Crypto++ 8.7
Free C++ class library of cryptographic schemes
sm4.cpp
1// sm4.cpp - written and placed in the public domain by Jeffrey Walton and Han Lulu
2//
3// We understand future ARMv8 enhancements are supposed
4// to include SM3 and SM4 related instructions so the function
5// is stubbed for an eventual SM4_Round_ARMV8.
6
7#include "pch.h"
8#include "config.h"
9
10#include "sm4.h"
11#include "misc.h"
12#include "cpu.h"
13
14#if CRYPTOPP_MSC_VERSION
15# pragma warning(disable: 4307)
16#endif
17
18ANONYMOUS_NAMESPACE_BEGIN
19
20using CryptoPP::byte;
23
24CRYPTOPP_ALIGN_DATA(4)
25const byte S[256] =
26{
27 0xD6, 0x90, 0xE9, 0xFE, 0xCC, 0xE1, 0x3D, 0xB7, 0x16, 0xB6, 0x14, 0xC2, 0x28, 0xFB, 0x2C, 0x05,
28 0x2B, 0x67, 0x9A, 0x76, 0x2A, 0xBE, 0x04, 0xC3, 0xAA, 0x44, 0x13, 0x26, 0x49, 0x86, 0x06, 0x99,
29 0x9C, 0x42, 0x50, 0xF4, 0x91, 0xEF, 0x98, 0x7A, 0x33, 0x54, 0x0B, 0x43, 0xED, 0xCF, 0xAC, 0x62,
30 0xE4, 0xB3, 0x1C, 0xA9, 0xC9, 0x08, 0xE8, 0x95, 0x80, 0xDF, 0x94, 0xFA, 0x75, 0x8F, 0x3F, 0xA6,
31 0x47, 0x07, 0xA7, 0xFC, 0xF3, 0x73, 0x17, 0xBA, 0x83, 0x59, 0x3C, 0x19, 0xE6, 0x85, 0x4F, 0xA8,
32 0x68, 0x6B, 0x81, 0xB2, 0x71, 0x64, 0xDA, 0x8B, 0xF8, 0xEB, 0x0F, 0x4B, 0x70, 0x56, 0x9D, 0x35,
33 0x1E, 0x24, 0x0E, 0x5E, 0x63, 0x58, 0xD1, 0xA2, 0x25, 0x22, 0x7C, 0x3B, 0x01, 0x21, 0x78, 0x87,
34 0xD4, 0x00, 0x46, 0x57, 0x9F, 0xD3, 0x27, 0x52, 0x4C, 0x36, 0x02, 0xE7, 0xA0, 0xC4, 0xC8, 0x9E,
35 0xEA, 0xBF, 0x8A, 0xD2, 0x40, 0xC7, 0x38, 0xB5, 0xA3, 0xF7, 0xF2, 0xCE, 0xF9, 0x61, 0x15, 0xA1,
36 0xE0, 0xAE, 0x5D, 0xA4, 0x9B, 0x34, 0x1A, 0x55, 0xAD, 0x93, 0x32, 0x30, 0xF5, 0x8C, 0xB1, 0xE3,
37 0x1D, 0xF6, 0xE2, 0x2E, 0x82, 0x66, 0xCA, 0x60, 0xC0, 0x29, 0x23, 0xAB, 0x0D, 0x53, 0x4E, 0x6F,
38 0xD5, 0xDB, 0x37, 0x45, 0xDE, 0xFD, 0x8E, 0x2F, 0x03, 0xFF, 0x6A, 0x72, 0x6D, 0x6C, 0x5B, 0x51,
39 0x8D, 0x1B, 0xAF, 0x92, 0xBB, 0xDD, 0xBC, 0x7F, 0x11, 0xD9, 0x5C, 0x41, 0x1F, 0x10, 0x5A, 0xD8,
40 0x0A, 0xC1, 0x31, 0x88, 0xA5, 0xCD, 0x7B, 0xBD, 0x2D, 0x74, 0xD0, 0x12, 0xB8, 0xE5, 0xB4, 0xB0,
41 0x89, 0x69, 0x97, 0x4A, 0x0C, 0x96, 0x77, 0x7E, 0x65, 0xB9, 0xF1, 0x09, 0xC5, 0x6E, 0xC6, 0x84,
42 0x18, 0xF0, 0x7D, 0xEC, 0x3A, 0xDC, 0x4D, 0x20, 0x79, 0xEE, 0x5F, 0x3E, 0xD7, 0xCB, 0x39, 0x48
43};
44
45const word32 CK[32] =
46{
47 0x00070E15, 0x1C232A31, 0x383F464D, 0x545B6269,
48 0x70777E85, 0x8C939AA1, 0xA8AFB6BD, 0xC4CBD2D9,
49 0xE0E7EEF5, 0xFC030A11, 0x181F262D, 0x343B4249,
50 0x50575E65, 0x6C737A81, 0x888F969D, 0xA4ABB2B9,
51 0xC0C7CED5, 0xDCE3EAF1, 0xF8FF060D, 0x141B2229,
52 0x30373E45, 0x4C535A61, 0x686F767D, 0x848B9299,
53 0xA0A7AEB5, 0xBCC3CAD1, 0xD8DFE6ED, 0xF4FB0209,
54 0x10171E25, 0x2C333A41, 0x484F565D, 0x646B7279
55};
56
57inline word32 SM4_H(word32 x)
58{
59 return (S[GETBYTE(x, 3)] << 24) | (S[GETBYTE(x, 2)] << 16) | (S[GETBYTE(x, 1)] << 8) | (S[GETBYTE(x, 0)]);
60}
61
62inline word32 SM4_G(word32 x)
63{
64 const word32 t = SM4_H(x);
65 return t ^ rotlConstant<13>(t) ^ rotlConstant<23>(t);
66}
67
68inline word32 SM4_F(word32 x)
69{
70 const word32 t = SM4_H(x);
71 return t ^ rotlConstant<2>(t) ^ rotlConstant<10>(t) ^ rotlConstant<18>(t) ^ rotlConstant<24>(t);
72}
73
74template <unsigned int R, bool FWD>
75inline void SM4_Round(word32 wspace[4], const word32 rkeys[32])
76{
77 if (FWD)
78 {
79 wspace[0] ^= SM4_F(wspace[1] ^ wspace[2] ^ wspace[3] ^ rkeys[R+0]);
80 wspace[1] ^= SM4_F(wspace[0] ^ wspace[2] ^ wspace[3] ^ rkeys[R+1]);
81 wspace[2] ^= SM4_F(wspace[0] ^ wspace[1] ^ wspace[3] ^ rkeys[R+2]);
82 wspace[3] ^= SM4_F(wspace[0] ^ wspace[1] ^ wspace[2] ^ rkeys[R+3]);
83 }
84 else
85 {
86 wspace[0] ^= SM4_F(wspace[1] ^ wspace[2] ^ wspace[3] ^ rkeys[R-0]);
87 wspace[1] ^= SM4_F(wspace[0] ^ wspace[2] ^ wspace[3] ^ rkeys[R-1]);
88 wspace[2] ^= SM4_F(wspace[0] ^ wspace[1] ^ wspace[3] ^ rkeys[R-2]);
89 wspace[3] ^= SM4_F(wspace[0] ^ wspace[1] ^ wspace[2] ^ rkeys[R-3]);
90 }
91}
92
93ANONYMOUS_NAMESPACE_END
94
95NAMESPACE_BEGIN(CryptoPP)
96
97#if CRYPTOPP_SM4_ADVANCED_PROCESS_BLOCKS
98# if defined(CRYPTOPP_AESNI_AVAILABLE)
99extern size_t SM4_Enc_AdvancedProcessBlocks_AESNI(const word32* subKeys, size_t rounds,
100 const byte *inBlocks, const byte *xorBlocks, byte *outBlocks, size_t length, word32 flags);
101# endif
102#endif
103
104std::string SM4::Enc::AlgorithmProvider() const
105{
106#if defined(CRYPTOPP_AESNI_AVAILABLE)
107 if (HasAESNI())
108 return "AESNI";
109#endif
110 return "C++";
111}
112
113void SM4::Base::UncheckedSetKey(const byte *userKey, unsigned int keyLength, const NameValuePairs &params)
114{
115 CRYPTOPP_ASSERT(keyLength == 16);
116 CRYPTOPP_UNUSED(params);
117
118 m_rkeys.New(32);
119 m_wspace.New(5);
120
121 GetUserKey(BIG_ENDIAN_ORDER, m_wspace.begin(), 4, userKey, keyLength);
122 m_wspace[0] ^= 0xa3b1bac6; m_wspace[1] ^= 0x56aa3350;
123 m_wspace[2] ^= 0x677d9197; m_wspace[3] ^= 0xb27022dc;
124
125 size_t i=0;
126 do
127 {
128 m_rkeys[i] = (m_wspace[0] ^= SM4_G(m_wspace[1] ^ m_wspace[2] ^ m_wspace[3] ^ CK[i])); i++;
129 m_rkeys[i] = (m_wspace[1] ^= SM4_G(m_wspace[2] ^ m_wspace[3] ^ m_wspace[0] ^ CK[i])); i++;
130 m_rkeys[i] = (m_wspace[2] ^= SM4_G(m_wspace[3] ^ m_wspace[0] ^ m_wspace[1] ^ CK[i])); i++;
131 m_rkeys[i] = (m_wspace[3] ^= SM4_G(m_wspace[0] ^ m_wspace[1] ^ m_wspace[2] ^ CK[i])); i++;
132 }
133 while (i < 32);
134}
135
136void SM4::Enc::ProcessAndXorBlock(const byte *inBlock, const byte *xorBlock, byte *outBlock) const
137{
138 // Reverse bytes on LittleEndian; align pointer on BigEndian
140 InBlock iblk(inBlock); iblk(m_wspace[0])(m_wspace[1])(m_wspace[2])(m_wspace[3]);
141
142 // Timing attack countermeasure, see comments in Rijndael for more details.
143 // The hardening does not materially affect benchmarks. SM4 runs at
144 // 30.5 cpb on a Core i5 Skylake with and without the code below.
145 const int cacheLineSize = GetCacheLineSize();
146 volatile word32 _u = 0;
147 word32 u = _u;
148
149 for (unsigned int i=0; i<sizeof(S); i+=cacheLineSize)
150 u |= *(const word32 *)(void*)(S+i);
151 m_wspace[4] = u;
152
153 SM4_Round< 0, true>(m_wspace, m_rkeys);
154 SM4_Round< 4, true>(m_wspace, m_rkeys);
155 SM4_Round< 8, true>(m_wspace, m_rkeys);
156 SM4_Round<12, true>(m_wspace, m_rkeys);
157 SM4_Round<16, true>(m_wspace, m_rkeys);
158 SM4_Round<20, true>(m_wspace, m_rkeys);
159 SM4_Round<24, true>(m_wspace, m_rkeys);
160 SM4_Round<28, true>(m_wspace, m_rkeys);
161
162 // Reverse bytes on LittleEndian; align pointer on BigEndian
163 typedef PutBlock<word32, BigEndian, false> OutBlock;
164 OutBlock oblk(xorBlock, outBlock); oblk(m_wspace[3])(m_wspace[2])(m_wspace[1])(m_wspace[0]);
165}
166
167void SM4::Dec::ProcessAndXorBlock(const byte *inBlock, const byte *xorBlock, byte *outBlock) const
168{
169 // Reverse bytes on LittleEndian; align pointer on BigEndian
171 InBlock iblk(inBlock); iblk(m_wspace[0])(m_wspace[1])(m_wspace[2])(m_wspace[3]);
172
173 // Timing attack countermeasure, see comments in Rijndael for more details.
174 // The hardening does not materially affect benchmarks. SM4 runs at
175 // 30.5 cpb on a Core i5 Skylake with and without the code below.
176 const int cacheLineSize = GetCacheLineSize();
177 volatile word32 _u = 0;
178 word32 u = _u;
179
180 for (unsigned int i=0; i<sizeof(S); i+=cacheLineSize)
181 u |= *(const word32 *)(void*)(S+i);
182 m_wspace[4] = u;
183
184 SM4_Round<31, false>(m_wspace, m_rkeys);
185 SM4_Round<27, false>(m_wspace, m_rkeys);
186 SM4_Round<23, false>(m_wspace, m_rkeys);
187 SM4_Round<19, false>(m_wspace, m_rkeys);
188 SM4_Round<15, false>(m_wspace, m_rkeys);
189 SM4_Round<11, false>(m_wspace, m_rkeys);
190 SM4_Round< 7, false>(m_wspace, m_rkeys);
191 SM4_Round< 3, false>(m_wspace, m_rkeys);
192
193 // Reverse bytes on LittleEndian; align pointer on BigEndian
194 typedef PutBlock<word32, BigEndian, false> OutBlock;
195 OutBlock oblk(xorBlock, outBlock); oblk(m_wspace[3])(m_wspace[2])(m_wspace[1])(m_wspace[0]);
196}
197
198#if CRYPTOPP_SM4_ADVANCED_PROCESS_BLOCKS
199size_t SM4::Enc::AdvancedProcessBlocks(const byte *inBlocks, const byte *xorBlocks,
200 byte *outBlocks, size_t length, word32 flags) const
201{
202#if defined(CRYPTOPP_AESNI_AVAILABLE)
203 if (HasAESNI()) {
204 return SM4_Enc_AdvancedProcessBlocks_AESNI(m_rkeys, 32,
205 inBlocks, xorBlocks, outBlocks, length, flags);
206 }
207#endif
208 return BlockTransformation::AdvancedProcessBlocks(inBlocks, xorBlocks, outBlocks, length, flags);
209}
210#endif // CRYPTOPP_SM4_ADVANCED_PROCESS_BLOCKS
211
212NAMESPACE_END
virtual size_t AdvancedProcessBlocks(const byte *inBlocks, const byte *xorBlocks, byte *outBlocks, size_t length, word32 flags) const
Encrypt and xor multiple blocks using additional flags.
Access a block of memory.
Definition: misc.h:2766
Interface for retrieving values given their names.
Definition: cryptlib.h:322
Access a block of memory.
Definition: misc.h:2807
Library configuration file.
unsigned char byte
8-bit unsigned datatype
Definition: config_int.h:56
unsigned int word32
32-bit unsigned datatype
Definition: config_int.h:62
Functions for CPU features and intrinsics.
@ BIG_ENDIAN_ORDER
byte order is big-endian
Definition: cryptlib.h:147
Utility functions for the Crypto++ library.
T rotlConstant(T x)
Performs a left rotate.
Definition: misc.h:1548
void GetUserKey(ByteOrder order, T *out, size_t outlen, const byte *in, size_t inlen)
Copy bytes in a buffer to an array of elements in big-endian order.
Definition: misc.h:2291
Crypto++ library namespace.
Precompiled header file.
Classes for the SM4 block cipher.
#define CRYPTOPP_ASSERT(exp)
Debugging and diagnostic assertion.
Definition: trap.h:68