crypto - Add Serpent implementation
authorAlex Hornung <ahornung@gmail.com>
Sat, 2 Jul 2011 12:07:01 +0000 (13:07 +0100)
committerAlex Hornung <ahornung@gmail.com>
Sat, 2 Jul 2011 12:24:53 +0000 (13:24 +0100)
sys/conf/files
sys/crypto/Makefile
sys/crypto/serpent/serpent.c [new file with mode: 0644]
sys/crypto/serpent/serpent.h [new file with mode: 0644]

index 983cb3c..f3563b7 100644 (file)
@@ -1910,6 +1910,7 @@ crypto/rijndael/rijndael-api.c            optional crypto
 crypto/sha1.c                          optional crypto
 crypto/sha2/sha2.c                             optional crypto
 crypto/twofish/twofish.c               optional crypto
+crypto/serpent/serpent.c               optional crypto
 net/zlib.c                             optional crypto
 opencrypto/cast.c                      optional crypto
 opencrypto/criov.c                     optional crypto
index 8f3ac87..6d0b70b 100644 (file)
@@ -8,6 +8,7 @@
 .PATH: ${.CURDIR}/../crypto/sha2
 .PATH: ${.CURDIR}/../crypto/camellia
 .PATH: ${.CURDIR}/../crypto/twofish
+.PATH: ${.CURDIR}/../crypto/serpent
 
 KMOD   = crypto
 SRCS   = crypto.c cryptodev_if.c
@@ -20,6 +21,7 @@ SRCS  += opt_param.h cryptodev_if.h bus_if.h device_if.h
 SRCS   += opt_ddb.h
 SRCS   += camellia.c camellia-api.c
 SRCS   += twofish.c
+SRCS   += serpent.c
 
 SUBDIR = rc4
 
diff --git a/sys/crypto/serpent/serpent.c b/sys/crypto/serpent/serpent.c
new file mode 100644 (file)
index 0000000..efc65de
--- /dev/null
@@ -0,0 +1,631 @@
+
+// Copyright in this code is held by Dr B.R. Gladman but free direct or
+// derivative use is permitted subject to acknowledgement of its origin
+// and subject to any constraints placed on the use of the algorithm by
+// its designers (if such constraints may exist, this will be indicated
+// below).
+//
+// Dr. B. R. Gladman                               . 25th January 2000.
+//
+// This is an implementation of Serpent, an encryption algorithm designed
+// by Anderson, Biham and Knudsen and submitted as a candidate for the
+// Advanced Encryption Standard programme of the US National Institute of
+// Standards and Technology.
+//
+// The designers of Serpent have not placed any constraints on the use of
+// this algorithm.
+
+#include <sys/types.h>
+#include <sys/endian.h>
+
+#include <crypto/serpent/serpent.h>
+
+// Partially optimised Serpent S Box boolean functions derived
+// using a recursive descent analyser but without a full search
+// of all subtrees. This set of S boxes is the result of work
+// by Sam Simpson and Brian Gladman using the spare time on a
+// cluster of high capacity servers to search for S boxes with
+// this customised search engine.
