static const u32 bytebit[8] =
{
- 0200, 0100, 040, 020, 010, 04, 02, 01
+ 0200, 0100, 040, 020, 010, 04, 02, 01
};
static const u32 bigbyte[24] =
0x8000UL, 0x4000UL, 0x2000UL, 0x1000UL,
0x800UL, 0x400UL, 0x200UL, 0x100UL,
0x80UL, 0x40UL, 0x20UL, 0x10UL,
- 0x8UL, 0x4UL, 0x2UL, 0x1L
+ 0x8UL, 0x4UL, 0x2UL, 0x1L
};
/* Use the key schedule specific in the standard (ANSI X3.92-1981) */
static const u8 pc1[56] = {
- 56, 48, 40, 32, 24, 16, 8, 0, 57, 49, 41, 33, 25, 17,
- 9, 1, 58, 50, 42, 34, 26, 18, 10, 2, 59, 51, 43, 35,
+ 56, 48, 40, 32, 24, 16, 8, 0, 57, 49, 41, 33, 25, 17,
+ 9, 1, 58, 50, 42, 34, 26, 18, 10, 2, 59, 51, 43, 35,
62, 54, 46, 38, 30, 22, 14, 6, 61, 53, 45, 37, 29, 21,
- 13, 5, 60, 52, 44, 36, 28, 20, 12, 4, 27, 19, 11, 3
+ 13, 5, 60, 52, 44, 36, 28, 20, 12, 4, 27, 19, 11, 3
};
static const u8 totrot[16] = {
1, 2, 4, 6,
- 8, 10, 12, 14,
- 15, 17, 19, 21,
+ 8, 10, 12, 14,
+ 15, 17, 19, 21,
23, 25, 27, 28
};
#define MP_PREC 32 /* default digits of precision */
#else
#define MP_PREC 8 /* default digits of precision */
- #endif
+ #endif
#endif
/* size of comba arrays, should be at least 2 * 2**(BITS_PER_WORD - BITS_PER_DIGIT*2) */
*tmpc++ &= MP_MASK;
}
- /* now copy higher words if any, that is in A+B
- * if A or B has more digits add those in
+ /* now copy higher words if any, that is in A+B
+ * if A or B has more digits add those in
*/
if (min != max) {
for (; i < max; i++) {
#ifdef BN_MP_TOOM_MUL_C
if (MIN (a->used, b->used) >= TOOM_MUL_CUTOFF) {
res = mp_toom_mul(a, b, c);
- } else
+ } else
#endif
#ifdef BN_MP_KARATSUBA_MUL_C
/* use Karatsuba? */
if (MIN (a->used, b->used) >= KARATSUBA_MUL_CUTOFF) {
res = mp_karatsuba_mul (a, b, c);
- } else
+ } else
#endif
{
/* can we use the fast multiplier?
*
- * The fast multiplier can be used if the output will
- * have less than MP_WARRAY digits and the number of
+ * The fast multiplier can be used if the output will
+ * have less than MP_WARRAY digits and the number of
* digits won't affect carry propagation
*/
#ifdef BN_FAST_S_MP_MUL_DIGS_C
int digs = a->used + b->used + 1;
if ((digs < MP_WARRAY) &&
- MIN(a->used, b->used) <=
+ MIN(a->used, b->used) <=
(1 << ((CHAR_BIT * sizeof (mp_word)) - (2 * DIGIT_BIT)))) {
res = fast_s_mp_mul_digs (a, b, c, digs);
- } else
+ } else
#endif
#ifdef BN_S_MP_MUL_DIGS_C
res = s_mp_mul (a, b, c); /* uses s_mp_mul_digs */
err = mp_exptmod(&tmpG, &tmpX, P, Y);
mp_clear_multi(&tmpG, &tmpX, NULL);
return err;
-#else
+#else
#error mp_exptmod would always fail
/* no invmod */
return MP_VAL;
dr = mp_reduce_is_2k(P) << 1;
}
#endif
-
+
/* if the modulus is odd or dr != 0 use the montgomery method */
#ifdef BN_MP_EXPTMOD_FAST_C
if (mp_isodd (P) == 1 || dr != 0) {
return MP_GT;
}
}
-
+
/* compare digits */
if (a->sign == MP_NEG) {
/* if negative compare opposite direction */
}
/* init temps */
- if ((res = mp_init_multi(&x, &y, &u, &v,
+ if ((res = mp_init_multi(&x, &y, &u, &v,
&A, &B, &C, &D, NULL)) != MP_OKAY) {
return res;
}
goto LBL_ERR;
}
}
-
+
/* too big */
while (mp_cmp_mag(&C, b) != MP_LT) {
if ((res = mp_sub(&C, b, &C)) != MP_OKAY) {
goto LBL_ERR;
}
}
-
+
/* C is now the inverse */
mp_exch (&C, c);
res = MP_OKAY;
if (a->used > b->used) {
return MP_GT;
}
-
+
if (a->used < b->used) {
return MP_LT;
}
/* top [offset into digits] */
top = a->dp + b;
- /* this is implemented as a sliding window where
- * the window is b-digits long and digits from
+ /* this is implemented as a sliding window where
+ * the window is b-digits long and digits from
* the top of the window are copied to the bottom
*
* e.g.
