re-format everything through astyle. Taken from libtom/libtommath (pull request [https://github.com/libtom/libtommath/pull/85|85])

129 files changed, 5787 insertions, 5696 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 a1015af..763dbb1 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 (mp_int * a, mp_int * b, mp_int * c, int digs)
+int fast_s_mp_mul_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 */
-  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 (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 
+      /* 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 (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;
-    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 f435af9..ceed82b 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 (mp_int * a, mp_int * b)
+int fast_s_mp_sqr(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 (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 (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..701144c 100644
--- a/libtommath/bn_mp_2expt.c
+++ b/libtommath/bn_mp_2expt.c
@@ -15,31 +15,30 @@
  * 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)
+int 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..d5fc012 100644
--- a/libtommath/bn_mp_abs.c
+++ b/libtommath/bn_mp_abs.c
@@ -15,26 +15,25 @@
  * 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)
+int 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 bdb166f..4df4c81 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 (mp_int * a, mp_int * b, mp_int * c)
+int mp_add(mp_int *a, 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 fd1a186..1e6ff63 100644
--- a/libtommath/bn_mp_add_d.c
+++ b/libtommath/bn_mp_add_d.c
@@ -16,93 +16,92 @@
  */
 
 /* single digit addition */
-int
-mp_add_d (mp_int * a, mp_digit b, mp_int * c)
+int mp_add_d(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))) {
-     /* temporarily fix sign of a */
-     a->sign = MP_ZPOS;
-
-     /* c = |a| - b */
-     res = mp_sub_d(a, b, c);
-
-     /* fix sign  */
-     a->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))) {
+      /* temporarily fix sign of a */
+      a->sign = MP_ZPOS;
+
+      /* c = |a| - b */
+      res = mp_sub_d(a, b, c);
+
+      /* fix sign  */
+      a->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
index dc06788..229a716 100644
--- a/libtommath/bn_mp_addmod.c
+++ b/libtommath/bn_mp_addmod.c
@@ -16,23 +16,22 @@
  */
 
 /* d = a + b (mod c) */
-int
-mp_addmod (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+int mp_addmod(mp_int *a, mp_int *b, mp_int *c, mp_int *d)
 {
-  int     res;
-  mp_int  t;
+   int     res;
+   mp_int  t;
 
-  if ((res = mp_init (&t)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init(&t)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_add (a, b, &t)) != MP_OKAY) {
-    mp_clear (&t);
-    return res;
-  }
-  res = mp_mod (&t, c, d);
-  mp_clear (&t);
-  return res;
+   if ((res = mp_add(a, b, &t)) != MP_OKAY) {
+      mp_clear(&t);
+      return res;
+   }
+   res = mp_mod(&t, c, d);
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_and.c b/libtommath/bn_mp_and.c
index 53008a5..2f1472a 100644
--- a/libtommath/bn_mp_and.c
+++ b/libtommath/bn_mp_and.c
@@ -16,39 +16,38 @@
  */
 
 /* AND two ints together */
-int
-mp_and (mp_int * a, mp_int * b, mp_int * c)
+int mp_and(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res, ix, px;
-  mp_int  t, *x;
+   int     res, ix, px;
+   mp_int  t, *x;
 
-  if (a->used > b->used) {
-    if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-      return res;
-    }
-    px = b->used;
-    x = b;
-  } else {
-    if ((res = mp_init_copy (&t, b)) != MP_OKAY) {
-      return res;
-    }
-    px = a->used;
-    x = a;
-  }
+   if (a->used > b->used) {
+      if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+         return res;
+      }
+      px = b->used;
+      x = b;
+   } else {
+      if ((res = mp_init_copy(&t, b)) != MP_OKAY) {
+         return res;
+      }
+      px = a->used;
+      x = a;
+   }
 
-  for (ix = 0; ix < px; ix++) {
-    t.dp[ix] &= x->dp[ix];
-  }
+   for (ix = 0; ix < px; ix++) {
+      t.dp[ix] &= x->dp[ix];
+   }
 
-  /* zero digits above the last from the smallest mp_int */
-  for (; ix < t.used; ix++) {
-    t.dp[ix] = 0;
-  }
+   /* zero digits above the last from the smallest mp_int */
+   for (; ix < t.used; ix++) {
+      t.dp[ix] = 0;
+   }
 
-  mp_clamp (&t);
-  mp_exch (c, &t);
-  mp_clear (&t);
-  return MP_OKAY;
+   mp_clamp(&t);
+   mp_exch(c, &t);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_clamp.c b/libtommath/bn_mp_clamp.c
index 2c0a1a6..3853914 100644
--- a/libtommath/bn_mp_clamp.c
+++ b/libtommath/bn_mp_clamp.c
@@ -15,27 +15,26 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* 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
  * Typically very fast.  Also fixes the sign if there
  * are no more leading digits
  */
-void
-mp_clamp (mp_int * a)
+void mp_clamp(mp_int *a)
 {
-  /* decrease used while the most significant digit is
-   * zero.
-   */
-  while ((a->used > 0) && (a->dp[a->used - 1] == 0)) {
-    --(a->used);
-  }
+   /* decrease used while the most significant digit is
+    * zero.
+    */
+   while ((a->used > 0) && (a->dp[a->used - 1] == 0)) {
+      --(a->used);
+   }
 
-  /* reset the sign flag if used == 0 */
-  if (a->used == 0) {
-    a->sign = MP_ZPOS;
-  }
+   /* reset the sign flag if used == 0 */
+   if (a->used == 0) {
+      a->sign = MP_ZPOS;
+   }
 }
 #endif
 
diff --git a/libtommath/bn_mp_clear.c b/libtommath/bn_mp_clear.c
index 97f3db0..fcf4d61 100644
--- a/libtommath/bn_mp_clear.c
+++ b/libtommath/bn_mp_clear.c
@@ -16,26 +16,25 @@
  */
 
 /* clear one (frees)  */
-void
-mp_clear (mp_int * a)
+void mp_clear(mp_int *a)
 {
-  int i;
+   int i;
 
-  /* only do anything if a hasn't been freed previously */
-  if (a->dp != NULL) {
-    /* first zero the digits */
-    for (i = 0; i < a->used; i++) {
-        a->dp[i] = 0;
-    }
+   /* only do anything if a hasn't been freed previously */
+   if (a->dp != NULL) {
+      /* first zero the digits */
+      for (i = 0; i < a->used; i++) {
+         a->dp[i] = 0;
+      }
 
-    /* free ram */
-    XFREE(a->dp);
+      /* free ram */
+      XFREE(a->dp);
 
-    /* reset members to make debugging easier */
-    a->dp    = NULL;
-    a->alloc = a->used = 0;
-    a->sign  = MP_ZPOS;
-  }
+      /* reset members to make debugging easier */
+      a->dp    = NULL;
+      a->alloc = a->used = 0;
+      a->sign  = MP_ZPOS;
+   }
 }
 #endif
 
diff --git a/libtommath/bn_mp_clear_multi.c b/libtommath/bn_mp_clear_multi.c
index bd4b232..284fab8 100644
--- a/libtommath/bn_mp_clear_multi.c
+++ b/libtommath/bn_mp_clear_multi.c
@@ -16,16 +16,16 @@
  */
 #include <stdarg.h>
 
-void mp_clear_multi(mp_int *mp, ...) 
+void mp_clear_multi(mp_int *mp, ...)
 {
-    mp_int* next_mp = mp;
-    va_list args;
-    va_start(args, mp);
-    while (next_mp != NULL) {
-        mp_clear(next_mp);
-        next_mp = va_arg(args, mp_int*);
-    }
-    va_end(args);
+   mp_int *next_mp = mp;
+   va_list args;
+   va_start(args, mp);
+   while (next_mp != NULL) {
+      mp_clear(next_mp);
+      next_mp = va_arg(args, mp_int *);
+   }
+   va_end(args);
 }
 #endif
 
diff --git a/libtommath/bn_mp_cmp.c b/libtommath/bn_mp_cmp.c
index e757ddf..9047060 100644
--- a/libtommath/bn_mp_cmp.c
+++ b/libtommath/bn_mp_cmp.c
@@ -16,25 +16,24 @@
  */
 
 /* compare two ints (signed)*/
-int
-mp_cmp (mp_int * a, mp_int * b)
+int mp_cmp(mp_int *a, mp_int *b)
 {
-  /* compare based on sign */
-  if (a->sign != b->sign) {
-     if (a->sign == MP_NEG) {
-        return MP_LT;
-     } else {
-        return MP_GT;
-     }
-  }
-  
-  /* compare digits */
-  if (a->sign == MP_NEG) {
-     /* if negative compare opposite direction */
-     return mp_cmp_mag(b, a);
-  } else {
-     return mp_cmp_mag(a, b);
-  }
+   /* compare based on sign */
+   if (a->sign != b->sign) {
+      if (a->sign == MP_NEG) {
+         return MP_LT;
+      } else {
+         return MP_GT;
+      }
+   }
+
+   /* compare digits */
+   if (a->sign == MP_NEG) {
+      /* if negative compare opposite direction */
+      return mp_cmp_mag(b, a);
+   } else {
+      return mp_cmp_mag(a, b);
+   }
 }
 #endif
 
diff --git a/libtommath/bn_mp_cmp_d.c b/libtommath/bn_mp_cmp_d.c
index 3f5ebae..27a546d 100644
--- a/libtommath/bn_mp_cmp_d.c
+++ b/libtommath/bn_mp_cmp_d.c
@@ -16,26 +16,26 @@
  */
 
 /* compare a digit */
-int mp_cmp_d(mp_int * a, mp_digit b)
+int mp_cmp_d(mp_int *a, mp_digit b)
 {
-  /* compare based on sign */
-  if (a->sign == MP_NEG) {
-    return MP_LT;
-  }
+   /* compare based on sign */
+   if (a->sign == MP_NEG) {
+      return MP_LT;
+   }
 
-  /* compare based on magnitude */
-  if (a->used > 1) {
-    return MP_GT;
-  }
+   /* compare based on magnitude */
+   if (a->used > 1) {
+      return MP_GT;
+   }
 
-  /* compare the only digit of a to b */
-  if (a->dp[0] > b) {
-    return MP_GT;
-  } else if (a->dp[0] < b) {
-    return MP_LT;
-  } else {
-    return MP_EQ;
-  }
+   /* compare the only digit of a to b */
+   if (a->dp[0] > b) {
+      return MP_GT;
+   } else if (a->dp[0] < b) {
+      return MP_LT;
+   } else {
+      return MP_EQ;
+   }
 }
 #endif
 
diff --git a/libtommath/bn_mp_cmp_mag.c b/libtommath/bn_mp_cmp_mag.c
index 7ceda97..ca2bc88 100644
--- a/libtommath/bn_mp_cmp_mag.c
+++ b/libtommath/bn_mp_cmp_mag.c
@@ -16,37 +16,37 @@
  */
 
 /* compare maginitude of two ints (unsigned) */
-int mp_cmp_mag (mp_int * a, mp_int * b)
+int mp_cmp_mag(mp_int *a, mp_int *b)
 {
-  int     n;
-  mp_digit *tmpa, *tmpb;
-
-  /* compare based on # of non-zero digits */
-  if (a->used > b->used) {
-    return MP_GT;
-  }
-  
-  if (a->used < b->used) {
-    return MP_LT;
-  }
-
-  /* alias for a */
-  tmpa = a->dp + (a->used - 1);
-
-  /* alias for b */
-  tmpb = b->dp + (a->used - 1);
-
-  /* compare based on digits  */
-  for (n = 0; n < a->used; ++n, --tmpa, --tmpb) {
-    if (*tmpa > *tmpb) {
+   int     n;
+   mp_digit *tmpa, *tmpb;
+
+   /* compare based on # of non-zero digits */
+   if (a->used > b->used) {
       return MP_GT;
-    }
+   }
 
-    if (*tmpa < *tmpb) {
+   if (a->used < b->used) {
       return MP_LT;
-    }
-  }
-  return MP_EQ;
+   }
+
+   /* alias for a */
+   tmpa = a->dp + (a->used - 1);
+
+   /* alias for b */
+   tmpb = b->dp + (a->used - 1);
+
+   /* compare based on digits  */
+   for (n = 0; n < a->used; ++n, --tmpa, --tmpb) {
+      if (*tmpa > *tmpb) {
+         return MP_GT;
+      }
+
+      if (*tmpa < *tmpb) {
+         return MP_LT;
+      }
+   }
+   return MP_EQ;
 }
 #endif
 
diff --git a/libtommath/bn_mp_cnt_lsb.c b/libtommath/bn_mp_cnt_lsb.c
index bf201b5..7273655 100644
--- a/libtommath/bn_mp_cnt_lsb.c
+++ b/libtommath/bn_mp_cnt_lsb.c
@@ -15,7 +15,7 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-static const int lnz[16] = { 
+static const int lnz[16] = {
    4, 0, 1, 0, 2, 0, 1, 0, 3, 0, 1, 0, 2, 0, 1, 0
 };
 
diff --git a/libtommath/bn_mp_copy.c b/libtommath/bn_mp_copy.c
index 84e839e..bed03b2 100644
--- a/libtommath/bn_mp_copy.c
+++ b/libtommath/bn_mp_copy.c
@@ -16,50 +16,49 @@
  */
 
 /* copy, b = a */
-int
-mp_copy (mp_int * a, mp_int * b)
+int mp_copy(mp_int *a, mp_int *b)
 {
-  int     res, n;
+   int     res, n;
 
-  /* if dst == src do nothing */
-  if (a == b) {
-    return MP_OKAY;
-  }
+   /* if dst == src do nothing */
+   if (a == b) {
+      return MP_OKAY;
+   }
 
-  /* grow dest */
-  if (b->alloc < a->used) {
-     if ((res = mp_grow (b, a->used)) != MP_OKAY) {
-        return res;
-     }
-  }
+   /* grow dest */
+   if (b->alloc < a->used) {
+      if ((res = mp_grow(b, a->used)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* zero b and copy the parameters over */
-  {
-    mp_digit *tmpa, *tmpb;
+   /* zero b and copy the parameters over */
+   {
+      mp_digit *tmpa, *tmpb;
 
-    /* pointer aliases */
+      /* pointer aliases */
 
-    /* source */
-    tmpa = a->dp;
+      /* source */
+      tmpa = a->dp;
 
-    /* destination */
-    tmpb = b->dp;
+      /* destination */
+      tmpb = b->dp;
 
-    /* copy all the digits */
-    for (n = 0; n < a->used; n++) {
-      *tmpb++ = *tmpa++;
-    }
+      /* copy all the digits */
+      for (n = 0; n < a->used; n++) {
+         *tmpb++ = *tmpa++;
+      }
 
-    /* clear high digits */
-    for (; n < b->used; n++) {
-      *tmpb++ = 0;
-    }
-  }
+      /* clear high digits */
+      for (; n < b->used; n++) {
+         *tmpb++ = 0;
+      }
+   }
 
-  /* copy used count and sign */
-  b->used = a->used;
-  b->sign = a->sign;
-  return MP_OKAY;
+   /* copy used count and sign */
+   b->used = a->used;
+   b->sign = a->sign;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_count_bits.c b/libtommath/bn_mp_count_bits.c
index ff558eb..dea364f 100644
--- a/libtommath/bn_mp_count_bits.c
+++ b/libtommath/bn_mp_count_bits.c
@@ -16,27 +16,26 @@
  */
 
 /* returns the number of bits in an int */
-int
-mp_count_bits (mp_int * a)
+int mp_count_bits(mp_int *a)
 {
-  int     r;
-  mp_digit q;
+   int     r;
+   mp_digit q;
 
-  /* shortcut */
-  if (a->used == 0) {
-    return 0;
-  }
+   /* shortcut */
+   if (a->used == 0) {
+      return 0;
+   }
 
-  /* 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)) {
-    ++r;
-    q >>= ((mp_digit) 1);
-  }
-  return r;
+   /* 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)) {
+      ++r;
+      q >>= ((mp_digit) 1);
+   }
+   return r;
 }
 #endif
 
diff --git a/libtommath/bn_mp_div.c b/libtommath/bn_mp_div.c
index 0890e65..fdb3453 100644
--- a/libtommath/bn_mp_div.c
+++ b/libtommath/bn_mp_div.c
@@ -18,68 +18,68 @@
 #ifdef BN_MP_DIV_SMALL
 
 /* slower bit-bang division... also smaller */
-int mp_div(mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+int mp_div(mp_int *a, mp_int *b, mp_int *c, mp_int *d)
 {
    mp_int ta, tb, tq, q;
    int    res, n, n2;
 
-  /* is divisor zero ? */
-  if (mp_iszero (b) == MP_YES) {
-    return MP_VAL;
-  }
-
-  /* if a < b then q=0, r = a */
-  if (mp_cmp_mag (a, b) == MP_LT) {
-    if (d != NULL) {
-      res = mp_copy (a, d);
-    } else {
-      res = MP_OKAY;
-    }
-    if (c != NULL) {
-      mp_zero (c);
-    }
-    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_mul_2d(&tb, n, &tb)) != MP_OKAY) ||
-      ((res = mp_mul_2d(&tq, n, &tq)) != MP_OKAY)) {
+   /* is divisor zero ? */
+   if (mp_iszero(b) == MP_YES) {
+      return MP_VAL;
+   }
+
+   /* if a < b then q=0, r = a */
+   if (mp_cmp_mag(a, b) == MP_LT) {
+      if (d != NULL) {
+         res = mp_copy(a, d);
+      } else {
+         res = MP_OKAY;
+      }
+      if (c != NULL) {
+         mp_zero(c);
+      }
+      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_mul_2d(&tb, n, &tb)) != MP_OKAY) ||
+       ((res = mp_mul_2d(&tq, n, &tq)) != MP_OKAY)) {
       goto LBL_ERR;
-  }
-
-  while (n-- >= 0) {
-     if (mp_cmp(&tb, &ta) != MP_GT) {
-        if (((res = mp_sub(&ta, &tb, &ta)) != MP_OKAY) ||
-            ((res = mp_add(&q, &tq, &q)) != MP_OKAY)) {
-           goto LBL_ERR;
-        }
-     }
-     if (((res = mp_div_2d(&tb, 1, &tb, NULL)) != MP_OKAY) ||
-         ((res = mp_div_2d(&tq, 1, &tq, NULL)) != MP_OKAY)) {
-           goto LBL_ERR;
-     }
-  }
-
-  /* now q == quotient and ta == remainder */
-  n  = a->sign;
-  n2 = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
-  if (c != NULL) {
-     mp_exch(c, &q);
-     c->sign  = (mp_iszero(c) == MP_YES) ? MP_ZPOS : n2;
-  }
-  if (d != NULL) {
-     mp_exch(d, &ta);
-     d->sign = (mp_iszero(d) == MP_YES) ? MP_ZPOS : n;
-  }
+   }
+
+   while (n-- >= 0) {
+      if (mp_cmp(&tb, &ta) != MP_GT) {
+         if (((res = mp_sub(&ta, &tb, &ta)) != MP_OKAY) ||
+             ((res = mp_add(&q, &tq, &q)) != MP_OKAY)) {
+            goto LBL_ERR;
+         }
+      }
+      if (((res = mp_div_2d(&tb, 1, &tb, NULL)) != MP_OKAY) ||
+          ((res = mp_div_2d(&tq, 1, &tq, NULL)) != MP_OKAY)) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* now q == quotient and ta == remainder */
+   n  = a->sign;
+   n2 = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
+   if (c != NULL) {
+      mp_exch(c, &q);
+      c->sign  = (mp_iszero(c) == MP_YES) ? MP_ZPOS : n2;
+   }
+   if (d != NULL) {
+      mp_exch(d, &ta);
+      d->sign = (mp_iszero(d) == MP_YES) ? MP_ZPOS : n;
+   }
 LBL_ERR:
    mp_clear_multi(&ta, &tb, &tq, &q, NULL);
    return res;
@@ -100,190 +100,195 @@ LBL_ERR:
  * The overall algorithm is as described as
  * 14.20 from HAC but fixed to treat these cases.
 */
-int mp_div (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+int mp_div(mp_int *a, mp_int *b, mp_int *c, mp_int *d)
 {
-  mp_int  q, x, y, t1, t2;
-  int     res, n, t, i, norm, neg;
-
-  /* is divisor zero ? */
-  if (mp_iszero (b) == MP_YES) {
-    return MP_VAL;
-  }
-
-  /* if a < b then q=0, r = a */
-  if (mp_cmp_mag (a, b) == MP_LT) {
-    if (d != NULL) {
-      res = mp_copy (a, d);
-    } else {
-      res = MP_OKAY;
-    }
-    if (c != NULL) {
-      mp_zero (c);
-    }
-    return res;
-  }
-
-  if ((res = mp_init_size (&q, a->used + 2)) != MP_OKAY) {
-    return res;
-  }
-  q.used = a->used + 2;
-
-  if ((res = mp_init (&t1)) != MP_OKAY) {
-    goto LBL_Q;
-  }
-
-  if ((res = mp_init (&t2)) != MP_OKAY) {
-    goto LBL_T1;
-  }
-
-  if ((res = mp_init_copy (&x, a)) != MP_OKAY) {
-    goto LBL_T2;
-  }
-
-  if ((res = mp_init_copy (&y, b)) != MP_OKAY) {
-    goto LBL_X;
-  }
-
-  /* fix the sign */
-  neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
-  x.sign = y.sign = MP_ZPOS;
-
-  /* normalize both x and y, ensure that y >= b/2, [b == 2**DIGIT_BIT] */
-  norm = mp_count_bits(&y) % DIGIT_BIT;
-  if (norm < (int)(DIGIT_BIT-1)) {
-     norm = (DIGIT_BIT-1) - norm;
-     if ((res = mp_mul_2d (&x, norm, &x)) != MP_OKAY) {
-       goto LBL_Y;
-     }
-     if ((res = mp_mul_2d (&y, norm, &y)) != MP_OKAY) {
-       goto LBL_Y;
-     }
-  } else {
-     norm = 0;
-  }
-
-  /* note hac does 0 based, so if used==5 then its 0,1,2,3,4, e.g. use 4 */
-  n = x.used - 1;
-  t = y.used - 1;
-
-  /* while (x >= y*b**n-t) do { q[n-t] += 1; x -= y*b**{n-t} } */
-  if ((res = mp_lshd (&y, n - t)) != MP_OKAY) { /* y = y*b**{n-t} */
-    goto LBL_Y;
-  }
-
-  while (mp_cmp (&x, &y) != MP_LT) {
-    ++(q.dp[n - t]);
-    if ((res = mp_sub (&x, &y, &x)) != MP_OKAY) {
-      goto LBL_Y;
-    }
-  }
-
-  /* reset y by shifting it back down */
-  mp_rshd (&y, n - t);
-
-  /* step 3. for i from n down to (t + 1) */
-  for (i = n; i >= (t + 1); i--) {
-    if (i > x.used) {
-      continue;
-    }
-
-    /* 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);
-    } else {
-      mp_word tmp;
-      tmp = ((mp_word) x.dp[i]) << ((mp_word) DIGIT_BIT);
-      tmp |= ((mp_word) x.dp[i - 1]);
-      tmp /= ((mp_word) y.dp[t]);
-      if (tmp > (mp_word) MP_MASK) {
-        tmp = MP_MASK;
+   mp_int  q, x, y, t1, t2;
+   int     res, n, t, i, norm, neg;
+
+   /* is divisor zero ? */
+   if (mp_iszero(b) == MP_YES) {
+      return MP_VAL;
+   }
+
+   /* if a < b then q=0, r = a */
+   if (mp_cmp_mag(a, b) == MP_LT) {
+      if (d != NULL) {
+         res = mp_copy(a, d);
+      } else {
+         res = MP_OKAY;
       }
-      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;
-    */
-    q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] + 1) & MP_MASK;
-    do {
-      q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1) & MP_MASK;
-
-      /* find left hand */
-      mp_zero (&t1);
-      t1.dp[0] = ((t - 1) < 0) ? 0 : y.dp[t - 1];
-      t1.dp[1] = y.dp[t];
-      t1.used = 2;
-      if ((res = mp_mul_d (&t1, q.dp[(i - t) - 1], &t1)) != MP_OKAY) {
-        goto LBL_Y;
+      if (c != NULL) {
+         mp_zero(c);
       }
+      return res;
+   }
+
+   if ((res = mp_init_size(&q, a->used + 2)) != MP_OKAY) {
+      return res;
+   }
+   q.used = a->used + 2;
+
+   if ((res = mp_init(&t1)) != MP_OKAY) {
+      goto LBL_Q;
+   }
+
+   if ((res = mp_init(&t2)) != MP_OKAY) {
+      goto LBL_T1;
+   }
+
+   if ((res = mp_init_copy(&x, a)) != MP_OKAY) {
+      goto LBL_T2;
+   }
+
+   if ((res = mp_init_copy(&y, b)) != MP_OKAY) {
+      goto LBL_X;
+   }
+
+   /* fix the sign */
+   neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
+   x.sign = y.sign = MP_ZPOS;
+
+   /* normalize both x and y, ensure that y >= b/2, [b == 2**DIGIT_BIT] */
+   norm = mp_count_bits(&y) % DIGIT_BIT;
+   if (norm < (int)(DIGIT_BIT-1)) {
+      norm = (DIGIT_BIT-1) - norm;
+      if ((res = mp_mul_2d(&x, norm, &x)) != MP_OKAY) {
+         goto LBL_Y;
+      }
+      if ((res = mp_mul_2d(&y, norm, &y)) != MP_OKAY) {
+         goto LBL_Y;
+      }
+   } else {
+      norm = 0;
+   }
 
-      /* find right hand */
-      t2.dp[0] = ((i - 2) < 0) ? 0 : x.dp[i - 2];
-      t2.dp[1] = ((i - 1) < 0) ? 0 : x.dp[i - 1];
-      t2.dp[2] = x.dp[i];
-      t2.used = 3;
-    } while (mp_cmp_mag(&t1, &t2) == MP_GT);
+   /* note hac does 0 based, so if used==5 then its 0,1,2,3,4, e.g. use 4 */
+   n = x.used - 1;
+   t = y.used - 1;
 
-    /* step 3.3 x = x - q{i-t-1} * y * b**{i-t-1} */
-    if ((res = mp_mul_d (&y, q.dp[(i - t) - 1], &t1)) != MP_OKAY) {
+   /* while (x >= y*b**n-t) do { q[n-t] += 1; x -= y*b**{n-t} } */
+   if ((res = mp_lshd(&y, n - t)) != MP_OKAY) { /* y = y*b**{n-t} */
       goto LBL_Y;
-    }
+   }
 
-    if ((res = mp_lshd (&t1, (i - t) - 1)) != MP_OKAY) {
-      goto LBL_Y;
-    }
+   while (mp_cmp(&x, &y) != MP_LT) {
+      ++(q.dp[n - t]);
+      if ((res = mp_sub(&x, &y, &x)) != MP_OKAY) {
+         goto LBL_Y;
+      }
+   }
 
-    if ((res = mp_sub (&x, &t1, &x)) != MP_OKAY) {
-      goto LBL_Y;
-    }
+   /* reset y by shifting it back down */
+   mp_rshd(&y, n - t);
 
-    /* if x < 0 then { x = x + y*b**{i-t-1}; q{i-t-1} -= 1; } */
-    if (x.sign == MP_NEG) {
-      if ((res = mp_copy (&y, &t1)) != MP_OKAY) {
-        goto LBL_Y;
+   /* step 3. for i from n down to (t + 1) */
+   for (i = n; i >= (t + 1); i--) {
+      if (i > x.used) {
+         continue;
       }
-      if ((res = mp_lshd (&t1, (i - t) - 1)) != MP_OKAY) {
-        goto LBL_Y;
+
+      /* 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);
+      } else {
+         mp_word tmp;
+         tmp = ((mp_word) x.dp[i]) << ((mp_word) DIGIT_BIT);
+         tmp |= ((mp_word) x.dp[i - 1]);
+         tmp /= ((mp_word) y.dp[t]);
+         if (tmp > (mp_word) MP_MASK) {
+            tmp = MP_MASK;
+         }
+         q.dp[(i - t) - 1] = (mp_digit)(tmp & (mp_word)(MP_MASK));
       }
-      if ((res = mp_add (&x, &t1, &x)) != MP_OKAY) {
-        goto LBL_Y;
+
+      /* 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 {
+         q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1) & MP_MASK;
+
+         /* find left hand */
+         mp_zero(&t1);
+         t1.dp[0] = ((t - 1) < 0) ? 0 : y.dp[t - 1];
+         t1.dp[1] = y.dp[t];
+         t1.used = 2;
+         if ((res = mp_mul_d(&t1, q.dp[(i - t) - 1], &t1)) != MP_OKAY) {
+            goto LBL_Y;
+         }
+
+         /* find right hand */
+         t2.dp[0] = ((i - 2) < 0) ? 0 : x.dp[i - 2];
+         t2.dp[1] = ((i - 1) < 0) ? 0 : x.dp[i - 1];
+         t2.dp[2] = x.dp[i];
+         t2.used = 3;
+      } while (mp_cmp_mag(&t1, &t2) == MP_GT);
+
+      /* step 3.3 x = x - q{i-t-1} * y * b**{i-t-1} */
+      if ((res = mp_mul_d(&y, q.dp[(i - t) - 1], &t1)) != MP_OKAY) {
+         goto LBL_Y;
       }
 
-      q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1UL) & MP_MASK;
-    }
-  }
+      if ((res = mp_lshd(&t1, (i - t) - 1)) != MP_OKAY) {
+         goto LBL_Y;
+      }
 
-  /* now q is the quotient and x is the remainder
-   * [which we have to normalize]
-   */
+      if ((res = mp_sub(&x, &t1, &x)) != MP_OKAY) {
+         goto LBL_Y;
+      }
 
-  /* get sign before writing to c */
-  x.sign = (x.used == 0) ? MP_ZPOS : a->sign;
+      /* if x < 0 then { x = x + y*b**{i-t-1}; q{i-t-1} -= 1; } */
+      if (x.sign == MP_NEG) {
+         if ((res = mp_copy(&y, &t1)) != MP_OKAY) {
+            goto LBL_Y;
+         }
+         if ((res = mp_lshd(&t1, (i - t) - 1)) != MP_OKAY) {
+            goto LBL_Y;
+         }
+         if ((res = mp_add(&x, &t1, &x)) != MP_OKAY) {
+            goto LBL_Y;
+         }
+
+         q.dp[(i - t) - 1] = (q.dp[(i - t) - 1] - 1UL) & MP_MASK;
+      }
+   }
 
-  if (c != NULL) {
-    mp_clamp (&q);
-    mp_exch (&q, c);
-    c->sign = neg;
-  }
+   /* now q is the quotient and x is the remainder
+    * [which we have to normalize]
+    */
 
-  if (d != NULL) {
-    if ((res = mp_div_2d (&x, norm, &x, NULL)) != MP_OKAY) {
-      goto LBL_Y;
-    }
-    mp_exch (&x, d);
-  }
-
-  res = MP_OKAY;
-
-LBL_Y:mp_clear (&y);
-LBL_X:mp_clear (&x);
-LBL_T2:mp_clear (&t2);
-LBL_T1:mp_clear (&t1);
-LBL_Q:mp_clear (&q);
-  return res;
+   /* get sign before writing to c */
+   x.sign = (x.used == 0) ? MP_ZPOS : a->sign;
+
+   if (c != NULL) {
+      mp_clamp(&q);
+      mp_exch(&q, c);
+      c->sign = neg;
+   }
+
+   if (d != NULL) {
+      if ((res = mp_div_2d(&x, norm, &x, NULL)) != MP_OKAY) {
+         goto LBL_Y;
+      }
+      mp_exch(&x, d);
+   }
+
+   res = MP_OKAY;
+
+LBL_Y:
+   mp_clear(&y);
+LBL_X:
+   mp_clear(&x);
+LBL_T2:
+   mp_clear(&t2);
+LBL_T1:
+   mp_clear(&t1);
+LBL_Q:
+   mp_clear(&q);
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_div_2.c b/libtommath/bn_mp_div_2.c
index 2b5bb49..b9d5339 100644
--- a/libtommath/bn_mp_div_2.c
+++ b/libtommath/bn_mp_div_2.c
@@ -16,50 +16,50 @@
  */
 
 /* b = a/2 */
-int mp_div_2(mp_int * a, mp_int * b)
+int mp_div_2(mp_int *a, mp_int *b)
 {
-  int     x, res, oldused;
+   int     x, res, oldused;
 
-  /* copy */
-  if (b->alloc < a->used) {
-    if ((res = mp_grow (b, a->used)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* copy */
+   if (b->alloc < a->used) {
+      if ((res = mp_grow(b, a->used)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  oldused = b->used;
-  b->used = a->used;
-  {
-    mp_digit r, rr, *tmpa, *tmpb;
+   oldused = b->used;
+   b->used = a->used;
+   {
+      mp_digit r, rr, *tmpa, *tmpb;
 
-    /* source alias */
-    tmpa = a->dp + b->used - 1;
+      /* source alias */
+      tmpa = a->dp + b->used - 1;
 
-    /* dest alias */
-    tmpb = b->dp + b->used - 1;
+      /* dest alias */
+      tmpb = b->dp + b->used - 1;
 
-    /* carry */
-    r = 0;
-    for (x = b->used - 1; x >= 0; x--) {
-      /* get the carry for the next iteration */
-      rr = *tmpa & 1;
+      /* carry */
+      r = 0;
+      for (x = b->used - 1; x >= 0; x--) {
+         /* get the carry for the next iteration */
+         rr = *tmpa & 1;
 
-      /* shift the current digit, add in carry and store */
-      *tmpb-- = (*tmpa-- >> 1) | (r << (DIGIT_BIT - 1));
+         /* shift the current digit, add in carry and store */
+         *tmpb-- = (*tmpa-- >> 1) | (r << (DIGIT_BIT - 1));
 
-      /* forward carry to next iteration */
-      r = rr;
-    }
+         /* forward carry to next iteration */
+         r = rr;
+      }
 
-    /* zero excess digits */
-    tmpb = b->dp + b->used;
-    for (x = b->used; x < oldused; x++) {
-      *tmpb++ = 0;
-    }
-  }
-  b->sign = a->sign;
-  mp_clamp (b);
-  return MP_OKAY;
+      /* zero excess digits */
+      tmpb = b->dp + b->used;
+      for (x = b->used; x < oldused; x++) {
+         *tmpb++ = 0;
+      }
+   }
+   b->sign = a->sign;
+   mp_clamp(b);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_div_2d.c b/libtommath/bn_mp_div_2d.c
index 635d374..d6723ee 100644
--- a/libtommath/bn_mp_div_2d.c
+++ b/libtommath/bn_mp_div_2d.c
@@ -16,68 +16,68 @@
  */
 
 /* shift right by a certain bit count (store quotient in c, optional remainder in d) */
-int mp_div_2d (mp_int * a, int b, mp_int * c, mp_int * d)
+int mp_div_2d(mp_int *a, int b, mp_int *c, mp_int *d)
 {
-  mp_digit D, r, rr;
-  int     x, res;
+   mp_digit D, r, rr;
+   int     x, res;
 
-  /* if the shift count is <= 0 then we do no work */
-  if (b <= 0) {
-    res = mp_copy (a, c);
-    if (d != NULL) {
-      mp_zero (d);
-    }
-    return res;
-  }
-
-  /* copy */
-  if ((res = mp_copy (a, c)) != MP_OKAY) {
-    return res;
-  }
-  /* 'a' should not be used after here - it might be the same as d */
+   /* if the shift count is <= 0 then we do no work */
+   if (b <= 0) {
+      res = mp_copy(a, c);
+      if (d != NULL) {
+         mp_zero(d);
+      }
+      return res;
+   }
 