+//
+// Copyright:   Dr B. R Gladman (gladman@seven77.demon.co.uk)
+//              and Sam Simpson (s.simpson@mia.co.uk)
+//              17th December 1998
+//
+// We hereby give permission for information in this file to be
+// used freely subject only to acknowledgement of its origin
+
+// 15 terms
+
+#define rotr(x,n)   (((x) >> ((int)((n) & 0x1f))) | ((x) << ((int)((32 - ((n) & 0x1f))))))
+#define rotl(x,n)   (((x) << ((int)((n) & 0x1f))) | ((x) >> ((int)((32 - ((n) & 0x1f))))))
+
+#define sb0(a,b,c,d,e,f,g,h)    \
+{   t1 = a ^ d;     \
+    t2 = a & d;     \
+    t3 = c ^ t1;    \
+    t6 = b & t1;    \
+    t4 = b ^ t3;    \
+    t10 = ~t3;      \
+    h = t2 ^ t4;    \
+    t7 = a ^ t6;    \
+    t14 = ~t7;      \
+    t8 = c | t7;    \
+    t11 = t3 ^ t7;  \
+    g = t4 ^ t8;    \
+    t12 = h & t11;  \
+    f = t10 ^ t12;  \
+    e = t12 ^ t14;  \
+}
+// 15 terms
+
+#define ib0(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = a ^ b;     \
+    t3 = t1 | t2;   \
+    t4 = d ^ t3;    \
+    t7 = d & t2;    \
+    t5 = c ^ t4;    \
+    t8 = t1 ^ t7;   \
+    g = t2 ^ t5;    \
+    t11 = a & t4;   \
+    t9 = g & t8;    \
+    t14 = t5 ^ t8;  \
+    f = t4 ^ t9;    \
+    t12 = t5 | f;   \
+    h = t11 ^ t12;  \
+    e = h ^ t14;    \
+}
+
+// 14 terms!
+
+#define sb1(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = b ^ t1;    \
+    t3 = a | t2;    \
+    t4 = d | t2;    \
+    t5 = c ^ t3;    \
+    g = d ^ t5;     \
+    t7 = b ^ t4;    \
+    t8 = t2 ^ g;    \
+    t9 = t5 & t7;   \
+    h = t8 ^ t9;    \
+    t11 = t5 ^ t7;  \
+    f = h ^ t11;    \
+    t13 = t8 & t11; \
+    e = t5 ^ t13;   \
+}
+
+// 17 terms
+
+#define ib1(a,b,c,d,e,f,g,h)    \
+{   t1 = a ^ d;     \
+    t2 = a & b;     \
+    t3 = b ^ c;     \
+    t4 = a ^ t3;    \
+    t5 = b | d;     \
+    t7 = c | t1;    \
+    h = t4 ^ t5;    \
+    t8 = b ^ t7;    \
+    t11 = ~t2;      \
+    t9 = t4 & t8;   \
+    f = t1 ^ t9;    \
+    t13 = t9 ^ t11; \
+    t12 = h & f;    \
+    g = t12 ^ t13;  \
+    t15 = a & d;    \
+    t16 = c ^ t13;  \
+    e = t15 ^ t16;  \
+}
+
+// 16 terms
+
+#define sb2(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = b ^ d;     \
+    t3 = c & t1;    \
+    t13 = d | t1;   \
+    e = t2 ^ t3;    \
+    t5 = c ^ t1;    \
+    t6 = c ^ e;     \
+    t7 = b & t6;    \
+    t10 = e | t5;   \
+    h = t5 ^ t7;    \
+    t9 = d | t7;    \
+    t11 = t9 & t10; \
+    t14 = t2 ^ h;   \
+    g = a ^ t11;    \
+    t15 = g ^ t13;  \
+    f = t14 ^ t15;  \
+}
+
+// 16 terms
+
+#define ib2(a,b,c,d,e,f,g,h)    \
+{   t1 = b ^ d;     \
+    t2 = ~t1;       \
+    t3 = a ^ c;     \
