mirror of
https://github.com/reactos/reactos.git
synced 2024-11-01 12:26:32 +00:00
318 lines
7.5 KiB
C
318 lines
7.5 KiB
C
/* rijndael-alg-ref.c v2.0 August '99
|
|
* Reference ANSI C code
|
|
* authors: Paulo Barreto
|
|
* Vincent Rijmen, K.U.Leuven
|
|
*
|
|
* This code is placed in the public domain.
|
|
*/
|
|
|
|
#include "mvOs.h"
|
|
|
|
#include "mvAesAlg.h"
|
|
|
|
#include "mvAesBoxes.dat"
|
|
|
|
|
|
MV_U8 mul1(MV_U8 aa, MV_U8 bb);
|
|
void KeyAddition(MV_U8 a[4][MAXBC], MV_U8 rk[4][MAXBC], MV_U8 BC);
|
|
void ShiftRow128Enc(MV_U8 a[4][MAXBC]);
|
|
void ShiftRow128Dec(MV_U8 a[4][MAXBC]);
|
|
void Substitution(MV_U8 a[4][MAXBC], MV_U8 box[256]);
|
|
void MixColumn(MV_U8 a[4][MAXBC], MV_U8 rk[4][MAXBC]);
|
|
void InvMixColumn(MV_U8 a[4][MAXBC]);
|
|
|
|
|
|
#define mul(aa, bb) (mask[bb] & Alogtable[aa + Logtable[bb]])
|
|
|
|
MV_U8 mul1(MV_U8 aa, MV_U8 bb)
|
|
{
|
|
return mask[bb] & Alogtable[aa + Logtable[bb]];
|
|
}
|
|
|
|
|
|
void KeyAddition(MV_U8 a[4][MAXBC], MV_U8 rk[4][MAXBC], MV_U8 BC)
|
|
{
|
|
/* Exor corresponding text input and round key input bytes
|
|
*/
|
|
((MV_U32*)(&(a[0][0])))[0] ^= ((MV_U32*)(&(rk[0][0])))[0];
|
|
((MV_U32*)(&(a[1][0])))[0] ^= ((MV_U32*)(&(rk[1][0])))[0];
|
|
((MV_U32*)(&(a[2][0])))[0] ^= ((MV_U32*)(&(rk[2][0])))[0];
|
|
((MV_U32*)(&(a[3][0])))[0] ^= ((MV_U32*)(&(rk[3][0])))[0];
|
|
|
|
}
|
|
|
|
void ShiftRow128Enc(MV_U8 a[4][MAXBC]) {
|
|
/* Row 0 remains unchanged
|
|
* The other three rows are shifted a variable amount
|
|
*/
|
|
MV_U8 tmp[MAXBC];
|
|
|
|
tmp[0] = a[1][1];
|
|
tmp[1] = a[1][2];
|
|
tmp[2] = a[1][3];
|
|
tmp[3] = a[1][0];
|
|
|
|
((MV_U32*)(&(a[1][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[1][0] = tmp[0];
|
|
a[1][1] = tmp[1];
|
|
a[1][2] = tmp[2];
|
|
a[1][3] = tmp[3];
|
|
*/
|
|
tmp[0] = a[2][2];
|
|
tmp[1] = a[2][3];
|
|
tmp[2] = a[2][0];
|
|
tmp[3] = a[2][1];
|
|
|
|
((MV_U32*)(&(a[2][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[2][0] = tmp[0];
|
|
a[2][1] = tmp[1];
|
|
a[2][2] = tmp[2];
|
|
a[2][3] = tmp[3];
|
|
*/
|
|
tmp[0] = a[3][3];
|
|
tmp[1] = a[3][0];
|
|
tmp[2] = a[3][1];
|
|
tmp[3] = a[3][2];
|
|
|
|
((MV_U32*)(&(a[3][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[3][0] = tmp[0];
|
|
a[3][1] = tmp[1];
|
|
a[3][2] = tmp[2];
|
|
a[3][3] = tmp[3];
|
|
*/
|
|
}
|
|
|
|
void ShiftRow128Dec(MV_U8 a[4][MAXBC]) {
|
|
/* Row 0 remains unchanged
|
|
* The other three rows are shifted a variable amount
|
|
*/
|
|
MV_U8 tmp[MAXBC];
|
|
|
|
tmp[0] = a[1][3];
|
|
tmp[1] = a[1][0];
|
|
tmp[2] = a[1][1];
|
|
tmp[3] = a[1][2];
|
|
|
|
((MV_U32*)(&(a[1][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[1][0] = tmp[0];
|
|
a[1][1] = tmp[1];
|
|
a[1][2] = tmp[2];
|
|
a[1][3] = tmp[3];
|
|
*/
|
|
|
|
tmp[0] = a[2][2];
|
|
tmp[1] = a[2][3];
|
|
tmp[2] = a[2][0];
