#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
+#include <linux/bitops.h>
+#include <asm/unaligned.h>
static const u32 camellia_sp1110[256] = {
0x70707000,0x82828200,0x2c2c2c00,0xececec00,
#define CAMELLIA_BLOCK_SIZE 16
#define CAMELLIA_TABLE_BYTE_LEN 272
+/*
+ * NB: L and R below stand for 'left' and 'right' as in written numbers.
+ * That is, in (xxxL,xxxR) pair xxxL holds most significant digits,
+ * _not_ least significant ones!
+ */
+
/* key constants */
/*
* macros
*/
-
-# define GETU32(v, pt) \
- do { \
- /* latest breed of gcc is clever enough to use move */ \
- memcpy(&(v), (pt), 4); \
- (v) = be32_to_cpu(v); \
- } while(0)
-
-/* rotation right shift 1byte */
-#define ROR8(x) (((x) >> 8) + ((x) << 24))
-/* rotation left shift 1bit */
-#define ROL1(x) (((x) << 1) + ((x) >> 31))
-/* rotation left shift 1byte */
-#define ROL8(x) (((x) << 8) + ((x) >> 24))
-
#define ROLDQ(ll, lr, rl, rr, w0, w1, bits) \
do { \
w0 = ll; \
rr = (w0 << (bits - 32)) + (w1 >> (64 - bits)); \
} while(0)
-
#define CAMELLIA_F(xl, xr, kl, kr, yl, yr, il, ir, t0, t1) \
do { \
il = xl ^ kl; \
ir = xr ^ kr; \
t0 = il >> 16; \
t1 = ir >> 16; \
- yl = camellia_sp1110[ir & 0xff] \
- ^ camellia_sp0222[(t1 >> 8) & 0xff] \
- ^ camellia_sp3033[t1 & 0xff] \
- ^ camellia_sp4404[(ir >> 8) & 0xff]; \
- yr = camellia_sp1110[(t0 >> 8) & 0xff] \
- ^ camellia_sp0222[t0 & 0xff] \
- ^ camellia_sp3033[(il >> 8) & 0xff] \
- ^ camellia_sp4404[il & 0xff]; \
+ yl = camellia_sp1110[(u8)(ir )] \
+ ^ camellia_sp0222[ (t1 >> 8)] \
+ ^ camellia_sp3033[(u8)(t1 )] \
+ ^ camellia_sp4404[(u8)(ir >> 8)]; \
+ yr = camellia_sp1110[ (t0 >> 8)] \
+ ^ camellia_sp0222[(u8)(t0 )] \
+ ^ camellia_sp3033[(u8)(il >> 8)] \
+ ^ camellia_sp4404[(u8)(il )]; \
yl ^= yr; \
- yr = ROR8(yr); \
+ yr = ror32(yr, 8); \
yr ^= yl; \
} while(0)
-
-/*
- * for speed up
- *
- */
-#define CAMELLIA_FLS(ll, lr, rl, rr, kll, klr, krl, krr, t0, t1, t2, t3) \
- do { \
- t0 = kll; \
- t2 = krr; \
- t0 &= ll; \
- t2 |= rr; \
- rl ^= t2; \
- lr ^= ROL1(t0); \
- t3 = krl; \
- t1 = klr; \
- t3 &= rl; \
- t1 |= lr; \
- ll ^= t1; \
- rr ^= ROL1(t3); \
- } while(0)
-
-#define CAMELLIA_ROUNDSM(xl, xr, kl, kr, yl, yr, il, ir, t0, t1) \
- do { \
- ir = camellia_sp1110[xr & 0xff]; \
- il = camellia_sp1110[(xl>>24) & 0xff]; \
- ir ^= camellia_sp0222[(xr>>24) & 0xff]; \
- il ^= camellia_sp0222[(xl>>16) & 0xff]; \
- ir ^= camellia_sp3033[(xr>>16) & 0xff]; \
- il ^= camellia_sp3033[(xl>>8) & 0xff]; \
- ir ^= camellia_sp4404[(xr>>8) & 0xff]; \
- il ^= camellia_sp4404[xl & 0xff]; \
- il ^= kl; \
- ir ^= il ^ kr; \
- yl ^= ir; \
- yr ^= ROR8(il) ^ ir; \
- } while(0)
-
-
#define SUBKEY_L(INDEX) (subkey[(INDEX)*2])
#define SUBKEY_R(INDEX) (subkey[(INDEX)*2 + 1])
-static void camellia_setup128(const unsigned char *key, u32 *subkey)
+static void camellia_setup_tail(u32 *subkey, u32 *subL, u32 *subR, int max)
{
- u32 kll, klr, krl, krr;
- u32 il, ir, t0, t1, w0, w1;
- u32 kw4l, kw4r, dw, tl, tr;
- u32 subL[26];
- u32 subR[26];
-
- /**
- * k == kll || klr || krl || krr (|| is concatination)
- */
- GETU32(kll, key );
- GETU32(klr, key + 4);
- GETU32(krl, key + 8);
- GETU32(krr, key + 12);
-
- /**
- * generate KL dependent subkeys
- */
- /* kw1 */
- subL[0] = kll; subR[0] = klr;
- /* kw2 */
- subL[1] = krl; subR[1] = krr;
- /* rotation left shift 15bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* k3 */
- subL[4] = kll; subR[4] = klr;
- /* k4 */
- subL[5] = krl; subR[5] = krr;
- /* rotation left shift 15+30bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 30);
- /* k7 */
- subL[10] = kll; subR[10] = klr;
- /* k8 */
- subL[11] = krl; subR[11] = krr;
- /* rotation left shift 15+30+15bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* k10 */
- subL[13] = krl; subR[13] = krr;
- /* rotation left shift 15+30+15+17 bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 17);
- /* kl3 */
- subL[16] = kll; subR[16] = klr;
- /* kl4 */
- subL[17] = krl; subR[17] = krr;
- /* rotation left shift 15+30+15+17+17 bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 17);
- /* k13 */
- subL[18] = kll; subR[18] = klr;
- /* k14 */
- subL[19] = krl; subR[19] = krr;
- /* rotation left shift 15+30+15+17+17+17 bit */
- ROLDQ(kll, klr, krl, krr, w0, w1, 17);
- /* k17 */
- subL[22] = kll; subR[22] = klr;
- /* k18 */
- subL[23] = krl; subR[23] = krr;
-
- /* generate KA */
- kll = subL[0]; klr = subR[0];
- krl = subL[1]; krr = subR[1];
- CAMELLIA_F(kll, klr,
- CAMELLIA_SIGMA1L, CAMELLIA_SIGMA1R,
- w0, w1, il, ir, t0, t1);
- krl ^= w0; krr ^= w1;
- CAMELLIA_F(krl, krr,
- CAMELLIA_SIGMA2L, CAMELLIA_SIGMA2R,
- kll, klr, il, ir, t0, t1);
- /* current status == (kll, klr, w0, w1) */
- CAMELLIA_F(kll, klr,
- CAMELLIA_SIGMA3L, CAMELLIA_SIGMA3R,
- krl, krr, il, ir, t0, t1);
- krl ^= w0; krr ^= w1;
- CAMELLIA_F(krl, krr,
- CAMELLIA_SIGMA4L, CAMELLIA_SIGMA4R,
- w0, w1, il, ir, t0, t1);
- kll ^= w0; klr ^= w1;
-
- /* generate KA dependent subkeys */
- /* k1, k2 */
- subL[2] = kll; subR[2] = klr;
- subL[3] = krl; subR[3] = krr;