+    t4 = c ^ t1;    \
+    t7 = a | t2;    \
+    t5 = b & t4;    \
+    t8 = d ^ t7;    \
+    t11 = ~t4;      \
+    e = t3 ^ t5;    \
+    t9 = t3 | t8;   \
+    t14 = d & t11;  \
+    h = t1 ^ t9;    \
+    t12 = e | h;    \
+    f = t11 ^ t12;  \
+    t15 = t3 ^ t12; \
+    g = t14 ^ t15;  \
+}
+
+// 17 terms
+
+#define sb3(a,b,c,d,e,f,g,h)    \
+{   t1 = a ^ c;     \
+    t2 = d ^ t1;    \
+    t3 = a & t2;    \
+    t4 = d ^ t3;    \
+    t5 = b & t4;    \
+    g = t2 ^ t5;    \
+    t7 = a | g;     \
+    t8 = b | d;     \
+    t11 = a | d;    \
+    t9 = t4 & t7;   \
+    f = t8 ^ t9;    \
+    t12 = b ^ t11;  \
+    t13 = g ^ t9;   \
+    t15 = t3 ^ t8;  \
+    h = t12 ^ t13;  \
+    t16 = c & t15;  \
+    e = t12 ^ t16;  \
+}
+
+// 16 term solution that performs less well than 17 term one
+// in my environment (PPro/PII)
+
+#if(0)
+
+#define sb3(a,b,c,d,e,f,g,h)    \
+{   t1 = a ^ b;     \
+    t2 = a & c;     \
+    t3 = a | d;     \
+    t4 = c ^ d;     \
+    t5 = t1 & t3;   \
+    t6 = t2 | t5;   \
+    g = t4 ^ t6;    \
+    t8 = b ^ t3;    \
+    t9 = t6 ^ t8;   \
+    t10 = t4 & t9;  \
+    e = t1 ^ t10;   \
+    t12 = g & e;    \
+    f = t9 ^ t12;   \
+    t14 = b | d;    \
+    t15 = t4 ^ t12; \
+    h = t14 ^ t15;  \
+}
+
+#endif
+
+// 17 terms
+
+#define ib3(a,b,c,d,e,f,g,h)    \
+{   t1 = b ^ c;     \
+    t2 = b | c;     \
+    t3 = a ^ c;     \
+    t7 = a ^ d;     \
+    t4 = t2 ^ t3;   \
+    t5 = d | t4;    \
+    t9 = t2 ^ t7;   \
+    e = t1 ^ t5;    \
+    t8 = t1 | t5;   \
+    t11 = a & t4;   \
+    g = t8 ^ t9;    \
+    t12 = e | t9;   \
+    f = t11 ^ t12;  \
+    t14 = a & g;    \
+    t15 = t2 ^ t14; \
+    t16 = e & t15;  \
+    h = t4 ^ t16;   \
+}
+
+// 15 terms
+
+#define sb4(a,b,c,d,e,f,g,h)    \
+{   t1 = a ^ d;     \
+    t2 = d & t1;    \
+    t3 = c ^ t2;    \
+    t4 = b | t3;    \
+    h = t1 ^ t4;    \
+    t6 = ~b;        \
+    t7 = t1 | t6;   \
+    e = t3 ^ t7;    \
+    t9 = a & e;     \
+    t10 = t1 ^ t6;  \
+    t11 = t4 & t10; \
+    g = t9 ^ t11;   \
+    t13 = a ^ t3;   \
+    t14 = t10 & g;  \
+    f = t13 ^ t14;  \
+}
+
+// 17 terms
+
+#define ib4(a,b,c,d,e,f,g,h)    \
+{   t1 = c ^ d;     \
+    t2 = c | d;     \
+    t3 = b ^ t2;    \
+    t4 = a & t3;    \
+    f = t1 ^ t4;    \
+    t6 = a ^ d;     \
+    t7 = b | d;     \
+    t8 = t6 & t7;   \
+    h = t3 ^ t8;    \
+    t10 = ~a;       \
+    t11 = c ^ h;    \
+    t12 = t10 | t11;\
+    e = t3 ^ t12;   \
+    t14 = c | t4;   \
+    t15 = t7 ^ t14; \
+    t16 = h | t10;  \
+    g = t15 ^ t16;  \
+}
+
+// 16 terms
+
+#define sb5(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = a ^ b;     \
