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

v3.6.1

Browse Source
This commit is contained in:
Jay D Dee
2017-03-31 11:59:04 -04:00
parent 2dcc4821b6
commit 9afa0d9820
63 changed files with 297 additions and 176 deletions
Download Patch File Download Diff File Expand all files Collapse all files

392
algo/sha/aes_helper.c Normal file
View File

@@ -0,0 +1,392 @@
/* $Id: aes_helper.c 220 2010-06-09 09:21:50Z tp $ */
/*
* AES tables. This file is not meant to be compiled by itself; it
* is included by some hash function implementations. It contains
* the precomputed tables and helper macros for evaluating an AES
* round, optionally with a final XOR with a subkey.
*
* By default, this file defines the tables and macros for little-endian
* processing (i.e. it is assumed that the input bytes have been read
* from memory and assembled with the little-endian convention). If
* the 'AES_BIG_ENDIAN' macro is defined (to a non-zero integer value)
* when this file is included, then the tables and macros for big-endian
* processing are defined instead. The big-endian tables and macros have
* names distinct from the little-endian tables and macros, hence it is
* possible to have both simultaneously, by including this file twice
* (with and without the AES_BIG_ENDIAN macro).
*
* ==========================(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 "sph_types.h"
#ifdef __cplusplus
extern "C"{
#endif
#if AES_BIG_ENDIAN
#define AESx(x) ( ((SPH_C32(x) >> 24) & SPH_C32(0x000000FF)) \
| ((SPH_C32(x) >> 8) & SPH_C32(0x0000FF00)) \
| ((SPH_C32(x) << 8) & SPH_C32(0x00FF0000)) \
| ((SPH_C32(x) << 24) & SPH_C32(0xFF000000)))
#define AES0 AES0_BE
#define AES1 AES1_BE
#define AES2 AES2_BE
#define AES3 AES3_BE
#define AES_ROUND_BE(X0, X1, X2, X3, K0, K1, K2, K3, Y0, Y1, Y2, Y3) do { \
(Y0) = AES0[((X0) >> 24) & 0xFF] \
^ AES1[((X1) >> 16) & 0xFF] \
^ AES2[((X2) >> 8) & 0xFF] \
^ AES3[(X3) & 0xFF] ^ (K0); \
(Y1) = AES0[((X1) >> 24) & 0xFF] \
^ AES1[((X2) >> 16) & 0xFF] \
^ AES2[((X3) >> 8) & 0xFF] \
^ AES3[(X0) & 0xFF] ^ (K1); \
(Y2) = AES0[((X2) >> 24) & 0xFF] \
^ AES1[((X3) >> 16) & 0xFF] \
^ AES2[((X0) >> 8) & 0xFF] \
^ AES3[(X1) & 0xFF] ^ (K2); \
(Y3) = AES0[((X3) >> 24) & 0xFF] \
^ AES1[((X0) >> 16) & 0xFF] \
^ AES2[((X1) >> 8) & 0xFF] \
^ AES3[(X2) & 0xFF] ^ (K3); \
} while (0)
#define AES_ROUND_NOKEY_BE(X0, X1, X2, X3, Y0, Y1, Y2, Y3) \
AES_ROUND_BE(X0, X1, X2, X3, 0, 0, 0, 0, Y0, Y1, Y2, Y3)
#else
#define AESx(x) SPH_C32(x)
#define AES0 AES0_LE
#define AES1 AES1_LE
#define AES2 AES2_LE
#define AES3 AES3_LE
#define AES_ROUND_LE(X0, X1, X2, X3, K0, K1, K2, K3, Y0, Y1, Y2, Y3) do { \
(Y0) = AES0[(X0) & 0xFF] \
^ AES1[((X1) >> 8) & 0xFF] \
^ AES2[((X2) >> 16) & 0xFF] \
^ AES3[((X3) >> 24) & 0xFF] ^ (K0); \
(Y1) = AES0[(X1) & 0xFF] \
^ AES1[((X2) >> 8) & 0xFF] \
^ AES2[((X3) >> 16) & 0xFF] \
^ AES3[((X0) >> 24) & 0xFF] ^ (K1); \
(Y2) = AES0[(X2) & 0xFF] \
^ AES1[((X3) >> 8) & 0xFF] \
^ AES2[((X0) >> 16) & 0xFF] \
^ AES3[((X1) >> 24) & 0xFF] ^ (K2); \
(Y3) = AES0[(X3) & 0xFF] \
^ AES1[((X0) >> 8) & 0xFF] \
^ AES2[((X1) >> 16) & 0xFF] \
^ AES3[((X2) >> 24) & 0xFF] ^ (K3); \
} while (0)
#define AES_ROUND_NOKEY_LE(X0, X1, X2, X3, Y0, Y1, Y2, Y3) \
AES_ROUND_LE(X0, X1, X2, X3, 0, 0, 0, 0, Y0, Y1, Y2, Y3)
#endif
/*
* The AES*[] tables allow us to perform a fast evaluation of an AES
* round; table AESi[] combines SubBytes for a byte at row i, and
* MixColumns for the column where that byte goes after ShiftRows.
*/
static const sph_u32 AES0[256] = {
AESx(0xA56363C6), AESx(0x847C7CF8), AESx(0x997777EE), AESx(0x8D7B7BF6),
AESx(0x0DF2F2FF), AESx(0xBD6B6BD6), AESx(0xB16F6FDE), AESx(0x54C5C591),
AESx(0x50303060), AESx(0x03010102), AESx(0xA96767CE), AESx(0x7D2B2B56),
AESx(0x19FEFEE7), AESx(0x62D7D7B5), AESx(0xE6ABAB4D), AESx(0x9A7676EC),
AESx(0x45CACA8F), AESx(0x9D82821F), AESx(0x40C9C989), AESx(0x877D7DFA),
AESx(0x15FAFAEF), AESx(0xEB5959B2), AESx(0xC947478E), AESx(0x0BF0F0FB),
AESx(0xECADAD41), AESx(0x67D4D4B3), AESx(0xFDA2A25F), AESx(0xEAAFAF45),
AESx(0xBF9C9C23), AESx(0xF7A4A453), AESx(0x967272E4), AESx(0x5BC0C09B),
AESx(0xC2B7B775), AESx(0x1CFDFDE1), AESx(0xAE93933D), AESx(0x6A26264C),
AESx(0x5A36366C), AESx(0x413F3F7E), AESx(0x02F7F7F5), AESx(0x4FCCCC83),
AESx(0x5C343468), AESx(0xF4A5A551), AESx(0x34E5E5D1), AESx(0x08F1F1F9),
AESx(0x937171E2), AESx(0x73D8D8AB), AESx(0x53313162), AESx(0x3F15152A),
AESx(0x0C040408), AESx(0x52C7C795), AESx(0x65232346), AESx(0x5EC3C39D),
AESx(0x28181830), AESx(0xA1969637), AESx(0x0F05050A), AESx(0xB59A9A2F),
AESx(0x0907070E), AESx(0x36121224), AESx(0x9B80801B), AESx(0x3DE2E2DF),
AESx(0x26EBEBCD), AESx(0x6927274E), AESx(0xCDB2B27F), AESx(0x9F7575EA),
AESx(0x1B090912), AESx(0x9E83831D), AESx(0x742C2C58), AESx(0x2E1A1A34),
AESx(0x2D1B1B36), AESx(0xB26E6EDC), AESx(0xEE5A5AB4), AESx(0xFBA0A05B),
AESx(0xF65252A4), AESx(0x4D3B3B76), AESx(0x61D6D6B7), AESx(0xCEB3B37D),
AESx(0x7B292952), AESx(0x3EE3E3DD), AESx(0x712F2F5E), AESx(0x97848413),
AESx(0xF55353A6), AESx(0x68D1D1B9), AESx(0x00000000), AESx(0x2CEDEDC1),
AESx(0x60202040), AESx(0x1FFCFCE3), AESx(0xC8B1B179), AESx(0xED5B5BB6),
AESx(0xBE6A6AD4), AESx(0x46CBCB8D), AESx(0xD9BEBE67), AESx(0x4B393972),
AESx(0xDE4A4A94), AESx(0xD44C4C98), AESx(0xE85858B0), AESx(0x4ACFCF85),
AESx(0x6BD0D0BB), AESx(0x2AEFEFC5), AESx(0xE5AAAA4F), AESx(0x16FBFBED),
AESx(0xC5434386), AESx(0xD74D4D9A), AESx(0x55333366), AESx(0x94858511),
AESx(0xCF45458A), AESx(0x10F9F9E9), AESx(0x06020204), AESx(0x817F7FFE),
AESx(0xF05050A0), AESx(0x443C3C78), AESx(0xBA9F9F25), AESx(0xE3A8A84B),
AESx(0xF35151A2), AESx(0xFEA3A35D), AESx(0xC0404080), AESx(0x8A8F8F05),
AESx(0xAD92923F), AESx(0xBC9D9D21), AESx(0x48383870), AESx(0x04F5F5F1),
AESx(0xDFBCBC63), AESx(0xC1B6B677), AESx(0x75DADAAF), AESx(0x63212142),
AESx(0x30101020), AESx(0x1AFFFFE5), AESx(0x0EF3F3FD), AESx(0x6DD2D2BF),
AESx(0x4CCDCD81), AESx(0x140C0C18), AESx(0x35131326), AESx(0x2FECECC3),
AESx(0xE15F5FBE), AESx(0xA2979735), AESx(0xCC444488), AESx(0x3917172E),
AESx(0x57C4C493), AESx(0xF2A7A755), AESx(0x827E7EFC), AESx(0x473D3D7A),
AESx(0xAC6464C8), AESx(0xE75D5DBA), AESx(0x2B191932), AESx(0x957373E6),
AESx(0xA06060C0), AESx(0x98818119), AESx(0xD14F4F9E), AESx(0x7FDCDCA3),
AESx(0x66222244), AESx(0x7E2A2A54), AESx(0xAB90903B), AESx(0x8388880B),
AESx(0xCA46468C), AESx(0x29EEEEC7), AESx(0xD3B8B86B), AESx(0x3C141428),
AESx(0x79DEDEA7), AESx(0xE25E5EBC), AESx(0x1D0B0B16), AESx(0x76DBDBAD),
AESx(0x3BE0E0DB), AESx(0x56323264), AESx(0x4E3A3A74), AESx(0x1E0A0A14),
AESx(0xDB494992), AESx(0x0A06060C), AESx(0x6C242448), AESx(0xE45C5CB8),
AESx(0x5DC2C29F), AESx(0x6ED3D3BD), AESx(0xEFACAC43), AESx(0xA66262C4),
AESx(0xA8919139), AESx(0xA4959531), AESx(0x37E4E4D3), AESx(0x8B7979F2),
AESx(0x32E7E7D5), AESx(0x43C8C88B), AESx(0x5937376E), AESx(0xB76D6DDA),
AESx(0x8C8D8D01), AESx(0x64D5D5B1), AESx(0xD24E4E9C), AESx(0xE0A9A949),
AESx(0xB46C6CD8), AESx(0xFA5656AC), AESx(0x07F4F4F3), AESx(0x25EAEACF),
AESx(0xAF6565CA), AESx(0x8E7A7AF4), AESx(0xE9AEAE47), AESx(0x18080810),
AESx(0xD5BABA6F), AESx(0x887878F0), AESx(0x6F25254A), AESx(0x722E2E5C),
AESx(0x241C1C38), AESx(0xF1A6A657), AESx(0xC7B4B473), AESx(0x51C6C697),
AESx(0x23E8E8CB), AESx(0x7CDDDDA1), AESx(0x9C7474E8), AESx(0x211F1F3E),
AESx(0xDD4B4B96), AESx(0xDCBDBD61), AESx(0x868B8B0D), AESx(0x858A8A0F),
AESx(0x907070E0), AESx(0x423E3E7C), AESx(0xC4B5B571), AESx(0xAA6666CC),
AESx(0xD8484890), AESx(0x05030306), AESx(0x01F6F6F7), AESx(0x120E0E1C),
AESx(0xA36161C2), AESx(0x5F35356A), AESx(0xF95757AE), AESx(0xD0B9B969),
AESx(0x91868617), AESx(0x58C1C199), AESx(0x271D1D3A), AESx(0xB99E9E27),
AESx(0x38E1E1D9), AESx(0x13F8F8EB), AESx(0xB398982B), AESx(0x33111122),
AESx(0xBB6969D2), AESx(0x70D9D9A9), AESx(0x898E8E07), AESx(0xA7949433),
AESx(0xB69B9B2D), AESx(0x221E1E3C), AESx(0x92878715), AESx(0x20E9E9C9),
AESx(0x49CECE87), AESx(0xFF5555AA), AESx(0x78282850), AESx(0x7ADFDFA5),
AESx(0x8F8C8C03), AESx(0xF8A1A159), AESx(0x80898909), AESx(0x170D0D1A),
AESx(0xDABFBF65), AESx(0x31E6E6D7), AESx(0xC6424284), AESx(0xB86868D0),
AESx(0xC3414182), AESx(0xB0999929), AESx(0x772D2D5A), AESx(0x110F0F1E),
AESx(0xCBB0B07B), AESx(0xFC5454A8), AESx(0xD6BBBB6D), AESx(0x3A16162C)
};
static const sph_u32 AES1[256] = {
AESx(0x6363C6A5), AESx(0x7C7CF884), AESx(0x7777EE99), AESx(0x7B7BF68D),
AESx(0xF2F2FF0D), AESx(0x6B6BD6BD), AESx(0x6F6FDEB1), AESx(0xC5C59154),
AESx(0x30306050), AESx(0x01010203), AESx(0x6767CEA9), AESx(0x2B2B567D),
AESx(0xFEFEE719), AESx(0xD7D7B562), AESx(0xABAB4DE6), AESx(0x7676EC9A),
AESx(0xCACA8F45), AESx(0x82821F9D), AESx(0xC9C98940), AESx(0x7D7DFA87),
AESx(0xFAFAEF15), AESx(0x5959B2EB), AESx(0x47478EC9), AESx(0xF0F0FB0B),
AESx(0xADAD41EC), AESx(0xD4D4B367), AESx(0xA2A25FFD), AESx(0xAFAF45EA),
AESx(0x9C9C23BF), AESx(0xA4A453F7), AESx(0x7272E496), AESx(0xC0C09B5B),
AESx(0xB7B775C2), AESx(0xFDFDE11C), AESx(0x93933DAE), AESx(0x26264C6A),
AESx(0x36366C5A), AESx(0x3F3F7E41), AESx(0xF7F7F502), AESx(0xCCCC834F),
AESx(0x3434685C), AESx(0xA5A551F4), AESx(0xE5E5D134), AESx(0xF1F1F908),
AESx(0x7171E293), AESx(0xD8D8AB73), AESx(0x31316253), AESx(0x15152A3F),
AESx(0x0404080C), AESx(0xC7C79552), AESx(0x23234665), AESx(0xC3C39D5E),
AESx(0x18183028), AESx(0x969637A1), AESx(0x05050A0F), AESx(0x9A9A2FB5),
AESx(0x07070E09), AESx(0x12122436), AESx(0x80801B9B), AESx(0xE2E2DF3D),
AESx(0xEBEBCD26), AESx(0x27274E69), AESx(0xB2B27FCD), AESx(0x7575EA9F),
AESx(0x0909121B), AESx(0x83831D9E), AESx(0x2C2C5874), AESx(0x1A1A342E),
AESx(0x1B1B362D), AESx(0x6E6EDCB2), AESx(0x5A5AB4EE), AESx(0xA0A05BFB),
AESx(0x5252A4F6), AESx(0x3B3B764D), AESx(0xD6D6B761), AESx(0xB3B37DCE),
AESx(0x2929527B), AESx(0xE3E3DD3E), AESx(0x2F2F5E71), AESx(0x84841397),
AESx(0x5353A6F5), AESx(0xD1D1B968), AESx(0x00000000), AESx(0xEDEDC12C),
AESx(0x20204060), AESx(0xFCFCE31F), AESx(0xB1B179C8), AESx(0x5B5BB6ED),
AESx(0x6A6AD4BE), AESx(0xCBCB8D46), AESx(0xBEBE67D9), AESx(0x3939724B),
AESx(0x4A4A94DE), AESx(0x4C4C98D4), AESx(0x5858B0E8), AESx(0xCFCF854A),
AESx(0xD0D0BB6B), AESx(0xEFEFC52A), AESx(0xAAAA4FE5), AESx(0xFBFBED16),
AESx(0x434386C5), AESx(0x4D4D9AD7), AESx(0x33336655), AESx(0x85851194),
AESx(0x45458ACF), AESx(0xF9F9E910), AESx(0x02020406), AESx(0x7F7FFE81),
AESx(0x5050A0F0), AESx(0x3C3C7844), AESx(0x9F9F25BA), AESx(0xA8A84BE3),
AESx(0x5151A2F3), AESx(0xA3A35DFE), AESx(0x404080C0), AESx(0x8F8F058A),
AESx(0x92923FAD), AESx(0x9D9D21BC), AESx(0x38387048), AESx(0xF5F5F104),
AESx(0xBCBC63DF), AESx(0xB6B677C1), AESx(0xDADAAF75), AESx(0x21214263),
AESx(0x10102030), AESx(0xFFFFE51A), AESx(0xF3F3FD0E), AESx(0xD2D2BF6D),
AESx(0xCDCD814C), AESx(0x0C0C1814), AESx(0x13132635), AESx(0xECECC32F),
AESx(0x5F5FBEE1), AESx(0x979735A2), AESx(0x444488CC), AESx(0x17172E39),
AESx(0xC4C49357), AESx(0xA7A755F2), AESx(0x7E7EFC82), AESx(0x3D3D7A47),
AESx(0x6464C8AC), AESx(0x5D5DBAE7), AESx(0x1919322B), AESx(0x7373E695),
AESx(0x6060C0A0), AESx(0x81811998), AESx(0x4F4F9ED1), AESx(0xDCDCA37F),
AESx(0x22224466), AESx(0x2A2A547E), AESx(0x90903BAB), AESx(0x88880B83),
AESx(0x46468CCA), AESx(0xEEEEC729), AESx(0xB8B86BD3), AESx(0x1414283C),
AESx(0xDEDEA779), AESx(0x5E5EBCE2), AESx(0x0B0B161D), AESx(0xDBDBAD76),