*bottom++ = 0;
}
}
-
+
/* remove excess digits */
a->used -= b;
}
-/* swap the elements of two integers, for cases where you can't simply swap the
+/* swap the elements of two integers, for cases where you can't simply swap the
* mp_int pointers around
*/
static void mp_exch (mp_int * a, mp_int * b)
}
-/* trim unused digits
+/* trim unused digits
*
* This is used to ensure that leading zero digits are
* trimed and the leading "used" digit will be non-zero
#ifdef BN_MP_ABS_C
-/* b = |a|
+/* b = |a|
*
* Simple function copies the input and fixes the sign to positive
*/
/* set the carry to the carry bits of the current word */
r = rr;
}
-
+
/* set final carry */
if (r != 0) {
c->dp[(c->used)++] = r;
#ifdef BN_MP_INIT_MULTI_C
-static int mp_init_multi(mp_int *mp, ...)
+static int mp_init_multi(mp_int *mp, ...)
{
mp_err res = MP_OKAY; /* Assume ok until proven otherwise */
int n = 0; /* Number of ok inits */
succeeded in init-ing, then return error.
*/
va_list clean_args;
-
+
/* end the current list */
va_end(args);
-
- /* now start cleaning up */
+
+ /* now start cleaning up */
cur_arg = mp;
va_start(clean_args, mp);
while (n--) {
#ifdef BN_MP_CLEAR_MULTI_C
-static void mp_clear_multi(mp_int *mp, ...)
+static void mp_clear_multi(mp_int *mp, ...)
{
mp_int* next_mp = mp;
va_list args;
/* get number of digits and add that */
r = (a->used - 1) * DIGIT_BIT;
-
+
/* take the last digit and count the bits in it */
q = a->dp[a->used - 1];
while (q > ((mp_digit) 0)) {
}
return res;
}
-
+
/* init our temps */
if ((res = mp_init_multi(&ta, &tb, &tq, &q, NULL)) != MP_OKAY) {
return res;
mp_set(&tq, 1);
n = mp_count_bits(a) - mp_count_bits(b);
if (((res = mp_abs(a, &ta)) != MP_OKAY) ||
- ((res = mp_abs(b, &tb)) != MP_OKAY) ||
+ ((res = mp_abs(b, &tb)) != MP_OKAY) ||
((res = mp_mul_2d(&tb, n, &tb)) != MP_OKAY) ||
((res = mp_mul_2d(&tq, n, &tq)) != MP_OKAY)) {
goto LBL_ERR;
#else
-/* integer signed division.
+/* integer signed division.
* c*b + d == a [e.g. a/b, c=quotient, d=remainder]
* HAC pp.598 Algorithm 14.20
*
- * Note that the description in HAC is horribly
- * incomplete. For example, it doesn't consider
- * the case where digits are removed from 'x' in
- * the inner loop. It also doesn't consider the
+ * Note that the description in HAC is horribly
+ * incomplete. For example, it doesn't consider
+ * the case where digits are removed from 'x' in
+ * the inner loop. It also doesn't consider the
* case that y has fewer than three digits, etc..
*
- * The overall algorithm is as described as
+ * The overall algorithm is as described as
* 14.20 from HAC but fixed to treat these cases.