+    t3 = a ^ d;     \
+    t4 = c ^ t1;    \
+    t5 = t2 | t3;   \
+    e = t4 ^ t5;    \
+    t7 = d & e;     \
+    t8 = t2 ^ e;    \
+    t10 = t1 | e;   \
+    f = t7 ^ t8;    \
+    t11 = t2 | t7;  \
+    t12 = t3 ^ t10; \
+    t14 = b ^ t7;   \
+    g = t11 ^ t12;  \
+    t15 = f & t12;  \
+    h = t14 ^ t15;  \
+}
+
+// 16 terms
+
+#define ib5(a,b,c,d,e,f,g,h)    \
+{   t1 = ~c;        \
+    t2 = b & t1;    \
+    t3 = d ^ t2;    \
+    t4 = a & t3;    \
+    t5 = b ^ t1;    \
+    h = t4 ^ t5;    \
+    t7 = b | h;     \
+    t8 = a & t7;    \
+    f = t3 ^ t8;    \
+    t10 = a | d;    \
+    t11 = t1 ^ t7;  \
+    e = t10 ^ t11;  \
+    t13 = a ^ c;    \
+    t14 = b & t10;  \
+    t15 = t4 | t13; \
+    g = t14 ^ t15;  \
+}
+
+// 15 terms
+
+#define sb6(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = a ^ d;     \
+    t3 = b ^ t2;    \
+    t4 = t1 | t2;   \
+    t5 = c ^ t4;    \
+    f = b ^ t5;     \
+    t13 = ~t5;      \
+    t7 = t2 | f;    \
+    t8 = d ^ t7;    \
+    t9 = t5 & t8;   \
+    g = t3 ^ t9;    \
+    t11 = t5 ^ t8;  \
+    e = g ^ t11;    \
+    t14 = t3 & t11; \
+    h = t13 ^ t14;  \
+}
+
+// 15 terms
+
+#define ib6(a,b,c,d,e,f,g,h)    \
+{   t1 = ~a;        \
+    t2 = a ^ b;     \
+    t3 = c ^ t2;    \
+    t4 = c | t1;    \
+    t5 = d ^ t4;    \
+    t13 = d & t1;   \
+    f = t3 ^ t5;    \
+    t7 = t3 & t5;   \
+    t8 = t2 ^ t7;   \
+    t9 = b | t8;    \
+    h = t5 ^ t9;    \
+    t11 = b | h;    \
+    e = t8 ^ t11;   \
+    t14 = t3 ^ t11; \
+    g = t13 ^ t14;  \
+}
+
+// 17 terms
+
+#define sb7(a,b,c,d,e,f,g,h)    \
+{   t1 = ~c;        \
+    t2 = b ^ c;     \
+    t3 = b | t1;    \
+    t4 = d ^ t3;    \
+    t5 = a & t4;    \
+    t7 = a ^ d;     \
+    h = t2 ^ t5;    \
+    t8 = b ^ t5;    \
+    t9 = t2 | t8;   \
+    t11 = d & t3;   \
+    f = t7 ^ t9;    \
+    t12 = t5 ^ f;   \
+    t15 = t1 | t4;  \
+    t13 = h & t12;  \
+    g = t11 ^ t13;  \
+    t16 = t12 ^ g;  \
+    e = t15 ^ t16;  \
+}
+
+// 17 terms
+
+#define ib7(a,b,c,d,e,f,g,h)    \
+{   t1 = a & b;     \
+    t2 = a | b;     \
+    t3 = c | t1;    \
+    t4 = d & t2;    \
+    h = t3 ^ t4;    \
+    t6 = ~d;        \
+    t7 = b ^ t4;    \
+    t8 = h ^ t6;    \
+    t11 = c ^ t7;   \
+    t9 = t7 | t8;   \
+    f = a ^ t9;     \
+    t12 = d | f;    \
+    e = t11 ^ t12;  \
+    t14 = a & h;    \
+    t15 = t3 ^ f;   \
+    t16 = e ^ t14;  \
+    g = t15 ^ t16;  \
+}
+
+#define k_xor(r,a,b,c,d)             \
+{   a ^= ctx->l_key[4 * r +  8]; \
+    b ^= ctx->l_key[4 * r +  9]; \
+    c ^= ctx->l_key[4 * r + 10]; \
+    d ^= ctx->l_key[4 * r + 11]; \
