = tmps[gid].h[3];
h[4] = tmps[gid].h[4];
h[5] = tmps[gid].h[5];
h[6] = tmps[gid].h[6];
h[7] = tmps[gid].h[7];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
w0[0] = 0x80000000;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = tmps[gid].len * 8;
sha256_transform (w0, w1, w2, w3, h);
hooks[gid].ukey[0] = hc_swap32_S (h[0]);
hooks[gid].ukey[1] = hc_swap32_S (h[1]);
hooks[gid].ukey[2] = hc_swap32_S (h[2]);
hooks[gid].ukey[3] = hc_swap32_S (h[3]);
hooks[gid].ukey[4] = hc_swap32_S (h[4]);
hooks[gid].ukey[5] = hc_swap32_S (h[5]);
hooks[gid].ukey[6] = hc_swap32_S (h[6]);
hooks[gid].ukey[7] = hc_swap32_S (h[7]);
}
KERNEL_FQ void m11600_comp (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_hook_t))
{
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
if (hooks[gid].hook_success == 1)
{
if (atomic_inc (&hashes_shown[digests_offset]) == 0)
{
mark_hash (plains_buf, d_return_buf, salt_pos, digests_cnt, 0, digests_offset + 0, gid, 0, 0, 0);
}
return;
}
}

73
OpenCL/m11600-pure.cl
View File

@ -289,46 +289,49 @@ KERNEL_FQ void m11600_hook23 (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_h
* context load
*/
sha256_ctx_t ctx;
u32 h[8];
ctx.h[0] = tmps[gid].h[0];
ctx.h[1] = tmps[gid].h[1];
ctx.h[2] = tmps[gid].h[2];
ctx.h[3] = tmps[gid].h[3];
ctx.h[4] = tmps[gid].h[4];
ctx.h[5] = tmps[gid].h[5];
ctx.h[6] = tmps[gid].h[6];
ctx.h[7] = tmps[gid].h[7];
h[0] = tmps[gid].h[0];
h[1] = tmps[gid].h[1];
h[2] = tmps[gid].h[2];
h[3] = tmps[gid].h[3];
h[4] = tmps[gid].h[4];
h[5] = tmps[gid].h[5];
h[6] = tmps[gid].h[6];
h[7] = tmps[gid].h[7];
ctx.w0[0] = tmps[gid].w0[0];
ctx.w0[1] = tmps[gid].w0[1];
ctx.w0[2] = tmps[gid].w0[2];
ctx.w0[3] = tmps[gid].w0[3];
ctx.w1[0] = tmps[gid].w1[0];
ctx.w1[1] = tmps[gid].w1[1];
ctx.w1[2] = tmps[gid].w1[2];
ctx.w1[3] = tmps[gid].w1[3];
ctx.w2[0] = tmps[gid].w2[0];
ctx.w2[1] = tmps[gid].w2[1];
ctx.w2[2] = tmps[gid].w2[2];
ctx.w2[3] = tmps[gid].w2[3];
ctx.w3[0] = tmps[gid].w3[0];
ctx.w3[1] = tmps[gid].w3[1];
ctx.w3[2] = tmps[gid].w3[2];
ctx.w3[3] = tmps[gid].w3[3];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
ctx.len = tmps[gid].len;
w0[0] = 0x80000000;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = tmps[gid].len * 8;
sha256_final (&ctx);
sha256_transform (w0, w1, w2, w3, h);
hooks[gid].ukey[0] = hc_swap32_S (ctx.h[0]);
hooks[gid].ukey[1] = hc_swap32_S (ctx.h[1]);
hooks[gid].ukey[2] = hc_swap32_S (ctx.h[2]);
hooks[gid].ukey[3] = hc_swap32_S (ctx.h[3]);
hooks[gid].ukey[4] = hc_swap32_S (ctx.h[4]);
hooks[gid].ukey[5] = hc_swap32_S (ctx.h[5]);
hooks[gid].ukey[6] = hc_swap32_S (ctx.h[6]);
hooks[gid].ukey[7] = hc_swap32_S (ctx.h[7]);
hooks[gid].ukey[0] = hc_swap32_S (h[0]);
hooks[gid].ukey[1] = hc_swap32_S (h[1]);
hooks[gid].ukey[2] = hc_swap32_S (h[2]);
hooks[gid].ukey[3] = hc_swap32_S (h[3]);
hooks[gid].ukey[4] = hc_swap32_S (h[4]);
hooks[gid].ukey[5] = hc_swap32_S (h[5]);
hooks[gid].ukey[6] = hc_swap32_S (h[6]);
hooks[gid].ukey[7] = hc_swap32_S (h[7]);
}
KERNEL_FQ void m11600_comp (KERN_ATTR_TMPS_HOOKS (seven_zip_tmp_t, seven_zip_hook_t))

34
OpenCL/m12400-pure.cl
View File

@ -726,40 +726,6 @@ KERNEL_FQ void m12400_loop (KERN_ATTR_TMPS (bsdicrypt_tmp_t))
_des_crypt_encrypt (iv, mask, loop_cnt, Kc, Kd, s_SPtrans);
tmps[gid].Kc[ 0] = Kc[ 0];
tmps[gid].Kc[ 1] = Kc[ 1];
tmps[gid].Kc[ 2] = Kc[ 2];
tmps[gid].Kc[ 3] = Kc[ 3];
tmps[gid].Kc[ 4] = Kc[ 4];
tmps[gid].Kc[ 5] = Kc[ 5];
tmps[gid].Kc[ 6] = Kc[ 6];
tmps[gid].Kc[ 7] = Kc[ 7];
tmps[gid].Kc[ 8] = Kc[ 8];
tmps[gid].Kc[ 9] = Kc[ 9];
tmps[gid].Kc[10] = Kc[10];
tmps[gid].Kc[11] = Kc[11];
tmps[gid].Kc[12] = Kc[12];
tmps[gid].Kc[13] = Kc[13];
tmps[gid].Kc[14] = Kc[14];
tmps[gid].Kc[15] = Kc[15];
tmps[gid].Kd[ 0] = Kd[ 0];
tmps[gid].Kd[ 1] = Kd[ 1];
tmps[gid].Kd[ 2] = Kd[ 2];
tmps[gid].Kd[ 3] = Kd[ 3];
tmps[gid].Kd[ 4] = Kd[ 4];
tmps[gid].Kd[ 5] = Kd[ 5];
tmps[gid].Kd[ 6] = Kd[ 6];
tmps[gid].Kd[ 7] = Kd[ 7];
tmps[gid].Kd[ 8] = Kd[ 8];
tmps[gid].Kd[ 9] = Kd[ 9];
tmps[gid].Kd[10] = Kd[10];
tmps[gid].Kd[11] = Kd[11];
tmps[gid].Kd[12] = Kd[12];
tmps[gid].Kd[13] = Kd[13];
tmps[gid].Kd[14] = Kd[14];
tmps[gid].Kd[15] = Kd[15];
tmps[gid].iv[0] = iv[0];
tmps[gid].iv[1] = iv[1];
}