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* k5,k6 */
- subL[6] = kll; subR[6] = klr;
- subL[7] = krl; subR[7] = krr;
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* kl1, kl2 */
- subL[8] = kll; subR[8] = klr;
- subL[9] = krl; subR[9] = krr;
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* k9 */
- subL[12] = kll; subR[12] = klr;
- ROLDQ(kll, klr, krl, krr, w0, w1, 15);
- /* k11, k12 */
- subL[14] = kll; subR[14] = klr;
- subL[15] = krl; subR[15] = krr;
- ROLDQo32(kll, klr, krl, krr, w0, w1, 34);
- /* k15, k16 */
- subL[20] = kll; subR[20] = klr;
- subL[21] = krl; subR[21] = krr;
- ROLDQ(kll, klr, krl, krr, w0, w1, 17);
- /* kw3, kw4 */
- subL[24] = kll; subR[24] = klr;
- subL[25] = krl; subR[25] = krr;
+ u32 dw, tl, tr;
+ u32 kw4l, kw4r;
+ int i;
/* absorb kw2 to other subkeys */
/* round 2 */
subL[7] ^= subL[1]; subR[7] ^= subR[1];
subL[1] ^= subR[1] & ~subR[9];
dw = subL[1] & subL[9],
- subR[1] ^= ROL1(dw); /* modified for FLinv(kl2) */
+ subR[1] ^= rol32(dw, 1); /* modified for FLinv(kl2) */
/* round 8 */
subL[11] ^= subL[1]; subR[11] ^= subR[1];
/* round 10 */
subL[15] ^= subL[1]; subR[15] ^= subR[1];
subL[1] ^= subR[1] & ~subR[17];
dw = subL[1] & subL[17],
- subR[1] ^= ROL1(dw); /* modified for FLinv(kl4) */
+ subR[1] ^= rol32(dw, 1); /* modified for FLinv(kl4) */
/* round 14 */
subL[19] ^= subL[1]; subR[19] ^= subR[1];
/* round 16 */
subL[21] ^= subL[1]; subR[21] ^= subR[1];
/* round 18 */
subL[23] ^= subL[1]; subR[23] ^= subR[1];
- /* kw3 */
- subL[24] ^= subL[1]; subR[24] ^= subR[1];
+ if (max == 24) {
+ /* kw3 */
+ subL[24] ^= subL[1]; subR[24] ^= subR[1];
/* absorb kw4 to other subkeys */
- kw4l = subL[25]; kw4r = subR[25];
+ kw4l = subL[25]; kw4r = subR[25];
+ } else {
+ subL[1] ^= subR[1] & ~subR[25];
+ dw = subL[1] & subL[25],
+ subR[1] ^= rol32(dw, 1); /* modified for FLinv(kl6) */
+ /* round 20 */
+ subL[27] ^= subL[1]; subR[27] ^= subR[1];
+ /* round 22 */
+ subL[29] ^= subL[1]; subR[29] ^= subR[1];
+ /* round 24 */
+ subL[31] ^= subL[1]; subR[31] ^= subR[1];
+ /* kw3 */
+ subL[32] ^= subL[1]; subR[32] ^= subR[1];
+
+ /* absorb kw4 to other subkeys */
+ kw4l = subL[33]; kw4r = subR[33];
+ /* round 23 */
+ subL[30] ^= kw4l; subR[30] ^= kw4r;
+ /* round 21 */
+ subL[28] ^= kw4l; subR[28] ^= kw4r;
+ /* round 19 */
+ subL[26] ^= kw4l; subR[26] ^= kw4r;
+ kw4l ^= kw4r & ~subR[24];
+ dw = kw4l & subL[24],
+ kw4r ^= rol32(dw, 1); /* modified for FL(kl5) */
+ }
/* round 17 */
subL[22] ^= kw4l; subR[22] ^= kw4r;
/* round 15 */
subL[18] ^= kw4l; subR[18] ^= kw4r;
kw4l ^= kw4r & ~subR[16];
dw = kw4l & subL[16],
- kw4r ^= ROL1(dw); /* modified for FL(kl3) */
+ kw4r ^= rol32(dw, 1); /* modified for FL(kl3) */
/* round 11 */
subL[14] ^= kw4l; subR[14] ^= kw4r;
/* round 9 */
subL[10] ^= kw4l; subR[10] ^= kw4r;
kw4l ^= kw4r & ~subR[8];
dw = kw4l & subL[8],
- kw4r ^= ROL1(dw); /* modified for FL(kl1) */
+ kw4r ^= rol32(dw, 1); /* modified for FL(kl1) */
/* round 5 */
subL[6] ^= kw4l; subR[6] ^= kw4r;
/* round 3 */
SUBKEY_R(6) = subR[5] ^ subR[7];
tl = subL[10] ^ (subR[10] & ~subR[8]);
dw = tl & subL[8], /* FL(kl1) */
- tr = subR[10] ^ ROL1(dw);
+ tr = subR[10] ^ rol32(dw, 1);
SUBKEY_L(7) = subL[6] ^ tl; /* round 6 */
SUBKEY_R(7) = subR[6] ^ tr;
SUBKEY_L(8) = subL[8]; /* FL(kl1) */
SUBKEY_R(9) = subR[9];
tl = subL[7] ^ (subR[7] & ~subR[9]);
dw = tl & subL[9], /* FLinv(kl2) */
- tr = subR[7] ^ ROL1(dw);
+ tr = subR[7] ^ rol32(dw, 1);
SUBKEY_L(10) = tl ^ subL[11]; /* round 7 */
SUBKEY_R(10) = tr ^ subR[11];
SUBKEY_L(11) = subL[10] ^ subL[12]; /* round 8 */
SUBKEY_R(14) = subR[13] ^ subR[15];
tl = subL[18] ^ (subR[18] & ~subR[16]);
dw = tl & subL[16], /* FL(kl3) */
- tr = subR[18] ^ ROL1(dw);
+ tr = subR[18] ^ rol32(dw, 1);
SUBKEY_L(15) = subL[14] ^ tl; /* round 12 */
SUBKEY_R(15) = subR[14] ^ tr;
SUBKEY_L(16) = subL[16]; /* FL(kl3) */
SUBKEY_R(17) = subR[17];
tl = subL[15] ^ (subR[15] & ~subR[17]);
dw = tl & subL[17], /* FLinv(kl4) */
- tr = subR[15] ^ ROL1(dw);
+ tr = subR[15] ^ rol32(dw, 1);
SUBKEY_L(18) = tl ^ subL[19]; /* round 13 */
SUBKEY_R(18) = tr ^ subR[19];
SUBKEY_L(19) = subL[18] ^ subL[20]; /* round 14 */
SUBKEY_R(21) = subR[20] ^ subR[22];
SUBKEY_L(22) = subL[21] ^ subL[23]; /* round 17 */
SUBKEY_R(22) = subR[21] ^ subR[23];
- SUBKEY_L(23) = subL[22]; /* round 18 */
- SUBKEY_R(23) = subR[22];
- SUBKEY_L(24) = subL[24] ^ subL[23]; /* kw3 */
- SUBKEY_R(24) = subR[24] ^ subR[23];
+ if (max == 24) {
+ SUBKEY_L(23) = subL[22]; /* round 18 */
+ SUBKEY_R(23) = subR[22];
+ SUBKEY_L(24) = subL[24] ^ subL[23]; /* kw3 */
+ SUBKEY_R(24) = subR[24] ^ subR[23];
+ } else {
+ tl = subL[26] ^ (subR[26] & ~subR[24]);
+ dw = tl & subL[24], /* FL(kl5) */
+ tr = subR[26] ^ rol32(dw, 1);
+ SUBKEY_L(23) = subL[22] ^ tl; /* round 18 */
+ SUBKEY_R(23) = subR[22] ^ tr;
+ SUBKEY_L(24) = subL[24]; /* FL(kl5) */
+ SUBKEY_R(24) = subR[24];
+ SUBKEY_L(25) = subL[25]; /* FLinv(kl6) */
+ SUBKEY_R(25) = subR[25];
+ tl = subL[23] ^ (subR[23] & ~subR[25]);
+ dw = tl & subL[25], /* FLinv(kl6) */