+}
+
+#define k_set(r,a,b,c,d)             \
+{   a = ctx->l_key[4 * r +  8];  \
+    b = ctx->l_key[4 * r +  9];  \
+    c = ctx->l_key[4 * r + 10];  \
+    d = ctx->l_key[4 * r + 11];  \
+}
+
+#define k_get(r,a,b,c,d)    \
+{   ctx->l_key[4 * r +  8] = a;  \
+    ctx->l_key[4 * r +  9] = b;  \
+    ctx->l_key[4 * r + 10] = c;  \
+    ctx->l_key[4 * r + 11] = d;  \
+}
+
+// the linear transformation and its inverse
+
+#define rot(a,b,c,d)    \
+{   a = rotl(a, 13);    \
+    c = rotl(c, 3);     \
+    d ^= c ^ (a << 3);  \
+    b ^= a ^ c;         \
+    d = rotl(d, 7);     \
+    b = rotl(b, 1);     \
+    a ^= b ^ d;         \
+    c ^= d ^ (b << 7);  \
+    a = rotl(a, 5);     \
+    c = rotl(c, 22);    \
+}
+
+#define irot(a,b,c,d)   \
+{   c = rotr(c, 22);    \
+    a = rotr(a, 5);     \
+    c ^= d ^ (b << 7);  \
+    a ^= b ^ d;         \
+    d = rotr(d, 7);     \
+    b = rotr(b, 1);     \
+    d ^= c ^ (a << 3);  \
+    b ^= a ^ c;         \
+    c = rotr(c, 3);     \
+    a = rotr(a, 13);    \
+}
+
+// initialise the key schedule from the user supplied key
+
+void serpent_set_key(serpent_ctx *ctx, const u_int8_t in_key[], int key_len)
+{
+    u_int32_t  i,lk,a,b,c,d,e,f,g,h;
+    u_int32_t  t1,t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14,t15,t16;
+
+    if(key_len > 256)
+        return;
+
+    i = 0; lk = (key_len + 31) / 32;
+
+    while(i < lk)
+    {
+        ctx->l_key[i] = htole32(*((const u_int32_t *)(in_key + 4 * i))); i++;
+    }
+
+    if(key_len < 256)
+    {
+        while(i < 8)
+
+            ctx->l_key[i++] = 0;
+
+        i = key_len / 32; lk = 1 << key_len % 32;
+
+        ctx->l_key[i] = (ctx->l_key[i] & (lk - 1)) | lk;
+    }
+
+    t1 = ctx->l_key[2] ^ ctx->l_key[4] ^ ctx->l_key[6] ^ 0x9e3779b9;
+    t2 = ctx->l_key[3] ^ ctx->l_key[5] ^ ctx->l_key[7] ^ 0x9e3779b9;
+
+    for(i = 0; i < 132; i += 2)
+    {
+        ctx->l_key[i + 8] = rotr(i ^ ctx->l_key[i] ^ t2, 21);
+
+        t1 ^= ctx->l_key[i + 2] ^ ctx->l_key[i + 8];
+
+        ctx->l_key[i + 9] = rotr((i + 1) ^ ctx->l_key[i + 1] ^ t1, 21);
+
+        t2 ^= ctx->l_key[i + 3] ^ ctx->l_key[i + 9];
+    }
+
+    k_set( 0,a,b,c,d);sb3(a,b,c,d,e,f,g,h);k_get( 0,e,f,g,h);
+    k_set( 1,a,b,c,d);sb2(a,b,c,d,e,f,g,h);k_get( 1,e,f,g,h);
+    k_set( 2,a,b,c,d);sb1(a,b,c,d,e,f,g,h);k_get( 2,e,f,g,h);
+    k_set( 3,a,b,c,d);sb0(a,b,c,d,e,f,g,h);k_get( 3,e,f,g,h);
+    k_set( 4,a,b,c,d);sb7(a,b,c,d,e,f,g,h);k_get( 4,e,f,g,h);
+    k_set( 5,a,b,c,d);sb6(a,b,c,d,e,f,g,h);k_get( 5,e,f,g,h);
+    k_set( 6,a,b,c,d);sb5(a,b,c,d,e,f,g,h);k_get( 6,e,f,g,h);
+    k_set( 7,a,b,c,d);sb4(a,b,c,d,e,f,g,h);k_get( 7,e,f,g,h);