397
OpenCL/m12500-optimized.cl Normal file
View File

@ -0,0 +1,397 @@
/**
* Author......: See docs/credits.txt
* License.....: MIT
*/
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
#include "inc_types.h"
#include "inc_platform.cl"
#include "inc_common.cl"
#include "inc_hash_sha1.cl"
#include "inc_cipher_aes.cl"
#endif
#define COMPARE_S "inc_comp_single.cl"
#define COMPARE_M "inc_comp_multi.cl"
#define ROUNDS 0x40000
#define PUTCHAR(a,p,c) ((u8 *)(a))[(p)] = (u8) (c)
#define GETCHAR(a,p) ((u8 *)(a))[(p)]
#define PUTCHAR_BE(a,p,c) ((u8 *)(a))[(p) ^ 3] = (u8) (c)
#define GETCHAR_BE(a,p) ((u8 *)(a))[(p) ^ 3]
#define MIN(a,b) (((a) < (b)) ? (a) : (b))
typedef struct pbkdf2_sha1
{
u32 salt_buf[64];
} pbkdf2_sha1_t;
typedef struct rar3_tmp
{
u32 dgst[17][5];
} rar3_tmp_t;
KERNEL_FQ void m12500_init (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
{
/**
* base
*/
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
tmps[gid].dgst[0][0] = SHA1M_A;
tmps[gid].dgst[0][1] = SHA1M_B;
tmps[gid].dgst[0][2] = SHA1M_C;
tmps[gid].dgst[0][3] = SHA1M_D;
tmps[gid].dgst[0][4] = SHA1M_E;
}
KERNEL_FQ void m12500_loop (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
{
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
u32 pw_buf[5];
pw_buf[0] = pws[gid].i[0];
pw_buf[1] = pws[gid].i[1];
pw_buf[2] = pws[gid].i[2];
pw_buf[3] = pws[gid].i[3];
pw_buf[4] = pws[gid].i[4];
const u32 pw_len = MIN (pws[gid].pw_len, 20);
u32 salt_buf[2];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
const u32 salt_len = 8;
// this is large enough to hold all possible w[] arrays for 64 iterations
#define LARGEBLOCK_ELEMS ((40 + 8 + 3) * 16)
u32 largeblock[LARGEBLOCK_ELEMS];
for (u32 i = 0; i < LARGEBLOCK_ELEMS; i++) largeblock[i] = 0;
for (u32 i = 0, p = 0; i < 64; i++)
{
for (u32 j = 0; j < pw_len; j++, p += 2)
{
PUTCHAR_BE (largeblock, p, GETCHAR (pw_buf, j));
}
for (u32 j = 0; j < salt_len; j++, p += 1)
{
PUTCHAR_BE (largeblock, p, GETCHAR (salt_buf, j));
}
PUTCHAR_BE (largeblock, p + 2, (loop_pos >> 16) & 0xff);
p += 3;
}
const u32 p3 = (pw_len * 2) + salt_len + 3;
const u32 init_pos = loop_pos / (ROUNDS / 16);
u32 dgst[5];
dgst[0] = tmps[gid].dgst[init_pos][0];
dgst[1] = tmps[gid].dgst[init_pos][1];
dgst[2] = tmps[gid].dgst[init_pos][2];
dgst[3] = tmps[gid].dgst[init_pos][3];
dgst[4] = tmps[gid].dgst[init_pos][4];
u32 iter = loop_pos;
for (u32 i = 0; i < 256; i += 4)
{
for (u32 j = 0; j < 64; j++)
{
const u32 p = ((j + 1) * p3) - 2;
PUTCHAR_BE (largeblock, p, iter >> 8);
}
for (u32 k = 0; k < 4; k++)
{
for (u32 j = 0; j < 64; j++)
{
const u32 p = ((j + 1) * p3) - 3;
PUTCHAR_BE (largeblock, p, iter >> 0);
iter++;
}
for (u32 j = 0; j < p3; j++)
{
const u32 j16 = j * 16;
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
w0[0] = largeblock[j16 + 0];
w0[1] = largeblock[j16 + 1];
w0[2] = largeblock[j16 + 2];
w0[3] = largeblock[j16 + 3];
w1[0] = largeblock[j16 + 4];
w1[1] = largeblock[j16 + 5];
w1[2] = largeblock[j16 + 6];
w1[3] = largeblock[j16 + 7];
w2[0] = largeblock[j16 + 8];
w2[1] = largeblock[j16 + 9];
w2[2] = largeblock[j16 + 10];
w2[3] = largeblock[j16 + 11];
w3[0] = largeblock[j16 + 12];
w3[1] = largeblock[j16 + 13];
w3[2] = largeblock[j16 + 14];
w3[3] = largeblock[j16 + 15];
sha1_transform (w0, w1, w2, w3, dgst);
}
}
}
tmps[gid].dgst[init_pos + 1][0] = dgst[0];
tmps[gid].dgst[init_pos + 1][1] = dgst[1];
tmps[gid].dgst[init_pos + 1][2] = dgst[2];
tmps[gid].dgst[init_pos + 1][3] = dgst[3];
tmps[gid].dgst[init_pos + 1][4] = dgst[4];
}
KERNEL_FQ void m12500_comp (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
/**
* base
*/
const u32 pw_len = MIN (pws[gid].pw_len, 20);
const u32 salt_len = 8;
const u32 p3 = (pw_len * 2) + salt_len + 3;
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
w0[0] = 0x80000000;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (p3 * ROUNDS) * 8;
u32 dgst[5];
dgst[0] = tmps[gid].dgst[16][0];
dgst[1] = tmps[gid].dgst[16][1];
dgst[2] = tmps[gid].dgst[16][2];
dgst[3] = tmps[gid].dgst[16][3];
dgst[4] = tmps[gid].dgst[16][4];
sha1_transform (w0, w1, w2, w3, dgst);
u32 ukey[4];
ukey[0] = hc_swap32_S (dgst[0]);
ukey[1] = hc_swap32_S (dgst[1]);
ukey[2] = hc_swap32_S (dgst[2]);
ukey[3] = hc_swap32_S (dgst[3]);
u32 ks[44];
AES128_set_decrypt_key (ks, ukey, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3);
u32 data[4];
data[0] = salt_bufs[salt_pos].salt_buf[2];
data[1] = salt_bufs[salt_pos].salt_buf[3];
data[2] = salt_bufs[salt_pos].salt_buf[4];
data[3] = salt_bufs[salt_pos].salt_buf[5];
u32 out[4];
AES128_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4);
u32 iv[4];
iv[0] = 0;
iv[1] = 0;
iv[2] = 0;
iv[3] = 0;
for (int i = 0; i < 16; i++)
{
u32 pw_buf[5];
pw_buf[0] = pws[gid].i[0];
pw_buf[1] = pws[gid].i[1];
pw_buf[2] = pws[gid].i[2];
pw_buf[3] = pws[gid].i[3];
pw_buf[4] = pws[gid].i[4];
//const u32 pw_len = pws[gid].pw_len;
u32 salt_buf[2];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
//const u32 salt_len = 8;
//const u32 p3 = (pw_len * 2) + salt_len + 3;
u32 w[16];
w[ 0] = 0;
w[ 1] = 0;
w[ 2] = 0;
w[ 3] = 0;
w[ 4] = 0;
w[ 5] = 0;
w[ 6] = 0;
w[ 7] = 0;
w[ 8] = 0;
w[ 9] = 0;
w[10] = 0;
w[11] = 0;
w[12] = 0;
w[13] = 0;
w[14] = 0;
w[15] = 0;
u32 p = 0;
for (u32 j = 0; j < pw_len; j++, p += 2)
{
PUTCHAR_BE (w, p, GETCHAR (pw_buf, j));
}
for (u32 j = 0; j < salt_len; j++, p += 1)
{
PUTCHAR_BE (w, p, GETCHAR (salt_buf, j));
}
const u32 iter_pos = i * (ROUNDS / 16);
PUTCHAR_BE (w, p + 0, (iter_pos >> 0) & 0xff);
PUTCHAR_BE (w, p + 1, (iter_pos >> 8) & 0xff);
PUTCHAR_BE (w, p + 2, (iter_pos >> 16) & 0xff);
PUTCHAR_BE (w, p3, 0x80);
w[15] = ((iter_pos + 1) * p3) * 8;
u32 dgst[5];
dgst[0] = tmps[gid].dgst[i][0];
dgst[1] = tmps[gid].dgst[i][1];
dgst[2] = tmps[gid].dgst[i][2];
dgst[3] = tmps[gid].dgst[i][3];
dgst[4] = tmps[gid].dgst[i][4];
sha1_transform (w + 0, w + 4, w + 8, w + 12, dgst);
PUTCHAR (iv, i, dgst[4] & 0xff);
}
out[0] ^= hc_swap32_S (iv[0]);
out[1] ^= hc_swap32_S (iv[1]);
out[2] ^= hc_swap32_S (iv[2]);
out[3] ^= hc_swap32_S (iv[3]);
const u32 r0 = out[0];
const u32 r1 = out[1];
const u32 r2 = 0;
const u32 r3 = 0;
#define il_pos 0
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}

