123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675/*
 * Cryptographic API.
 *
 * Serpent Cipher Algorithm.
 *
 * Copyright (C) 2002 Dag Arne Osvik <osvik@ii.uib.no>
 *               2003 Herbert Valerio Riedel <hvr@gnu.org>
 *
 * Added tnepres support:
 *		Ruben Jesus Garcia Hernandez <ruben@ugr.es>, 18.10.2004
 *              Based on code by hvr
 *
 * 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.
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/errno.h>
#include <asm/byteorder.h>
#include <linux/crypto.h>
#include <linux/types.h>
#include <crypto/serpent.h>

/* Key is padded to the maximum of 256 bits before round key generation.
 * Any key length <= 256 bits (32 bytes) is allowed by the algorithm.
 */

#define PHI 0x9e3779b9UL

#define keyiter(a, b, c, d, i, j) \
	({ b ^= d; b ^= c; b ^= a; b ^= PHI ^ i; b = rol32(b, 11); k[j] = b; })

#define loadkeys(x0, x1, x2, x3, i) \
	({ x0 = k[i]; x1 = k[i+1]; x2 = k[i+2]; x3 = k[i+3]; })

#define storekeys(x0, x1, x2, x3, i) \
	({ k[i] = x0; k[i+1] = x1; k[i+2] = x2; k[i+3] = x3; })

#define store_and_load_keys(x0, x1, x2, x3, s, l) \
	({ storekeys(x0, x1, x2, x3, s); loadkeys(x0, x1, x2, x3, l); })

#define K(x0, x1, x2, x3, i) ({				\
	x3 ^= k[4*(i)+3];        x2 ^= k[4*(i)+2];	\
	x1 ^= k[4*(i)+1];        x0 ^= k[4*(i)+0];	\
	})

#define LK(x0, x1, x2, x3, x4, i) ({					   \
							x0 = rol32(x0, 13);\
	x2 = rol32(x2, 3);	x1 ^= x0;		x4  = x0 << 3;	   \
	x3 ^= x2;		x1 ^= x2;				   \
	x1 = rol32(x1, 1);	x3 ^= x4;				   \
	x3 = rol32(x3, 7);	x4  = x1;				   \
	x0 ^= x1;		x4 <<= 7;		x2 ^= x3;	   \
	x0 ^= x3;		x2 ^= x4;		x3 ^= k[4*i+3];	   \
	x1 ^= k[4*i+1];		x0 = rol32(x0, 5);	x2 = rol32(x2, 22);\
	x0 ^= k[4*i+0];		x2 ^= k[4*i+2];				   \
	})

#define KL(x0, x1, x2, x3, x4, i) ({					   \
	x0 ^= k[4*i+0];		x1 ^= k[4*i+1];		x2 ^= k[4*i+2];	   \
	x3 ^= k[4*i+3];		x0 = ror32(x0, 5);	x2 = ror32(x2, 22);\
	x4 =  x1;		x2 ^= x3;		x0 ^= x3;	   \
	x4 <<= 7;		x0 ^= x1;		x1 = ror32(x1, 1); \
	x2 ^= x4;		x3 = ror32(x3, 7);	x4 = x0 << 3;	   \
	x1 ^= x0;		x3 ^= x4;		x0 = ror32(x0, 13);\
	x1 ^= x2;		x3 ^= x2;		x2 = ror32(x2, 3); \
	})

#define S0(x0, x1, x2, x3, x4) ({			\
					x4  = x3;	\
	x3 |= x0;	x0 ^= x4;	x4 ^= x2;	\
	x4 = ~x4;	x3 ^= x1;	x1 &= x0;	\
	x1 ^= x4;	x2 ^= x0;	x0 ^= x3;	\
	x4 |= x0;	x0 ^= x2;	x2 &= x1;	\
	x3 ^= x2;	x1 = ~x1;	x2 ^= x4;	\
	x1 ^= x2;					\
	})

#define S1(x0, x1, x2, x3, x4) ({			\
					x4  = x1;	\
	x1 ^= x0;	x0 ^= x3;	x3 = ~x3;	\
	x4 &= x1;	x0 |= x1;	x3 ^= x2;	\
	x0 ^= x3;	x1 ^= x3;	x3 ^= x4;	\
	x1 |= x4;	x4 ^= x2;	x2 &= x0;	\
	x2 ^= x1;	x1 |= x0;	x0 = ~x0;	\
	x0 ^= x2;	x4 ^= x1;			\
	})

#define S2(x0, x1, x2, x3, x4) ({			\
					x3 = ~x3;	\
	x1 ^= x0;	x4  = x0;	x0 &= x2;	\
	x0 ^= x3;	x3 |= x4;	x2 ^= x1;	\
	x3 ^= x1;	x1 &= x0;	x0 ^= x2;	\
	x2 &= x3;	x3 |= x1;	x0 = ~x0;	\
	x3 ^= x0;	x4 ^= x0;	x0 ^= x2;	\
	x1 |= x2;					\
	})