-  /* get the remainder */
-  if (d != NULL) {
-    if ((res = mp_mod_2d (a, b, d)) != MP_OKAY) {
+   /* copy */
+   if ((res = mp_copy(a, c)) != MP_OKAY) {
       return res;
-    }
-  }
+   }
+   /* 'a' should not be used after here - it might be the same as d */
+
+   /* get the remainder */
+   if (d != NULL) {
+      if ((res = mp_mod_2d(a, b, d)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* shift by as many digits in the bit count */
-  if (b >= (int)DIGIT_BIT) {
-    mp_rshd (c, b / DIGIT_BIT);
-  }
+   /* shift by as many digits in the bit count */
+   if (b >= (int)DIGIT_BIT) {
+      mp_rshd(c, b / DIGIT_BIT);
+   }
 
-  /* shift any bit count < DIGIT_BIT */
-  D = (mp_digit) (b % DIGIT_BIT);
-  if (D != 0) {
-    mp_digit *tmpc, mask, shift;
+   /* shift any bit count < DIGIT_BIT */
+   D = (mp_digit)(b % DIGIT_BIT);
+   if (D != 0) {
+      mp_digit *tmpc, mask, shift;
 
-    /* mask */
-    mask = (((mp_digit)1) << D) - 1;
+      /* mask */
+      mask = (((mp_digit)1) << D) - 1;
 
-    /* shift for lsb */
-    shift = DIGIT_BIT - D;
+      /* shift for lsb */
+      shift = DIGIT_BIT - D;
 
-    /* alias */
-    tmpc = c->dp + (c->used - 1);
+      /* alias */
+      tmpc = c->dp + (c->used - 1);
 
-    /* carry */
-    r = 0;
-    for (x = c->used - 1; x >= 0; x--) {
-      /* get the lower  bits of this word in a temp */
-      rr = *tmpc & mask;
+      /* carry */
+      r = 0;
+      for (x = c->used - 1; x >= 0; x--) {
+         /* get the lower  bits of this word in a temp */
+         rr = *tmpc & mask;
 
-      /* shift the current word and mix in the carry bits from the previous word */
-      *tmpc = (*tmpc >> D) | (r << shift);
-      --tmpc;
+         /* shift the current word and mix in the carry bits from the previous word */
+         *tmpc = (*tmpc >> D) | (r << shift);
+         --tmpc;
 
-      /* set the carry to the carry bits of the current word found above */
-      r = rr;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+         /* set the carry to the carry bits of the current word found above */
+         r = rr;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_div_3.c b/libtommath/bn_mp_div_3.c
index e8504ea..c3a023a 100644
--- a/libtommath/bn_mp_div_3.c
+++ b/libtommath/bn_mp_div_3.c
@@ -16,60 +16,59 @@
  */
 
 /* divide by three (based on routine from MPI and the GMP manual) */
-int
-mp_div_3 (mp_int * a, mp_int *c, mp_digit * d)
+int mp_div_3(mp_int *a, mp_int *c, mp_digit *d)
 {
-  mp_int   q;
-  mp_word  w, t;
-  mp_digit b;
-  int      res, ix;
-  
-  /* b = 2**DIGIT_BIT / 3 */
-  b = (((mp_word)1) << ((mp_word)DIGIT_BIT)) / ((mp_word)3);
+   mp_int   q;
+   mp_word  w, t;
+   mp_digit b;
+   int      res, ix;
 
-  if ((res = mp_init_size(&q, a->used)) != MP_OKAY) {
-     return res;
-  }
-  
-  q.used = a->used;
-  q.sign = a->sign;
-  w = 0;
-  for (ix = a->used - 1; ix >= 0; ix--) {
-     w = (w << ((mp_word)DIGIT_BIT)) | ((mp_word)a->dp[ix]);
+   /* b = 2**DIGIT_BIT / 3 */
+   b = (((mp_word)1) << ((mp_word)DIGIT_BIT)) / ((mp_word)3);
 
-     if (w >= 3) {
-        /* multiply w by [1/3] */
-        t = (w * ((mp_word)b)) >> ((mp_word)DIGIT_BIT);
+   if ((res = mp_init_size(&q, a->used)) != MP_OKAY) {
+      return res;
+   }
 
-        /* now subtract 3 * [w/3] from w, to get the remainder */
-        w -= t+t+t;
+   q.used = a->used;
+   q.sign = a->sign;
+   w = 0;
+   for (ix = a->used - 1; ix >= 0; ix--) {
+      w = (w << ((mp_word)DIGIT_BIT)) | ((mp_word)a->dp[ix]);
 
-        /* fixup the remainder as required since
-         * the optimization is not exact.
-         */
-        while (w >= 3) {
-           t += 1;
-           w -= 3;
-        }
+      if (w >= 3) {
+         /* multiply w by [1/3] */
+         t = (w * ((mp_word)b)) >> ((mp_word)DIGIT_BIT);
+
+         /* now subtract 3 * [w/3] from w, to get the remainder */
+         w -= t+t+t;
+
+         /* fixup the remainder as required since
+          * the optimization is not exact.
+          */
+         while (w >= 3) {
+            t += 1;
+            w -= 3;
+         }
       } else {
-        t = 0;
+         t = 0;
       }
       q.dp[ix] = (mp_digit)t;
-  }
+   }
+
+   /* [optional] store the remainder */
+   if (d != NULL) {
+      *d = (mp_digit)w;
+   }
 
-  /* [optional] store the remainder */
-  if (d != NULL) {
-     *d = (mp_digit)w;
-  }
+   /* [optional] store the quotient */
+   if (c != NULL) {
+      mp_clamp(&q);
+      mp_exch(&q, c);
+   }
+   mp_clear(&q);
 
-  /* [optional] store the quotient */
-  if (c != NULL) {
-     mp_clamp(&q);
-     mp_exch(&q, c);
-  }
-  mp_clear(&q);
-  
-  return res;
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_div_d.c b/libtommath/bn_mp_div_d.c
index a5dbc59..141db1d 100644
--- a/libtommath/bn_mp_div_d.c
+++ b/libtommath/bn_mp_div_d.c
@@ -34,78 +34,78 @@ static int s_is_power_of_two(mp_digit b, int *p)
 }
 
 /* single digit division (based on routine from MPI) */
-int mp_div_d (mp_int * a, mp_digit b, mp_int * c, mp_digit * d)
+int mp_div_d(mp_int *a, mp_digit b, mp_int *c, mp_digit *d)
 {
-  mp_int  q;
-  mp_word w;
-  mp_digit t;
-  int     res, ix;
-
-  /* cannot divide by zero */
-  if (b == 0) {
-     return MP_VAL;
-  }
-
-  /* quick outs */
-  if ((b == 1) || (mp_iszero(a) == MP_YES)) {
-     if (d != NULL) {
-        *d = 0;
-     }
-     if (c != NULL) {
-        return mp_copy(a, c);
-     }
-     return MP_OKAY;
-  }
-
-  /* power of two ? */
-  if (s_is_power_of_two(b, &ix) == 1) {
-     if (d != NULL) {
-        *d = a->dp[0] & ((((mp_digit)1)<<ix) - 1);
-     }
-     if (c != NULL) {
-        return mp_div_2d(a, ix, c, NULL);
-     }
-     return MP_OKAY;
-  }
+   mp_int  q;
+   mp_word w;
+   mp_digit t;
+   int     res, ix;
+
+   /* cannot divide by zero */
+   if (b == 0) {
+      return MP_VAL;
+   }
+
+   /* quick outs */
+   if ((b == 1) || (mp_iszero(a) == MP_YES)) {
+      if (d != NULL) {
+         *d = 0;
+      }
+      if (c != NULL) {
+         return mp_copy(a, c);
+      }
+      return MP_OKAY;
+   }
+
+   /* power of two ? */
+   if (s_is_power_of_two(b, &ix) == 1) {
+      if (d != NULL) {
+         *d = a->dp[0] & ((((mp_digit)1)<<ix) - 1);
+      }
+      if (c != NULL) {
+         return mp_div_2d(a, ix, c, NULL);
+      }
+      return MP_OKAY;
+   }
 
 #ifdef BN_MP_DIV_3_C
-  /* three? */
-  if (b == 3) {
-     return mp_div_3(a, c, d);
-  }
+   /* three? */
+   if (b == 3) {
+      return mp_div_3(a, c, d);
+   }
 #endif
 
-  /* no easy answer [c'est la vie].  Just division */
-  if ((res = mp_init_size(&q, a->used)) != MP_OKAY) {
-     return res;
-  }
-  
-  q.used = a->used;
-  q.sign = a->sign;
-  w = 0;
-  for (ix = a->used - 1; ix >= 0; ix--) {
-     w = (w << ((mp_word)DIGIT_BIT)) | ((mp_word)a->dp[ix]);
-     
-     if (w >= b) {
-        t = (mp_digit)(w / b);
-        w -= ((mp_word)t) * ((mp_word)b);
+   /* no easy answer [c'est la vie].  Just division */
+   if ((res = mp_init_size(&q, a->used)) != MP_OKAY) {
+      return res;
+   }
+
+   q.used = a->used;
+   q.sign = a->sign;
+   w = 0;
+   for (ix = a->used - 1; ix >= 0; ix--) {
+      w = (w << ((mp_word)DIGIT_BIT)) | ((mp_word)a->dp[ix]);
+
+      if (w >= b) {
+         t = (mp_digit)(w / b);
+         w -= ((mp_word)t) * ((mp_word)b);
       } else {
-        t = 0;
+         t = 0;
       }
       q.dp[ix] = (mp_digit)t;
-  }
-  
-  if (d != NULL) {
-     *d = (mp_digit)w;
-  }
-  
-  if (c != NULL) {
-     mp_clamp(&q);
-     mp_exch(&q, c);
-  }
-  mp_clear(&q);
-  
-  return res;
+   }
+
+   if (d != NULL) {
+      *d = (mp_digit)w;
+   }
+
+   if (c != NULL) {
+      mp_clamp(&q);
+      mp_exch(&q, c);
+   }
+   mp_clear(&q);
+
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_dr_is_modulus.c b/libtommath/bn_mp_dr_is_modulus.c
index ced330c..4631daa 100644
--- a/libtommath/bn_mp_dr_is_modulus.c
+++ b/libtommath/bn_mp_dr_is_modulus.c
@@ -29,9 +29,9 @@ int mp_dr_is_modulus(mp_int *a)
     * but the first digit must be equal to -1 (mod b).
     */
    for (ix = 1; ix < a->used; ix++) {
-       if (a->dp[ix] != MP_MASK) {
-          return 0;
-       }
+      if (a->dp[ix] != MP_MASK) {
+         return 0;
+      }
    }
    return 1;
 }
diff --git a/libtommath/bn_mp_dr_reduce.c b/libtommath/bn_mp_dr_reduce.c
index c85ee77..25079be 100644
--- a/libtommath/bn_mp_dr_reduce.c
+++ b/libtommath/bn_mp_dr_reduce.c
@@ -29,65 +29,64 @@
  *
  * Input x must be in the range 0 <= x <= (n-1)**2
  */
-int
-mp_dr_reduce (mp_int * x, mp_int * n, mp_digit k)
+int mp_dr_reduce(mp_int *x, mp_int *n, mp_digit k)
 {
-  int      err, i, m;
-  mp_word  r;
-  mp_digit mu, *tmpx1, *tmpx2;
+   int      err, i, m;
+   mp_word  r;
+   mp_digit mu, *tmpx1, *tmpx2;
 
-  /* m = digits in modulus */
-  m = n->used;
+   /* m = digits in modulus */
+   m = n->used;
 
-  /* ensure that "x" has at least 2m digits */
-  if (x->alloc < (m + m)) {
-    if ((err = mp_grow (x, m + m)) != MP_OKAY) {
-      return err;
-    }
-  }
+   /* ensure that "x" has at least 2m digits */
+   if (x->alloc < (m + m)) {
+      if ((err = mp_grow(x, m + m)) != MP_OKAY) {
+         return err;
+      }
+   }
 
-/* top of loop, this is where the code resumes if
- * another reduction pass is required.
- */
+   /* top of loop, this is where the code resumes if
+    * another reduction pass is required.
+    */
 top:
-  /* aliases for digits */
-  /* alias for lower half of x */
-  tmpx1 = x->dp;
+   /* aliases for digits */
+   /* alias for lower half of x */
+   tmpx1 = x->dp;
 
-  /* alias for upper half of x, or x/B**m */
-  tmpx2 = x->dp + m;
+   /* alias for upper half of x, or x/B**m */
+   tmpx2 = x->dp + m;
 
-  /* set carry to zero */
-  mu = 0;
+   /* set carry to zero */
+   mu = 0;
 
-  /* compute (x mod B**m) + k * [x/B**m] inline and inplace */
-  for (i = 0; i < m; i++) {
+   /* compute (x mod B**m) + k * [x/B**m] inline and inplace */
+   for (i = 0; i < m; i++) {
       r         = (((mp_word)*tmpx2++) * (mp_word)k) + *tmpx1 + mu;
       *tmpx1++  = (mp_digit)(r & MP_MASK);
       mu        = (mp_digit)(r >> ((mp_word)DIGIT_BIT));
-  }
+   }
 
-  /* set final carry */
-  *tmpx1++ = mu;
+   /* set final carry */
+   *tmpx1++ = mu;
 
-  /* zero words above m */
-  for (i = m + 1; i < x->used; i++) {
+   /* zero words above m */
+   for (i = m + 1; i < x->used; i++) {
       *tmpx1++ = 0;
-  }
+   }
 
-  /* clamp, sub and return */
-  mp_clamp (x);
+   /* clamp, sub and return */
+   mp_clamp(x);
 
-  /* if x >= n then subtract and reduce again
-   * Each successive "recursion" makes the input smaller and smaller.
-   */
-  if (mp_cmp_mag (x, n) != MP_LT) {
-    if ((err = s_mp_sub(x, n, x)) != MP_OKAY) {
-      return err;
-    }
-    goto top;
-  }
-  return MP_OKAY;
+   /* if x >= n then subtract and reduce again
+    * Each successive "recursion" makes the input smaller and smaller.
+    */
+   if (mp_cmp_mag(x, n) != MP_LT) {
+      if ((err = s_mp_sub(x, n, x)) != MP_OKAY) {
+         return err;
+      }
+      goto top;
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_dr_setup.c b/libtommath/bn_mp_dr_setup.c
index b0d4a14..97f31ba 100644
--- a/libtommath/bn_mp_dr_setup.c
+++ b/libtommath/bn_mp_dr_setup.c
@@ -21,8 +21,7 @@ void mp_dr_setup(mp_int *a, mp_digit *d)
    /* the casts are required if DIGIT_BIT is one less than
     * the number of bits in a mp_digit [e.g. DIGIT_BIT==31]
     */
-   *d = (mp_digit)((((mp_word)1) << ((mp_word)DIGIT_BIT)) - 
-        ((mp_word)a->dp[0]));
+   *d = (mp_digit)((((mp_word)1) << ((mp_word)DIGIT_BIT)) - ((mp_word)a->dp[0]));
 }
 
 #endif
diff --git a/libtommath/bn_mp_exch.c b/libtommath/bn_mp_exch.c
index fc26bae..2bc635f 100644
--- a/libtommath/bn_mp_exch.c
+++ b/libtommath/bn_mp_exch.c
@@ -15,17 +15,16 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* 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
  */
-void
-mp_exch (mp_int * a, mp_int * b)
+void mp_exch(mp_int *a, mp_int *b)
 {
-  mp_int  t;
+   mp_int  t;
 
-  t  = *a;
-  *a = *b;
-  *b = t;
+   t  = *a;
+   *a = *b;
+   *b = t;
 }
 #endif
 
diff --git a/libtommath/bn_mp_export.c b/libtommath/bn_mp_export.c
index 4bbc8c5..b69a4fb 100644
--- a/libtommath/bn_mp_export.c
+++ b/libtommath/bn_mp_export.c
@@ -18,67 +18,66 @@
 /* based on gmp's mpz_export.
  * see http://gmplib.org/manual/Integer-Import-and-Export.html
  */
-int mp_export(void* rop, size_t* countp, int order, size_t size, 
-                                int endian, size_t nails, mp_int* op) {
-	int result;
-	size_t odd_nails, nail_bytes, i, j, bits, count;
-	unsigned char odd_nail_mask;
-
-	mp_int t;
-
-	if ((result = mp_init_copy(&t, op)) != MP_OKAY) {
-		return result;
-	}
-
-	if (endian == 0) {
-		union {
-			unsigned int i;
-			char c[4];
-		} lint;
-		lint.i = 0x01020304;
-
-		endian = (lint.c[0] == 4) ? -1 : 1;
-	}
-
-	odd_nails = (nails % 8);
-	odd_nail_mask = 0xff;
-	for (i = 0; i < odd_nails; ++i) {
-		odd_nail_mask ^= (1 << (7 - i));
-	}
-	nail_bytes = nails / 8;
-
-	bits = mp_count_bits(&t);
-	count = (bits / ((size * 8) - nails)) + (((bits % ((size * 8) - nails)) != 0) ? 1 : 0);
-
-	for (i = 0; i < count; ++i) {
-		for (j = 0; j < size; ++j) {
-			unsigned char* byte = (
-				(unsigned char*)rop + 
-				(((order == -1) ? i : ((count - 1) - i)) * size) +
-				((endian == -1) ? j : ((size - 1) - j))
-			);
-
-			if (j >= (size - nail_bytes)) {
-				*byte = 0;
-				continue;
-			}
-
-			*byte = (unsigned char)((j == ((size - nail_bytes) - 1)) ? (t.dp[0] & odd_nail_mask) : (t.dp[0] & 0xFF));
-
-			if ((result = mp_div_2d(&t, ((j == ((size - nail_bytes) - 1)) ? (8 - odd_nails) : 8), &t, NULL)) != MP_OKAY) {
-				mp_clear(&t);
-				return result;
-			}
-		}
-	}
-
-	mp_clear(&t);
-
-	if (countp != NULL) {
-		*countp = count;
-	}
-
-	return MP_OKAY;
+int mp_export(void *rop, size_t *countp, int order, size_t size,
+              int endian, size_t nails, mp_int *op)
+{
+   int result;
+   size_t odd_nails, nail_bytes, i, j, bits, count;
+   unsigned char odd_nail_mask;
+
+   mp_int t;
+
+   if ((result = mp_init_copy(&t, op)) != MP_OKAY) {
+      return result;
+   }
+
+   if (endian == 0) {
+      union {
+         unsigned int i;
+         char c[4];
+      } lint;
+      lint.i = 0x01020304;
+
+      endian = (lint.c[0] == 4) ? -1 : 1;
+   }
+
+   odd_nails = (nails % 8);
+   odd_nail_mask = 0xff;
+   for (i = 0; i < odd_nails; ++i) {
+      odd_nail_mask ^= (1 << (7 - i));
+   }
+   nail_bytes = nails / 8;
+
+   bits = mp_count_bits(&t);
+   count = (bits / ((size * 8) - nails)) + (((bits % ((size * 8) - nails)) != 0) ? 1 : 0);
+
+   for (i = 0; i < count; ++i) {
+      for (j = 0; j < size; ++j) {
+         unsigned char *byte = (unsigned char *)rop +
+                               (((order == -1) ? i : ((count - 1) - i)) * size) +
+                               ((endian == -1) ? j : ((size - 1) - j));
+
+         if (j >= (size - nail_bytes)) {
+            *byte = 0;
+            continue;
+         }
+
+         *byte = (unsigned char)((j == ((size - nail_bytes) - 1)) ? (t.dp[0] & odd_nail_mask) : (t.dp[0] & 0xFF));
+
+         if ((result = mp_div_2d(&t, ((j == ((size - nail_bytes) - 1)) ? (8 - odd_nails) : 8), &t, NULL)) != MP_OKAY) {
+            mp_clear(&t);
+            return result;
+         }
+      }
+   }
+
+   mp_clear(&t);
+
+   if (countp != NULL) {
+      *countp = count;
+   }
+
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_expt_d.c b/libtommath/bn_mp_expt_d.c
index a311926..38bf679 100644
--- a/libtommath/bn_mp_expt_d.c
+++ b/libtommath/bn_mp_expt_d.c
@@ -16,9 +16,9 @@
  */
 
 /* wrapper function for mp_expt_d_ex() */
-int mp_expt_d (mp_int * a, mp_digit b, mp_int * c)
+int mp_expt_d(mp_int *a, mp_digit b, mp_int *c)
 {
-  return mp_expt_d_ex(a, b, c, 0);
+   return mp_expt_d_ex(a, b, c, 0);
 }
 
 #endif
diff --git a/libtommath/bn_mp_expt_d_ex.c b/libtommath/bn_mp_expt_d_ex.c
index c361b27..bece77b 100644
--- a/libtommath/bn_mp_expt_d_ex.c
+++ b/libtommath/bn_mp_expt_d_ex.c
@@ -16,65 +16,64 @@
  */
 
 /* calculate c = a**b  using a square-multiply algorithm */
-int mp_expt_d_ex (mp_int * a, mp_digit b, mp_int * c, int fast)
+int mp_expt_d_ex(mp_int *a, mp_digit b, mp_int *c, int fast)
 {
-  int     res;
-  unsigned int x;
+   int     res;
+   unsigned int x;
 
-  mp_int  g;
+   mp_int  g;
 
-  if ((res = mp_init_copy (&g, a)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_copy(&g, a)) != MP_OKAY) {
+      return res;
+   }
 
-  /* set initial result */
-  mp_set (c, 1);
+   /* set initial result */
+   mp_set(c, 1);
 
-  if (fast != 0) {
-    while (b > 0) {
-      /* if the bit is set multiply */
-      if ((b & 1) != 0) {
-        if ((res = mp_mul (c, &g, c)) != MP_OKAY) {
-          mp_clear (&g);
-          return res;
-        }
-      }
+   if (fast != 0) {
+      while (b > 0) {
+         /* if the bit is set multiply */
+         if ((b & 1) != 0) {
+            if ((res = mp_mul(c, &g, c)) != MP_OKAY) {
+               mp_clear(&g);
+               return res;
+            }
+         }
 
-      /* square */
-      if (b > 1) {
-        if ((res = mp_sqr (&g, &g)) != MP_OKAY) {
-          mp_clear (&g);
-          return res;
-        }
-      }
+         /* square */
+         if (b > 1) {
+            if ((res = mp_sqr(&g, &g)) != MP_OKAY) {
+               mp_clear(&g);
+               return res;
+            }
+         }
 
-      /* shift to next bit */
-      b >>= 1;
-    }
-  }
-  else {
-    for (x = 0; x < DIGIT_BIT; x++) {
-      /* square */
-      if ((res = mp_sqr (c, c)) != MP_OKAY) {
-        mp_clear (&g);
-        return res;
+         /* shift to next bit */
+         b >>= 1;
       }
+   } else {
+      for (x = 0; x < DIGIT_BIT; x++) {
+         /* square */
+         if ((res = mp_sqr(c, c)) != MP_OKAY) {
+            mp_clear(&g);
+            return res;
+         }
 
-      /* if the bit is set multiply */
-      if ((b & (mp_digit) (((mp_digit)1) << (DIGIT_BIT - 1))) != 0) {
-        if ((res = mp_mul (c, &g, c)) != MP_OKAY) {
-           mp_clear (&g);
-           return res;
-        }
-      }
+         /* if the bit is set multiply */
+         if ((b & (mp_digit)(((mp_digit)1) << (DIGIT_BIT - 1))) != 0) {
+            if ((res = mp_mul(c, &g, c)) != MP_OKAY) {
+               mp_clear(&g);
+               return res;
+            }
+         }
 
-      /* shift to next bit */
-      b <<= 1;
-    }
-  } /* if ... else */
+         /* shift to next bit */
+         b <<= 1;
+      }
+   } /* if ... else */
 
-  mp_clear (&g);
-  return MP_OKAY;
+   mp_clear(&g);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_exptmod.c b/libtommath/bn_mp_exptmod.c
index 25c389d..c4f392b 100644
--- a/libtommath/bn_mp_exptmod.c
+++ b/libtommath/bn_mp_exptmod.c
@@ -21,87 +21,87 @@
  * embedded in the normal function but that wasted alot of stack space
  * for nothing (since 99% of the time the Montgomery code would be called)
  */
-int mp_exptmod (mp_int * G, mp_int * X, mp_int * P, mp_int * Y)
+int mp_exptmod(mp_int *G, mp_int *X, mp_int *P, mp_int *Y)
 {
-  int dr;
+   int dr;
 
-  /* modulus P must be positive */
-  if (P->sign == MP_NEG) {
-     return MP_VAL;
-  }
+   /* modulus P must be positive */
+   if (P->sign == MP_NEG) {
+      return MP_VAL;
+   }
 
-  /* if exponent X is negative we have to recurse */
-  if (X->sign == MP_NEG) {
+   /* if exponent X is negative we have to recurse */
+   if (X->sign == MP_NEG) {
 #ifdef BN_MP_INVMOD_C
-     mp_int tmpG, tmpX;
-     int err;
+      mp_int tmpG, tmpX;
+      int err;
 
-     /* first compute 1/G mod P */
-     if ((err = mp_init(&tmpG)) != MP_OKAY) {
-        return err;
-     }
-     if ((err = mp_invmod(G, P, &tmpG)) != MP_OKAY) {
-        mp_clear(&tmpG);
-        return err;
-     }
+      /* first compute 1/G mod P */
+      if ((err = mp_init(&tmpG)) != MP_OKAY) {
+         return err;
+      }
+      if ((err = mp_invmod(G, P, &tmpG)) != MP_OKAY) {
+         mp_clear(&tmpG);
+         return err;
+      }
 
-     /* now get |X| */
-     if ((err = mp_init(&tmpX)) != MP_OKAY) {
-        mp_clear(&tmpG);
-        return err;
-     }
-     if ((err = mp_abs(X, &tmpX)) != MP_OKAY) {
-        mp_clear_multi(&tmpG, &tmpX, NULL);
-        return err;
-     }
+      /* now get |X| */
+      if ((err = mp_init(&tmpX)) != MP_OKAY) {
+         mp_clear(&tmpG);
+         return err;
+      }
+      if ((err = mp_abs(X, &tmpX)) != MP_OKAY) {
+         mp_clear_multi(&tmpG, &tmpX, NULL);
+         return err;
+      }
 
-     /* and now compute (1/G)**|X| instead of G**X [X < 0] */
-     err = mp_exptmod(&tmpG, &tmpX, P, Y);
-     mp_clear_multi(&tmpG, &tmpX, NULL);
-     return err;
-#else 
-     /* no invmod */
-     return MP_VAL;
+      /* and now compute (1/G)**|X| instead of G**X [X < 0] */
+      err = mp_exptmod(&tmpG, &tmpX, P, Y);
+      mp_clear_multi(&tmpG, &tmpX, NULL);
+      return err;
+#else
+      /* no invmod */
+      return MP_VAL;
 #endif
-  }
+   }
 
-/* modified diminished radix reduction */
+   /* modified diminished radix reduction */
 #if defined(BN_MP_REDUCE_IS_2K_L_C) && defined(BN_MP_REDUCE_2K_L_C) && defined(BN_S_MP_EXPTMOD_C)
-  if (mp_reduce_is_2k_l(P) == MP_YES) {
-     return s_mp_exptmod(G, X, P, Y, 1);
-  }
+   if (mp_reduce_is_2k_l(P) == MP_YES) {
+      return s_mp_exptmod(G, X, P, Y, 1);
+   }
 #endif
 
 #ifdef BN_MP_DR_IS_MODULUS_C
-  /* is it a DR modulus? */
-  dr = mp_dr_is_modulus(P);
+   /* is it a DR modulus? */
+   dr = mp_dr_is_modulus(P);
 #else
-  /* default to no */
-  dr = 0;
+   /* default to no */
+   dr = 0;
 #endif
 
 #ifdef BN_MP_REDUCE_IS_2K_C
-  /* if not, is it a unrestricted DR modulus? */
-  if (dr == 0) {
-     dr = mp_reduce_is_2k(P) << 1;
-  }
+   /* if not, is it a unrestricted DR modulus? */
+   if (dr == 0) {
+      dr = mp_reduce_is_2k(P) << 1;
+   }
 #endif
-    
-  /* if the modulus is odd or dr != 0 use the montgomery method */
+
+   /* if the modulus is odd or dr != 0 use the montgomery method */
 #ifdef BN_MP_EXPTMOD_FAST_C
-  if ((mp_isodd (P) == MP_YES) || (dr !=  0)) {
-    return mp_exptmod_fast (G, X, P, Y, dr);
-  } else {
+   if ((mp_isodd(P) == MP_YES) || (dr !=  0)) {
+      return mp_exptmod_fast(G, X, P, Y, dr);
+   } else {
 #endif
 #ifdef BN_S_MP_EXPTMOD_C
-    /* otherwise use the generic Barrett reduction technique */
-    return s_mp_exptmod (G, X, P, Y, 0);
+      /* otherwise use the generic Barrett reduction technique */
+      return s_mp_exptmod(G, X, P, Y, 0);
 #else
-    /* no exptmod for evens */
-    return MP_VAL;
+      /* no exptmod for evens */
+      return MP_VAL;
 #endif
 #ifdef BN_MP_EXPTMOD_FAST_C
-  }
+   }
 #endif
 }
 
diff --git a/libtommath/bn_mp_exptmod_fast.c b/libtommath/bn_mp_exptmod_fast.c
index 5e5c7f2..38e0265 100644
--- a/libtommath/bn_mp_exptmod_fast.c
+++ b/libtommath/bn_mp_exptmod_fast.c
@@ -24,294 +24,295 @@
  */
 
 #ifdef MP_LOW_MEM
-   #define TAB_SIZE 32
+#   define TAB_SIZE 32
 #else
-   #define TAB_SIZE 256
+#   define TAB_SIZE 256
 #endif
 
-int mp_exptmod_fast (mp_int * G, mp_int * X, mp_int * P, mp_int * Y, int redmode)
+int mp_exptmod_fast(mp_int *G, mp_int *X, mp_int *P, mp_int *Y, int redmode)
 {
-  mp_int  M[TAB_SIZE], res;
-  mp_digit buf, mp;
-  int     err, bitbuf, bitcpy, bitcnt, mode, digidx, x, y, winsize;
-
-  /* use a pointer to the reduction algorithm.  This allows us to use
-   * one of many reduction algorithms without modding the guts of
-   * the code with if statements everywhere.
-   */
-  int     (*redux)(mp_int*,mp_int*,mp_digit);
-
-  /* find window size */
-  x = mp_count_bits (X);
-  if (x <= 7) {
-    winsize = 2;
-  } else if (x <= 36) {
-    winsize = 3;
-  } else if (x <= 140) {
-    winsize = 4;
-  } else if (x <= 450) {
-    winsize = 5;
-  } else if (x <= 1303) {
-    winsize = 6;
-  } else if (x <= 3529) {
-    winsize = 7;
-  } else {
-    winsize = 8;
-  }
+   mp_int  M[TAB_SIZE], res;
+   mp_digit buf, mp;
+   int     err, bitbuf, bitcpy, bitcnt, mode, digidx, x, y, winsize;
+
+   /* use a pointer to the reduction algorithm.  This allows us to use
+    * one of many reduction algorithms without modding the guts of
+    * the code with if statements everywhere.
+    */
+   int (*redux)(mp_int *,mp_int *,mp_digit);
+
+   /* find window size */
+   x = mp_count_bits(X);
+   if (x <= 7) {
+      winsize = 2;
+   } else if (x <= 36) {
+      winsize = 3;
+   } else if (x <= 140) {
+      winsize = 4;
+   } else if (x <= 450) {
+      winsize = 5;
+   } else if (x <= 1303) {
+      winsize = 6;
+   } else if (x <= 3529) {
+      winsize = 7;
+   } else {
+      winsize = 8;
+   }
 
 #ifdef MP_LOW_MEM
-  if (winsize > 5) {
-     winsize = 5;
-  }
+   if (winsize > 5) {
+      winsize = 5;
+   }
 #endif
 
-  /* init M array */
-  /* init first cell */
-  if ((err = mp_init_size(&M[1], P->alloc)) != MP_OKAY) {
-     return err;
-  }
-
-  /* now init the second half of the array */
-  for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
-    if ((err = mp_init_size(&M[x], P->alloc)) != MP_OKAY) {
-      for (y = 1<<(winsize-1); y < x; y++) {
-        mp_clear (&M[y]);
-      }
-      mp_clear(&M[1]);
+   /* init M array */
+   /* init first cell */
+   if ((err = mp_init_size(&M[1], P->alloc)) != MP_OKAY) {
       return err;
-    }
-  }
-
-  /* determine and setup reduction code */
-  if (redmode == 0) {
-#ifdef BN_MP_MONTGOMERY_SETUP_C     
-     /* now setup montgomery  */
-     if ((err = mp_montgomery_setup (P, &mp)) != MP_OKAY) {
-        goto LBL_M;
-     }
+   }
+
+   /* now init the second half of the array */
+   for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
+      if ((err = mp_init_size(&M[x], P->alloc)) != MP_OKAY) {
+         for (y = 1<<(winsize-1); y < x; y++) {
+            mp_clear(&M[y]);
+         }
+         mp_clear(&M[1]);
+         return err;
+      }
+   }
+
+   /* determine and setup reduction code */
+   if (redmode == 0) {
+#ifdef BN_MP_MONTGOMERY_SETUP_C
+      /* now setup montgomery  */
+      if ((err = mp_montgomery_setup(P, &mp)) != MP_OKAY) {
+         goto LBL_M;
+      }
 #else
-     err = MP_VAL;
-     goto LBL_M;
+      err = MP_VAL;
+      goto LBL_M;
 #endif
 
-     /* automatically pick the comba one if available (saves quite a few calls/ifs) */
+      /* automatically pick the comba one if available (saves quite a few calls/ifs) */
 #ifdef BN_FAST_MP_MONTGOMERY_REDUCE_C
-     if ((((P->used * 2) + 1) < MP_WARRAY) &&
+      if ((((P->used * 2) + 1) < MP_WARRAY) &&
           (P->used < (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
-        redux = fast_mp_montgomery_reduce;
-     } else 
+         redux = fast_mp_montgomery_reduce;
+      } else
 #endif
-     {
+      {
 #ifdef BN_MP_MONTGOMERY_REDUCE_C
-        /* use slower baseline Montgomery method */
-        redux = mp_montgomery_reduce;
+         /* use slower baseline Montgomery method */
+         redux = mp_montgomery_reduce;
 #else
-        err = MP_VAL;
-        goto LBL_M;
+         err = MP_VAL;
+         goto LBL_M;
 #endif
-     }
-  } else if (redmode == 1) {
+      }
+   } else if (redmode == 1) {
 #if defined(BN_MP_DR_SETUP_C) && defined(BN_MP_DR_REDUCE_C)
-     /* setup DR reduction for moduli of the form B**k - b */
-     mp_dr_setup(P, &mp);
-     redux = mp_dr_reduce;
+      /* setup DR reduction for moduli of the form B**k - b */
+      mp_dr_setup(P, &mp);
+      redux = mp_dr_reduce;
 #else
-     err = MP_VAL;
-     goto LBL_M;
+      err = MP_VAL;
+      goto LBL_M;
 #endif
-  } else {
+   } else {
 #if defined(BN_MP_REDUCE_2K_SETUP_C) && defined(BN_MP_REDUCE_2K_C)
-     /* setup DR reduction for moduli of the form 2**k - b */
-     if ((err = mp_reduce_2k_setup(P, &mp)) != MP_OKAY) {
-        goto LBL_M;
-     }
-     redux = mp_reduce_2k;
+      /* setup DR reduction for moduli of the form 2**k - b */
+      if ((err = mp_reduce_2k_setup(P, &mp)) != MP_OKAY) {
+         goto LBL_M;
+      }
+      redux = mp_reduce_2k;
 #else
-     err = MP_VAL;
-     goto LBL_M;
+      err = MP_VAL;
+      goto LBL_M;
 #endif
-  }
+   }
 