*/
static int mp_div (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
continue;
}
- /* step 3.1 if xi == yt then set q{i-t-1} to b-1,
+ /* step 3.1 if xi == yt then set q{i-t-1} to b-1,
* otherwise set q{i-t-1} to (xi*b + x{i-1})/yt */
if (x.dp[i] == y.dp[t]) {
q.dp[i - t - 1] = ((((mp_digit)1) << DIGIT_BIT) - 1);
q.dp[i - t - 1] = (mp_digit) (tmp & (mp_word) (MP_MASK));
}
- /* while (q{i-t-1} * (yt * b + y{t-1})) >
- xi * b**2 + xi-1 * b + xi-2
-
- do q{i-t-1} -= 1;
+ /* while (q{i-t-1} * (yt * b + y{t-1})) >
+ xi * b**2 + xi-1 * b + xi-2
+
+ do q{i-t-1} -= 1;
*/
q.dp[i - t - 1] = (q.dp[i - t - 1] + 1) & MP_MASK;
do {
}
}
- /* now q is the quotient and x is the remainder
- * [which we have to normalize]
+ /* now q is the quotient and x is the remainder
+ * [which we have to normalize]
*/
-
+
/* get sign before writing to c */
x.sign = x.used == 0 ? MP_ZPOS : a->sign;
/* init M array */
/* init first cell */
if ((err = mp_init(&M[1])) != MP_OKAY) {
- return err;
+ return err;
}
/* now init the second half of the array */
if ((err = mp_init (&mu)) != MP_OKAY) {
goto LBL_M;
}
-
+
if (redmode == 0) {
if ((err = mp_reduce_setup (&mu, P)) != MP_OKAY) {
goto LBL_MU;
goto LBL_MU;
}
redux = mp_reduce_2k_l;
- }
+ }
/* create M table
*
- * The M table contains powers of the base,
+ * The M table contains powers of the base,
* e.g. M[x] = G**x mod P
*
- * The first half of the table is not
+ * The first half of the table is not
* computed though accept for M[0] and M[1]
*/
if ((err = mp_mod (G, P, &M[1])) != MP_OKAY) {
goto LBL_MU;
}
- /* compute the value at M[1<<(winsize-1)] by squaring
- * M[1] (winsize-1) times
+ /* compute the value at M[1<<(winsize-1)] by squaring
+ * M[1] (winsize-1) times
*/
if ((err = mp_copy (&M[1], &M[1 << (winsize - 1)])) != MP_OKAY) {
goto LBL_MU;
for (x = 0; x < (winsize - 1); x++) {
/* square it */
- if ((err = mp_sqr (&M[1 << (winsize - 1)],
+ if ((err = mp_sqr (&M[1 << (winsize - 1)],
&M[1 << (winsize - 1)])) != MP_OKAY) {
goto LBL_MU;
}
if (a->used >= TOOM_SQR_CUTOFF) {
res = mp_toom_sqr(a, b);
/* Karatsuba? */
- } else
+ } else
#endif
#ifdef BN_MP_KARATSUBA_SQR_C
if (a->used >= KARATSUBA_SQR_CUTOFF) {
res = mp_karatsuba_sqr (a, b);
- } else
+ } else
#endif
{
#ifdef BN_FAST_S_MP_SQR_C
/* can we use the fast comba multiplier? */
- if ((a->used * 2 + 1) < MP_WARRAY &&
- a->used <
+ if ((a->used * 2 + 1) < MP_WARRAY &&
+ a->used <
(1 << (sizeof(mp_word) * CHAR_BIT - 2*DIGIT_BIT - 1))) {
res = fast_s_mp_sqr (a, b);
} else
}
-/* reduces a modulo n where n is of the form 2**p - d
+/* reduces a modulo n where n is of the form 2**p - d
This differs from reduce_2k since "d" can be larger
than a single digit.