#define S3(x0, x1, x2, x3, x4) ({			\
					x4  = x1;	\
	x1 ^= x3;	x3 |= x0;	x4 &= x0;	\
	x0 ^= x2;	x2 ^= x1;	x1 &= x3;	\
	x2 ^= x3;	x0 |= x4;	x4 ^= x3;	\
	x1 ^= x0;	x0 &= x3;	x3 &= x4;	\
	x3 ^= x2;	x4 |= x1;	x2 &= x1;	\
	x4 ^= x3;	x0 ^= x3;	x3 ^= x2;	\
	})

#define S4(x0, x1, x2, x3, x4) ({			\
					x4  = x3;	\
	x3 &= x0;	x0 ^= x4;			\
	x3 ^= x2;	x2 |= x4;	x0 ^= x1;	\
	x4 ^= x3;	x2 |= x0;			\
	x2 ^= x1;	x1 &= x0;			\
	x1 ^= x4;	x4 &= x2;	x2 ^= x3;	\
	x4 ^= x0;	x3 |= x1;	x1 = ~x1;	\
	x3 ^= x0;					\
	})

#define S5(x0, x1, x2, x3, x4) ({			\
	x4  = x1;	x1 |= x0;			\
	x2 ^= x1;	x3 = ~x3;	x4 ^= x0;	\
	x0 ^= x2;	x1 &= x4;	x4 |= x3;	\
	x4 ^= x0;	x0 &= x3;	x1 ^= x3;	\
	x3 ^= x2;	x0 ^= x1;	x2 &= x4;	\
	x1 ^= x2;	x2 &= x0;			\
	x3 ^= x2;					\
	})

#define S6(x0, x1, x2, x3, x4) ({			\
					x4  = x1;	\
	x3 ^= x0;	x1 ^= x2;	x2 ^= x0;	\
	x0 &= x3;	x1 |= x3;	x4 = ~x4;	\
	x0 ^= x1;	x1 ^= x2;			\
	x3 ^= x4;	x4 ^= x0;	x2 &= x0;	\
	x4 ^= x1;	x2 ^= x3;	x3 &= x1;	\
	x3 ^= x0;	x1 ^= x2;			\
	})

#define S7(x0, x1, x2, x3, x4) ({			\
					x1 = ~x1;	\
	x4  = x1;	x0 = ~x0;	x1 &= x2;	\
	x1 ^= x3;	x3 |= x4;	x4 ^= x2;	\
	x2 ^= x3;	x3 ^= x0;	x0 |= x1;	\
	x2 &= x0;	x0 ^= x4;	x4 ^= x3;	\
	x3 &= x0;	x4 ^= x1;			\
	x2 ^= x4;	x3 ^= x1;	x4 |= x0;	\
	x4 ^= x1;					\
	})

#define SI0(x0, x1, x2, x3, x4) ({			\
			x4  = x3;	x1 ^= x0;	\
	x3 |= x1;	x4 ^= x1;	x0 = ~x0;	\
	x2 ^= x3;	x3 ^= x0;	x0 &= x1;	\
	x0 ^= x2;	x2 &= x3;	x3 ^= x4;	\
	x2 ^= x3;	x1 ^= x3;	x3 &= x0;	\
	x1 ^= x0;	x0 ^= x2;	x4 ^= x3;	\
	})

#define SI1(x0, x1, x2, x3, x4) ({			\
	x1 ^= x3;	x4  = x0;			\
	x0 ^= x2;	x2 = ~x2;	x4 |= x1;	\
	x4 ^= x3;	x3 &= x1;	x1 ^= x2;	\
	x2 &= x4;	x4 ^= x1;	x1 |= x3;	\
	x3 ^= x0;	x2 ^= x0;	x0 |= x4;	\
	x2 ^= x4;	x1 ^= x0;			\
	x4 ^= x1;					\
	})

#define SI2(x0, x1, x2, x3, x4) ({			\
	x2 ^= x1;	x4  = x3;	x3 = ~x3;	\
	x3 |= x2;	x2 ^= x4;	x4 ^= x0;	\
	x3 ^= x1;	x1 |= x2;	x2 ^= x0;	\
	x1 ^= x4;	x4 |= x3;	x2 ^= x3;	\
	x4 ^= x2;	x2 &= x1;			\
	x2 ^= x3;	x3 ^= x4;	x4 ^= x0;	\
	})

#define SI3(x0, x1, x2, x3, x4) ({			\
					x2 ^= x1;	\
	x4  = x1;	x1 &= x2;			\
	x1 ^= x0;	x0 |= x4;	x4 ^= x3;	\
	x0 ^= x3;	x3 |= x1;	x1 ^= x2;	\
	x1 ^= x3;	x0 ^= x2;	x2 ^= x3;	\
	x3 &= x1;	x1 ^= x0;	x0 &= x2;	\
	x4 ^= x3;	x3 ^= x0;	x0 ^= x1;	\
	})

