135 files changed, 12115 insertions, 5705 deletions

diff --git a/libtommath/astylerc b/libtommath/astylerc
new file mode 100644
index 0000000..5d63f7a
--- /dev/null
+++ b/libtommath/astylerc
@@ -0,0 +1,27 @@
+# Artistic Style, see http://astyle.sourceforge.net/
+# full documentation, see: http://astyle.sourceforge.net/astyle.html
+#
+# usage:
+#       astyle --options=astylerc *.[ch]
+
+## Bracket Style Options
+style=kr
+
+## Tab Options
+indent=spaces=3
+
+## Bracket Modify Options
+
+## Indentation Options
+min-conditional-indent=0
+
+## Padding Options
+pad-header
+unpad-paren
+align-pointer=name
+
+## Formatting Options
+break-after-logical
+max-code-length=120
+convert-tabs
+mode=c
diff --git a/libtommath/bn_error.c b/libtommath/bn_error.c
index 0d77411..a51d712 100644
--- a/libtommath/bn_error.c
+++ b/libtommath/bn_error.c
@@ -16,12 +16,12 @@
  */
 
 static const struct {
-     int code;
-     const char *msg;
+   int code;
+   const char *msg;
 } msgs[] = {
-     { MP_OKAY, "Successful" },
-     { MP_MEM,  "Out of heap" },
-     { MP_VAL,  "Value out of range" }
+   { MP_OKAY, "Successful" },
+   { MP_MEM,  "Out of heap" },
+   { MP_VAL,  "Value out of range" }
 };
 
 /* return a char * string for a given code */
@@ -31,9 +31,9 @@ const char *mp_error_to_string(int code)
 
    /* scan the lookup table for the given message */
    for (x = 0; x < (int)(sizeof(msgs) / sizeof(msgs[0])); x++) {
-       if (msgs[x].code == code) {
-          return msgs[x].msg;
-       }
+      if (msgs[x].code == code) {
+         return msgs[x].msg;
+      }
    }
 
    /* generic reply for invalid code */
diff --git a/libtommath/bn_fast_mp_invmod.c b/libtommath/bn_fast_mp_invmod.c
index 12f42de..7771136 100644
--- a/libtommath/bn_fast_mp_invmod.c
+++ b/libtommath/bn_fast_mp_invmod.c
@@ -15,131 +15,132 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* computes the modular inverse via binary extended euclidean algorithm, 
- * that is c = 1/a mod b 
+/* computes the modular inverse via binary extended euclidean algorithm,
+ * that is c = 1/a mod b
  *
- * Based on slow invmod except this is optimized for the case where b is 
+ * Based on slow invmod except this is optimized for the case where b is
  * odd as per HAC Note 14.64 on pp. 610
  */
-int fast_mp_invmod (mp_int * a, mp_int * b, mp_int * c)
+int fast_mp_invmod(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  x, y, u, v, B, D;
-  int     res, neg;
-
-  /* 2. [modified] b must be odd   */
-  if (mp_iseven (b) == MP_YES) {
-    return MP_VAL;
-  }
-
-  /* init all our temps */
-  if ((res = mp_init_multi(&x, &y, &u, &v, &B, &D, NULL)) != MP_OKAY) {
-     return res;
-  }
-
-  /* x == modulus, y == value to invert */
-  if ((res = mp_copy (b, &x)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-
-  /* we need y = |a| */
-  if ((res = mp_mod (a, b, &y)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-
-  /* 3. u=x, v=y, A=1, B=0, C=0,D=1 */
-  if ((res = mp_copy (&x, &u)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-  if ((res = mp_copy (&y, &v)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-  mp_set (&D, 1);
+   mp_int  x, y, u, v, B, D;
+   int     res, neg;
 
-top:
-  /* 4.  while u is even do */
-  while (mp_iseven (&u) == MP_YES) {
-    /* 4.1 u = u/2 */
-    if ((res = mp_div_2 (&u, &u)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    /* 4.2 if B is odd then */
-    if (mp_isodd (&B) == MP_YES) {
-      if ((res = mp_sub (&B, &x, &B)) != MP_OKAY) {
-        goto LBL_ERR;
-      }
-    }
-    /* B = B/2 */
-    if ((res = mp_div_2 (&B, &B)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
+   /* 2. [modified] b must be odd   */
+   if (mp_iseven(b) == MP_YES) {
+      return MP_VAL;
+   }
 