*/
{
mp_int q;
int p, res;
-
+
if ((res = mp_init(&q)) != MP_OKAY) {
return res;
}
-
- p = mp_count_bits(n);
+
+ p = mp_count_bits(n);
top:
/* q = a/2**p, a = a mod 2**p */
if ((res = mp_div_2d(a, p, &q, a)) != MP_OKAY) {
goto ERR;
}
-
+
/* q = q * d */
- if ((res = mp_mul(&q, d, &q)) != MP_OKAY) {
+ if ((res = mp_mul(&q, d, &q)) != MP_OKAY) {
goto ERR;
}
-
+
/* a = a + q */
if ((res = s_mp_add(a, &q, a)) != MP_OKAY) {
goto ERR;
}
-
+
if (mp_cmp_mag(a, n) != MP_LT) {
s_mp_sub(a, n, a);
goto top;
}
-
+
ERR:
mp_clear(&q);
return res;
{
int res;
mp_int tmp;
-
+
if ((res = mp_init(&tmp)) != MP_OKAY) {
return res;
}
-
+
if ((res = mp_2expt(&tmp, mp_count_bits(a))) != MP_OKAY) {
goto ERR;
}
-
+
if ((res = s_mp_sub(&tmp, a, d)) != MP_OKAY) {
goto ERR;
}
-
+
ERR:
mp_clear(&tmp);
return res;
}
-/* computes a = 2**b
+/* computes a = 2**b
*
* Simple algorithm which zeroes the int, grows it then just sets one bit
* as required.
static int mp_reduce_setup (mp_int * a, mp_int * b)
{
int res;
-
+
if ((res = mp_2expt (a, b->used * 2 * DIGIT_BIT)) != MP_OKAY) {
return res;
}
}
-/* reduces x mod m, assumes 0 < x < m**2, mu is
+/* reduces x mod m, assumes 0 < x < m**2, mu is
* precomputed via mp_reduce_setup.
* From HAC pp.604 Algorithm 14.42
*/
}
/* q1 = x / b**(k-1) */
- mp_rshd (&q, um - 1);
+ mp_rshd (&q, um - 1);
/* according to HAC this optimization is ok */
if (((unsigned long) um) > (((mp_digit)1) << (DIGIT_BIT - 1))) {
if ((res = fast_s_mp_mul_high_digs (&q, mu, &q, um)) != MP_OKAY) {
goto CLEANUP;
}
-#else
- {
+#else
+ {
#error mp_reduce would always fail
res = MP_VAL;
goto CLEANUP;
}
/* q3 = q2 / b**(k+1) */
- mp_rshd (&q, um + 1);
+ mp_rshd (&q, um + 1);
/* x = x mod b**(k+1), quick (no division) */
if ((res = mp_mod_2d (x, DIGIT_BIT * (um + 1), x)) != MP_OKAY) {
goto CLEANUP;
}
}
-
+
CLEANUP:
mp_clear (&q);
/* multiplies |a| * |b| and only computes up to digs digits of result
- * HAC pp. 595, Algorithm 14.12 Modified so you can control how
+ * HAC pp. 595, Algorithm 14.12 Modified so you can control how
* many digits of output are created.
*/
static int s_mp_mul_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
#ifdef BN_FAST_S_MP_MUL_DIGS_C
/* can we use the fast multiplier? */
if (((digs) < MP_WARRAY) &&
- MIN (a->used, b->used) <
+ MIN (a->used, b->used) <
(1 << ((CHAR_BIT * sizeof (mp_word)) - (2 * DIGIT_BIT)))) {
return fast_s_mp_mul_digs (a, b, c, digs);
}
/* setup some aliases */
/* copy of the digit from a used within the nested loop */
tmpx = a->dp[ix];
-
+
/* an alias for the destination shifted ix places */
tmpt = t.dp + ix;
-
+
/* an alias for the digits of b */
tmpy = b->dp;
#ifdef BN_FAST_S_MP_MUL_DIGS_C
/* Fast (comba) multiplier
*
- * This is the fast column-array [comba] multiplier. It is
- * designed to compute the columns of the product first
- * then handle the carries afterwards. This has the effect
+ * This is the fast column-array [comba] multiplier. It is
+ * designed to compute the columns of the product first
+ * then handle the carries afterwards. This has the effect
* of making the nested loops that compute the columns very
* simple and schedulable on super-scalar processors.
*
- * This has been modified to produce a variable number of
- * digits of output so if say only a half-product is required
- * you don't have to compute the upper half (a feature
+ * This has been modified to produce a variable number of
+ * digits of output so if say only a half-product is required
+ * you don't have to compute the upper half (a feature
* required for fast Barrett reduction).