#define SI4(x0, x1, x2, x3, x4) ({			\
	x2 ^= x3;	x4  = x0;	x0 &= x1;	\
	x0 ^= x2;	x2 |= x3;	x4 = ~x4;	\
	x1 ^= x0;	x0 ^= x2;	x2 &= x4;	\
	x2 ^= x0;	x0 |= x4;			\
	x0 ^= x3;	x3 &= x2;			\
	x4 ^= x3;	x3 ^= x1;	x1 &= x0;	\
	x4 ^= x1;	x0 ^= x3;			\
	})

#define SI5(x0, x1, x2, x3, x4) ({			\
			x4  = x1;	x1 |= x2;	\
	x2 ^= x4;	x1 ^= x3;	x3 &= x4;	\
	x2 ^= x3;	x3 |= x0;	x0 = ~x0;	\
	x3 ^= x2;	x2 |= x0;	x4 ^= x1;	\
	x2 ^= x4;	x4 &= x0;	x0 ^= x1;	\
	x1 ^= x3;	x0 &= x2;	x2 ^= x3;	\
	x0 ^= x2;	x2 ^= x4;	x4 ^= x3;	\
	})

#define SI6(x0, x1, x2, x3, x4) ({			\
			x0 ^= x2;			\
	x4  = x0;	x0 &= x3;	x2 ^= x3;	\
	x0 ^= x2;	x3 ^= x1;	x2 |= x4;	\
	x2 ^= x3;	x3 &= x0;	x0 = ~x0;	\
	x3 ^= x1;	x1 &= x2;	x4 ^= x0;	\
	x3 ^= x4;	x4 ^= x2;	x0 ^= x1;	\
	x2 ^= x0;					\
	})

#define SI7(x0, x1, x2, x3, x4) ({			\
	x4  = x3;	x3 &= x0;	x0 ^= x2;	\
	x2 |= x4;	x4 ^= x1;	x0 = ~x0;	\
	x1 |= x3;	x4 ^= x0;	x0 &= x2;	\
	x0 ^= x1;	x1 &= x2;	x3 ^= x2;	\
	x4 ^= x3;	x2 &= x3;	x3 |= x0;	\
	x1 ^= x4;	x3 ^= x4;	x4 &= x0;	\
	x4 ^= x2;					\
	})

static void __serpent_setkey_sbox(u32 r0, u32 r1, u32 r2, u32 r3, u32 r4, u32 *k)
{
	k += 100;
	S3(r3, r4, r0, r1, r2); store_and_load_keys(r1, r2, r4, r3, 28, 24);
	S4(r1, r2, r4, r3, r0); store_and_load_keys(r2, r4, r3, r0, 24, 20);
	S5(r2, r4, r3, r0, r1); store_and_load_keys(r1, r2, r4, r0, 20, 16);
	S6(r1, r2, r4, r0, r3); store_and_load_keys(r4, r3, r2, r0, 16, 12);
	S7(r4, r3, r2, r0, r1); store_and_load_keys(r1, r2, r0, r4, 12, 8);
	S0(r1, r2, r0, r4, r3); store_and_load_keys(r0, r2, r4, r1, 8, 4);
	S1(r0, r2, r4, r1, r3); store_and_load_keys(r3, r4, r1, r0, 4, 0);
	S2(r3, r4, r1, r0, r2); store_and_load_keys(r2, r4, r3, r0, 0, -4);
	S3(r2, r4, r3, r0, r1); store_and_load_keys(r0, r1, r4, r2, -4, -8);
	S4(r0, r1, r4, r2, r3); store_and_load_keys(r1, r4, r2, r3, -8, -12);
	S5(r1, r4, r2, r3, r0); store_and_load_keys(r0, r1, r4, r3, -12, -16);
	S6(r0, r1, r4, r3, r2); store_and_load_keys(r4, r2, r1, r3, -16, -20);
	S7(r4, r2, r1, r3, r0); store_and_load_keys(r0, r1, r3, r4, -20, -24);
	S0(r0, r1, r3, r4, r2); store_and_load_keys(r3, r1, r4, r0, -24, -28);
	k -= 50;
	S1(r3, r1, r4, r0, r2); store_and_load_keys(r2, r4, r0, r3, 22, 18);
	S2(r2, r4, r0, r3, r1); store_and_load_keys(r1, r4, r2, r3, 18, 14);
	S3(r1, r4, r2, r3, r0); store_and_load_keys(r3, r0, r4, r1, 14, 10);
	S4(r3, r0, r4, r1, r2); store_and_load_keys(r0, r4, r1, r2, 10, 6);
	S5(r0, r4, r1, r2, r3); store_and_load_keys(r3, r0, r4, r2, 6, 2);
	S6(r3, r0, r4, r2, r1); store_and_load_keys(r4, r1, r0, r2, 2, -2);
	S7(r4, r1, r0, r2, r3); store_and_load_keys(r3, r0, r2, r4, -2, -6);
	S0(r3, r0, r2, r4, r1); store_and_load_keys(r2, r0, r4, r3, -6, -10);
	S1(r2, r0, r4, r3, r1); store_and_load_keys(r1, r4, r3, r2, -10, -14);
	S2(r1, r4, r3, r2, r0); store_and_load_keys(r0, r4, r1, r2, -14, -18);
	S3(r0, r4, r1, r2, r3); store_and_load_keys(r2, r3, r4, r0, -18, -22);
	k -= 50;
	S4(r2, r3, r4, r0, r1); store_and_load_keys(r3, r4, r0, r1, 28, 24);
	S5(r3, r4, r0, r1, r2); store_and_load_keys(r2, r3, r4, r1, 24, 20);
	S6(r2, r3, r4, r1, r0); store_and_load_keys(r4, r0, r3, r1, 20, 16);
	S7(r4, r0, r3, r1, r2); store_and_load_keys(r2, r3, r1, r4, 16, 12);
	S0(r2, r3, r1, r4, r0); store_and_load_keys(r1, r3, r4, r2, 12, 8);
	S1(r1, r3, r4, r2, r0); store_and_load_keys(r0, r4, r2, r1, 8, 4);
	S2(r0, r4, r2, r1, r3); store_and_load_keys(r3, r4, r0, r1, 4, 0);
	S3(r3, r4, r0, r1, r2); storekeys(r1, r2, r4, r3, 0);
}