488
OpenCL/m12500-pure.cl
View File

@ -37,132 +37,105 @@ typedef struct rar3_tmp
} rar3_tmp_t;
DECLSPEC void sha1_transform_intern (const u32 *w, u32 *digest)
DECLSPEC void memcat8c_be (u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 len, const u32 append, u32 *digest)
{
u32 A = digest[0];
u32 B = digest[1];
u32 C = digest[2];
u32 D = digest[3];
u32 E = digest[4];
const u32 func_len = len & 63;
u32 w0_t = w[ 0];
u32 w1_t = w[ 1];
u32 w2_t = w[ 2];
u32 w3_t = w[ 3];
u32 w4_t = w[ 4];
u32 w5_t = w[ 5];
u32 w6_t = w[ 6];
u32 w7_t = w[ 7];
u32 w8_t = w[ 8];
u32 w9_t = w[ 9];
u32 wa_t = w[10];
u32 wb_t = w[11];
u32 wc_t = w[12];
u32 wd_t = w[13];
u32 we_t = w[14];
u32 wf_t = w[15];
//const u32 mod = func_len & 3;
const u32 div = func_len / 4;
#undef K
#define K SHA1C00
u32 tmp0;
u32 tmp1;
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 = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
#if defined IS_AMD || defined IS_GENERIC
tmp0 = hc_bytealign_be (0, append, func_len);
tmp1 = hc_bytealign_be (append, 0, func_len);
#endif
#undef K
#define K SHA1C01
#ifdef IS_NV
const int selector = (0x76543210 >> ((func_len & 3) * 4)) & 0xffff;
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
tmp0 = hc_byte_perm (append, 0, selector);
tmp1 = hc_byte_perm (0, append, selector);
#endif
#undef K
#define K SHA1C02
u32 carry = 0;
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
switch (div)
{
case 0: w0[0] |= tmp0;
w0[1] = tmp1;
break;
case 1: w0[1] |= tmp0;
w0[2] = tmp1;
break;
case 2: w0[2] |= tmp0;
w0[3] = tmp1;
break;
case 3: w0[3] |= tmp0;
w1[0] = tmp1;
break;
case 4: w1[0] |= tmp0;
w1[1] = tmp1;
break;
case 5: w1[1] |= tmp0;
w1[2] = tmp1;
break;
case 6: w1[2] |= tmp0;
w1[3] = tmp1;
break;
case 7: w1[3] |= tmp0;
w2[0] = tmp1;
break;
case 8: w2[0] |= tmp0;
w2[1] = tmp1;
break;
case 9: w2[1] |= tmp0;
w2[2] = tmp1;
break;
case 10: w2[2] |= tmp0;
w2[3] = tmp1;
break;
case 11: w2[3] |= tmp0;
w3[0] = tmp1;
break;
case 12: w3[0] |= tmp0;
w3[1] = tmp1;
break;
case 13: w3[1] |= tmp0;
w3[2] = tmp1;
break;
case 14: w3[2] |= tmp0;
w3[3] = tmp1;
break;
case 15: w3[3] |= tmp0;
carry = tmp1;
break;
}
#undef K
#define K SHA1C03
const u32 new_len = func_len + 3;
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
w0_t = hc_rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
w1_t = hc_rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
w2_t = hc_rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
w3_t = hc_rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
w4_t = hc_rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
w5_t = hc_rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
w6_t = hc_rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
w7_t = hc_rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
w8_t = hc_rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
w9_t = hc_rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
wa_t = hc_rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
wb_t = hc_rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
wc_t = hc_rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
wd_t = hc_rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
we_t = hc_rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
wf_t = hc_rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
if (new_len >= 64)
{
sha1_transform (w0, w1, w2, w3, digest);
digest[0] += A;
digest[1] += B;
digest[2] += C;
digest[3] += D;
digest[4] += E;
w0[0] = carry;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
}
}
KERNEL_FQ void m12500_init (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
@ -180,6 +153,14 @@ KERNEL_FQ void m12500_init (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
tmps[gid].dgst[0][2] = SHA1M_C;
tmps[gid].dgst[0][3] = SHA1M_D;
tmps[gid].dgst[0][4] = SHA1M_E;
/**
* context save
*/
sha1_ctx_t ctx;
sha1_init (&ctx);
}
KERNEL_FQ void m12500_loop (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
@ -188,96 +169,68 @@ KERNEL_FQ void m12500_loop (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
if (gid >= gid_max) return;
u32 pw_buf[5];
/**
* base
*/
pw_buf[0] = pws[gid].i[0];
pw_buf[1] = pws[gid].i[1];
pw_buf[2] = pws[gid].i[2];
pw_buf[3] = pws[gid].i[3];
pw_buf[4] = pws[gid].i[4];
const u32 pw_len = pws[gid].pw_len;
const u32 pw_len = MIN (pws[gid].pw_len, 20);
u32 w[64] = { 0 };
u32 salt_buf[2];
for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1)
{
w[idx] = pws[gid].i[idx];
}
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
u32 salt_buf[16];
salt_buf[ 0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[ 1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[ 2] = 0;
salt_buf[ 3] = 0;
salt_buf[ 4] = 0;
salt_buf[ 5] = 0;
salt_buf[ 6] = 0;
salt_buf[ 7] = 0;
salt_buf[ 8] = 0;
salt_buf[ 9] = 0;
salt_buf[10] = 0;
salt_buf[11] = 0;
salt_buf[12] = 0;
salt_buf[13] = 0;
salt_buf[14] = 0;
salt_buf[15] = 0;
const u32 salt_len = 8;
// this is large enough to hold all possible w[] arrays for 64 iterations
#define LARGEBLOCK_ELEMS ((40 + 8 + 3) * 16)
u32 largeblock[LARGEBLOCK_ELEMS];
for (u32 i = 0; i < LARGEBLOCK_ELEMS; i++) largeblock[i] = 0;
for (u32 i = 0, p = 0; i < 64; i++)
{
for (u32 j = 0; j < pw_len; j++, p += 2)
{
PUTCHAR_BE (largeblock, p, GETCHAR (pw_buf, j));
}
for (u32 j = 0; j < salt_len; j++, p += 1)
{
PUTCHAR_BE (largeblock, p, GETCHAR (salt_buf, j));
}
PUTCHAR_BE (largeblock, p + 2, (loop_pos >> 16) & 0xff);
p += 3;
}
const u32 p3 = (pw_len * 2) + salt_len + 3;