|
|
tmp[3] = a[2][1];
|
|
|
|
((MV_U32*)(&(a[2][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[2][0] = tmp[0];
|
|
a[2][1] = tmp[1];
|
|
a[2][2] = tmp[2];
|
|
a[2][3] = tmp[3];
|
|
*/
|
|
|
|
tmp[0] = a[3][1];
|
|
tmp[1] = a[3][2];
|
|
tmp[2] = a[3][3];
|
|
tmp[3] = a[3][0];
|
|
|
|
((MV_U32*)(&(a[3][0])))[0] = ((MV_U32*)(&(tmp[0])))[0];
|
|
/*
|
|
a[3][0] = tmp[0];
|
|
a[3][1] = tmp[1];
|
|
a[3][2] = tmp[2];
|
|
a[3][3] = tmp[3];
|
|
*/
|
|
}
|
|
|
|
void Substitution(MV_U8 a[4][MAXBC], MV_U8 box[256]) {
|
|
/* Replace every byte of the input by the byte at that place
|
|
* in the nonlinear S-box
|
|
*/
|
|
int i, j;
|
|
|
|
for(i = 0; i < 4; i++)
|
|
for(j = 0; j < 4; j++) a[i][j] = box[a[i][j]] ;
|
|
}
|
|
|
|
void MixColumn(MV_U8 a[4][MAXBC], MV_U8 rk[4][MAXBC]) {
|
|
/* Mix the four bytes of every column in a linear way
|
|
*/
|
|
MV_U8 b[4][MAXBC];
|
|
int i, j;
|
|
|
|
for(j = 0; j < 4; j++){
|
|
b[0][j] = mul(25,a[0][j]) ^ mul(1,a[1][j]) ^ a[2][j] ^ a[3][j];
|
|
b[1][j] = mul(25,a[1][j]) ^ mul(1,a[2][j]) ^ a[3][j] ^ a[0][j];
|
|
b[2][j] = mul(25,a[2][j]) ^ mul(1,a[3][j]) ^ a[0][j] ^ a[1][j];
|
|
b[3][j] = mul(25,a[3][j]) ^ mul(1,a[0][j]) ^ a[1][j] ^ a[2][j];
|
|
}
|
|
for(i = 0; i < 4; i++)
|
|
/*for(j = 0; j < BC; j++) a[i][j] = b[i][j];*/
|
|
((MV_U32*)(&(a[i][0])))[0] = ((MV_U32*)(&(b[i][0])))[0] ^ ((MV_U32*)(&(rk[i][0])))[0];
|
|
}
|
|
|
|
void InvMixColumn(MV_U8 a[4][MAXBC]) {
|
|
/* Mix the four bytes of every column in a linear way
|
|
* This is the opposite operation of Mixcolumn
|
|
*/
|
|
MV_U8 b[4][MAXBC];
|
|
int i, j;
|
|
|
|
for(j = 0; j < 4; j++){
|
|
b[0][j] = mul(223,a[0][j]) ^ mul(104,a[1][j]) ^ mul(238,a[2][j]) ^ mul(199,a[3][j]);
|
|
b[1][j] = mul(223,a[1][j]) ^ mul(104,a[2][j]) ^ mul(238,a[3][j]) ^ mul(199,a[0][j]);
|
|
b[2][j] = mul(223,a[2][j]) ^ mul(104,a[3][j]) ^ mul(238,a[0][j]) ^ mul(199,a[1][j]);
|
|
b[3][j] = mul(223,a[3][j]) ^ mul(104,a[0][j]) ^ mul(238,a[1][j]) ^ mul(199,a[2][j]);
|
|
}
|
|
for(i = 0; i < 4; i++)
|
|
/*for(j = 0; j < BC; j++) a[i][j] = b[i][j];*/
|
|
((MV_U32*)(&(a[i][0])))[0] = ((MV_U32*)(&(b[i][0])))[0];
|
|
}
|
|
|
|
int rijndaelKeySched (MV_U8 k[4][MAXKC], int keyBits, int blockBits, MV_U8 W[MAXROUNDS+1][4][MAXBC])
|
|
{
|
|
/* Calculate the necessary round keys
|
|
* The number of calculations depends on keyBits and blockBits
|
|
*/
|
|
int KC, BC, ROUNDS;
|
|
int i, j, t, rconpointer = 0;
|
|
MV_U8 tk[4][MAXKC];
|
|
|
|
switch (keyBits) {
|
|
case 128: KC = 4; break;
|
|
case 192: KC = 6; break;
|
|
case 256: KC = 8; break;
|
|
default : return (-1);
|
|
}
|
|
|
|
switch (blockBits) {
|
|
case 128: BC = 4; break;
|
|
case 192: BC = 6; break;
|
|
case 256: BC = 8; break;
|
|
default : return (-2);
|
|
}
|
|
|
|
switch (keyBits >= blockBits ? keyBits : blockBits) {
|
|