int __serpent_setkey(struct serpent_ctx *ctx, const u8 *key,
		     unsigned int keylen)
{
	u32 *k = ctx->expkey;
	u8  *k8 = (u8 *)k;
	u32 r0, r1, r2, r3, r4;
	int i;

	/* Copy key, add padding */

	for (i = 0; i < keylen; ++i)
		k8[i] = key[i];
	if (i < SERPENT_MAX_KEY_SIZE)
		k8[i++] = 1;
	while (i < SERPENT_MAX_KEY_SIZE)
		k8[i++] = 0;

	/* Expand key using polynomial */

	r0 = le32_to_cpu(k[3]);
	r1 = le32_to_cpu(k[4]);
	r2 = le32_to_cpu(k[5]);
	r3 = le32_to_cpu(k[6]);
	r4 = le32_to_cpu(k[7]);

	keyiter(le32_to_cpu(k[0]), r0, r4, r2, 0, 0);
	keyiter(le32_to_cpu(k[1]), r1, r0, r3, 1, 1);
	keyiter(le32_to_cpu(k[2]), r2, r1, r4, 2, 2);
	keyiter(le32_to_cpu(k[3]), r3, r2, r0, 3, 3);
	keyiter(le32_to_cpu(k[4]), r4, r3, r1, 4, 4);
	keyiter(le32_to_cpu(k[5]), r0, r4, r2, 5, 5);
	keyiter(le32_to_cpu(k[6]), r1, r0, r3, 6, 6);
	keyiter(le32_to_cpu(k[7]), r2, r1, r4, 7, 7);

	keyiter(k[0], r3, r2, r0, 8, 8);
	keyiter(k[1], r4, r3, r1, 9, 9);
	keyiter(k[2], r0, r4, r2, 10, 10);
	keyiter(k[3], r1, r0, r3, 11, 11);
	keyiter(k[4], r2, r1, r4, 12, 12);
	keyiter(k[5], r3, r2, r0, 13, 13);
	keyiter(k[6], r4, r3, r1, 14, 14);
	keyiter(k[7], r0, r4, r2, 15, 15);
	keyiter(k[8], r1, r0, r3, 16, 16);
	keyiter(k[9], r2, r1, r4, 17, 17);
	keyiter(k[10], r3, r2, r0, 18, 18);
	keyiter(k[11], r4, r3, r1, 19, 19);
	keyiter(k[12], r0, r4, r2, 20, 20);
	keyiter(k[13], r1, r0, r3, 21, 21);
	keyiter(k[14], r2, r1, r4, 22, 22);
	keyiter(k[15], r3, r2, r0, 23, 23);
	keyiter(k[16], r4, r3, r1, 24, 24);
	keyiter(k[17], r0, r4, r2, 25, 25);
	keyiter(k[18], r1, r0, r3, 26, 26);
	keyiter(k[19], r2, r1, r4, 27, 27);
	keyiter(k[20], r3, r2, r0, 28, 28);
	keyiter(k[21], r4, r3, r1, 29, 29);
	keyiter(k[22], r0, r4, r2, 30, 30);
	keyiter(k[23], r1, r0, r3, 31, 31);

	k += 50;