const u32 init_pos = loop_pos / (ROUNDS / 16);
u32 dgst[5];
sha1_ctx_t ctx;
dgst[0] = tmps[gid].dgst[init_pos][0];
dgst[1] = tmps[gid].dgst[init_pos][1];
dgst[2] = tmps[gid].dgst[init_pos][2];
dgst[3] = tmps[gid].dgst[init_pos][3];
dgst[4] = tmps[gid].dgst[init_pos][4];
sha1_init (&ctx);
u32 iter = loop_pos;
ctx.h[0] = tmps[gid].dgst[init_pos][0];
ctx.h[1] = tmps[gid].dgst[init_pos][1];
ctx.h[2] = tmps[gid].dgst[init_pos][2];
ctx.h[3] = tmps[gid].dgst[init_pos][3];
ctx.h[4] = tmps[gid].dgst[init_pos][4];
for (u32 i = 0; i < 256; i += 4)
for (u32 i = 0, j = loop_pos; i < 16384; i++, j++)
{
for (u32 j = 0; j < 64; j++)
{
const u32 p = ((j + 1) * p3) - 2;
sha1_update_utf16le_swap (&ctx, w, pw_len);
PUTCHAR_BE (largeblock, p, iter >> 8);
}
sha1_update_swap (&ctx, salt_buf, salt_len);
for (u32 k = 0; k < 4; k++)
{
for (u32 j = 0; j < 64; j++)
{
const u32 p = ((j + 1) * p3) - 3;
memcat8c_be (ctx.w0, ctx.w1, ctx.w2, ctx.w3, ctx.len, hc_swap32_S (j), ctx.h);
PUTCHAR_BE (largeblock, p, iter >> 0);
iter++;
}
for (u32 j = 0; j < p3; j++)
{
const u32 j16 = j * 16;
sha1_transform_intern (&largeblock[j16], dgst);
}
}
ctx.len += 3;
}
tmps[gid].dgst[init_pos + 1][0] = dgst[0];
tmps[gid].dgst[init_pos + 1][1] = dgst[1];
tmps[gid].dgst[init_pos + 1][2] = dgst[2];
tmps[gid].dgst[init_pos + 1][3] = dgst[3];
tmps[gid].dgst[init_pos + 1][4] = dgst[4];
tmps[gid].dgst[init_pos + 1][0] = ctx.h[0];
tmps[gid].dgst[init_pos + 1][1] = ctx.h[1];
tmps[gid].dgst[init_pos + 1][2] = ctx.h[2];
tmps[gid].dgst[init_pos + 1][3] = ctx.h[3];
tmps[gid].dgst[init_pos + 1][4] = ctx.h[4];
}
KERNEL_FQ void m12500_comp (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
@ -343,47 +296,76 @@ KERNEL_FQ void m12500_comp (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
* base
*/
const u32 pw_len = MIN (pws[gid].pw_len, 20);
const u32 pw_len = pws[gid].pw_len;
u32 w[64] = { 0 };
for (u32 i = 0, idx = 0; i < pw_len; i += 4, idx += 1)
{
w[idx] = pws[gid].i[idx];
}
u32 salt_buf[16];
salt_buf[ 0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[ 1] = salt_bufs[salt_pos].salt_buf[1];
salt_buf[ 2] = 0;
salt_buf[ 3] = 0;
salt_buf[ 4] = 0;
salt_buf[ 5] = 0;
salt_buf[ 6] = 0;
salt_buf[ 7] = 0;
salt_buf[ 8] = 0;
salt_buf[ 9] = 0;
salt_buf[10] = 0;
salt_buf[11] = 0;
salt_buf[12] = 0;
salt_buf[13] = 0;
salt_buf[14] = 0;
salt_buf[15] = 0;
const u32 salt_len = 8;
const u32 p3 = (pw_len * 2) + salt_len + 3;
u32 w_buf[16];
u32 h[5];
w_buf[ 0] = 0x80000000;
w_buf[ 1] = 0;
w_buf[ 2] = 0;
w_buf[ 3] = 0;
w_buf[ 4] = 0;
w_buf[ 5] = 0;
w_buf[ 6] = 0;
w_buf[ 7] = 0;
w_buf[ 8] = 0;
w_buf[ 9] = 0;
w_buf[10] = 0;
w_buf[11] = 0;
w_buf[12] = 0;
w_buf[13] = 0;
w_buf[14] = 0;
w_buf[15] = (p3 * ROUNDS) * 8;
h[0] = tmps[gid].dgst[16][0];
h[1] = tmps[gid].dgst[16][1];
h[2] = tmps[gid].dgst[16][2];
h[3] = tmps[gid].dgst[16][3];
h[4] = tmps[gid].dgst[16][4];
u32 dgst[5];
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
dgst[0] = tmps[gid].dgst[16][0];
dgst[1] = tmps[gid].dgst[16][1];
dgst[2] = tmps[gid].dgst[16][2];
dgst[3] = tmps[gid].dgst[16][3];
dgst[4] = tmps[gid].dgst[16][4];
w0[0] = 0x80000000;
w0[1] = 0;
w0[2] = 0;
w0[3] = 0;
w1[0] = 0;
w1[1] = 0;
w1[2] = 0;
w1[3] = 0;
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = (ROUNDS * p3) * 8;
sha1_transform_intern (w_buf, dgst);
sha1_transform (w0, w1, w2, w3, h);
u32 ukey[4];
ukey[0] = hc_swap32_S (dgst[0]);
ukey[1] = hc_swap32_S (dgst[1]);
ukey[2] = hc_swap32_S (dgst[2]);
ukey[3] = hc_swap32_S (dgst[3]);
ukey[0] = hc_swap32_S (h[0]);
ukey[1] = hc_swap32_S (h[1]);
ukey[2] = hc_swap32_S (h[2]);
ukey[3] = hc_swap32_S (h[3]);
u32 ks[44];
@ -409,77 +391,31 @@ KERNEL_FQ void m12500_comp (KERN_ATTR_TMPS_ESALT (rar3_tmp_t, pbkdf2_sha1_t))
for (int i = 0; i < 16; i++)
{
u32 pw_buf[5];
sha1_ctx_t ctx;
pw_buf[0] = pws[gid].i[0];
pw_buf[1] = pws[gid].i[1];
pw_buf[2] = pws[gid].i[2];
pw_buf[3] = pws[gid].i[3];
pw_buf[4] = pws[gid].i[4];
sha1_init (&ctx);
//const u32 pw_len = pws[gid].pw_len;
u32 salt_buf[2];
salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
//const u32 salt_len = 8;
//const u32 p3 = (pw_len * 2) + salt_len + 3;
u32 w[16];
w[ 0] = 0;
w[ 1] = 0;
w[ 2] = 0;
w[ 3] = 0;
w[ 4] = 0;
w[ 5] = 0;
w[ 6] = 0;
w[ 7] = 0;
w[ 8] = 0;
w[ 9] = 0;
w[10] = 0;
w[11] = 0;
w[12] = 0;
w[13] = 0;
w[14] = 0;
w[15] = 0;
u32 p = 0;
for (u32 j = 0; j < pw_len; j++, p += 2)
{
PUTCHAR_BE (w, p, GETCHAR (pw_buf, j));
}
for (u32 j = 0; j < salt_len; j++, p += 1)
{
PUTCHAR_BE (w, p, GETCHAR (salt_buf, j));
}
ctx.h[0] = tmps[gid].dgst[i][0];
ctx.h[1] = tmps[gid].dgst[i][1];
ctx.h[2] = tmps[gid].dgst[i][2];
ctx.h[3] = tmps[gid].dgst[i][3];
ctx.h[4] = tmps[gid].dgst[i][4];
const u32 iter_pos = i * (ROUNDS / 16);
PUTCHAR_BE (w, p + 0, (iter_pos >> 0) & 0xff);
PUTCHAR_BE (w, p + 1, (iter_pos >> 8) & 0xff);
PUTCHAR_BE (w, p + 2, (iter_pos >> 16) & 0xff);
ctx.len = iter_pos * p3;
PUTCHAR_BE (w, p3, 0x80);
sha1_update_utf16le_swap (&ctx, w, pw_len);
w[15] = ((iter_pos + 1) * p3) * 8;
sha1_update_swap (&ctx, salt_buf, salt_len);
u32 dgst[5];
memcat8c_be (ctx.w0, ctx.w1, ctx.w2, ctx.w3, ctx.len, hc_swap32_S (iter_pos), ctx.h);
dgst[0] = tmps[gid].dgst[i][0];
dgst[1] = tmps[gid].dgst[i][1];
dgst[2] = tmps[gid].dgst[i][2];
dgst[3] = tmps[gid].dgst[i][3];
dgst[4] = tmps[gid].dgst[i][4];
ctx.len += 3;
sha1_transform_intern (w, dgst);
sha1_final (&ctx);
PUTCHAR (iv, i, dgst[4] & 0xff);
PUTCHAR (iv, i, ctx.h[4] & 0xff);
}
out[0] ^= hc_swap32_S (iv[0]);