AESx(0xE0E0DB3B), AESx(0x32326456), AESx(0x3A3A744E), AESx(0x0A0A141E),
AESx(0x494992DB), AESx(0x06060C0A), AESx(0x2424486C), AESx(0x5C5CB8E4),
AESx(0xC2C29F5D), AESx(0xD3D3BD6E), AESx(0xACAC43EF), AESx(0x6262C4A6),
AESx(0x919139A8), AESx(0x959531A4), AESx(0xE4E4D337), AESx(0x7979F28B),
AESx(0xE7E7D532), AESx(0xC8C88B43), AESx(0x37376E59), AESx(0x6D6DDAB7),
AESx(0x8D8D018C), AESx(0xD5D5B164), AESx(0x4E4E9CD2), AESx(0xA9A949E0),
AESx(0x6C6CD8B4), AESx(0x5656ACFA), AESx(0xF4F4F307), AESx(0xEAEACF25),
AESx(0x6565CAAF), AESx(0x7A7AF48E), AESx(0xAEAE47E9), AESx(0x08081018),
AESx(0xBABA6FD5), AESx(0x7878F088), AESx(0x25254A6F), AESx(0x2E2E5C72),
AESx(0x1C1C3824), AESx(0xA6A657F1), AESx(0xB4B473C7), AESx(0xC6C69751),
AESx(0xE8E8CB23), AESx(0xDDDDA17C), AESx(0x7474E89C), AESx(0x1F1F3E21),
AESx(0x4B4B96DD), AESx(0xBDBD61DC), AESx(0x8B8B0D86), AESx(0x8A8A0F85),
AESx(0x7070E090), AESx(0x3E3E7C42), AESx(0xB5B571C4), AESx(0x6666CCAA),
AESx(0x484890D8), AESx(0x03030605), AESx(0xF6F6F701), AESx(0x0E0E1C12),
AESx(0x6161C2A3), AESx(0x35356A5F), AESx(0x5757AEF9), AESx(0xB9B969D0),
AESx(0x86861791), AESx(0xC1C19958), AESx(0x1D1D3A27), AESx(0x9E9E27B9),
AESx(0xE1E1D938), AESx(0xF8F8EB13), AESx(0x98982BB3), AESx(0x11112233),
AESx(0x6969D2BB), AESx(0xD9D9A970), AESx(0x8E8E0789), AESx(0x949433A7),
AESx(0x9B9B2DB6), AESx(0x1E1E3C22), AESx(0x87871592), AESx(0xE9E9C920),
AESx(0xCECE8749), AESx(0x5555AAFF), AESx(0x28285078), AESx(0xDFDFA57A),
AESx(0x8C8C038F), AESx(0xA1A159F8), AESx(0x89890980), AESx(0x0D0D1A17),
AESx(0xBFBF65DA), AESx(0xE6E6D731), AESx(0x424284C6), AESx(0x6868D0B8),
AESx(0x414182C3), AESx(0x999929B0), AESx(0x2D2D5A77), AESx(0x0F0F1E11),
AESx(0xB0B07BCB), AESx(0x5454A8FC), AESx(0xBBBB6DD6), AESx(0x16162C3A)
};
static const sph_u32 AES2[256] = {
AESx(0x63C6A563), AESx(0x7CF8847C), AESx(0x77EE9977), AESx(0x7BF68D7B),
AESx(0xF2FF0DF2), AESx(0x6BD6BD6B), AESx(0x6FDEB16F), AESx(0xC59154C5),
AESx(0x30605030), AESx(0x01020301), AESx(0x67CEA967), AESx(0x2B567D2B),
AESx(0xFEE719FE), AESx(0xD7B562D7), AESx(0xAB4DE6AB), AESx(0x76EC9A76),
AESx(0xCA8F45CA), AESx(0x821F9D82), AESx(0xC98940C9), AESx(0x7DFA877D),
AESx(0xFAEF15FA), AESx(0x59B2EB59), AESx(0x478EC947), AESx(0xF0FB0BF0),
AESx(0xAD41ECAD), AESx(0xD4B367D4), AESx(0xA25FFDA2), AESx(0xAF45EAAF),
AESx(0x9C23BF9C), AESx(0xA453F7A4), AESx(0x72E49672), AESx(0xC09B5BC0),
AESx(0xB775C2B7), AESx(0xFDE11CFD), AESx(0x933DAE93), AESx(0x264C6A26),
AESx(0x366C5A36), AESx(0x3F7E413F), AESx(0xF7F502F7), AESx(0xCC834FCC),
AESx(0x34685C34), AESx(0xA551F4A5), AESx(0xE5D134E5), AESx(0xF1F908F1),
AESx(0x71E29371), AESx(0xD8AB73D8), AESx(0x31625331), AESx(0x152A3F15),
AESx(0x04080C04), AESx(0xC79552C7), AESx(0x23466523), AESx(0xC39D5EC3),
AESx(0x18302818), AESx(0x9637A196), AESx(0x050A0F05), AESx(0x9A2FB59A),
AESx(0x070E0907), AESx(0x12243612), AESx(0x801B9B80), AESx(0xE2DF3DE2),
AESx(0xEBCD26EB), AESx(0x274E6927), AESx(0xB27FCDB2), AESx(0x75EA9F75),
AESx(0x09121B09), AESx(0x831D9E83), AESx(0x2C58742C), AESx(0x1A342E1A),
AESx(0x1B362D1B), AESx(0x6EDCB26E), AESx(0x5AB4EE5A), AESx(0xA05BFBA0),
AESx(0x52A4F652), AESx(0x3B764D3B), AESx(0xD6B761D6), AESx(0xB37DCEB3),
AESx(0x29527B29), AESx(0xE3DD3EE3), AESx(0x2F5E712F), AESx(0x84139784),
AESx(0x53A6F553), AESx(0xD1B968D1), AESx(0x00000000), AESx(0xEDC12CED),
AESx(0x20406020), AESx(0xFCE31FFC), AESx(0xB179C8B1), AESx(0x5BB6ED5B),
AESx(0x6AD4BE6A), AESx(0xCB8D46CB), AESx(0xBE67D9BE), AESx(0x39724B39),
AESx(0x4A94DE4A), AESx(0x4C98D44C), AESx(0x58B0E858), AESx(0xCF854ACF),
AESx(0xD0BB6BD0), AESx(0xEFC52AEF), AESx(0xAA4FE5AA), AESx(0xFBED16FB),
AESx(0x4386C543), AESx(0x4D9AD74D), AESx(0x33665533), AESx(0x85119485),
AESx(0x458ACF45), AESx(0xF9E910F9), AESx(0x02040602), AESx(0x7FFE817F),
AESx(0x50A0F050), AESx(0x3C78443C), AESx(0x9F25BA9F), AESx(0xA84BE3A8),
AESx(0x51A2F351), AESx(0xA35DFEA3), AESx(0x4080C040), AESx(0x8F058A8F),
AESx(0x923FAD92), AESx(0x9D21BC9D), AESx(0x38704838), AESx(0xF5F104F5),
AESx(0xBC63DFBC), AESx(0xB677C1B6), AESx(0xDAAF75DA), AESx(0x21426321),
AESx(0x10203010), AESx(0xFFE51AFF), AESx(0xF3FD0EF3), AESx(0xD2BF6DD2),
AESx(0xCD814CCD), AESx(0x0C18140C), AESx(0x13263513), AESx(0xECC32FEC),
AESx(0x5FBEE15F), AESx(0x9735A297), AESx(0x4488CC44), AESx(0x172E3917),
AESx(0xC49357C4), AESx(0xA755F2A7), AESx(0x7EFC827E), AESx(0x3D7A473D),
AESx(0x64C8AC64), AESx(0x5DBAE75D), AESx(0x19322B19), AESx(0x73E69573),
AESx(0x60C0A060), AESx(0x81199881), AESx(0x4F9ED14F), AESx(0xDCA37FDC),
AESx(0x22446622), AESx(0x2A547E2A), AESx(0x903BAB90), AESx(0x880B8388),
AESx(0x468CCA46), AESx(0xEEC729EE), AESx(0xB86BD3B8), AESx(0x14283C14),
AESx(0xDEA779DE), AESx(0x5EBCE25E), AESx(0x0B161D0B), AESx(0xDBAD76DB),
AESx(0xE0DB3BE0), AESx(0x32645632), AESx(0x3A744E3A), AESx(0x0A141E0A),
AESx(0x4992DB49), AESx(0x060C0A06), AESx(0x24486C24), AESx(0x5CB8E45C),
AESx(0xC29F5DC2), AESx(0xD3BD6ED3), AESx(0xAC43EFAC), AESx(0x62C4A662),
AESx(0x9139A891), AESx(0x9531A495), AESx(0xE4D337E4), AESx(0x79F28B79),
AESx(0xE7D532E7), AESx(0xC88B43C8), AESx(0x376E5937), AESx(0x6DDAB76D),
AESx(0x8D018C8D), AESx(0xD5B164D5), AESx(0x4E9CD24E), AESx(0xA949E0A9),
AESx(0x6CD8B46C), AESx(0x56ACFA56), AESx(0xF4F307F4), AESx(0xEACF25EA),
AESx(0x65CAAF65), AESx(0x7AF48E7A), AESx(0xAE47E9AE), AESx(0x08101808),
AESx(0xBA6FD5BA), AESx(0x78F08878), AESx(0x254A6F25), AESx(0x2E5C722E),
AESx(0x1C38241C), AESx(0xA657F1A6), AESx(0xB473C7B4), AESx(0xC69751C6),
AESx(0xE8CB23E8), AESx(0xDDA17CDD), AESx(0x74E89C74), AESx(0x1F3E211F),
AESx(0x4B96DD4B), AESx(0xBD61DCBD), AESx(0x8B0D868B), AESx(0x8A0F858A),
AESx(0x70E09070), AESx(0x3E7C423E), AESx(0xB571C4B5), AESx(0x66CCAA66),
AESx(0x4890D848), AESx(0x03060503), AESx(0xF6F701F6), AESx(0x0E1C120E),
AESx(0x61C2A361), AESx(0x356A5F35), AESx(0x57AEF957), AESx(0xB969D0B9),
AESx(0x86179186), AESx(0xC19958C1), AESx(0x1D3A271D), AESx(0x9E27B99E),
AESx(0xE1D938E1), AESx(0xF8EB13F8), AESx(0x982BB398), AESx(0x11223311),
AESx(0x69D2BB69), AESx(0xD9A970D9), AESx(0x8E07898E), AESx(0x9433A794),
AESx(0x9B2DB69B), AESx(0x1E3C221E), AESx(0x87159287), AESx(0xE9C920E9),
AESx(0xCE8749CE), AESx(0x55AAFF55), AESx(0x28507828), AESx(0xDFA57ADF),
AESx(0x8C038F8C), AESx(0xA159F8A1), AESx(0x89098089), AESx(0x0D1A170D),
AESx(0xBF65DABF), AESx(0xE6D731E6), AESx(0x4284C642), AESx(0x68D0B868),
AESx(0x4182C341), AESx(0x9929B099), AESx(0x2D5A772D), AESx(0x0F1E110F),
AESx(0xB07BCBB0), AESx(0x54A8FC54), AESx(0xBB6DD6BB), AESx(0x162C3A16)
};
static const sph_u32 AES3[256] = {
AESx(0xC6A56363), AESx(0xF8847C7C), AESx(0xEE997777), AESx(0xF68D7B7B),
AESx(0xFF0DF2F2), AESx(0xD6BD6B6B), AESx(0xDEB16F6F), AESx(0x9154C5C5),
AESx(0x60503030), AESx(0x02030101), AESx(0xCEA96767), AESx(0x567D2B2B),
AESx(0xE719FEFE), AESx(0xB562D7D7), AESx(0x4DE6ABAB), AESx(0xEC9A7676),
AESx(0x8F45CACA), AESx(0x1F9D8282), AESx(0x8940C9C9), AESx(0xFA877D7D),
AESx(0xEF15FAFA), AESx(0xB2EB5959), AESx(0x8EC94747), AESx(0xFB0BF0F0),
AESx(0x41ECADAD), AESx(0xB367D4D4), AESx(0x5FFDA2A2), AESx(0x45EAAFAF),
AESx(0x23BF9C9C), AESx(0x53F7A4A4), AESx(0xE4967272), AESx(0x9B5BC0C0),
AESx(0x75C2B7B7), AESx(0xE11CFDFD), AESx(0x3DAE9393), AESx(0x4C6A2626),
AESx(0x6C5A3636), AESx(0x7E413F3F), AESx(0xF502F7F7), AESx(0x834FCCCC),
AESx(0x685C3434), AESx(0x51F4A5A5), AESx(0xD134E5E5), AESx(0xF908F1F1),
AESx(0xE2937171), AESx(0xAB73D8D8), AESx(0x62533131), AESx(0x2A3F1515),
AESx(0x080C0404), AESx(0x9552C7C7), AESx(0x46652323), AESx(0x9D5EC3C3),
AESx(0x30281818), AESx(0x37A19696), AESx(0x0A0F0505), AESx(0x2FB59A9A),
AESx(0x0E090707), AESx(0x24361212), AESx(0x1B9B8080), AESx(0xDF3DE2E2),
AESx(0xCD26EBEB), AESx(0x4E692727), AESx(0x7FCDB2B2), AESx(0xEA9F7575),
AESx(0x121B0909), AESx(0x1D9E8383), AESx(0x58742C2C), AESx(0x342E1A1A),
AESx(0x362D1B1B), AESx(0xDCB26E6E), AESx(0xB4EE5A5A), AESx(0x5BFBA0A0),
AESx(0xA4F65252), AESx(0x764D3B3B), AESx(0xB761D6D6), AESx(0x7DCEB3B3),
AESx(0x527B2929), AESx(0xDD3EE3E3), AESx(0x5E712F2F), AESx(0x13978484),
AESx(0xA6F55353), AESx(0xB968D1D1), AESx(0x00000000), AESx(0xC12CEDED),
AESx(0x40602020), AESx(0xE31FFCFC), AESx(0x79C8B1B1), AESx(0xB6ED5B5B),
AESx(0xD4BE6A6A), AESx(0x8D46CBCB), AESx(0x67D9BEBE), AESx(0x724B3939),
AESx(0x94DE4A4A), AESx(0x98D44C4C), AESx(0xB0E85858), AESx(0x854ACFCF),
AESx(0xBB6BD0D0), AESx(0xC52AEFEF), AESx(0x4FE5AAAA), AESx(0xED16FBFB),
AESx(0x86C54343), AESx(0x9AD74D4D), AESx(0x66553333), AESx(0x11948585),
AESx(0x8ACF4545), AESx(0xE910F9F9), AESx(0x04060202), AESx(0xFE817F7F),
AESx(0xA0F05050), AESx(0x78443C3C), AESx(0x25BA9F9F), AESx(0x4BE3A8A8),
AESx(0xA2F35151), AESx(0x5DFEA3A3), AESx(0x80C04040), AESx(0x058A8F8F),
AESx(0x3FAD9292), AESx(0x21BC9D9D), AESx(0x70483838), AESx(0xF104F5F5),
AESx(0x63DFBCBC), AESx(0x77C1B6B6), AESx(0xAF75DADA), AESx(0x42632121),
AESx(0x20301010), AESx(0xE51AFFFF), AESx(0xFD0EF3F3), AESx(0xBF6DD2D2),
AESx(0x814CCDCD), AESx(0x18140C0C), AESx(0x26351313), AESx(0xC32FECEC),
AESx(0xBEE15F5F), AESx(0x35A29797), AESx(0x88CC4444), AESx(0x2E391717),
AESx(0x9357C4C4), AESx(0x55F2A7A7), AESx(0xFC827E7E), AESx(0x7A473D3D),
AESx(0xC8AC6464), AESx(0xBAE75D5D), AESx(0x322B1919), AESx(0xE6957373),
AESx(0xC0A06060), AESx(0x19988181), AESx(0x9ED14F4F), AESx(0xA37FDCDC),
AESx(0x44662222), AESx(0x547E2A2A), AESx(0x3BAB9090), AESx(0x0B838888),
AESx(0x8CCA4646), AESx(0xC729EEEE), AESx(0x6BD3B8B8), AESx(0x283C1414),
AESx(0xA779DEDE), AESx(0xBCE25E5E), AESx(0x161D0B0B), AESx(0xAD76DBDB),
AESx(0xDB3BE0E0), AESx(0x64563232), AESx(0x744E3A3A), AESx(0x141E0A0A),
AESx(0x92DB4949), AESx(0x0C0A0606), AESx(0x486C2424), AESx(0xB8E45C5C),
AESx(0x9F5DC2C2), AESx(0xBD6ED3D3), AESx(0x43EFACAC), AESx(0xC4A66262),
AESx(0x39A89191), AESx(0x31A49595), AESx(0xD337E4E4), AESx(0xF28B7979),
AESx(0xD532E7E7), AESx(0x8B43C8C8), AESx(0x6E593737), AESx(0xDAB76D6D),
AESx(0x018C8D8D), AESx(0xB164D5D5), AESx(0x9CD24E4E), AESx(0x49E0A9A9),
AESx(0xD8B46C6C), AESx(0xACFA5656), AESx(0xF307F4F4), AESx(0xCF25EAEA),
AESx(0xCAAF6565), AESx(0xF48E7A7A), AESx(0x47E9AEAE), AESx(0x10180808),
AESx(0x6FD5BABA), AESx(0xF0887878), AESx(0x4A6F2525), AESx(0x5C722E2E),
AESx(0x38241C1C), AESx(0x57F1A6A6), AESx(0x73C7B4B4), AESx(0x9751C6C6),
AESx(0xCB23E8E8), AESx(0xA17CDDDD), AESx(0xE89C7474), AESx(0x3E211F1F),
AESx(0x96DD4B4B), AESx(0x61DCBDBD), AESx(0x0D868B8B), AESx(0x0F858A8A),
AESx(0xE0907070), AESx(0x7C423E3E), AESx(0x71C4B5B5), AESx(0xCCAA6666),
AESx(0x90D84848), AESx(0x06050303), AESx(0xF701F6F6), AESx(0x1C120E0E),
AESx(0xC2A36161), AESx(0x6A5F3535), AESx(0xAEF95757), AESx(0x69D0B9B9),
AESx(0x17918686), AESx(0x9958C1C1), AESx(0x3A271D1D), AESx(0x27B99E9E),
AESx(0xD938E1E1), AESx(0xEB13F8F8), AESx(0x2BB39898), AESx(0x22331111),
AESx(0xD2BB6969), AESx(0xA970D9D9), AESx(0x07898E8E), AESx(0x33A79494),
AESx(0x2DB69B9B), AESx(0x3C221E1E), AESx(0x15928787), AESx(0xC920E9E9),
AESx(0x8749CECE), AESx(0xAAFF5555), AESx(0x50782828), AESx(0xA57ADFDF),
AESx(0x038F8C8C), AESx(0x59F8A1A1), AESx(0x09808989), AESx(0x1A170D0D),
AESx(0x65DABFBF), AESx(0xD731E6E6), AESx(0x84C64242), AESx(0xD0B86868),
AESx(0x82C34141), AESx(0x29B09999), AESx(0x5A772D2D), AESx(0x1E110F0F),
AESx(0x7BCBB0B0), AESx(0xA8FC5454), AESx(0x6DD6BBBB), AESx(0x2C3A1616)
};
#ifdef __cplusplus
}
#endif