	keyiter(k[-26], r2, r1, r4, 32, -18);
	keyiter(k[-25], r3, r2, r0, 33, -17);
	keyiter(k[-24], r4, r3, r1, 34, -16);
	keyiter(k[-23], r0, r4, r2, 35, -15);
	keyiter(k[-22], r1, r0, r3, 36, -14);
	keyiter(k[-21], r2, r1, r4, 37, -13);
	keyiter(k[-20], r3, r2, r0, 38, -12);
	keyiter(k[-19], r4, r3, r1, 39, -11);
	keyiter(k[-18], r0, r4, r2, 40, -10);
	keyiter(k[-17], r1, r0, r3, 41, -9);
	keyiter(k[-16], r2, r1, r4, 42, -8);
	keyiter(k[-15], r3, r2, r0, 43, -7);
	keyiter(k[-14], r4, r3, r1, 44, -6);
	keyiter(k[-13], r0, r4, r2, 45, -5);
	keyiter(k[-12], r1, r0, r3, 46, -4);
	keyiter(k[-11], r2, r1, r4, 47, -3);
	keyiter(k[-10], r3, r2, r0, 48, -2);
	keyiter(k[-9], r4, r3, r1, 49, -1);
	keyiter(k[-8], r0, r4, r2, 50, 0);
	keyiter(k[-7], r1, r0, r3, 51, 1);
	keyiter(k[-6], r2, r1, r4, 52, 2);
	keyiter(k[-5], r3, r2, r0, 53, 3);
	keyiter(k[-4], r4, r3, r1, 54, 4);
	keyiter(k[-3], r0, r4, r2, 55, 5);
	keyiter(k[-2], r1, r0, r3, 56, 6);
	keyiter(k[-1], r2, r1, r4, 57, 7);
	keyiter(k[0], r3, r2, r0, 58, 8);
	keyiter(k[1], r4, r3, r1, 59, 9);
	keyiter(k[2], r0, r4, r2, 60, 10);
	keyiter(k[3], r1, r0, r3, 61, 11);
	keyiter(k[4], r2, r1, r4, 62, 12);
	keyiter(k[5], r3, r2, r0, 63, 13);
	keyiter(k[6], r4, r3, r1, 64, 14);
	keyiter(k[7], r0, r4, r2, 65, 15);
	keyiter(k[8], r1, r0, r3, 66, 16);
	keyiter(k[9], r2, r1, r4, 67, 17);
	keyiter(k[10], r3, r2, r0, 68, 18);
	keyiter(k[11], r4, r3, r1, 69, 19);
	keyiter(k[12], r0, r4, r2, 70, 20);
	keyiter(k[13], r1, r0, r3, 71, 21);
	keyiter(k[14], r2, r1, r4, 72, 22);
	keyiter(k[15], r3, r2, r0, 73, 23);
	keyiter(k[16], r4, r3, r1, 74, 24);
	keyiter(k[17], r0, r4, r2, 75, 25);
	keyiter(k[18], r1, r0, r3, 76, 26);
	keyiter(k[19], r2, r1, r4, 77, 27);
	keyiter(k[20], r3, r2, r0, 78, 28);
	keyiter(k[21], r4, r3, r1, 79, 29);
	keyiter(k[22], r0, r4, r2, 80, 30);
	keyiter(k[23], r1, r0, r3, 81, 31);

	k += 50;

	keyiter(k[-26], r2, r1, r4, 82, -18);
	keyiter(k[-25], r3, r2, r0, 83, -17);
	keyiter(k[-24], r4, r3, r1, 84, -16);
	keyiter(k[-23], r0, r4, r2, 85, -15);
	keyiter(k[-22], r1, r0, r3, 86, -14);
	keyiter(k[-21], r2, r1, r4, 87, -13);
	keyiter(k[-20], r3, r2, r0, 88, -12);
	keyiter(k[-19], r4, r3, r1, 89, -11);
	keyiter(k[-18], r0, r4, r2, 90, -10);
	keyiter(k[-17], r1, r0, r3, 91, -9);
	keyiter(k[-16], r2, r1, r4, 92, -8);
	keyiter(k[-15], r3, r2, r0, 93, -7);
	keyiter(k[-14], r4, r3, r1, 94, -6);
	keyiter(k[-13], r0, r4, r2, 95, -5);
	keyiter(k[-12], r1, r0, r3, 96, -4);
	keyiter(k[-11], r2, r1, r4, 97, -3);
	keyiter(k[-10], r3, r2, r0, 98, -2);
	keyiter(k[-9], r4, r3, r1, 99, -1);
	keyiter(k[-8], r0, r4, r2, 100, 0);
	keyiter(k[-7], r1, r0, r3, 101, 1);
	keyiter(k[-6], r2, r1, r4, 102, 2);
	keyiter(k[-5], r3, r2, r0, 103, 3);
	keyiter(k[-4], r4, r3, r1, 104, 4);
	keyiter(k[-3], r0, r4, r2, 105, 5);
	keyiter(k[-2], r1, r0, r3, 106, 6);
	keyiter(k[-1], r2, r1, r4, 107, 7);
	keyiter(k[0], r3, r2, r0, 108, 8);
	keyiter(k[1], r4, r3, r1, 109, 9);
	keyiter(k[2], r0, r4, r2, 110, 10);
	keyiter(k[3], r1, r0, r3, 111, 11);
	keyiter(k[4], r2, r1, r4, 112, 12);
	keyiter(k[5], r3, r2, r0, 113, 13);
	keyiter(k[6], r4, r3, r1, 114, 14);
	keyiter(k[7], r0, r4, r2, 115, 15);
	keyiter(k[8], r1, r0, r3, 116, 16);
	keyiter(k[9], r2, r1, r4, 117, 17);
	keyiter(k[10], r3, r2, r0, 118, 18);
	keyiter(k[11], r4, r3, r1, 119, 19);
	keyiter(k[12], r0, r4, r2, 120, 20);
	keyiter(k[13], r1, r0, r3, 121, 21);
	keyiter(k[14], r2, r1, r4, 122, 22);
	keyiter(k[15], r3, r2, r0, 123, 23);
	keyiter(k[16], r4, r3, r1, 124, 24);
	keyiter(k[17], r0, r4, r2, 125, 25);
	keyiter(k[18], r1, r0, r3, 126, 26);
	keyiter(k[19], r2, r1, r4, 127, 27);
	keyiter(k[20], r3, r2, r0, 128, 28);
	keyiter(k[21], r4, r3, r1, 129, 29);
	keyiter(k[22], r0, r4, r2, 130, 30);
	keyiter(k[23], r1, r0, r3, 131, 31);