-  /* setup result */
-  if ((err = mp_init_size (&res, P->alloc)) != MP_OKAY) {
-    goto LBL_M;
-  }
+   /* setup result */
+   if ((err = mp_init_size(&res, P->alloc)) != MP_OKAY) {
+      goto LBL_M;
+   }
 
-  /* create M table
-   *
+   /* create M table
+    *
 
-   *
-   * The first half of the table is not computed though accept for M[0] and M[1]
-   */
+    *
+    * The first half of the table is not computed though accept for M[0] and M[1]
+    */
 
-  if (redmode == 0) {
+   if (redmode == 0) {
 #ifdef BN_MP_MONTGOMERY_CALC_NORMALIZATION_C
-     /* now we need R mod m */
-     if ((err = mp_montgomery_calc_normalization (&res, P)) != MP_OKAY) {
-       goto LBL_RES;
-     }
-
-     /* now set M[1] to G * R mod m */
-     if ((err = mp_mulmod (G, &res, P, &M[1])) != MP_OKAY) {
-       goto LBL_RES;
-     }
+      /* now we need R mod m */
+      if ((err = mp_montgomery_calc_normalization(&res, P)) != MP_OKAY) {
+         goto LBL_RES;
+      }
+
+      /* now set M[1] to G * R mod m */
+      if ((err = mp_mulmod(G, &res, P, &M[1])) != MP_OKAY) {
+         goto LBL_RES;
+      }
 #else
-     err = MP_VAL;
-     goto LBL_RES;
-#endif
-  } else {
-     mp_set(&res, 1);
-     if ((err = mp_mod(G, P, &M[1])) != MP_OKAY) {
-        goto LBL_RES;
-     }
-  }
-
-  /* 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_RES;
-  }
-
-  for (x = 0; x < (winsize - 1); x++) {
-    if ((err = mp_sqr (&M[1 << (winsize - 1)], &M[1 << (winsize - 1)])) != MP_OKAY) {
-      goto LBL_RES;
-    }
-    if ((err = redux (&M[1 << (winsize - 1)], P, mp)) != MP_OKAY) {
+      err = MP_VAL;
       goto LBL_RES;
-    }
-  }
+#endif
+   } else {
+      mp_set(&res, 1);
+      if ((err = mp_mod(G, P, &M[1])) != MP_OKAY) {
+         goto LBL_RES;
+      }
+   }
 
-  /* create upper table */
-  for (x = (1 << (winsize - 1)) + 1; x < (1 << winsize); x++) {
-    if ((err = mp_mul (&M[x - 1], &M[1], &M[x])) != MP_OKAY) {
-      goto LBL_RES;
-    }
-    if ((err = redux (&M[x], P, mp)) != MP_OKAY) {
+   /* 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_RES;
-    }
-  }
-
-  /* set initial mode and bit cnt */
-  mode   = 0;
-  bitcnt = 1;
-  buf    = 0;
-  digidx = X->used - 1;
-  bitcpy = 0;
-  bitbuf = 0;
-
-  for (;;) {
-    /* grab next digit as required */
-    if (--bitcnt == 0) {
-      /* if digidx == -1 we are out of digits so break */
-      if (digidx == -1) {
-        break;
+   }
+
+   for (x = 0; x < (winsize - 1); x++) {
+      if ((err = mp_sqr(&M[1 << (winsize - 1)], &M[1 << (winsize - 1)])) != MP_OKAY) {
+         goto LBL_RES;
       }
-      /* read next digit and reset bitcnt */
-      buf    = X->dp[digidx--];
-      bitcnt = (int)DIGIT_BIT;
-    }
-
-    /* grab the next msb from the exponent */
-    y     = (mp_digit)(buf >> (DIGIT_BIT - 1)) & 1;
-    buf <<= (mp_digit)1;
-
-    /* if the bit is zero and mode == 0 then we ignore it
-     * These represent the leading zero bits before the first 1 bit
-     * in the exponent.  Technically this opt is not required but it
-     * does lower the # of trivial squaring/reductions used
-     */
-    if ((mode == 0) && (y == 0)) {
-      continue;
-    }
-
-    /* if the bit is zero and mode == 1 then we square */
-    if ((mode == 1) && (y == 0)) {
-      if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-        goto LBL_RES;
+      if ((err = redux(&M[1 << (winsize - 1)], P, mp)) != MP_OKAY) {
+         goto LBL_RES;
       }
-      if ((err = redux (&res, P, mp)) != MP_OKAY) {
-        goto LBL_RES;
+   }
+
+   /* create upper table */
+   for (x = (1 << (winsize - 1)) + 1; x < (1 << winsize); x++) {
+      if ((err = mp_mul(&M[x - 1], &M[1], &M[x])) != MP_OKAY) {
+         goto LBL_RES;
       }
-      continue;
-    }
-
-    /* else we add it to the window */
-    bitbuf |= (y << (winsize - ++bitcpy));
-    mode    = 2;
-
-    if (bitcpy == winsize) {
-      /* ok window is filled so square as required and multiply  */
-      /* square first */
-      for (x = 0; x < winsize; x++) {
-        if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-          goto LBL_RES;
-        }
-        if ((err = redux (&res, P, mp)) != MP_OKAY) {
-          goto LBL_RES;
-        }
+      if ((err = redux(&M[x], P, mp)) != MP_OKAY) {
+         goto LBL_RES;
+      }
+   }
+
+   /* set initial mode and bit cnt */
+   mode   = 0;
+   bitcnt = 1;
+   buf    = 0;
+   digidx = X->used - 1;
+   bitcpy = 0;
+   bitbuf = 0;
+
+   for (;;) {
+      /* grab next digit as required */
+      if (--bitcnt == 0) {
+         /* if digidx == -1 we are out of digits so break */
+         if (digidx == -1) {
+            break;
+         }
+         /* read next digit and reset bitcnt */
+         buf    = X->dp[digidx--];
+         bitcnt = (int)DIGIT_BIT;
       }
 
-      /* then multiply */
-      if ((err = mp_mul (&res, &M[bitbuf], &res)) != MP_OKAY) {
-        goto LBL_RES;
+      /* grab the next msb from the exponent */
+      y     = (mp_digit)(buf >> (DIGIT_BIT - 1)) & 1;
+      buf <<= (mp_digit)1;
+
+      /* if the bit is zero and mode == 0 then we ignore it
+       * These represent the leading zero bits before the first 1 bit
+       * in the exponent.  Technically this opt is not required but it
+       * does lower the # of trivial squaring/reductions used
+       */
+      if ((mode == 0) && (y == 0)) {
+         continue;
       }
-      if ((err = redux (&res, P, mp)) != MP_OKAY) {
-        goto LBL_RES;
+
+      /* if the bit is zero and mode == 1 then we square */
+      if ((mode == 1) && (y == 0)) {
+         if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, mp)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         continue;
       }
 
-      /* empty window and reset */
-      bitcpy = 0;
-      bitbuf = 0;
-      mode   = 1;
-    }
-  }
-
-  /* if bits remain then square/multiply */
-  if ((mode == 2) && (bitcpy > 0)) {
-    /* square then multiply if the bit is set */
-    for (x = 0; x < bitcpy; x++) {
-      if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-        goto LBL_RES;
+      /* else we add it to the window */
+      bitbuf |= (y << (winsize - ++bitcpy));
+      mode    = 2;
+
+      if (bitcpy == winsize) {
+         /* ok window is filled so square as required and multiply  */
+         /* square first */
+         for (x = 0; x < winsize; x++) {
+            if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+            if ((err = redux(&res, P, mp)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+         }
+
+         /* then multiply */
+         if ((err = mp_mul(&res, &M[bitbuf], &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, mp)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+
+         /* empty window and reset */
+         bitcpy = 0;
+         bitbuf = 0;
+         mode   = 1;
       }
-      if ((err = redux (&res, P, mp)) != MP_OKAY) {
-        goto LBL_RES;
+   }
+
+   /* if bits remain then square/multiply */
+   if ((mode == 2) && (bitcpy > 0)) {
+      /* square then multiply if the bit is set */
+      for (x = 0; x < bitcpy; x++) {
+         if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, mp)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+
+         /* get next bit of the window */
+         bitbuf <<= 1;
+         if ((bitbuf & (1 << winsize)) != 0) {
+            /* then multiply */
+            if ((err = mp_mul(&res, &M[1], &res)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+            if ((err = redux(&res, P, mp)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+         }
       }
-
-      /* get next bit of the window */
-      bitbuf <<= 1;
-      if ((bitbuf & (1 << winsize)) != 0) {
-        /* then multiply */
-        if ((err = mp_mul (&res, &M[1], &res)) != MP_OKAY) {
-          goto LBL_RES;
-        }
-        if ((err = redux (&res, P, mp)) != MP_OKAY) {
-          goto LBL_RES;
-        }
+   }
+
+   if (redmode == 0) {
+      /* fixup result if Montgomery reduction is used
+       * recall that any value in a Montgomery system is
+       * actually multiplied by R mod n.  So we have
+       * to reduce one more time to cancel out the factor
+       * of R.
+       */
+      if ((err = redux(&res, P, mp)) != MP_OKAY) {
+         goto LBL_RES;
       }
-    }
-  }
-
-  if (redmode == 0) {
-     /* fixup result if Montgomery reduction is used
-      * recall that any value in a Montgomery system is
-      * actually multiplied by R mod n.  So we have
-      * to reduce one more time to cancel out the factor
-      * of R.
-      */
-     if ((err = redux(&res, P, mp)) != MP_OKAY) {
-       goto LBL_RES;
-     }
-  }
-
-  /* swap res with Y */
-  mp_exch (&res, Y);
-  err = MP_OKAY;
-LBL_RES:mp_clear (&res);
+   }
+
+   /* swap res with Y */
+   mp_exch(&res, Y);
+   err = MP_OKAY;
+LBL_RES:
+   mp_clear(&res);
 LBL_M:
-  mp_clear(&M[1]);
-  for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
-    mp_clear (&M[x]);
-  }
-  return err;
+   mp_clear(&M[1]);
+   for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
+      mp_clear(&M[x]);
+   }
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_exteuclid.c b/libtommath/bn_mp_exteuclid.c
index 3c9612e..98eef76 100644
--- a/libtommath/bn_mp_exteuclid.c
+++ b/libtommath/bn_mp_exteuclid.c
@@ -20,7 +20,7 @@
  */
 int mp_exteuclid(mp_int *a, mp_int *b, mp_int *U1, mp_int *U2, mp_int *U3)
 {
-   mp_int u1,u2,u3,v1,v2,v3,t1,t2,t3,q,tmp;
+   mp_int u1, u2, u3, v1, v2, v3, t1, t2, t3, q, tmp;
    int err;
 
    if ((err = mp_init_multi(&u1, &u2, &u3, &v1, &v2, &v3, &t1, &t2, &t3, &q, &tmp, NULL)) != MP_OKAY) {
@@ -29,47 +29,89 @@ int mp_exteuclid(mp_int *a, mp_int *b, mp_int *U1, mp_int *U2, mp_int *U3)
 
    /* initialize, (u1,u2,u3) = (1,0,a) */
    mp_set(&u1, 1);
-   if ((err = mp_copy(a, &u3)) != MP_OKAY)                                        { goto LBL_ERR; }
+   if ((err = mp_copy(a, &u3)) != MP_OKAY)                                        {
+      goto LBL_ERR;
+   }
 
    /* initialize, (v1,v2,v3) = (0,1,b) */
    mp_set(&v2, 1);
-   if ((err = mp_copy(b, &v3)) != MP_OKAY)                                        { goto LBL_ERR; }
+   if ((err = mp_copy(b, &v3)) != MP_OKAY)                                        {
+      goto LBL_ERR;
+   }
 
    /* loop while v3 != 0 */
    while (mp_iszero(&v3) == MP_NO) {
-       /* q = u3/v3 */
-       if ((err = mp_div(&u3, &v3, &q, NULL)) != MP_OKAY)                         { goto LBL_ERR; }
+      /* q = u3/v3 */
+      if ((err = mp_div(&u3, &v3, &q, NULL)) != MP_OKAY)                         {
+         goto LBL_ERR;
+      }
 
-       /* (t1,t2,t3) = (u1,u2,u3) - (v1,v2,v3)q */
-       if ((err = mp_mul(&v1, &q, &tmp)) != MP_OKAY)                              { goto LBL_ERR; }
-       if ((err = mp_sub(&u1, &tmp, &t1)) != MP_OKAY)                             { goto LBL_ERR; }
-       if ((err = mp_mul(&v2, &q, &tmp)) != MP_OKAY)                              { goto LBL_ERR; }
-       if ((err = mp_sub(&u2, &tmp, &t2)) != MP_OKAY)                             { goto LBL_ERR; }
-       if ((err = mp_mul(&v3, &q, &tmp)) != MP_OKAY)                              { goto LBL_ERR; }
-       if ((err = mp_sub(&u3, &tmp, &t3)) != MP_OKAY)                             { goto LBL_ERR; }
+      /* (t1,t2,t3) = (u1,u2,u3) - (v1,v2,v3)q */
+      if ((err = mp_mul(&v1, &q, &tmp)) != MP_OKAY)                              {
+         goto LBL_ERR;
+      }
+      if ((err = mp_sub(&u1, &tmp, &t1)) != MP_OKAY)                             {
+         goto LBL_ERR;
+      }
+      if ((err = mp_mul(&v2, &q, &tmp)) != MP_OKAY)                              {
+         goto LBL_ERR;
+      }
+      if ((err = mp_sub(&u2, &tmp, &t2)) != MP_OKAY)                             {
+         goto LBL_ERR;
+      }
+      if ((err = mp_mul(&v3, &q, &tmp)) != MP_OKAY)                              {
+         goto LBL_ERR;
+      }
+      if ((err = mp_sub(&u3, &tmp, &t3)) != MP_OKAY)                             {
+         goto LBL_ERR;
+      }
 
-       /* (u1,u2,u3) = (v1,v2,v3) */
-       if ((err = mp_copy(&v1, &u1)) != MP_OKAY)                                  { goto LBL_ERR; }
-       if ((err = mp_copy(&v2, &u2)) != MP_OKAY)                                  { goto LBL_ERR; }
-       if ((err = mp_copy(&v3, &u3)) != MP_OKAY)                                  { goto LBL_ERR; }
+      /* (u1,u2,u3) = (v1,v2,v3) */
+      if ((err = mp_copy(&v1, &u1)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
+      if ((err = mp_copy(&v2, &u2)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
+      if ((err = mp_copy(&v3, &u3)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
 
-       /* (v1,v2,v3) = (t1,t2,t3) */
-       if ((err = mp_copy(&t1, &v1)) != MP_OKAY)                                  { goto LBL_ERR; }
-       if ((err = mp_copy(&t2, &v2)) != MP_OKAY)                                  { goto LBL_ERR; }
-       if ((err = mp_copy(&t3, &v3)) != MP_OKAY)                                  { goto LBL_ERR; }
+      /* (v1,v2,v3) = (t1,t2,t3) */
+      if ((err = mp_copy(&t1, &v1)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
+      if ((err = mp_copy(&t2, &v2)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
+      if ((err = mp_copy(&t3, &v3)) != MP_OKAY)                                  {
+         goto LBL_ERR;
+      }
    }
 
    /* make sure U3 >= 0 */
    if (u3.sign == MP_NEG) {
-       if ((err = mp_neg(&u1, &u1)) != MP_OKAY)                                   { goto LBL_ERR; }
-       if ((err = mp_neg(&u2, &u2)) != MP_OKAY)                                   { goto LBL_ERR; }
-       if ((err = mp_neg(&u3, &u3)) != MP_OKAY)                                   { goto LBL_ERR; }
+      if ((err = mp_neg(&u1, &u1)) != MP_OKAY)                                   {
+         goto LBL_ERR;
+      }
+      if ((err = mp_neg(&u2, &u2)) != MP_OKAY)                                   {
+         goto LBL_ERR;
+      }
+      if ((err = mp_neg(&u3, &u3)) != MP_OKAY)                                   {
+         goto LBL_ERR;
+      }
    }
 
    /* copy result out */
-   if (U1 != NULL) { mp_exch(U1, &u1); }
-   if (U2 != NULL) { mp_exch(U2, &u2); }
-   if (U3 != NULL) { mp_exch(U3, &u3); }
+   if (U1 != NULL) {
+      mp_exch(U1, &u1);
+   }
+   if (U2 != NULL) {
+      mp_exch(U2, &u2);
+   }
+   if (U3 != NULL) {
+      mp_exch(U3, &u3);
+   }
 
    err = MP_OKAY;
 LBL_ERR:
diff --git a/libtommath/bn_mp_fread.c b/libtommath/bn_mp_fread.c
index 140721b..d0de595 100644
--- a/libtommath/bn_mp_fread.c
+++ b/libtommath/bn_mp_fread.c
@@ -20,10 +20,10 @@
 int mp_fread(mp_int *a, int radix, FILE *stream)
 {
    int err, ch, neg, y;
-   
+
    /* clear a */
    mp_zero(a);
-   
+
    /* if first digit is - then set negative */
    ch = fgetc(stream);
    if (ch == '-') {
@@ -32,18 +32,18 @@ int mp_fread(mp_int *a, int radix, FILE *stream)
    } else {
       neg = MP_ZPOS;
    }
-   
+
    for (;;) {
       /* find y in the radix map */
       for (y = 0; y < radix; y++) {
-          if (mp_s_rmap[y] == ch) {
-             break;
-          }
+         if (mp_s_rmap[y] == ch) {
+            break;
+         }
       }
       if (y == radix) {
          break;
       }
-      
+
       /* shift up and add */
       if ((err = mp_mul_d(a, radix, a)) != MP_OKAY) {
          return err;
@@ -51,13 +51,13 @@ int mp_fread(mp_int *a, int radix, FILE *stream)
       if ((err = mp_add_d(a, y, a)) != MP_OKAY) {
          return err;
       }
-      
+
       ch = fgetc(stream);
    }
    if (mp_cmp_d(a, 0) != MP_EQ) {
       a->sign = neg;
    }
-   
+
    return MP_OKAY;
 }
 #endif
diff --git a/libtommath/bn_mp_fwrite.c b/libtommath/bn_mp_fwrite.c
index 23b5f64..3641823 100644
--- a/libtommath/bn_mp_fwrite.c
+++ b/libtommath/bn_mp_fwrite.c
@@ -20,29 +20,29 @@ int mp_fwrite(mp_int *a, int radix, FILE *stream)
 {
    char *buf;
    int err, len, x;
-   
+
    if ((err = mp_radix_size(a, radix, &len)) != MP_OKAY) {
       return err;
    }
 
-   buf = OPT_CAST(char) XMALLOC (len);
+   buf = OPT_CAST(char) XMALLOC(len);
    if (buf == NULL) {
       return MP_MEM;
    }
-   
+
    if ((err = mp_toradix(a, buf, radix)) != MP_OKAY) {
-      XFREE (buf);
+      XFREE(buf);
       return err;
    }
-   
+
    for (x = 0; x < len; x++) {
-       if (fputc(buf[x], stream) == EOF) {
-          XFREE (buf);
-          return MP_VAL;
-       }
+      if (fputc(buf[x], stream) == EOF) {
+         XFREE(buf);
+         return MP_VAL;
+      }
    }
-   
-   XFREE (buf);
+
+   XFREE(buf);
    return MP_OKAY;
 }
 #endif
diff --git a/libtommath/bn_mp_gcd.c b/libtommath/bn_mp_gcd.c
index b0be8fb..18f6dc3 100644
--- a/libtommath/bn_mp_gcd.c
+++ b/libtommath/bn_mp_gcd.c
@@ -16,87 +16,89 @@
  */
 
 /* Greatest Common Divisor using the binary method */
-int mp_gcd (mp_int * a, mp_int * b, mp_int * c)
+int mp_gcd(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  u, v;
-  int     k, u_lsb, v_lsb, res;
+   mp_int  u, v;
+   int     k, u_lsb, v_lsb, res;
 
-  /* either zero than gcd is the largest */
-  if (mp_iszero (a) == MP_YES) {
-    return mp_abs (b, c);
-  }
-  if (mp_iszero (b) == MP_YES) {
-    return mp_abs (a, c);
-  }
+   /* either zero than gcd is the largest */
+   if (mp_iszero(a) == MP_YES) {
+      return mp_abs(b, c);
+   }
+   if (mp_iszero(b) == MP_YES) {
+      return mp_abs(a, c);
+   }
 
-  /* get copies of a and b we can modify */
-  if ((res = mp_init_copy (&u, a)) != MP_OKAY) {
-    return res;
-  }
+   /* get copies of a and b we can modify */
+   if ((res = mp_init_copy(&u, a)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_init_copy (&v, b)) != MP_OKAY) {
-    goto LBL_U;
-  }
+   if ((res = mp_init_copy(&v, b)) != MP_OKAY) {
+      goto LBL_U;
+   }
 
-  /* must be positive for the remainder of the algorithm */
-  u.sign = v.sign = MP_ZPOS;
+   /* must be positive for the remainder of the algorithm */
+   u.sign = v.sign = MP_ZPOS;
 
-  /* B1.  Find the common power of two for u and v */
-  u_lsb = mp_cnt_lsb(&u);
-  v_lsb = mp_cnt_lsb(&v);
-  k     = MIN(u_lsb, v_lsb);
+   /* B1.  Find the common power of two for u and v */
+   u_lsb = mp_cnt_lsb(&u);
+   v_lsb = mp_cnt_lsb(&v);
+   k     = MIN(u_lsb, v_lsb);
 
-  if (k > 0) {
-     /* divide the power of two out */
-     if ((res = mp_div_2d(&u, k, &u, NULL)) != MP_OKAY) {
-        goto LBL_V;
-     }
+   if (k > 0) {
+      /* divide the power of two out */
+      if ((res = mp_div_2d(&u, k, &u, NULL)) != MP_OKAY) {
+         goto LBL_V;
+      }
 
-     if ((res = mp_div_2d(&v, k, &v, NULL)) != MP_OKAY) {
-        goto LBL_V;
-     }
-  }
+      if ((res = mp_div_2d(&v, k, &v, NULL)) != MP_OKAY) {
+         goto LBL_V;
+      }
+   }
 
-  /* divide any remaining factors of two out */
-  if (u_lsb != k) {
-     if ((res = mp_div_2d(&u, u_lsb - k, &u, NULL)) != MP_OKAY) {
-        goto LBL_V;
-     }
-  }
+   /* divide any remaining factors of two out */
+   if (u_lsb != k) {
+      if ((res = mp_div_2d(&u, u_lsb - k, &u, NULL)) != MP_OKAY) {
+         goto LBL_V;
+      }
+   }
 
-  if (v_lsb != k) {
-     if ((res = mp_div_2d(&v, v_lsb - k, &v, NULL)) != MP_OKAY) {
-        goto LBL_V;
-     }
-  }
+   if (v_lsb != k) {
+      if ((res = mp_div_2d(&v, v_lsb - k, &v, NULL)) != MP_OKAY) {
+         goto LBL_V;
+      }
+   }
 
-  while (mp_iszero(&v) == MP_NO) {
-     /* make sure v is the largest */
-     if (mp_cmp_mag(&u, &v) == MP_GT) {
-        /* swap u and v to make sure v is >= u */
-        mp_exch(&u, &v);
-     }
-     
-     /* subtract smallest from largest */
-     if ((res = s_mp_sub(&v, &u, &v)) != MP_OKAY) {
-        goto LBL_V;
-     }
-     
-     /* Divide out all factors of two */
-     if ((res = mp_div_2d(&v, mp_cnt_lsb(&v), &v, NULL)) != MP_OKAY) {
-        goto LBL_V;
-     } 
-  } 
+   while (mp_iszero(&v) == MP_NO) {
+      /* make sure v is the largest */
+      if (mp_cmp_mag(&u, &v) == MP_GT) {
+         /* swap u and v to make sure v is >= u */
+         mp_exch(&u, &v);
+      }
 
-  /* multiply by 2**k which we divided out at the beginning */
-  if ((res = mp_mul_2d (&u, k, c)) != MP_OKAY) {
-     goto LBL_V;
-  }
-  c->sign = MP_ZPOS;
-  res = MP_OKAY;
-LBL_V:mp_clear (&u);
-LBL_U:mp_clear (&v);
-  return res;
+      /* subtract smallest from largest */
+      if ((res = s_mp_sub(&v, &u, &v)) != MP_OKAY) {
+         goto LBL_V;
+      }
+
+      /* Divide out all factors of two */
+      if ((res = mp_div_2d(&v, mp_cnt_lsb(&v), &v, NULL)) != MP_OKAY) {
+         goto LBL_V;
+      }
+   }
+
+   /* multiply by 2**k which we divided out at the beginning */
+   if ((res = mp_mul_2d(&u, k, c)) != MP_OKAY) {
+      goto LBL_V;
+   }
+   c->sign = MP_ZPOS;
+   res = MP_OKAY;
+LBL_V:
+   mp_clear(&u);
+LBL_U:
+   mp_clear(&v);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_get_int.c b/libtommath/bn_mp_get_int.c
index 5c820f8..a3d1602 100644
--- a/libtommath/bn_mp_get_int.c
+++ b/libtommath/bn_mp_get_int.c
@@ -16,27 +16,27 @@
  */
 
 /* get the lower 32-bits of an mp_int */
-unsigned long mp_get_int(mp_int * a)
+unsigned long mp_get_int(mp_int *a)
 {
-  int i;
-  mp_min_u32 res;
+   int i;
+   mp_min_u32 res;
 
-  if (a->used == 0) {
-     return 0;
-  }
+   if (a->used == 0) {
+      return 0;
+   }
 
-  /* get number of digits of the lsb we have to read */
-  i = MIN(a->used,(int)(((sizeof(unsigned long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
+   /* get number of digits of the lsb we have to read */
+   i = MIN(a->used, (int)(((sizeof(unsigned long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
 
-  /* get most significant digit of result */
-  res = DIGIT(a,i);
+   /* get most significant digit of result */
+   res = DIGIT(a, i);
 
-  while (--i >= 0) {
-    res = (res << DIGIT_BIT) | DIGIT(a,i);
-  }
+   while (--i >= 0) {
+      res = (res << DIGIT_BIT) | DIGIT(a, i);
+   }
 
-  /* force result to 32-bits always so it is consistent on non 32-bit platforms */
-  return res & 0xFFFFFFFFUL;
+   /* force result to 32-bits always so it is consistent on non 32-bit platforms */
+   return res & 0xFFFFFFFFUL;
 }
 #endif
 
diff --git a/libtommath/bn_mp_get_long.c b/libtommath/bn_mp_get_long.c
index 7c3d0fe..053930c 100644
--- a/libtommath/bn_mp_get_long.c
+++ b/libtommath/bn_mp_get_long.c
@@ -16,26 +16,26 @@
  */
 
 /* get the lower unsigned long of an mp_int, platform dependent */
-unsigned long mp_get_long(mp_int * a)
+unsigned long mp_get_long(mp_int *a)
 {
-  int i;
-  unsigned long res;
+   int i;
+   unsigned long res;
 
-  if (a->used == 0) {
-     return 0;
-  }
+   if (a->used == 0) {
+      return 0;
+   }
 
-  /* get number of digits of the lsb we have to read */
-  i = MIN(a->used,(int)(((sizeof(unsigned long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
+   /* get number of digits of the lsb we have to read */
+   i = MIN(a->used, (int)(((sizeof(unsigned long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
 
-  /* get most significant digit of result */
-  res = DIGIT(a,i);
+   /* get most significant digit of result */
+   res = DIGIT(a, i);
 
 #if (ULONG_MAX != 0xffffffffuL) || (DIGIT_BIT < 32)
-  while (--i >= 0) {
-    res = (res << DIGIT_BIT) | DIGIT(a,i);
-  }
+   while (--i >= 0) {
+      res = (res << DIGIT_BIT) | DIGIT(a, i);
+   }
 #endif
-  return res;
+   return res;
 }
 #endif
diff --git a/libtommath/bn_mp_get_long_long.c b/libtommath/bn_mp_get_long_long.c
index 4b959e6..131571a 100644
--- a/libtommath/bn_mp_get_long_long.c
+++ b/libtommath/bn_mp_get_long_long.c
@@ -16,26 +16,26 @@
  */
 
 /* get the lower unsigned long long of an mp_int, platform dependent */
-unsigned long long mp_get_long_long (mp_int * a)
+unsigned long long mp_get_long_long(mp_int *a)
 {
-  int i;
-  unsigned long long res;
+   int i;
+   unsigned long long res;
 
-  if (a->used == 0) {
-     return 0;
-  }
+   if (a->used == 0) {
+      return 0;
+   }
 
-  /* get number of digits of the lsb we have to read */
-  i = MIN(a->used,(int)(((sizeof(unsigned long long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
+   /* get number of digits of the lsb we have to read */
+   i = MIN(a->used, (int)(((sizeof(unsigned long long) * CHAR_BIT) + DIGIT_BIT - 1) / DIGIT_BIT)) - 1;
 
-  /* get most significant digit of result */
-  res = DIGIT(a,i);
+   /* get most significant digit of result */
+   res = DIGIT(a, i);
 
 #if DIGIT_BIT < 64
-  while (--i >= 0) {
-    res = (res << DIGIT_BIT) | DIGIT(a,i);
-  }
+   while (--i >= 0) {
+      res = (res << DIGIT_BIT) | DIGIT(a, i);
+   }
 #endif
-  return res;
+   return res;
 }
 #endif
diff --git a/libtommath/bn_mp_grow.c b/libtommath/bn_mp_grow.c
index 74e07b1..0030931 100644
--- a/libtommath/bn_mp_grow.c
+++ b/libtommath/bn_mp_grow.c
@@ -16,39 +16,39 @@
  */
 
 /* grow as required */
-int mp_grow (mp_int * a, int size)
+int mp_grow(mp_int *a, int size)
 {
-  int     i;
-  mp_digit *tmp;
+   int     i;
+   mp_digit *tmp;
 
-  /* if the alloc size is smaller alloc more ram */
-  if (a->alloc < size) {
-    /* ensure there are always at least MP_PREC digits extra on top */
-    size += (MP_PREC * 2) - (size % MP_PREC);
+   /* if the alloc size is smaller alloc more ram */
+   if (a->alloc < size) {
+      /* ensure there are always at least MP_PREC digits extra on top */
+      size += (MP_PREC * 2) - (size % MP_PREC);
 
-    /* reallocate the array a->dp
-     *
-     * We store the return in a temporary variable
-     * in case the operation failed we don't want
-     * to overwrite the dp member of a.
-     */
-    tmp = OPT_CAST(mp_digit) XREALLOC (a->dp, sizeof (mp_digit) * size);
-    if (tmp == NULL) {
-      /* reallocation failed but "a" is still valid [can be freed] */
-      return MP_MEM;
-    }
+      /* reallocate the array a->dp
+       *
+       * We store the return in a temporary variable
+       * in case the operation failed we don't want
+       * to overwrite the dp member of a.
+       */
+      tmp = OPT_CAST(mp_digit) XREALLOC(a->dp, sizeof(mp_digit) * size);
+      if (tmp == NULL) {
+         /* reallocation failed but "a" is still valid [can be freed] */
+         return MP_MEM;
+      }
 
-    /* reallocation succeeded so set a->dp */
-    a->dp = tmp;
+      /* reallocation succeeded so set a->dp */
+      a->dp = tmp;
 
-    /* zero excess digits */
-    i        = a->alloc;
-    a->alloc = size;
-    for (; i < a->alloc; i++) {
-      a->dp[i] = 0;
-    }
-  }
-  return MP_OKAY;
+      /* zero excess digits */
+      i        = a->alloc;
+      a->alloc = size;
+      for (; i < a->alloc; i++) {
+         a->dp[i] = 0;
+      }
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_import.c b/libtommath/bn_mp_import.c
index df29389..afd735e 100644
--- a/libtommath/bn_mp_import.c
+++ b/libtommath/bn_mp_import.c
@@ -18,52 +18,50 @@
 /* based on gmp's mpz_import.
  * see http://gmplib.org/manual/Integer-Import-and-Export.html
  */
-int mp_import(mp_int* rop, size_t count, int order, size_t size, 
-                            int endian, size_t nails, const void* op) {
-	int result;
-	size_t odd_nails, nail_bytes, i, j;
-	unsigned char odd_nail_mask;
+int mp_import(mp_int *rop, size_t count, int order, size_t size,
+              int endian, size_t nails, const void *op)
+{
+   int result;
+   size_t odd_nails, nail_bytes, i, j;
+   unsigned char odd_nail_mask;
 
-	mp_zero(rop);
+   mp_zero(rop);
 
-	if (endian == 0) {
-		union {
-			unsigned int i;
-			char c[4];
-		} lint;
-		lint.i = 0x01020304;
+   if (endian == 0) {
+      union {
+         unsigned int i;
+         char c[4];
+      } lint;
+      lint.i = 0x01020304;
 
-		endian = (lint.c[0] == 4) ? -1 : 1;
-	}
+      endian = (lint.c[0] == 4) ? -1 : 1;
+   }
 
-	odd_nails = (nails % 8);
-	odd_nail_mask = 0xff;
-	for (i = 0; i < odd_nails; ++i) {
-		odd_nail_mask ^= (1 << (7 - i));
-	}
-	nail_bytes = nails / 8;
+   odd_nails = (nails % 8);
+   odd_nail_mask = 0xff;
+   for (i = 0; i < odd_nails; ++i) {
+      odd_nail_mask ^= (1 << (7 - i));
+   }
+   nail_bytes = nails / 8;
 
-	for (i = 0; i < count; ++i) {
-		for (j = 0; j < (size - nail_bytes); ++j) {
-			unsigned char byte = *(
-					(unsigned char*)op + 
-					(((order == 1) ? i : ((count - 1) - i)) * size) +
-					((endian == 1) ? (j + nail_bytes) : (((size - 1) - j) - nail_bytes))
-				);
+   for (i = 0; i < count; ++i) {
+      for (j = 0; j < (size - nail_bytes); ++j) {
+         unsigned char byte = *((unsigned char *)op +
+                                (((order == 1) ? i : ((count - 1) - i)) * size) +
+                                ((endian == 1) ? (j + nail_bytes) : (((size - 1) - j) - nail_bytes)));
 
-			if (
-				(result = mp_mul_2d(rop, ((j == 0) ? (8 - odd_nails) : 8), rop)) != MP_OKAY) {
-				return result;
-			}
+         if ((result = mp_mul_2d(rop, ((j == 0) ? (8 - odd_nails) : 8), rop)) != MP_OKAY) {
+            return result;
+         }
 
-			rop->dp[0] |= (j == 0) ? (byte & odd_nail_mask) : byte;
-			rop->used  += 1;
-		}
-	}
+         rop->dp[0] |= (j == 0) ? (byte & odd_nail_mask) : byte;
+         rop->used  += 1;
+      }
+   }
 
-	mp_clamp(rop);
+   mp_clamp(rop);
 
-	return MP_OKAY;
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_init.c b/libtommath/bn_mp_init.c
index ee374ae..0556aeb 100644
--- a/libtommath/bn_mp_init.c
+++ b/libtommath/bn_mp_init.c
@@ -16,28 +16,28 @@
  */
 