346
algo/sha/md_helper.c Normal file
View File

@@ -0,0 +1,346 @@
/* $Id: md_helper.c 216 2010-06-08 09:46:57Z tp $ */
/*
* This file contains some functions which implement the external data
* handling and padding for Merkle-Damgard hash functions which follow
* the conventions set out by MD4 (little-endian) or SHA-1 (big-endian).
*
* API: this file is meant to be included, not compiled as a stand-alone
* file. Some macros must be defined:
* RFUN name for the round function
* HASH "short name" for the hash function
* BE32 defined for big-endian, 32-bit based (e.g. SHA-1)
* LE32 defined for little-endian, 32-bit based (e.g. MD5)
* BE64 defined for big-endian, 64-bit based (e.g. SHA-512)
* LE64 defined for little-endian, 64-bit based (no example yet)
* PW01 if defined, append 0x01 instead of 0x80 (for Tiger)
* BLEN if defined, length of a message block (in bytes)
* PLW1 if defined, length is defined on one 64-bit word only (for Tiger)
* PLW4 if defined, length is defined on four 64-bit words (for WHIRLPOOL)
* SVAL if defined, reference to the context state information
*
* BLEN is used when a message block is not 16 (32-bit or 64-bit) words:
* this is used for instance for Tiger, which works on 64-bit words but
* uses 512-bit message blocks (eight 64-bit words). PLW1 and PLW4 are
* ignored if 32-bit words are used; if 64-bit words are used and PLW1 is
* set, then only one word (64 bits) will be used to encode the input
* message length (in bits), otherwise two words will be used (as in
* SHA-384 and SHA-512). If 64-bit words are used and PLW4 is defined (but
* not PLW1), four 64-bit words will be used to encode the message length
* (in bits). Note that regardless of those settings, only 64-bit message
* lengths are supported (in bits): messages longer than 2 Exabytes will be
* improperly hashed (this is unlikely to happen soon: 2 Exabytes is about
* 2 millions Terabytes, which is huge).
*
* If CLOSE_ONLY is defined, then this file defines only the sph_XXX_close()
* function. This is used for Tiger2, which is identical to Tiger except
* when it comes to the padding (Tiger2 uses the standard 0x80 byte instead
* of the 0x01 from original Tiger).
*
* The RFUN function is invoked with two arguments, the first pointing to
* aligned data (as a "const void *"), the second being state information
* from the context structure. By default, this state information is the
* "val" field from the context, and this field is assumed to be an array
* of words ("sph_u32" or "sph_u64", depending on BE32/LE32/BE64/LE64).
* from the context structure. The "val" field can have any type, except
* for the output encoding which assumes that it is an array of "sph_u32"
* values. By defining NO_OUTPUT, this last step is deactivated; the
* includer code is then responsible for writing out the hash result. When
* NO_OUTPUT is defined, the third parameter to the "close()" function is
* ignored.
*
* ==========================(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>
*/
#ifdef _MSC_VER
#pragma warning (disable: 4146)
#endif
#undef SPH_XCAT
#define SPH_XCAT(a, b) SPH_XCAT_(a, b)
#undef SPH_XCAT_
#define SPH_XCAT_(a, b) a ## b
#undef SPH_BLEN
#undef SPH_WLEN
#if defined BE64 || defined LE64
#define SPH_BLEN 128U
#define SPH_WLEN 8U
#else
#define SPH_BLEN 64U
#define SPH_WLEN 4U
#endif
#ifdef BLEN
#undef SPH_BLEN
#define SPH_BLEN BLEN
#endif
#undef SPH_MAXPAD
#if defined PLW1
#define SPH_MAXPAD (SPH_BLEN - SPH_WLEN)
#elif defined PLW4
#define SPH_MAXPAD (SPH_BLEN - (SPH_WLEN << 2))
#else
#define SPH_MAXPAD (SPH_BLEN - (SPH_WLEN << 1))
#endif
#undef SPH_VAL
#undef SPH_NO_OUTPUT
#ifdef SVAL
#define SPH_VAL SVAL
#define SPH_NO_OUTPUT 1
#else
#define SPH_VAL sc->val
#endif
#ifndef CLOSE_ONLY
#ifdef SPH_UPTR
static void
SPH_XCAT(HASH, _short)(void *cc, const void *data, size_t len)
#else
void
SPH_XCAT(sph_, HASH)(void *cc, const void *data, size_t len)
#endif
{
SPH_XCAT(sph_, SPH_XCAT(HASH, _context)) *sc;
size_t current;
sc = cc;
#if SPH_64
current = (unsigned)sc->count & (SPH_BLEN - 1U);
#else
current = (unsigned)sc->count_low & (SPH_BLEN - 1U);
#endif
while (len > 0) {
size_t clen;
#if !SPH_64
sph_u32 clow, clow2;
#endif
clen = SPH_BLEN - current;
if (clen > len)
clen = len;
memcpy(sc->buf + current, data, clen);
data = (const unsigned char *)data + clen;
current += clen;
len -= clen;
if (current == SPH_BLEN) {
RFUN(sc->buf, SPH_VAL);
current = 0;
}
#if SPH_64
sc->count += clen;
#else
clow = sc->count_low;
clow2 = SPH_T32(clow + clen);
sc->count_low = clow2;
if (clow2 < clow)
sc->count_high ++;
#endif
}
}
#ifdef SPH_UPTR
void
SPH_XCAT(sph_, HASH)(void *cc, const void *data, size_t len)
{
SPH_XCAT(sph_, SPH_XCAT(HASH, _context)) *sc;
unsigned current;
size_t orig_len;
#if !SPH_64
sph_u32 clow, clow2;
#endif
if (len < (2 * SPH_BLEN)) {
SPH_XCAT(HASH, _short)(cc, data, len);
return;
}
sc = cc;
#if SPH_64
current = (unsigned)sc->count & (SPH_BLEN - 1U);
#else
current = (unsigned)sc->count_low & (SPH_BLEN - 1U);
#endif
if (current > 0) {
unsigned t;
t = SPH_BLEN - current;
SPH_XCAT(HASH, _short)(cc, data, t);
data = (const unsigned char *)data + t;
len -= t;
}
#if !SPH_UNALIGNED
if (((SPH_UPTR)data & (SPH_WLEN - 1U)) != 0) {
SPH_XCAT(HASH, _short)(cc, data, len);
return;
}
#endif
orig_len = len;
while (len >= SPH_BLEN) {
RFUN(data, SPH_VAL);
len -= SPH_BLEN;
data = (const unsigned char *)data + SPH_BLEN;
}
if (len > 0)
memcpy(sc->buf, data, len);
#if SPH_64
sc->count += (sph_u64)orig_len;
#else
clow = sc->count_low;
clow2 = SPH_T32(clow + orig_len);
sc->count_low = clow2;
if (clow2 < clow)
sc->count_high ++;
/*
* This code handles the improbable situation where "size_t" is
* greater than 32 bits, and yet we do not have a 64-bit type.
*/
orig_len >>= 12;
orig_len >>= 10;
orig_len >>= 10;
sc->count_high += orig_len;
#endif
}
#endif
#endif
/*
* Perform padding and produce result. The context is NOT reinitialized
* by this function.
*/
static void
SPH_XCAT(HASH, _addbits_and_close)(void *cc,
unsigned ub, unsigned n, void *dst, unsigned rnum)
{
SPH_XCAT(sph_, SPH_XCAT(HASH, _context)) *sc;
unsigned current, u;
#if !SPH_64
sph_u32 low, high;
#endif
sc = cc;
#if SPH_64
current = (unsigned)sc->count & (SPH_BLEN - 1U);
#else
current = (unsigned)sc->count_low & (SPH_BLEN - 1U);
#endif
#ifdef PW01
sc->buf[current ++] = (0x100 | (ub & 0xFF)) >> (8 - n);
#else
{
unsigned z;
z = 0x80 >> n;
sc->buf[current ++] = ((ub & -z) | z) & 0xFF;
}
#endif
if (current > SPH_MAXPAD) {
memset(sc->buf + current, 0, SPH_BLEN - current);
RFUN(sc->buf, SPH_VAL);
memset(sc->buf, 0, SPH_MAXPAD);
} else {
memset(sc->buf + current, 0, SPH_MAXPAD - current);
}
#if defined BE64
#if defined PLW1
sph_enc64be_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
#elif defined PLW4
memset(sc->buf + SPH_MAXPAD, 0, 2 * SPH_WLEN);
sph_enc64be_aligned(sc->buf + SPH_MAXPAD + 2 * SPH_WLEN,
sc->count >> 61);
sph_enc64be_aligned(sc->buf + SPH_MAXPAD + 3 * SPH_WLEN,
SPH_T64(sc->count << 3) + (sph_u64)n);
#else
sph_enc64be_aligned(sc->buf + SPH_MAXPAD, sc->count >> 61);
sph_enc64be_aligned(sc->buf + SPH_MAXPAD + SPH_WLEN,
SPH_T64(sc->count << 3) + (sph_u64)n);
#endif
#elif defined LE64
#if defined PLW1
sph_enc64le_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
#elif defined PLW1
sph_enc64le_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
sph_enc64le_aligned(sc->buf + SPH_MAXPAD + SPH_WLEN, sc->count >> 61);
memset(sc->buf + SPH_MAXPAD + 2 * SPH_WLEN, 0, 2 * SPH_WLEN);
#else
sph_enc64le_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
sph_enc64le_aligned(sc->buf + SPH_MAXPAD + SPH_WLEN, sc->count >> 61);
#endif
#else
#if SPH_64
#ifdef BE32
sph_enc64be_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
#else
sph_enc64le_aligned(sc->buf + SPH_MAXPAD,
SPH_T64(sc->count << 3) + (sph_u64)n);
#endif
#else
low = sc->count_low;
high = SPH_T32((sc->count_high << 3) | (low >> 29));
low = SPH_T32(low << 3) + (sph_u32)n;
#ifdef BE32
sph_enc32be(sc->buf + SPH_MAXPAD, high);
sph_enc32be(sc->buf + SPH_MAXPAD + SPH_WLEN, low);
#else
sph_enc32le(sc->buf + SPH_MAXPAD, low);
sph_enc32le(sc->buf + SPH_MAXPAD + SPH_WLEN, high);
#endif
#endif
#endif
RFUN(sc->buf, SPH_VAL);
#ifdef SPH_NO_OUTPUT
(void)dst;
(void)rnum;
(void)u;
#else
for (u = 0; u < rnum; u ++) {
#if defined BE64
sph_enc64be((unsigned char *)dst + 8 * u, sc->val[u]);
#elif defined LE64
sph_enc64le((unsigned char *)dst + 8 * u, sc->val[u]);
#elif defined BE32
sph_enc32be((unsigned char *)dst + 4 * u, sc->val[u]);
#else
sph_enc32le((unsigned char *)dst + 4 * u, sc->val[u]);
#endif
}
#endif
}
static void
SPH_XCAT(HASH, _close)(void *cc, void *dst, unsigned rnum)
{
SPH_XCAT(HASH, _addbits_and_close)(cc, 0, 0, dst, rnum);
}