*
* Based on Algorithm 14.12 on pp.595 of HAC.
/* clear the carry */
_W = 0;
- for (ix = 0; ix < pa; ix++) {
+ for (ix = 0; ix < pa; ix++) {
int tx, ty;
int iy;
mp_digit *tmpx, *tmpy;
tmpx = a->dp + tx;
tmpy = b->dp + ty;
- /* this is the number of times the loop will iterrate, essentially
+ /* this is the number of times the loop will iterrate, essentially
while (tx++ < a->used && ty-- >= 0) { ... }
*/
iy = MIN(a->used-tx, ty+1);
int x;
/* pad size so there are always extra digits */
- size += (MP_PREC * 2) - (size % MP_PREC);
-
+ size += (MP_PREC * 2) - (size % MP_PREC);
+
/* alloc mem */
a->dp = OPT_CAST(mp_digit) XMALLOC (sizeof (mp_digit) * size);
if (a->dp == NULL) {
/* alias for where to store the results */
tmpt = t.dp + (2*ix + 1);
-
+
for (iy = ix + 1; iy < pa; iy++) {
/* first calculate the product */
r = ((mp_word)tmpx) * ((mp_word)a->dp[iy]);
/* alias for source */
tmpa = a->dp;
-
+
/* alias for dest */
tmpb = b->dp;
/* carry */
r = 0;
for (x = 0; x < a->used; x++) {
-
- /* get what will be the *next* carry bit from the
- * MSB of the current digit
+
+ /* get what will be the *next* carry bit from the
+ * MSB of the current digit
*/
rr = *tmpa >> ((mp_digit)(DIGIT_BIT - 1));
-
+
/* now shift up this digit, add in the carry [from the previous] */
*tmpb++ = ((*tmpa++ << ((mp_digit)1)) | r) & MP_MASK;
-
- /* copy the carry that would be from the source
- * digit into the next iteration
+
+ /* copy the carry that would be from the source
+ * digit into the next iteration
*/
r = rr;
}
++(b->used);
}
- /* now zero any excess digits on the destination
- * that we didn't write to
+ /* now zero any excess digits on the destination
+ * that we didn't write to
*/
tmpb = b->dp + b->used;
for (x = b->used; x < oldused; x++) {
/* determine and setup reduction code */
if (redmode == 0) {
-#ifdef BN_MP_MONTGOMERY_SETUP_C
+#ifdef BN_MP_MONTGOMERY_SETUP_C
/* now setup montgomery */
if ((err = mp_montgomery_setup (P, &mp)) != MP_OKAY) {
goto LBL_M;
if (((P->used * 2 + 1) < MP_WARRAY) &&
P->used < (1 << ((CHAR_BIT * sizeof (mp_word)) - (2 * DIGIT_BIT)))) {
redux = fast_mp_montgomery_reduce;
- } else
+ } else
#endif
{
#ifdef BN_MP_MONTGOMERY_REDUCE_C
if ((err = mp_montgomery_calc_normalization (&res, P)) != MP_OKAY) {
goto LBL_RES;
}
-#else
+#else
err = MP_VAL;
goto LBL_RES;
#endif
#ifdef BN_FAST_S_MP_SQR_C
/* the jist of squaring...
- * you do like mult except the offset of the tmpx [one that
- * starts closer to zero] can't equal the offset of tmpy.
+ * you do like mult except the offset of the tmpx [one that
+ * starts closer to zero] can't equal the offset of tmpy.
* So basically you set up iy like before then you min it with
- * (ty-tx) so that it never happens. You double all those
+ * (ty-tx) so that it never happens. You double all those
* you add in the inner loop
After that loop you do the squares and add them in.
/* number of output digits to produce */
W1 = 0;
- for (ix = 0; ix < pa; ix++) {
+ for (ix = 0; ix < pa; ix++) {
int tx, ty, iy;
mp_word _W;
mp_digit *tmpy;
*/
iy = MIN(a->used-tx, ty+1);
- /* now for squaring tx can never equal ty
+ /* now for squaring tx can never equal ty
* we halve the distance since they approach at a rate of 2x
* and we have to round because odd cases need to be executed
*/