 /* init a new mp_int */
-int mp_init (mp_int * a)
+int mp_init(mp_int *a)
 {
-  int i;
+   int i;
 
-  /* allocate memory required and clear it */
-  a->dp = OPT_CAST(mp_digit) XMALLOC (sizeof (mp_digit) * MP_PREC);
-  if (a->dp == NULL) {
-    return MP_MEM;
-  }
+   /* allocate memory required and clear it */
+   a->dp = OPT_CAST(mp_digit) XMALLOC(sizeof(mp_digit) * MP_PREC);
+   if (a->dp == NULL) {
+      return MP_MEM;
+   }
 
-  /* set the digits to zero */
-  for (i = 0; i < MP_PREC; i++) {
+   /* set the digits to zero */
+   for (i = 0; i < MP_PREC; i++) {
       a->dp[i] = 0;
-  }
+   }
 
-  /* set the used to zero, allocated digits to the default precision
-   * and sign to positive */
-  a->used  = 0;
-  a->alloc = MP_PREC;
-  a->sign  = MP_ZPOS;
+   /* set the used to zero, allocated digits to the default precision
+    * and sign to positive */
+   a->used  = 0;
+   a->alloc = MP_PREC;
+   a->sign  = MP_ZPOS;
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_init_copy.c b/libtommath/bn_mp_init_copy.c
index 37a57ec..c711e06 100644
--- a/libtommath/bn_mp_init_copy.c
+++ b/libtommath/bn_mp_init_copy.c
@@ -16,19 +16,19 @@
  */
 
 /* creates "a" then copies b into it */
-int mp_init_copy (mp_int * a, mp_int * b)
+int mp_init_copy(mp_int *a, mp_int *b)
 {
-  int     res;
+   int     res;
 
-  if ((res = mp_init_size (a, b->used)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_size(a, b->used)) != MP_OKAY) {
+      return res;
+   }
 
-  if((res = mp_copy (b, a)) != MP_OKAY) {
-    mp_clear(a);
-  }
+   if ((res = mp_copy(b, a)) != MP_OKAY) {
+      mp_clear(a);
+   }
 
-  return res;
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_init_multi.c b/libtommath/bn_mp_init_multi.c
index 73d6a0f..0da7803 100644
--- a/libtommath/bn_mp_init_multi.c
+++ b/libtommath/bn_mp_init_multi.c
@@ -16,37 +16,37 @@
  */
 #include <stdarg.h>
 
-int mp_init_multi(mp_int *mp, ...) 
+int mp_init_multi(mp_int *mp, ...)
 {
-    mp_err res = MP_OKAY;      /* Assume ok until proven otherwise */
-    int n = 0;                 /* Number of ok inits */
-    mp_int* cur_arg = mp;
-    va_list args;
+   mp_err res = MP_OKAY;      /* Assume ok until proven otherwise */
+   int n = 0;                 /* Number of ok inits */
+   mp_int *cur_arg = mp;
+   va_list args;
 
-    va_start(args, mp);        /* init args to next argument from caller */
-    while (cur_arg != NULL) {
-        if (mp_init(cur_arg) != MP_OKAY) {
-            /* Oops - error! Back-track and mp_clear what we already
-               succeeded in init-ing, then return error.
-            */
-            va_list clean_args;
-            
-            /* now start cleaning up */            
-            cur_arg = mp;
-            va_start(clean_args, mp);
-            while (n-- != 0) {
-                mp_clear(cur_arg);
-                cur_arg = va_arg(clean_args, mp_int*);
-            }
-            va_end(clean_args);
-            res = MP_MEM;
-            break;
-        }
-        n++;
-        cur_arg = va_arg(args, mp_int*);
-    }
-    va_end(args);
-    return res;                /* Assumed ok, if error flagged above. */
+   va_start(args, mp);        /* init args to next argument from caller */
+   while (cur_arg != NULL) {
+      if (mp_init(cur_arg) != MP_OKAY) {
+         /* Oops - error! Back-track and mp_clear what we already
+            succeeded in init-ing, then return error.
+         */
+         va_list clean_args;
+
+         /* now start cleaning up */
+         cur_arg = mp;
+         va_start(clean_args, mp);
+         while (n-- != 0) {
+            mp_clear(cur_arg);
+            cur_arg = va_arg(clean_args, mp_int *);
+         }
+         va_end(clean_args);
+         res = MP_MEM;
+         break;
+      }
+      n++;
+      cur_arg = va_arg(args, mp_int *);
+   }
+   va_end(args);
+   return res;                /* Assumed ok, if error flagged above. */
 }
 
 #endif
diff --git a/libtommath/bn_mp_init_set.c b/libtommath/bn_mp_init_set.c
index ed4955c..e9c1b12 100644
--- a/libtommath/bn_mp_init_set.c
+++ b/libtommath/bn_mp_init_set.c
@@ -16,14 +16,14 @@
  */
 
 /* initialize and set a digit */
-int mp_init_set (mp_int * a, mp_digit b)
+int mp_init_set(mp_int *a, mp_digit b)
 {
-  int err;
-  if ((err = mp_init(a)) != MP_OKAY) {
-     return err;
-  }
-  mp_set(a, b);
-  return err;
+   int err;
+   if ((err = mp_init(a)) != MP_OKAY) {
+      return err;
+   }
+   mp_set(a, b);
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_init_set_int.c b/libtommath/bn_mp_init_set_int.c
index 1bc1942..8e7441a 100644
--- a/libtommath/bn_mp_init_set_int.c
+++ b/libtommath/bn_mp_init_set_int.c
@@ -16,13 +16,13 @@
  */
 
 /* initialize and set a digit */
-int mp_init_set_int (mp_int * a, unsigned long b)
+int mp_init_set_int(mp_int *a, unsigned long b)
 {
-  int err;
-  if ((err = mp_init(a)) != MP_OKAY) {
-     return err;
-  }
-  return mp_set_int(a, b);
+   int err;
+   if ((err = mp_init(a)) != MP_OKAY) {
+      return err;
+   }
+   return mp_set_int(a, b);
 }
 #endif
 
diff --git a/libtommath/bn_mp_init_size.c b/libtommath/bn_mp_init_size.c
index 4446773..623a03f 100644
--- a/libtommath/bn_mp_init_size.c
+++ b/libtommath/bn_mp_init_size.c
@@ -16,30 +16,30 @@
  */
 
 /* init an mp_init for a given size */
-int mp_init_size (mp_int * a, int size)
+int mp_init_size(mp_int *a, int size)
 {
-  int x;
+   int x;
 
-  /* pad size so there are always extra digits */
-  size += (MP_PREC * 2) - (size % MP_PREC);	
-  
-  /* alloc mem */
-  a->dp = OPT_CAST(mp_digit) XMALLOC (sizeof (mp_digit) * size);
-  if (a->dp == NULL) {
-    return MP_MEM;
-  }
+   /* pad size so there are always extra digits */
+   size += (MP_PREC * 2) - (size % MP_PREC);
 
-  /* set the members */
-  a->used  = 0;
-  a->alloc = size;
-  a->sign  = MP_ZPOS;
+   /* alloc mem */
+   a->dp = OPT_CAST(mp_digit) XMALLOC(sizeof(mp_digit) * size);
+   if (a->dp == NULL) {
+      return MP_MEM;
+   }
 
-  /* zero the digits */
-  for (x = 0; x < size; x++) {
+   /* set the members */
+   a->used  = 0;
+   a->alloc = size;
+   a->sign  = MP_ZPOS;
+
+   /* zero the digits */
+   for (x = 0; x < size; x++) {
       a->dp[x] = 0;
-  }
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_invmod.c b/libtommath/bn_mp_invmod.c
index 36011d0..b70fe18 100644
--- a/libtommath/bn_mp_invmod.c
+++ b/libtommath/bn_mp_invmod.c
@@ -16,24 +16,24 @@
  */
 
 /* hac 14.61, pp608 */
-int mp_invmod (mp_int * a, mp_int * b, mp_int * c)
+int mp_invmod(mp_int *a, mp_int *b, mp_int *c)
 {
-  /* b cannot be negative */
-  if ((b->sign == MP_NEG) || (mp_iszero(b) == MP_YES)) {
-    return MP_VAL;
-  }
+   /* b cannot be negative */
+   if ((b->sign == MP_NEG) || (mp_iszero(b) == MP_YES)) {
+      return MP_VAL;
+   }
 
 #ifdef BN_FAST_MP_INVMOD_C
-  /* if the modulus is odd we can use a faster routine instead */
-  if ((mp_isodd(b) == MP_YES) && (mp_cmp_d(b, 1) != MP_EQ)) {
-    return fast_mp_invmod (a, b, c);
-  }
+   /* if the modulus is odd we can use a faster routine instead */
+   if ((mp_isodd(b) == MP_YES) && (mp_cmp_d(b, 1) != MP_EQ)) {
+      return fast_mp_invmod(a, b, c);
+   }
 #endif
 
 #ifdef BN_MP_INVMOD_SLOW_C
-  return mp_invmod_slow(a, b, c);
+   return mp_invmod_slow(a, b, c);
 #else
-  return MP_VAL;
+   return MP_VAL;
 #endif
 }
 #endif
diff --git a/libtommath/bn_mp_invmod_slow.c b/libtommath/bn_mp_invmod_slow.c
index ff0d5ae..2bdd2b1 100644
--- a/libtommath/bn_mp_invmod_slow.c
+++ b/libtommath/bn_mp_invmod_slow.c
@@ -16,157 +16,158 @@
  */
 
 /* hac 14.61, pp608 */
-int mp_invmod_slow (mp_int * a, mp_int * b, mp_int * c)
+int mp_invmod_slow(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  x, y, u, v, A, B, C, D;
-  int     res;
-
-  /* b cannot be negative */
-  if ((b->sign == MP_NEG) || (mp_iszero(b) == MP_YES)) {
-    return MP_VAL;
-  }
-
-  /* init temps */
-  if ((res = mp_init_multi(&x, &y, &u, &v, 
-                           &A, &B, &C, &D, NULL)) != MP_OKAY) {
-     return res;
-  }
-
-  /* x = a, y = b */
-  if ((res = mp_mod(a, b, &x)) != MP_OKAY) {
+   mp_int  x, y, u, v, A, B, C, D;
+   int     res;
+
+   /* b cannot be negative */
+   if ((b->sign == MP_NEG) || (mp_iszero(b) == MP_YES)) {
+      return MP_VAL;
+   }
+
+   /* init temps */
+   if ((res = mp_init_multi(&x, &y, &u, &v,
+                            &A, &B, &C, &D, NULL)) != MP_OKAY) {
+      return res;
+   }
+
+   /* x = a, y = b */
+   if ((res = mp_mod(a, b, &x)) != MP_OKAY) {
       goto LBL_ERR;
-  }
-  if ((res = mp_copy (b, &y)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-
-  /* 2. [modified] if x,y are both even then return an error! */
-  if ((mp_iseven (&x) == MP_YES) && (mp_iseven (&y) == MP_YES)) {
-    res = MP_VAL;
-    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 (&A, 1);
-  mp_set (&D, 1);
+   }
+   if ((res = mp_copy(b, &y)) != MP_OKAY) {
+      goto LBL_ERR;
+   }
 
-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) {
+   /* 2. [modified] if x,y are both even then return an error! */
+   if ((mp_iseven(&x) == MP_YES) && (mp_iseven(&y) == MP_YES)) {
+      res = MP_VAL;
+      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;
-    }
-    /* 4.2 if A or B is odd then */
-    if ((mp_isodd (&A) == MP_YES) || (mp_isodd (&B) == MP_YES)) {
-      /* A = (A+y)/2, B = (B-x)/2 */
-      if ((res = mp_add (&A, &y, &A)) != MP_OKAY) {
+   }
+   mp_set(&A, 1);
+   mp_set(&D, 1);
+
+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;
       }
-      if ((res = mp_sub (&B, &x, &B)) != MP_OKAY) {
+      /* 4.2 if A or B is odd then */
+      if ((mp_isodd(&A) == MP_YES) || (mp_isodd(&B) == MP_YES)) {
+         /* A = (A+y)/2, B = (B-x)/2 */
+         if ((res = mp_add(&A, &y, &A)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+         if ((res = mp_sub(&B, &x, &B)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* A = A/2, B = B/2 */
+      if ((res = mp_div_2(&A, &A)) != MP_OKAY) {
          goto LBL_ERR;
       }
-    }
-    /* A = A/2, B = B/2 */
-    if ((res = mp_div_2 (&A, &A)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    if ((res = mp_div_2 (&B, &B)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
+      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 C or D is odd then */
-    if ((mp_isodd (&C) == MP_YES) || (mp_isodd (&D) == MP_YES)) {
-      /* C = (C+y)/2, D = (D-x)/2 */
-      if ((res = mp_add (&C, &y, &C)) != MP_OKAY) {
+   /* 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;
       }
-      if ((res = mp_sub (&D, &x, &D)) != MP_OKAY) {
+      /* 5.2 if C or D is odd then */
+      if ((mp_isodd(&C) == MP_YES) || (mp_isodd(&D) == MP_YES)) {
+         /* C = (C+y)/2, D = (D-x)/2 */
+         if ((res = mp_add(&C, &y, &C)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+         if ((res = mp_sub(&D, &x, &D)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* C = C/2, D = D/2 */
+      if ((res = mp_div_2(&C, &C)) != MP_OKAY) {
          goto LBL_ERR;
       }
-    }
-    /* C = C/2, D = D/2 */
-    if ((res = mp_div_2 (&C, &C)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    if ((res = mp_div_2 (&D, &D)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
+      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, A = A - C, B = B - D */
-    if ((res = mp_sub (&u, &v, &u)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
+   /* 6.  if u >= v then */
+   if (mp_cmp(&u, &v) != MP_LT) {
+      /* u = u - v, A = A - C, B = B - D */
+      if ((res = mp_sub(&u, &v, &u)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
 
-    if ((res = mp_sub (&A, &C, &A)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
+      if ((res = mp_sub(&A, &C, &A)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
 
-    if ((res = mp_sub (&B, &D, &B)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  } else {
-    /* v - v - u, C = C - A, D = D - B */
-    if ((res = mp_sub (&v, &u, &v)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
+      if ((res = mp_sub(&B, &D, &B)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   } else {
+      /* v - v - u, C = C - A, D = D - B */
+      if ((res = mp_sub(&v, &u, &v)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
 
-    if ((res = mp_sub (&C, &A, &C)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
+      if ((res = mp_sub(&C, &A, &C)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
 
-    if ((res = mp_sub (&D, &B, &D)) != 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;
+   /* if not zero goto step 4 */
+   if (mp_iszero(&u) == MP_NO)
+      goto top;
 
-  /* now a = C, b = D, gcd == g*v */
+   /* 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;
-  }
+   /* if v != 1 then there is no inverse */
+   if (mp_cmp_d(&v, 1) != MP_EQ) {
+      res = MP_VAL;
+      goto LBL_ERR;
+   }
 
-  /* if its too low */
-  while (mp_cmp_d(&C, 0) == MP_LT) {
+   /* if its too low */
+   while (mp_cmp_d(&C, 0) == MP_LT) {
       if ((res = mp_add(&C, b, &C)) != MP_OKAY) {
          goto LBL_ERR;
       }
-  }
-  
-  /* too big */
-  while (mp_cmp_mag(&C, b) != MP_LT) {
+   }
+
+   /* 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;
-LBL_ERR:mp_clear_multi (&x, &y, &u, &v, &A, &B, &C, &D, NULL);
-  return res;
+   }
+
+   /* C is now the inverse */
+   mp_exch(&C, c);
+   res = MP_OKAY;
+LBL_ERR:
+   mp_clear_multi(&x, &y, &u, &v, &A, &B, &C, &D, NULL);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_is_square.c b/libtommath/bn_mp_is_square.c
index dd08d58..303fab6 100644
--- a/libtommath/bn_mp_is_square.c
+++ b/libtommath/bn_mp_is_square.c
@@ -17,90 +17,91 @@
 
 /* Check if remainders are possible squares - fast exclude non-squares */
 static const char rem_128[128] = {
- 0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1
+   0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1
 };
 
 static const char rem_105[105] = {
- 0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1,
- 0, 0, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1,
- 0, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1,
- 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,
- 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1,
- 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 1, 1,
- 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1
+   0, 0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1,
+   0, 0, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1,
+   0, 1, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1,
+   1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,
+   0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1,
+   1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 1, 1,
+   1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1
 };
 
 /* Store non-zero to ret if arg is square, and zero if not */
-int mp_is_square(mp_int *arg,int *ret) 
+int mp_is_square(mp_int *arg, int *ret)
 {
-  int           res;
-  mp_digit      c;
-  mp_int        t;
-  unsigned long r;
+   int           res;
+   mp_digit      c;
+   mp_int        t;
+   unsigned long r;
 
-  /* Default to Non-square :) */
-  *ret = MP_NO; 
+   /* Default to Non-square :) */
+   *ret = MP_NO;
 
-  if (arg->sign == MP_NEG) {
-    return MP_VAL;
-  }
+   if (arg->sign == MP_NEG) {
+      return MP_VAL;
+   }
 
-  /* digits used?  (TSD) */
-  if (arg->used == 0) {
-     return MP_OKAY;
-  }
+   /* digits used?  (TSD) */
+   if (arg->used == 0) {
+      return MP_OKAY;
+   }
 
-  /* First check mod 128 (suppose that DIGIT_BIT is at least 7) */
-  if (rem_128[127 & DIGIT(arg,0)] == 1) {
-     return MP_OKAY;
-  }
+   /* First check mod 128 (suppose that DIGIT_BIT is at least 7) */
+   if (rem_128[127 & DIGIT(arg, 0)] == 1) {
+      return MP_OKAY;
+   }
 
-  /* Next check mod 105 (3*5*7) */
-  if ((res = mp_mod_d(arg,105,&c)) != MP_OKAY) {
-     return res;
-  }
-  if (rem_105[c] == 1) {
-     return MP_OKAY;
-  }
+   /* Next check mod 105 (3*5*7) */
+   if ((res = mp_mod_d(arg, 105, &c)) != MP_OKAY) {
+      return res;
+   }
+   if (rem_105[c] == 1) {
+      return MP_OKAY;
+   }
 
 
-  if ((res = mp_init_set_int(&t,11L*13L*17L*19L*23L*29L*31L)) != MP_OKAY) {
-     return res;
-  }
-  if ((res = mp_mod(arg,&t,&t)) != MP_OKAY) {
-     goto ERR;
-  }
-  r = mp_get_int(&t);
-  /* Check for other prime modules, note it's not an ERROR but we must
-   * free "t" so the easiest way is to goto ERR.  We know that res
-   * is already equal to MP_OKAY from the mp_mod call 
-   */ 
-  if (((1L<<(r%11)) & 0x5C4L) != 0L)       goto ERR;
-  if (((1L<<(r%13)) & 0x9E4L) != 0L)       goto ERR;
-  if (((1L<<(r%17)) & 0x5CE8L) != 0L)      goto ERR;
-  if (((1L<<(r%19)) & 0x4F50CL) != 0L)     goto ERR;
-  if (((1L<<(r%23)) & 0x7ACCA0L) != 0L)    goto ERR;
-  if (((1L<<(r%29)) & 0xC2EDD0CL) != 0L)   goto ERR;
-  if (((1L<<(r%31)) & 0x6DE2B848L) != 0L)  goto ERR;
+   if ((res = mp_init_set_int(&t, 11L*13L*17L*19L*23L*29L*31L)) != MP_OKAY) {
+      return res;
+   }
+   if ((res = mp_mod(arg, &t, &t)) != MP_OKAY) {
+      goto ERR;
+   }
+   r = mp_get_int(&t);
+   /* Check for other prime modules, note it's not an ERROR but we must
+    * free "t" so the easiest way is to goto ERR.  We know that res
+    * is already equal to MP_OKAY from the mp_mod call
+    */
+   if (((1L<<(r%11)) & 0x5C4L) != 0L)       goto ERR;
+   if (((1L<<(r%13)) & 0x9E4L) != 0L)       goto ERR;
+   if (((1L<<(r%17)) & 0x5CE8L) != 0L)      goto ERR;
+   if (((1L<<(r%19)) & 0x4F50CL) != 0L)     goto ERR;
+   if (((1L<<(r%23)) & 0x7ACCA0L) != 0L)    goto ERR;
+   if (((1L<<(r%29)) & 0xC2EDD0CL) != 0L)   goto ERR;
+   if (((1L<<(r%31)) & 0x6DE2B848L) != 0L)  goto ERR;
 
-  /* Final check - is sqr(sqrt(arg)) == arg ? */
-  if ((res = mp_sqrt(arg,&t)) != MP_OKAY) {
-     goto ERR;
-  }
-  if ((res = mp_sqr(&t,&t)) != MP_OKAY) {
-     goto ERR;
-  }
+   /* Final check - is sqr(sqrt(arg)) == arg ? */
+   if ((res = mp_sqrt(arg, &t)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sqr(&t, &t)) != MP_OKAY) {
+      goto ERR;
+   }
 
-  *ret = (mp_cmp_mag(&t,arg) == MP_EQ) ? MP_YES : MP_NO;
-ERR:mp_clear(&t);
-  return res;
+   *ret = (mp_cmp_mag(&t, arg) == MP_EQ) ? MP_YES : MP_NO;
+ERR:
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_jacobi.c b/libtommath/bn_mp_jacobi.c
index 5fc8593..8981393 100644
--- a/libtommath/bn_mp_jacobi.c
+++ b/libtommath/bn_mp_jacobi.c
@@ -20,95 +20,97 @@
  * HAC is wrong here, as the special case of (0 | 1) is not
  * handled correctly.
  */
-int mp_jacobi (mp_int * a, mp_int * n, int *c)
+int mp_jacobi(mp_int *a, mp_int *n, int *c)
 {
-  mp_int  a1, p1;
-  int     k, s, r, res;
-  mp_digit residue;
+   mp_int  a1, p1;
+   int     k, s, r, res;
+   mp_digit residue;
 
-  /* if a < 0 return MP_VAL */
-  if (mp_isneg(a) == MP_YES) {
-     return MP_VAL;
-  }
+   /* if a < 0 return MP_VAL */
+   if (mp_isneg(a) == MP_YES) {
+      return MP_VAL;
+   }
 
-  /* if n <= 0 return MP_VAL */
-  if (mp_cmp_d(n, 0) != MP_GT) {
-     return MP_VAL;
-  }
+   /* if n <= 0 return MP_VAL */
+   if (mp_cmp_d(n, 0) != MP_GT) {
+      return MP_VAL;
+   }
 
-  /* step 1. handle case of a == 0 */
-  if (mp_iszero (a) == MP_YES) {
-     /* special case of a == 0 and n == 1 */
-     if (mp_cmp_d (n, 1) == MP_EQ) {
-       *c = 1;
-     } else {
-       *c = 0;
-     }
-     return MP_OKAY;
-  }
+   /* step 1. handle case of a == 0 */
+   if (mp_iszero(a) == MP_YES) {
+      /* special case of a == 0 and n == 1 */
+      if (mp_cmp_d(n, 1) == MP_EQ) {
+         *c = 1;
+      } else {
+         *c = 0;
+      }
+      return MP_OKAY;
+   }
 
-  /* step 2.  if a == 1, return 1 */
-  if (mp_cmp_d (a, 1) == MP_EQ) {
-    *c = 1;
-    return MP_OKAY;
-  }
+   /* step 2.  if a == 1, return 1 */
+   if (mp_cmp_d(a, 1) == MP_EQ) {
+      *c = 1;
+      return MP_OKAY;
+   }
 
-  /* default */
-  s = 0;
+   /* default */
+   s = 0;
 
-  /* step 3.  write a = a1 * 2**k  */
-  if ((res = mp_init_copy (&a1, a)) != MP_OKAY) {
-    return res;
-  }
+   /* step 3.  write a = a1 * 2**k  */
+   if ((res = mp_init_copy(&a1, a)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_init (&p1)) != MP_OKAY) {
-    goto LBL_A1;
-  }
+   if ((res = mp_init(&p1)) != MP_OKAY) {
+      goto LBL_A1;
+   }
 
-  /* divide out larger power of two */
-  k = mp_cnt_lsb(&a1);
-  if ((res = mp_div_2d(&a1, k, &a1, NULL)) != MP_OKAY) {
-     goto LBL_P1;
-  }
-
-  /* step 4.  if e is even set s=1 */
-  if ((k & 1) == 0) {
-    s = 1;
-  } else {
-    /* else set s=1 if p = 1/7 (mod 8) or s=-1 if p = 3/5 (mod 8) */
-    residue = n->dp[0] & 7;
+   /* divide out larger power of two */
+   k = mp_cnt_lsb(&a1);
+   if ((res = mp_div_2d(&a1, k, &a1, NULL)) != MP_OKAY) {
+      goto LBL_P1;
+   }
 
-    if ((residue == 1) || (residue == 7)) {
+   /* step 4.  if e is even set s=1 */
+   if ((k & 1) == 0) {
       s = 1;
-    } else if ((residue == 3) || (residue == 5)) {
-      s = -1;
-    }
-  }
+   } else {
+      /* else set s=1 if p = 1/7 (mod 8) or s=-1 if p = 3/5 (mod 8) */
+      residue = n->dp[0] & 7;
 
-  /* step 5.  if p == 3 (mod 4) *and* a1 == 3 (mod 4) then s = -s */
-  if ( ((n->dp[0] & 3) == 3) && ((a1.dp[0] & 3) == 3)) {
-    s = -s;
-  }
+      if ((residue == 1) || (residue == 7)) {
+         s = 1;
+      } else if ((residue == 3) || (residue == 5)) {
+         s = -1;
+      }
+   }
 
-  /* if a1 == 1 we're done */
-  if (mp_cmp_d (&a1, 1) == MP_EQ) {
-    *c = s;
-  } else {
-    /* n1 = n mod a1 */
-    if ((res = mp_mod (n, &a1, &p1)) != MP_OKAY) {
-      goto LBL_P1;
-    }
-    if ((res = mp_jacobi (&p1, &a1, &r)) != MP_OKAY) {
-      goto LBL_P1;
-    }
-    *c = s * r;
-  }
+   /* step 5.  if p == 3 (mod 4) *and* a1 == 3 (mod 4) then s = -s */
+   if (((n->dp[0] & 3) == 3) && ((a1.dp[0] & 3) == 3)) {
+      s = -s;
+   }
+
+   /* if a1 == 1 we're done */
+   if (mp_cmp_d(&a1, 1) == MP_EQ) {
+      *c = s;
+   } else {
+      /* n1 = n mod a1 */
+      if ((res = mp_mod(n, &a1, &p1)) != MP_OKAY) {
+         goto LBL_P1;
+      }
+      if ((res = mp_jacobi(&p1, &a1, &r)) != MP_OKAY) {
+         goto LBL_P1;
+      }
+      *c = s * r;
+   }
 
