popov/cpuminer-opt-gpu
1
0
Fork 0
mirror of https://github.com/JayDDee/cpuminer-opt.git synced 2025-09-17 23:44:27 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
badc80f071912faaa0d7fe2f09202e903de491df
cpuminer-opt-gpu/algo/skein/sse2/skein.c
Jay D Dee badc80f071 v3.5.0
2017-01-12 19:40:17 -05:00

483 lines
11 KiB
C
Raw Blame History

/* $Id: skein.c 254 2011-06-07 19:38:58Z tp $ */
/*
* Skein implementation.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#include <stddef.h>
#include <string.h>
#include "../sph_skein.h"
#ifdef __cplusplus
extern "C"{
#endif
#ifdef _MSC_VER
#pragma warning (disable: 4146)
#endif
/*
* M9_ ## s ## _ ## i evaluates to s+i mod 9 (0 <= s <= 18, 0 <= i <= 7).
*/
#define M9_0_0 0
#define M9_0_1 1
#define M9_0_2 2
#define M9_0_3 3
#define M9_0_4 4
#define M9_0_5 5
#define M9_0_6 6
#define M9_0_7 7
#define M9_1_0 1
#define M9_1_1 2
#define M9_1_2 3
#define M9_1_3 4
#define M9_1_4 5
#define M9_1_5 6
#define M9_1_6 7
#define M9_1_7 8
#define M9_2_0 2
#define M9_2_1 3
#define M9_2_2 4
#define M9_2_3 5
#define M9_2_4 6
#define M9_2_5 7
#define M9_2_6 8
#define M9_2_7 0
#define M9_3_0 3
#define M9_3_1 4
#define M9_3_2 5
#define M9_3_3 6
#define M9_3_4 7
#define M9_3_5 8
#define M9_3_6 0
#define M9_3_7 1
#define M9_4_0 4
#define M9_4_1 5
#define M9_4_2 6
#define M9_4_3 7
#define M9_4_4 8
#define M9_4_5 0
#define M9_4_6 1
#define M9_4_7 2
#define M9_5_0 5
#define M9_5_1 6
#define M9_5_2 7
#define M9_5_3 8
#define M9_5_4 0
#define M9_5_5 1
#define M9_5_6 2
#define M9_5_7 3
#define M9_6_0 6
#define M9_6_1 7
#define M9_6_2 8
#define M9_6_3 0
#define M9_6_4 1
#define M9_6_5 2
#define M9_6_6 3
#define M9_6_7 4
#define M9_7_0 7
#define M9_7_1 8
#define M9_7_2 0
#define M9_7_3 1
#define M9_7_4 2
#define M9_7_5 3
#define M9_7_6 4
#define M9_7_7 5
#define M9_8_0 8
#define M9_8_1 0
#define M9_8_2 1
#define M9_8_3 2
#define M9_8_4 3
#define M9_8_5 4
#define M9_8_6 5
#define M9_8_7 6
#define M9_9_0 0
#define M9_9_1 1
#define M9_9_2 2
#define M9_9_3 3
#define M9_9_4 4
#define M9_9_5 5
#define M9_9_6 6
#define M9_9_7 7
#define M9_10_0 1
#define M9_10_1 2
#define M9_10_2 3
#define M9_10_3 4
#define M9_10_4 5
#define M9_10_5 6
#define M9_10_6 7
#define M9_10_7 8
#define M9_11_0 2
#define M9_11_1 3
#define M9_11_2 4
#define M9_11_3 5
#define M9_11_4 6
#define M9_11_5 7
#define M9_11_6 8
#define M9_11_7 0
#define M9_12_0 3
#define M9_12_1 4
#define M9_12_2 5
#define M9_12_3 6
#define M9_12_4 7
#define M9_12_5 8
#define M9_12_6 0
#define M9_12_7 1
#define M9_13_0 4
#define M9_13_1 5
#define M9_13_2 6
#define M9_13_3 7
#define M9_13_4 8
#define M9_13_5 0
#define M9_13_6 1
#define M9_13_7 2
#define M9_14_0 5
#define M9_14_1 6
#define M9_14_2 7
#define M9_14_3 8
#define M9_14_4 0
#define M9_14_5 1
#define M9_14_6 2
#define M9_14_7 3
#define M9_15_0 6
#define M9_15_1 7
#define M9_15_2 8
#define M9_15_3 0
#define M9_15_4 1
#define M9_15_5 2
#define M9_15_6 3
#define M9_15_7 4
#define M9_16_0 7
#define M9_16_1 8
#define M9_16_2 0
#define M9_16_3 1
#define M9_16_4 2
#define M9_16_5 3
#define M9_16_6 4
#define M9_16_7 5
#define M9_17_0 8
#define M9_17_1 0
#define M9_17_2 1
#define M9_17_3 2
#define M9_17_4 3
#define M9_17_5 4
#define M9_17_6 5
#define M9_17_7 6
#define M9_18_0 0
#define M9_18_1 1
#define M9_18_2 2
#define M9_18_3 3
#define M9_18_4 4
#define M9_18_5 5
#define M9_18_6 6
#define M9_18_7 7
/*
* M3_ ## s ## _ ## i evaluates to s+i mod 3 (0 <= s <= 18, 0 <= i <= 1).