+ tr = subR[23] ^ rol32(dw, 1);
+ SUBKEY_L(26) = tl ^ subL[27]; /* round 19 */
+ SUBKEY_R(26) = tr ^ subR[27];
+ SUBKEY_L(27) = subL[26] ^ subL[28]; /* round 20 */
+ SUBKEY_R(27) = subR[26] ^ subR[28];
+ SUBKEY_L(28) = subL[27] ^ subL[29]; /* round 21 */
+ SUBKEY_R(28) = subR[27] ^ subR[29];
+ SUBKEY_L(29) = subL[28] ^ subL[30]; /* round 22 */
+ SUBKEY_R(29) = subR[28] ^ subR[30];
+ SUBKEY_L(30) = subL[29] ^ subL[31]; /* round 23 */
+ SUBKEY_R(30) = subR[29] ^ subR[31];
+ SUBKEY_L(31) = subL[30]; /* round 24 */
+ SUBKEY_R(31) = subR[30];
+ SUBKEY_L(32) = subL[32] ^ subL[31]; /* kw3 */
+ SUBKEY_R(32) = subR[32] ^ subR[31];
+ }
/* apply the inverse of the last half of P-function */
- dw = SUBKEY_L(2) ^ SUBKEY_R(2); dw = ROL8(dw);/* round 1 */
- SUBKEY_R(2) = SUBKEY_L(2) ^ dw; SUBKEY_L(2) = dw;
- dw = SUBKEY_L(3) ^ SUBKEY_R(3); dw = ROL8(dw);/* round 2 */
- SUBKEY_R(3) = SUBKEY_L(3) ^ dw; SUBKEY_L(3) = dw;
- dw = SUBKEY_L(4) ^ SUBKEY_R(4); dw = ROL8(dw);/* round 3 */
- SUBKEY_R(4) = SUBKEY_L(4) ^ dw; SUBKEY_L(4) = dw;
- dw = SUBKEY_L(5) ^ SUBKEY_R(5); dw = ROL8(dw);/* round 4 */
- SUBKEY_R(5) = SUBKEY_L(5) ^ dw; SUBKEY_L(5) = dw;
- dw = SUBKEY_L(6) ^ SUBKEY_R(6); dw = ROL8(dw);/* round 5 */
- SUBKEY_R(6) = SUBKEY_L(6) ^ dw; SUBKEY_L(6) = dw;
- dw = SUBKEY_L(7) ^ SUBKEY_R(7); dw = ROL8(dw);/* round 6 */
- SUBKEY_R(7) = SUBKEY_L(7) ^ dw; SUBKEY_L(7) = dw;
- dw = SUBKEY_L(10) ^ SUBKEY_R(10); dw = ROL8(dw);/* round 7 */
- SUBKEY_R(10) = SUBKEY_L(10) ^ dw; SUBKEY_L(10) = dw;
- dw = SUBKEY_L(11) ^ SUBKEY_R(11); dw = ROL8(dw);/* round 8 */
- SUBKEY_R(11) = SUBKEY_L(11) ^ dw; SUBKEY_L(11) = dw;
- dw = SUBKEY_L(12) ^ SUBKEY_R(12); dw = ROL8(dw);/* round 9 */
- SUBKEY_R(12) = SUBKEY_L(12) ^ dw; SUBKEY_L(12) = dw;
- dw = SUBKEY_L(13) ^ SUBKEY_R(13); dw = ROL8(dw);/* round 10 */
- SUBKEY_R(13) = SUBKEY_L(13) ^ dw; SUBKEY_L(13) = dw;
- dw = SUBKEY_L(14) ^ SUBKEY_R(14); dw = ROL8(dw);/* round 11 */
- SUBKEY_R(14) = SUBKEY_L(14) ^ dw; SUBKEY_L(14) = dw;
- dw = SUBKEY_L(15) ^ SUBKEY_R(15); dw = ROL8(dw);/* round 12 */
- SUBKEY_R(15) = SUBKEY_L(15) ^ dw; SUBKEY_L(15) = dw;
- dw = SUBKEY_L(18) ^ SUBKEY_R(18); dw = ROL8(dw);/* round 13 */
- SUBKEY_R(18) = SUBKEY_L(18) ^ dw; SUBKEY_L(18) = dw;
- dw = SUBKEY_L(19) ^ SUBKEY_R(19); dw = ROL8(dw);/* round 14 */
- SUBKEY_R(19) = SUBKEY_L(19) ^ dw; SUBKEY_L(19) = dw;
- dw = SUBKEY_L(20) ^ SUBKEY_R(20); dw = ROL8(dw);/* round 15 */
- SUBKEY_R(20) = SUBKEY_L(20) ^ dw; SUBKEY_L(20) = dw;
- dw = SUBKEY_L(21) ^ SUBKEY_R(21); dw = ROL8(dw);/* round 16 */
- SUBKEY_R(21) = SUBKEY_L(21) ^ dw; SUBKEY_L(21) = dw;
- dw = SUBKEY_L(22) ^ SUBKEY_R(22); dw = ROL8(dw);/* round 17 */
- SUBKEY_R(22) = SUBKEY_L(22) ^ dw; SUBKEY_L(22) = dw;
- dw = SUBKEY_L(23) ^ SUBKEY_R(23); dw = ROL8(dw);/* round 18 */
- SUBKEY_R(23) = SUBKEY_L(23) ^ dw; SUBKEY_L(23) = dw;
+ i = 2;
+ do {
+ dw = SUBKEY_L(i + 0) ^ SUBKEY_R(i + 0); dw = rol32(dw, 8);/* round 1 */
+ SUBKEY_R(i + 0) = SUBKEY_L(i + 0) ^ dw; SUBKEY_L(i + 0) = dw;
+ dw = SUBKEY_L(i + 1) ^ SUBKEY_R(i + 1); dw = rol32(dw, 8);/* round 2 */
+ SUBKEY_R(i + 1) = SUBKEY_L(i + 1) ^ dw; SUBKEY_L(i + 1) = dw;
+ dw = SUBKEY_L(i + 2) ^ SUBKEY_R(i + 2); dw = rol32(dw, 8);/* round 3 */
+ SUBKEY_R(i + 2) = SUBKEY_L(i + 2) ^ dw; SUBKEY_L(i + 2) = dw;
+ dw = SUBKEY_L(i + 3) ^ SUBKEY_R(i + 3); dw = rol32(dw, 8);/* round 4 */
+ SUBKEY_R(i + 3) = SUBKEY_L(i + 3) ^ dw; SUBKEY_L(i + 3) = dw;
+ dw = SUBKEY_L(i + 4) ^ SUBKEY_R(i + 4); dw = rol32(dw, 8);/* round 5 */
+ SUBKEY_R(i + 4) = SUBKEY_L(i + 4) ^ dw; SUBKEY_L(i + 4) = dw;
+ dw = SUBKEY_L(i + 5) ^ SUBKEY_R(i + 5); dw = rol32(dw, 8);/* round 6 */
+ SUBKEY_R(i + 5) = SUBKEY_L(i + 5) ^ dw; SUBKEY_L(i + 5) = dw;
+ i += 8;
+ } while (i < max);
+}
+
+static void camellia_setup128(const unsigned char *key, u32 *subkey)
+{
+ u32 kll, klr, krl, krr;
+ u32 il, ir, t0, t1, w0, w1;
+ u32 subL[26];
+ u32 subR[26];
+
+ /**
+ * k == kll || klr || krl || krr (|| is concatenation)
+ */
+ kll = get_unaligned_be32(key);
+ klr = get_unaligned_be32(key + 4);
+ krl = get_unaligned_be32(key + 8);
+ krr = get_unaligned_be32(key + 12);
+
+ /* generate KL dependent subkeys */
+ /* kw1 */
+ subL[0] = kll; subR[0] = klr;
+ /* kw2 */
+ subL[1] = krl; subR[1] = krr;
+ /* rotation left shift 15bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* k3 */
+ subL[4] = kll; subR[4] = klr;
+ /* k4 */
+ subL[5] = krl; subR[5] = krr;
+ /* rotation left shift 15+30bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 30);
+ /* k7 */
+ subL[10] = kll; subR[10] = klr;
+ /* k8 */
+ subL[11] = krl; subR[11] = krr;
+ /* rotation left shift 15+30+15bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* k10 */
+ subL[13] = krl; subR[13] = krr;
+ /* rotation left shift 15+30+15+17 bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 17);
+ /* kl3 */
+ subL[16] = kll; subR[16] = klr;
+ /* kl4 */
+ subL[17] = krl; subR[17] = krr;
+ /* rotation left shift 15+30+15+17+17 bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 17);
+ /* k13 */
+ subL[18] = kll; subR[18] = klr;