-  /* done */
-  res = MP_OKAY;
-LBL_P1:mp_clear (&p1);
-LBL_A1:mp_clear (&a1);
-  return res;
+   /* done */
+   res = MP_OKAY;
+LBL_P1:
+   mp_clear(&p1);
+LBL_A1:
+   mp_clear(&a1);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_karatsuba_mul.c b/libtommath/bn_mp_karatsuba_mul.c
index 4d982c7..353c37c 100644
--- a/libtommath/bn_mp_karatsuba_mul.c
+++ b/libtommath/bn_mp_karatsuba_mul.c
@@ -15,150 +15,157 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* c = |a| * |b| using Karatsuba Multiplication using 
+/* c = |a| * |b| using Karatsuba Multiplication using
  * three half size multiplications
  *
- * Let B represent the radix [e.g. 2**DIGIT_BIT] and 
- * let n represent half of the number of digits in 
+ * Let B represent the radix [e.g. 2**DIGIT_BIT] and
+ * let n represent half of the number of digits in
  * the min(a,b)
  *
  * a = a1 * B**n + a0
  * b = b1 * B**n + b0
  *
- * Then, a * b => 
+ * Then, a * b =>
    a1b1 * B**2n + ((a1 + a0)(b1 + b0) - (a0b0 + a1b1)) * B + a0b0
  *
- * Note that a1b1 and a0b0 are used twice and only need to be 
- * computed once.  So in total three half size (half # of 
- * digit) multiplications are performed, a0b0, a1b1 and 
+ * Note that a1b1 and a0b0 are used twice and only need to be
+ * computed once.  So in total three half size (half # of
+ * digit) multiplications are performed, a0b0, a1b1 and
  * (a1+b1)(a0+b0)
  *
  * Note that a multiplication of half the digits requires
- * 1/4th the number of single precision multiplications so in 
- * total after one call 25% of the single precision multiplications 
- * are saved.  Note also that the call to mp_mul can end up back 
- * in this function if the a0, a1, b0, or b1 are above the threshold.  
- * This is known as divide-and-conquer and leads to the famous 
- * O(N**lg(3)) or O(N**1.584) work which is asymptopically lower than 
- * the standard O(N**2) that the baseline/comba methods use.  
- * Generally though the overhead of this method doesn't pay off 
+ * 1/4th the number of single precision multiplications so in
+ * total after one call 25% of the single precision multiplications
+ * are saved.  Note also that the call to mp_mul can end up back
+ * in this function if the a0, a1, b0, or b1 are above the threshold.
+ * This is known as divide-and-conquer and leads to the famous
+ * O(N**lg(3)) or O(N**1.584) work which is asymptopically lower than
+ * the standard O(N**2) that the baseline/comba methods use.
+ * Generally though the overhead of this method doesn't pay off
  * until a certain size (N ~ 80) is reached.
  */
-int mp_karatsuba_mul (mp_int * a, mp_int * b, mp_int * c)
+int mp_karatsuba_mul(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  x0, x1, y0, y1, t1, x0y0, x1y1;
-  int     B, err;
-
-  /* default the return code to an error */
-  err = MP_MEM;
-
-  /* min # of digits */
-  B = MIN (a->used, b->used);
-
-  /* now divide in two */
-  B = B >> 1;
-
-  /* init copy all the temps */
-  if (mp_init_size (&x0, B) != MP_OKAY)
-    goto ERR;
-  if (mp_init_size (&x1, a->used - B) != MP_OKAY)
-    goto X0;
-  if (mp_init_size (&y0, B) != MP_OKAY)
-    goto X1;
-  if (mp_init_size (&y1, b->used - B) != MP_OKAY)
-    goto Y0;
-
-  /* init temps */
-  if (mp_init_size (&t1, B * 2) != MP_OKAY)
-    goto Y1;
-  if (mp_init_size (&x0y0, B * 2) != MP_OKAY)
-    goto T1;
-  if (mp_init_size (&x1y1, B * 2) != MP_OKAY)
-    goto X0Y0;
-
-  /* now shift the digits */
-  x0.used = y0.used = B;
-  x1.used = a->used - B;
-  y1.used = b->used - B;
-
-  {
-    int x;
-    mp_digit *tmpa, *tmpb, *tmpx, *tmpy;
-
-    /* we copy the digits directly instead of using higher level functions
-     * since we also need to shift the digits
-     */
-    tmpa = a->dp;
-    tmpb = b->dp;
-
-    tmpx = x0.dp;
-    tmpy = y0.dp;
-    for (x = 0; x < B; x++) {
-      *tmpx++ = *tmpa++;
-      *tmpy++ = *tmpb++;
-    }
-
-    tmpx = x1.dp;
-    for (x = B; x < a->used; x++) {
-      *tmpx++ = *tmpa++;
-    }
-
-    tmpy = y1.dp;
-    for (x = B; x < b->used; x++) {
-      *tmpy++ = *tmpb++;
-    }
-  }
-
-  /* only need to clamp the lower words since by definition the 
-   * upper words x1/y1 must have a known number of digits
-   */
-  mp_clamp (&x0);
-  mp_clamp (&y0);
-
-  /* now calc the products x0y0 and x1y1 */
-  /* after this x0 is no longer required, free temp [x0==t2]! */
-  if (mp_mul (&x0, &y0, &x0y0) != MP_OKAY)  
-    goto X1Y1;          /* x0y0 = x0*y0 */
-  if (mp_mul (&x1, &y1, &x1y1) != MP_OKAY)
-    goto X1Y1;          /* x1y1 = x1*y1 */
-
-  /* now calc x1+x0 and y1+y0 */
-  if (s_mp_add (&x1, &x0, &t1) != MP_OKAY)
-    goto X1Y1;          /* t1 = x1 - x0 */
-  if (s_mp_add (&y1, &y0, &x0) != MP_OKAY)
-    goto X1Y1;          /* t2 = y1 - y0 */
-  if (mp_mul (&t1, &x0, &t1) != MP_OKAY)
-    goto X1Y1;          /* t1 = (x1 + x0) * (y1 + y0) */
-
-  /* add x0y0 */
-  if (mp_add (&x0y0, &x1y1, &x0) != MP_OKAY)
-    goto X1Y1;          /* t2 = x0y0 + x1y1 */
-  if (s_mp_sub (&t1, &x0, &t1) != MP_OKAY)
-    goto X1Y1;          /* t1 = (x1+x0)*(y1+y0) - (x1y1 + x0y0) */
-
-  /* shift by B */
-  if (mp_lshd (&t1, B) != MP_OKAY)
-    goto X1Y1;          /* t1 = (x0y0 + x1y1 - (x1-x0)*(y1-y0))<<B */
-  if (mp_lshd (&x1y1, B * 2) != MP_OKAY)
-    goto X1Y1;          /* x1y1 = x1y1 << 2*B */
-
-  if (mp_add (&x0y0, &t1, &t1) != MP_OKAY)
-    goto X1Y1;          /* t1 = x0y0 + t1 */
-  if (mp_add (&t1, &x1y1, c) != MP_OKAY)
-    goto X1Y1;          /* t1 = x0y0 + t1 + x1y1 */
-
-  /* Algorithm succeeded set the return code to MP_OKAY */
-  err = MP_OKAY;
-
-X1Y1:mp_clear (&x1y1);
-X0Y0:mp_clear (&x0y0);
-T1:mp_clear (&t1);
-Y1:mp_clear (&y1);
-Y0:mp_clear (&y0);
-X1:mp_clear (&x1);
-X0:mp_clear (&x0);
+   mp_int  x0, x1, y0, y1, t1, x0y0, x1y1;
+   int     B, err;
+
+   /* default the return code to an error */
+   err = MP_MEM;
+
+   /* min # of digits */
+   B = MIN(a->used, b->used);
+
+   /* now divide in two */
+   B = B >> 1;
+
+   /* init copy all the temps */
+   if (mp_init_size(&x0, B) != MP_OKAY)
+      goto ERR;
+   if (mp_init_size(&x1, a->used - B) != MP_OKAY)
+      goto X0;
+   if (mp_init_size(&y0, B) != MP_OKAY)
+      goto X1;
+   if (mp_init_size(&y1, b->used - B) != MP_OKAY)
+      goto Y0;
+
+   /* init temps */
+   if (mp_init_size(&t1, B * 2) != MP_OKAY)
+      goto Y1;
+   if (mp_init_size(&x0y0, B * 2) != MP_OKAY)
+      goto T1;
+   if (mp_init_size(&x1y1, B * 2) != MP_OKAY)
+      goto X0Y0;
+
+   /* now shift the digits */
+   x0.used = y0.used = B;
+   x1.used = a->used - B;
+   y1.used = b->used - B;
+
+   {
+      int x;
+      mp_digit *tmpa, *tmpb, *tmpx, *tmpy;
+
+      /* we copy the digits directly instead of using higher level functions
+       * since we also need to shift the digits
+       */
+      tmpa = a->dp;
+      tmpb = b->dp;
+
+      tmpx = x0.dp;
+      tmpy = y0.dp;
+      for (x = 0; x < B; x++) {
+         *tmpx++ = *tmpa++;
+         *tmpy++ = *tmpb++;
+      }
+
+      tmpx = x1.dp;
+      for (x = B; x < a->used; x++) {
+         *tmpx++ = *tmpa++;
+      }
+
+      tmpy = y1.dp;
+      for (x = B; x < b->used; x++) {
+         *tmpy++ = *tmpb++;
+      }
+   }
+
+   /* only need to clamp the lower words since by definition the
+    * upper words x1/y1 must have a known number of digits
+    */
+   mp_clamp(&x0);
+   mp_clamp(&y0);
+
+   /* now calc the products x0y0 and x1y1 */
+   /* after this x0 is no longer required, free temp [x0==t2]! */
+   if (mp_mul(&x0, &y0, &x0y0) != MP_OKAY)
+      goto X1Y1;          /* x0y0 = x0*y0 */
+   if (mp_mul(&x1, &y1, &x1y1) != MP_OKAY)
+      goto X1Y1;          /* x1y1 = x1*y1 */
+
+   /* now calc x1+x0 and y1+y0 */
+   if (s_mp_add(&x1, &x0, &t1) != MP_OKAY)
+      goto X1Y1;          /* t1 = x1 - x0 */
+   if (s_mp_add(&y1, &y0, &x0) != MP_OKAY)
+      goto X1Y1;          /* t2 = y1 - y0 */
+   if (mp_mul(&t1, &x0, &t1) != MP_OKAY)
+      goto X1Y1;          /* t1 = (x1 + x0) * (y1 + y0) */
+
+   /* add x0y0 */
+   if (mp_add(&x0y0, &x1y1, &x0) != MP_OKAY)
+      goto X1Y1;          /* t2 = x0y0 + x1y1 */
+   if (s_mp_sub(&t1, &x0, &t1) != MP_OKAY)
+      goto X1Y1;          /* t1 = (x1+x0)*(y1+y0) - (x1y1 + x0y0) */
+
+   /* shift by B */
+   if (mp_lshd(&t1, B) != MP_OKAY)
+      goto X1Y1;          /* t1 = (x0y0 + x1y1 - (x1-x0)*(y1-y0))<<B */
+   if (mp_lshd(&x1y1, B * 2) != MP_OKAY)
+      goto X1Y1;          /* x1y1 = x1y1 << 2*B */
+
+   if (mp_add(&x0y0, &t1, &t1) != MP_OKAY)
+      goto X1Y1;          /* t1 = x0y0 + t1 */
+   if (mp_add(&t1, &x1y1, c) != MP_OKAY)
+      goto X1Y1;          /* t1 = x0y0 + t1 + x1y1 */
+
+   /* Algorithm succeeded set the return code to MP_OKAY */
+   err = MP_OKAY;
+
+X1Y1:
+   mp_clear(&x1y1);
+X0Y0:
+   mp_clear(&x0y0);
+T1:
+   mp_clear(&t1);
+Y1:
+   mp_clear(&y1);
+Y0:
+   mp_clear(&y0);
+X1:
+   mp_clear(&x1);
+X0:
+   mp_clear(&x0);
 ERR:
-  return err;
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_karatsuba_sqr.c b/libtommath/bn_mp_karatsuba_sqr.c
index 764e85a..fe39a33 100644
--- a/libtommath/bn_mp_karatsuba_sqr.c
+++ b/libtommath/bn_mp_karatsuba_sqr.c
@@ -15,104 +15,110 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* Karatsuba squaring, computes b = a*a using three 
+/* Karatsuba squaring, computes b = a*a using three
  * half size squarings
  *
- * See comments of karatsuba_mul for details.  It 
- * is essentially the same algorithm but merely 
+ * See comments of karatsuba_mul for details.  It
+ * is essentially the same algorithm but merely
  * tuned to perform recursive squarings.
  */
-int mp_karatsuba_sqr (mp_int * a, mp_int * b)
+int mp_karatsuba_sqr(mp_int *a, mp_int *b)
 {
-  mp_int  x0, x1, t1, t2, x0x0, x1x1;
-  int     B, err;
-
-  err = MP_MEM;
-
-  /* min # of digits */
-  B = a->used;
-
-  /* now divide in two */
-  B = B >> 1;
-
-  /* init copy all the temps */
-  if (mp_init_size (&x0, B) != MP_OKAY)
-    goto ERR;
-  if (mp_init_size (&x1, a->used - B) != MP_OKAY)
-    goto X0;
-
-  /* init temps */
-  if (mp_init_size (&t1, a->used * 2) != MP_OKAY)
-    goto X1;
-  if (mp_init_size (&t2, a->used * 2) != MP_OKAY)
-    goto T1;
-  if (mp_init_size (&x0x0, B * 2) != MP_OKAY)
-    goto T2;
-  if (mp_init_size (&x1x1, (a->used - B) * 2) != MP_OKAY)
-    goto X0X0;
-
-  {
-    int x;
-    mp_digit *dst, *src;
-
-    src = a->dp;
-
-    /* now shift the digits */
-    dst = x0.dp;
-    for (x = 0; x < B; x++) {
-      *dst++ = *src++;
-    }
-
-    dst = x1.dp;
-    for (x = B; x < a->used; x++) {
-      *dst++ = *src++;
-    }
-  }
-
-  x0.used = B;
-  x1.used = a->used - B;
-
-  mp_clamp (&x0);
-
-  /* now calc the products x0*x0 and x1*x1 */
-  if (mp_sqr (&x0, &x0x0) != MP_OKAY)
-    goto X1X1;           /* x0x0 = x0*x0 */
-  if (mp_sqr (&x1, &x1x1) != MP_OKAY)
-    goto X1X1;           /* x1x1 = x1*x1 */
-
-  /* now calc (x1+x0)**2 */
-  if (s_mp_add (&x1, &x0, &t1) != MP_OKAY)
-    goto X1X1;           /* t1 = x1 - x0 */
-  if (mp_sqr (&t1, &t1) != MP_OKAY)
-    goto X1X1;           /* t1 = (x1 - x0) * (x1 - x0) */
-
-  /* add x0y0 */
-  if (s_mp_add (&x0x0, &x1x1, &t2) != MP_OKAY)
-    goto X1X1;           /* t2 = x0x0 + x1x1 */
-  if (s_mp_sub (&t1, &t2, &t1) != MP_OKAY)
-    goto X1X1;           /* t1 = (x1+x0)**2 - (x0x0 + x1x1) */
-
-  /* shift by B */
-  if (mp_lshd (&t1, B) != MP_OKAY)
-    goto X1X1;           /* t1 = (x0x0 + x1x1 - (x1-x0)*(x1-x0))<<B */
-  if (mp_lshd (&x1x1, B * 2) != MP_OKAY)
-    goto X1X1;           /* x1x1 = x1x1 << 2*B */
-
-  if (mp_add (&x0x0, &t1, &t1) != MP_OKAY)
-    goto X1X1;           /* t1 = x0x0 + t1 */
-  if (mp_add (&t1, &x1x1, b) != MP_OKAY)
-    goto X1X1;           /* t1 = x0x0 + t1 + x1x1 */
-
-  err = MP_OKAY;
-
-X1X1:mp_clear (&x1x1);
-X0X0:mp_clear (&x0x0);
-T2:mp_clear (&t2);
-T1:mp_clear (&t1);
-X1:mp_clear (&x1);
-X0:mp_clear (&x0);
+   mp_int  x0, x1, t1, t2, x0x0, x1x1;
+   int     B, err;
+
+   err = MP_MEM;
+
+   /* min # of digits */
+   B = a->used;
+
+   /* now divide in two */
+   B = B >> 1;
+
+   /* init copy all the temps */
+   if (mp_init_size(&x0, B) != MP_OKAY)
+      goto ERR;
+   if (mp_init_size(&x1, a->used - B) != MP_OKAY)
+      goto X0;
+
+   /* init temps */
+   if (mp_init_size(&t1, a->used * 2) != MP_OKAY)
+      goto X1;
+   if (mp_init_size(&t2, a->used * 2) != MP_OKAY)
+      goto T1;
+   if (mp_init_size(&x0x0, B * 2) != MP_OKAY)
+      goto T2;
+   if (mp_init_size(&x1x1, (a->used - B) * 2) != MP_OKAY)
+      goto X0X0;
+
+   {
+      int x;
+      mp_digit *dst, *src;
+
+      src = a->dp;
+
+      /* now shift the digits */
+      dst = x0.dp;
+      for (x = 0; x < B; x++) {
+         *dst++ = *src++;
+      }
+
+      dst = x1.dp;
+      for (x = B; x < a->used; x++) {
+         *dst++ = *src++;
+      }
+   }
+
+   x0.used = B;
+   x1.used = a->used - B;
+
+   mp_clamp(&x0);
+
+   /* now calc the products x0*x0 and x1*x1 */
+   if (mp_sqr(&x0, &x0x0) != MP_OKAY)
+      goto X1X1;           /* x0x0 = x0*x0 */
+   if (mp_sqr(&x1, &x1x1) != MP_OKAY)
+      goto X1X1;           /* x1x1 = x1*x1 */
+
+   /* now calc (x1+x0)**2 */
+   if (s_mp_add(&x1, &x0, &t1) != MP_OKAY)
+      goto X1X1;           /* t1 = x1 - x0 */
+   if (mp_sqr(&t1, &t1) != MP_OKAY)
+      goto X1X1;           /* t1 = (x1 - x0) * (x1 - x0) */
+
+   /* add x0y0 */
+   if (s_mp_add(&x0x0, &x1x1, &t2) != MP_OKAY)
+      goto X1X1;           /* t2 = x0x0 + x1x1 */
+   if (s_mp_sub(&t1, &t2, &t1) != MP_OKAY)
+      goto X1X1;           /* t1 = (x1+x0)**2 - (x0x0 + x1x1) */
+
+   /* shift by B */
+   if (mp_lshd(&t1, B) != MP_OKAY)
+      goto X1X1;           /* t1 = (x0x0 + x1x1 - (x1-x0)*(x1-x0))<<B */
+   if (mp_lshd(&x1x1, B * 2) != MP_OKAY)
+      goto X1X1;           /* x1x1 = x1x1 << 2*B */
+
+   if (mp_add(&x0x0, &t1, &t1) != MP_OKAY)
+      goto X1X1;           /* t1 = x0x0 + t1 */
+   if (mp_add(&t1, &x1x1, b) != MP_OKAY)
+      goto X1X1;           /* t1 = x0x0 + t1 + x1x1 */
+
+   err = MP_OKAY;
+
+X1X1:
+   mp_clear(&x1x1);
+X0X0:
+   mp_clear(&x0x0);
+T2:
+   mp_clear(&t2);
+T1:
+   mp_clear(&t1);
+X1:
+   mp_clear(&x1);
+X0:
+   mp_clear(&x0);
 ERR:
-  return err;
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_lcm.c b/libtommath/bn_mp_lcm.c
index 512e8ec..dc661f3 100644
--- a/libtommath/bn_mp_lcm.c
+++ b/libtommath/bn_mp_lcm.c
@@ -16,42 +16,42 @@
  */
 
 /* computes least common multiple as |a*b|/(a, b) */
-int mp_lcm (mp_int * a, mp_int * b, mp_int * c)
+int mp_lcm(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res;
-  mp_int  t1, t2;
-
-
-  if ((res = mp_init_multi (&t1, &t2, NULL)) != MP_OKAY) {
-    return res;
-  }
-
-  /* t1 = get the GCD of the two inputs */
-  if ((res = mp_gcd (a, b, &t1)) != MP_OKAY) {
-    goto LBL_T;
-  }
-
-  /* divide the smallest by the GCD */
-  if (mp_cmp_mag(a, b) == MP_LT) {
-     /* store quotient in t2 such that t2 * b is the LCM */
-     if ((res = mp_div(a, &t1, &t2, NULL)) != MP_OKAY) {
-        goto LBL_T;
-     }
-     res = mp_mul(b, &t2, c);
-  } else {
-     /* store quotient in t2 such that t2 * a is the LCM */
-     if ((res = mp_div(b, &t1, &t2, NULL)) != MP_OKAY) {
-        goto LBL_T;
-     }
-     res = mp_mul(a, &t2, c);
-  }
-
-  /* fix the sign to positive */
-  c->sign = MP_ZPOS;
+   int     res;
+   mp_int  t1, t2;
+
+
+   if ((res = mp_init_multi(&t1, &t2, NULL)) != MP_OKAY) {
+      return res;
+   }
+
+   /* t1 = get the GCD of the two inputs */
+   if ((res = mp_gcd(a, b, &t1)) != MP_OKAY) {
+      goto LBL_T;
+   }
+
+   /* divide the smallest by the GCD */
+   if (mp_cmp_mag(a, b) == MP_LT) {
+      /* store quotient in t2 such that t2 * b is the LCM */
+      if ((res = mp_div(a, &t1, &t2, NULL)) != MP_OKAY) {
+         goto LBL_T;
+      }
+      res = mp_mul(b, &t2, c);
+   } else {
+      /* store quotient in t2 such that t2 * a is the LCM */
+      if ((res = mp_div(b, &t1, &t2, NULL)) != MP_OKAY) {
+         goto LBL_T;
+      }
+      res = mp_mul(a, &t2, c);
+   }
+
+   /* fix the sign to positive */
+   c->sign = MP_ZPOS;
 
 LBL_T:
-  mp_clear_multi (&t1, &t2, NULL);
-  return res;
+   mp_clear_multi(&t1, &t2, NULL);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_lshd.c b/libtommath/bn_mp_lshd.c
index 0143e94..888989a 100644
--- a/libtommath/bn_mp_lshd.c
+++ b/libtommath/bn_mp_lshd.c
@@ -16,49 +16,49 @@
  */
 
 /* shift left a certain amount of digits */
-int mp_lshd (mp_int * a, int b)
+int mp_lshd(mp_int *a, int b)
 {
-  int     x, res;
+   int     x, res;
 
-  /* if its less than zero return */
-  if (b <= 0) {
-    return MP_OKAY;
-  }
+   /* if its less than zero return */
+   if (b <= 0) {
+      return MP_OKAY;
+   }
 
-  /* grow to fit the new digits */
-  if (a->alloc < (a->used + b)) {
-     if ((res = mp_grow (a, a->used + b)) != MP_OKAY) {
-       return res;
-     }
-  }
+   /* grow to fit the new digits */
+   if (a->alloc < (a->used + b)) {
+      if ((res = mp_grow(a, a->used + b)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  {
-    mp_digit *top, *bottom;
+   {
+      mp_digit *top, *bottom;
 
-    /* increment the used by the shift amount then copy upwards */
-    a->used += b;
+      /* increment the used by the shift amount then copy upwards */
+      a->used += b;
 
-    /* top */
-    top = a->dp + a->used - 1;
+      /* top */
+      top = a->dp + a->used - 1;
 
-    /* base */
-    bottom = (a->dp + a->used - 1) - b;
+      /* base */
+      bottom = (a->dp + a->used - 1) - b;
 
-    /* much like mp_rshd this is implemented using a sliding window
-     * except the window goes the otherway around.  Copying from
-     * the bottom to the top.  see bn_mp_rshd.c for more info.
-     */
-    for (x = a->used - 1; x >= b; x--) {
-      *top-- = *bottom--;
-    }
+      /* much like mp_rshd this is implemented using a sliding window
+       * except the window goes the otherway around.  Copying from
+       * the bottom to the top.  see bn_mp_rshd.c for more info.
+       */
+      for (x = a->used - 1; x >= b; x--) {
+         *top-- = *bottom--;
+      }
 
-    /* zero the lower digits */
-    top = a->dp;
-    for (x = 0; x < b; x++) {
-      *top++ = 0;
-    }
-  }
-  return MP_OKAY;
+      /* zero the lower digits */
+      top = a->dp;
+      for (x = 0; x < b; x++) {
+         *top++ = 0;
+      }
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mod.c b/libtommath/bn_mp_mod.c
index 06240a0..267688c 100644
--- a/libtommath/bn_mp_mod.c
+++ b/libtommath/bn_mp_mod.c
@@ -16,30 +16,29 @@
  */
 
 /* c = a mod b, 0 <= c < b if b > 0, b < c <= 0 if b < 0 */
-int
-mp_mod (mp_int * a, mp_int * b, mp_int * c)
+int mp_mod(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  t;
-  int     res;
+   mp_int  t;
+   int     res;
 
-  if ((res = mp_init_size (&t, b->used)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_size(&t, b->used)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_div (a, b, NULL, &t)) != MP_OKAY) {
-    mp_clear (&t);
-    return res;
-  }
+   if ((res = mp_div(a, b, NULL, &t)) != MP_OKAY) {
+      mp_clear(&t);
+      return res;
+   }
 
-  if ((mp_iszero(&t) != MP_NO) || (t.sign == b->sign)) {
-    res = MP_OKAY;
-    mp_exch (&t, c);
-  } else {
-    res = mp_add (b, &t, c);
-  }
+   if ((mp_iszero(&t) != MP_NO) || (t.sign == b->sign)) {
+      res = MP_OKAY;
+      mp_exch(&t, c);
+   } else {
+      res = mp_add(b, &t, c);
+   }
 
-  mp_clear (&t);
-  return res;
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mod_2d.c b/libtommath/bn_mp_mod_2d.c
index 2bb86da..59270b9 100644
--- a/libtommath/bn_mp_mod_2d.c
+++ b/libtommath/bn_mp_mod_2d.c
@@ -16,37 +16,36 @@
  */
 
 /* calc a value mod 2**b */
-int
-mp_mod_2d (mp_int * a, int b, mp_int * c)
+int mp_mod_2d(mp_int *a, int b, mp_int *c)
 {
-  int     x, res;
+   int     x, res;
 
-  /* if b is <= 0 then zero the int */
-  if (b <= 0) {
-    mp_zero (c);
-    return MP_OKAY;
-  }
+   /* if b is <= 0 then zero the int */
+   if (b <= 0) {
+      mp_zero(c);
+      return MP_OKAY;
+   }
 
-  /* if the modulus is larger than the value than return */
-  if (b >= (int) (a->used * DIGIT_BIT)) {
-    res = mp_copy (a, c);
-    return res;
-  }
+   /* if the modulus is larger than the value than return */
+   if (b >= (int)(a->used * DIGIT_BIT)) {
+      res = mp_copy(a, c);
+      return res;
+   }
 
-  /* copy */
-  if ((res = mp_copy (a, c)) != MP_OKAY) {
-    return res;
-  }
+   /* copy */
+   if ((res = mp_copy(a, c)) != MP_OKAY) {
+      return res;
+   }
 
-  /* zero digits above the last digit of the modulus */
-  for (x = (b / DIGIT_BIT) + (((b % DIGIT_BIT) == 0) ? 0 : 1); x < c->used; x++) {
-    c->dp[x] = 0;
-  }
-  /* clear the digit that is not completely outside/inside the modulus */
-  c->dp[b / DIGIT_BIT] &=
-    (mp_digit) ((((mp_digit) 1) << (((mp_digit) b) % DIGIT_BIT)) - ((mp_digit) 1));
-  mp_clamp (c);
-  return MP_OKAY;
+   /* zero digits above the last digit of the modulus */
+   for (x = (b / DIGIT_BIT) + (((b % DIGIT_BIT) == 0) ? 0 : 1); x < c->used; x++) {
+      c->dp[x] = 0;
+   }
+   /* clear the digit that is not completely outside/inside the modulus */
+   c->dp[b / DIGIT_BIT] &=
+      (mp_digit)((((mp_digit) 1) << (((mp_digit) b) % DIGIT_BIT)) - ((mp_digit) 1));
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mod_d.c b/libtommath/bn_mp_mod_d.c
index bf2ccaa..ff77346 100644
--- a/libtommath/bn_mp_mod_d.c
+++ b/libtommath/bn_mp_mod_d.c
@@ -15,10 +15,9 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-int
-mp_mod_d (mp_int * a, mp_digit b, mp_digit * c)
+int mp_mod_d(mp_int *a, mp_digit b, mp_digit *c)
 {
-  return mp_div_d(a, b, NULL, c);
+   return mp_div_d(a, b, NULL, c);
 }
 #endif
 
diff --git a/libtommath/bn_mp_montgomery_calc_normalization.c b/libtommath/bn_mp_montgomery_calc_normalization.c
index 679a871..2d95140 100644
--- a/libtommath/bn_mp_montgomery_calc_normalization.c
+++ b/libtommath/bn_mp_montgomery_calc_normalization.c
@@ -21,36 +21,36 @@
  * The method is slightly modified to shift B unconditionally upto just under
  * the leading bit of b.  This saves alot of multiple precision shifting.
  */
-int mp_montgomery_calc_normalization (mp_int * a, mp_int * b)
+int mp_montgomery_calc_normalization(mp_int *a, mp_int *b)
 {
-  int     x, bits, res;
+   int     x, bits, res;
 
-  /* how many bits of last digit does b use */
-  bits = mp_count_bits (b) % DIGIT_BIT;
+   /* how many bits of last digit does b use */
+   bits = mp_count_bits(b) % DIGIT_BIT;
 
-  if (b->used > 1) {
-     if ((res = mp_2expt (a, ((b->used - 1) * DIGIT_BIT) + bits - 1)) != MP_OKAY) {
-        return res;
-     }
-  } else {
-     mp_set(a, 1);
-     bits = 1;
-  }
+   if (b->used > 1) {
+      if ((res = mp_2expt(a, ((b->used - 1) * DIGIT_BIT) + bits - 1)) != MP_OKAY) {
+         return res;
+      }
+   } else {
+      mp_set(a, 1);
+      bits = 1;
+   }
 
 
-  /* now compute C = A * B mod b */
-  for (x = bits - 1; x < (int)DIGIT_BIT; x++) {
-    if ((res = mp_mul_2 (a, a)) != MP_OKAY) {
-      return res;
-    }
-    if (mp_cmp_mag (a, b) != MP_LT) {
-      if ((res = s_mp_sub (a, b, a)) != MP_OKAY) {
-        return res;
+   /* now compute C = A * B mod b */
+   for (x = bits - 1; x < (int)DIGIT_BIT; x++) {
+      if ((res = mp_mul_2(a, a)) != MP_OKAY) {
+         return res;
+      }
+      if (mp_cmp_mag(a, b) != MP_LT) {
+         if ((res = s_mp_sub(a, b, a)) != MP_OKAY) {
+            return res;
+         }
       }
-    }
-  }
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_montgomery_reduce.c b/libtommath/bn_mp_montgomery_reduce.c
index 05e8bfa..1909997 100644
--- a/libtommath/bn_mp_montgomery_reduce.c
+++ b/libtommath/bn_mp_montgomery_reduce.c
@@ -16,100 +16,99 @@
  */
 
 /* computes xR**-1 == x (mod N) via Montgomery Reduction */
-int
-mp_montgomery_reduce (mp_int * x, mp_int * n, mp_digit rho)
+int mp_montgomery_reduce(mp_int *x, mp_int *n, mp_digit rho)
 {
-  int     ix, res, digs;
-  mp_digit mu;
-
-  /* can the fast reduction [comba] method be used?
-   *
-   * Note that unlike in mul you're safely allowed *less*
-   * than the available columns [255 per default] since carries
-   * are fixed up in the inner loop.
-   */
-  digs = (n->used * 2) + 1;
-  if ((digs < MP_WARRAY) &&
-      (n->used <
-      (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
-    return fast_mp_montgomery_reduce (x, n, rho);
-  }
-
-  /* grow the input as required */
-  if (x->alloc < digs) {
-    if ((res = mp_grow (x, digs)) != MP_OKAY) {
-      return res;
-    }
-  }
-  x->used = digs;
-
-  for (ix = 0; ix < n->used; ix++) {
-    /* mu = ai * rho mod b
-     *
-     * The value of rho must be precalculated via
-     * montgomery_setup() such that
-     * it equals -1/n0 mod b this allows the
-     * following inner loop to reduce the
-     * input one digit at a time
-     */
-    mu = (mp_digit) (((mp_word)x->dp[ix] * (mp_word)rho) & MP_MASK);
-
-    /* a = a + mu * m * b**i */
-    {
-      int iy;
-      mp_digit *tmpn, *tmpx, u;
-      mp_word r;
-
-      /* alias for digits of the modulus */
-      tmpn = n->dp;
-
-      /* alias for the digits of x [the input] */
-      tmpx = x->dp + ix;
-
-      /* set the carry to zero */
-      u = 0;
-
-      /* Multiply and add in place */
-      for (iy = 0; iy < n->used; iy++) {
-        /* compute product and sum */
-        r       = ((mp_word)mu * (mp_word)*tmpn++) +
-                   (mp_word) u + (mp_word) *tmpx;
-
-        /* get carry */
-        u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
-
-        /* fix digit */
-        *tmpx++ = (mp_digit)(r & ((mp_word) MP_MASK));
+   int     ix, res, digs;
+   mp_digit mu;
+
+   /* can the fast reduction [comba] method be used?
+    *
+    * Note that unlike in mul you're safely allowed *less*
+    * than the available columns [255 per default] since carries
+    * are fixed up in the inner loop.
+    */
+   digs = (n->used * 2) + 1;
+   if ((digs < MP_WARRAY) &&
+       (n->used <
+        (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
+      return fast_mp_montgomery_reduce(x, n, rho);
+   }
+
+   /* grow the input as required */
+   if (x->alloc < digs) {
+      if ((res = mp_grow(x, digs)) != MP_OKAY) {
+         return res;
       }
-      /* At this point the ix'th digit of x should be zero */
+   }
+   x->used = digs;
+
+   for (ix = 0; ix < n->used; ix++) {
+      /* mu = ai * rho mod b
+       *
+       * The value of rho must be precalculated via
+       * montgomery_setup() such that
+       * it equals -1/n0 mod b this allows the
+       * following inner loop to reduce the
+       * input one digit at a time
+       */
+      mu = (mp_digit)(((mp_word)x->dp[ix] * (mp_word)rho) & MP_MASK);
+
+      /* a = a + mu * m * b**i */
+      {
+         int iy;
+         mp_digit *tmpn, *tmpx, u;
+         mp_word r;
+
+         /* alias for digits of the modulus */
+         tmpn = n->dp;
+
+         /* alias for the digits of x [the input] */
+         tmpx = x->dp + ix;
+
+         /* set the carry to zero */
+         u = 0;
+
+         /* Multiply and add in place */
+         for (iy = 0; iy < n->used; iy++) {
+            /* compute product and sum */
+            r       = ((mp_word)mu * (mp_word)*tmpn++) +
+                      (mp_word) u + (mp_word) *tmpx;
+
+            /* get carry */
+            u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+
+            /* fix digit */
+            *tmpx++ = (mp_digit)(r & ((mp_word) MP_MASK));
+         }
+         /* At this point the ix'th digit of x should be zero */
+
+
+         /* propagate carries upwards as required*/
+         while (u != 0) {
+            *tmpx   += u;
+            u        = *tmpx >> DIGIT_BIT;
+            *tmpx++ &= MP_MASK;
+         }
+      }
+   }
 
+   /* at this point the n.used'th least
+    * significant digits of x are all zero
+    * which means we can shift x to the
+    * right by n.used digits and the
+    * residue is unchanged.
+    */
 
-      /* propagate carries upwards as required*/
-      while (u != 0) {
-        *tmpx   += u;
-        u        = *tmpx >> DIGIT_BIT;
-        *tmpx++ &= MP_MASK;
-      }
-    }
-  }
-
-  /* at this point the n.used'th least
-   * significant digits of x are all zero
-   * which means we can shift x to the
-   * right by n.used digits and the
-   * residue is unchanged.
-   */
-
-  /* x = x/b**n.used */
-  mp_clamp(x);
-  mp_rshd (x, n->used);
-
-  /* if x >= n then x = x - n */
-  if (mp_cmp_mag (x, n) != MP_LT) {
-    return s_mp_sub (x, n, x);
-  }
-
-  return MP_OKAY;
+   /* x = x/b**n.used */
+   mp_clamp(x);
+   mp_rshd(x, n->used);
+
+   /* if x >= n then x = x - n */
+   if (mp_cmp_mag(x, n) != MP_LT) {
+      return s_mp_sub(x, n, x);
+   }
+
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_montgomery_setup.c b/libtommath/bn_mp_montgomery_setup.c
index 1c17445..46f560d 100644
--- a/libtommath/bn_mp_montgomery_setup.c
+++ b/libtommath/bn_mp_montgomery_setup.c
@@ -16,41 +16,40 @@
  */
 
 /* setups the montgomery reduction stuff */
-int
-mp_montgomery_setup (mp_int * n, mp_digit * rho)
+int mp_montgomery_setup(mp_int *n, mp_digit *rho)
 {
-  mp_digit x, b;
+   mp_digit x, b;
 
-/* fast inversion mod 2**k
- *
- * Based on the fact that
- *
- * XA = 1 (mod 2**n)  =>  (X(2-XA)) A = 1 (mod 2**2n)
- *                    =>  2*X*A - X*X*A*A = 1
- *                    =>  2*(1) - (1)     = 1
- */
-  b = n->dp[0];
+   /* fast inversion mod 2**k
+    *
+    * Based on the fact that
+    *
+    * XA = 1 (mod 2**n)  =>  (X(2-XA)) A = 1 (mod 2**2n)
+    *                    =>  2*X*A - X*X*A*A = 1
+    *                    =>  2*(1) - (1)     = 1
+    */
+   b = n->dp[0];
 
-  if ((b & 1) == 0) {
-    return MP_VAL;
-  }
+   if ((b & 1) == 0) {
+      return MP_VAL;
+   }
 
-  x = (((b + 2) & 4) << 1) + b; /* here x*a==1 mod 2**4 */
-  x *= 2 - (b * x);             /* here x*a==1 mod 2**8 */
+   x = (((b + 2) & 4) << 1) + b; /* here x*a==1 mod 2**4 */
+   x *= 2 - (b * x);             /* here x*a==1 mod 2**8 */
 #if !defined(MP_8BIT)
-  x *= 2 - (b * x);             /* here x*a==1 mod 2**16 */
+   x *= 2 - (b * x);             /* here x*a==1 mod 2**16 */
 #endif
 #if defined(MP_64BIT) || !(defined(MP_8BIT) || defined(MP_16BIT))
-  x *= 2 - (b * x);             /* here x*a==1 mod 2**32 */
+   x *= 2 - (b * x);             /* here x*a==1 mod 2**32 */
 #endif
 #ifdef MP_64BIT
-  x *= 2 - (b * x);             /* here x*a==1 mod 2**64 */
+   x *= 2 - (b * x);             /* here x*a==1 mod 2**64 */
 #endif
 
-  /* rho = -1/m mod b */
-  *rho = (mp_digit)(((mp_word)1 << ((mp_word) DIGIT_BIT)) - x) & MP_MASK;
+   /* rho = -1/m mod b */
+   *rho = (mp_digit)(((mp_word)1 << ((mp_word) DIGIT_BIT)) - x) & MP_MASK;
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mul.c b/libtommath/bn_mp_mul.c
index cc3b9c8..315a520 100644
--- a/libtommath/bn_mp_mul.c
+++ b/libtommath/bn_mp_mul.c
@@ -16,49 +16,49 @@
  */
 
 /* high level multiplication (handles sign) */
-int mp_mul (mp_int * a, mp_int * b, mp_int * c)
+int mp_mul(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res, neg;
-  neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
+   int     res, neg;
+   neg = (a->sign == b->sign) ? MP_ZPOS : MP_NEG;
 
-  /* use Toom-Cook? */
+   /* use Toom-Cook? */
 #ifdef BN_MP_TOOM_MUL_C
-  if (MIN (a->used, b->used) >= TOOM_MUL_CUTOFF) {
-    res = mp_toom_mul(a, b, c);
-  } else 
+   if (MIN(a->used, b->used) >= TOOM_MUL_CUTOFF) {
+      res = mp_toom_mul(a, b, c);
+   } 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 
+      /* use Karatsuba? */
+      if (MIN(a->used, b->used) >= KARATSUBA_MUL_CUTOFF) {
+         res = mp_karatsuba_mul(a, b, c);
+      } 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 
-     * digits won't affect carry propagation
-     */
-    int     digs = a->used + b->used + 1;
+      {
+         /* 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
+          * digits won't affect carry propagation
+          */
+         int     digs = a->used + b->used + 1;
 
 #ifdef BN_FAST_S_MP_MUL_DIGS_C
-    if ((digs < MP_WARRAY) &&
-        (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 
+         if ((digs < MP_WARRAY) &&
+             (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
 #endif
-    {
+         {
 #ifdef BN_S_MP_MUL_DIGS_C
-      res = s_mp_mul (a, b, c); /* uses s_mp_mul_digs */
+            res = s_mp_mul(a, b, c); /* uses s_mp_mul_digs */
 #else
-      res = MP_VAL;
+            res = MP_VAL;
 #endif
-    }
-  }
-  c->sign = (c->used > 0) ? neg : MP_ZPOS;
-  return res;
+         }
+      }
+   c->sign = (c->used > 0) ? neg : MP_ZPOS;
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mul_2.c b/libtommath/bn_mp_mul_2.c
index d22fd89..33bdc4c 100644
--- a/libtommath/bn_mp_mul_2.c
+++ b/libtommath/bn_mp_mul_2.c
@@ -16,64 +16,64 @@
  */
 
 /* b = a*2 */
-int mp_mul_2(mp_int * a, mp_int * b)
+int mp_mul_2(mp_int *a, mp_int *b)
 {
-  int     x, res, oldused;
+   int     x, res, oldused;
 
-  /* grow to accomodate result */
-  if (b->alloc < (a->used + 1)) {
-    if ((res = mp_grow (b, a->used + 1)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* grow to accomodate result */
+   if (b->alloc < (a->used + 1)) {
+      if ((res = mp_grow(b, a->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  oldused = b->used;
-  b->used = a->used;
+   oldused = b->used;
+   b->used = a->used;
 
-  {
-    mp_digit r, rr, *tmpa, *tmpb;
+   {
+      mp_digit r, rr, *tmpa, *tmpb;
 
-    /* alias for source */
-    tmpa = a->dp;
-    
-    /* alias for dest */
-    tmpb = b->dp;
+      /* alias for source */
+      tmpa = a->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 
-       */
-      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 
-       */
-      r = rr;
-    }
+      /* 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
+          */
+         rr = *tmpa >> ((mp_digit)(DIGIT_BIT - 1));
 
-    /* new leading digit? */
-    if (r != 0) {
-      /* add a MSB which is always 1 at this point */
-      *tmpb = 1;
-      ++(b->used);
-    }
+         /* now shift up this digit, add in the carry [from the previous] */
+         *tmpb++ = ((*tmpa++ << ((mp_digit)1)) | r) & MP_MASK;
 
-    /* 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++) {
-      *tmpb++ = 0;
-    }
-  }
-  b->sign = a->sign;
-  return MP_OKAY;
+         /* copy the carry that would be from the source
+          * digit into the next iteration
+          */
+         r = rr;
+      }
+
+      /* new leading digit? */
+      if (r != 0) {
+         /* add a MSB which is always 1 at this point */
+         *tmpb = 1;
+         ++(b->used);
+      }
+
+      /* 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++) {
+         *tmpb++ = 0;
+      }
+   }
+   b->sign = a->sign;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mul_2d.c b/libtommath/bn_mp_mul_2d.c
index c00fd7e..b3d92a9 100644
--- a/libtommath/bn_mp_mul_2d.c
+++ b/libtommath/bn_mp_mul_2d.c
@@ -16,67 +16,67 @@
  */
 
 /* shift left by a certain bit count */
-int mp_mul_2d (mp_int * a, int b, mp_int * c)
+int mp_mul_2d(mp_int *a, int b, mp_int *c)
 {
-  mp_digit d;
-  int      res;
+   mp_digit d;
+   int      res;
 