-  /* 5.  while v is even do */
-  while (mp_iseven (&v) == MP_YES) {
-    /* 5.1 v = v/2 */
-    if ((res = mp_div_2 (&v, &v)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    /* 5.2 if D is odd then */
-    if (mp_isodd (&D) == MP_YES) {
-      /* D = (D-x)/2 */
-      if ((res = mp_sub (&D, &x, &D)) != MP_OKAY) {
-        goto LBL_ERR;
-      }
-    }
-    /* D = D/2 */
-    if ((res = mp_div_2 (&D, &D)) != MP_OKAY) {
+   /* init all our temps */
+   if ((res = mp_init_multi(&x, &y, &u, &v, &B, &D, NULL)) != MP_OKAY) {
+      return res;
+   }
+
+   /* x == modulus, y == value to invert */
+   if ((res = mp_copy(b, &x)) != MP_OKAY) {
       goto LBL_ERR;
-    }
-  }
+   }
 
-  /* 6.  if u >= v then */
-  if (mp_cmp (&u, &v) != MP_LT) {
-    /* u = u - v, B = B - D */
-    if ((res = mp_sub (&u, &v, &u)) != MP_OKAY) {
+   /* we need y = |a| */
+   if ((res = mp_mod(a, b, &y)) != MP_OKAY) {
       goto LBL_ERR;
-    }
+   }
 
-    if ((res = mp_sub (&B, &D, &B)) != MP_OKAY) {
+   /* 3. u=x, v=y, A=1, B=0, C=0,D=1 */
+   if ((res = mp_copy(&x, &u)) != MP_OKAY) {
       goto LBL_ERR;
-    }
-  } else {
-    /* v - v - u, D = D - B */
-    if ((res = mp_sub (&v, &u, &v)) != MP_OKAY) {
+   }
+   if ((res = mp_copy(&y, &v)) != MP_OKAY) {
       goto LBL_ERR;
-    }
+   }
+   mp_set(&D, 1);
 