+ /* k14 */
+ subL[19] = krl; subR[19] = krr;
+ /* rotation left shift 15+30+15+17+17+17 bit */
+ ROLDQ(kll, klr, krl, krr, w0, w1, 17);
+ /* k17 */
+ subL[22] = kll; subR[22] = klr;
+ /* k18 */
+ subL[23] = krl; subR[23] = krr;
+
+ /* generate KA */
+ kll = subL[0]; klr = subR[0];
+ krl = subL[1]; krr = subR[1];
+ CAMELLIA_F(kll, klr,
+ CAMELLIA_SIGMA1L, CAMELLIA_SIGMA1R,
+ w0, w1, il, ir, t0, t1);
+ krl ^= w0; krr ^= w1;
+ CAMELLIA_F(krl, krr,
+ CAMELLIA_SIGMA2L, CAMELLIA_SIGMA2R,
+ kll, klr, il, ir, t0, t1);
+ /* current status == (kll, klr, w0, w1) */
+ CAMELLIA_F(kll, klr,
+ CAMELLIA_SIGMA3L, CAMELLIA_SIGMA3R,
+ krl, krr, il, ir, t0, t1);
+ krl ^= w0; krr ^= w1;
+ CAMELLIA_F(krl, krr,
+ CAMELLIA_SIGMA4L, CAMELLIA_SIGMA4R,
+ w0, w1, il, ir, t0, t1);
+ kll ^= w0; klr ^= w1;
+
+ /* generate KA dependent subkeys */
+ /* k1, k2 */
+ subL[2] = kll; subR[2] = klr;
+ subL[3] = krl; subR[3] = krr;
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* k5,k6 */
+ subL[6] = kll; subR[6] = klr;
+ subL[7] = krl; subR[7] = krr;
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* kl1, kl2 */
+ subL[8] = kll; subR[8] = klr;
+ subL[9] = krl; subR[9] = krr;
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* k9 */
+ subL[12] = kll; subR[12] = klr;
+ ROLDQ(kll, klr, krl, krr, w0, w1, 15);
+ /* k11, k12 */
+ subL[14] = kll; subR[14] = klr;
+ subL[15] = krl; subR[15] = krr;
+ ROLDQo32(kll, klr, krl, krr, w0, w1, 34);
+ /* k15, k16 */
+ subL[20] = kll; subR[20] = klr;
+ subL[21] = krl; subR[21] = krr;
+ ROLDQ(kll, klr, krl, krr, w0, w1, 17);
+ /* kw3, kw4 */
+ subL[24] = kll; subR[24] = klr;
+ subL[25] = krl; subR[25] = krr;
+
+ camellia_setup_tail(subkey, subL, subR, 24);
}
static void camellia_setup256(const unsigned char *key, u32 *subkey)
u32 kll, klr, krl, krr; /* left half of key */
u32 krll, krlr, krrl, krrr; /* right half of key */
u32 il, ir, t0, t1, w0, w1; /* temporary variables */
- u32 kw4l, kw4r, dw, tl, tr;
u32 subL[34];
u32 subR[34];
/**
* key = (kll || klr || krl || krr || krll || krlr || krrl || krrr)
- * (|| is concatination)
+ * (|| is concatenation)
*/
- GETU32(kll, key );
- GETU32(klr, key + 4);
- GETU32(krl, key + 8);
- GETU32(krr, key + 12);
- GETU32(krll, key + 16);
- GETU32(krlr, key + 20);
- GETU32(krrl, key + 24);
- GETU32(krrr, key + 28);
+ kll = get_unaligned_be32(key);
+ klr = get_unaligned_be32(key + 4);
+ krl = get_unaligned_be32(key + 8);
+ krr = get_unaligned_be32(key + 12);
+ krll = get_unaligned_be32(key + 16);
+ krlr = get_unaligned_be32(key + 20);
+ krrl = get_unaligned_be32(key + 24);
+ krrr = get_unaligned_be32(key + 28);
/* generate KL dependent subkeys */
/* kw1 */
/* kw4 */
subL[33] = krrl; subR[33] = krrr;
- /* absorb kw2 to other subkeys */
- /* round 2 */
- subL[3] ^= subL[1]; subR[3] ^= subR[1];
- /* round 4 */
- subL[5] ^= subL[1]; subR[5] ^= subR[1];
- /* round 6 */
- subL[7] ^= subL[1]; subR[7] ^= subR[1];
- subL[1] ^= subR[1] & ~subR[9];
- dw = subL[1] & subL[9],
- subR[1] ^= ROL1(dw); /* modified for FLinv(kl2) */
- /* round 8 */
- subL[11] ^= subL[1]; subR[11] ^= subR[1];
- /* round 10 */
- subL[13] ^= subL[1]; subR[13] ^= subR[1];
- /* round 12 */
- subL[15] ^= subL[1]; subR[15] ^= subR[1];
- subL[1] ^= subR[1] & ~subR[17];
- dw = subL[1] & subL[17],
- subR[1] ^= ROL1(dw); /* modified for FLinv(kl4) */
- /* round 14 */
- subL[19] ^= subL[1]; subR[19] ^= subR[1];
- /* round 16 */
- subL[21] ^= subL[1]; subR[21] ^= subR[1];
- /* round 18 */
- subL[23] ^= subL[1]; subR[23] ^= subR[1];
- subL[1] ^= subR[1] & ~subR[25];
- dw = subL[1] & subL[25],
- subR[1] ^= ROL1(dw); /* modified for FLinv(kl6) */
- /* round 20 */
- subL[27] ^= subL[1]; subR[27] ^= subR[1];
- /* round 22 */
- subL[29] ^= subL[1]; subR[29] ^= subR[1];
- /* round 24 */
- subL[31] ^= subL[1]; subR[31] ^= subR[1];
- /* kw3 */
- subL[32] ^= subL[1]; subR[32] ^= subR[1];
-
- /* absorb kw4 to other subkeys */
- kw4l = subL[33]; kw4r = subR[33];
- /* round 23 */
- subL[30] ^= kw4l; subR[30] ^= kw4r;
- /* round 21 */
- subL[28] ^= kw4l; subR[28] ^= kw4r;
- /* round 19 */
- subL[26] ^= kw4l; subR[26] ^= kw4r;
- kw4l ^= kw4r & ~subR[24];
- dw = kw4l & subL[24],
- kw4r ^= ROL1(dw); /* modified for FL(kl5) */
- /* round 17 */
- subL[22] ^= kw4l; subR[22] ^= kw4r;
- /* round 15 */
- subL[20] ^= kw4l; subR[20] ^= kw4r;
- /* round 13 */
- subL[18] ^= kw4l; subR[18] ^= kw4r;
- kw4l ^= kw4r & ~subR[16];
- dw = kw4l & subL[16],
- kw4r ^= ROL1(dw); /* modified for FL(kl3) */
- /* round 11 */
- subL[14] ^= kw4l; subR[14] ^= kw4r;
- /* round 9 */
- subL[12] ^= kw4l; subR[12] ^= kw4r;
- /* round 7 */
- subL[10] ^= kw4l; subR[10] ^= kw4r;
- kw4l ^= kw4r & ~subR[8];
- dw = kw4l & subL[8],
- kw4r ^= ROL1(dw); /* modified for FL(kl1) */
- /* round 5 */
- subL[6] ^= kw4l; subR[6] ^= kw4r;
- /* round 3 */
- subL[4] ^= kw4l; subR[4] ^= kw4r;
- /* round 1 */
- subL[2] ^= kw4l; subR[2] ^= kw4r;
- /* kw1 */
- subL[0] ^= kw4l; subR[0] ^= kw4r;
-
- /* key XOR is end of F-function */
- SUBKEY_L(0) = subL[0] ^ subL[2];/* kw1 */