	/* Apply S-boxes */
	__serpent_setkey_sbox(r0, r1, r2, r3, r4, ctx->expkey);

	return 0;
}
EXPORT_SYMBOL_GPL(__serpent_setkey);

int serpent_setkey(struct crypto_tfm *tfm, const u8 *key, unsigned int keylen)
{
	return __serpent_setkey(crypto_tfm_ctx(tfm), key, keylen);
}
EXPORT_SYMBOL_GPL(serpent_setkey);

void __serpent_encrypt(struct serpent_ctx *ctx, u8 *dst, const u8 *src)
{
	const u32 *k = ctx->expkey;
	const __le32 *s = (const __le32 *)src;
	__le32	*d = (__le32 *)dst;
	u32	r0, r1, r2, r3, r4;

/*
 * Note: The conversions between u8* and u32* might cause trouble
 * on architectures with stricter alignment rules than x86
 */

	r0 = le32_to_cpu(s[0]);
	r1 = le32_to_cpu(s[1]);
	r2 = le32_to_cpu(s[2]);
	r3 = le32_to_cpu(s[3]);

					K(r0, r1, r2, r3, 0);
	S0(r0, r1, r2, r3, r4);		LK(r2, r1, r3, r0, r4, 1);
	S1(r2, r1, r3, r0, r4);		LK(r4, r3, r0, r2, r1, 2);
	S2(r4, r3, r0, r2, r1);		LK(r1, r3, r4, r2, r0, 3);
	S3(r1, r3, r4, r2, r0);		LK(r2, r0, r3, r1, r4, 4);
	S4(r2, r0, r3, r1, r4);		LK(r0, r3, r1, r4, r2, 5);
	S5(r0, r3, r1, r4, r2);		LK(r2, r0, r3, r4, r1, 6);
	S6(r2, r0, r3, r4, r1);		LK(r3, r1, r0, r4, r2, 7);
	S7(r3, r1, r0, r4, r2);		LK(r2, r0, r4, r3, r1, 8);
	S0(r2, r0, r4, r3, r1);		LK(r4, r0, r3, r2, r1, 9);
	S1(r4, r0, r3, r2, r1);		LK(r1, r3, r2, r4, r0, 10);
	S2(r1, r3, r2, r4, r0);		LK(r0, r3, r1, r4, r2, 11);
	S3(r0, r3, r1, r4, r2);		LK(r4, r2, r3, r0, r1, 12);
	S4(r4, r2, r3, r0, r1);		LK(r2, r3, r0, r1, r4, 13);
	S5(r2, r3, r0, r1, r4);		LK(r4, r2, r3, r1, r0, 14);
	S6(r4, r2, r3, r1, r0);		LK(r3, r0, r2, r1, r4, 15);
	S7(r3, r0, r2, r1, r4);		LK(r4, r2, r1, r3, r0, 16);
	S0(r4, r2, r1, r3, r0);		LK(r1, r2, r3, r4, r0, 17);
	S1(r1, r2, r3, r4, r0);		LK(r0, r3, r4, r1, r2, 18);
	S2(r0, r3, r4, r1, r2);		LK(r2, r3, r0, r1, r4, 19);
	S3(r2, r3, r0, r1, r4);		LK(r1, r4, r3, r2, r0, 20);
	S4(r1, r4, r3, r2, r0);		LK(r4, r3, r2, r0, r1, 21);
	S5(r4, r3, r2, r0, r1);		LK(r1, r4, r3, r0, r2, 22);
	S6(r1, r4, r3, r0, r2);		LK(r3, r2, r4, r0, r1, 23);
	S7(r3, r2, r4, r0, r1);		LK(r1, r4, r0, r3, r2, 24);
	S0(r1, r4, r0, r3, r2);		LK(r0, r4, r3, r1, r2, 25);
	S1(r0, r4, r3, r1, r2);		LK(r2, r3, r1, r0, r4, 26);
	S2(r2, r3, r1, r0, r4);		LK(r4, r3, r2, r0, r1, 27);
	S3(r4, r3, r2, r0, r1);		LK(r0, r1, r3, r4, r2, 28);
	S4(r0, r1, r3, r4, r2);		LK(r1, r3, r4, r2, r0, 29);
	S5(r1, r3, r4, r2, r0);		LK(r0, r1, r3, r2, r4, 30);
	S6(r0, r1, r3, r2, r4);		LK(r3, r4, r1, r2, r0, 31);
	S7(r3, r4, r1, r2, r0);		K(r0, r1, r2, r3, 32);