-    if ((res = mp_sub (&D, &B, &D)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
-
-  /* if not zero goto step 4 */
-  if (mp_iszero (&u) == MP_NO) {
-    goto top;
-  }
-
-  /* now a = C, b = D, gcd == g*v */
-
-  /* if v != 1 then there is no inverse */
-  if (mp_cmp_d (&v, 1) != MP_EQ) {
-    res = MP_VAL;
-    goto LBL_ERR;
-  }
-
-  /* b is now the inverse */
-  neg = a->sign;
-  while (D.sign == MP_NEG) {
-    if ((res = mp_add (&D, b, &D)) != MP_OKAY) {
+top:
+   /* 4.  while u is even do */
+   while (mp_iseven(&u) == MP_YES) {
+      /* 4.1 u = u/2 */
+      if ((res = mp_div_2(&u, &u)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+      /* 4.2 if B is odd then */
+      if (mp_isodd(&B) == MP_YES) {
+         if ((res = mp_sub(&B, &x, &B)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* B = B/2 */
+      if ((res = mp_div_2(&B, &B)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* 5.  while v is even do */
+   while (mp_iseven(&v) == MP_YES) {
+      /* 5.1 v = v/2 */
+      if ((res = mp_div_2(&v, &v)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+      /* 5.2 if D is odd then */
+      if (mp_isodd(&D) == MP_YES) {
+         /* D = (D-x)/2 */
+         if ((res = mp_sub(&D, &x, &D)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* D = D/2 */
+      if ((res = mp_div_2(&D, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* 6.  if u >= v then */
+   if (mp_cmp(&u, &v) != MP_LT) {
+      /* u = u - v, B = B - D */
+      if ((res = mp_sub(&u, &v, &u)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+
+      if ((res = mp_sub(&B, &D, &B)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   } else {
+      /* v - v - u, D = D - B */
+      if ((res = mp_sub(&v, &u, &v)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+
+      if ((res = mp_sub(&D, &B, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* if not zero goto step 4 */
+   if (mp_iszero(&u) == MP_NO) {
+      goto top;
+   }
+
+   /* now a = C, b = D, gcd == g*v */
+
+   /* if v != 1 then there is no inverse */
+   if (mp_cmp_d(&v, 1) != MP_EQ) {
+      res = MP_VAL;
       goto LBL_ERR;
-    }
-  }
-  mp_exch (&D, c);
-  c->sign = neg;
-  res = MP_OKAY;
-
-LBL_ERR:mp_clear_multi (&x, &y, &u, &v, &B, &D, NULL);
-  return res;
+   }
+
+   /* b is now the inverse */
+   neg = a->sign;
+   while (D.sign == MP_NEG) {
+      if ((res = mp_add(&D, b, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+   mp_exch(&D, c);
+   c->sign = neg;
+   res = MP_OKAY;
+
+LBL_ERR:
+   mp_clear_multi(&x, &y, &u, &v, &B, &D, NULL);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_fast_mp_montgomery_reduce.c b/libtommath/bn_fast_mp_montgomery_reduce.c
index 16d5ff7..f2c38bf 100644
--- a/libtommath/bn_fast_mp_montgomery_reduce.c
+++ b/libtommath/bn_fast_mp_montgomery_reduce.c
@@ -23,147 +23,147 @@
  *
  * Based on Algorithm 14.32 on pp.601 of HAC.
 */
-int fast_mp_montgomery_reduce (mp_int * x, mp_int * n, mp_digit rho)
+int fast_mp_montgomery_reduce(mp_int *x, mp_int *n, mp_digit rho)
 {
-  int     ix, res, olduse;
-  mp_word W[MP_WARRAY];
-
-  /* get old used count */
-  olduse = x->used;
-
-  /* grow a as required */
-  if (x->alloc < (n->used + 1)) {
-    if ((res = mp_grow (x, n->used + 1)) != MP_OKAY) {
-      return res;
-    }
-  }
-
-  /* first we have to get the digits of the input into
-   * an array of double precision words W[...]
-   */
-  {
-    mp_word *_W;
-    mp_digit *tmpx;
-
-    /* alias for the W[] array */
-    _W   = W;
-
-    /* alias for the digits of  x*/
-    tmpx = x->dp;
-
-    /* copy the digits of a into W[0..a->used-1] */
-    for (ix = 0; ix < x->used; ix++) {
-      *_W++ = *tmpx++;
-    }
-
-    /* zero the high words of W[a->used..m->used*2] */
-    for (; ix < ((n->used * 2) + 1); ix++) {
-      *_W++ = 0;
-    }
-  }
-
-  /* now we proceed to zero successive digits
-   * from the least significant upwards
-   */
-  for (ix = 0; ix < n->used; ix++) {
-    /* mu = ai * m' mod b
-     *
-     * We avoid a double precision multiplication (which isn't required)
-     * by casting the value down to a mp_digit.  Note this requires
-     * that W[ix-1] have  the carry cleared (see after the inner loop)
-     */
-    mp_digit mu;
-    mu = (mp_digit) (((W[ix] & MP_MASK) * rho) & MP_MASK);
-
-    /* a = a + mu * m * b**i
-     *
-     * This is computed in place and on the fly.  The multiplication
-     * by b**i is handled by offseting which columns the results
-     * are added to.
-     *
-     * Note the comba method normally doesn't handle carries in the
-     * inner loop In this case we fix the carry from the previous
-     * column since the Montgomery reduction requires digits of the
-     * result (so far) [see above] to work.  This is
-     * handled by fixing up one carry after the inner loop.  The
-     * carry fixups are done in order so after these loops the
-     * first m->used words of W[] have the carries fixed
-     */
-    {
-      int iy;
-      mp_digit *tmpn;
+   int     ix, res, olduse;
+   mp_word W[MP_WARRAY];
+
+   /* get old used count */
+   olduse = x->used;
+
+   /* grow a as required */
+   if (x->alloc < (n->used + 1)) {
+      if ((res = mp_grow(x, n->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* first we have to get the digits of the input into
+    * an array of double precision words W[...]
+    */
+   {
       mp_word *_W;
+      mp_digit *tmpx;
+
+      /* alias for the W[] array */
+      _W   = W;
 
-      /* alias for the digits of the modulus */
-      tmpn = n->dp;
+      /* alias for the digits of  x*/
+      tmpx = x->dp;
 
-      /* Alias for the columns set by an offset of ix */
-      _W = W + ix;
+      /* copy the digits of a into W[0..a->used-1] */
+      for (ix = 0; ix < x->used; ix++) {
+         *_W++ = *tmpx++;
+      }
 