- SUBKEY_R(0) = subR[0] ^ subR[2];
- SUBKEY_L(2) = subL[3]; /* round 1 */
- SUBKEY_R(2) = subR[3];
- SUBKEY_L(3) = subL[2] ^ subL[4]; /* round 2 */
- SUBKEY_R(3) = subR[2] ^ subR[4];
- SUBKEY_L(4) = subL[3] ^ subL[5]; /* round 3 */
- SUBKEY_R(4) = subR[3] ^ subR[5];
- SUBKEY_L(5) = subL[4] ^ subL[6]; /* round 4 */
- SUBKEY_R(5) = subR[4] ^ subR[6];
- SUBKEY_L(6) = subL[5] ^ subL[7]; /* round 5 */
- SUBKEY_R(6) = subR[5] ^ subR[7];
- tl = subL[10] ^ (subR[10] & ~subR[8]);
- dw = tl & subL[8], /* FL(kl1) */
- tr = subR[10] ^ ROL1(dw);
- SUBKEY_L(7) = subL[6] ^ tl; /* round 6 */
- SUBKEY_R(7) = subR[6] ^ tr;
- SUBKEY_L(8) = subL[8]; /* FL(kl1) */
- SUBKEY_R(8) = subR[8];
- SUBKEY_L(9) = subL[9]; /* FLinv(kl2) */
- SUBKEY_R(9) = subR[9];
- tl = subL[7] ^ (subR[7] & ~subR[9]);
- dw = tl & subL[9], /* FLinv(kl2) */
- tr = subR[7] ^ ROL1(dw);
- SUBKEY_L(10) = tl ^ subL[11]; /* round 7 */
- SUBKEY_R(10) = tr ^ subR[11];
- SUBKEY_L(11) = subL[10] ^ subL[12]; /* round 8 */
- SUBKEY_R(11) = subR[10] ^ subR[12];
- SUBKEY_L(12) = subL[11] ^ subL[13]; /* round 9 */
- SUBKEY_R(12) = subR[11] ^ subR[13];
- SUBKEY_L(13) = subL[12] ^ subL[14]; /* round 10 */
- SUBKEY_R(13) = subR[12] ^ subR[14];
- SUBKEY_L(14) = subL[13] ^ subL[15]; /* round 11 */
- SUBKEY_R(14) = subR[13] ^ subR[15];
- tl = subL[18] ^ (subR[18] & ~subR[16]);
- dw = tl & subL[16], /* FL(kl3) */
- tr = subR[18] ^ ROL1(dw);
- SUBKEY_L(15) = subL[14] ^ tl; /* round 12 */
- SUBKEY_R(15) = subR[14] ^ tr;
- SUBKEY_L(16) = subL[16]; /* FL(kl3) */
- SUBKEY_R(16) = subR[16];
- SUBKEY_L(17) = subL[17]; /* FLinv(kl4) */
- SUBKEY_R(17) = subR[17];
- tl = subL[15] ^ (subR[15] & ~subR[17]);
- dw = tl & subL[17], /* FLinv(kl4) */
- tr = subR[15] ^ ROL1(dw);
- SUBKEY_L(18) = tl ^ subL[19]; /* round 13 */
- SUBKEY_R(18) = tr ^ subR[19];
- SUBKEY_L(19) = subL[18] ^ subL[20]; /* round 14 */
- SUBKEY_R(19) = subR[18] ^ subR[20];
- SUBKEY_L(20) = subL[19] ^ subL[21]; /* round 15 */
- SUBKEY_R(20) = subR[19] ^ subR[21];
- SUBKEY_L(21) = subL[20] ^ subL[22]; /* round 16 */
- SUBKEY_R(21) = subR[20] ^ subR[22];
- SUBKEY_L(22) = subL[21] ^ subL[23]; /* round 17 */
- SUBKEY_R(22) = subR[21] ^ subR[23];
- tl = subL[26] ^ (subR[26] & ~subR[24]);
- dw = tl & subL[24], /* FL(kl5) */
- tr = subR[26] ^ ROL1(dw);
- SUBKEY_L(23) = subL[22] ^ tl; /* round 18 */
- SUBKEY_R(23) = subR[22] ^ tr;
- SUBKEY_L(24) = subL[24]; /* FL(kl5) */
- SUBKEY_R(24) = subR[24];
- SUBKEY_L(25) = subL[25]; /* FLinv(kl6) */
- SUBKEY_R(25) = subR[25];
- tl = subL[23] ^ (subR[23] & ~subR[25]);
- dw = tl & subL[25], /* FLinv(kl6) */
- tr = subR[23] ^ ROL1(dw);
- SUBKEY_L(26) = tl ^ subL[27]; /* round 19 */
- SUBKEY_R(26) = tr ^ subR[27];
- SUBKEY_L(27) = subL[26] ^ subL[28]; /* round 20 */
- SUBKEY_R(27) = subR[26] ^ subR[28];
- SUBKEY_L(28) = subL[27] ^ subL[29]; /* round 21 */
- SUBKEY_R(28) = subR[27] ^ subR[29];
- SUBKEY_L(29) = subL[28] ^ subL[30]; /* round 22 */
- SUBKEY_R(29) = subR[28] ^ subR[30];
- SUBKEY_L(30) = subL[29] ^ subL[31]; /* round 23 */
- SUBKEY_R(30) = subR[29] ^ subR[31];
- SUBKEY_L(31) = subL[30]; /* round 24 */
- SUBKEY_R(31) = subR[30];
- SUBKEY_L(32) = subL[32] ^ subL[31]; /* kw3 */
- SUBKEY_R(32) = subR[32] ^ subR[31];
-
- /* apply the inverse of the last half of P-function */
- dw = SUBKEY_L(2) ^ SUBKEY_R(2); dw = ROL8(dw);/* round 1 */
- SUBKEY_R(2) = SUBKEY_L(2) ^ dw; SUBKEY_L(2) = dw;
- dw = SUBKEY_L(3) ^ SUBKEY_R(3); dw = ROL8(dw);/* round 2 */
- SUBKEY_R(3) = SUBKEY_L(3) ^ dw; SUBKEY_L(3) = dw;
- dw = SUBKEY_L(4) ^ SUBKEY_R(4); dw = ROL8(dw);/* round 3 */
- SUBKEY_R(4) = SUBKEY_L(4) ^ dw; SUBKEY_L(4) = dw;
- dw = SUBKEY_L(5) ^ SUBKEY_R(5); dw = ROL8(dw);/* round 4 */
- SUBKEY_R(5) = SUBKEY_L(5) ^ dw; SUBKEY_L(5) = dw;
- dw = SUBKEY_L(6) ^ SUBKEY_R(6); dw = ROL8(dw);/* round 5 */
- SUBKEY_R(6) = SUBKEY_L(6) ^ dw; SUBKEY_L(6) = dw;
- dw = SUBKEY_L(7) ^ SUBKEY_R(7); dw = ROL8(dw);/* round 6 */
- SUBKEY_R(7) = SUBKEY_L(7) ^ dw; SUBKEY_L(7) = dw;
- dw = SUBKEY_L(10) ^ SUBKEY_R(10); dw = ROL8(dw);/* round 7 */
- SUBKEY_R(10) = SUBKEY_L(10) ^ dw; SUBKEY_L(10) = dw;
- dw = SUBKEY_L(11) ^ SUBKEY_R(11); dw = ROL8(dw);/* round 8 */
- SUBKEY_R(11) = SUBKEY_L(11) ^ dw; SUBKEY_L(11) = dw;
- dw = SUBKEY_L(12) ^ SUBKEY_R(12); dw = ROL8(dw);/* round 9 */
- SUBKEY_R(12) = SUBKEY_L(12) ^ dw; SUBKEY_L(12) = dw;
- dw = SUBKEY_L(13) ^ SUBKEY_R(13); dw = ROL8(dw);/* round 10 */
- SUBKEY_R(13) = SUBKEY_L(13) ^ dw; SUBKEY_L(13) = dw;
- dw = SUBKEY_L(14) ^ SUBKEY_R(14); dw = ROL8(dw);/* round 11 */
- SUBKEY_R(14) = SUBKEY_L(14) ^ dw; SUBKEY_L(14) = dw;
- dw = SUBKEY_L(15) ^ SUBKEY_R(15); dw = ROL8(dw);/* round 12 */
- SUBKEY_R(15) = SUBKEY_L(15) ^ dw; SUBKEY_L(15) = dw;
- dw = SUBKEY_L(18) ^ SUBKEY_R(18); dw = ROL8(dw);/* round 13 */
- SUBKEY_R(18) = SUBKEY_L(18) ^ dw; SUBKEY_L(18) = dw;
- dw = SUBKEY_L(19) ^ SUBKEY_R(19); dw = ROL8(dw);/* round 14 */
- SUBKEY_R(19) = SUBKEY_L(19) ^ dw; SUBKEY_L(19) = dw;
- dw = SUBKEY_L(20) ^ SUBKEY_R(20); dw = ROL8(dw);/* round 15 */