*/
#define M3_0_0 0
#define M3_0_1 1
#define M3_1_0 1
#define M3_1_1 2
#define M3_2_0 2
#define M3_2_1 0
#define M3_3_0 0
#define M3_3_1 1
#define M3_4_0 1
#define M3_4_1 2
#define M3_5_0 2
#define M3_5_1 0
#define M3_6_0 0
#define M3_6_1 1
#define M3_7_0 1
#define M3_7_1 2
#define M3_8_0 2
#define M3_8_1 0
#define M3_9_0 0
#define M3_9_1 1
#define M3_10_0 1
#define M3_10_1 2
#define M3_11_0 2
#define M3_11_1 0
#define M3_12_0 0
#define M3_12_1 1
#define M3_13_0 1
#define M3_13_1 2
#define M3_14_0 2
#define M3_14_1 0
#define M3_15_0 0
#define M3_15_1 1
#define M3_16_0 1
#define M3_16_1 2
#define M3_17_0 2
#define M3_17_1 0
#define M3_18_0 0
#define M3_18_1 1
#define XCAT(x, y) XCAT_(x, y)
#define XCAT_(x, y) x ## y
#define SKBI(k, s, i) XCAT(k, XCAT(XCAT(XCAT(M9_, s), _), i))
#define SKBT(t, s, v) XCAT(t, XCAT(XCAT(XCAT(M3_, s), _), v))
#define TFBIG_KINIT(k0, k1, k2, k3, k4, k5, k6, k7, k8, t0, t1, t2) do { \
k8 = ((k0 ^ k1) ^ (k2 ^ k3)) ^ ((k4 ^ k5) ^ (k6 ^ k7)) \
^ SPH_C64(0x1BD11BDAA9FC1A22); \
t2 = t0 ^ t1; \
} while (0)
#define TFBIG_ADDKEY(w0, w1, w2, w3, w4, w5, w6, w7, k, t, s) do { \
w0 = SPH_T64(w0 + SKBI(k, s, 0)); \
w1 = SPH_T64(w1 + SKBI(k, s, 1)); \
w2 = SPH_T64(w2 + SKBI(k, s, 2)); \
w3 = SPH_T64(w3 + SKBI(k, s, 3)); \
w4 = SPH_T64(w4 + SKBI(k, s, 4)); \
w5 = SPH_T64(w5 + SKBI(k, s, 5) + SKBT(t, s, 0)); \
w6 = SPH_T64(w6 + SKBI(k, s, 6) + SKBT(t, s, 1)); \
w7 = SPH_T64(w7 + SKBI(k, s, 7) + (sph_u64)s); \
} while (0)
#define TFBIG_MIX(x0, x1, rc) do { \
x0 = SPH_T64(x0 + x1); \
x1 = SPH_ROTL64(x1, rc) ^ x0; \
} while (0)
#define TFBIG_MIX8(w0, w1, w2, w3, w4, w5, w6, w7, rc0, rc1, rc2, rc3) do { \
TFBIG_MIX(w0, w1, rc0); \
TFBIG_MIX(w2, w3, rc1); \
TFBIG_MIX(w4, w5, rc2); \
TFBIG_MIX(w6, w7, rc3); \
} while (0)
#define TFBIG_4e(s) do { \
TFBIG_ADDKEY(p0, p1, p2, p3, p4, p5, p6, p7, sknh, t, s); \
TFBIG_MIX8(p0, p1, p2, p3, p4, p5, p6, p7, 46, 36, 19, 37); \
TFBIG_MIX8(p2, p1, p4, p7, p6, p5, p0, p3, 33, 27, 14, 42); \
TFBIG_MIX8(p4, p1, p6, p3, p0, p5, p2, p7, 17, 49, 36, 39); \
TFBIG_MIX8(p6, p1, p0, p7, p2, p5, p4, p3, 44, 9, 54, 56); \