-      /* inner loop */
-      for (iy = 0; iy < n->used; iy++) {
-          *_W++ += ((mp_word)mu) * ((mp_word)*tmpn++);
+      /* zero the high words of W[a->used..m->used*2] */
+      for (; ix < ((n->used * 2) + 1); ix++) {
+         *_W++ = 0;
       }
-    }
-
-    /* now fix carry for next digit, W[ix+1] */
-    W[ix + 1] += W[ix] >> ((mp_word) DIGIT_BIT);
-  }
-
-  /* now we have to propagate the carries and
-   * shift the words downward [all those least
-   * significant digits we zeroed].
-   */
-  {
-    mp_digit *tmpx;
-    mp_word *_W, *_W1;
-
-    /* nox fix rest of carries */
-
-    /* alias for current word */
-    _W1 = W + ix;
-
-    /* alias for next word, where the carry goes */
-    _W = W + ++ix;
-
-    for (; ix <= ((n->used * 2) + 1); ix++) {
-      *_W++ += *_W1++ >> ((mp_word) DIGIT_BIT);
-    }
-
-    /* copy out, A = A/b**n
-     *
-     * The result is A/b**n but instead of converting from an
-     * array of mp_word to mp_digit than calling mp_rshd
-     * we just copy them in the right order
-     */
-
-    /* alias for destination word */
-    tmpx = x->dp;
-
-    /* alias for shifted double precision result */
-    _W = W + n->used;
-
-    for (ix = 0; ix < (n->used + 1); ix++) {
-      *tmpx++ = (mp_digit)(*_W++ & ((mp_word) MP_MASK));
-    }
-
-    /* zero oldused digits, if the input a was larger than
-     * m->used+1 we'll have to clear the digits
-     */
-    for (; ix < olduse; ix++) {
-      *tmpx++ = 0;
-    }
-  }
-
-  /* set the max used and clamp */
-  x->used = n->used + 1;
-  mp_clamp (x);
-
-  /* if A >= m then A = A - m */
-  if (mp_cmp_mag (x, n) != MP_LT) {
-    return s_mp_sub (x, n, x);
-  }
-  return MP_OKAY;
+   }
+
+   /* now we proceed to zero successive digits
+    * from the least significant upwards
+    */
+   for (ix = 0; ix < n->used; ix++) {
+      /* mu = ai * m' mod b
+       *
+       * We avoid a double precision multiplication (which isn't required)
+       * by casting the value down to a mp_digit.  Note this requires
+       * that W[ix-1] have  the carry cleared (see after the inner loop)
+       */
+      mp_digit mu;
+      mu = (mp_digit)(((W[ix] & MP_MASK) * rho) & MP_MASK);
+
+      /* a = a + mu * m * b**i
+       *
+       * This is computed in place and on the fly.  The multiplication
+       * by b**i is handled by offseting which columns the results
+       * are added to.
+       *
+       * Note the comba method normally doesn't handle carries in the
+       * inner loop In this case we fix the carry from the previous
+       * column since the Montgomery reduction requires digits of the
+       * result (so far) [see above] to work.  This is
+       * handled by fixing up one carry after the inner loop.  The
+       * carry fixups are done in order so after these loops the
+       * first m->used words of W[] have the carries fixed
+       */
+      {
+         int iy;
+         mp_digit *tmpn;
+         mp_word *_W;
+
+         /* alias for the digits of the modulus */
+         tmpn = n->dp;
+
+         /* Alias for the columns set by an offset of ix */
+         _W = W + ix;
+
+         /* inner loop */
+         for (iy = 0; iy < n->used; iy++) {
+            *_W++ += ((mp_word)mu) * ((mp_word)*tmpn++);
+         }
+      }
+
+      /* now fix carry for next digit, W[ix+1] */
+      W[ix + 1] += W[ix] >> ((mp_word) DIGIT_BIT);
+   }
+
+   /* now we have to propagate the carries and
+    * shift the words downward [all those least
+    * significant digits we zeroed].
+    */
+   {
+      mp_digit *tmpx;
+      mp_word *_W, *_W1;
+
+      /* nox fix rest of carries */
+
+      /* alias for current word */
+      _W1 = W + ix;
+
+      /* alias for next word, where the carry goes */
+      _W = W + ++ix;
+
+      for (; ix <= ((n->used * 2) + 1); ix++) {
+         *_W++ += *_W1++ >> ((mp_word) DIGIT_BIT);
+      }
+
+      /* copy out, A = A/b**n
+       *
+       * The result is A/b**n but instead of converting from an
+       * array of mp_word to mp_digit than calling mp_rshd
+       * we just copy them in the right order
+       */
+
+      /* alias for destination word */
+      tmpx = x->dp;
+
+      /* alias for shifted double precision result */
+      _W = W + n->used;
+
+      for (ix = 0; ix < (n->used + 1); ix++) {
+         *tmpx++ = (mp_digit)(*_W++ & ((mp_word) MP_MASK));
+      }
+
+      /* zero oldused digits, if the input a was larger than
+       * m->used+1 we'll have to clear the digits
+       */
+      for (; ix < olduse; ix++) {
+         *tmpx++ = 0;
+      }
+   }
+
+   /* set the max used and clamp */
+   x->used = n->used + 1;
+   mp_clamp(x);
+
+   /* if A >= m then A = A - m */
+   if (mp_cmp_mag(x, n) != MP_LT) {
+      return s_mp_sub(x, n, x);
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_mul_digs.c b/libtommath/bn_fast_s_mp_mul_digs.c
index 641b574..558d151 100644
--- a/libtommath/bn_fast_s_mp_mul_digs.c
+++ b/libtommath/bn_fast_s_mp_mul_digs.c
@@ -17,39 +17,39 @@
 