646
algo/sha/sha2.c Normal file
View File

@@ -0,0 +1,646 @@
/*
* Copyright 2011 ArtForz
* Copyright 2011-2013 pooler
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the Free
* Software Foundation; either version 2 of the License, or (at your option)
* any later version. See COPYING for more details.
*/
#include "miner.h"
#include "algo-gate-api.h"
#include <string.h>
#include <inttypes.h>
#if defined(USE_ASM) && defined(__arm__) && defined(__APCS_32__)
#define EXTERN_SHA256
#endif
static const uint32_t sha256_h[8] = {
0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a,
0x510e527f, 0x9b05688c, 0x1f83d9ab, 0x5be0cd19
};
static const uint32_t sha256_k[64] = {
0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
};
void sha256_init(uint32_t *state)
{
memcpy(state, sha256_h, 32);
}
/* Elementary functions used by SHA256 */
#define Ch(x, y, z) ((x & (y ^ z)) ^ z)
#define Maj(x, y, z) ((x & (y | z)) | (y & z))
#define ROTR(x, n) ((x >> n) | (x << (32 - n)))
#define S0(x) (ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
#define S1(x) (ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
#define s0(x) (ROTR(x, 7) ^ ROTR(x, 18) ^ (x >> 3))
#define s1(x) (ROTR(x, 17) ^ ROTR(x, 19) ^ (x >> 10))
/* SHA256 round function */
#define RND(a, b, c, d, e, f, g, h, k) \
do { \
t0 = h + S1(e) + Ch(e, f, g) + k; \
t1 = S0(a) + Maj(a, b, c); \
d += t0; \
h = t0 + t1; \
} while (0)
/* Adjusted round function for rotating state */
#define RNDr(S, W, i) \
RND(S[(64 - i) % 8], S[(65 - i) % 8], \
S[(66 - i) % 8], S[(67 - i) % 8], \
S[(68 - i) % 8], S[(69 - i) % 8], \
S[(70 - i) % 8], S[(71 - i) % 8], \
W[i] + sha256_k[i])
#ifndef EXTERN_SHA256
/*
* SHA256 block compression function. The 256-bit state is transformed via
* the 512-bit input block to produce a new state.
*/
void sha256_transform(uint32_t *state, const uint32_t *block, int swap)
{
uint32_t W[64];
uint32_t S[8];
uint32_t t0, t1;
int i;
/* 1. Prepare message schedule W. */
if (swap) {
for (i = 0; i < 16; i++)
W[i] = swab32(block[i]);
} else
memcpy(W, block, 64);
for (i = 16; i < 64; i += 2) {
W[i] = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
W[i+1] = s1(W[i - 1]) + W[i - 6] + s0(W[i - 14]) + W[i - 15];
}
/* 2. Initialize working variables. */
memcpy(S, state, 32);
/* 3. Mix. */
RNDr(S, W, 0);
RNDr(S, W, 1);
RNDr(S, W, 2);
RNDr(S, W, 3);
RNDr(S, W, 4);
RNDr(S, W, 5);
RNDr(S, W, 6);
RNDr(S, W, 7);
RNDr(S, W, 8);
RNDr(S, W, 9);
RNDr(S, W, 10);
RNDr(S, W, 11);
RNDr(S, W, 12);
RNDr(S, W, 13);
RNDr(S, W, 14);
RNDr(S, W, 15);
RNDr(S, W, 16);
RNDr(S, W, 17);
RNDr(S, W, 18);
RNDr(S, W, 19);
RNDr(S, W, 20);
RNDr(S, W, 21);
RNDr(S, W, 22);
RNDr(S, W, 23);
RNDr(S, W, 24);
RNDr(S, W, 25);
RNDr(S, W, 26);
RNDr(S, W, 27);
RNDr(S, W, 28);
RNDr(S, W, 29);
RNDr(S, W, 30);
RNDr(S, W, 31);
RNDr(S, W, 32);
RNDr(S, W, 33);
RNDr(S, W, 34);
RNDr(S, W, 35);
RNDr(S, W, 36);
RNDr(S, W, 37);
RNDr(S, W, 38);
RNDr(S, W, 39);
RNDr(S, W, 40);
RNDr(S, W, 41);
RNDr(S, W, 42);
RNDr(S, W, 43);
RNDr(S, W, 44);
RNDr(S, W, 45);
RNDr(S, W, 46);
RNDr(S, W, 47);
RNDr(S, W, 48);
RNDr(S, W, 49);
RNDr(S, W, 50);
RNDr(S, W, 51);
RNDr(S, W, 52);
RNDr(S, W, 53);
RNDr(S, W, 54);
RNDr(S, W, 55);
RNDr(S, W, 56);
RNDr(S, W, 57);
RNDr(S, W, 58);
RNDr(S, W, 59);
RNDr(S, W, 60);
RNDr(S, W, 61);
RNDr(S, W, 62);
RNDr(S, W, 63);
/* 4. Mix local working variables into global state */
for (i = 0; i < 8; i++)
state[i] += S[i];
}
#endif /* EXTERN_SHA256 */
static const uint32_t sha256d_hash1[16] = {
0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000000,
0x80000000, 0x00000000, 0x00000000, 0x00000000,
0x00000000, 0x00000000, 0x00000000, 0x00000100
};
static void sha256d_80_swap(uint32_t *hash, const uint32_t *data)
{
uint32_t S[16];
int i;
sha256_init(S);
sha256_transform(S, data, 0);
sha256_transform(S, data + 16, 0);
memcpy(S + 8, sha256d_hash1 + 8, 32);
sha256_init(hash);
sha256_transform(hash, S, 0);
for (i = 0; i < 8; i++)
hash[i] = swab32(hash[i]);
}
extern void sha256d(unsigned char *hash, const unsigned char *data, int len)
{
uint32_t S[16], T[16];
int i, r;
sha256_init(S);
for (r = len; r > -9; r -= 64) {
if (r < 64)
memset(T, 0, 64);
memcpy(T, data + len - r, r > 64 ? 64 : (r < 0 ? 0 : r));
if (r >= 0 && r < 64)
((unsigned char *)T)[r] = 0x80;
for (i = 0; i < 16; i++)
T[i] = be32dec(T + i);
if (r < 56)
T[15] = 8 * len;
sha256_transform(S, T, 0);
}
memcpy(S + 8, sha256d_hash1 + 8, 32);
sha256_init(T);
sha256_transform(T, S, 0);
for (i = 0; i < 8; i++)
be32enc((uint32_t *)hash + i, T[i]);
}
static inline void sha256d_preextend(uint32_t *W)
{
W[16] = s1(W[14]) + W[ 9] + s0(W[ 1]) + W[ 0];
W[17] = s1(W[15]) + W[10] + s0(W[ 2]) + W[ 1];
W[18] = s1(W[16]) + W[11] + W[ 2];
W[19] = s1(W[17]) + W[12] + s0(W[ 4]);
W[20] = W[13] + s0(W[ 5]) + W[ 4];
W[21] = W[14] + s0(W[ 6]) + W[ 5];
W[22] = W[15] + s0(W[ 7]) + W[ 6];
W[23] = W[16] + s0(W[ 8]) + W[ 7];
W[24] = W[17] + s0(W[ 9]) + W[ 8];
W[25] = s0(W[10]) + W[ 9];
W[26] = s0(W[11]) + W[10];
W[27] = s0(W[12]) + W[11];
W[28] = s0(W[13]) + W[12];
W[29] = s0(W[14]) + W[13];
W[30] = s0(W[15]) + W[14];
W[31] = s0(W[16]) + W[15];
}
static inline void sha256d_prehash(uint32_t *S, const uint32_t *W)
{
uint32_t t0, t1;
RNDr(S, W, 0);
RNDr(S, W, 1);
RNDr(S, W, 2);
}
#ifdef EXTERN_SHA256
void sha256d_ms(uint32_t *hash, uint32_t *W,
const uint32_t *midstate, const uint32_t *prehash);
#else
static inline void sha256d_ms(uint32_t *hash, uint32_t *W,
const uint32_t *midstate, const uint32_t *prehash)
{
uint32_t S[64];
uint32_t t0, t1;
int i;
S[18] = W[18];
S[19] = W[19];
S[20] = W[20];
S[22] = W[22];
S[23] = W[23];
S[24] = W[24];
S[30] = W[30];
S[31] = W[31];
W[18] += s0(W[3]);
W[19] += W[3];
W[20] += s1(W[18]);
W[21] = s1(W[19]);
W[22] += s1(W[20]);
W[23] += s1(W[21]);
W[24] += s1(W[22]);
W[25] = s1(W[23]) + W[18];
W[26] = s1(W[24]) + W[19];
W[27] = s1(W[25]) + W[20];
W[28] = s1(W[26]) + W[21];
W[29] = s1(W[27]) + W[22];
W[30] += s1(W[28]) + W[23];
W[31] += s1(W[29]) + W[24];
for (i = 32; i < 64; i += 2) {
W[i] = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
W[i+1] = s1(W[i - 1]) + W[i - 6] + s0(W[i - 14]) + W[i - 15];
}
memcpy(S, prehash, 32);
RNDr(S, W, 3);
RNDr(S, W, 4);
RNDr(S, W, 5);
RNDr(S, W, 6);
RNDr(S, W, 7);
RNDr(S, W, 8);
RNDr(S, W, 9);
RNDr(S, W, 10);
RNDr(S, W, 11);
RNDr(S, W, 12);
RNDr(S, W, 13);
RNDr(S, W, 14);
RNDr(S, W, 15);
RNDr(S, W, 16);
RNDr(S, W, 17);
RNDr(S, W, 18);
RNDr(S, W, 19);
RNDr(S, W, 20);
RNDr(S, W, 21);
RNDr(S, W, 22);
RNDr(S, W, 23);
RNDr(S, W, 24);
RNDr(S, W, 25);
RNDr(S, W, 26);
RNDr(S, W, 27);
RNDr(S, W, 28);
RNDr(S, W, 29);
RNDr(S, W, 30);
RNDr(S, W, 31);
RNDr(S, W, 32);
RNDr(S, W, 33);
RNDr(S, W, 34);
RNDr(S, W, 35);
RNDr(S, W, 36);
RNDr(S, W, 37);
RNDr(S, W, 38);
RNDr(S, W, 39);
RNDr(S, W, 40);
RNDr(S, W, 41);
RNDr(S, W, 42);
RNDr(S, W, 43);
RNDr(S, W, 44);
RNDr(S, W, 45);
RNDr(S, W, 46);
RNDr(S, W, 47);
RNDr(S, W, 48);
RNDr(S, W, 49);
RNDr(S, W, 50);
RNDr(S, W, 51);
RNDr(S, W, 52);
RNDr(S, W, 53);
RNDr(S, W, 54);
RNDr(S, W, 55);
RNDr(S, W, 56);
RNDr(S, W, 57);
RNDr(S, W, 58);
RNDr(S, W, 59);
RNDr(S, W, 60);
RNDr(S, W, 61);
RNDr(S, W, 62);
RNDr(S, W, 63);
for (i = 0; i < 8; i++)
S[i] += midstate[i];
W[18] = S[18];
W[19] = S[19];
W[20] = S[20];
W[22] = S[22];
W[23] = S[23];
W[24] = S[24];
W[30] = S[30];
W[31] = S[31];
memcpy(S + 8, sha256d_hash1 + 8, 32);
S[16] = s1(sha256d_hash1[14]) + sha256d_hash1[ 9] + s0(S[ 1]) + S[ 0];
S[17] = s1(sha256d_hash1[15]) + sha256d_hash1[10] + s0(S[ 2]) + S[ 1];
S[18] = s1(S[16]) + sha256d_hash1[11] + s0(S[ 3]) + S[ 2];
S[19] = s1(S[17]) + sha256d_hash1[12] + s0(S[ 4]) + S[ 3];
S[20] = s1(S[18]) + sha256d_hash1[13] + s0(S[ 5]) + S[ 4];
S[21] = s1(S[19]) + sha256d_hash1[14] + s0(S[ 6]) + S[ 5];
S[22] = s1(S[20]) + sha256d_hash1[15] + s0(S[ 7]) + S[ 6];
S[23] = s1(S[21]) + S[16] + s0(sha256d_hash1[ 8]) + S[ 7];
S[24] = s1(S[22]) + S[17] + s0(sha256d_hash1[ 9]) + sha256d_hash1[ 8];
S[25] = s1(S[23]) + S[18] + s0(sha256d_hash1[10]) + sha256d_hash1[ 9];
S[26] = s1(S[24]) + S[19] + s0(sha256d_hash1[11]) + sha256d_hash1[10];
S[27] = s1(S[25]) + S[20] + s0(sha256d_hash1[12]) + sha256d_hash1[11];
S[28] = s1(S[26]) + S[21] + s0(sha256d_hash1[13]) + sha256d_hash1[12];
S[29] = s1(S[27]) + S[22] + s0(sha256d_hash1[14]) + sha256d_hash1[13];
S[30] = s1(S[28]) + S[23] + s0(sha256d_hash1[15]) + sha256d_hash1[14];
S[31] = s1(S[29]) + S[24] + s0(S[16]) + sha256d_hash1[15];
for (i = 32; i < 60; i += 2) {
S[i] = s1(S[i - 2]) + S[i - 7] + s0(S[i - 15]) + S[i - 16];
S[i+1] = s1(S[i - 1]) + S[i - 6] + s0(S[i - 14]) + S[i - 15];
}
S[60] = s1(S[58]) + S[53] + s0(S[45]) + S[44];
sha256_init(hash);
RNDr(hash, S, 0);
RNDr(hash, S, 1);
RNDr(hash, S, 2);
RNDr(hash, S, 3);
RNDr(hash, S, 4);
RNDr(hash, S, 5);
RNDr(hash, S, 6);
RNDr(hash, S, 7);
RNDr(hash, S, 8);
RNDr(hash, S, 9);
RNDr(hash, S, 10);
RNDr(hash, S, 11);
RNDr(hash, S, 12);
RNDr(hash, S, 13);
RNDr(hash, S, 14);
RNDr(hash, S, 15);
RNDr(hash, S, 16);
RNDr(hash, S, 17);
RNDr(hash, S, 18);
RNDr(hash, S, 19);
RNDr(hash, S, 20);
RNDr(hash, S, 21);
RNDr(hash, S, 22);
RNDr(hash, S, 23);
RNDr(hash, S, 24);
RNDr(hash, S, 25);
RNDr(hash, S, 26);
RNDr(hash, S, 27);
RNDr(hash, S, 28);
RNDr(hash, S, 29);
RNDr(hash, S, 30);
RNDr(hash, S, 31);
RNDr(hash, S, 32);
RNDr(hash, S, 33);
RNDr(hash, S, 34);
RNDr(hash, S, 35);
RNDr(hash, S, 36);
RNDr(hash, S, 37);
RNDr(hash, S, 38);
RNDr(hash, S, 39);
RNDr(hash, S, 40);
RNDr(hash, S, 41);
RNDr(hash, S, 42);
RNDr(hash, S, 43);
RNDr(hash, S, 44);
RNDr(hash, S, 45);
RNDr(hash, S, 46);
RNDr(hash, S, 47);
RNDr(hash, S, 48);
RNDr(hash, S, 49);
RNDr(hash, S, 50);
RNDr(hash, S, 51);
RNDr(hash, S, 52);
RNDr(hash, S, 53);
RNDr(hash, S, 54);
RNDr(hash, S, 55);
RNDr(hash, S, 56);
hash[2] += hash[6] + S1(hash[3]) + Ch(hash[3], hash[4], hash[5])
+ S[57] + sha256_k[57];
hash[1] += hash[5] + S1(hash[2]) + Ch(hash[2], hash[3], hash[4])
+ S[58] + sha256_k[58];
hash[0] += hash[4] + S1(hash[1]) + Ch(hash[1], hash[2], hash[3])
+ S[59] + sha256_k[59];
hash[7] += hash[3] + S1(hash[0]) + Ch(hash[0], hash[1], hash[2])
+ S[60] + sha256_k[60]
+ sha256_h[7];
}
#endif /* EXTERN_SHA256 */
#ifdef HAVE_SHA256_4WAY
void sha256d_ms_4way(uint32_t *hash, uint32_t *data,
const uint32_t *midstate, const uint32_t *prehash);
static inline int scanhash_sha256d_4way(int thr_id, struct work *work,
uint32_t max_nonce, uint64_t *hashes_done)
{
uint32_t *pdata = work->data;
uint32_t *ptarget = work->target;
uint32_t _ALIGN(128) data[4 * 64];
uint32_t _ALIGN(32) hash[4 * 8];
uint32_t _ALIGN(32) midstate[4 * 8];
uint32_t _ALIGN(32) prehash[4 * 8];
uint32_t n = pdata[19] - 1;
const uint32_t first_nonce = pdata[19];
const uint32_t Htarg = ptarget[7];
int i, j;
memcpy(data, pdata + 16, 64);
sha256d_preextend(data);
for (i = 31; i >= 0; i--)
for (j = 0; j < 4; j++)
data[i * 4 + j] = data[i];
sha256_init(midstate);
sha256_transform(midstate, pdata, 0);
memcpy(prehash, midstate, 32);
sha256d_prehash(prehash, pdata + 16);
for (i = 7; i >= 0; i--) {
for (j = 0; j < 4; j++) {
midstate[i * 4 + j] = midstate[i];
prehash[i * 4 + j] = prehash[i];
}
}
do {
for (i = 0; i < 4; i++)
data[4 * 3 + i] = ++n;
sha256d_ms_4way(hash, data, midstate, prehash);
for (i = 0; i < 4; i++) {
if (swab32(hash[4 * 7 + i]) <= Htarg) {
pdata[19] = data[4 * 3 + i];
sha256d_80_swap(hash, pdata);
if (fulltest(hash, ptarget)) {
*hashes_done = n - first_nonce + 1;
return 1;
}
}
}
} while (n < max_nonce && !work_restart[thr_id].restart);
*hashes_done = n - first_nonce + 1;
pdata[19] = n;
return 0;
}
#endif /* HAVE_SHA256_4WAY */
#ifdef HAVE_SHA256_8WAY
void sha256d_ms_8way(uint32_t *hash, uint32_t *data,
const uint32_t *midstate, const uint32_t *prehash);
static inline int scanhash_sha256d_8way(int thr_id, struct work *work,
uint32_t max_nonce, uint64_t *hashes_done)
{
uint32_t *pdata = work->data;
uint32_t *ptarget = work->target;
uint32_t _ALIGN(128) data[8 * 64];
uint32_t _ALIGN(32) hash[8 * 8];
uint32_t _ALIGN(32) midstate[8 * 8];
uint32_t _ALIGN(32) prehash[8 * 8];
uint32_t n = pdata[19] - 1;
const uint32_t first_nonce = pdata[19];
const uint32_t Htarg = ptarget[7];
int i, j;
memcpy(data, pdata + 16, 64);
sha256d_preextend(data);
for (i = 31; i >= 0; i--)
for (j = 0; j < 8; j++)
data[i * 8 + j] = data[i];
sha256_init(midstate);
sha256_transform(midstate, pdata, 0);
memcpy(prehash, midstate, 32);
sha256d_prehash(prehash, pdata + 16);
for (i = 7; i >= 0; i--) {
for (j = 0; j < 8; j++) {
midstate[i * 8 + j] = midstate[i];
prehash[i * 8 + j] = prehash[i];
}
}
do {
for (i = 0; i < 8; i++)
data[8 * 3 + i] = ++n;
sha256d_ms_8way(hash, data, midstate, prehash);
for (i = 0; i < 8; i++) {
if (swab32(hash[8 * 7 + i]) <= Htarg) {
pdata[19] = data[8 * 3 + i];
sha256d_80_swap(hash, pdata);
if (fulltest(hash, ptarget)) {
*hashes_done = n - first_nonce + 1;
return 1;
}
}
}
} while (n < max_nonce && !work_restart[thr_id].restart);
*hashes_done = n - first_nonce + 1;
pdata[19] = n;
return 0;
}
#endif /* HAVE_SHA256_8WAY */
int scanhash_sha256d(int thr_id, struct work *work,
uint32_t max_nonce, uint64_t *hashes_done)
{
uint32_t *pdata = work->data;
uint32_t *ptarget = work->target;
uint32_t _ALIGN(128) data[64];
uint32_t _ALIGN(32) hash[8];
uint32_t _ALIGN(32) midstate[8];
uint32_t _ALIGN(32) prehash[8];
uint32_t n = pdata[19] - 1;
const uint32_t first_nonce = pdata[19];
const uint32_t Htarg = ptarget[7];
#ifdef HAVE_SHA256_8WAY
if (sha256_use_8way())
return scanhash_sha256d_8way(thr_id, work,
max_nonce, hashes_done);
#endif
#ifdef HAVE_SHA256_4WAY
if (sha256_use_4way())
return scanhash_sha256d_4way(thr_id, work,
max_nonce, hashes_done);
#endif
memcpy(data, pdata + 16, 64);
sha256d_preextend(data);
sha256_init(midstate);
sha256_transform(midstate, pdata, 0);
memcpy(prehash, midstate, 32);
sha256d_prehash(prehash, pdata + 16);
do {
data[3] = ++n;
sha256d_ms(hash, data, midstate, prehash);
if (unlikely(swab32(hash[7]) <= Htarg)) {
pdata[19] = data[3];
sha256d_80_swap(hash, pdata);
if (fulltest(hash, ptarget)) {
*hashes_done = n - first_nonce + 1;
return 1;
}
}
} while (likely(n < max_nonce && !work_restart[thr_id].restart));
*hashes_done = n - first_nonce + 1;
pdata[19] = n;
return 0;
}
bool register_sha256d_algo( algo_gate_t* gate )
{
gate->scanhash = (void*)&scanhash_sha256d;
gate->hash = (void*)&sha256d;
return true;
};