- SUBKEY_R(20) = SUBKEY_L(20) ^ dw; SUBKEY_L(20) = dw;
- dw = SUBKEY_L(21) ^ SUBKEY_R(21); dw = ROL8(dw);/* round 16 */
- SUBKEY_R(21) = SUBKEY_L(21) ^ dw; SUBKEY_L(21) = dw;
- dw = SUBKEY_L(22) ^ SUBKEY_R(22); dw = ROL8(dw);/* round 17 */
- SUBKEY_R(22) = SUBKEY_L(22) ^ dw; SUBKEY_L(22) = dw;
- dw = SUBKEY_L(23) ^ SUBKEY_R(23); dw = ROL8(dw);/* round 18 */
- SUBKEY_R(23) = SUBKEY_L(23) ^ dw; SUBKEY_L(23) = dw;
- dw = SUBKEY_L(26) ^ SUBKEY_R(26); dw = ROL8(dw);/* round 19 */
- SUBKEY_R(26) = SUBKEY_L(26) ^ dw; SUBKEY_L(26) = dw;
- dw = SUBKEY_L(27) ^ SUBKEY_R(27); dw = ROL8(dw);/* round 20 */
- SUBKEY_R(27) = SUBKEY_L(27) ^ dw; SUBKEY_L(27) = dw;
- dw = SUBKEY_L(28) ^ SUBKEY_R(28); dw = ROL8(dw);/* round 21 */
- SUBKEY_R(28) = SUBKEY_L(28) ^ dw; SUBKEY_L(28) = dw;
- dw = SUBKEY_L(29) ^ SUBKEY_R(29); dw = ROL8(dw);/* round 22 */
- SUBKEY_R(29) = SUBKEY_L(29) ^ dw; SUBKEY_L(29) = dw;
- dw = SUBKEY_L(30) ^ SUBKEY_R(30); dw = ROL8(dw);/* round 23 */
- SUBKEY_R(30) = SUBKEY_L(30) ^ dw; SUBKEY_L(30) = dw;
- dw = SUBKEY_L(31) ^ SUBKEY_R(31); dw = ROL8(dw);/* round 24 */
- SUBKEY_R(31) = SUBKEY_L(31) ^ dw; SUBKEY_L(31) = dw;
+ camellia_setup_tail(subkey, subL, subR, 32);
}
static void camellia_setup192(const unsigned char *key, u32 *subkey)
}
-static void camellia_encrypt128(const u32 *subkey, u32 *io_text)
-{
- u32 il,ir,t0,t1; /* temporary variables */
-
- u32 io[4];
-
- /* pre whitening but absorb kw2 */
- io[0] = io_text[0] ^ SUBKEY_L(0);
- io[1] = io_text[1] ^ SUBKEY_R(0);
- io[2] = io_text[2];
- io[3] = io_text[3];
-
- /* main iteration */
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(2),SUBKEY_R(2),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(3),SUBKEY_R(3),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(4),SUBKEY_R(4),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(5),SUBKEY_R(5),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(6),SUBKEY_R(6),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(7),SUBKEY_R(7),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(8),SUBKEY_R(8),
- SUBKEY_L(9),SUBKEY_R(9),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(10),SUBKEY_R(10),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(11),SUBKEY_R(11),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(12),SUBKEY_R(12),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(13),SUBKEY_R(13),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(14),SUBKEY_R(14),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(15),SUBKEY_R(15),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(16),SUBKEY_R(16),
- SUBKEY_L(17),SUBKEY_R(17),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(18),SUBKEY_R(18),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(19),SUBKEY_R(19),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(20),SUBKEY_R(20),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(21),SUBKEY_R(21),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(22),SUBKEY_R(22),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(23),SUBKEY_R(23),
- io[0],io[1],il,ir,t0,t1);
+/*
+ * Encrypt/decrypt
+ */
+#define CAMELLIA_FLS(ll, lr, rl, rr, kll, klr, krl, krr, t0, t1, t2, t3) \
+ do { \
+ t0 = kll; \
+ t2 = krr; \
+ t0 &= ll; \
+ t2 |= rr; \
+ rl ^= t2; \
+ lr ^= rol32(t0, 1); \
+ t3 = krl; \
+ t1 = klr; \
+ t3 &= rl; \
+ t1 |= lr; \
+ ll ^= t1; \
+ rr ^= rol32(t3, 1); \
+ } while(0)
- /* post whitening but kw4 */
- io_text[0] = io[2] ^ SUBKEY_L(24);
- io_text[1] = io[3] ^ SUBKEY_R(24);
- io_text[2] = io[0];
- io_text[3] = io[1];
-}
+#define CAMELLIA_ROUNDSM(xl, xr, kl, kr, yl, yr, il, ir) \
+ do { \
+ ir = camellia_sp1110[(u8)xr]; \
+ il = camellia_sp1110[ (xl >> 24)]; \
+ ir ^= camellia_sp0222[ (xr >> 24)]; \
+ il ^= camellia_sp0222[(u8)(xl >> 16)]; \
+ ir ^= camellia_sp3033[(u8)(xr >> 16)]; \
+ il ^= camellia_sp3033[(u8)(xl >> 8)]; \
+ ir ^= camellia_sp4404[(u8)(xr >> 8)]; \
+ il ^= camellia_sp4404[(u8)xl]; \
+ il ^= kl; \
+ ir ^= il ^ kr; \
+ yl ^= ir; \
+ yr ^= ror32(il, 8) ^ ir; \
+ } while(0)
-static void camellia_decrypt128(const u32 *subkey, u32 *io_text)
+/* max = 24: 128bit encrypt, max = 32: 256bit encrypt */
+static void camellia_do_encrypt(const u32 *subkey, u32 *io, unsigned max)
{
u32 il,ir,t0,t1; /* temporary variables */
- u32 io[4];
-
/* pre whitening but absorb kw2 */
- io[0] = io_text[0] ^ SUBKEY_L(24);
- io[1] = io_text[1] ^ SUBKEY_R(24);
- io[2] = io_text[2];
- io[3] = io_text[3];
+ io[0] ^= SUBKEY_L(0);
+ io[1] ^= SUBKEY_R(0);
/* main iteration */
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(23),SUBKEY_R(23),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(22),SUBKEY_R(22),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(21),SUBKEY_R(21),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(20),SUBKEY_R(20),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(19),SUBKEY_R(19),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(18),SUBKEY_R(18),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(17),SUBKEY_R(17),