	d[0] = cpu_to_le32(r0);
	d[1] = cpu_to_le32(r1);
	d[2] = cpu_to_le32(r2);
	d[3] = cpu_to_le32(r3);
}
EXPORT_SYMBOL_GPL(__serpent_encrypt);

static void serpent_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
{
	struct serpent_ctx *ctx = crypto_tfm_ctx(tfm);

	__serpent_encrypt(ctx, dst, src);
}

void __serpent_decrypt(struct serpent_ctx *ctx, u8 *dst, const u8 *src)
{
	const u32 *k = ctx->expkey;
	const __le32 *s = (const __le32 *)src;
	__le32	*d = (__le32 *)dst;
	u32	r0, r1, r2, r3, r4;

	r0 = le32_to_cpu(s[0]);
	r1 = le32_to_cpu(s[1]);
	r2 = le32_to_cpu(s[2]);
	r3 = le32_to_cpu(s[3]);

					K(r0, r1, r2, r3, 32);
	SI7(r0, r1, r2, r3, r4);	KL(r1, r3, r0, r4, r2, 31);
	SI6(r1, r3, r0, r4, r2);	KL(r0, r2, r4, r1, r3, 30);
	SI5(r0, r2, r4, r1, r3);	KL(r2, r3, r0, r4, r1, 29);
	SI4(r2, r3, r0, r4, r1);	KL(r2, r0, r1, r4, r3, 28);
	SI3(r2, r0, r1, r4, r3);	KL(r1, r2, r3, r4, r0, 27);
	SI2(r1, r2, r3, r4, r0);	KL(r2, r0, r4, r3, r1, 26);
	SI1(r2, r0, r4, r3, r1);	KL(r1, r0, r4, r3, r2, 25);
	SI0(r1, r0, r4, r3, r2);	KL(r4, r2, r0, r1, r3, 24);
	SI7(r4, r2, r0, r1, r3);	KL(r2, r1, r4, r3, r0, 23);
	SI6(r2, r1, r4, r3, r0);	KL(r4, r0, r3, r2, r1, 22);
	SI5(r4, r0, r3, r2, r1);	KL(r0, r1, r4, r3, r2, 21);
	SI4(r0, r1, r4, r3, r2);	KL(r0, r4, r2, r3, r1, 20);
	SI3(r0, r4, r2, r3, r1);	KL(r2, r0, r1, r3, r4, 19);
	SI2(r2, r0, r1, r3, r4);	KL(r0, r4, r3, r1, r2, 18);
	SI1(r0, r4, r3, r1, r2);	KL(r2, r4, r3, r1, r0, 17);
	SI0(r2, r4, r3, r1, r0);	KL(r3, r0, r4, r2, r1, 16);
	SI7(r3, r0, r4, r2, r1);	KL(r0, r2, r3, r1, r4, 15);
	SI6(r0, r2, r3, r1, r4);	KL(r3, r4, r1, r0, r2, 14);
	SI5(r3, r4, r1, r0, r2);	KL(r4, r2, r3, r1, r0, 13);
	SI4(r4, r2, r3, r1, r0);	KL(r4, r3, r0, r1, r2, 12);
	SI3(r4, r3, r0, r1, r2);	KL(r0, r4, r2, r1, r3, 11);
	SI2(r0, r4, r2, r1, r3);	KL(r4, r3, r1, r2, r0, 10);
	SI1(r4, r3, r1, r2, r0);	KL(r0, r3, r1, r2, r4, 9);
	SI0(r0, r3, r1, r2, r4);	KL(r1, r4, r3, r0, r2, 8);
	SI7(r1, r4, r3, r0, r2);	KL(r4, r0, r1, r2, r3, 7);
	SI6(r4, r0, r1, r2, r3);	KL(r1, r3, r2, r4, r0, 6);
	SI5(r1, r3, r2, r4, r0);	KL(r3, r0, r1, r2, r4, 5);
	SI4(r3, r0, r1, r2, r4);	KL(r3, r1, r4, r2, r0, 4);
	SI3(r3, r1, r4, r2, r0);	KL(r4, r3, r0, r2, r1, 3);
	SI2(r4, r3, r0, r2, r1);	KL(r3, r1, r2, r0, r4, 2);
	SI1(r3, r1, r2, r0, r4);	KL(r4, r1, r2, r0, r3, 1);
	SI0(r4, r1, r2, r0, r3);	K(r2, r3, r1, r4, 0);