164
algo/sha/sha256t.c Normal file
View File

@@ -0,0 +1,164 @@
#include "miner.h"
#include "algo-gate-api.h"
#include <stdlib.h>
#include <stdint.h>
#include <string.h>
#include <stdio.h>
#include "sph_sha2.h"
#if defined (SHA_NI)
#include <openssl/sha.h>
static SHA256_CTX sha256t_ctx __attribute__ ((aligned (64)));
static __thread SHA256_CTX sha256t_mid __attribute__ ((aligned (64)));
#else
static sph_sha256_context sha256t_ctx __attribute__ ((aligned (64)));
static __thread sph_sha256_context sha256t_mid __attribute__ ((aligned (64)));
#endif
void sha256t_midstate( const void* input )
{
memcpy( &sha256t_mid, &sha256t_ctx, sizeof sha256t_mid );
#if defined (SHA_NI)
SHA256_Update( &sha256t_mid, input, 64 );
#else
sph_sha256( &sha256t_mid, input, 64 );
#endif
}
void sha256t_hash(void* output, const void* input, uint32_t len)
{
uint32_t _ALIGN(64) hashA[16];
const int midlen = 64; // bytes
const int tail = 80 - midlen; // 16
#if defined (SHA_NI)
SHA256_CTX ctx_sha256 __attribute__ ((aligned (64)));
memcpy( &ctx_sha256, &sha256t_mid, sizeof sha256t_mid );
SHA256_Update( &ctx_sha256, input + midlen, tail );
SHA256_Final( hashA, &ctx_sha256 );
memcpy( &ctx_sha256, &sha256t_ctx, sizeof sha256t_ctx );
SHA256_Update( &ctx_sha256, hashA, 32 );
SHA256_Final( hashA, &ctx_sha256 );
memcpy( &ctx_sha256, &sha256t_ctx, sizeof sha256t_ctx );
SHA256_Update( &ctx_sha256, hashA, 32 );
SHA256_Final( hashA, &ctx_sha256 );
#else
sph_sha256_context ctx_sha256 __attribute__ ((aligned (64)));
memcpy( &ctx_sha256, &sha256t_mid, sizeof sha256t_mid );
sph_sha256( &ctx_sha256, input + midlen, tail );
sph_sha256_close( &ctx_sha256, hashA );
memcpy( &ctx_sha256, &sha256t_ctx, sizeof sha256t_ctx );
sph_sha256( &ctx_sha256, hashA, 32 );
sph_sha256_close( &ctx_sha256, hashA );
memcpy( &ctx_sha256, &sha256t_ctx, sizeof sha256t_ctx );
sph_sha256( &ctx_sha256, hashA, 32 );
sph_sha256_close( &ctx_sha256, hashA );
#endif
memcpy( output, hashA, 32 );
}
int scanhash_sha256t(int thr_id, struct work *work,
uint32_t max_nonce, uint64_t *hashes_done)
{
uint32_t *pdata = work->data;
uint32_t *ptarget = work->target;
uint32_t len = 80;
uint32_t n = pdata[19] - 1;
const uint32_t first_nonce = pdata[19];
const uint32_t Htarg = ptarget[7];
#ifdef _MSC_VER
uint32_t __declspec(align(32)) hash64[8];
#else
uint32_t hash64[8] __attribute__((aligned(32)));
#endif
uint32_t endiandata[32];
uint64_t htmax[] = {
0,
0xF,
0xFF,
0xFFF,
0xFFFF,
0x10000000
};
uint32_t masks[] = {
0xFFFFFFFF,
0xFFFFFFF0,
0xFFFFFF00,
0xFFFFF000,
0xFFFF0000,
0
};
// we need bigendian data...
for (int k = 0; k < 19; k++)
be32enc(&endiandata[k], pdata[k]);
sha256t_midstate( endiandata );
#ifdef DEBUG_ALGO
if (Htarg != 0)
printf("[%d] Htarg=%X\n", thr_id, Htarg);
#endif
for (int m=0; m < 6; m++) {
if (Htarg <= htmax[m]) {
uint32_t mask = masks[m];
do {
pdata[19] = ++n;
be32enc(&endiandata[19], n);
sha256t_hash(hash64, endiandata, len);
#ifndef DEBUG_ALGO
if ((!(hash64[7] & mask)) && fulltest(hash64, ptarget)) {
*hashes_done = n - first_nonce + 1;
return true;
}
#else
if (!(n % 0x1000) && !thr_id) printf(".");
if (!(hash64[7] & mask)) {
printf("[%d]",thr_id);
if (fulltest(hash64, ptarget)) {
*hashes_done = n - first_nonce + 1;
return true;
}
}
#endif
} while (n < max_nonce && !work_restart[thr_id].restart);
// see blake.c if else to understand the loop on htmax => mask
break;
}
}
*hashes_done = n - first_nonce + 1;
pdata[19] = n;
return 0;
}
void sha256t_set_target( struct work* work, double job_diff )
{
work_set_target( work, job_diff / (256.0 * opt_diff_factor) );
}
bool register_sha256t_algo( algo_gate_t* gate )
{
#if defined (SHA_NI)
SHA256_Init( &sha256t_ctx );
#else
sph_sha256_init( &sha256t_ctx );
#endif
gate->optimizations = SSE2_OPT | SHA_OPT;
gate->scanhash = (void*)&scanhash_sha256t;
gate->hash = (void*)&sha256t_hash;
gate->set_target = (void*)&sha256t_set_target;
gate->get_max64 = (void*)&get_max64_0x3ffff;
return true;
}

22
algo/sha/sha3-defs.h Normal file
View File

@@ -0,0 +1,22 @@
#ifndef DEFS_X5_H__
#define DEFS_X5_H__
#include <emmintrin.h>
typedef unsigned char BitSequence;
typedef unsigned long long DataLength;
typedef enum { SUCCESS = 0, FAIL = 1, BAD_HASHBITLEN = 2} HashReturn;
typedef unsigned char uint8;
typedef unsigned int uint32;
typedef unsigned long long uint64;
//typedef struct {
// uint32 buffer[8]; /* Buffer to be hashed */
// __m128i chainv[10]; /* Chaining values */
// uint64 bitlen[2]; /* Message length in bits */
// uint32 rembitlen; /* Length of buffer data to be hashed */
// int hashbitlen;
//} hashState_luffa;
typedef unsigned char byte;
#endif

31
algo/sha/sha3_common.h Normal file
View File

@@ -0,0 +1,31 @@
/*
* file : sha3_common.h
* version : 1.0.208
* date : 14.12.2010
*
* Common declarations
*
* Cagdas Calik
* ccalik@metu.edu.tr
* Institute of Applied Mathematics, Middle East Technical University, Turkey.
*
*/
#include "sha3-defs.h"
#ifndef SHA3_COMMON_H
#define SHA3_COMMON_H
#ifdef __GNUC__
#define MYALIGN __attribute__((aligned(16)))
#else
#define MYALIGN __declspec(align(16))
#endif
#define M128(x) *((__m128i*)x)
//typedef unsigned char BitSequence;
//typedef unsigned long long DataLength;
//typedef enum {SUCCESS = 0, FAIL = 1, BAD_HASHBITLEN = 2} HashReturn;
#endif // SHA3_COMMON_H