 /* 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.
  *
  */
-int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs)
+int fast_s_mp_mul_digs(const mp_int *a, const mp_int *b, mp_int *c, int digs)
 {
-  int     olduse, res, pa, ix, iz;
-  mp_digit W[MP_WARRAY];
-  mp_word  _W;
+   int     olduse, res, pa, ix, iz;
+   mp_digit W[MP_WARRAY];
+   mp_word  _W;
 
-  /* grow the destination as required */
-  if (c->alloc < digs) {
-    if ((res = mp_grow (c, digs)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* grow the destination as required */
+   if (c->alloc < digs) {
+      if ((res = mp_grow(c, digs)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* number of output digits to produce */
-  pa = MIN(digs, a->used + b->used);
+   /* number of output digits to produce */
+   pa = MIN(digs, a->used + b->used);
 
-  /* clear the carry */
-  _W = 0;
-  for (ix = 0; ix < pa; ix++) { 
+   /* clear the carry */
+   _W = 0;
+   for (ix = 0; ix < pa; ix++) {
       int      tx, ty;
       int      iy;
       mp_digit *tmpx, *tmpy;
@@ -62,7 +62,7 @@ int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs
       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);
@@ -78,27 +78,27 @@ int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs
 
       /* make next carry */
       _W = _W >> ((mp_word)DIGIT_BIT);
-  }
+   }
 
-  /* setup dest */
-  olduse  = c->used;
-  c->used = pa;
+   /* setup dest */
+   olduse  = c->used;
+   c->used = pa;
 
-  {
-    mp_digit *tmpc;
-    tmpc = c->dp;
-    for (ix = 0; ix < (pa + 1); ix++) {
-      /* now extract the previous digit [below the carry] */
-      *tmpc++ = W[ix];
-    }
+   {
+      mp_digit *tmpc;
+      tmpc = c->dp;
+      for (ix = 0; ix < (pa + 1); ix++) {
+         /* now extract the previous digit [below the carry] */
+         *tmpc++ = W[ix];
+      }
 
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpc++ = 0;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpc++ = 0;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_mul_high_digs.c b/libtommath/bn_fast_s_mp_mul_high_digs.c
index 08f0355..588d80b 100644
--- a/libtommath/bn_fast_s_mp_mul_high_digs.c
+++ b/libtommath/bn_fast_s_mp_mul_high_digs.c
@@ -24,24 +24,24 @@
  *
  * Based on Algorithm 14.12 on pp.595 of HAC.
  */
-int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
+int fast_s_mp_mul_high_digs(mp_int *a, mp_int *b, mp_int *c, int digs)
 {
-  int     olduse, res, pa, ix, iz;
-  mp_digit W[MP_WARRAY];
-  mp_word  _W;
+   int     olduse, res, pa, ix, iz;
+   mp_digit W[MP_WARRAY];
+   mp_word  _W;
 
-  /* grow the destination as required */
-  pa = a->used + b->used;
-  if (c->alloc < pa) {
-    if ((res = mp_grow (c, pa)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* grow the destination as required */
+   pa = a->used + b->used;
+   if (c->alloc < pa) {
+      if ((res = mp_grow(c, pa)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* number of output digits to produce */
-  pa = a->used + b->used;
-  _W = 0;
-  for (ix = digs; ix < pa; ix++) { 
+   /* number of output digits to produce */
+   pa = a->used + b->used;
+   _W = 0;
+   for (ix = digs; ix < pa; ix++) {
       int      tx, ty, iy;
       mp_digit *tmpx, *tmpy;
 