} while (0)
#define TFBIG_4o(s) do { \
TFBIG_ADDKEY(p0, p1, p2, p3, p4, p5, p6, p7, sknh, t, s); \
TFBIG_MIX8(p0, p1, p2, p3, p4, p5, p6, p7, 39, 30, 34, 24); \
TFBIG_MIX8(p2, p1, p4, p7, p6, p5, p0, p3, 13, 50, 10, 17); \
TFBIG_MIX8(p4, p1, p6, p3, p0, p5, p2, p7, 25, 29, 39, 43); \
TFBIG_MIX8(p6, p1, p0, p7, p2, p5, p4, p3, 8, 35, 56, 22); \
} while (0)
#define UBI_BIG(etype, extra) do { \
sph_u64 sknh8, t0, t1, t2; \
sph_u64 m0 = sph_dec64le_aligned(buf + 0); \
sph_u64 m1 = sph_dec64le_aligned(buf + 8); \
sph_u64 m2 = sph_dec64le_aligned(buf + 16); \
sph_u64 m3 = sph_dec64le_aligned(buf + 24); \
sph_u64 m4 = sph_dec64le_aligned(buf + 32); \
sph_u64 m5 = sph_dec64le_aligned(buf + 40); \
sph_u64 m6 = sph_dec64le_aligned(buf + 48); \
sph_u64 m7 = sph_dec64le_aligned(buf + 56); \
sph_u64 p0 = m0; \
sph_u64 p1 = m1; \
sph_u64 p2 = m2; \
sph_u64 p3 = m3; \
sph_u64 p4 = m4; \
sph_u64 p5 = m5; \
sph_u64 p6 = m6; \
sph_u64 p7 = m7; \
t0 = SPH_T64(hashctA << 6) + (sph_u64)(extra); \
t1 = (hashctA >> 58) + ((sph_u64)(etype) << 55); \
TFBIG_KINIT(sknh0, sknh1, sknh2, sknh3, sknh4, sknh5, sknh6, sknh7, sknh8, t0, t1, t2); \
TFBIG_4e(0); \
TFBIG_4o(1); \
TFBIG_4e(2); \
TFBIG_4o(3); \
TFBIG_4e(4); \
TFBIG_4o(5); \
TFBIG_4e(6); \
TFBIG_4o(7); \
TFBIG_4e(8); \
TFBIG_4o(9); \
TFBIG_4e(10); \
TFBIG_4o(11); \
TFBIG_4e(12); \
TFBIG_4o(13); \
TFBIG_4e(14); \
TFBIG_4o(15); \
TFBIG_4e(16); \
TFBIG_4o(17); \
TFBIG_ADDKEY(p0, p1, p2, p3, p4, p5, p6, p7, sknh, t, 18); \
sknh0 = m0 ^ p0; \
sknh1 = m1 ^ p1; \
sknh2 = m2 ^ p2; \
sknh3 = m3 ^ p3; \
sknh4 = m4 ^ p4; \
sknh5 = m5 ^ p5; \
sknh6 = m6 ^ p6; \
sknh7 = m7 ^ p7; \
} while (0)
#define sknDECL_STATE_BIG \
sph_u64 sknh0, sknh1, sknh2, sknh3, sknh4, sknh5, sknh6, sknh7; \
#define DECL_SKN \
sph_u64 sknh0, sknh1, sknh2, sknh3, sknh4, sknh5, sknh6, sknh7; \
#define sknREAD_STATE_BIG(sc) do { \
sknh0 = (sc)->sknh0; \
sknh1 = (sc)->sknh1; \
sknh2 = (sc)->sknh2; \
sknh3 = (sc)->sknh3; \
sknh4 = (sc)->sknh4; \
sknh5 = (sc)->sknh5; \
sknh6 = (sc)->sknh6; \
sknh7 = (sc)->sknh7; \
} while (0)
#define sknWRITE_STATE_BIG(sc) do { \