693
algo/sha/sph-sha2.c.hide Normal file
View File

@@ -0,0 +1,693 @@
/* $Id: sha2.c 227 2010-06-16 17:28:38Z tp $ */
/*
* SHA-224 / SHA-256 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-sha2.h"
#if SPH_SMALL_FOOTPRINT && !defined SPH_SMALL_FOOTPRINT_SHA2
#define SPH_SMALL_FOOTPRINT_SHA2 1
#endif
#define CH(X, Y, Z) ((((Y) ^ (Z)) & (X)) ^ (Z))
#define MAJ(X, Y, Z) (((Y) & (Z)) | (((Y) | (Z)) & (X)))
#define ROTR SPH_ROTR32
#define BSG2_0(x) (ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
#define BSG2_1(x) (ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
#define SSG2_0(x) (ROTR(x, 7) ^ ROTR(x, 18) ^ SPH_T32((x) >> 3))
#define SSG2_1(x) (ROTR(x, 17) ^ ROTR(x, 19) ^ SPH_T32((x) >> 10))
static const sph_u32 H224[8] = {
SPH_C32(0xC1059ED8), SPH_C32(0x367CD507), SPH_C32(0x3070DD17),
SPH_C32(0xF70E5939), SPH_C32(0xFFC00B31), SPH_C32(0x68581511),
SPH_C32(0x64F98FA7), SPH_C32(0xBEFA4FA4)
};
static const sph_u32 H256[8] = {
SPH_C32(0x6A09E667), SPH_C32(0xBB67AE85), SPH_C32(0x3C6EF372),
SPH_C32(0xA54FF53A), SPH_C32(0x510E527F), SPH_C32(0x9B05688C),
SPH_C32(0x1F83D9AB), SPH_C32(0x5BE0CD19)
};
/*
* The SHA2_ROUND_BODY defines the body for a SHA-224 / SHA-256
* compression function implementation. The "in" parameter should
* evaluate, when applied to a numerical input parameter from 0 to 15,
* to an expression which yields the corresponding input block. The "r"
* parameter should evaluate to an array or pointer expression
* designating the array of 8 words which contains the input and output
* of the compression function.
*/
#if SPH_SMALL_FOOTPRINT_SHA2
static const sph_u32 K[64] = {
SPH_C32(0x428A2F98), SPH_C32(0x71374491),
SPH_C32(0xB5C0FBCF), SPH_C32(0xE9B5DBA5),
SPH_C32(0x3956C25B), SPH_C32(0x59F111F1),
SPH_C32(0x923F82A4), SPH_C32(0xAB1C5ED5),
SPH_C32(0xD807AA98), SPH_C32(0x12835B01),
SPH_C32(0x243185BE), SPH_C32(0x550C7DC3),
SPH_C32(0x72BE5D74), SPH_C32(0x80DEB1FE),
SPH_C32(0x9BDC06A7), SPH_C32(0xC19BF174),
SPH_C32(0xE49B69C1), SPH_C32(0xEFBE4786),
SPH_C32(0x0FC19DC6), SPH_C32(0x240CA1CC),
SPH_C32(0x2DE92C6F), SPH_C32(0x4A7484AA),
SPH_C32(0x5CB0A9DC), SPH_C32(0x76F988DA),
SPH_C32(0x983E5152), SPH_C32(0xA831C66D),
SPH_C32(0xB00327C8), SPH_C32(0xBF597FC7),
SPH_C32(0xC6E00BF3), SPH_C32(0xD5A79147),
SPH_C32(0x06CA6351), SPH_C32(0x14292967),
SPH_C32(0x27B70A85), SPH_C32(0x2E1B2138),
SPH_C32(0x4D2C6DFC), SPH_C32(0x53380D13),
SPH_C32(0x650A7354), SPH_C32(0x766A0ABB),
SPH_C32(0x81C2C92E), SPH_C32(0x92722C85),
SPH_C32(0xA2BFE8A1), SPH_C32(0xA81A664B),
SPH_C32(0xC24B8B70), SPH_C32(0xC76C51A3),
SPH_C32(0xD192E819), SPH_C32(0xD6990624),
SPH_C32(0xF40E3585), SPH_C32(0x106AA070),
SPH_C32(0x19A4C116), SPH_C32(0x1E376C08),
SPH_C32(0x2748774C), SPH_C32(0x34B0BCB5),
SPH_C32(0x391C0CB3), SPH_C32(0x4ED8AA4A),
SPH_C32(0x5B9CCA4F), SPH_C32(0x682E6FF3),
SPH_C32(0x748F82EE), SPH_C32(0x78A5636F),
SPH_C32(0x84C87814), SPH_C32(0x8CC70208),
SPH_C32(0x90BEFFFA), SPH_C32(0xA4506CEB),
SPH_C32(0xBEF9A3F7), SPH_C32(0xC67178F2)
};
#define SHA2_MEXP1(in, pc) do { \
W[pc] = in(pc); \
} while (0)
#define SHA2_MEXP2(in, pc) do { \
W[(pc) & 0x0F] = SPH_T32(SSG2_1(W[((pc) - 2) & 0x0F]) \
+ W[((pc) - 7) & 0x0F] \
+ SSG2_0(W[((pc) - 15) & 0x0F]) + W[(pc) & 0x0F]); \
} while (0)
#define SHA2_STEPn(n, a, b, c, d, e, f, g, h, in, pc) do { \
sph_u32 t1, t2; \
SHA2_MEXP ## n(in, pc); \
t1 = SPH_T32(h + BSG2_1(e) + CH(e, f, g) \
+ K[pcount + (pc)] + W[(pc) & 0x0F]); \
t2 = SPH_T32(BSG2_0(a) + MAJ(a, b, c)); \
d = SPH_T32(d + t1); \
h = SPH_T32(t1 + t2); \
} while (0)
#define SHA2_STEP1(a, b, c, d, e, f, g, h, in, pc) \
SHA2_STEPn(1, a, b, c, d, e, f, g, h, in, pc)
#define SHA2_STEP2(a, b, c, d, e, f, g, h, in, pc) \
SHA2_STEPn(2, a, b, c, d, e, f, g, h, in, pc)
#define SHA2_ROUND_BODY(in, r) do { \
sph_u32 A, B, C, D, E, F, G, H; \
sph_u32 W[16]; \
unsigned pcount; \
\
A = (r)[0]; \
B = (r)[1]; \
C = (r)[2]; \
D = (r)[3]; \
E = (r)[4]; \
F = (r)[5]; \
G = (r)[6]; \
H = (r)[7]; \
pcount = 0; \
SHA2_STEP1(A, B, C, D, E, F, G, H, in, 0); \
SHA2_STEP1(H, A, B, C, D, E, F, G, in, 1); \
SHA2_STEP1(G, H, A, B, C, D, E, F, in, 2); \
SHA2_STEP1(F, G, H, A, B, C, D, E, in, 3); \
SHA2_STEP1(E, F, G, H, A, B, C, D, in, 4); \
SHA2_STEP1(D, E, F, G, H, A, B, C, in, 5); \
SHA2_STEP1(C, D, E, F, G, H, A, B, in, 6); \
SHA2_STEP1(B, C, D, E, F, G, H, A, in, 7); \
SHA2_STEP1(A, B, C, D, E, F, G, H, in, 8); \
SHA2_STEP1(H, A, B, C, D, E, F, G, in, 9); \
SHA2_STEP1(G, H, A, B, C, D, E, F, in, 10); \
SHA2_STEP1(F, G, H, A, B, C, D, E, in, 11); \
SHA2_STEP1(E, F, G, H, A, B, C, D, in, 12); \
SHA2_STEP1(D, E, F, G, H, A, B, C, in, 13); \
SHA2_STEP1(C, D, E, F, G, H, A, B, in, 14); \
SHA2_STEP1(B, C, D, E, F, G, H, A, in, 15); \
for (pcount = 16; pcount < 64; pcount += 16) { \
SHA2_STEP2(A, B, C, D, E, F, G, H, in, 0); \
SHA2_STEP2(H, A, B, C, D, E, F, G, in, 1); \
SHA2_STEP2(G, H, A, B, C, D, E, F, in, 2); \
SHA2_STEP2(F, G, H, A, B, C, D, E, in, 3); \
SHA2_STEP2(E, F, G, H, A, B, C, D, in, 4); \
SHA2_STEP2(D, E, F, G, H, A, B, C, in, 5); \
SHA2_STEP2(C, D, E, F, G, H, A, B, in, 6); \
SHA2_STEP2(B, C, D, E, F, G, H, A, in, 7); \
SHA2_STEP2(A, B, C, D, E, F, G, H, in, 8); \
SHA2_STEP2(H, A, B, C, D, E, F, G, in, 9); \
SHA2_STEP2(G, H, A, B, C, D, E, F, in, 10); \
SHA2_STEP2(F, G, H, A, B, C, D, E, in, 11); \
SHA2_STEP2(E, F, G, H, A, B, C, D, in, 12); \
SHA2_STEP2(D, E, F, G, H, A, B, C, in, 13); \
SHA2_STEP2(C, D, E, F, G, H, A, B, in, 14); \
SHA2_STEP2(B, C, D, E, F, G, H, A, in, 15); \
} \
(r)[0] = SPH_T32((r)[0] + A); \
(r)[1] = SPH_T32((r)[1] + B); \
(r)[2] = SPH_T32((r)[2] + C); \
(r)[3] = SPH_T32((r)[3] + D); \
(r)[4] = SPH_T32((r)[4] + E); \
(r)[5] = SPH_T32((r)[5] + F); \
(r)[6] = SPH_T32((r)[6] + G); \
(r)[7] = SPH_T32((r)[7] + H); \
} while (0)
#else
#define SHA2_ROUND_BODY(in, r) do { \
sph_u32 A, B, C, D, E, F, G, H, T1, T2; \
sph_u32 W00, W01, W02, W03, W04, W05, W06, W07; \
sph_u32 W08, W09, W10, W11, W12, W13, W14, W15; \
int i; \
\
/* for (i=0;i<8;i++) {printf("in[%d]=%08x in[%d]=%08x \n",2*i,in(2*i),2*i+1,in(2*i+1));} */ \
A = (r)[0]; \
B = (r)[1]; \
C = (r)[2]; \
D = (r)[3]; \
E = (r)[4]; \
F = (r)[5]; \
G = (r)[6]; \
H = (r)[7]; \
W00 = in(0); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x428A2F98) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = in(1); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x71374491) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = in(2); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xB5C0FBCF) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = in(3); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xE9B5DBA5) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = in(4); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x3956C25B) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = in(5); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x59F111F1) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = in(6); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x923F82A4) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = in(7); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xAB1C5ED5) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = in(8); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xD807AA98) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = in(9); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x12835B01) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = in(10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x243185BE) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = in(11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x550C7DC3) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = in(12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x72BE5D74) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = in(13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x80DEB1FE) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = in(14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x9BDC06A7) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = in(15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xC19BF174) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xE49B69C1) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xEFBE4786) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x0FC19DC6) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x240CA1CC) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x2DE92C6F) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x4A7484AA) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x5CB0A9DC) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x76F988DA) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x983E5152) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xA831C66D) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xB00327C8) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xBF597FC7) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0xC6E00BF3) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xD5A79147) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x06CA6351) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x14292967) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x27B70A85) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x2E1B2138) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x4D2C6DFC) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x53380D13) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x650A7354) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x766A0ABB) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x81C2C92E) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x92722C85) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xA2BFE8A1) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xA81A664B) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xC24B8B70) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xC76C51A3) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0xD192E819) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xD6990624) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0xF40E3585) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x106AA070) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x19A4C116) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x1E376C08) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x2748774C) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x34B0BCB5) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x391C0CB3) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x4ED8AA4A) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x5B9CCA4F) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x682E6FF3) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x748F82EE) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x78A5636F) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x84C87814) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x8CC70208) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x90BEFFFA) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xA4506CEB) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0xBEF9A3F7) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xC67178F2) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
(r)[0] = SPH_T32((r)[0] + A); \
(r)[1] = SPH_T32((r)[1] + B); \
(r)[2] = SPH_T32((r)[2] + C); \
(r)[3] = SPH_T32((r)[3] + D); \
(r)[4] = SPH_T32((r)[4] + E); \
(r)[5] = SPH_T32((r)[5] + F); \
(r)[6] = SPH_T32((r)[6] + G); \
(r)[7] = SPH_T32((r)[7] + H); \
/* for (i=0;i<4;i++) {printf("r[%d]=%08x r[%d]=%08x\n",2*i,(r)[2*i],2*i+1,(r)[2*i+1]);} */ \
} while (0)
#endif
/*
* One round of SHA-224 / SHA-256. The data must be aligned for 32-bit access.
*/
static void
sha2_round(const unsigned char *data, sph_u32 r[8])
{
#define SHA2_IN(x) sph_dec32be_aligned(data + (4 * (x)))
SHA2_ROUND_BODY(SHA2_IN, r);
#undef SHA2_IN
}
/* see sph_sha2.h */
void
sph_sha224_init(void *cc)
{
sph_sha224_context *sc;
sc = cc;
memcpy(sc->val, H224, sizeof H224);
#if SPH_64
sc->count = 0;
#else
sc->count_high = sc->count_low = 0;
#endif
}
/* see sph_sha2.h */
void
sph_sha256_init(void *cc)
{
sph_sha256_context *sc;
sc = cc;
memcpy(sc->val, H256, sizeof H256);
#if SPH_64
sc->count = 0;
#else
sc->count_high = sc->count_low = 0;
#endif
}
#define RFUN sha2_round
#define HASH sha224
#define BE32 1
#include "md_helper.c"
/* see sph_sha2.h */
void
sph_sha224_close(void *cc, void *dst)
{
sha224_close(cc, dst, 7);
// sph_sha224_init(cc);
}
/* see sph_sha2.h */
void
sph_sha224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha224_addbits_and_close(cc, ub, n, dst, 7);
// sph_sha224_init(cc);
}
/* see sph_sha2.h */
void
sph_sha256_close(void *cc, void *dst)
{
sha224_close(cc, dst, 8);
// sph_sha256_init(cc);
}
/* see sph_sha2.h */
void
sph_sha256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha224_addbits_and_close(cc, ub, n, dst, 8);
// sph_sha256_init(cc);
}
/* see sph_sha2.h */
void
sph_sha224_comp(const sph_u32 msg[16], sph_u32 val[8])
{
#define SHA2_IN(x) msg[x]
SHA2_ROUND_BODY(SHA2_IN, val);
#undef SHA2_IN
}

378
algo/sha/sph-sha2.h.hide Normal file
View File

@@ -0,0 +1,378 @@
/* $Id: sph_sha2.h 216 2010-06-08 09:46:57Z tp $ */
/**
* SHA-224, SHA-256, SHA-384 and SHA-512 interface.
*
* SHA-256 has been published in FIPS 180-2, now amended with a change
* notice to include SHA-224 as well (which is a simple variation on
* SHA-256). SHA-384 and SHA-512 are also defined in FIPS 180-2. FIPS
* standards can be found at:
* http://csrc.nist.gov/publications/fips/
*
* ==========================(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)=============================
*
* @file sph_sha2.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_SHA2_H__
#define SPH_SHA2_H__
#include <stddef.h>
#include "sph_types.h"
#ifdef __cplusplus
extern "C"{
#endif
/**
* Output size (in bits) for SHA-224.
*/
#define SPH_SIZE_sha224 224
/**
* Output size (in bits) for SHA-256.
*/
#define SPH_SIZE_sha256 256
/**
* This structure is a context for SHA-224 computations: it contains the
* intermediate values and some data from the last entered block. Once
* a SHA-224 computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running SHA-224 computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
sph_u32 val[8];
#if SPH_64
sph_u64 count;
#else
sph_u32 count_high, count_low;
#endif
#endif
} sph_sha224_context;
/**
* This structure is a context for SHA-256 computations. It is identical
* to the SHA-224 context. However, a context is initialized for SHA-224
* <strong>or</strong> SHA-256, but not both (the internal IV is not the
* same).
*/
typedef sph_sha224_context sph_sha256_context;
/**
* Initialize a SHA-224 context. This process performs no memory allocation.
*
* @param cc the SHA-224 context (pointer to
* a <code>sph_sha224_context</code>)
*/
void sph_sha224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHA-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha224(void *cc, const void *data, size_t len);
/**
* Terminate the current SHA-224 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-224 context
* @param dst the destination buffer
*/
void sph_sha224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
/**
* Apply the SHA-224 compression function on the provided data. The
* <code>msg</code> parameter contains the 16 32-bit input blocks,
* as numerical values (hence after the big-endian decoding). The
* <code>val</code> parameter contains the 8 32-bit input blocks for
* the compression function; the output is written in place in this
* array.
*
* @param msg the message block (16 values)
* @param val the function 256-bit input and output
*/
void sph_sha224_comp(const sph_u32 msg[16], sph_u32 val[8]);
/**
* Initialize a SHA-256 context. This process performs no memory allocation.
*
* @param cc the SHA-256 context (pointer to
* a <code>sph_sha256_context</code>)
*/
void sph_sha256_init(void *cc);
#ifdef DOXYGEN_IGNORE
/**
* Process some data bytes, for SHA-256. This function is identical to
* <code>sha_224()</code>
*
* @param cc the SHA-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha256(void *cc, const void *data, size_t len);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha256 sph_sha224
#endif
/**
* Terminate the current SHA-256 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-256 context
* @param dst the destination buffer
*/
void sph_sha256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
#ifdef DOXYGEN_IGNORE
/**
* Apply the SHA-256 compression function on the provided data. This
* function is identical to <code>sha224_comp()</code>.
*
* @param msg the message block (16 values)
* @param val the function 256-bit input and output
*/
void sph_sha256_comp(const sph_u32 msg[16], sph_u32 val[8]);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha256_comp sph_sha224_comp
#endif
#if SPH_64
/**
* Output size (in bits) for SHA-384.
*/
#define SPH_SIZE_sha384 384
/**
* Output size (in bits) for SHA-512.
*/
#define SPH_SIZE_sha512 512
/**
* This structure is a context for SHA-384 computations: it contains the
* intermediate values and some data from the last entered block. Once
* a SHA-384 computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running SHA-384 computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
sph_u64 val[8];
sph_u64 count;
#endif
} sph_sha384_context;
/**
* Initialize a SHA-384 context. This process performs no memory allocation.
*
* @param cc the SHA-384 context (pointer to
* a <code>sph_sha384_context</code>)
*/
void sph_sha384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHA-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha384(void *cc, const void *data, size_t len);
/**
* Terminate the current SHA-384 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-384 context
* @param dst the destination buffer
*/
void sph_sha384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha384_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
/**
* Apply the SHA-384 compression function on the provided data. The
* <code>msg</code> parameter contains the 16 64-bit input blocks,
* as numerical values (hence after the big-endian decoding). The
* <code>val</code> parameter contains the 8 64-bit input blocks for
* the compression function; the output is written in place in this
* array.
*
* @param msg the message block (16 values)
* @param val the function 512-bit input and output
*/
void sph_sha384_comp(const sph_u64 msg[16], sph_u64 val[8]);
/**
* This structure is a context for SHA-512 computations. It is identical
* to the SHA-384 context. However, a context is initialized for SHA-384
* <strong>or</strong> SHA-512, but not both (the internal IV is not the
* same).
*/
typedef sph_sha384_context sph_sha512_context;
/**
* Initialize a SHA-512 context. This process performs no memory allocation.
*
* @param cc the SHA-512 context (pointer to
* a <code>sph_sha512_context</code>)
*/
void sph_sha512_init(void *cc);
#ifdef DOXYGEN_IGNORE
/**
* Process some data bytes, for SHA-512. This function is identical to
* <code>sph_sha384()</code>.
*
* @param cc the SHA-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha512(void *cc, const void *data, size_t len);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha512 sph_sha384
#endif
/**
* Terminate the current SHA-512 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-512 context
* @param dst the destination buffer
*/
void sph_sha512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha512_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
#ifdef DOXYGEN_IGNORE
/**
* Apply the SHA-512 compression function. This function is identical to
* <code>sph_sha384_comp()</code>.
*
* @param msg the message block (16 values)
* @param val the function 512-bit input and output
*/
void sph_sha512_comp(const sph_u64 msg[16], sph_u64 val[8]);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha512_comp sph_sha384_comp
#endif
#endif
#endif
#ifdef __cplusplus
}
#endif