- SUBKEY_L(16),SUBKEY_R(16),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(15),SUBKEY_R(15),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(14),SUBKEY_R(14),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(13),SUBKEY_R(13),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(12),SUBKEY_R(12),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(11),SUBKEY_R(11),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(10),SUBKEY_R(10),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(9),SUBKEY_R(9),
- SUBKEY_L(8),SUBKEY_R(8),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(7),SUBKEY_R(7),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(6),SUBKEY_R(6),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(5),SUBKEY_R(5),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(4),SUBKEY_R(4),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(3),SUBKEY_R(3),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(2),SUBKEY_R(2),
- io[0],io[1],il,ir,t0,t1);
-
- /* post whitening but kw4 */
- io_text[0] = io[2] ^ SUBKEY_L(0);
- io_text[1] = io[3] ^ SUBKEY_R(0);
- io_text[2] = io[0];
- io_text[3] = io[1];
-}
-
-static void camellia_encrypt256(const u32 *subkey, u32 *io_text)
-{
- u32 il,ir,t0,t1; /* temporary variables */
-
- u32 io[4];
-
- /* pre whitening but absorb kw2 */
- io[0] = io_text[0] ^ SUBKEY_L(0);
- io[1] = io_text[1] ^ SUBKEY_R(0);
- io[2] = io_text[2];
- io[3] = io_text[3];
+#define ROUNDS(i) do { \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 2),SUBKEY_R(i + 2), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 3),SUBKEY_R(i + 3), \
+ io[0],io[1],il,ir); \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 4),SUBKEY_R(i + 4), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 5),SUBKEY_R(i + 5), \
+ io[0],io[1],il,ir); \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 6),SUBKEY_R(i + 6), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 7),SUBKEY_R(i + 7), \
+ io[0],io[1],il,ir); \
+} while (0)
+#define FLS(i) do { \
+ CAMELLIA_FLS(io[0],io[1],io[2],io[3], \
+ SUBKEY_L(i + 0),SUBKEY_R(i + 0), \
+ SUBKEY_L(i + 1),SUBKEY_R(i + 1), \
+ t0,t1,il,ir); \
+} while (0)
+
+ ROUNDS(0);
+ FLS(8);
+ ROUNDS(8);
+ FLS(16);
+ ROUNDS(16);
+ if (max == 32) {
+ FLS(24);
+ ROUNDS(24);
+ }
- /* main iteration */
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(2),SUBKEY_R(2),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(3),SUBKEY_R(3),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(4),SUBKEY_R(4),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(5),SUBKEY_R(5),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(6),SUBKEY_R(6),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(7),SUBKEY_R(7),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(8),SUBKEY_R(8),
- SUBKEY_L(9),SUBKEY_R(9),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(10),SUBKEY_R(10),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(11),SUBKEY_R(11),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(12),SUBKEY_R(12),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(13),SUBKEY_R(13),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(14),SUBKEY_R(14),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(15),SUBKEY_R(15),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(16),SUBKEY_R(16),
- SUBKEY_L(17),SUBKEY_R(17),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(18),SUBKEY_R(18),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(19),SUBKEY_R(19),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(20),SUBKEY_R(20),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(21),SUBKEY_R(21),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(22),SUBKEY_R(22),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(23),SUBKEY_R(23),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(24),SUBKEY_R(24),
- SUBKEY_L(25),SUBKEY_R(25),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(26),SUBKEY_R(26),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(27),SUBKEY_R(27),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(28),SUBKEY_R(28),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(29),SUBKEY_R(29),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(30),SUBKEY_R(30),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(31),SUBKEY_R(31),
- io[0],io[1],il,ir,t0,t1);
+#undef ROUNDS
+#undef FLS
/* post whitening but kw4 */
- io_text[0] = io[2] ^ SUBKEY_L(32);
- io_text[1] = io[3] ^ SUBKEY_R(32);
- io_text[2] = io[0];
- io_text[3] = io[1];
+ io[2] ^= SUBKEY_L(max);
+ io[3] ^= SUBKEY_R(max);
+ /* NB: io[0],[1] should be swapped with [2],[3] by caller! */
}
-static void camellia_decrypt256(const u32 *subkey, u32 *io_text)
+static void camellia_do_decrypt(const u32 *subkey, u32 *io, unsigned i)
{
- u32 il,ir,t0,t1; /* temporary variables */
-
- u32 io[4];
+ u32 il,ir,t0,t1; /* temporary variables */
/* pre whitening but absorb kw2 */
- io[0] = io_text[0] ^ SUBKEY_L(32);
- io[1] = io_text[1] ^ SUBKEY_R(32);
- io[2] = io_text[2];
- io[3] = io_text[3];
+ io[0] ^= SUBKEY_L(i);