157
OpenCL/m12700-pure.cl
View File

@ -28,6 +28,20 @@ typedef struct mywallet_tmp
} mywallet_tmp_t;
DECLSPEC int is_valid_char (const u32 v)
{
if ((v & 0xff000000) < 0x09000000) return 0;
if ((v & 0xff000000) > 0x7e000000) return 0;
if ((v & 0x00ff0000) < 0x00090000) return 0;
if ((v & 0x00ff0000) > 0x007e0000) return 0;
if ((v & 0x0000ff00) < 0x00000900) return 0;
if ((v & 0x0000ff00) > 0x00007e00) return 0;
if ((v & 0x000000ff) < 0x00000009) return 0;
if ((v & 0x000000ff) > 0x0000007e) return 0;
return 1;
}
DECLSPEC void hmac_sha1_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest)
{
digest[0] = ipad[0];
@ -318,121 +332,52 @@ KERNEL_FQ void m12700_comp (KERN_ATTR_TMPS (mywallet_tmp_t))
AES256_set_decrypt_key (ks, ukey, s_te0, s_te1, s_te2, s_te3, s_td0, s_td1, s_td2, s_td3);
u32 data[4];
u32 iv[4];
data[0] = salt_bufs[salt_pos].salt_buf[4];
data[1] = salt_bufs[salt_pos].salt_buf[5];
data[2] = salt_bufs[salt_pos].salt_buf[6];
data[3] = salt_bufs[salt_pos].salt_buf[7];
iv[0] = salt_bufs[salt_pos].salt_buf[0];
iv[1] = salt_bufs[salt_pos].salt_buf[1];
iv[2] = salt_bufs[salt_pos].salt_buf[2];
iv[3] = salt_bufs[salt_pos].salt_buf[3];
u32 out[4];
// decrypted data should be a JSON string consisting only of ASCII chars (0x09-0x7e)
AES256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4);
out[0] ^= salt_bufs[salt_pos].salt_buf[0];
out[1] ^= salt_bufs[salt_pos].salt_buf[1];
out[2] ^= salt_bufs[salt_pos].salt_buf[2];
out[3] ^= salt_bufs[salt_pos].salt_buf[3];
out[0] = hc_swap32_S (out[0]);
out[1] = hc_swap32_S (out[1]);
out[2] = hc_swap32_S (out[2]);
out[3] = hc_swap32_S (out[3]);
if ((out[0] & 0xff) != '{') return;
char *pt = (char *) out;
for (int i = 1; i < 16 - 6; i++)
for (u32 i = 4; i < 16; i += 4)
{
// "guid"
if ((pt[i + 0] == '"') && (pt[i + 1] == 'g') && (pt[i + 2] == 'u') && (pt[i + 3] == 'i') && (pt[i + 4] == 'd') && (pt[i + 5] == '"'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
u32 data[4];
#ifdef KERNEL_STATIC
#define il_pos 0
#endif
data[0] = salt_bufs[salt_pos].salt_buf[i + 0];
data[1] = salt_bufs[salt_pos].salt_buf[i + 1];
data[2] = salt_bufs[salt_pos].salt_buf[i + 2];
data[3] = salt_bufs[salt_pos].salt_buf[i + 3];
#include COMPARE_M
}
u32 out[4];
// "tx_no
if ((pt[i + 0] == '"') && (pt[i + 1] == 't') && (pt[i + 2] == 'x') && (pt[i + 3] == '_') && (pt[i + 4] == 'n') && (pt[i + 5] == 'o'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
AES256_decrypt (ks, data, out, s_td0, s_td1, s_td2, s_td3, s_td4);
#define il_pos 0
out[0] ^= iv[0];
out[1] ^= iv[1];
out[2] ^= iv[2];
out[3] ^= iv[3];
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}
if (is_valid_char (out[0]) == 0) return;
if (is_valid_char (out[1]) == 0) return;
if (is_valid_char (out[2]) == 0) return;
if (is_valid_char (out[3]) == 0) return;
// "share
if ((pt[i + 0] == '"') && (pt[i + 1] == 's') && (pt[i + 2] == 'h') && (pt[i + 3] == 'a') && (pt[i + 4] == 'r') && (pt[i + 5] == 'e'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
#ifdef KERNEL_STATIC
#define il_pos 0
#endif
#include COMPARE_M
}
// "doubl
if ((pt[i + 0] == '"') && (pt[i + 1] == 'd') && (pt[i + 2] == 'o') && (pt[i + 3] == 'u') && (pt[i + 4] == 'b') && (pt[i + 5] == 'l'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
#define il_pos 0
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}
// "addre
if ((pt[i + 0] == '"') && (pt[i + 1] == 'a') && (pt[i + 2] == 'd') && (pt[i + 3] == 'd') && (pt[i + 4] == 'r') && (pt[i + 5] == 'a'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
#define il_pos 0
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}
// "keys"
if ((pt[i + 0] == '"') && (pt[i + 1] == 'k') && (pt[i + 2] == 'e') && (pt[i + 3] == 'y') && (pt[i + 4] == 's') && (pt[i + 5] == '"'))
{
const u32 r0 = data[0];
const u32 r1 = data[1];
const u32 r2 = data[2];
const u32 r3 = data[3];
#define il_pos 0
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}
iv[0] = data[0];
iv[1] = data[1];
iv[2] = data[2];
iv[3] = data[3];
}
const u32 r0 = salt_bufs[salt_pos].salt_buf[4];
const u32 r1 = salt_bufs[salt_pos].salt_buf[5];
const u32 r2 = salt_bufs[salt_pos].salt_buf[6];
const u32 r3 = salt_bufs[salt_pos].salt_buf[7];
#define il_pos 0
#ifdef KERNEL_STATIC
#include COMPARE_M
#endif
}

94
OpenCL/m13600-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -43,7 +43,7 @@ typedef struct zip2
} zip2_t;