690
algo/sha/sph_sha2.c Normal file
View File

@@ -0,0 +1,690 @@
/* $Id: sha2.c 227 2010-06-16 17:28:38Z tp $ */
/*
* SHA-224 / SHA-256 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_sha2.h"
#if SPH_SMALL_FOOTPRINT && !defined SPH_SMALL_FOOTPRINT_SHA2
#define SPH_SMALL_FOOTPRINT_SHA2 1
#endif
#define CH(X, Y, Z) ((((Y) ^ (Z)) & (X)) ^ (Z))
#define MAJ(X, Y, Z) (((Y) & (Z)) | (((Y) | (Z)) & (X)))
#define ROTR SPH_ROTR32
#define BSG2_0(x) (ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
#define BSG2_1(x) (ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
#define SSG2_0(x) (ROTR(x, 7) ^ ROTR(x, 18) ^ SPH_T32((x) >> 3))
#define SSG2_1(x) (ROTR(x, 17) ^ ROTR(x, 19) ^ SPH_T32((x) >> 10))
static const sph_u32 H224[8] = {
SPH_C32(0xC1059ED8), SPH_C32(0x367CD507), SPH_C32(0x3070DD17),
SPH_C32(0xF70E5939), SPH_C32(0xFFC00B31), SPH_C32(0x68581511),
SPH_C32(0x64F98FA7), SPH_C32(0xBEFA4FA4)
};
static const sph_u32 H256[8] = {
SPH_C32(0x6A09E667), SPH_C32(0xBB67AE85), SPH_C32(0x3C6EF372),
SPH_C32(0xA54FF53A), SPH_C32(0x510E527F), SPH_C32(0x9B05688C),
SPH_C32(0x1F83D9AB), SPH_C32(0x5BE0CD19)
};
/*
* The SHA2_ROUND_BODY defines the body for a SHA-224 / SHA-256
* compression function implementation. The "in" parameter should
* evaluate, when applied to a numerical input parameter from 0 to 15,
* to an expression which yields the corresponding input block. The "r"
* parameter should evaluate to an array or pointer expression
* designating the array of 8 words which contains the input and output
* of the compression function.
*/
#if SPH_SMALL_FOOTPRINT_SHA2
static const sph_u32 K[64] = {
SPH_C32(0x428A2F98), SPH_C32(0x71374491),
SPH_C32(0xB5C0FBCF), SPH_C32(0xE9B5DBA5),
SPH_C32(0x3956C25B), SPH_C32(0x59F111F1),
SPH_C32(0x923F82A4), SPH_C32(0xAB1C5ED5),
SPH_C32(0xD807AA98), SPH_C32(0x12835B01),
SPH_C32(0x243185BE), SPH_C32(0x550C7DC3),
SPH_C32(0x72BE5D74), SPH_C32(0x80DEB1FE),
SPH_C32(0x9BDC06A7), SPH_C32(0xC19BF174),
SPH_C32(0xE49B69C1), SPH_C32(0xEFBE4786),
SPH_C32(0x0FC19DC6), SPH_C32(0x240CA1CC),
SPH_C32(0x2DE92C6F), SPH_C32(0x4A7484AA),
SPH_C32(0x5CB0A9DC), SPH_C32(0x76F988DA),
SPH_C32(0x983E5152), SPH_C32(0xA831C66D),
SPH_C32(0xB00327C8), SPH_C32(0xBF597FC7),
SPH_C32(0xC6E00BF3), SPH_C32(0xD5A79147),
SPH_C32(0x06CA6351), SPH_C32(0x14292967),
SPH_C32(0x27B70A85), SPH_C32(0x2E1B2138),
SPH_C32(0x4D2C6DFC), SPH_C32(0x53380D13),
SPH_C32(0x650A7354), SPH_C32(0x766A0ABB),
SPH_C32(0x81C2C92E), SPH_C32(0x92722C85),
SPH_C32(0xA2BFE8A1), SPH_C32(0xA81A664B),
SPH_C32(0xC24B8B70), SPH_C32(0xC76C51A3),
SPH_C32(0xD192E819), SPH_C32(0xD6990624),
SPH_C32(0xF40E3585), SPH_C32(0x106AA070),
SPH_C32(0x19A4C116), SPH_C32(0x1E376C08),
SPH_C32(0x2748774C), SPH_C32(0x34B0BCB5),
SPH_C32(0x391C0CB3), SPH_C32(0x4ED8AA4A),
SPH_C32(0x5B9CCA4F), SPH_C32(0x682E6FF3),
SPH_C32(0x748F82EE), SPH_C32(0x78A5636F),
SPH_C32(0x84C87814), SPH_C32(0x8CC70208),
SPH_C32(0x90BEFFFA), SPH_C32(0xA4506CEB),
SPH_C32(0xBEF9A3F7), SPH_C32(0xC67178F2)
};
#define SHA2_MEXP1(in, pc) do { \
W[pc] = in(pc); \
} while (0)
#define SHA2_MEXP2(in, pc) do { \
W[(pc) & 0x0F] = SPH_T32(SSG2_1(W[((pc) - 2) & 0x0F]) \
+ W[((pc) - 7) & 0x0F] \
+ SSG2_0(W[((pc) - 15) & 0x0F]) + W[(pc) & 0x0F]); \
} while (0)
#define SHA2_STEPn(n, a, b, c, d, e, f, g, h, in, pc) do { \
sph_u32 t1, t2; \
SHA2_MEXP ## n(in, pc); \
t1 = SPH_T32(h + BSG2_1(e) + CH(e, f, g) \
+ K[pcount + (pc)] + W[(pc) & 0x0F]); \
t2 = SPH_T32(BSG2_0(a) + MAJ(a, b, c)); \
d = SPH_T32(d + t1); \
h = SPH_T32(t1 + t2); \
} while (0)
#define SHA2_STEP1(a, b, c, d, e, f, g, h, in, pc) \
SHA2_STEPn(1, a, b, c, d, e, f, g, h, in, pc)
#define SHA2_STEP2(a, b, c, d, e, f, g, h, in, pc) \
SHA2_STEPn(2, a, b, c, d, e, f, g, h, in, pc)
#define SHA2_ROUND_BODY(in, r) do { \
sph_u32 A, B, C, D, E, F, G, H; \
sph_u32 W[16]; \
unsigned pcount; \
\
A = (r)[0]; \
B = (r)[1]; \
C = (r)[2]; \
D = (r)[3]; \
E = (r)[4]; \
F = (r)[5]; \
G = (r)[6]; \
H = (r)[7]; \
pcount = 0; \
SHA2_STEP1(A, B, C, D, E, F, G, H, in, 0); \
SHA2_STEP1(H, A, B, C, D, E, F, G, in, 1); \
SHA2_STEP1(G, H, A, B, C, D, E, F, in, 2); \
SHA2_STEP1(F, G, H, A, B, C, D, E, in, 3); \
SHA2_STEP1(E, F, G, H, A, B, C, D, in, 4); \
SHA2_STEP1(D, E, F, G, H, A, B, C, in, 5); \
SHA2_STEP1(C, D, E, F, G, H, A, B, in, 6); \
SHA2_STEP1(B, C, D, E, F, G, H, A, in, 7); \
SHA2_STEP1(A, B, C, D, E, F, G, H, in, 8); \
SHA2_STEP1(H, A, B, C, D, E, F, G, in, 9); \
SHA2_STEP1(G, H, A, B, C, D, E, F, in, 10); \
SHA2_STEP1(F, G, H, A, B, C, D, E, in, 11); \
SHA2_STEP1(E, F, G, H, A, B, C, D, in, 12); \
SHA2_STEP1(D, E, F, G, H, A, B, C, in, 13); \
SHA2_STEP1(C, D, E, F, G, H, A, B, in, 14); \
SHA2_STEP1(B, C, D, E, F, G, H, A, in, 15); \
for (pcount = 16; pcount < 64; pcount += 16) { \
SHA2_STEP2(A, B, C, D, E, F, G, H, in, 0); \
SHA2_STEP2(H, A, B, C, D, E, F, G, in, 1); \
SHA2_STEP2(G, H, A, B, C, D, E, F, in, 2); \
SHA2_STEP2(F, G, H, A, B, C, D, E, in, 3); \
SHA2_STEP2(E, F, G, H, A, B, C, D, in, 4); \
SHA2_STEP2(D, E, F, G, H, A, B, C, in, 5); \
SHA2_STEP2(C, D, E, F, G, H, A, B, in, 6); \
SHA2_STEP2(B, C, D, E, F, G, H, A, in, 7); \
SHA2_STEP2(A, B, C, D, E, F, G, H, in, 8); \
SHA2_STEP2(H, A, B, C, D, E, F, G, in, 9); \
SHA2_STEP2(G, H, A, B, C, D, E, F, in, 10); \
SHA2_STEP2(F, G, H, A, B, C, D, E, in, 11); \
SHA2_STEP2(E, F, G, H, A, B, C, D, in, 12); \
SHA2_STEP2(D, E, F, G, H, A, B, C, in, 13); \
SHA2_STEP2(C, D, E, F, G, H, A, B, in, 14); \
SHA2_STEP2(B, C, D, E, F, G, H, A, in, 15); \
} \
(r)[0] = SPH_T32((r)[0] + A); \
(r)[1] = SPH_T32((r)[1] + B); \
(r)[2] = SPH_T32((r)[2] + C); \
(r)[3] = SPH_T32((r)[3] + D); \
(r)[4] = SPH_T32((r)[4] + E); \
(r)[5] = SPH_T32((r)[5] + F); \
(r)[6] = SPH_T32((r)[6] + G); \
(r)[7] = SPH_T32((r)[7] + H); \
} while (0)
#else
#define SHA2_ROUND_BODY(in, r) do { \
sph_u32 A, B, C, D, E, F, G, H, T1, T2; \
sph_u32 W00, W01, W02, W03, W04, W05, W06, W07; \
sph_u32 W08, W09, W10, W11, W12, W13, W14, W15; \
\
A = (r)[0]; \
B = (r)[1]; \
C = (r)[2]; \
D = (r)[3]; \
E = (r)[4]; \
F = (r)[5]; \
G = (r)[6]; \
H = (r)[7]; \
W00 = in(0); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x428A2F98) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = in(1); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x71374491) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = in(2); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xB5C0FBCF) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = in(3); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xE9B5DBA5) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = in(4); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x3956C25B) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = in(5); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x59F111F1) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = in(6); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x923F82A4) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = in(7); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xAB1C5ED5) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = in(8); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xD807AA98) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = in(9); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x12835B01) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = in(10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x243185BE) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = in(11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x550C7DC3) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = in(12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x72BE5D74) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = in(13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x80DEB1FE) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = in(14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x9BDC06A7) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = in(15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xC19BF174) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xE49B69C1) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xEFBE4786) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x0FC19DC6) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x240CA1CC) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x2DE92C6F) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x4A7484AA) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x5CB0A9DC) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x76F988DA) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x983E5152) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xA831C66D) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xB00327C8) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xBF597FC7) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0xC6E00BF3) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xD5A79147) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x06CA6351) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x14292967) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x27B70A85) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x2E1B2138) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x4D2C6DFC) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x53380D13) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x650A7354) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x766A0ABB) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x81C2C92E) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x92722C85) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0xA2BFE8A1) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0xA81A664B) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0xC24B8B70) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0xC76C51A3) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0xD192E819) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xD6990624) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0xF40E3585) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x106AA070) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W00 = SPH_T32(SSG2_1(W14) + W09 + SSG2_0(W01) + W00); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x19A4C116) + W00); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W01 = SPH_T32(SSG2_1(W15) + W10 + SSG2_0(W02) + W01); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x1E376C08) + W01); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W02 = SPH_T32(SSG2_1(W00) + W11 + SSG2_0(W03) + W02); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x2748774C) + W02); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W03 = SPH_T32(SSG2_1(W01) + W12 + SSG2_0(W04) + W03); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x34B0BCB5) + W03); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W04 = SPH_T32(SSG2_1(W02) + W13 + SSG2_0(W05) + W04); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x391C0CB3) + W04); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W05 = SPH_T32(SSG2_1(W03) + W14 + SSG2_0(W06) + W05); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0x4ED8AA4A) + W05); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W06 = SPH_T32(SSG2_1(W04) + W15 + SSG2_0(W07) + W06); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0x5B9CCA4F) + W06); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W07 = SPH_T32(SSG2_1(W05) + W00 + SSG2_0(W08) + W07); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0x682E6FF3) + W07); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
W08 = SPH_T32(SSG2_1(W06) + W01 + SSG2_0(W09) + W08); \
T1 = SPH_T32(H + BSG2_1(E) + CH(E, F, G) \
+ SPH_C32(0x748F82EE) + W08); \
T2 = SPH_T32(BSG2_0(A) + MAJ(A, B, C)); \
D = SPH_T32(D + T1); \
H = SPH_T32(T1 + T2); \
W09 = SPH_T32(SSG2_1(W07) + W02 + SSG2_0(W10) + W09); \
T1 = SPH_T32(G + BSG2_1(D) + CH(D, E, F) \
+ SPH_C32(0x78A5636F) + W09); \
T2 = SPH_T32(BSG2_0(H) + MAJ(H, A, B)); \
C = SPH_T32(C + T1); \
G = SPH_T32(T1 + T2); \
W10 = SPH_T32(SSG2_1(W08) + W03 + SSG2_0(W11) + W10); \
T1 = SPH_T32(F + BSG2_1(C) + CH(C, D, E) \
+ SPH_C32(0x84C87814) + W10); \
T2 = SPH_T32(BSG2_0(G) + MAJ(G, H, A)); \
B = SPH_T32(B + T1); \
F = SPH_T32(T1 + T2); \
W11 = SPH_T32(SSG2_1(W09) + W04 + SSG2_0(W12) + W11); \
T1 = SPH_T32(E + BSG2_1(B) + CH(B, C, D) \
+ SPH_C32(0x8CC70208) + W11); \
T2 = SPH_T32(BSG2_0(F) + MAJ(F, G, H)); \
A = SPH_T32(A + T1); \
E = SPH_T32(T1 + T2); \
W12 = SPH_T32(SSG2_1(W10) + W05 + SSG2_0(W13) + W12); \
T1 = SPH_T32(D + BSG2_1(A) + CH(A, B, C) \
+ SPH_C32(0x90BEFFFA) + W12); \
T2 = SPH_T32(BSG2_0(E) + MAJ(E, F, G)); \
H = SPH_T32(H + T1); \
D = SPH_T32(T1 + T2); \
W13 = SPH_T32(SSG2_1(W11) + W06 + SSG2_0(W14) + W13); \
T1 = SPH_T32(C + BSG2_1(H) + CH(H, A, B) \
+ SPH_C32(0xA4506CEB) + W13); \
T2 = SPH_T32(BSG2_0(D) + MAJ(D, E, F)); \
G = SPH_T32(G + T1); \
C = SPH_T32(T1 + T2); \
W14 = SPH_T32(SSG2_1(W12) + W07 + SSG2_0(W15) + W14); \
T1 = SPH_T32(B + BSG2_1(G) + CH(G, H, A) \
+ SPH_C32(0xBEF9A3F7) + W14); \
T2 = SPH_T32(BSG2_0(C) + MAJ(C, D, E)); \
F = SPH_T32(F + T1); \
B = SPH_T32(T1 + T2); \
W15 = SPH_T32(SSG2_1(W13) + W08 + SSG2_0(W00) + W15); \
T1 = SPH_T32(A + BSG2_1(F) + CH(F, G, H) \
+ SPH_C32(0xC67178F2) + W15); \
T2 = SPH_T32(BSG2_0(B) + MAJ(B, C, D)); \
E = SPH_T32(E + T1); \
A = SPH_T32(T1 + T2); \
(r)[0] = SPH_T32((r)[0] + A); \
(r)[1] = SPH_T32((r)[1] + B); \
(r)[2] = SPH_T32((r)[2] + C); \
(r)[3] = SPH_T32((r)[3] + D); \
(r)[4] = SPH_T32((r)[4] + E); \
(r)[5] = SPH_T32((r)[5] + F); \
(r)[6] = SPH_T32((r)[6] + G); \
(r)[7] = SPH_T32((r)[7] + H); \
} while (0)
#endif
/*
* One round of SHA-224 / SHA-256. The data must be aligned for 32-bit access.
*/
static void
sha2_round(const unsigned char *data, sph_u32 r[8])
{
#define SHA2_IN(x) sph_dec32be_aligned(data + (4 * (x)))
SHA2_ROUND_BODY(SHA2_IN, r);
#undef SHA2_IN
}
/* see sph_sha2.h */
void
sph_sha224_init(void *cc)
{
sph_sha224_context *sc;
sc = cc;
memcpy(sc->val, H224, sizeof H224);
#if SPH_64
sc->count = 0;
#else
sc->count_high = sc->count_low = 0;
#endif
}
/* see sph_sha2.h */
void
sph_sha256_init(void *cc)
{
sph_sha256_context *sc;
sc = cc;
memcpy(sc->val, H256, sizeof H256);
#if SPH_64
sc->count = 0;
#else
sc->count_high = sc->count_low = 0;
#endif
}
#define RFUN sha2_round
#define HASH sha224
#define BE32 1
#include "md_helper.c"
/* see sph_sha2.h */
void
sph_sha224_close(void *cc, void *dst)
{
sha224_close(cc, dst, 7);
sph_sha224_init(cc);
}
/* see sph_sha2.h */
void
sph_sha224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha224_addbits_and_close(cc, ub, n, dst, 7);
sph_sha224_init(cc);
}
/* see sph_sha2.h */
void
sph_sha256_close(void *cc, void *dst)
{
sha224_close(cc, dst, 8);
sph_sha256_init(cc);
}
/* see sph_sha2.h */
void
sph_sha256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha224_addbits_and_close(cc, ub, n, dst, 8);
sph_sha256_init(cc);
}
/* see sph_sha2.h */
void
sph_sha224_comp(const sph_u32 msg[16], sph_u32 val[8])
{
#define SHA2_IN(x) msg[x]
SHA2_ROUND_BODY(SHA2_IN, val);
#undef SHA2_IN
}

370
algo/sha/sph_sha2.h Normal file
View File

@@ -0,0 +1,370 @@
/* $Id: sph_sha2.h 216 2010-06-08 09:46:57Z tp $ */
/**
* SHA-224, SHA-256, SHA-384 and SHA-512 interface.
*
* SHA-256 has been published in FIPS 180-2, now amended with a change
* notice to include SHA-224 as well (which is a simple variation on
* SHA-256). SHA-384 and SHA-512 are also defined in FIPS 180-2. FIPS
* standards can be found at:
* http://csrc.nist.gov/publications/fips/
*
* ==========================(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)=============================
*
* @file sph_sha2.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_SHA2_H__
#define SPH_SHA2_H__
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for SHA-224.
*/
#define SPH_SIZE_sha224 224
/**
* Output size (in bits) for SHA-256.
*/
#define SPH_SIZE_sha256 256
/**
* This structure is a context for SHA-224 computations: it contains the
* intermediate values and some data from the last entered block. Once
* a SHA-224 computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running SHA-224 computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
sph_u32 val[8];
#if SPH_64
sph_u64 count;
#else
sph_u32 count_high, count_low;
#endif
#endif
} sph_sha224_context;
/**
* This structure is a context for SHA-256 computations. It is identical
* to the SHA-224 context. However, a context is initialized for SHA-224
* <strong>or</strong> SHA-256, but not both (the internal IV is not the
* same).
*/
typedef sph_sha224_context sph_sha256_context;
/**
* Initialize a SHA-224 context. This process performs no memory allocation.
*
* @param cc the SHA-224 context (pointer to
* a <code>sph_sha224_context</code>)
*/
void sph_sha224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHA-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha224(void *cc, const void *data, size_t len);
/**
* Terminate the current SHA-224 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-224 context
* @param dst the destination buffer
*/
void sph_sha224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
/**
* Apply the SHA-224 compression function on the provided data. The
* <code>msg</code> parameter contains the 16 32-bit input blocks,
* as numerical values (hence after the big-endian decoding). The
* <code>val</code> parameter contains the 8 32-bit input blocks for
* the compression function; the output is written in place in this
* array.
*
* @param msg the message block (16 values)
* @param val the function 256-bit input and output
*/
void sph_sha224_comp(const sph_u32 msg[16], sph_u32 val[8]);
/**
* Initialize a SHA-256 context. This process performs no memory allocation.
*
* @param cc the SHA-256 context (pointer to
* a <code>sph_sha256_context</code>)
*/
void sph_sha256_init(void *cc);
#ifdef DOXYGEN_IGNORE
/**
* Process some data bytes, for SHA-256. This function is identical to
* <code>sha_224()</code>
*
* @param cc the SHA-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha256(void *cc, const void *data, size_t len);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha256 sph_sha224
#endif
/**
* Terminate the current SHA-256 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-256 context
* @param dst the destination buffer
*/
void sph_sha256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
#ifdef DOXYGEN_IGNORE
/**
* Apply the SHA-256 compression function on the provided data. This
* function is identical to <code>sha224_comp()</code>.
*
* @param msg the message block (16 values)
* @param val the function 256-bit input and output
*/
void sph_sha256_comp(const sph_u32 msg[16], sph_u32 val[8]);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha256_comp sph_sha224_comp
#endif
#if SPH_64
/**
* Output size (in bits) for SHA-384.
*/
#define SPH_SIZE_sha384 384
/**
* Output size (in bits) for SHA-512.
*/
#define SPH_SIZE_sha512 512
/**
* This structure is a context for SHA-384 computations: it contains the
* intermediate values and some data from the last entered block. Once
* a SHA-384 computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running SHA-384 computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
sph_u64 val[8];
sph_u64 count;
#endif
} sph_sha384_context;
/**
* Initialize a SHA-384 context. This process performs no memory allocation.
*
* @param cc the SHA-384 context (pointer to
* a <code>sph_sha384_context</code>)
*/
void sph_sha384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHA-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha384(void *cc, const void *data, size_t len);
/**
* Terminate the current SHA-384 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-384 context
* @param dst the destination buffer
*/
void sph_sha384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha384_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
/**
* Apply the SHA-384 compression function on the provided data. The
* <code>msg</code> parameter contains the 16 64-bit input blocks,
* as numerical values (hence after the big-endian decoding). The
* <code>val</code> parameter contains the 8 64-bit input blocks for
* the compression function; the output is written in place in this
* array.
*
* @param msg the message block (16 values)
* @param val the function 512-bit input and output
*/
void sph_sha384_comp(const sph_u64 msg[16], sph_u64 val[8]);
/**
* This structure is a context for SHA-512 computations. It is identical
* to the SHA-384 context. However, a context is initialized for SHA-384
* <strong>or</strong> SHA-512, but not both (the internal IV is not the
* same).
*/
typedef sph_sha384_context sph_sha512_context;
/**
* Initialize a SHA-512 context. This process performs no memory allocation.
*
* @param cc the SHA-512 context (pointer to
* a <code>sph_sha512_context</code>)
*/
void sph_sha512_init(void *cc);
#ifdef DOXYGEN_IGNORE
/**
* Process some data bytes, for SHA-512. This function is identical to
* <code>sph_sha384()</code>.
*
* @param cc the SHA-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_sha512(void *cc, const void *data, size_t len);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha512 sph_sha384
#endif
/**
* Terminate the current SHA-512 computation and output the result into the
* provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHA-512 context
* @param dst the destination buffer
*/
void sph_sha512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHA-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_sha512_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst);
#ifdef DOXYGEN_IGNORE
/**
* Apply the SHA-512 compression function. This function is identical to
* <code>sph_sha384_comp()</code>.
*
* @param msg the message block (16 values)
* @param val the function 512-bit input and output
*/
void sph_sha512_comp(const sph_u64 msg[16], sph_u64 val[8]);
#endif
#ifndef DOXYGEN_IGNORE
#define sph_sha512_comp sph_sha384_comp
#endif
#endif
#endif

247
algo/sha/sph_sha2big.c Normal file
View File

@@ -0,0 +1,247 @@
/* $Id: sha2big.c 216 2010-06-08 09:46:57Z tp $ */
/*
* SHA-384 / SHA-512 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_sha2.h"
#if SPH_64
#define CH(X, Y, Z) ((((Y) ^ (Z)) & (X)) ^ (Z))
#define MAJ(X, Y, Z) (((X) & (Y)) | (((X) | (Y)) & (Z)))
#define ROTR64 SPH_ROTR64
#define BSG5_0(x) (ROTR64(x, 28) ^ ROTR64(x, 34) ^ ROTR64(x, 39))
#define BSG5_1(x) (ROTR64(x, 14) ^ ROTR64(x, 18) ^ ROTR64(x, 41))
#define SSG5_0(x) (ROTR64(x, 1) ^ ROTR64(x, 8) ^ SPH_T64((x) >> 7))
#define SSG5_1(x) (ROTR64(x, 19) ^ ROTR64(x, 61) ^ SPH_T64((x) >> 6))
static const sph_u64 K512[80] = {
SPH_C64(0x428A2F98D728AE22), SPH_C64(0x7137449123EF65CD),
SPH_C64(0xB5C0FBCFEC4D3B2F), SPH_C64(0xE9B5DBA58189DBBC),
SPH_C64(0x3956C25BF348B538), SPH_C64(0x59F111F1B605D019),
SPH_C64(0x923F82A4AF194F9B), SPH_C64(0xAB1C5ED5DA6D8118),
SPH_C64(0xD807AA98A3030242), SPH_C64(0x12835B0145706FBE),
SPH_C64(0x243185BE4EE4B28C), SPH_C64(0x550C7DC3D5FFB4E2),
SPH_C64(0x72BE5D74F27B896F), SPH_C64(0x80DEB1FE3B1696B1),
SPH_C64(0x9BDC06A725C71235), SPH_C64(0xC19BF174CF692694),
SPH_C64(0xE49B69C19EF14AD2), SPH_C64(0xEFBE4786384F25E3),
SPH_C64(0x0FC19DC68B8CD5B5), SPH_C64(0x240CA1CC77AC9C65),
SPH_C64(0x2DE92C6F592B0275), SPH_C64(0x4A7484AA6EA6E483),
SPH_C64(0x5CB0A9DCBD41FBD4), SPH_C64(0x76F988DA831153B5),
SPH_C64(0x983E5152EE66DFAB), SPH_C64(0xA831C66D2DB43210),
SPH_C64(0xB00327C898FB213F), SPH_C64(0xBF597FC7BEEF0EE4),
SPH_C64(0xC6E00BF33DA88FC2), SPH_C64(0xD5A79147930AA725),
SPH_C64(0x06CA6351E003826F), SPH_C64(0x142929670A0E6E70),
SPH_C64(0x27B70A8546D22FFC), SPH_C64(0x2E1B21385C26C926),
SPH_C64(0x4D2C6DFC5AC42AED), SPH_C64(0x53380D139D95B3DF),
SPH_C64(0x650A73548BAF63DE), SPH_C64(0x766A0ABB3C77B2A8),
SPH_C64(0x81C2C92E47EDAEE6), SPH_C64(0x92722C851482353B),
SPH_C64(0xA2BFE8A14CF10364), SPH_C64(0xA81A664BBC423001),
SPH_C64(0xC24B8B70D0F89791), SPH_C64(0xC76C51A30654BE30),
SPH_C64(0xD192E819D6EF5218), SPH_C64(0xD69906245565A910),
SPH_C64(0xF40E35855771202A), SPH_C64(0x106AA07032BBD1B8),
SPH_C64(0x19A4C116B8D2D0C8), SPH_C64(0x1E376C085141AB53),
SPH_C64(0x2748774CDF8EEB99), SPH_C64(0x34B0BCB5E19B48A8),
SPH_C64(0x391C0CB3C5C95A63), SPH_C64(0x4ED8AA4AE3418ACB),
SPH_C64(0x5B9CCA4F7763E373), SPH_C64(0x682E6FF3D6B2B8A3),
SPH_C64(0x748F82EE5DEFB2FC), SPH_C64(0x78A5636F43172F60),
SPH_C64(0x84C87814A1F0AB72), SPH_C64(0x8CC702081A6439EC),
SPH_C64(0x90BEFFFA23631E28), SPH_C64(0xA4506CEBDE82BDE9),
SPH_C64(0xBEF9A3F7B2C67915), SPH_C64(0xC67178F2E372532B),
SPH_C64(0xCA273ECEEA26619C), SPH_C64(0xD186B8C721C0C207),
SPH_C64(0xEADA7DD6CDE0EB1E), SPH_C64(0xF57D4F7FEE6ED178),
SPH_C64(0x06F067AA72176FBA), SPH_C64(0x0A637DC5A2C898A6),
SPH_C64(0x113F9804BEF90DAE), SPH_C64(0x1B710B35131C471B),
SPH_C64(0x28DB77F523047D84), SPH_C64(0x32CAAB7B40C72493),
SPH_C64(0x3C9EBE0A15C9BEBC), SPH_C64(0x431D67C49C100D4C),
SPH_C64(0x4CC5D4BECB3E42B6), SPH_C64(0x597F299CFC657E2A),
SPH_C64(0x5FCB6FAB3AD6FAEC), SPH_C64(0x6C44198C4A475817)
};
static const sph_u64 H384[8] = {
SPH_C64(0xCBBB9D5DC1059ED8), SPH_C64(0x629A292A367CD507),
SPH_C64(0x9159015A3070DD17), SPH_C64(0x152FECD8F70E5939),
SPH_C64(0x67332667FFC00B31), SPH_C64(0x8EB44A8768581511),
SPH_C64(0xDB0C2E0D64F98FA7), SPH_C64(0x47B5481DBEFA4FA4)
};
static const sph_u64 H512[8] = {
SPH_C64(0x6A09E667F3BCC908), SPH_C64(0xBB67AE8584CAA73B),
SPH_C64(0x3C6EF372FE94F82B), SPH_C64(0xA54FF53A5F1D36F1),
SPH_C64(0x510E527FADE682D1), SPH_C64(0x9B05688C2B3E6C1F),
SPH_C64(0x1F83D9ABFB41BD6B), SPH_C64(0x5BE0CD19137E2179)
};
/*
* This macro defines the body for a SHA-384 / SHA-512 compression function
* implementation. The "in" parameter should evaluate, when applied to a
* numerical input parameter from 0 to 15, to an expression which yields
* the corresponding input block. The "r" parameter should evaluate to
* an array or pointer expression designating the array of 8 words which
* contains the input and output of the compression function.
*
* SHA-512 is hard for the compiler. If the loop is completely unrolled,
* then the code will be quite huge (possibly more than 100 kB), and the
* performance will be degraded due to cache misses on the code. We
* unroll only eight steps, which avoids all needless copies when
* 64-bit registers are swapped.
*/
#define SHA3_STEP(A, B, C, D, E, F, G, H, i) do { \
sph_u64 T1, T2; \
T1 = SPH_T64(H + BSG5_1(E) + CH(E, F, G) + K512[i] + W[i]); \
T2 = SPH_T64(BSG5_0(A) + MAJ(A, B, C)); \
D = SPH_T64(D + T1); \
H = SPH_T64(T1 + T2); \
} while (0)
#define SHA3_ROUND_BODY(in, r) do { \
int i; \
sph_u64 A, B, C, D, E, F, G, H; \
sph_u64 W[80]; \
\
for (i = 0; i < 16; i ++) \
W[i] = in(i); \
for (i = 16; i < 80; i ++) \
W[i] = SPH_T64(SSG5_1(W[i - 2]) + W[i - 7] \
+ SSG5_0(W[i - 15]) + W[i - 16]); \
A = (r)[0]; \
B = (r)[1]; \
C = (r)[2]; \
D = (r)[3]; \
E = (r)[4]; \
F = (r)[5]; \
G = (r)[6]; \
H = (r)[7]; \
for (i = 0; i < 80; i += 8) { \
SHA3_STEP(A, B, C, D, E, F, G, H, i + 0); \
SHA3_STEP(H, A, B, C, D, E, F, G, i + 1); \
SHA3_STEP(G, H, A, B, C, D, E, F, i + 2); \
SHA3_STEP(F, G, H, A, B, C, D, E, i + 3); \
SHA3_STEP(E, F, G, H, A, B, C, D, i + 4); \
SHA3_STEP(D, E, F, G, H, A, B, C, i + 5); \
SHA3_STEP(C, D, E, F, G, H, A, B, i + 6); \
SHA3_STEP(B, C, D, E, F, G, H, A, i + 7); \
} \
(r)[0] = SPH_T64((r)[0] + A); \
(r)[1] = SPH_T64((r)[1] + B); \
(r)[2] = SPH_T64((r)[2] + C); \
(r)[3] = SPH_T64((r)[3] + D); \
(r)[4] = SPH_T64((r)[4] + E); \
(r)[5] = SPH_T64((r)[5] + F); \
(r)[6] = SPH_T64((r)[6] + G); \
(r)[7] = SPH_T64((r)[7] + H); \
} while (0)
/*
* One round of SHA-384 / SHA-512. The data must be aligned for 64-bit access.
*/
static void
sha3_round(const unsigned char *data, sph_u64 r[8])
{
#define SHA3_IN(x) sph_dec64be_aligned(data + (8 * (x)))
SHA3_ROUND_BODY(SHA3_IN, r);
#undef SHA3_IN
}
/* see sph_sha3.h */
void
sph_sha384_init(void *cc)
{
sph_sha384_context *sc;
sc = cc;
memcpy(sc->val, H384, sizeof H384);
sc->count = 0;
}
/* see sph_sha3.h */
void
sph_sha512_init(void *cc)
{
sph_sha512_context *sc;
sc = cc;
memcpy(sc->val, H512, sizeof H512);
sc->count = 0;
}
#define RFUN sha3_round
#define HASH sha384
#define BE64 1
#include "md_helper.c"
/* see sph_sha3.h */
void
sph_sha384_close(void *cc, void *dst)
{
sha384_close(cc, dst, 6);
// sph_sha384_init(cc);
}
/* see sph_sha3.h */
void
sph_sha384_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha384_addbits_and_close(cc, ub, n, dst, 6);
// sph_sha384_init(cc);
}
/* see sph_sha3.h */
void
sph_sha512_close(void *cc, void *dst)
{
sha384_close(cc, dst, 8);
// sph_sha512_init(cc);
}
/* see sph_sha3.h */
void
sph_sha512_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
sha384_addbits_and_close(cc, ub, n, dst, 8);
// sph_sha512_init(cc);
}
/* see sph_sha3.h */
void
sph_sha384_comp(const sph_u64 msg[16], sph_u64 val[8])
{
#define SHA3_IN(x) msg[x]
SHA3_ROUND_BODY(SHA3_IN, val);
#undef SHA3_IN
}
#endif

1976
algo/sha/sph_types.h Normal file
View File

File diff suppressed because it is too large Load Diff