@@ -53,7 +53,7 @@ int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
       tmpx = a->dp + tx;
       tmpy = b->dp + ty;
 
-      /* this is the number of times the loop will iterrate, essentially its 
+      /* this is the number of times the loop will iterrate, essentially its
          while (tx++ < a->used && ty-- >= 0) { ... }
        */
       iy = MIN(a->used-tx, ty+1);
@@ -68,28 +68,28 @@ int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
 
       /* make next carry */
       _W = _W >> ((mp_word)DIGIT_BIT);
-  }
-  
-  /* setup dest */
-  olduse  = c->used;
-  c->used = pa;
+   }
+
+   /* setup dest */
+   olduse  = c->used;
+   c->used = pa;
 
-  {
-    mp_digit *tmpc;
+   {
+      mp_digit *tmpc;
 
-    tmpc = c->dp + digs;
-    for (ix = digs; ix < pa; ix++) {
-      /* now extract the previous digit [below the carry] */
-      *tmpc++ = W[ix];
-    }
+      tmpc = c->dp + digs;
+      for (ix = digs; ix < pa; ix++) {
+         /* now extract the previous digit [below the carry] */
+         *tmpc++ = W[ix];
+      }
 
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpc++ = 0;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpc++ = 0;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_sqr.c b/libtommath/bn_fast_s_mp_sqr.c
index 282d303..161f785 100644
--- a/libtommath/bn_fast_s_mp_sqr.c
+++ b/libtommath/bn_fast_s_mp_sqr.c
@@ -16,32 +16,32 @@
  */
 
 /* 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.
 */
 
-int fast_s_mp_sqr (const mp_int * a, mp_int * b)
+int fast_s_mp_sqr(const mp_int *a, mp_int *b)
 {
-  int       olduse, res, pa, ix, iz;
-  mp_digit   W[MP_WARRAY], *tmpx;
-  mp_word   W1;
-
-  /* grow the destination as required */
-  pa = a->used + a->used;
-  if (b->alloc < pa) {
-    if ((res = mp_grow (b, pa)) != MP_OKAY) {
-      return res;
-    }
-  }
-
-  /* number of output digits to produce */
-  W1 = 0;
-  for (ix = 0; ix < pa; ix++) { 
+   int       olduse, res, pa, ix, iz;
+   mp_digit   W[MP_WARRAY], *tmpx;
+   mp_word   W1;
+
+   /* grow the destination as required */
+   pa = a->used + a->used;
+   if (b->alloc < pa) {
+      if ((res = mp_grow(b, pa)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* number of output digits to produce */
+   W1 = 0;
+   for (ix = 0; ix < pa; ix++) {
       int      tx, ty, iy;
       mp_word  _W;
       mp_digit *tmpy;
@@ -62,7 +62,7 @@ int fast_s_mp_sqr (const mp_int * a, mp_int * b)
        */
       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
        */
@@ -86,26 +86,26 @@ int fast_s_mp_sqr (const mp_int * a, mp_int * b)
 
       /* make next carry */
       W1 = _W >> ((mp_word)DIGIT_BIT);
-  }
-
-  /* setup dest */
-  olduse  = b->used;
-  b->used = a->used+a->used;
-
-  {
-    mp_digit *tmpb;
-    tmpb = b->dp;
-    for (ix = 0; ix < pa; ix++) {
-      *tmpb++ = W[ix] & MP_MASK;
-    }
-
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpb++ = 0;
-    }
-  }
-  mp_clamp (b);
-  return MP_OKAY;
+   }
+
+   /* setup dest */
+   olduse  = b->used;
+   b->used = a->used+a->used;
+
+   {
+      mp_digit *tmpb;
+      tmpb = b->dp;
+      for (ix = 0; ix < pa; ix++) {
+         *tmpb++ = W[ix] & MP_MASK;
+      }
+
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpb++ = 0;
+      }
+   }
+   mp_clamp(b);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_2expt.c b/libtommath/bn_mp_2expt.c
index 989bb9f..bdc868f 100644
--- a/libtommath/bn_mp_2expt.c
+++ b/libtommath/bn_mp_2expt.c
@@ -15,31 +15,31 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* computes a = 2**b 
+/* computes a = 2**b
  *
  * Simple algorithm which zeroes the int, grows it then just sets one bit
  * as required.
  */
 int
-mp_2expt (mp_int * a, int b)
+mp_2expt(mp_int *a, int b)
 {
-  int     res;
+   int     res;
 