+    k_set( 8,a,b,c,d);sb3(a,b,c,d,e,f,g,h);k_get( 8,e,f,g,h);
+    k_set( 9,a,b,c,d);sb2(a,b,c,d,e,f,g,h);k_get( 9,e,f,g,h);
+    k_set(10,a,b,c,d);sb1(a,b,c,d,e,f,g,h);k_get(10,e,f,g,h);
+    k_set(11,a,b,c,d);sb0(a,b,c,d,e,f,g,h);k_get(11,e,f,g,h);
+    k_set(12,a,b,c,d);sb7(a,b,c,d,e,f,g,h);k_get(12,e,f,g,h);
+    k_set(13,a,b,c,d);sb6(a,b,c,d,e,f,g,h);k_get(13,e,f,g,h);
+    k_set(14,a,b,c,d);sb5(a,b,c,d,e,f,g,h);k_get(14,e,f,g,h);
+    k_set(15,a,b,c,d);sb4(a,b,c,d,e,f,g,h);k_get(15,e,f,g,h);
+    k_set(16,a,b,c,d);sb3(a,b,c,d,e,f,g,h);k_get(16,e,f,g,h);
+    k_set(17,a,b,c,d);sb2(a,b,c,d,e,f,g,h);k_get(17,e,f,g,h);
+    k_set(18,a,b,c,d);sb1(a,b,c,d,e,f,g,h);k_get(18,e,f,g,h);
+    k_set(19,a,b,c,d);sb0(a,b,c,d,e,f,g,h);k_get(19,e,f,g,h);
+    k_set(20,a,b,c,d);sb7(a,b,c,d,e,f,g,h);k_get(20,e,f,g,h);
+    k_set(21,a,b,c,d);sb6(a,b,c,d,e,f,g,h);k_get(21,e,f,g,h);
+    k_set(22,a,b,c,d);sb5(a,b,c,d,e,f,g,h);k_get(22,e,f,g,h);
+    k_set(23,a,b,c,d);sb4(a,b,c,d,e,f,g,h);k_get(23,e,f,g,h);
+    k_set(24,a,b,c,d);sb3(a,b,c,d,e,f,g,h);k_get(24,e,f,g,h);
+    k_set(25,a,b,c,d);sb2(a,b,c,d,e,f,g,h);k_get(25,e,f,g,h);
+    k_set(26,a,b,c,d);sb1(a,b,c,d,e,f,g,h);k_get(26,e,f,g,h);
+    k_set(27,a,b,c,d);sb0(a,b,c,d,e,f,g,h);k_get(27,e,f,g,h);
+    k_set(28,a,b,c,d);sb7(a,b,c,d,e,f,g,h);k_get(28,e,f,g,h);
+    k_set(29,a,b,c,d);sb6(a,b,c,d,e,f,g,h);k_get(29,e,f,g,h);
+    k_set(30,a,b,c,d);sb5(a,b,c,d,e,f,g,h);k_get(30,e,f,g,h);
+    k_set(31,a,b,c,d);sb4(a,b,c,d,e,f,g,h);k_get(31,e,f,g,h);
+    k_set(32,a,b,c,d);sb3(a,b,c,d,e,f,g,h);k_get(32,e,f,g,h);
+
+    return;
+}
+
+// encrypt a block of text
+
+void serpent_encrypt(serpent_ctx *ctx, const u_int8_t in_blk[],
+                    u_int8_t out_blk[])
+{
+    u_int32_t  a,b,c,d,e,f,g,h;
+    u_int32_t  t1,t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14,t15,t16;
+
+    a = htole32(*((const u_int32_t *)(in_blk     )));
+    b = htole32(*((const u_int32_t *)(in_blk + 4 )));
+    c = htole32(*((const u_int32_t *)(in_blk + 8 )));
+    d = htole32(*((const u_int32_t *)(in_blk + 12)));
+
+    k_xor( 0,a,b,c,d); sb0(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor( 1,e,f,g,h); sb1(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor( 2,a,b,c,d); sb2(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor( 3,e,f,g,h); sb3(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor( 4,a,b,c,d); sb4(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor( 5,e,f,g,h); sb5(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor( 6,a,b,c,d); sb6(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor( 7,e,f,g,h); sb7(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor( 