DECLSPEC void hmac_sha1_run (u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *ipad, u32 *opad, u32 *digest)
DECLSPEC void hmac_sha1_run_V (u32x *w0, u32x *w1, u32x *w2, u32x *w3, u32x *ipad, u32x *opad, u32x *digest)
{
digest[0] = ipad[0];
digest[1] = ipad[1];
@ -51,7 +51,7 @@ DECLSPEC void hmac_sha1_run (u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *ipad, u32
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];
@ -76,7 +76,7 @@ DECLSPEC void hmac_sha1_run (u32 *w0, u32 *w1, u32 *w2, u32 *w3, u32 *ipad, u32
digest[3] = opad[3];
digest[4] = opad[4];
sha1_transform (w0, w1, w2, w3, digest);
sha1_transform_vector (w0, w1, w2, w3, digest);
}
KERNEL_FQ void m13600_init (KERN_ATTR_TMPS_ESALT (pbkdf2_sha1_tmp_t, zip2_t))
@ -192,22 +192,22 @@ KERNEL_FQ void m13600_loop (KERN_ATTR_TMPS_ESALT (pbkdf2_sha1_tmp_t, zip2_t))
{
const u64 gid = get_global_id (0);
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
u32 ipad[5];
u32 opad[5];
u32x ipad[5];
u32x opad[5];
ipad[0] = tmps[gid].ipad[0];
ipad[1] = tmps[gid].ipad[1];
ipad[2] = tmps[gid].ipad[2];
ipad[3] = tmps[gid].ipad[3];
ipad[4] = tmps[gid].ipad[4];
ipad[0] = packv (tmps, ipad, gid, 0);
ipad[1] = packv (tmps, ipad, gid, 1);
ipad[2] = packv (tmps, ipad, gid, 2);
ipad[3] = packv (tmps, ipad, gid, 3);
ipad[4] = packv (tmps, ipad, gid, 4);
opad[0] = tmps[gid].opad[0];
opad[1] = tmps[gid].opad[1];
opad[2] = tmps[gid].opad[2];
opad[3] = tmps[gid].opad[3];
opad[4] = tmps[gid].opad[4];
opad[0] = packv (tmps, opad, gid, 0);
opad[1] = packv (tmps, opad, gid, 1);
opad[2] = packv (tmps, opad, gid, 2);
opad[3] = packv (tmps, opad, gid, 3);
opad[4] = packv (tmps, opad, gid, 4);
const u32 verify_bytes = esalt_bufs[digests_offset].verify_bytes;
@ -233,27 +233,27 @@ KERNEL_FQ void m13600_loop (KERN_ATTR_TMPS_ESALT (pbkdf2_sha1_tmp_t, zip2_t))
{
const u32 i5 = i * 5;
u32 dgst[5];
u32 out[5];
u32x dgst[5];
u32x out[5];
dgst[0] = tmps[gid].dgst[i5 + 0];
dgst[1] = tmps[gid].dgst[i5 + 1];
dgst[2] = tmps[gid].dgst[i5 + 2];
dgst[3] = tmps[gid].dgst[i5 + 3];
dgst[4] = tmps[gid].dgst[i5 + 4];
dgst[0] = packv (tmps, dgst, gid, i5 + 0);
dgst[1] = packv (tmps, dgst, gid, i5 + 1);
dgst[2] = packv (tmps, dgst, gid, i5 + 2);
dgst[3] = packv (tmps, dgst, gid, i5 + 3);
dgst[4] = packv (tmps, dgst, gid, i5 + 4);
out[0] = tmps[gid].out[i5 + 0];
out[1] = tmps[gid].out[i5 + 1];
out[2] = tmps[gid].out[i5 + 2];
out[3] = tmps[gid].out[i5 + 3];
out[4] = tmps[gid].out[i5 + 4];
out[0] = packv (tmps, out, gid, i5 + 0);
out[1] = packv (tmps, out, gid, i5 + 1);
out[2] = packv (tmps, out, gid, i5 + 2);
out[3] = packv (tmps, out, gid, i5 + 3);
out[4] = packv (tmps, out, gid, i5 + 4);
for (u32 j = 0; j < loop_cnt; j++)
{
u32 w0[4];
u32 w1[4];
u32 w2[4];
u32 w3[4];
u32x w0[4];
u32x w1[4];
u32x w2[4];
u32x w3[4];
w0[0] = dgst[0];
w0[1] = dgst[1];
@ -272,7 +272,7 @@ KERNEL_FQ void m13600_loop (KERN_ATTR_TMPS_ESALT (pbkdf2_sha1_tmp_t, zip2_t))
w3[2] = 0;
w3[3] = (64 + 20) * 8;
hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
hmac_sha1_run_V (w0, w1, w2, w3, ipad, opad, dgst);
out[0] ^= dgst[0];
out[1] ^= dgst[1];
@ -283,22 +283,22 @@ KERNEL_FQ void m13600_loop (KERN_ATTR_TMPS_ESALT (pbkdf2_sha1_tmp_t, zip2_t))
if (i == iter_stop - 1) // 2 byte optimization check
{
if (mode == 1) if ((out[3] >> 16) != verify_bytes) break;
if (mode == 2) if ((out[2] >> 16) != verify_bytes) break;
if (mode == 3) if ((out[1] >> 16) != verify_bytes) break;
if (mode == 1) if (MATCHES_NONE_VS ((out[3] >> 16), verify_bytes)) break;
if (mode == 2) if (MATCHES_NONE_VS ((out[2] >> 16), verify_bytes)) break;
if (mode == 3) if (MATCHES_NONE_VS ((out[1] >> 16), verify_bytes)) break;
}
tmps[gid].dgst[i5 + 0] = dgst[0];
tmps[gid].dgst[i5 + 1] = dgst[1];
tmps[gid].dgst[i5 + 2] = dgst[2];
tmps[gid].dgst[i5 + 3] = dgst[3];
tmps[gid].dgst[i5 + 4] = dgst[4];
unpackv (tmps, dgst, gid, i5 + 0, dgst[0]);
unpackv (tmps, dgst, gid, i5 + 1, dgst[1]);
unpackv (tmps, dgst, gid, i5 + 2, dgst[2]);
unpackv (tmps, dgst, gid, i5 + 3, dgst[3]);
unpackv (tmps, dgst, gid, i5 + 4, dgst[4]);
tmps[gid].out[i5 + 0] = out[0];
tmps[gid].out[i5 + 1] = out[1];
tmps[gid].out[i5 + 2] = out[2];
tmps[gid].out[i5 + 3] = out[3];
tmps[gid].out[i5 + 4] = out[4];
unpackv (tmps, out, gid, i5 + 0, out[0]);
unpackv (tmps, out, gid, i5 + 1, out[1]);
unpackv (tmps, out, gid, i5 + 2, out[2]);
unpackv (tmps, out, gid, i5 + 3, out[3]);
unpackv (tmps, out, gid, i5 + 4, out[4]);
}
}