-  /* copy */
-  if (a != c) {
-     if ((res = mp_copy (a, c)) != MP_OKAY) {
-       return res;
-     }
-  }
+   /* copy */
+   if (a != c) {
+      if ((res = mp_copy(a, c)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  if (c->alloc < (int)(c->used + (b / DIGIT_BIT) + 1)) {
-     if ((res = mp_grow (c, c->used + (b / DIGIT_BIT) + 1)) != MP_OKAY) {
-       return res;
-     }
-  }
+   if (c->alloc < (int)(c->used + (b / DIGIT_BIT) + 1)) {
+      if ((res = mp_grow(c, c->used + (b / DIGIT_BIT) + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* shift by as many digits in the bit count */
-  if (b >= (int)DIGIT_BIT) {
-    if ((res = mp_lshd (c, b / DIGIT_BIT)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* shift by as many digits in the bit count */
+   if (b >= (int)DIGIT_BIT) {
+      if ((res = mp_lshd(c, b / DIGIT_BIT)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* shift any bit count < DIGIT_BIT */
-  d = (mp_digit) (b % DIGIT_BIT);
-  if (d != 0) {
-    mp_digit *tmpc, shift, mask, r, rr;
-    int x;
+   /* shift any bit count < DIGIT_BIT */
+   d = (mp_digit)(b % DIGIT_BIT);
+   if (d != 0) {
+      mp_digit *tmpc, shift, mask, r, rr;
+      int x;
 
-    /* bitmask for carries */
-    mask = (((mp_digit)1) << d) - 1;
+      /* bitmask for carries */
+      mask = (((mp_digit)1) << d) - 1;
 
-    /* shift for msbs */
-    shift = DIGIT_BIT - d;
+      /* shift for msbs */
+      shift = DIGIT_BIT - d;
 
-    /* alias */
-    tmpc = c->dp;
+      /* alias */
+      tmpc = c->dp;
 
-    /* carry */
-    r    = 0;
-    for (x = 0; x < c->used; x++) {
-      /* get the higher bits of the current word */
-      rr = (*tmpc >> shift) & mask;
+      /* carry */
+      r    = 0;
+      for (x = 0; x < c->used; x++) {
+         /* get the higher bits of the current word */
+         rr = (*tmpc >> shift) & mask;
 
-      /* shift the current word and OR in the carry */
-      *tmpc = ((*tmpc << d) | r) & MP_MASK;
-      ++tmpc;
+         /* shift the current word and OR in the carry */
+         *tmpc = ((*tmpc << d) | r) & MP_MASK;
+         ++tmpc;
 
-      /* set the carry to the carry bits of the current word */
-      r = rr;
-    }
-    
-    /* set final carry */
-    if (r != 0) {
-       c->dp[(c->used)++] = r;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+         /* set the carry to the carry bits of the current word */
+         r = rr;
+      }
+
+      /* set final carry */
+      if (r != 0) {
+         c->dp[(c->used)++] = r;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mul_d.c b/libtommath/bn_mp_mul_d.c
index 6954ed3..1aa448c 100644
--- a/libtommath/bn_mp_mul_d.c
+++ b/libtommath/bn_mp_mul_d.c
@@ -16,61 +16,60 @@
  */
 
 /* multiply by a digit */
-int
-mp_mul_d (mp_int * a, mp_digit b, mp_int * c)
+int mp_mul_d(mp_int *a, mp_digit b, mp_int *c)
 {
-  mp_digit u, *tmpa, *tmpc;
-  mp_word  r;
-  int      ix, res, olduse;
+   mp_digit u, *tmpa, *tmpc;
+   mp_word  r;
+   int      ix, res, olduse;
 
-  /* make sure c is big enough to hold a*b */
-  if (c->alloc < (a->used + 1)) {
-    if ((res = mp_grow (c, a->used + 1)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* make sure c is big enough to hold a*b */
+   if (c->alloc < (a->used + 1)) {
+      if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* get the original destinations used count */
-  olduse = c->used;
+   /* get the original destinations used count */
+   olduse = c->used;
 
-  /* set the sign */
-  c->sign = a->sign;
+   /* set the sign */
+   c->sign = a->sign;
 
-  /* alias for a->dp [source] */
-  tmpa = a->dp;
+   /* alias for a->dp [source] */
+   tmpa = a->dp;
 
-  /* alias for c->dp [dest] */
-  tmpc = c->dp;
+   /* alias for c->dp [dest] */
+   tmpc = c->dp;
 
-  /* zero carry */
-  u = 0;
+   /* zero carry */
+   u = 0;
 
-  /* compute columns */
-  for (ix = 0; ix < a->used; ix++) {
-    /* compute product and carry sum for this term */
-    r       = (mp_word)u + ((mp_word)*tmpa++ * (mp_word)b);
+   /* compute columns */
+   for (ix = 0; ix < a->used; ix++) {
+      /* compute product and carry sum for this term */
+      r       = (mp_word)u + ((mp_word)*tmpa++ * (mp_word)b);
 
-    /* mask off higher bits to get a single digit */
-    *tmpc++ = (mp_digit) (r & ((mp_word) MP_MASK));
+      /* mask off higher bits to get a single digit */
+      *tmpc++ = (mp_digit)(r & ((mp_word)MP_MASK));
 
-    /* send carry into next iteration */
-    u       = (mp_digit) (r >> ((mp_word) DIGIT_BIT));
-  }
+      /* send carry into next iteration */
+      u       = (mp_digit)(r >> ((mp_word)DIGIT_BIT));
+   }
 
-  /* store final carry [if any] and increment ix offset  */
-  *tmpc++ = u;
-  ++ix;
+   /* store final carry [if any] and increment ix offset  */
+   *tmpc++ = u;
+   ++ix;
 
-  /* now zero digits above the top */
-  while (ix++ < olduse) {
-     *tmpc++ = 0;
-  }
+   /* now zero digits above the top */
+   while (ix++ < olduse) {
+      *tmpc++ = 0;
+   }
 
-  /* set used count */
-  c->used = a->used + 1;
-  mp_clamp(c);
+   /* set used count */
+   c->used = a->used + 1;
+   mp_clamp(c);
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_mulmod.c b/libtommath/bn_mp_mulmod.c
index d7a4d3c..b1e6a33 100644
--- a/libtommath/bn_mp_mulmod.c
+++ b/libtommath/bn_mp_mulmod.c
@@ -16,22 +16,22 @@
  */
 
 /* d = a * b (mod c) */
-int mp_mulmod (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+int mp_mulmod(mp_int *a, mp_int *b, mp_int *c, mp_int *d)
 {
-  int     res;
-  mp_int  t;
+   int     res;
+   mp_int  t;
 
-  if ((res = mp_init_size (&t, c->used)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_size(&t, c->used)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_mul (a, b, &t)) != MP_OKAY) {
-    mp_clear (&t);
-    return res;
-  }
-  res = mp_mod (&t, c, d);
-  mp_clear (&t);
-  return res;
+   if ((res = mp_mul(a, b, &t)) != MP_OKAY) {
+      mp_clear(&t);
+      return res;
+   }
+   res = mp_mod(&t, c, d);
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_n_root.c b/libtommath/bn_mp_n_root.c
index f717f17..8211c0a 100644
--- a/libtommath/bn_mp_n_root.c
+++ b/libtommath/bn_mp_n_root.c
@@ -18,9 +18,9 @@
 /* wrapper function for mp_n_root_ex()
  * computes c = (a)**(1/b) such that (c)**b <= a and (c+1)**b > a
  */
-int mp_n_root (mp_int * a, mp_digit b, mp_int * c)
+int mp_n_root(mp_int *a, mp_digit b, mp_int *c)
 {
-  return mp_n_root_ex(a, b, c, 0);
+   return mp_n_root_ex(a, b, c, 0);
 }
 
 #endif
diff --git a/libtommath/bn_mp_n_root_ex.c b/libtommath/bn_mp_n_root_ex.c
index 079b4f3..9546745 100644
--- a/libtommath/bn_mp_n_root_ex.c
+++ b/libtommath/bn_mp_n_root_ex.c
@@ -25,105 +25,108 @@
  * each step involves a fair bit.  This is not meant to
  * find huge roots [square and cube, etc].
  */
-int mp_n_root_ex (mp_int * a, mp_digit b, mp_int * c, int fast)
+int mp_n_root_ex(mp_int *a, mp_digit b, mp_int *c, int fast)
 {
-  mp_int  t1, t2, t3;
-  int     res, neg;
-
-  /* input must be positive if b is even */
-  if (((b & 1) == 0) && (a->sign == MP_NEG)) {
-    return MP_VAL;
-  }
-
-  if ((res = mp_init (&t1)) != MP_OKAY) {
-    return res;
-  }
-
-  if ((res = mp_init (&t2)) != MP_OKAY) {
-    goto LBL_T1;
-  }
-
-  if ((res = mp_init (&t3)) != MP_OKAY) {
-    goto LBL_T2;
-  }
-
-  /* if a is negative fudge the sign but keep track */
-  neg     = a->sign;
-  a->sign = MP_ZPOS;
-
-  /* t2 = 2 */
-  mp_set (&t2, 2);
-
-  do {
-    /* t1 = t2 */
-    if ((res = mp_copy (&t2, &t1)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    /* t2 = t1 - ((t1**b - a) / (b * t1**(b-1))) */
-
-    /* t3 = t1**(b-1) */
-    if ((res = mp_expt_d_ex (&t1, b - 1, &t3, fast)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    /* numerator */
-    /* t2 = t1**b */
-    if ((res = mp_mul (&t3, &t1, &t2)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    /* t2 = t1**b - a */
-    if ((res = mp_sub (&t2, a, &t2)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    /* denominator */
-    /* t3 = t1**(b-1) * b  */
-    if ((res = mp_mul_d (&t3, b, &t3)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    /* t3 = (t1**b - a)/(b * t1**(b-1)) */
-    if ((res = mp_div (&t2, &t3, &t3, NULL)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    if ((res = mp_sub (&t1, &t3, &t2)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-  }  while (mp_cmp (&t1, &t2) != MP_EQ);
-
-  /* result can be off by a few so check */
-  for (;;) {
-    if ((res = mp_expt_d_ex (&t1, b, &t2, fast)) != MP_OKAY) {
-      goto LBL_T3;
-    }
-
-    if (mp_cmp (&t2, a) == MP_GT) {
-      if ((res = mp_sub_d (&t1, 1, &t1)) != MP_OKAY) {
+   mp_int  t1, t2, t3;
+   int     res, neg;
+
+   /* input must be positive if b is even */
+   if (((b & 1) == 0) && (a->sign == MP_NEG)) {
+      return MP_VAL;
+   }
+
+   if ((res = mp_init(&t1)) != MP_OKAY) {
+      return res;
+   }
+
+   if ((res = mp_init(&t2)) != MP_OKAY) {
+      goto LBL_T1;
+   }
+
+   if ((res = mp_init(&t3)) != MP_OKAY) {
+      goto LBL_T2;
+   }
+
+   /* if a is negative fudge the sign but keep track */
+   neg     = a->sign;
+   a->sign = MP_ZPOS;
+
+   /* t2 = 2 */
+   mp_set(&t2, 2);
+
+   do {
+      /* t1 = t2 */
+      if ((res = mp_copy(&t2, &t1)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      /* t2 = t1 - ((t1**b - a) / (b * t1**(b-1))) */
+
+      /* t3 = t1**(b-1) */
+      if ((res = mp_expt_d_ex(&t1, b - 1, &t3, fast)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      /* numerator */
+      /* t2 = t1**b */
+      if ((res = mp_mul(&t3, &t1, &t2)) != MP_OKAY) {
          goto LBL_T3;
       }
-    } else {
-      break;
-    }
-  }
 
-  /* reset the sign of a first */
-  a->sign = neg;
+      /* t2 = t1**b - a */
+      if ((res = mp_sub(&t2, a, &t2)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      /* denominator */
+      /* t3 = t1**(b-1) * b  */
+      if ((res = mp_mul_d(&t3, b, &t3)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      /* t3 = (t1**b - a)/(b * t1**(b-1)) */
+      if ((res = mp_div(&t2, &t3, &t3, NULL)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      if ((res = mp_sub(&t1, &t3, &t2)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+   }  while (mp_cmp(&t1, &t2) != MP_EQ);
+
+   /* result can be off by a few so check */
+   for (;;) {
+      if ((res = mp_expt_d_ex(&t1, b, &t2, fast)) != MP_OKAY) {
+         goto LBL_T3;
+      }
+
+      if (mp_cmp(&t2, a) == MP_GT) {
+         if ((res = mp_sub_d(&t1, 1, &t1)) != MP_OKAY) {
+            goto LBL_T3;
+         }
+      } else {
+         break;
+      }
+   }
+
+   /* reset the sign of a first */
+   a->sign = neg;
 
-  /* set the result */
-  mp_exch (&t1, c);
+   /* set the result */
+   mp_exch(&t1, c);
 
-  /* set the sign of the result */
-  c->sign = neg;
+   /* set the sign of the result */
+   c->sign = neg;
 
-  res = MP_OKAY;
+   res = MP_OKAY;
 
-LBL_T3:mp_clear (&t3);
-LBL_T2:mp_clear (&t2);
-LBL_T1:mp_clear (&t1);
-  return res;
+LBL_T3:
+   mp_clear(&t3);
+LBL_T2:
+   mp_clear(&t2);
+LBL_T1:
+   mp_clear(&t1);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_neg.c b/libtommath/bn_mp_neg.c
index d03e92e..b30d044 100644
--- a/libtommath/bn_mp_neg.c
+++ b/libtommath/bn_mp_neg.c
@@ -16,22 +16,22 @@
  */
 
 /* b = -a */
-int mp_neg (mp_int * a, mp_int * b)
+int mp_neg(mp_int *a, mp_int *b)
 {
-  int     res;
-  if (a != b) {
-     if ((res = mp_copy (a, b)) != MP_OKAY) {
-        return res;
-     }
-  }
+   int     res;
+   if (a != b) {
+      if ((res = mp_copy(a, b)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  if (mp_iszero(b) != MP_YES) {
-     b->sign = (a->sign == MP_ZPOS) ? MP_NEG : MP_ZPOS;
-  } else {
-     b->sign = MP_ZPOS;
-  }
+   if (mp_iszero(b) != MP_YES) {
+      b->sign = (a->sign == MP_ZPOS) ? MP_NEG : MP_ZPOS;
+   } else {
+      b->sign = MP_ZPOS;
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_or.c b/libtommath/bn_mp_or.c
index b9775ce..2318c79 100644
--- a/libtommath/bn_mp_or.c
+++ b/libtommath/bn_mp_or.c
@@ -16,32 +16,32 @@
  */
 
 /* OR two ints together */
-int mp_or (mp_int * a, mp_int * b, mp_int * c)
+int mp_or(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res, ix, px;
-  mp_int  t, *x;
+   int     res, ix, px;
+   mp_int  t, *x;
 
-  if (a->used > b->used) {
-    if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-      return res;
-    }
-    px = b->used;
-    x = b;
-  } else {
-    if ((res = mp_init_copy (&t, b)) != MP_OKAY) {
-      return res;
-    }
-    px = a->used;
-    x = a;
-  }
+   if (a->used > b->used) {
+      if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+         return res;
+      }
+      px = b->used;
+      x = b;
+   } else {
+      if ((res = mp_init_copy(&t, b)) != MP_OKAY) {
+         return res;
+      }
+      px = a->used;
+      x = a;
+   }
 
-  for (ix = 0; ix < px; ix++) {
-    t.dp[ix] |= x->dp[ix];
-  }
-  mp_clamp (&t);
-  mp_exch (c, &t);
-  mp_clear (&t);
-  return MP_OKAY;
+   for (ix = 0; ix < px; ix++) {
+      t.dp[ix] |= x->dp[ix];
+   }
+   mp_clamp(&t);
+   mp_exch(c, &t);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_prime_fermat.c b/libtommath/bn_mp_prime_fermat.c
index d99feeb..37ad6ec 100644
--- a/libtommath/bn_mp_prime_fermat.c
+++ b/libtommath/bn_mp_prime_fermat.c
@@ -16,44 +16,45 @@
  */
 
 /* performs one Fermat test.
- * 
+ *
  * If "a" were prime then b**a == b (mod a) since the order of
  * the multiplicative sub-group would be phi(a) = a-1.  That means
  * it would be the same as b**(a mod (a-1)) == b**1 == b (mod a).
  *
  * Sets result to 1 if the congruence holds, or zero otherwise.
  */
-int mp_prime_fermat (mp_int * a, mp_int * b, int *result)
+int mp_prime_fermat(mp_int *a, mp_int *b, int *result)
 {
-  mp_int  t;
-  int     err;
-
-  /* default to composite  */
-  *result = MP_NO;
-
-  /* ensure b > 1 */
-  if (mp_cmp_d(b, 1) != MP_GT) {
-     return MP_VAL;
-  }
-
-  /* init t */
-  if ((err = mp_init (&t)) != MP_OKAY) {
-    return err;
-  }
-
-  /* compute t = b**a mod a */
-  if ((err = mp_exptmod (b, a, a, &t)) != MP_OKAY) {
-    goto LBL_T;
-  }
-
-  /* is it equal to b? */
-  if (mp_cmp (&t, b) == MP_EQ) {
-    *result = MP_YES;
-  }
-
-  err = MP_OKAY;
-LBL_T:mp_clear (&t);
-  return err;
+   mp_int  t;
+   int     err;
+
+   /* default to composite  */
+   *result = MP_NO;
+
+   /* ensure b > 1 */
+   if (mp_cmp_d(b, 1) != MP_GT) {
+      return MP_VAL;
+   }
+
+   /* init t */
+   if ((err = mp_init(&t)) != MP_OKAY) {
+      return err;
+   }
+
+   /* compute t = b**a mod a */
+   if ((err = mp_exptmod(b, a, a, &t)) != MP_OKAY) {
+      goto LBL_T;
+   }
+
+   /* is it equal to b? */
+   if (mp_cmp(&t, b) == MP_EQ) {
+      *result = MP_YES;
+   }
+
+   err = MP_OKAY;
+LBL_T:
+   mp_clear(&t);
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_prime_is_divisible.c b/libtommath/bn_mp_prime_is_divisible.c
index eea4a27..92af330 100644
--- a/libtommath/bn_mp_prime_is_divisible.c
+++ b/libtommath/bn_mp_prime_is_divisible.c
@@ -15,33 +15,33 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* determines if an integers is divisible by one 
+/* determines if an integers is divisible by one
  * of the first PRIME_SIZE primes or not
  *
  * sets result to 0 if not, 1 if yes
  */
-int mp_prime_is_divisible (mp_int * a, int *result)
+int mp_prime_is_divisible(mp_int *a, int *result)
 {
-  int     err, ix;
-  mp_digit res;
+   int     err, ix;
+   mp_digit res;
 
-  /* default to not */
-  *result = MP_NO;
+   /* default to not */
+   *result = MP_NO;
 
-  for (ix = 0; ix < PRIME_SIZE; ix++) {
-    /* what is a mod LBL_prime_tab[ix] */
-    if ((err = mp_mod_d (a, ltm_prime_tab[ix], &res)) != MP_OKAY) {
-      return err;
-    }
+   for (ix = 0; ix < PRIME_SIZE; ix++) {
+      /* what is a mod LBL_prime_tab[ix] */
+      if ((err = mp_mod_d(a, ltm_prime_tab[ix], &res)) != MP_OKAY) {
+         return err;
+      }
 
-    /* is the residue zero? */
-    if (res == 0) {
-      *result = MP_YES;
-      return MP_OKAY;
-    }
-  }
+      /* is the residue zero? */
+      if (res == 0) {
+         *result = MP_YES;
+         return MP_OKAY;
+      }
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_prime_is_prime.c b/libtommath/bn_mp_prime_is_prime.c
index 3eda4fd..20a7d1f 100644
--- a/libtommath/bn_mp_prime_is_prime.c
+++ b/libtommath/bn_mp_prime_is_prime.c
@@ -22,59 +22,60 @@
  *
  * Sets result to 1 if probably prime, 0 otherwise
  */
-int mp_prime_is_prime (mp_int * a, int t, int *result)
+int mp_prime_is_prime(mp_int *a, int t, int *result)
 {
-  mp_int  b;
-  int     ix, err, res;
+   mp_int  b;
+   int     ix, err, res;
 
-  /* default to no */
-  *result = MP_NO;
+   /* default to no */
+   *result = MP_NO;
 
-  /* valid value of t? */
-  if ((t <= 0) || (t > PRIME_SIZE)) {
-    return MP_VAL;
-  }
+   /* valid value of t? */
+   if ((t <= 0) || (t > PRIME_SIZE)) {
+      return MP_VAL;
+   }
 
-  /* is the input equal to one of the primes in the table? */
-  for (ix = 0; ix < PRIME_SIZE; ix++) {
+   /* is the input equal to one of the primes in the table? */
+   for (ix = 0; ix < PRIME_SIZE; ix++) {
       if (mp_cmp_d(a, ltm_prime_tab[ix]) == MP_EQ) {
          *result = 1;
          return MP_OKAY;
       }
-  }
+   }
 
-  /* first perform trial division */
-  if ((err = mp_prime_is_divisible (a, &res)) != MP_OKAY) {
-    return err;
-  }
+   /* first perform trial division */
+   if ((err = mp_prime_is_divisible(a, &res)) != MP_OKAY) {
+      return err;
+   }
 
-  /* return if it was trivially divisible */
-  if (res == MP_YES) {
-    return MP_OKAY;
-  }
+   /* return if it was trivially divisible */
+   if (res == MP_YES) {
+      return MP_OKAY;
+   }
 
-  /* now perform the miller-rabin rounds */
-  if ((err = mp_init (&b)) != MP_OKAY) {
-    return err;
-  }
+   /* now perform the miller-rabin rounds */
+   if ((err = mp_init(&b)) != MP_OKAY) {
+      return err;
+   }
 
-  for (ix = 0; ix < t; ix++) {
-    /* set the prime */
-    mp_set (&b, ltm_prime_tab[ix]);
+   for (ix = 0; ix < t; ix++) {
+      /* set the prime */
+      mp_set(&b, ltm_prime_tab[ix]);
 
-    if ((err = mp_prime_miller_rabin (a, &b, &res)) != MP_OKAY) {
-      goto LBL_B;
-    }
+      if ((err = mp_prime_miller_rabin(a, &b, &res)) != MP_OKAY) {
+         goto LBL_B;
+      }
 
-    if (res == MP_NO) {
-      goto LBL_B;
-    }
-  }
+      if (res == MP_NO) {
+         goto LBL_B;
+      }
+   }
 
-  /* passed the test */
-  *result = MP_YES;
-LBL_B:mp_clear (&b);
-  return err;
+   /* passed the test */
+   *result = MP_YES;
+LBL_B:
+   mp_clear(&b);
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_prime_miller_rabin.c b/libtommath/bn_mp_prime_miller_rabin.c
index 7de0634..917dc01 100644
--- a/libtommath/bn_mp_prime_miller_rabin.c
+++ b/libtommath/bn_mp_prime_miller_rabin.c
@@ -15,86 +15,89 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* Miller-Rabin test of "a" to the base of "b" as described in 
+/* Miller-Rabin test of "a" to the base of "b" as described in
  * HAC pp. 139 Algorithm 4.24
  *
  * Sets result to 0 if definitely composite or 1 if probably prime.
- * Randomly the chance of error is no more than 1/4 and often 
+ * Randomly the chance of error is no more than 1/4 and often
  * very much lower.
  */
-int mp_prime_miller_rabin (mp_int * a, mp_int * b, int *result)
+int mp_prime_miller_rabin(mp_int *a, mp_int *b, int *result)
 {
-  mp_int  n1, y, r;
-  int     s, j, err;
+   mp_int  n1, y, r;
+   int     s, j, err;
 
-  /* default */
-  *result = MP_NO;
+   /* default */
+   *result = MP_NO;
 
-  /* ensure b > 1 */
-  if (mp_cmp_d(b, 1) != MP_GT) {
-     return MP_VAL;
-  }     
+   /* ensure b > 1 */
+   if (mp_cmp_d(b, 1) != MP_GT) {
+      return MP_VAL;
+   }
 
-  /* get n1 = a - 1 */
-  if ((err = mp_init_copy (&n1, a)) != MP_OKAY) {
-    return err;
-  }
-  if ((err = mp_sub_d (&n1, 1, &n1)) != MP_OKAY) {
-    goto LBL_N1;
-  }
+   /* get n1 = a - 1 */
+   if ((err = mp_init_copy(&n1, a)) != MP_OKAY) {
+      return err;
+   }
+   if ((err = mp_sub_d(&n1, 1, &n1)) != MP_OKAY) {
+      goto LBL_N1;
+   }
 
-  /* set 2**s * r = n1 */
-  if ((err = mp_init_copy (&r, &n1)) != MP_OKAY) {
-    goto LBL_N1;
-  }
+   /* set 2**s * r = n1 */
+   if ((err = mp_init_copy(&r, &n1)) != MP_OKAY) {
+      goto LBL_N1;
+   }
 
-  /* count the number of least significant bits
-   * which are zero
-   */
-  s = mp_cnt_lsb(&r);
+   /* count the number of least significant bits
+    * which are zero
+    */
+   s = mp_cnt_lsb(&r);
 
-  /* now divide n - 1 by 2**s */
-  if ((err = mp_div_2d (&r, s, &r, NULL)) != MP_OKAY) {
-    goto LBL_R;
-  }
+   /* now divide n - 1 by 2**s */
+   if ((err = mp_div_2d(&r, s, &r, NULL)) != MP_OKAY) {
+      goto LBL_R;
+   }
 
-  /* compute y = b**r mod a */
-  if ((err = mp_init (&y)) != MP_OKAY) {
-    goto LBL_R;
-  }
-  if ((err = mp_exptmod (b, &r, a, &y)) != MP_OKAY) {
-    goto LBL_Y;
-  }
+   /* compute y = b**r mod a */
+   if ((err = mp_init(&y)) != MP_OKAY) {
+      goto LBL_R;
+   }
+   if ((err = mp_exptmod(b, &r, a, &y)) != MP_OKAY) {
+      goto LBL_Y;
+   }
 
-  /* if y != 1 and y != n1 do */
-  if ((mp_cmp_d (&y, 1) != MP_EQ) && (mp_cmp (&y, &n1) != MP_EQ)) {
-    j = 1;
-    /* while j <= s-1 and y != n1 */
-    while ((j <= (s - 1)) && (mp_cmp (&y, &n1) != MP_EQ)) {
-      if ((err = mp_sqrmod (&y, a, &y)) != MP_OKAY) {
-         goto LBL_Y;
+   /* if y != 1 and y != n1 do */
+   if ((mp_cmp_d(&y, 1) != MP_EQ) && (mp_cmp(&y, &n1) != MP_EQ)) {
+      j = 1;
+      /* while j <= s-1 and y != n1 */
+      while ((j <= (s - 1)) && (mp_cmp(&y, &n1) != MP_EQ)) {
+         if ((err = mp_sqrmod(&y, a, &y)) != MP_OKAY) {
+            goto LBL_Y;
+         }
+
+         /* if y == 1 then composite */
+         if (mp_cmp_d(&y, 1) == MP_EQ) {
+            goto LBL_Y;
+         }
+
+         ++j;
       }
 
-      /* if y == 1 then composite */
-      if (mp_cmp_d (&y, 1) == MP_EQ) {
+      /* if y != n1 then composite */
+      if (mp_cmp(&y, &n1) != MP_EQ) {
          goto LBL_Y;
       }
+   }
 
-      ++j;
-    }
-
-    /* if y != n1 then composite */
-    if (mp_cmp (&y, &n1) != MP_EQ) {
-      goto LBL_Y;
-    }
-  }
-
-  /* probably prime now */
-  *result = MP_YES;
-LBL_Y:mp_clear (&y);
-LBL_R:mp_clear (&r);
-LBL_N1:mp_clear (&n1);
-  return err;
+   /* probably prime now */
+   *result = MP_YES;
+LBL_Y:
+   mp_clear(&y);
+LBL_R:
+   mp_clear(&r);
+LBL_N1:
+   mp_clear(&n1);
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_mp_prime_next_prime.c b/libtommath/bn_mp_prime_next_prime.c
index 7a32d9b..f383cbb 100644
--- a/libtommath/bn_mp_prime_next_prime.c
+++ b/libtommath/bn_mp_prime_next_prime.c
@@ -38,28 +38,28 @@ int mp_prime_next_prime(mp_int *a, int t, int bbs_style)
    if (mp_cmp_d(a, ltm_prime_tab[PRIME_SIZE-1]) == MP_LT) {
       /* find which prime it is bigger than */
       for (x = PRIME_SIZE - 2; x >= 0; x--) {
-          if (mp_cmp_d(a, ltm_prime_tab[x]) != MP_LT) {
-             if (bbs_style == 1) {
-                /* ok we found a prime smaller or
-                 * equal [so the next is larger]
-                 *
-                 * however, the prime must be
-                 * congruent to 3 mod 4
-                 */
-                if ((ltm_prime_tab[x + 1] & 3) != 3) {
-                   /* scan upwards for a prime congruent to 3 mod 4 */
-                   for (y = x + 1; y < PRIME_SIZE; y++) {
-                       if ((ltm_prime_tab[y] & 3) == 3) {
-                          mp_set(a, ltm_prime_tab[y]);
-                          return MP_OKAY;
-                       }
-                   }
-                }
-             } else {
-                mp_set(a, ltm_prime_tab[x + 1]);
-                return MP_OKAY;
-             }
-          }
+         if (mp_cmp_d(a, ltm_prime_tab[x]) != MP_LT) {
+            if (bbs_style == 1) {
+               /* ok we found a prime smaller or
+                * equal [so the next is larger]
+                *
+                * however, the prime must be
+                * congruent to 3 mod 4
+                */
+               if ((ltm_prime_tab[x + 1] & 3) != 3) {
+                  /* scan upwards for a prime congruent to 3 mod 4 */
+                  for (y = x + 1; y < PRIME_SIZE; y++) {
+                     if ((ltm_prime_tab[y] & 3) == 3) {
+                        mp_set(a, ltm_prime_tab[y]);
+                        return MP_OKAY;
+                     }
+                  }
+               }
+            } else {
+               mp_set(a, ltm_prime_tab[x + 1]);
+               return MP_OKAY;
+            }
+         }
       }
       /* at this point a maybe 1 */
       if (mp_cmp_d(a, 1) == MP_EQ) {
@@ -81,7 +81,9 @@ int mp_prime_next_prime(mp_int *a, int t, int bbs_style)
    if (bbs_style == 1) {
       /* if a mod 4 != 3 subtract the correct value to make it so */
       if ((a->dp[0] & 3) != 3) {
-         if ((err = mp_sub_d(a, (a->dp[0] & 3) + 1, a)) != MP_OKAY) { return err; };
+         if ((err = mp_sub_d(a, (a->dp[0] & 3) + 1, a)) != MP_OKAY) {
+            return err;
+         };
       }
    } else {
       if (mp_iseven(a) == MP_YES) {
@@ -116,18 +118,18 @@ int mp_prime_next_prime(mp_int *a, int t, int bbs_style)
 
          /* compute the new residue without using division */
          for (x = 1; x < PRIME_SIZE; x++) {
-             /* add the step to each residue */
-             res_tab[x] += kstep;
-
-             /* subtract the modulus [instead of using division] */
-             if (res_tab[x] >= ltm_prime_tab[x]) {
-                res_tab[x]  -= ltm_prime_tab[x];
-             }
-
-             /* set flag if zero */
-             if (res_tab[x] == 0) {
-                y = 1;
-             }
+            /* add the step to each residue */
+            res_tab[x] += kstep;
+
+            /* subtract the modulus [instead of using division] */
+            if (res_tab[x] >= ltm_prime_tab[x]) {
+               res_tab[x]  -= ltm_prime_tab[x];
+            }
+
+            /* set flag if zero */
+            if (res_tab[x] == 0) {
+               y = 1;
+            }
          }
       } while ((y == 1) && (step < ((((mp_digit)1) << DIGIT_BIT) - kstep)));
 
@@ -143,13 +145,13 @@ int mp_prime_next_prime(mp_int *a, int t, int bbs_style)
 
       /* is this prime? */
       for (x = 0; x < t; x++) {
-          mp_set(&b, ltm_prime_tab[x]);
-          if ((err = mp_prime_miller_rabin(a, &b, &res)) != MP_OKAY) {
-             goto LBL_ERR;
-          }
-          if (res == MP_NO) {
-             break;
-          }
+         mp_set(&b, ltm_prime_tab[x]);
+         if ((err = mp_prime_miller_rabin(a, &b, &res)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+         if (res == MP_NO) {
+            break;
+         }
       }
 
       if (res == MP_YES) {
diff --git a/libtommath/bn_mp_prime_rabin_miller_trials.c b/libtommath/bn_mp_prime_rabin_miller_trials.c
index 378ceb2..cde309a 100644
--- a/libtommath/bn_mp_prime_rabin_miller_trials.c
+++ b/libtommath/bn_mp_prime_rabin_miller_trials.c
@@ -19,14 +19,14 @@
 static const struct {
    int k, t;
 } sizes[] = {
-{   128,    28 },
-{   256,    16 },
-{   384,    10 },
-{   512,     7 },
-{   640,     6 },
-{   768,     5 },
-{   896,     4 },
-{  1024,     4 }
+   {   128,    28 },
+   {   256,    16 },
+   {   384,    10 },
+   {   512,     7 },
+   {   640,     6 },
+   {   768,     5 },
+   {   896,     4 },
+   {  1024,     4 }
 };
 
 /* returns # of RM trials required for a given bit size */
@@ -35,11 +35,11 @@ int mp_prime_rabin_miller_trials(int size)
    int x;
 
    for (x = 0; x < (int)(sizeof(sizes)/(sizeof(sizes[0]))); x++) {
-       if (sizes[x].k == size) {
-          return sizes[x].t;
-       } else if (sizes[x].k > size) {
-          return (x == 0) ? sizes[0].t : sizes[x - 1].t;
-       }
+      if (sizes[x].k == size) {
+         return sizes[x].t;
+      } else if (sizes[x].k > size) {
+         return (x == 0) ? sizes[0].t : sizes[x - 1].t;
+      }
    }
    return sizes[x-1].t + 1;
 }
diff --git a/libtommath/bn_mp_prime_random_ex.c b/libtommath/bn_mp_prime_random_ex.c
index cf5272e..d3d6f3d 100644
--- a/libtommath/bn_mp_prime_random_ex.c
+++ b/libtommath/bn_mp_prime_random_ex.c
@@ -18,7 +18,7 @@
 /* makes a truly random prime of a given size (bits),
  *
  * Flags are as follows:
- * 
+ *
  *   LTM_PRIME_BBS      - make prime congruent to 3 mod 4
  *   LTM_PRIME_SAFE     - make sure (p-1)/2 is prime as well (implies LTM_PRIME_BBS)
  *   LTM_PRIME_2MSB_ON  - make the 2nd highest bit one
@@ -62,7 +62,7 @@ int mp_prime_random_ex(mp_int *a, int t, int size, int flags, ltm_prime_callback
    maskOR_msb_offset = ((size & 7) == 1) ? 1 : 0;
    if ((flags & LTM_PRIME_2MSB_ON) != 0) {
       maskOR_msb       |= 0x80 >> ((9 - size) & 7);
-   }  
+   }
 
    /* get the maskOR_lsb */
    maskOR_lsb         = 1;
@@ -76,7 +76,7 @@ int mp_prime_random_ex(mp_int *a, int t, int size, int flags, ltm_prime_callback
          err = MP_VAL;
          goto error;
       }
- 
+
       /* work over the MSbyte */
       tmp[0]    &= maskAND;
       tmp[0]    |= 1 << ((size - 1) & 7);
@@ -86,28 +86,42 @@ int mp_prime_random_ex(mp_int *a, int t, int size, int flags, ltm_prime_callback
       tmp[bsize-1]             |= maskOR_lsb;
 
       /* read it in */
-      if ((err = mp_read_unsigned_bin(a, tmp, bsize)) != MP_OKAY)     { goto error; }
+      if ((err = mp_read_unsigned_bin(a, tmp, bsize)) != MP_OKAY)     {
+         goto error;
+      }
 
       /* is it prime? */
-      if ((err = mp_prime_is_prime(a, t, &res)) != MP_OKAY)           { goto error; }
-      if (res == MP_NO) {  
+      if ((err = mp_prime_is_prime(a, t, &res)) != MP_OKAY)           {
+         goto error;
+      }
+      if (res == MP_NO) {
          continue;
       }
 
       if ((flags & LTM_PRIME_SAFE) != 0) {
          /* see if (a-1)/2 is prime */
-         if ((err = mp_sub_d(a, 1, a)) != MP_OKAY)                    { goto error; }
-         if ((err = mp_div_2(a, a)) != MP_OKAY)                       { goto error; }
- 
+         if ((err = mp_sub_d(a, 1, a)) != MP_OKAY)                    {
+            goto error;
+         }
+         if ((err = mp_div_2(a, a)) != MP_OKAY)                       {
+            goto error;
+         }
+
          /* is it prime? */
-         if ((err = mp_prime_is_prime(a, t, &res)) != MP_OKAY)        { goto error; }
+         if ((err = mp_prime_is_prime(a, t, &res)) != MP_OKAY)        {
+            goto error;
+         }
       }
    } while (res == MP_NO);
 
    if ((flags & LTM_PRIME_SAFE) != 0) {
       /* restore a to the original value */
-      if ((err = mp_mul_2(a, a)) != MP_OKAY)                          { goto error; }
-      if ((err = mp_add_d(a, 1, a)) != MP_OKAY)                       { goto error; }
+      if ((err = mp_mul_2(a, a)) != MP_OKAY)                          {
+         goto error;
+      }
+      if ((err = mp_add_d(a, 1, a)) != MP_OKAY)                       {
+         goto error;
+      }
    }
 
    err = MP_OKAY;
diff --git a/libtommath/bn_mp_radix_size.c b/libtommath/bn_mp_radix_size.c
index cb0c134..a6e2f04 100644
--- a/libtommath/bn_mp_radix_size.c
+++ b/libtommath/bn_mp_radix_size.c
@@ -16,59 +16,59 @@
  */
 
 /* returns size of ASCII reprensentation */
-int mp_radix_size (mp_int * a, int radix, int *size)
+int mp_radix_size(mp_int *a, int radix, int *size)
 {
-  int     res, digs;
-  mp_int  t;
-  mp_digit d;
+   int     res, digs;
+   mp_int  t;
+   mp_digit d;
 
-  *size = 0;
+   *size = 0;
 
-  /* make sure the radix is in range */
-  if ((radix < 2) || (radix > 64)) {
-    return MP_VAL;
-  }
+   /* make sure the radix is in range */
+   if ((radix < 2) || (radix > 64)) {
+      return MP_VAL;
+   }
 
-  if (mp_iszero(a) == MP_YES) {
-    *size = 2;
-    return MP_OKAY;
-  }
+   if (mp_iszero(a) == MP_YES) {
+      *size = 2;
+      return MP_OKAY;
+   }
 
-  /* special case for binary */
-  if (radix == 2) {
-    *size = mp_count_bits (a) + ((a->sign == MP_NEG) ? 1 : 0) + 1;
-    return MP_OKAY;
-  }
+   /* special case for binary */
+   if (radix == 2) {
+      *size = mp_count_bits(a) + ((a->sign == MP_NEG) ? 1 : 0) + 1;
+      return MP_OKAY;
+   }
 
-  /* digs is the digit count */
-  digs = 0;
+   /* digs is the digit count */
+   digs = 0;
 
-  /* if it's negative add one for the sign */
-  if (a->sign == MP_NEG) {
-    ++digs;
-  }
+   /* if it's negative add one for the sign */
+   if (a->sign == MP_NEG) {
+      ++digs;
+   }
 
-  /* init a copy of the input */
-  if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-    return res;
-  }
+   /* init a copy of the input */
+   if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+      return res;
+   }
 
-  /* force temp to positive */
-  t.sign = MP_ZPOS; 
+   /* force temp to positive */
+   t.sign = MP_ZPOS;
 
-  /* fetch out all of the digits */
-  while (mp_iszero (&t) == MP_NO) {
-    if ((res = mp_div_d (&t, (mp_digit) radix, &t, &d)) != MP_OKAY) {
-      mp_clear (&t);
-      return res;
-    }
-    ++digs;
-  }
-  mp_clear (&t);
+   /* fetch out all of the digits */
+   while (mp_iszero(&t) == MP_NO) {
+      if ((res = mp_div_d(&t, (mp_digit)radix, &t, &d)) != MP_OKAY) {
+         mp_clear(&t);
+         return res;
+      }
+      ++digs;
+   }
+   mp_clear(&t);
 
-  /* return digs + 1, the 1 is for the NULL byte that would be required. */
-  *size = digs + 1;
-  return MP_OKAY;
+   /* return digs + 1, the 1 is for the NULL byte that would be required. */
+   *size = digs + 1;
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_rand.c b/libtommath/bn_mp_rand.c
index 93e255a..92a9a97 100644
--- a/libtommath/bn_mp_rand.c
+++ b/libtommath/bn_mp_rand.c
@@ -16,13 +16,13 @@
  */
 
 #if MP_GEN_RANDOM_MAX == 0xffffffff
-  #define MP_GEN_RANDOM_SHIFT  32
+#define MP_GEN_RANDOM_SHIFT  32
 #elif MP_GEN_RANDOM_MAX == 32767
-  /* SHRT_MAX */
-  #define MP_GEN_RANDOM_SHIFT  15
+/* SHRT_MAX */
+#define MP_GEN_RANDOM_SHIFT  15
 #elif MP_GEN_RANDOM_MAX == 2147483647
-  /* INT_MAX */
-  #define MP_GEN_RANDOM_SHIFT  31
+/* INT_MAX */
+#define MP_GEN_RANDOM_SHIFT  31
 #elif !defined(MP_GEN_RANDOM_SHIFT)
 #error Thou shalt define their own valid MP_GEN_RANDOM_SHIFT
 #endif
@@ -30,48 +30,47 @@
 /* makes a pseudo-random int of a given size */
 static mp_digit s_gen_random(void)
 {
-  mp_digit d = 0, msk = 0;
-  do {
-    d <<= MP_GEN_RANDOM_SHIFT;
-    d |= ((mp_digit) MP_GEN_RANDOM());
-    msk <<= MP_GEN_RANDOM_SHIFT;
-    msk |= (MP_MASK & MP_GEN_RANDOM_MAX);
-  } while ((MP_MASK & msk) != MP_MASK);
-  d &= MP_MASK;
-  return d;
+   mp_digit d = 0, msk = 0;
+   do {
+      d <<= MP_GEN_RANDOM_SHIFT;
+      d |= ((mp_digit) MP_GEN_RANDOM());
+      msk <<= MP_GEN_RANDOM_SHIFT;
+      msk |= (MP_MASK & MP_GEN_RANDOM_MAX);
+   } while ((MP_MASK & msk) != MP_MASK);
+   d &= MP_MASK;
+   return d;
 }
 
-int
-mp_rand (mp_int * a, int digits)
+int mp_rand(mp_int *a, int digits)
 {
-  int     res;
-  mp_digit d;
+   int     res;
+   mp_digit d;
 
-  mp_zero (a);
-  if (digits <= 0) {
-    return MP_OKAY;
-  }
+   mp_zero(a);
+   if (digits <= 0) {
+      return MP_OKAY;
+   }
 
-  /* first place a random non-zero digit */
-  do {
-    d = s_gen_random();
-  } while (d == 0);
+   /* first place a random non-zero digit */
+   do {
+      d = s_gen_random();
+   } while (d == 0);
 
-  if ((res = mp_add_d (a, d, a)) != MP_OKAY) {
-    return res;
-  }
-
-  while (--digits > 0) {
-    if ((res = mp_lshd (a, 1)) != MP_OKAY) {
+   if ((res = mp_add_d(a, d, a)) != MP_OKAY) {
       return res;
-    }
+   }
 
-    if ((res = mp_add_d (a, s_gen_random(), a)) != MP_OKAY) {
-      return res;
-    }
-  }
+   while (--digits > 0) {
+      if ((res = mp_lshd(a, 1)) != MP_OKAY) {
+         return res;
+      }
+
+      if ((res = mp_add_d(a, s_gen_random(), a)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_read_radix.c b/libtommath/bn_mp_read_radix.c
index 12aa499..bc31cc5 100644
--- a/libtommath/bn_mp_read_radix.c
+++ b/libtommath/bn_mp_read_radix.c
@@ -16,73 +16,73 @@
  */
 
 /* read a string [ASCII] in a given radix */
-int mp_read_radix (mp_int * a, const char *str, int radix)
+int mp_read_radix(mp_int *a, const char *str, int radix)
 {
-  int     y, res, neg;
-  char    ch;
+   int     y, res, neg;
+   char    ch;
 
-  /* zero the digit bignum */
-  mp_zero(a);
+   /* zero the digit bignum */
+   mp_zero(a);
 
-  /* make sure the radix is ok */
-  if ((radix < 2) || (radix > 64)) {
-    return MP_VAL;
-  }
+   /* make sure the radix is ok */
+   if ((radix < 2) || (radix > 64)) {
+      return MP_VAL;
+   }
 
-  /* if the leading digit is a 
-   * minus set the sign to negative. 
-   */
-  if (*str == '-') {
-    ++str;
-    neg = MP_NEG;
-  } else {
-    neg = MP_ZPOS;
-  }
+   /* if the leading digit is a
+    * minus set the sign to negative.
+    */
+   if (*str == '-') {
+      ++str;
+      neg = MP_NEG;
+   } else {
+      neg = MP_ZPOS;
+   }
 
-  /* set the integer to the default of zero */
-  mp_zero (a);
-  
-  /* process each digit of the string */
-  while (*str != '\0') {
-    /* if the radix <= 36 the conversion is case insensitive
-     * this allows numbers like 1AB and 1ab to represent the same  value
-     * [e.g. in hex]
-     */
-    ch = (radix <= 36) ? (char)toupper((int)*str) : *str;
-    for (y = 0; y < 64; y++) {
-      if (ch == mp_s_rmap[y]) {
-         break;
-      }
-    }
+   /* set the integer to the default of zero */
+   mp_zero(a);
 
-    /* if the char was found in the map 
-     * and is less than the given radix add it
-     * to the number, otherwise exit the loop. 
-     */
-    if (y < radix) {
-      if ((res = mp_mul_d (a, (mp_digit) radix, a)) != MP_OKAY) {
-         return res;
+   /* process each digit of the string */
+   while (*str != '\0') {
+      /* if the radix <= 36 the conversion is case insensitive
+       * this allows numbers like 1AB and 1ab to represent the same  value
+       * [e.g. in hex]
+       */
+      ch = (radix <= 36) ? (char)toupper((int)*str) : *str;
+      for (y = 0; y < 64; y++) {
+         if (ch == mp_s_rmap[y]) {
+            break;
+         }
       }
-      if ((res = mp_add_d (a, (mp_digit) y, a)) != MP_OKAY) {
-         return res;
+
+      /* if the char was found in the map
+       * and is less than the given radix add it
+       * to the number, otherwise exit the loop.
+       */
+      if (y < radix) {
+         if ((res = mp_mul_d(a, (mp_digit)radix, a)) != MP_OKAY) {
+            return res;
+         }
+         if ((res = mp_add_d(a, (mp_digit)y, a)) != MP_OKAY) {
+            return res;
+         }
+      } else {
+         break;
       }
-    } else {
-      break;
-    }
-    ++str;
-  }
+      ++str;
+   }
 
-  /* if an illegal character was found, fail. */
-  if (!(*str == '\0' || *str == '\r' || *str == '\n')) {
+   /* if an illegal character was found, fail. */
+   if (!(*str == '\0' || *str == '\r' || *str == '\n')) {
       mp_zero(a);
       return MP_VAL;
-  }
+   }
 
-  /* set the sign only if a != 0 */
-  if (mp_iszero(a) != MP_YES) {
-     a->sign = neg;
-  }
-  return MP_OKAY;
+   /* set the sign only if a != 0 */
+   if (mp_iszero(a) != MP_YES) {
+      a->sign = neg;
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_read_signed_bin.c b/libtommath/bn_mp_read_signed_bin.c
index 363e11f..eabc803 100644
--- a/libtommath/bn_mp_read_signed_bin.c
+++ b/libtommath/bn_mp_read_signed_bin.c
@@ -16,23 +16,23 @@
  */
 
 /* read signed bin, big endian, first byte is 0==positive or 1==negative */
-int mp_read_signed_bin (mp_int * a, const unsigned char *b, int c)
+int mp_read_signed_bin(mp_int *a, const unsigned char *b, int c)
 {
-  int     res;
+   int     res;
 
-  /* read magnitude */
-  if ((res = mp_read_unsigned_bin (a, b + 1, c - 1)) != MP_OKAY) {
-    return res;
-  }
+   /* read magnitude */
+   if ((res = mp_read_unsigned_bin(a, b + 1, c - 1)) != MP_OKAY) {
+      return res;
+   }
 
-  /* first byte is 0 for positive, non-zero for negative */
-  if (b[0] == 0) {
-     a->sign = MP_ZPOS;
-  } else {
-     a->sign = MP_NEG;
-  }
+   /* first byte is 0 for positive, non-zero for negative */
+   if (b[0] == 0) {
+      a->sign = MP_ZPOS;
+   } else {
+      a->sign = MP_NEG;
+   }
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_read_unsigned_bin.c b/libtommath/bn_mp_read_unsigned_bin.c
index 1a50b58..ad9f05f 100644
--- a/libtommath/bn_mp_read_unsigned_bin.c
+++ b/libtommath/bn_mp_read_unsigned_bin.c
@@ -16,37 +16,37 @@
  */
 
 /* reads a unsigned char array, assumes the msb is stored first [big endian] */
-int mp_read_unsigned_bin (mp_int * a, const unsigned char *b, int c)
+int mp_read_unsigned_bin(mp_int *a, const unsigned char *b, int c)
 {
-  int     res;
+   int     res;
 
-  /* make sure there are at least two digits */
-  if (a->alloc < 2) {
-     if ((res = mp_grow(a, 2)) != MP_OKAY) {
-        return res;
-     }
-  }
+   /* make sure there are at least two digits */
+   if (a->alloc < 2) {
+      if ((res = mp_grow(a, 2)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* zero the int */
-  mp_zero (a);
+   /* zero the int */
+   mp_zero(a);
 
-  /* read the bytes in */
-  while (c-- > 0) {
-    if ((res = mp_mul_2d (a, 8, a)) != MP_OKAY) {
-      return res;
-    }
+   /* read the bytes in */
+   while (c-- > 0) {
+      if ((res = mp_mul_2d(a, 8, a)) != MP_OKAY) {
+         return res;
+      }
 
 #ifndef MP_8BIT
-    a->dp[0] |= *b++;
-    a->used += 1;
+      a->dp[0] |= *b++;
+      a->used += 1;
 #else
-    a->dp[0] = (*b & MP_MASK);
-    a->dp[1] |= ((*b++ >> 7U) & 1);
-    a->used += 2;
+      a->dp[0] = (*b & MP_MASK);
+      a->dp[1] |= ((*b++ >> 7U) & 1);
+      a->used += 2;
 #endif
-  }
-  mp_clamp (a);
-  return MP_OKAY;
+   }
+   mp_clamp(a);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_reduce.c b/libtommath/bn_mp_reduce.c
index 367383f..a2b9bf7 100644
--- a/libtommath/bn_mp_reduce.c
+++ b/libtommath/bn_mp_reduce.c
@@ -19,79 +19,79 @@
  * precomputed via mp_reduce_setup.
  * From HAC pp.604 Algorithm 14.42
  */
-int mp_reduce (mp_int * x, mp_int * m, mp_int * mu)
+int mp_reduce(mp_int *x, mp_int *m, mp_int *mu)
 {
-  mp_int  q;
-  int     res, um = m->used;
+   mp_int  q;
+   int     res, um = m->used;
 
-  /* q = x */
-  if ((res = mp_init_copy (&q, x)) != MP_OKAY) {
-    return res;
-  }
+   /* q = x */
+   if ((res = mp_init_copy(&q, x)) != MP_OKAY) {
+      return res;
+   }
 
-  /* q1 = x / b**(k-1)  */
-  mp_rshd (&q, um - 1);
+   /* q1 = x / b**(k-1)  */
+   mp_rshd(&q, um - 1);
 
-  /* according to HAC this optimization is ok */
-  if (((mp_digit) um) > (((mp_digit)1) << (DIGIT_BIT - 1))) {
-    if ((res = mp_mul (&q, mu, &q)) != MP_OKAY) {
-      goto CLEANUP;
-    }
-  } else {
+   /* according to HAC this optimization is ok */
+   if (((mp_digit) um) > (((mp_digit)1) << (DIGIT_BIT - 1))) {
+      if ((res = mp_mul(&q, mu, &q)) != MP_OKAY) {
+         goto CLEANUP;
+      }
+   } else {
 #ifdef BN_S_MP_MUL_HIGH_DIGS_C
-    if ((res = s_mp_mul_high_digs (&q, mu, &q, um)) != MP_OKAY) {
-      goto CLEANUP;
-    }
+      if ((res = s_mp_mul_high_digs(&q, mu, &q, um)) != MP_OKAY) {
+         goto CLEANUP;
+      }
 #elif defined(BN_FAST_S_MP_MUL_HIGH_DIGS_C)
-    if ((res = fast_s_mp_mul_high_digs (&q, mu, &q, um)) != MP_OKAY) {
-      goto CLEANUP;
-    }
+      if ((res = fast_s_mp_mul_high_digs(&q, mu, &q, um)) != MP_OKAY) {
+         goto CLEANUP;
+      }
 #else
-    {
-      res = MP_VAL;
-      goto CLEANUP;
-    }
+      {
+         res = MP_VAL;
+         goto CLEANUP;
+      }
 #endif
-  }
-
-  /* q3 = q2 / b**(k+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;
-  }
+   /* q3 = q2 / b**(k+1) */
+   mp_rshd(&q, um + 1);
 
-  /* q = q * m mod b**(k+1), quick (no division) */
-  if ((res = s_mp_mul_digs (&q, m, &q, um + 1)) != MP_OKAY) {
-    goto CLEANUP;
-  }
-
-  /* x = x - q */
-  if ((res = mp_sub (x, &q, x)) != MP_OKAY) {
-    goto CLEANUP;
-  }
-
-  /* If x < 0, add b**(k+1) to it */
-  if (mp_cmp_d (x, 0) == MP_LT) {
-    mp_set (&q, 1);
-    if ((res = mp_lshd (&q, um + 1)) != MP_OKAY)
+   /* x = x mod b**(k+1), quick (no division) */
+   if ((res = mp_mod_2d(x, DIGIT_BIT * (um + 1), x)) != MP_OKAY) {
       goto CLEANUP;
-    if ((res = mp_add (x, &q, x)) != MP_OKAY)
+   }
+
+   /* q = q * m mod b**(k+1), quick (no division) */
+   if ((res = s_mp_mul_digs(&q, m, &q, um + 1)) != MP_OKAY) {
       goto CLEANUP;
-  }
+   }
 
-  /* Back off if it's too big */
-  while (mp_cmp (x, m) != MP_LT) {
-    if ((res = s_mp_sub (x, m, x)) != MP_OKAY) {
+   /* x = x - q */
+   if ((res = mp_sub(x, &q, x)) != MP_OKAY) {
       goto CLEANUP;
-    }
-  }
+   }
+
+   /* If x < 0, add b**(k+1) to it */
+   if (mp_cmp_d(x, 0) == MP_LT) {
+      mp_set(&q, 1);
+      if ((res = mp_lshd(&q, um + 1)) != MP_OKAY)
+         goto CLEANUP;
+      if ((res = mp_add(x, &q, x)) != MP_OKAY)
+         goto CLEANUP;
+   }
+
+   /* Back off if it's too big */
+   while (mp_cmp(x, m) != MP_LT) {
+      if ((res = s_mp_sub(x, m, x)) != MP_OKAY) {
+         goto CLEANUP;
+      }
+   }
 
 CLEANUP:
-  mp_clear (&q);
+   mp_clear(&q);
 
-  return res;
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_reduce_2k_setup.c b/libtommath/bn_mp_reduce_2k_setup.c
index bf810c0..802a5ba 100644
--- a/libtommath/bn_mp_reduce_2k_setup.c
+++ b/libtommath/bn_mp_reduce_2k_setup.c
@@ -20,22 +20,22 @@ int mp_reduce_2k_setup(mp_int *a, mp_digit *d)
 {
    int res, p;
    mp_int tmp;
-   
+
    if ((res = mp_init(&tmp)) != MP_OKAY) {
       return res;
    }
-   
+
    p = mp_count_bits(a);
    if ((res = mp_2expt(&tmp, p)) != MP_OKAY) {
       mp_clear(&tmp);
       return res;
    }
-   
+
    if ((res = s_mp_sub(&tmp, a, &tmp)) != MP_OKAY) {
       mp_clear(&tmp);
       return res;
    }
-   
+
    *d = tmp.dp[0];
    mp_clear(&tmp);
    return MP_OKAY;
diff --git a/libtommath/bn_mp_reduce_2k_setup_l.c b/libtommath/bn_mp_reduce_2k_setup_l.c
index 56d1ba8..34367ed 100644
--- a/libtommath/bn_mp_reduce_2k_setup_l.c
+++ b/libtommath/bn_mp_reduce_2k_setup_l.c
@@ -20,19 +20,19 @@ int mp_reduce_2k_setup_l(mp_int *a, mp_int *d)
 {
    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;
diff --git a/libtommath/bn_mp_reduce_is_2k.c b/libtommath/bn_mp_reduce_is_2k.c
index 0499e83..c733ca9 100644
--- a/libtommath/bn_mp_reduce_is_2k.c
+++ b/libtommath/bn_mp_reduce_is_2k.c
@@ -20,7 +20,7 @@ int mp_reduce_is_2k(mp_int *a)
 {
    int ix, iy, iw;
    mp_digit iz;
-   
+
    if (a->used == 0) {
       return MP_NO;
    } else if (a->used == 1) {
@@ -29,17 +29,17 @@ int mp_reduce_is_2k(mp_int *a)
       iy = mp_count_bits(a);
       iz = 1;
       iw = 1;
-    
+
       /* Test every bit from the second digit up, must be 1 */
       for (ix = DIGIT_BIT; ix < iy; ix++) {
-          if ((a->dp[iw] & iz) == 0) {
-             return MP_NO;
-          }
-          iz <<= 1;
-          if (iz > (mp_digit)MP_MASK) {
-             ++iw;
-             iz = 1;
-          }
+         if ((a->dp[iw] & iz) == 0) {
+            return MP_NO;
+         }
+         iz <<= 1;
+         if (iz > (mp_digit)MP_MASK) {
+            ++iw;
+            iz = 1;
+         }
       }
    }
    return MP_YES;
diff --git a/libtommath/bn_mp_reduce_is_2k_l.c b/libtommath/bn_mp_reduce_is_2k_l.c
index 48b3498..d4804d5 100644
--- a/libtommath/bn_mp_reduce_is_2k_l.c
+++ b/libtommath/bn_mp_reduce_is_2k_l.c
@@ -19,7 +19,7 @@
 int mp_reduce_is_2k_l(mp_int *a)
 {
    int ix, iy;
-   
+
    if (a->used == 0) {
       return MP_NO;
    } else if (a->used == 1) {
@@ -27,12 +27,12 @@ int mp_reduce_is_2k_l(mp_int *a)
    } else if (a->used > 1) {
       /* if more than half of the digits are -1 we're sold */
       for (iy = ix = 0; ix < a->used; ix++) {
-          if (a->dp[ix] == MP_MASK) {
-              ++iy;
-          }
+         if (a->dp[ix] == MP_MASK) {
+            ++iy;
+         }
       }
       return (iy >= (a->used/2)) ? MP_YES : MP_NO;
-      
+
    }
    return MP_NO;
 }
diff --git a/libtommath/bn_mp_reduce_setup.c b/libtommath/bn_mp_reduce_setup.c
index 8875698..00ff61c 100644
--- a/libtommath/bn_mp_reduce_setup.c
+++ b/libtommath/bn_mp_reduce_setup.c
@@ -18,14 +18,14 @@
 /* pre-calculate the value required for Barrett reduction
  * For a given modulus "b" it calulates the value required in "a"
  */
-int mp_reduce_setup (mp_int * a, mp_int * b)
+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;
-  }
-  return mp_div (a, b, a, NULL);
+   int     res;
+
+   if ((res = mp_2expt(a, b->used * 2 * DIGIT_BIT)) != MP_OKAY) {
+      return res;
+   }
+   return mp_div(a, b, a, NULL);
 }
 #endif
 
diff --git a/libtommath/bn_mp_rshd.c b/libtommath/bn_mp_rshd.c
index 4b598de..fd06735 100644
--- a/libtommath/bn_mp_rshd.c
+++ b/libtommath/bn_mp_rshd.c
@@ -16,54 +16,54 @@
  */
 
 /* shift right a certain amount of digits */
-void mp_rshd (mp_int * a, int b)
+void mp_rshd(mp_int *a, int b)
 {
-  int     x;
+   int     x;
 
-  /* if b <= 0 then ignore it */
-  if (b <= 0) {
-    return;
-  }
+   /* if b <= 0 then ignore it */
+   if (b <= 0) {
+      return;
+   }
 
-  /* if b > used then simply zero it and return */
-  if (a->used <= b) {
-    mp_zero (a);
-    return;
-  }
+   /* if b > used then simply zero it and return */
+   if (a->used <= b) {
+      mp_zero(a);
+      return;
+   }
 
-  {
-    mp_digit *bottom, *top;
+   {
+      mp_digit *bottom, *top;
 
-    /* shift the digits down */
+      /* shift the digits down */
 
-    /* bottom */
-    bottom = a->dp;
+      /* bottom */
+      bottom = a->dp;
 
-    /* top [offset into digits] */
-    top = a->dp + b;
+      /* 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 
-     * the top of the window are copied to the bottom
-     *
-     * e.g.
+      /* 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.
 
-     b-2 | b-1 | b0 | b1 | b2 | ... | bb |   ---->
-                 /\                   |      ---->
-                  \-------------------/      ---->
-     */
-    for (x = 0; x < (a->used - b); x++) {
-      *bottom++ = *top++;
-    }
+       b-2 | b-1 | b0 | b1 | b2 | ... | bb |   ---->
+                   /\                   |      ---->
+                    \-------------------/      ---->
+       */
+      for (x = 0; x < (a->used - b); x++) {
+         *bottom++ = *top++;
+      }
 
-    /* zero the top digits */
-    for (; x < a->used; x++) {
-      *bottom++ = 0;
-    }
-  }
-  
-  /* remove excess digits */
-  a->used -= b;
+      /* zero the top digits */
+      for (; x < a->used; x++) {
+         *bottom++ = 0;
+      }
+   }
+
+   /* remove excess digits */
+   a->used -= b;
 }
 #endif
 
diff --git a/libtommath/bn_mp_set.c b/libtommath/bn_mp_set.c
index dd4de3c..eaf7fed 100644
--- a/libtommath/bn_mp_set.c
+++ b/libtommath/bn_mp_set.c
@@ -16,11 +16,11 @@
  */
 
 /* set to a digit */
-void mp_set (mp_int * a, mp_digit b)
+void mp_set(mp_int *a, mp_digit b)
 {
-  mp_zero (a);
-  a->dp[0] = b & MP_MASK;
-  a->used  = (a->dp[0] != 0) ? 1 : 0;
+   mp_zero(a);
+   a->dp[0] = b & MP_MASK;
+   a->used  = (a->dp[0] != 0) ? 1 : 0;
 }
 #endif
 
diff --git a/libtommath/bn_mp_set_int.c b/libtommath/bn_mp_set_int.c
index 3aafec9..4c71180 100644
--- a/libtommath/bn_mp_set_int.c
+++ b/libtommath/bn_mp_set_int.c
@@ -16,30 +16,30 @@
  */
 
 /* set a 32-bit const */
-int mp_set_int (mp_int * a, unsigned long b)
+int mp_set_int(mp_int *a, unsigned long b)
 {
-  int     x, res;
+   int     x, res;
 
-  mp_zero (a);
-  
-  /* set four bits at a time */
-  for (x = 0; x < 8; x++) {
-    /* shift the number up four bits */
-    if ((res = mp_mul_2d (a, 4, a)) != MP_OKAY) {
-      return res;
-    }
+   mp_zero(a);
 
-    /* OR in the top four bits of the source */
-    a->dp[0] |= (b >> 28) & 15;
+   /* set four bits at a time */
+   for (x = 0; x < 8; x++) {
+      /* shift the number up four bits */
+      if ((res = mp_mul_2d(a, 4, a)) != MP_OKAY) {
+         return res;
+      }
 
-    /* shift the source up to the next four bits */
-    b <<= 4;
+      /* OR in the top four bits of the source */
+      a->dp[0] |= (b >> 28) & 15;
 
-    /* ensure that digits are not clamped off */
-    a->used += 1;
-  }
-  mp_clamp (a);
-  return MP_OKAY;
+      /* shift the source up to the next four bits */
+      b <<= 4;
+
+      /* ensure that digits are not clamped off */
+      a->used += 1;
+   }
+   mp_clamp(a);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_shrink.c b/libtommath/bn_mp_shrink.c
index 0712c2b..8ac1f33 100644
--- a/libtommath/bn_mp_shrink.c
+++ b/libtommath/bn_mp_shrink.c
@@ -16,23 +16,23 @@
  */
 
 /* shrink a bignum */
-int mp_shrink (mp_int * a)
+int mp_shrink(mp_int *a)
 {
-  mp_digit *tmp;
-  int used = 1;
-  
-  if(a->used > 0) {
-    used = a->used;
-  }
-  
-  if (a->alloc != used) {
-    if ((tmp = OPT_CAST(mp_digit) XREALLOC (a->dp, sizeof (mp_digit) * used)) == NULL) {
-      return MP_MEM;
-    }
-    a->dp    = tmp;
-    a->alloc = used;
-  }
-  return MP_OKAY;
+   mp_digit *tmp;
+   int used = 1;
+
+   if (a->used > 0) {
+      used = a->used;
+   }
+
+   if (a->alloc != used) {
+      if ((tmp = OPT_CAST(mp_digit) XREALLOC(a->dp, sizeof(mp_digit) * used)) == NULL) {
+         return MP_MEM;
+      }
+      a->dp    = tmp;
+      a->alloc = used;
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_signed_bin_size.c b/libtommath/bn_mp_signed_bin_size.c
index 0910333..082aeca 100644
--- a/libtommath/bn_mp_signed_bin_size.c
+++ b/libtommath/bn_mp_signed_bin_size.c
@@ -16,9 +16,9 @@
  */
 
 /* get the size for an signed equivalent */
-int mp_signed_bin_size (mp_int * a)
+int mp_signed_bin_size(mp_int *a)
 {
-  return 1 + mp_unsigned_bin_size (a);
+   return 1 + mp_unsigned_bin_size(a);
 }
 #endif
 
diff --git a/libtommath/bn_mp_sqr.c b/libtommath/bn_mp_sqr.c
index ffe94e2..e2e8641 100644
--- a/libtommath/bn_mp_sqr.c
+++ b/libtommath/bn_mp_sqr.c
@@ -16,42 +16,41 @@
  */
 
 /* computes b = a*a */
-int
-mp_sqr (mp_int * a, mp_int * b)
+int mp_sqr(mp_int *a, mp_int *b)
 {
-  int     res;
+   int     res;
 
 #ifdef BN_MP_TOOM_SQR_C
-  /* use Toom-Cook? */
-  if (a->used >= TOOM_SQR_CUTOFF) {
-    res = mp_toom_sqr(a, b);
-  /* Karatsuba? */
-  } else 
+   /* use Toom-Cook? */
+   if (a->used >= TOOM_SQR_CUTOFF) {
+      res = mp_toom_sqr(a, b);
+      /* Karatsuba? */
+   } else
 #endif
 #ifdef BN_MP_KARATSUBA_SQR_C
-  if (a->used >= KARATSUBA_SQR_CUTOFF) {
-    res = mp_karatsuba_sqr (a, b);
-  } else 
+      if (a->used >= KARATSUBA_SQR_CUTOFF) {
+         res = mp_karatsuba_sqr(a, b);
+      } 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 <
-         (1 << (((sizeof(mp_word) * CHAR_BIT) - (2 * DIGIT_BIT)) - 1)))) {
-      res = fast_s_mp_sqr (a, b);
-    } else
+         /* can we use the fast comba multiplier? */
+         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
 #endif
-    {
+         {
 #ifdef BN_S_MP_SQR_C
-      res = s_mp_sqr (a, b);
+            res = s_mp_sqr(a, b);
 #else
-      res = MP_VAL;
+            res = MP_VAL;
 #endif
-    }
-  }
-  b->sign = MP_ZPOS;
-  return res;
+         }
+      }
+   b->sign = MP_ZPOS;
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_sqrmod.c b/libtommath/bn_mp_sqrmod.c
index 3b29fbc..96a7574 100644
--- a/libtommath/bn_mp_sqrmod.c
+++ b/libtommath/bn_mp_sqrmod.c
@@ -16,23 +16,22 @@
  */
 
 /* c = a * a (mod b) */
-int
-mp_sqrmod (mp_int * a, mp_int * b, mp_int * c)
+int mp_sqrmod(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res;
-  mp_int  t;
+   int     res;
+   mp_int  t;
 
-  if ((res = mp_init (&t)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init(&t)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_sqr (a, &t)) != MP_OKAY) {
-    mp_clear (&t);
-    return res;
-  }
-  res = mp_mod (&t, b, c);
-  mp_clear (&t);
-  return res;
+   if ((res = mp_sqr(a, &t)) != MP_OKAY) {
+      mp_clear(&t);
+      return res;
+   }
+   res = mp_mod(&t, b, c);
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_sqrt.c b/libtommath/bn_mp_sqrt.c
index d3b5d62..95a6892 100644
--- a/libtommath/bn_mp_sqrt.c
+++ b/libtommath/bn_mp_sqrt.c
@@ -16,62 +16,64 @@
  */
 
 /* this function is less generic than mp_n_root, simpler and faster */
-int mp_sqrt(mp_int *arg, mp_int *ret) 
+int mp_sqrt(mp_int *arg, mp_int *ret)
 {
-  int res;
-  mp_int t1,t2;
+   int res;
+   mp_int t1, t2;
 
-  /* must be positive */
-  if (arg->sign == MP_NEG) {
-    return MP_VAL;
-  }
+   /* must be positive */
+   if (arg->sign == MP_NEG) {
+      return MP_VAL;
+   }
 
-  /* easy out */
-  if (mp_iszero(arg) == MP_YES) {
-    mp_zero(ret);
-    return MP_OKAY;
-  }
+   /* easy out */
+   if (mp_iszero(arg) == MP_YES) {
+      mp_zero(ret);
+      return MP_OKAY;
+   }
 