-  /* zero a as per default */
-  mp_zero (a);
+   /* zero a as per default */
+   mp_zero(a);
 
-  /* grow a to accomodate the single bit */
-  if ((res = mp_grow (a, (b / DIGIT_BIT) + 1)) != MP_OKAY) {
-    return res;
-  }
+   /* grow a to accomodate the single bit */
+   if ((res = mp_grow(a, (b / DIGIT_BIT) + 1)) != MP_OKAY) {
+      return res;
+   }
 
-  /* set the used count of where the bit will go */
-  a->used = (b / DIGIT_BIT) + 1;
+   /* set the used count of where the bit will go */
+   a->used = (b / DIGIT_BIT) + 1;
 
-  /* put the single bit in its place */
-  a->dp[b / DIGIT_BIT] = ((mp_digit)1) << (b % DIGIT_BIT);
+   /* put the single bit in its place */
+   a->dp[b / DIGIT_BIT] = ((mp_digit)1) << (b % DIGIT_BIT);
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_abs.c b/libtommath/bn_mp_abs.c
index e7c5e25..343a102 100644
--- a/libtommath/bn_mp_abs.c
+++ b/libtommath/bn_mp_abs.c
@@ -15,26 +15,26 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* b = |a| 
+/* b = |a|
  *
  * Simple function copies the input and fixes the sign to positive
  */
 int
-mp_abs (mp_int * a, mp_int * b)
+mp_abs(mp_int *a, mp_int *b)
 {
-  int     res;
+   int     res;
 
-  /* copy a to b */
-  if (a != b) {
-     if ((res = mp_copy (a, b)) != MP_OKAY) {
-       return res;
-     }
-  }
+   /* copy a to b */
+   if (a != b) {
+      if ((res = mp_copy(a, b)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* force the sign of b to positive */
-  b->sign = MP_ZPOS;
+   /* force the sign of b to positive */
+   b->sign = MP_ZPOS;
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_add.c b/libtommath/bn_mp_add.c
index a68c02e..d31d5a0 100644
--- a/libtommath/bn_mp_add.c
+++ b/libtommath/bn_mp_add.c
@@ -16,34 +16,34 @@
  */
 
 /* high level addition (handles signs) */
-int mp_add (const mp_int * a, const mp_int * b, mp_int * c)
+int mp_add(const mp_int *a, const mp_int *b, mp_int *c)
 {
-  int     sa, sb, res;
+   int     sa, sb, res;
 
-  /* get sign of both inputs */
-  sa = a->sign;
-  sb = b->sign;
+   /* get sign of both inputs */
+   sa = a->sign;
+   sb = b->sign;
 
-  /* handle two cases, not four */
-  if (sa == sb) {
-    /* both positive or both negative */
-    /* add their magnitudes, copy the sign */
-    c->sign = sa;
-    res = s_mp_add (a, b, c);
-  } else {
-    /* one positive, the other negative */
-    /* subtract the one with the greater magnitude from */
-    /* the one of the lesser magnitude.  The result gets */
-    /* the sign of the one with the greater magnitude. */
-    if (mp_cmp_mag (a, b) == MP_LT) {
-      c->sign = sb;
-      res = s_mp_sub (b, a, c);
-    } else {
+   /* handle two cases, not four */
+   if (sa == sb) {
+      /* both positive or both negative */
+      /* add their magnitudes, copy the sign */
       c->sign = sa;
-      res = s_mp_sub (a, b, c);
-    }
-  }
-  return res;
+      res = s_mp_add(a, b, c);
+   } else {
+      /* one positive, the other negative */
+      /* subtract the one with the greater magnitude from */
+      /* the one of the lesser magnitude.  The result gets */
+      /* the sign of the one with the greater magnitude. */
+      if (mp_cmp_mag(a, b) == MP_LT) {
+         c->sign = sb;
+         res = s_mp_sub(b, a, c);
+      } else {
+         c->sign = sa;
+         res = s_mp_sub(a, b, c);
+      }
+   }
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_add_d.c b/libtommath/bn_mp_add_d.c
index c2f2e0e..5270d27 100644
--- a/libtommath/bn_mp_add_d.c
+++ b/libtommath/bn_mp_add_d.c
@@ -17,93 +17,93 @@
 