142
OpenCL/m13711-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc_tmp
u32 pim_key[64];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc_tmp_t;
@ -252,61 +253,8 @@ KERNEL_FQ void m13711_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
KERNEL_FQ void m13711_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -407,11 +355,13 @@ KERNEL_FQ void m13711_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
unpackv (tmps, pim_key, gid, i + 0, out[0]);
unpackv (tmps, pim_key, gid, i + 1, out[1]);
unpackv (tmps, pim_key, gid, i + 2, out[2]);
unpackv (tmps, pim_key, gid, i + 3, out[3]);
unpackv (tmps, pim_key, gid, i + 4, out[4]);
tmps[gid].pim_check = pim;
}
}
@ -427,10 +377,78 @@ KERNEL_FQ void m13711_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
unpackv (tmps, out, gid, i + 3, out[3]);
unpackv (tmps, out, gid, i + 4, out[4]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13711_loop_extended (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13711_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))

148
OpenCL/m13712-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc_tmp
u32 pim_key[64];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc_tmp_t;
@ -303,61 +304,8 @@ KERNEL_FQ void m13712_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
KERNEL_FQ void m13712_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -458,11 +406,13 @@ KERNEL_FQ void m13712_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
unpackv (tmps, pim_key, gid, i + 0, out[0]);
unpackv (tmps, pim_key, gid, i + 1, out[1]);
unpackv (tmps, pim_key, gid, i + 2, out[2]);
unpackv (tmps, pim_key, gid, i + 3, out[3]);
unpackv (tmps, pim_key, gid, i + 4, out[4]);
tmps[gid].pim_check = pim;
}
}
@ -478,11 +428,83 @@ KERNEL_FQ void m13712_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
unpackv (tmps, out, gid, i + 3, out[3]);
unpackv (tmps, out, gid, i + 4, out[4]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13712_loop_extended (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13712_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))

154
OpenCL/m13713-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc_tmp
u32 pim_key[64];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc_tmp_t;
@ -368,61 +369,8 @@ KERNEL_FQ void m13713_init (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
KERNEL_FQ void m13713_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -523,11 +471,13 @@ KERNEL_FQ void m13713_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
unpackv (tmps, pim_key, gid, i + 0, out[0]);
unpackv (tmps, pim_key, gid, i + 1, out[1]);
unpackv (tmps, pim_key, gid, i + 2, out[2]);
unpackv (tmps, pim_key, gid, i + 3, out[3]);
unpackv (tmps, pim_key, gid, i + 4, out[4]);
tmps[gid].pim_check = pim;
}
}
@ -543,12 +493,88 @@ KERNEL_FQ void m13713_loop (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
unpackv (tmps, out, gid, i + 3, out[3]);
unpackv (tmps, out, gid, i + 4, out[4]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13713_loop_extended (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1536 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1536 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13713_comp (KERN_ATTR_TMPS_ESALT (vc_tmp_t, vc_t))