case 128: ROUNDS = 10; break;
|
|
case 192: ROUNDS = 12; break;
|
|
case 256: ROUNDS = 14; break;
|
|
default : return (-3); /* this cannot happen */
|
|
}
|
|
|
|
|
|
for(j = 0; j < KC; j++)
|
|
for(i = 0; i < 4; i++)
|
|
tk[i][j] = k[i][j];
|
|
t = 0;
|
|
/* copy values into round key array */
|
|
for(j = 0; (j < KC) && (t < (ROUNDS+1)*BC); j++, t++)
|
|
for(i = 0; i < 4; i++) W[t / BC][i][t % BC] = tk[i][j];
|
|
|
|
while (t < (ROUNDS+1)*BC) { /* while not enough round key material calculated */
|
|
/* calculate new values */
|
|
for(i = 0; i < 4; i++)
|
|
tk[i][0] ^= S[tk[(i+1)%4][KC-1]];
|
|
tk[0][0] ^= rcon[rconpointer++];
|
|
|
|
if (KC != 8)
|
|
for(j = 1; j < KC; j++)
|
|
for(i = 0; i < 4; i++) tk[i][j] ^= tk[i][j-1];
|
|
else {
|
|
for(j = 1; j < KC/2; j++)
|
|
for(i = 0; i < 4; i++) tk[i][j] ^= tk[i][j-1];
|
|
for(i = 0; i < 4; i++) tk[i][KC/2] ^= S[tk[i][KC/2 - 1]];
|
|
for(j = KC/2 + 1; j < KC; j++)
|
|
for(i = 0; i < 4; i++) tk[i][j] ^= tk[i][j-1];
|
|
}
|
|
/* copy values into round key array */
|
|
for(j = 0; (j < KC) && (t < (ROUNDS+1)*BC); j++, t++)
|
|
for(i = 0; i < 4; i++) W[t / BC][i][t % BC] = tk[i][j];
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
|
|
|
|
int rijndaelEncrypt128(MV_U8 a[4][MAXBC], MV_U8 rk[MAXROUNDS+1][4][MAXBC], int rounds)
|
|
{
|
|
/* Encryption of one block.
|
|
*/
|
|
int r, BC, ROUNDS;
|
|
|
|
BC = 4;
|
|
ROUNDS = rounds;
|
|
|
|
/* begin with a key addition
|
|
*/
|
|
|
|
KeyAddition(a,rk[0],BC);
|
|
|
|
/* ROUNDS-1 ordinary rounds
|
|
*/
|
|
for(r = 1; r < ROUNDS; r++) {
|
|
Substitution(a,S);
|
|
ShiftRow128Enc(a);
|
|
MixColumn(a, rk[r]);
|
|
/*KeyAddition(a,rk[r],BC);*/
|
|
}
|
|
|
|
/* Last round is special: there is no MixColumn
|
|
*/
|
|
Substitution(a,S);
|
|
ShiftRow128Enc(a);
|
|
KeyAddition(a,rk[ROUNDS],BC);
|
|
|
|
return 0;
|
|
}
|
|
|
|
|
|
int rijndaelDecrypt128(MV_U8 a[4][MAXBC], MV_U8 rk[MAXROUNDS+1][4][MAXBC], int rounds)
|
|
{
|
|
int r, BC, ROUNDS;
|
|
|
|
BC = 4;
|
|
ROUNDS = rounds;
|
|
|
|
/* To decrypt: apply the inverse operations of the encrypt routine,
|
|
* in opposite order
|
|
*
|
|
* (KeyAddition is an involution: it 's equal to its inverse)
|
|
* (the inverse of Substitution with table S is Substitution with the inverse table of S)
|
|
* (the inverse of Shiftrow is Shiftrow over a suitable distance)
|
|
*/
|
|
|
|
/* First the special round:
|
|
* without InvMixColumn
|
|
* with extra KeyAddition
|
|
*/
|
|
KeyAddition(a,rk[ROUNDS],BC);
|
|
ShiftRow128Dec(a);
|
|
Substitution(a,Si);
|
|
|
|
/* ROUNDS-1 ordinary rounds
|
|
*/
|
|
for(r = ROUNDS-1; r > 0; r--) {
|
|
KeyAddition(a,rk[r],BC);
|
|
InvMixColumn(a);
|
|
ShiftRow128Dec(a);
|
|
Substitution(a,Si);
|
|
|
|
}
|
|
|
|
/* End with the extra key addition
|
|
*/
|
|
|
|
KeyAddition(a,rk[0],BC);
|
|
|
|
return 0;
|
|
}
|
|
|