 /* single digit addition */
 int
-mp_add_d (const mp_int * a, mp_digit b, mp_int * c)
+mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
 {
-  int     res, ix, oldused;
-  mp_digit *tmpa, *tmpc, mu;
-
-  /* grow c as required */
-  if (c->alloc < (a->used + 1)) {
-     if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
-        return res;
-     }
-  }
-
-  /* if a is negative and |a| >= b, call c = |a| - b */
-  if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) {
-     mp_int a_ = *a;
-     /* temporarily fix sign of a */
-     a_.sign = MP_ZPOS;
-
-     /* c = |a| - b */
-     res = mp_sub_d(&a_, b, c);
-
-     /* fix sign  */
-     c->sign = MP_NEG;
-
-     /* clamp */
-     mp_clamp(c);
-
-     return res;
-  }
-
-  /* old number of used digits in c */
-  oldused = c->used;
-
-  /* source alias */
-  tmpa    = a->dp;
-
-  /* destination alias */
-  tmpc    = c->dp;
-
-  /* if a is positive */
-  if (a->sign == MP_ZPOS) {
-     /* add digit, after this we're propagating
-      * the carry.
-      */
-     *tmpc   = *tmpa++ + b;
-     mu      = *tmpc >> DIGIT_BIT;
-     *tmpc++ &= MP_MASK;
-
-     /* now handle rest of the digits */
-     for (ix = 1; ix < a->used; ix++) {
-        *tmpc   = *tmpa++ + mu;
-        mu      = *tmpc >> DIGIT_BIT;
-        *tmpc++ &= MP_MASK;
-     }
-     /* set final carry */
-     ix++;
-     *tmpc++  = mu;
-
-     /* setup size */
-     c->used = a->used + 1;
-  } else {
-     /* a was negative and |a| < b */
-     c->used  = 1;
-
-     /* the result is a single digit */
-     if (a->used == 1) {
-        *tmpc++  =  b - a->dp[0];
-     } else {
-        *tmpc++  =  b;
-     }
-
-     /* setup count so the clearing of oldused
-      * can fall through correctly
-      */
-     ix       = 1;
-  }
-
-  /* sign always positive */
-  c->sign = MP_ZPOS;
-
-  /* now zero to oldused */
-  while (ix++ < oldused) {
-     *tmpc++ = 0;
-  }
-  mp_clamp(c);
-
-  return MP_OKAY;
+   int     res, ix, oldused;
+   mp_digit *tmpa, *tmpc, mu;
+
+   /* grow c as required */
+   if (c->alloc < (a->used + 1)) {
+      if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* if a is negative and |a| >= b, call c = |a| - b */
+   if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) {
+      mp_int a_ = *a;
+      /* temporarily fix sign of a */
+      a_.sign = MP_ZPOS;
+
+      /* c = |a| - b */
+      res = mp_sub_d(&a_, b, c);
+
+      /* fix sign  */
+      c->sign = MP_NEG;
+
+      /* clamp */
+      mp_clamp(c);
+
+      return res;
+   }
+
+   /* old number of used digits in c */
+   oldused = c->used;
+
+   /* source alias */
+   tmpa    = a->dp;
+
+   /* destination alias */
+   tmpc    = c->dp;
+
+   /* if a is positive */
+   if (a->sign == MP_ZPOS) {
+      /* add digit, after this we're propagating
+       * the carry.
+       */
+      *tmpc   = *tmpa++ + b;
+      mu      = *tmpc >> DIGIT_BIT;
+      *tmpc++ &= MP_MASK;
+
+      /* now handle rest of the digits */
+      for (ix = 1; ix < a->used; ix++) {
+         *tmpc   = *tmpa++ + mu;
+         mu      = *tmpc >> DIGIT_BIT;
+         *tmpc++ &= MP_MASK;
+      }
+      /* set final carry */
+      ix++;
+      *tmpc++  = mu;
+
+      /* setup size */
+      c->used = a->used + 1;
+   } else {
+      /* a was negative and |a| < b */
+      c->used  = 1;
+
+      /* the result is a single digit */
+      if (a->used == 1) {
+         *tmpc++  =  b - a->dp[0];
+      } else {
+         *tmpc++  =  b;
+      }
+
+      /* setup count so the clearing of oldused
+       * can fall through correctly
+       */
+      ix       = 1;
+   }
+
+   /* sign always positive */
+   c->sign = MP_ZPOS;
+
+   /* now zero to oldused */
+   while (ix++ < oldused) {
+      *tmpc++ = 0;
+   }
+   mp_clamp(c);
+
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_addmod.c b/libtommath/bn_mp_addmod.c