(sc)->sknh0 = sknh0; \
(sc)->sknh1 = sknh1; \
(sc)->sknh2 = sknh2; \
(sc)->sknh3 = sknh3; \
(sc)->sknh4 = sknh4; \
(sc)->sknh5 = sknh5; \
(sc)->sknh6 = sknh6; \
(sc)->sknh7 = sknh7; \
} while (0)
/* not used */
#define SKN_H \
do { \
sph_skein512_init(&ctx_skein); \
skein_big_core(&ctx_skein, hash,64); \
sph_skein512_close(&ctx_skein, hash); \
} while (0)
/* load initial constants */
#define SKN_I \
do { \
sknh0 = sknIV512[0]; \
sknh1 = sknIV512[1]; \
sknh2 = sknIV512[2]; \
sknh3 = sknIV512[3]; \
sknh4 = sknIV512[4]; \
sknh5 = sknIV512[5]; \
sknh6 = sknIV512[6]; \
sknh7 = sknIV512[7]; \
hashctA = 0; \
hashptr = 0; \
} while (0)
/* load hash for loop */
#define SKN_U \
do { \
unsigned char *buf; \
size_t ptr; \
size_t len = 64; \
const void *data = hash; \
buf = hashbuf; \
ptr = hashptr; \
memcpy(buf + ptr, data, len); \
ptr += len; \
hashptr = ptr; \
} while (0)
/* skein512 hash loaded */
/* hash = skein512(loaded) */
#define SKN_C \
do { \
unsigned char *buf; \
size_t ptr; \
unsigned et; \
\
buf = hashbuf; \
ptr = hashptr; \
\
memset(buf + ptr, 0, (sizeof(char)*64) - ptr); \
/* for break loop */ \
/* one copy of inline UBI_BIG */ \
et = 352 + ((hashctA == 0) << 7) + (0 != 0); \
for (;;) { \
UBI_BIG(et, ptr); \
/* et gets changed for 2nd run */ \
if (et == 510) break; \
memset(buf, 0, (sizeof(char)*64)); \
hashctA = 0; \
et = 510; \
ptr = 8; \
} \
\
sph_enc64le_aligned(buf + 0, sknh0); \
sph_enc64le_aligned(buf + 8, sknh1); \
sph_enc64le_aligned(buf + 16, sknh2); \
sph_enc64le_aligned(buf + 24, sknh3); \
sph_enc64le_aligned(buf + 32, sknh4); \
sph_enc64le_aligned(buf + 40, sknh5); \
sph_enc64le_aligned(buf + 48, sknh6); \
sph_enc64le_aligned(buf + 56, sknh7); \
memcpy(hash, buf, 64); \
\
} while (0)
static const sph_u64 sknIV512[] = {
SPH_C64(0x4903ADFF749C51CE), SPH_C64(0x0D95DE399746DF03),
SPH_C64(0x8FD1934127C79BCE), SPH_C64(0x9A255629FF352CB1),
SPH_C64(0x5DB62599DF6CA7B0), SPH_C64(0xEABE394CA9D5C3F4),
SPH_C64(0x991112C71A75B523), SPH_C64(0xAE18A40B660FCC33)
};
#ifdef __cplusplus
}
#endif
Reference in New Issue View Git Blame Copy Permalink