	d[0] = cpu_to_le32(r2);
	d[1] = cpu_to_le32(r3);
	d[2] = cpu_to_le32(r1);
	d[3] = cpu_to_le32(r4);
}
EXPORT_SYMBOL_GPL(__serpent_decrypt);

static void serpent_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
{
	struct serpent_ctx *ctx = crypto_tfm_ctx(tfm);

	__serpent_decrypt(ctx, dst, src);
}

static int tnepres_setkey(struct crypto_tfm *tfm, const u8 *key,
			  unsigned int keylen)
{
	u8 rev_key[SERPENT_MAX_KEY_SIZE];
	int i;

	for (i = 0; i < keylen; ++i)
		rev_key[keylen - i - 1] = key[i];

	return serpent_setkey(tfm, rev_key, keylen);
}

static void tnepres_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
{
	const u32 * const s = (const u32 * const)src;
	u32 * const d = (u32 * const)dst;

	u32 rs[4], rd[4];

	rs[0] = swab32(s[3]);
	rs[1] = swab32(s[2]);
	rs[2] = swab32(s[1]);
	rs[3] = swab32(s[0]);

	serpent_encrypt(tfm, (u8 *)rd, (u8 *)rs);

	d[0] = swab32(rd[3]);
	d[1] = swab32(rd[2]);
	d[2] = swab32(rd[1]);
	d[3] = swab32(rd[0]);
}

static void tnepres_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
{
	const u32 * const s = (const u32 * const)src;
	u32 * const d = (u32 * const)dst;

	u32 rs[4], rd[4];

	rs[0] = swab32(s[3]);
	rs[1] = swab32(s[2]);
	rs[2] = swab32(s[1]);
	rs[3] = swab32(s[0]);

	serpent_decrypt(tfm, (u8 *)rd, (u8 *)rs);

	d[0] = swab32(rd[3]);
	d[1] = swab32(rd[2]);
	d[2] = swab32(rd[1]);
	d[3] = swab32(rd[0]);
}

static struct crypto_alg srp_algs[2] = { {
	.cra_name		=	"serpent",
	.cra_driver_name	=	"serpent-generic",
	.cra_priority		=	100,
	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
	.cra_blocksize		=	SERPENT_BLOCK_SIZE,
	.cra_ctxsize		=	sizeof(struct serpent_ctx),
	.cra_alignmask		=	3,
	.cra_module		=	THIS_MODULE,
	.cra_u			=	{ .cipher = {
	.cia_min_keysize	=	SERPENT_MIN_KEY_SIZE,
	.cia_max_keysize	=	SERPENT_MAX_KEY_SIZE,
	.cia_setkey		=	serpent_setkey,
	.cia_encrypt		=	serpent_encrypt,
	.cia_decrypt		=	serpent_decrypt } }
}, {
	.cra_name		=	"tnepres",
	.cra_flags		=	CRYPTO_ALG_TYPE_CIPHER,
	.cra_blocksize		=	SERPENT_BLOCK_SIZE,
	.cra_ctxsize		=	sizeof(struct serpent_ctx),
	.cra_alignmask		=	3,
	.cra_module		=	THIS_MODULE,
	.cra_u			=	{ .cipher = {
	.cia_min_keysize	=	SERPENT_MIN_KEY_SIZE,
	.cia_max_keysize	=	SERPENT_MAX_KEY_SIZE,
	.cia_setkey		=	tnepres_setkey,
	.cia_encrypt		=	tnepres_encrypt,
	.cia_decrypt		=	tnepres_decrypt } }
} };

static int __init serpent_mod_init(void)
{
	return crypto_register_algs(srp_algs, ARRAY_SIZE(srp_algs));
}

static void __exit serpent_mod_fini(void)
{
	crypto_unregister_algs(srp_algs, ARRAY_SIZE(srp_algs));
}

module_init(serpent_mod_init);
module_exit(serpent_mod_fini);

MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("Serpent and tnepres (kerneli compatible serpent reversed) Cipher Algorithm");
MODULE_AUTHOR("Dag Arne Osvik <osvik@ii.uib.no>");
MODULE_ALIAS_CRYPTO("tnepres");
MODULE_ALIAS_CRYPTO("serpent");
MODULE_ALIAS_CRYPTO("serpent-generic"