148
OpenCL/m13721-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc64_tmp
u64 pim_key[32];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc64_tmp_t;
@ -355,61 +356,8 @@ KERNEL_FQ void m13721_init (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
KERNEL_FQ void m13721_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* shared lookup table
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -545,14 +493,16 @@ KERNEL_FQ void m13721_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
tmps[gid].pim_key[i + 5] = out[5];
tmps[gid].pim_key[i + 6] = out[6];
tmps[gid].pim_key[i + 7] = out[7];
unpack64v (tmps, pim_key, gid, i + 0, out[0]);
unpack64v (tmps, pim_key, gid, i + 1, out[1]);
unpack64v (tmps, pim_key, gid, i + 2, out[2]);
unpack64v (tmps, pim_key, gid, i + 3, out[3]);
unpack64v (tmps, pim_key, gid, i + 4, out[4]);
unpack64v (tmps, pim_key, gid, i + 5, out[5]);
unpack64v (tmps, pim_key, gid, i + 6, out[6]);
unpack64v (tmps, pim_key, gid, i + 7, out[7]);
tmps[gid].pim_check = pim;
}
}
@ -574,10 +524,78 @@ KERNEL_FQ void m13721_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
unpack64v (tmps, out, gid, i + 6, out[6]);
unpack64v (tmps, out, gid, i + 7, out[7]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13721_loop_extended (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13721_comp (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))

154
OpenCL/m13722-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc64_tmp
u64 pim_key[32];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc64_tmp_t;
@ -406,61 +407,8 @@ KERNEL_FQ void m13722_init (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
KERNEL_FQ void m13722_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* shared lookup table
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -596,14 +544,16 @@ KERNEL_FQ void m13722_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
tmps[gid].pim_key[i + 5] = out[5];
tmps[gid].pim_key[i + 6] = out[6];
tmps[gid].pim_key[i + 7] = out[7];
unpack64v (tmps, pim_key, gid, i + 0, out[0]);
unpack64v (tmps, pim_key, gid, i + 1, out[1]);
unpack64v (tmps, pim_key, gid, i + 2, out[2]);
unpack64v (tmps, pim_key, gid, i + 3, out[3]);
unpack64v (tmps, pim_key, gid, i + 4, out[4]);
unpack64v (tmps, pim_key, gid, i + 5, out[5]);
unpack64v (tmps, pim_key, gid, i + 6, out[6]);
unpack64v (tmps, pim_key, gid, i + 7, out[7]);
tmps[gid].pim_check = pim;
}
}
@ -625,11 +575,83 @@ KERNEL_FQ void m13722_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
unpack64v (tmps, out, gid, i + 6, out[6]);
unpack64v (tmps, out, gid, i + 7, out[7]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13722_loop_extended (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13722_comp (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))

160
OpenCL/m13723-pure.cl
View File

@ -3,7 +3,7 @@
* License.....: MIT
*/
//#define NEW_SIMD_CODE
#define NEW_SIMD_CODE
#ifdef KERNEL_STATIC
#include "inc_vendor.h"
@ -52,6 +52,7 @@ typedef struct vc64_tmp
u64 pim_key[32];
int pim; // marker for cracked
int pim_check; // marker for _extended kernel
} vc64_tmp_t;
@ -471,61 +472,8 @@ KERNEL_FQ void m13723_init (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
KERNEL_FQ void m13723_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
/**
* shared lookup table
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
if ((gid * VECT_SIZE) >= gid_max) return;
// this is the pim range check
// it is guaranteed that only 0 or 1 innerloops will match a "pim" mark (each 1000 iterations)
@ -661,14 +609,16 @@ KERNEL_FQ void m13723_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
if (j == pim_at)
{
tmps[gid].pim_key[i + 0] = out[0];
tmps[gid].pim_key[i + 1] = out[1];
tmps[gid].pim_key[i + 2] = out[2];
tmps[gid].pim_key[i + 3] = out[3];
tmps[gid].pim_key[i + 4] = out[4];
tmps[gid].pim_key[i + 5] = out[5];
tmps[gid].pim_key[i + 6] = out[6];
tmps[gid].pim_key[i + 7] = out[7];
unpack64v (tmps, pim_key, gid, i + 0, out[0]);
unpack64v (tmps, pim_key, gid, i + 1, out[1]);
unpack64v (tmps, pim_key, gid, i + 2, out[2]);
unpack64v (tmps, pim_key, gid, i + 3, out[3]);
unpack64v (tmps, pim_key, gid, i + 4, out[4]);
unpack64v (tmps, pim_key, gid, i + 5, out[5]);
unpack64v (tmps, pim_key, gid, i + 6, out[6]);
unpack64v (tmps, pim_key, gid, i + 7, out[7]);
tmps[gid].pim_check = pim;
}
}
@ -690,12 +640,88 @@ KERNEL_FQ void m13723_loop (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
unpack64v (tmps, out, gid, i + 6, out[6]);
unpack64v (tmps, out, gid, i + 7, out[7]);
}
}
if (pim == 0) return;
KERNEL_FQ void m13723_loop_extended (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))
{
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
const u64 lsz = get_local_size (0);
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
if (check_header_1536 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1) tmps[gid].pim = pim;
/**
* aes shared
*/
#ifdef REAL_SHM
LOCAL_VK u32 s_td0[256];
LOCAL_VK u32 s_td1[256];
LOCAL_VK u32 s_td2[256];
LOCAL_VK u32 s_td3[256];
LOCAL_VK u32 s_td4[256];
LOCAL_VK u32 s_te0[256];
LOCAL_VK u32 s_te1[256];
LOCAL_VK u32 s_te2[256];
LOCAL_VK u32 s_te3[256];
LOCAL_VK u32 s_te4[256];
for (u32 i = lid; i < 256; i += lsz)
{
s_td0[i] = td0[i];
s_td1[i] = td1[i];
s_td2[i] = td2[i];
s_td3[i] = td3[i];
s_td4[i] = td4[i];
s_te0[i] = te0[i];
s_te1[i] = te1[i];
s_te2[i] = te2[i];
s_te3[i] = te3[i];
s_te4[i] = te4[i];
}
SYNC_THREADS ();
#else
CONSTANT_AS u32a *s_td0 = td0;
CONSTANT_AS u32a *s_td1 = td1;
CONSTANT_AS u32a *s_td2 = td2;
CONSTANT_AS u32a *s_td3 = td3;
CONSTANT_AS u32a *s_td4 = td4;
CONSTANT_AS u32a *s_te0 = te0;
CONSTANT_AS u32a *s_te1 = te1;
CONSTANT_AS u32a *s_te2 = te2;
CONSTANT_AS u32a *s_te3 = te3;
CONSTANT_AS u32a *s_te4 = te4;
#endif
if (gid >= gid_max) return;
const u32 pim_check = tmps[gid].pim_check;
if (pim_check)
{
if (check_header_0512 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1024 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
if (check_header_1536 (esalt_bufs, tmps[gid].pim_key, s_te0, s_te1, s_te2, s_te3, s_te4, s_td0, s_td1, s_td2, s_td3, s_td4) != -1)
{
tmps[gid].pim = pim_check;
}
tmps[gid].pim_check = 0;
}
}
KERNEL_FQ void m13723_comp (KERN_ATTR_TMPS_ESALT (vc64_tmp_t, vc_t))

Some files were not shown because too many files have changed in this diff Show More