+ io[1] ^= SUBKEY_R(i);
/* main iteration */
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(31),SUBKEY_R(31),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(30),SUBKEY_R(30),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(29),SUBKEY_R(29),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(28),SUBKEY_R(28),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(27),SUBKEY_R(27),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(26),SUBKEY_R(26),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(25),SUBKEY_R(25),
- SUBKEY_L(24),SUBKEY_R(24),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(23),SUBKEY_R(23),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(22),SUBKEY_R(22),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(21),SUBKEY_R(21),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(20),SUBKEY_R(20),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(19),SUBKEY_R(19),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(18),SUBKEY_R(18),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(17),SUBKEY_R(17),
- SUBKEY_L(16),SUBKEY_R(16),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(15),SUBKEY_R(15),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(14),SUBKEY_R(14),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(13),SUBKEY_R(13),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(12),SUBKEY_R(12),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(11),SUBKEY_R(11),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(10),SUBKEY_R(10),
- io[0],io[1],il,ir,t0,t1);
-
- CAMELLIA_FLS(io[0],io[1],io[2],io[3],
- SUBKEY_L(9),SUBKEY_R(9),
- SUBKEY_L(8),SUBKEY_R(8),
- t0,t1,il,ir);
-
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(7),SUBKEY_R(7),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(6),SUBKEY_R(6),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(5),SUBKEY_R(5),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(4),SUBKEY_R(4),
- io[0],io[1],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[0],io[1],
- SUBKEY_L(3),SUBKEY_R(3),
- io[2],io[3],il,ir,t0,t1);
- CAMELLIA_ROUNDSM(io[2],io[3],
- SUBKEY_L(2),SUBKEY_R(2),
- io[0],io[1],il,ir,t0,t1);
+#define ROUNDS(i) do { \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 7),SUBKEY_R(i + 7), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 6),SUBKEY_R(i + 6), \
+ io[0],io[1],il,ir); \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 5),SUBKEY_R(i + 5), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 4),SUBKEY_R(i + 4), \
+ io[0],io[1],il,ir); \
+ CAMELLIA_ROUNDSM(io[0],io[1], \
+ SUBKEY_L(i + 3),SUBKEY_R(i + 3), \
+ io[2],io[3],il,ir); \
+ CAMELLIA_ROUNDSM(io[2],io[3], \
+ SUBKEY_L(i + 2),SUBKEY_R(i + 2), \
+ io[0],io[1],il,ir); \
+} while (0)
+#define FLS(i) do { \
+ CAMELLIA_FLS(io[0],io[1],io[2],io[3], \
+ SUBKEY_L(i + 1),SUBKEY_R(i + 1), \
+ SUBKEY_L(i + 0),SUBKEY_R(i + 0), \
+ t0,t1,il,ir); \
+} while (0)
+
+ if (i == 32) {
+ ROUNDS(24);
+ FLS(24);
+ }
+ ROUNDS(16);
+ FLS(16);
+ ROUNDS(8);
+ FLS(8);
+ ROUNDS(0);
+
+#undef ROUNDS
+#undef FLS
/* post whitening but kw4 */
- io_text[0] = io[2] ^ SUBKEY_L(0);
- io_text[1] = io[3] ^ SUBKEY_R(0);
- io_text[2] = io[0];
- io_text[3] = io[1];
+ io[2] ^= SUBKEY_L(0);
+ io[3] ^= SUBKEY_R(0);
+ /* NB: 0,1 should be swapped with 2,3 by caller! */
}
struct camellia_ctx {
int key_length;
- u32 key_table[CAMELLIA_TABLE_BYTE_LEN / 4];
+ u32 key_table[CAMELLIA_TABLE_BYTE_LEN / sizeof(u32)];
};
static int
tmp[2] = be32_to_cpu(src[2]);
tmp[3] = be32_to_cpu(src[3]);
- switch (cctx->key_length) {
- case 16:
- camellia_encrypt128(cctx->key_table, tmp);
- break;
- case 24:
- /* fall through */
- case 32:
- camellia_encrypt256(cctx->key_table, tmp);
- break;
- }
+ camellia_do_encrypt(cctx->key_table, tmp,
+ cctx->key_length == 16 ? 24 : 32 /* for key lengths of 24 and 32 */
+ );
- dst[0] = cpu_to_be32(tmp[0]);
- dst[1] = cpu_to_be32(tmp[1]);
- dst[2] = cpu_to_be32(tmp[2]);
- dst[3] = cpu_to_be32(tmp[3]);
+ /* do_encrypt returns 0,1 swapped with 2,3 */
+ dst[0] = cpu_to_be32(tmp[2]);
+ dst[1] = cpu_to_be32(tmp[3]);
+ dst[2] = cpu_to_be32(tmp[0]);
+ dst[3] = cpu_to_be32(tmp[1]);
}
static void camellia_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in)
tmp[2] = be32_to_cpu(src[2]);
tmp[3] = be32_to_cpu(src[3]);
- switch (cctx->key_length) {
- case 16:
- camellia_decrypt128(cctx->key_table, tmp);
- break;
- case 24:
- /* fall through */
- case 32:
- camellia_decrypt256(cctx->key_table, tmp);
- break;
- }
+ camellia_do_decrypt(cctx->key_table, tmp,
+ cctx->key_length == 16 ? 24 : 32 /* for key lengths of 24 and 32 */
+ );
- dst[0] = cpu_to_be32(tmp[0]);
- dst[1] = cpu_to_be32(tmp[1]);
- dst[2] = cpu_to_be32(tmp[2]);
- dst[3] = cpu_to_be32(tmp[3]);
+ /* do_decrypt returns 0,1 swapped with 2,3 */
+ dst[0] = cpu_to_be32(tmp[2]);
+ dst[1] = cpu_to_be32(tmp[3]);
+ dst[2] = cpu_to_be32(tmp[0]);
+ dst[3] = cpu_to_be32(tmp[1]);
}
static struct crypto_alg camellia_alg = {
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff