SM4

分组/密钥:128 位 / 128 位,32 轮(不平衡 Feistel,SM4 风格)

概述

  • 结构:基于 32 位字的不平衡 Feistel
  • 字节序:标准大端(big-endian),16 字节装成 4 个 32 位字
  • 解密:与加密相同,轮密钥倒序使用即可

轮函数

轮函数定义为

T(x)=L(τ(x)) T(x) = L(\tau(x))
  • τ\tau:对 32 位字的 4 个字节逐字节过 888\to 8 的 S 盒
  • LL:线性扩散
L(x)=x(x2)(x10)(x18)(x24) L(x) = x \oplus (x\lll2) \oplus (x\lll10) \oplus (x\lll18) \oplus (x\lll24)

密钥扩展(Key Schedule)

  • 同一个 S 盒,线性层换成 LL'
L(x)=x(x13)(x23) L'(x) = x \oplus (x\lll13) \oplus (x\lll23)
  • FK[4]FK[4]CK[32]CK[32] 生成 32 个轮密钥 rk[031]\mathrm{rk}[0\ldots31]

加/解密与密钥扩展(Python 参考实现)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
# SM4 reference implementation (single-block)

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

FK = [0xa3b1bac6, 0x56aa3350, 0x677d9197, 0xb27022dc]
CK = [
0x00070e15,0x1c232a31,0x383f464d,0x545b6269,
0x70777e85,0x8c939aa1,0xa8afb6bd,0xc4cbd2d9,
0xe0e7eef5,0xfc030a11,0x181f262d,0x343b4249,
0x50575e65,0x6c737a81,0x888f969d,0xa4abb2b9,
0xc0c7ced5,0xdce3eaf1,0xf8ff060d,0x141b2229,
0x30373e45,0x4c535a61,0x686f767d,0x848b9299,
0xa0a7aeb5,0xbcc3cad1,0xd8dfe6ed,0xf4fb0209,
0x10171e25,0x2c333a41,0x484f565d,0x646b7279,
]

def rotl32(x, r): return ((x << r) | (x >> (32 - r))) & 0xffffffff

def tau(x):
b0 = SBOX[(x >> 24) & 0xff]
b1 = SBOX[(x >> 16) & 0xff]
b2 = SBOX[(x >> 8) & 0xff]
b3 = SBOX[(x >> 0) & 0xff]
return ((b0 << 24) | (b1 << 16) | (b2 << 8) | b3) & 0xffffffff

def L(x): return x ^ rotl32(x,2) ^ rotl32(x,10) ^ rotl32(x,18) ^ rotl32(x,24)
def Lp(x): return x ^ rotl32(x,13) ^ rotl32(x,23) # L'

def T(x): return L(tau(x))
def Tp(x): return Lp(tau(x)) # T'

def bytes_to_u32_be(b, i): return int.from_bytes(b[i:i+4], 'big')
def u32_to_bytes_be(x): return x.to_bytes(4, 'big')

def key_schedule(key16: bytes):
MK0, MK1, MK2, MK3 = [bytes_to_u32_be(key16, i) for i in (0,4,8,12)]
K0, K1, K2, K3 = MK0 ^ FK[0], MK1 ^ FK[1], MK2 ^ FK[2], MK3 ^ FK[3]
K = [K0, K1, K2, K3]
rk = []
for i in range(32):
Ki4 = K[i] ^ Tp(K[i+1] ^ K[i+2] ^ K[i+3] ^ CK[i])
K.append(Ki4)
rk.append(Ki4)
return rk

def sm4_encrypt_block(plain16: bytes, key16: bytes) -> bytes:
rk = key_schedule(key16)
X = [bytes_to_u32_be(plain16, i) for i in (0,4,8,12)]
# expand to X0..X35
for i in range(32):
Xi4 = X[i] ^ T(X[i+1] ^ X[i+2] ^ X[i+3] ^ rk[i])
X.append(Xi4)
# output is X35,X34,X33,X32
out = b''.join(u32_to_bytes_be(X[35 - i]) for i in range(4))
return out

def sm4_decrypt_block(cipher16: bytes, key16: bytes) -> bytes:
rk = key_schedule(key16)[::-1] # reverse order
X = [bytes_to_u32_be(cipher16, i) for i in (0,4,8,12)]
for i in range(32):
Xi4 = X[i] ^ T(X[i+1] ^ X[i+2] ^ X[i+3] ^ rk[i])
X.append(Xi4)
out = b''.join(u32_to_bytes_be(X[35 - i]) for i in range(4))
return out

# --- self-test (KAT) ---
if __name__ == "__main__":
key = bytes.fromhex("0123456789abcdeffedcba9876543210")
pt = bytes.fromhex("0123456789abcdeffedcba9876543210")
ct_expect = bytes.fromhex("681edf34d206965e86b3e94f536e4246")
ct = sm4_encrypt_block(pt, key)
assert ct == ct_expect, (ct.hex(), ct_expect.hex())
assert sm4_decrypt_block(ct, key) == pt
print("SM4 KAT ok:", ct.hex())

细节说明

  1. 每一轮先用 S 盒提供非线性(混淆),再用多次循环左移 + 异或提供线性扩散(扩散)。

  2. 轮函数 TT 分为两步 τ\tau(S 盒)+ LL(扩散)。

    • τ\tau:逐字节过 S 盒(对 32 位字 x=b0b1b2b3x=b_0\|b_1\|b_2\|b_3,把每个字节丢进 8→8 的 S 盒得到 τ(x)\tau(x))。
    • LL:把一个 32 位字做若干循环左移后异或: L(y)=y(y2)(y10)(y18)(y24). L(y) = y \oplus (y\lll2) \oplus (y\lll10) \oplus (y\lll18) \oplus (y\lll24).
  3. 一轮流程重述:

    • step1: s=X[i+1]X[i+2]X[i+3]rk[i]s = X[i+1] \oplus X[i+2] \oplus X[i+3] \oplus rk[i]
    • step2: t=T(s)t = T(s)(即先 τ\tauLL
    • step3: X[i+4]=X[i]tX[i+4] = X[i] \oplus t
      32 轮后输出按反序拼接为 C=X[35]X[34]X[33]X[32]C = X[35]\|X[34]\|X[33]\|X[32]
  4. 轮密钥来源:主密钥按大端拆成 MK0..MK3,与 FK[0..3] 异或得到 K0..K3,随后迭代生成 K4..K35(每步用 TT' 与 CK[i]),并把 Ki+4K_{i+4} 作为 rk[i]rk[i]

设计目的与安全考量

  • S-盒(τ\tau:提供非线性,阻断简单线性/差分传播。
  • LLLL'(多次循环左移 + 异或):提供快速扩散,字内和字间影响传播。
  • Feistel 架构:保证可逆性并允许复杂轮函数。
  • CK / FK:破坏对称性,降低相关密钥攻击风险。
  • 32 轮:通过多轮累积将局部非线性与简单扩散叠加为全局混合。