8,a,b,c,d); sb0(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor( 9,e,f,g,h); sb1(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(10,a,b,c,d); sb2(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(11,e,f,g,h); sb3(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(12,a,b,c,d); sb4(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(13,e,f,g,h); sb5(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(14,a,b,c,d); sb6(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(15,e,f,g,h); sb7(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(16,a,b,c,d); sb0(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(17,e,f,g,h); sb1(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(18,a,b,c,d); sb2(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(19,e,f,g,h); sb3(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(20,a,b,c,d); sb4(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(21,e,f,g,h); sb5(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(22,a,b,c,d); sb6(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(23,e,f,g,h); sb7(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(24,a,b,c,d); sb0(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(25,e,f,g,h); sb1(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(26,a,b,c,d); sb2(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(27,e,f,g,h); sb3(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(28,a,b,c,d); sb4(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(29,e,f,g,h); sb5(e,f,g,h,a,b,c,d); rot(a,b,c,d);
+    k_xor(30,a,b,c,d); sb6(a,b,c,d,e,f,g,h); rot(e,f,g,h);
+    k_xor(31,e,f,g,h); sb7(e,f,g,h,a,b,c,d); k_xor(32,a,b,c,d);
+
+    *((u_int32_t *)(out_blk     )) = le32toh(a);
+    *((u_int32_t *)(out_blk +  4)) = le32toh(b);
+    *((u_int32_t *)(out_blk +  8)) = le32toh(c);
+    *((u_int32_t *)(out_blk + 12)) = le32toh(d);
+}
+
+// decrypt a block of text
+
+void serpent_decrypt(serpent_ctx *ctx, const u_int8_t in_blk[],
+                    u_int8_t out_blk[])
+{
+    u_int32_t  a,b,c,d,e,f,g,h;
+    u_int32_t  t1,t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14,t15,t16;
+
+    a = htole32(*((const u_int32_t *)(in_blk     )));
+    b = htole32(*((const u_int32_t *)(in_blk + 4 )));
+    c = htole32(*((const u_int32_t *)(in_blk + 8 )));
+    d = htole32(*((const u_int32_t *)(in_blk + 12)));
+
+    k_xor(32,a,b,c,d); ib7(a,b,c,d,e,f,g,h); k_xor(31,e,f,g,h);
+    irot(e,f,g,h); ib6(e,f,g,h,a,b,c,d); k_xor(30,a,b,c,d);
+    irot(a,b,c,d); ib5(a,b,c,d,e,f,g,h); k_xor(29,e,f,g,h);
+    irot(e,f,g,h); ib4(e,f,g,h,a,b,c,d); k_xor(28,a,b,c,d);
+    irot(a,b,c,d); ib3(a,b,c,d,e,f,g,h); k_xor(27,e,f,g,h);
+    irot(e,f,g,h); ib2(e,f,g,h,a,b,c,d); k_xor(26,a,b,c,d);
+    irot(a,b,c,d); ib1(a,b,c,d,e,f,g,h); k_xor(25,e,f,g,h);
+    irot(e,f,g,h); ib0(e,f,g,h,a,b,c,d); k_xor(24,a,b,c,d);
+    irot(a,b,c,d); ib7(a,b,c,d,e,f,g,h); k_xor(23,e,f,g,h);
+    irot(e,f,g,h); ib6(e,f,g,h,a,b,c,d); k_xor(22,a,b,c,d);
+    irot(a,b,c,d); ib5(a,b,c,d,e,f,g,h); k_xor(21,e,f,g,h);
+    irot(e,f,g,h); ib4(e,f,g,h,a,b,c,d); k_xor(20,a,b,c,d);
+    irot(a,b,c,d); ib3(a,b,c,d,e,f,g,h); k_xor(19,e,f,g,h);
+    irot(e,f,g,h); ib2(e,f,g,h,a,b,c,d); k_xor(18,a,b,c,d);
+    irot(a,b,c,d); ib1(a,b,c,d,e,f,g,h); k_xor(17,e,f,g,h);
+    irot(e,f,g,h); ib0(e,f,g,h,a,b,c,d); k_xor(16,a,b,c,d);
+    irot(a,b,c,d); ib7(a,b,c,d,e,f,g,h); k_xor(15,e,f,g,h);
+    irot(e,f,g,h); ib6(e,f,g,h,a,b,c,d); k_xor(14,a,b,c,d);
+    irot(a,b,c,d); ib5(a,b,c,d,e,f,g,h); k_xor(13,e,f,g,h);
+    irot(e,f,g,h); ib4(e,f,g,h,a,b,c,d); k_xor(12,a,b,c,d);
+    irot(a,b,c,d); ib3(a,b,c,d,e,f,g,h); k_xor(11,e,f,g,h);
+    irot(e,f,g,h); ib2(e,f,g,h,a,b,c,d); k_xor(10,a,b,c,d);
+    irot(a,b,c,d); ib1(a,b,c,d,e,f,g,h); k_xor( 9,e,f,g,h);
+    irot(e,f,g,h); ib0(e,f,g,h,a,b,c,d); k_xor( 8,a,b,c,d);
+    irot(a,b,c,d); ib7(a,b,c,d,e,f,g,h); k_xor( 7,e,f,g,h);
+    irot(e,f,g,h); ib6(e,f,g,h,a,b,c,d); k_xor( 6,a,b,c,d);
+    irot(a,b,c,d); ib5(a,b,c,d,e,f,g,h); k_xor( 5,e,f,g,h);
+    irot(e,f,g,h); ib4(e,f,g,h,a,b,c,d); k_xor( 4,a,b,c,d);
+    irot(a,b,c,d); ib3(a,b,c,d,e,f,g,h); k_xor( 3,e,f,g,h);
+    irot(e,f,g,h); ib2(e,f,g,h,a,b,c,d); k_xor( 2,a,b,c,d);
+    irot(a,b,c,d); ib1(a,b,c,d,e,f,g,h); k_xor( 1,e,f,g,h);
+    irot(e,f,g,h); ib0(e,f,g,h,a,b,c,d); k_xor( 0,a,b,c,d);
+
+    *((u_int32_t *)(out_blk     )) = le32toh(a);
+    *((u_int32_t *)(out_blk +  4)) = le32toh(b);
+    *((u_int32_t *)(out_blk +  8)) = le32toh(c);
+    *((u_int32_t *)(out_blk + 12)) = le32toh(d);
+}
+
diff --git a/sys/crypto/serpent/serpent.h b/sys/crypto/serpent/serpent.h
new file mode 100644 (file)
index 0000000..7319cb7
--- /dev/null
@@ -0,0 +1,16 @@
+
+// Copyright in this code is held by Dr B. R. Gladman but free direct or
+// derivative use is permitted subject to acknowledgement of its origin.
+// Dr B. R. Gladman                               .   25th January 2000.
+
+typedef struct
+{
+       uint32_t l_key[140];
+} serpent_ctx;
+
+void serpent_set_key(serpent_ctx *ctx, const u_int8_t in_key[], int key_len);
+void serpent_encrypt(serpent_ctx *ctx, const u_int8_t in_blk[],
+                    u_int8_t out_blk[]);
+void serpent_decrypt(serpent_ctx *ctx, const u_int8_t in_blk[],
+                    u_int8_t out_blk[]);
+