-  if ((res = mp_init_copy(&t1, arg)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_copy(&t1, arg)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_init(&t2)) != MP_OKAY) {
-    goto E2;
-  }
+   if ((res = mp_init(&t2)) != MP_OKAY) {
+      goto E2;
+   }
 
-  /* First approx. (not very bad for large arg) */
-  mp_rshd (&t1,t1.used/2);
+   /* First approx. (not very bad for large arg) */
+   mp_rshd(&t1, t1.used/2);
 
-  /* t1 > 0  */ 
-  if ((res = mp_div(arg,&t1,&t2,NULL)) != MP_OKAY) {
-    goto E1;
-  }
-  if ((res = mp_add(&t1,&t2,&t1)) != MP_OKAY) {
-    goto E1;
-  }
-  if ((res = mp_div_2(&t1,&t1)) != MP_OKAY) {
-    goto E1;
-  }
-  /* And now t1 > sqrt(arg) */
-  do { 
-    if ((res = mp_div(arg,&t1,&t2,NULL)) != MP_OKAY) {
+   /* t1 > 0  */
+   if ((res = mp_div(arg, &t1, &t2, NULL)) != MP_OKAY) {
       goto E1;
-    }
-    if ((res = mp_add(&t1,&t2,&t1)) != MP_OKAY) {
+   }
+   if ((res = mp_add(&t1, &t2, &t1)) != MP_OKAY) {
       goto E1;
-    }
-    if ((res = mp_div_2(&t1,&t1)) != MP_OKAY) {
+   }
+   if ((res = mp_div_2(&t1, &t1)) != MP_OKAY) {
       goto E1;
-    }
-    /* t1 >= sqrt(arg) >= t2 at this point */
-  } while (mp_cmp_mag(&t1,&t2) == MP_GT);
+   }
+   /* And now t1 > sqrt(arg) */
+   do {
+      if ((res = mp_div(arg, &t1, &t2, NULL)) != MP_OKAY) {
+         goto E1;
+      }
+      if ((res = mp_add(&t1, &t2, &t1)) != MP_OKAY) {
+         goto E1;
+      }
+      if ((res = mp_div_2(&t1, &t1)) != MP_OKAY) {
+         goto E1;
+      }
+      /* t1 >= sqrt(arg) >= t2 at this point */
+   } while (mp_cmp_mag(&t1, &t2) == MP_GT);
 
-  mp_exch(&t1,ret);
+   mp_exch(&t1, ret);
 
-E1: mp_clear(&t2);
-E2: mp_clear(&t1);
-  return res;
+E1:
+   mp_clear(&t2);
+E2:
+   mp_clear(&t1);
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_sqrtmod_prime.c b/libtommath/bn_mp_sqrtmod_prime.c
index 968729e..12b427c 100644
--- a/libtommath/bn_mp_sqrtmod_prime.c
+++ b/libtommath/bn_mp_sqrtmod_prime.c
@@ -17,108 +17,108 @@
 
 int mp_sqrtmod_prime(mp_int *n, mp_int *prime, mp_int *ret)
 {
-  int res, legendre;
-  mp_int t1, C, Q, S, Z, M, T, R, two;
-  mp_digit i;
+   int res, legendre;
+   mp_int t1, C, Q, S, Z, M, T, R, two;
+   mp_digit i;
 
-  /* first handle the simple cases */
-  if (mp_cmp_d(n, 0) == MP_EQ) {
-    mp_zero(ret);
-    return MP_OKAY;
-  }
-  if (mp_cmp_d(prime, 2) == MP_EQ)                              return MP_VAL; /* prime must be odd */
-  if ((res = mp_jacobi(n, prime, &legendre)) != MP_OKAY)        return res;
-  if (legendre == -1)                                           return MP_VAL; /* quadratic non-residue mod prime */
+   /* first handle the simple cases */
+   if (mp_cmp_d(n, 0) == MP_EQ) {
+      mp_zero(ret);
+      return MP_OKAY;
+   }
+   if (mp_cmp_d(prime, 2) == MP_EQ)                              return MP_VAL; /* prime must be odd */
+   if ((res = mp_jacobi(n, prime, &legendre)) != MP_OKAY)        return res;
+   if (legendre == -1)                                           return MP_VAL; /* quadratic non-residue mod prime */
 
-  if ((res = mp_init_multi(&t1, &C, &Q, &S, &Z, &M, &T, &R, &two, NULL)) != MP_OKAY) {
-	return res;
-  }
+   if ((res = mp_init_multi(&t1, &C, &Q, &S, &Z, &M, &T, &R, &two, NULL)) != MP_OKAY) {
+      return res;
+   }
 
-  /* SPECIAL CASE: if prime mod 4 == 3
-   * compute directly: res = n^(prime+1)/4 mod prime
-   * Handbook of Applied Cryptography algorithm 3.36
-   */
-  if ((res = mp_mod_d(prime, 4, &i)) != MP_OKAY)                goto cleanup;
-  if (i == 3) {
-    if ((res = mp_add_d(prime, 1, &t1)) != MP_OKAY)             goto cleanup;
-    if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                  goto cleanup;
-    if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                  goto cleanup;
-    if ((res = mp_exptmod(n, &t1, prime, ret)) != MP_OKAY)      goto cleanup;
-    res = MP_OKAY;
-    goto cleanup;
-  }
+   /* SPECIAL CASE: if prime mod 4 == 3
+    * compute directly: res = n^(prime+1)/4 mod prime
+    * Handbook of Applied Cryptography algorithm 3.36
+    */
+   if ((res = mp_mod_d(prime, 4, &i)) != MP_OKAY)                goto cleanup;
+   if (i == 3) {
+      if ((res = mp_add_d(prime, 1, &t1)) != MP_OKAY)             goto cleanup;
+      if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                  goto cleanup;
+      if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                  goto cleanup;
+      if ((res = mp_exptmod(n, &t1, prime, ret)) != MP_OKAY)      goto cleanup;
+      res = MP_OKAY;
+      goto cleanup;
+   }
 
-  /* NOW: Tonelli-Shanks algorithm */
+   /* NOW: Tonelli-Shanks algorithm */
 
-  /* factor out powers of 2 from prime-1, defining Q and S as: prime-1 = Q*2^S */
-  if ((res = mp_copy(prime, &Q)) != MP_OKAY)                    goto cleanup;
-  if ((res = mp_sub_d(&Q, 1, &Q)) != MP_OKAY)                   goto cleanup;
-  /* Q = prime - 1 */
-  mp_zero(&S);
-  /* S = 0 */
-  while (mp_iseven(&Q) != MP_NO) {
-    if ((res = mp_div_2(&Q, &Q)) != MP_OKAY)                    goto cleanup;
-    /* Q = Q / 2 */
-    if ((res = mp_add_d(&S, 1, &S)) != MP_OKAY)                 goto cleanup;
-    /* S = S + 1 */
-  }
+   /* factor out powers of 2 from prime-1, defining Q and S as: prime-1 = Q*2^S */
+   if ((res = mp_copy(prime, &Q)) != MP_OKAY)                    goto cleanup;
+   if ((res = mp_sub_d(&Q, 1, &Q)) != MP_OKAY)                   goto cleanup;
+   /* Q = prime - 1 */
+   mp_zero(&S);
+   /* S = 0 */
+   while (mp_iseven(&Q) != MP_NO) {
+      if ((res = mp_div_2(&Q, &Q)) != MP_OKAY)                    goto cleanup;
+      /* Q = Q / 2 */
+      if ((res = mp_add_d(&S, 1, &S)) != MP_OKAY)                 goto cleanup;
+      /* S = S + 1 */
+   }
 
-  /* find a Z such that the Legendre symbol (Z|prime) == -1 */
-  if ((res = mp_set_int(&Z, 2)) != MP_OKAY)                     goto cleanup;
-  /* Z = 2 */
-  while(1) {
-    if ((res = mp_jacobi(&Z, prime, &legendre)) != MP_OKAY)     goto cleanup;
-    if (legendre == -1) break;
-    if ((res = mp_add_d(&Z, 1, &Z)) != MP_OKAY)                 goto cleanup;
-    /* Z = Z + 1 */
-  }
+   /* find a Z such that the Legendre symbol (Z|prime) == -1 */
+   if ((res = mp_set_int(&Z, 2)) != MP_OKAY)                     goto cleanup;
+   /* Z = 2 */
+   while (1) {
+      if ((res = mp_jacobi(&Z, prime, &legendre)) != MP_OKAY)     goto cleanup;
+      if (legendre == -1) break;
+      if ((res = mp_add_d(&Z, 1, &Z)) != MP_OKAY)                 goto cleanup;
+      /* Z = Z + 1 */
+   }
 
-  if ((res = mp_exptmod(&Z, &Q, prime, &C)) != MP_OKAY)         goto cleanup;
-  /* C = Z ^ Q mod prime */
-  if ((res = mp_add_d(&Q, 1, &t1)) != MP_OKAY)                  goto cleanup;
-  if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                    goto cleanup;
-  /* t1 = (Q + 1) / 2 */
-  if ((res = mp_exptmod(n, &t1, prime, &R)) != MP_OKAY)         goto cleanup;
-  /* R = n ^ ((Q + 1) / 2) mod prime */
-  if ((res = mp_exptmod(n, &Q, prime, &T)) != MP_OKAY)          goto cleanup;
-  /* T = n ^ Q mod prime */
-  if ((res = mp_copy(&S, &M)) != MP_OKAY)                       goto cleanup;
-  /* M = S */
-  if ((res = mp_set_int(&two, 2)) != MP_OKAY)                   goto cleanup;
+   if ((res = mp_exptmod(&Z, &Q, prime, &C)) != MP_OKAY)         goto cleanup;
+   /* C = Z ^ Q mod prime */
+   if ((res = mp_add_d(&Q, 1, &t1)) != MP_OKAY)                  goto cleanup;
+   if ((res = mp_div_2(&t1, &t1)) != MP_OKAY)                    goto cleanup;
+   /* t1 = (Q + 1) / 2 */
+   if ((res = mp_exptmod(n, &t1, prime, &R)) != MP_OKAY)         goto cleanup;
+   /* R = n ^ ((Q + 1) / 2) mod prime */
+   if ((res = mp_exptmod(n, &Q, prime, &T)) != MP_OKAY)          goto cleanup;
+   /* T = n ^ Q mod prime */
+   if ((res = mp_copy(&S, &M)) != MP_OKAY)                       goto cleanup;
+   /* M = S */
+   if ((res = mp_set_int(&two, 2)) != MP_OKAY)                   goto cleanup;
 
-  res = MP_VAL;
-  while (1) {
-    if ((res = mp_copy(&T, &t1)) != MP_OKAY)                    goto cleanup;
-    i = 0;
-    while (1) {
-      if (mp_cmp_d(&t1, 1) == MP_EQ) break;
-      if ((res = mp_exptmod(&t1, &two, prime, &t1)) != MP_OKAY) goto cleanup;
-      i++;
-    }
-    if (i == 0) {
-      if ((res = mp_copy(&R, ret)) != MP_OKAY)                  goto cleanup;
-      res = MP_OKAY;
-      goto cleanup;
-    }
-    if ((res = mp_sub_d(&M, i, &t1)) != MP_OKAY)                goto cleanup;
-    if ((res = mp_sub_d(&t1, 1, &t1)) != MP_OKAY)               goto cleanup;
-    if ((res = mp_exptmod(&two, &t1, prime, &t1)) != MP_OKAY)   goto cleanup;
-    /* t1 = 2 ^ (M - i - 1) */
-    if ((res = mp_exptmod(&C, &t1, prime, &t1)) != MP_OKAY)     goto cleanup;
-    /* t1 = C ^ (2 ^ (M - i - 1)) mod prime */
-    if ((res = mp_sqrmod(&t1, prime, &C)) != MP_OKAY)           goto cleanup;
-    /* C = (t1 * t1) mod prime */
-    if ((res = mp_mulmod(&R, &t1, prime, &R)) != MP_OKAY)       goto cleanup;
-    /* R = (R * t1) mod prime */
-    if ((res = mp_mulmod(&T, &C, prime, &T)) != MP_OKAY)        goto cleanup;
-    /* T = (T * C) mod prime */
-    mp_set(&M, i);
-    /* M = i */
-  }
+   res = MP_VAL;
+   while (1) {
+      if ((res = mp_copy(&T, &t1)) != MP_OKAY)                    goto cleanup;
+      i = 0;
+      while (1) {
+         if (mp_cmp_d(&t1, 1) == MP_EQ) break;
+         if ((res = mp_exptmod(&t1, &two, prime, &t1)) != MP_OKAY) goto cleanup;
+         i++;
+      }
+      if (i == 0) {
+         if ((res = mp_copy(&R, ret)) != MP_OKAY)                  goto cleanup;
+         res = MP_OKAY;
+         goto cleanup;
+      }
+      if ((res = mp_sub_d(&M, i, &t1)) != MP_OKAY)                goto cleanup;
+      if ((res = mp_sub_d(&t1, 1, &t1)) != MP_OKAY)               goto cleanup;
+      if ((res = mp_exptmod(&two, &t1, prime, &t1)) != MP_OKAY)   goto cleanup;
+      /* t1 = 2 ^ (M - i - 1) */
+      if ((res = mp_exptmod(&C, &t1, prime, &t1)) != MP_OKAY)     goto cleanup;
+      /* t1 = C ^ (2 ^ (M - i - 1)) mod prime */
+      if ((res = mp_sqrmod(&t1, prime, &C)) != MP_OKAY)           goto cleanup;
+      /* C = (t1 * t1) mod prime */
+      if ((res = mp_mulmod(&R, &t1, prime, &R)) != MP_OKAY)       goto cleanup;
+      /* R = (R * t1) mod prime */
+      if ((res = mp_mulmod(&T, &C, prime, &T)) != MP_OKAY)        goto cleanup;
+      /* T = (T * C) mod prime */
+      mp_set(&M, i);
+      /* M = i */
+   }
 
 cleanup:
-  mp_clear_multi(&t1, &C, &Q, &S, &Z, &M, &T, &R, &two, NULL);
-  return res;
+   mp_clear_multi(&t1, &C, &Q, &S, &Z, &M, &T, &R, &two, NULL);
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_sub.c b/libtommath/bn_mp_sub.c
index 2f73faa..75c7c2d 100644
--- a/libtommath/bn_mp_sub.c
+++ b/libtommath/bn_mp_sub.c
@@ -16,40 +16,39 @@
  */
 
 /* high level subtraction (handles signs) */
-int
-mp_sub (mp_int * a, mp_int * b, mp_int * c)
+int mp_sub(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     sa, sb, res;
+   int     sa, sb, res;
 
-  sa = a->sign;
-  sb = b->sign;
+   sa = a->sign;
+   sb = b->sign;
 
-  if (sa != sb) {
-    /* subtract a negative from a positive, OR */
-    /* subtract a positive from a negative. */
-    /* In either case, ADD their magnitudes, */
-    /* and use the sign of the first number. */
-    c->sign = sa;
-    res = s_mp_add (a, b, c);
-  } else {
-    /* subtract a positive from a positive, OR */
-    /* subtract a negative from a negative. */
-    /* First, take the difference between their */
-    /* magnitudes, then... */
-    if (mp_cmp_mag (a, b) != MP_LT) {
-      /* Copy the sign from the first */
+   if (sa != sb) {
+      /* subtract a negative from a positive, OR */
+      /* subtract a positive from a negative. */
+      /* In either case, ADD their magnitudes, */
+      /* and use the sign of the first number. */
       c->sign = sa;
-      /* The first has a larger or equal magnitude */
-      res = s_mp_sub (a, b, c);
-    } else {
-      /* The result has the *opposite* sign from */
-      /* the first number. */
-      c->sign = (sa == MP_ZPOS) ? MP_NEG : MP_ZPOS;
-      /* The second has a larger magnitude */
-      res = s_mp_sub (b, a, c);
-    }
-  }
-  return res;
+      res = s_mp_add(a, b, c);
+   } else {
+      /* subtract a positive from a positive, OR */
+      /* subtract a negative from a negative. */
+      /* First, take the difference between their */
+      /* magnitudes, then... */
+      if (mp_cmp_mag(a, b) != MP_LT) {
+         /* Copy the sign from the first */
+         c->sign = sa;
+         /* The first has a larger or equal magnitude */
+         res = s_mp_sub(a, b, c);
+      } else {
+         /* The result has the *opposite* sign from */
+         /* the first number. */
+         c->sign = (sa == MP_ZPOS) ? MP_NEG : MP_ZPOS;
+         /* The second has a larger magnitude */
+         res = s_mp_sub(b, a, c);
+      }
+   }
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_sub_d.c b/libtommath/bn_mp_sub_d.c
index 5e96030..7016abc 100644
--- a/libtommath/bn_mp_sub_d.c
+++ b/libtommath/bn_mp_sub_d.c
@@ -16,74 +16,73 @@
  */
 
 /* single digit subtraction */
-int
-mp_sub_d (mp_int * a, mp_digit b, mp_int * c)
+int mp_sub_d(mp_int *a, mp_digit b, mp_int *c)
 {
-  mp_digit *tmpa, *tmpc, mu;
-  int       res, ix, oldused;
+   mp_digit *tmpa, *tmpc, mu;
+   int       res, ix, oldused;
 
-  /* grow c as required */
-  if (c->alloc < (a->used + 1)) {
-     if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
-        return res;
-     }
-  }
+   /* 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 just do an unsigned
-   * addition [with fudged signs]
-   */
-  if (a->sign == MP_NEG) {
-     a->sign = MP_ZPOS;
-     res     = mp_add_d(a, b, c);
-     a->sign = c->sign = MP_NEG;
+   /* if a is negative just do an unsigned
+    * addition [with fudged signs]
+    */
+   if (a->sign == MP_NEG) {
+      a->sign = MP_ZPOS;
+      res     = mp_add_d(a, b, c);
+      a->sign = c->sign = MP_NEG;
 
-     /* clamp */
-     mp_clamp(c);
+      /* clamp */
+      mp_clamp(c);
 
-     return res;
-  }
+      return res;
+   }
 
-  /* setup regs */
-  oldused = c->used;
-  tmpa    = a->dp;
-  tmpc    = c->dp;
+   /* setup regs */
+   oldused = c->used;
+   tmpa    = a->dp;
+   tmpc    = c->dp;
 
-  /* if a <= b simply fix the single digit */
-  if (((a->used == 1) && (a->dp[0] <= b)) || (a->used == 0)) {
-     if (a->used == 1) {
-        *tmpc++ = b - *tmpa;
-     } else {
-        *tmpc++ = b;
-     }
-     ix      = 1;
+   /* if a <= b simply fix the single digit */
+   if (((a->used == 1) && (a->dp[0] <= b)) || (a->used == 0)) {
+      if (a->used == 1) {
+         *tmpc++ = b - *tmpa;
+      } else {
+         *tmpc++ = b;
+      }
+      ix      = 1;
 
-     /* negative/1digit */
-     c->sign = MP_NEG;
-     c->used = 1;
-  } else {
-     /* positive/size */
-     c->sign = MP_ZPOS;
-     c->used = a->used;
+      /* negative/1digit */
+      c->sign = MP_NEG;
+      c->used = 1;
+   } else {
+      /* positive/size */
+      c->sign = MP_ZPOS;
+      c->used = a->used;
 
-     /* subtract first digit */
-     *tmpc    = *tmpa++ - b;
-     mu       = *tmpc >> ((sizeof(mp_digit) * CHAR_BIT) - 1);
-     *tmpc++ &= MP_MASK;
+      /* subtract first digit */
+      *tmpc    = *tmpa++ - b;
+      mu       = *tmpc >> ((sizeof(mp_digit) * CHAR_BIT) - 1);
+      *tmpc++ &= MP_MASK;
 
-     /* handle rest of the digits */
-     for (ix = 1; ix < a->used; ix++) {
-        *tmpc    = *tmpa++ - mu;
-        mu       = *tmpc >> ((sizeof(mp_digit) * CHAR_BIT) - 1);
-        *tmpc++ &= MP_MASK;
-     }
-  }
+      /* handle rest of the digits */
+      for (ix = 1; ix < a->used; ix++) {
+         *tmpc    = *tmpa++ - mu;
+         mu       = *tmpc >> ((sizeof(mp_digit) * CHAR_BIT) - 1);
+         *tmpc++ &= MP_MASK;
+      }
+   }
 
-  /* zero excess digits */
-  while (ix++ < oldused) {
-     *tmpc++ = 0;
-  }
-  mp_clamp(c);
-  return MP_OKAY;
+   /* zero excess digits */
+   while (ix++ < oldused) {
+      *tmpc++ = 0;
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_submod.c b/libtommath/bn_mp_submod.c
index 138863c..510fb19 100644
--- a/libtommath/bn_mp_submod.c
+++ b/libtommath/bn_mp_submod.c
@@ -16,24 +16,23 @@
  */
 
 /* d = a - b (mod c) */
-int
-mp_submod (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+int mp_submod(mp_int *a, mp_int *b, mp_int *c, mp_int *d)
 {
-  int     res;
-  mp_int  t;
+   int     res;
+   mp_int  t;
 
 
-  if ((res = mp_init (&t)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init(&t)) != MP_OKAY) {
+      return res;
+   }
 
-  if ((res = mp_sub (a, b, &t)) != MP_OKAY) {
-    mp_clear (&t);
-    return res;
-  }
-  res = mp_mod (&t, c, d);
-  mp_clear (&t);
-  return res;
+   if ((res = mp_sub(a, b, &t)) != MP_OKAY) {
+      mp_clear(&t);
+      return res;
+   }
+   res = mp_mod(&t, c, d);
+   mp_clear(&t);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_mp_to_signed_bin.c b/libtommath/bn_mp_to_signed_bin.c
index c49c87d..615bf32 100644
--- a/libtommath/bn_mp_to_signed_bin.c
+++ b/libtommath/bn_mp_to_signed_bin.c
@@ -16,15 +16,15 @@
  */
 
 /* store in signed [big endian] format */
-int mp_to_signed_bin (mp_int * a, unsigned char *b)
+int mp_to_signed_bin(mp_int *a, unsigned char *b)
 {
-  int     res;
+   int     res;
 
-  if ((res = mp_to_unsigned_bin (a, b + 1)) != MP_OKAY) {
-    return res;
-  }
-  b[0] = (a->sign == MP_ZPOS) ? (unsigned char)0 : (unsigned char)1;
-  return MP_OKAY;
+   if ((res = mp_to_unsigned_bin(a, b + 1)) != MP_OKAY) {
+      return res;
+   }
+   b[0] = (a->sign == MP_ZPOS) ? (unsigned char)0 : (unsigned char)1;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_to_signed_bin_n.c b/libtommath/bn_mp_to_signed_bin_n.c
index dc5ec26..501f849 100644
--- a/libtommath/bn_mp_to_signed_bin_n.c
+++ b/libtommath/bn_mp_to_signed_bin_n.c
@@ -16,7 +16,7 @@
  */
 
 /* store in signed [big endian] format */
-int mp_to_signed_bin_n (mp_int * a, unsigned char *b, unsigned long *outlen)
+int mp_to_signed_bin_n(mp_int *a, unsigned char *b, unsigned long *outlen)
 {
    if (*outlen < (unsigned long)mp_signed_bin_size(a)) {
       return MP_VAL;
diff --git a/libtommath/bn_mp_to_unsigned_bin.c b/libtommath/bn_mp_to_unsigned_bin.c
index d249359..e103383 100644
--- a/libtommath/bn_mp_to_unsigned_bin.c
+++ b/libtommath/bn_mp_to_unsigned_bin.c
@@ -16,30 +16,30 @@
  */
 
 /* store in unsigned [big endian] format */
-int mp_to_unsigned_bin (mp_int * a, unsigned char *b)
+int mp_to_unsigned_bin(mp_int *a, unsigned char *b)
 {
-  int     x, res;
-  mp_int  t;
+   int     x, res;
+   mp_int  t;
 
-  if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+      return res;
+   }
 
-  x = 0;
-  while (mp_iszero (&t) == MP_NO) {
+   x = 0;
+   while (mp_iszero(&t) == MP_NO) {
 #ifndef MP_8BIT
-      b[x++] = (unsigned char) (t.dp[0] & 255);
+      b[x++] = (unsigned char)(t.dp[0] & 255);
 #else
-      b[x++] = (unsigned char) (t.dp[0] | ((t.dp[1] & 0x01) << 7));
+      b[x++] = (unsigned char)(t.dp[0] | ((t.dp[1] & 0x01) << 7));
 #endif
-    if ((res = mp_div_2d (&t, 8, &t, NULL)) != MP_OKAY) {
-      mp_clear (&t);
-      return res;
-    }
-  }
-  bn_reverse (b, x);
-  mp_clear (&t);
-  return MP_OKAY;
+      if ((res = mp_div_2d(&t, 8, &t, NULL)) != MP_OKAY) {
+         mp_clear(&t);
+         return res;
+      }
+   }
+   bn_reverse(b, x);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_to_unsigned_bin_n.c b/libtommath/bn_mp_to_unsigned_bin_n.c
index f671621..5ee28f1 100644
--- a/libtommath/bn_mp_to_unsigned_bin_n.c
+++ b/libtommath/bn_mp_to_unsigned_bin_n.c
@@ -16,7 +16,7 @@
  */
 
 /* store in unsigned [big endian] format */
-int mp_to_unsigned_bin_n (mp_int * a, unsigned char *b, unsigned long *outlen)
+int mp_to_unsigned_bin_n(mp_int *a, unsigned char *b, unsigned long *outlen)
 {
    if (*outlen < (unsigned long)mp_unsigned_bin_size(a)) {
       return MP_VAL;
diff --git a/libtommath/bn_mp_toom_mul.c b/libtommath/bn_mp_toom_mul.c
index 4a574fc..8b771bc 100644
--- a/libtommath/bn_mp_toom_mul.c
+++ b/libtommath/bn_mp_toom_mul.c
@@ -24,259 +24,259 @@
 */
 int mp_toom_mul(mp_int *a, mp_int *b, mp_int *c)
 {
-    mp_int w0, w1, w2, w3, w4, tmp1, tmp2, a0, a1, a2, b0, b1, b2;
-    int res, B;
+   mp_int w0, w1, w2, w3, w4, tmp1, tmp2, a0, a1, a2, b0, b1, b2;
+   int res, B;
 
-    /* init temps */
-    if ((res = mp_init_multi(&w0, &w1, &w2, &w3, &w4,
-                             &a0, &a1, &a2, &b0, &b1,
-                             &b2, &tmp1, &tmp2, NULL)) != MP_OKAY) {
-       return res;
-    }
+   /* init temps */
+   if ((res = mp_init_multi(&w0, &w1, &w2, &w3, &w4,
+                            &a0, &a1, &a2, &b0, &b1,
+                            &b2, &tmp1, &tmp2, NULL)) != MP_OKAY) {
+      return res;
+   }
 
-    /* B */
-    B = MIN(a->used, b->used) / 3;
+   /* B */
+   B = MIN(a->used, b->used) / 3;
 
-    /* a = a2 * B**2 + a1 * B + a0 */
-    if ((res = mp_mod_2d(a, DIGIT_BIT * B, &a0)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* a = a2 * B**2 + a1 * B + a0 */
+   if ((res = mp_mod_2d(a, DIGIT_BIT * B, &a0)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_copy(a, &a1)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&a1, B);
-    if ((res = mp_mod_2d(&a1, DIGIT_BIT * B, &a1)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_copy(a, &a1)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&a1, B);
+   if ((res = mp_mod_2d(&a1, DIGIT_BIT * B, &a1)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_copy(a, &a2)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&a2, B*2);
+   if ((res = mp_copy(a, &a2)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&a2, B*2);
 
-    /* b = b2 * B**2 + b1 * B + b0 */
-    if ((res = mp_mod_2d(b, DIGIT_BIT * B, &b0)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* b = b2 * B**2 + b1 * B + b0 */
+   if ((res = mp_mod_2d(b, DIGIT_BIT * B, &b0)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_copy(b, &b1)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&b1, B);
-    (void)mp_mod_2d(&b1, DIGIT_BIT * B, &b1);
+   if ((res = mp_copy(b, &b1)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&b1, B);
+   (void)mp_mod_2d(&b1, DIGIT_BIT * B, &b1);
 
-    if ((res = mp_copy(b, &b2)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&b2, B*2);
+   if ((res = mp_copy(b, &b2)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&b2, B*2);
 
-    /* w0 = a0*b0 */
-    if ((res = mp_mul(&a0, &b0, &w0)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* w0 = a0*b0 */
+   if ((res = mp_mul(&a0, &b0, &w0)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    /* w4 = a2 * b2 */
-    if ((res = mp_mul(&a2, &b2, &w4)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* w4 = a2 * b2 */
+   if ((res = mp_mul(&a2, &b2, &w4)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    /* w1 = (a2 + 2(a1 + 2a0))(b2 + 2(b1 + 2b0)) */
-    if ((res = mp_mul_2(&a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a2, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* w1 = (a2 + 2(a1 + 2a0))(b2 + 2(b1 + 2b0)) */
+   if ((res = mp_mul_2(&a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a2, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_mul_2(&b0, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp2, &b2, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_mul_2(&b0, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp2, &b2, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_mul(&tmp1, &tmp2, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_mul(&tmp1, &tmp2, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    /* w3 = (a0 + 2(a1 + 2a2))(b0 + 2(b1 + 2b2)) */
-    if ((res = mp_mul_2(&a2, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* w3 = (a0 + 2(a1 + 2a2))(b0 + 2(b1 + 2b2)) */
+   if ((res = mp_mul_2(&a2, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_mul_2(&b2, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_mul_2(&b2, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp2, &b1, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp2, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_mul(&tmp1, &tmp2, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_mul(&tmp1, &tmp2, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
 
 
-    /* w2 = (a2 + a1 + a0)(b2 + b1 + b0) */
-    if ((res = mp_add(&a2, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&b2, &b1, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul(&tmp1, &tmp2, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* w2 = (a2 + a1 + a0)(b2 + b1 + b0) */
+   if ((res = mp_add(&a2, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&b2, &b1, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp2, &b0, &tmp2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul(&tmp1, &tmp2, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    /* now solve the matrix
+   /* now solve the matrix
 
-       0  0  0  0  1
-       1  2  4  8  16
-       1  1  1  1  1
-       16 8  4  2  1
-       1  0  0  0  0
+      0  0  0  0  1
+      1  2  4  8  16
+      1  1  1  1  1
+      16 8  4  2  1
+      1  0  0  0  0
 
-       using 12 subtractions, 4 shifts,
-              2 small divisions and 1 small multiplication
-     */
+      using 12 subtractions, 4 shifts,
+             2 small divisions and 1 small multiplication
+    */
 
-    /* r1 - r4 */
-    if ((res = mp_sub(&w1, &w4, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r0 */
-    if ((res = mp_sub(&w3, &w0, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1/2 */
-    if ((res = mp_div_2(&w1, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3/2 */
-    if ((res = mp_div_2(&w3, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r2 - r0 - r4 */
-    if ((res = mp_sub(&w2, &w0, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w4, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - r2 */
-    if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r2 */
-    if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - 8r0 */
-    if ((res = mp_mul_2d(&w0, 3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w1, &tmp1, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - 8r4 */
-    if ((res = mp_mul_2d(&w4, 3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w3, &tmp1, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* 3r2 - r1 - r3 */
-    if ((res = mp_mul_d(&w2, 3, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w1, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w3, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - r2 */
-    if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r2 */
-    if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1/3 */
-    if ((res = mp_div_3(&w1, &w1, NULL)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3/3 */
-    if ((res = mp_div_3(&w3, &w3, NULL)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* r1 - r4 */
+   if ((res = mp_sub(&w1, &w4, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r0 */
+   if ((res = mp_sub(&w3, &w0, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1/2 */
+   if ((res = mp_div_2(&w1, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3/2 */
+   if ((res = mp_div_2(&w3, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r2 - r0 - r4 */
+   if ((res = mp_sub(&w2, &w0, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w4, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - r2 */
+   if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r2 */
+   if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - 8r0 */
+   if ((res = mp_mul_2d(&w0, 3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w1, &tmp1, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - 8r4 */
+   if ((res = mp_mul_2d(&w4, 3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w3, &tmp1, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* 3r2 - r1 - r3 */
+   if ((res = mp_mul_d(&w2, 3, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w1, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w3, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - r2 */
+   if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r2 */
+   if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1/3 */
+   if ((res = mp_div_3(&w1, &w1, NULL)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3/3 */
+   if ((res = mp_div_3(&w3, &w3, NULL)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    /* at this point shift W[n] by B*n */
-    if ((res = mp_lshd(&w1, 1*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w2, 2*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w3, 3*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w4, 4*B)) != MP_OKAY) {
-       goto ERR;
-    }
+   /* at this point shift W[n] by B*n */
+   if ((res = mp_lshd(&w1, 1*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w2, 2*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w3, 3*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w4, 4*B)) != MP_OKAY) {
+      goto ERR;
+   }
 
-    if ((res = mp_add(&w0, &w1, c)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&w2, &w3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&w4, &tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, c, c)) != MP_OKAY) {
-       goto ERR;
-    }
+   if ((res = mp_add(&w0, &w1, c)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&w2, &w3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&w4, &tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, c, c)) != MP_OKAY) {
+      goto ERR;
+   }
 
 ERR:
-    mp_clear_multi(&w0, &w1, &w2, &w3, &w4,
-                   &a0, &a1, &a2, &b0, &b1,
-                   &b2, &tmp1, &tmp2, NULL);
-    return res;
+   mp_clear_multi(&w0, &w1, &w2, &w3, &w4,
+                  &a0, &a1, &a2, &b0, &b1,
+                  &b2, &tmp1, &tmp2, NULL);
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_toom_sqr.c b/libtommath/bn_mp_toom_sqr.c
index 0a38192..5e1e452 100644
--- a/libtommath/bn_mp_toom_sqr.c
+++ b/libtommath/bn_mp_toom_sqr.c
@@ -16,209 +16,208 @@
  */
 
 /* squaring using Toom-Cook 3-way algorithm */
-int
-mp_toom_sqr(mp_int *a, mp_int *b)
+int mp_toom_sqr(mp_int *a, mp_int *b)
 {
-    mp_int w0, w1, w2, w3, w4, tmp1, a0, a1, a2;
-    int res, B;
-
-    /* init temps */
-    if ((res = mp_init_multi(&w0, &w1, &w2, &w3, &w4, &a0, &a1, &a2, &tmp1, NULL)) != MP_OKAY) {
-       return res;
-    }
-
-    /* B */
-    B = a->used / 3;
-
-    /* a = a2 * B**2 + a1 * B + a0 */
-    if ((res = mp_mod_2d(a, DIGIT_BIT * B, &a0)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    if ((res = mp_copy(a, &a1)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&a1, B);
-    if ((res = mp_mod_2d(&a1, DIGIT_BIT * B, &a1)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    if ((res = mp_copy(a, &a2)) != MP_OKAY) {
-       goto ERR;
-    }
-    mp_rshd(&a2, B*2);
-
-    /* w0 = a0*a0 */
-    if ((res = mp_sqr(&a0, &w0)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    /* w4 = a2 * a2 */
-    if ((res = mp_sqr(&a2, &w4)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    /* w1 = (a2 + 2(a1 + 2a0))**2 */
-    if ((res = mp_mul_2(&a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a2, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    if ((res = mp_sqr(&tmp1, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    /* w3 = (a0 + 2(a1 + 2a2))**2 */
-    if ((res = mp_mul_2(&a2, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    if ((res = mp_sqr(&tmp1, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-
-
-    /* w2 = (a2 + a1 + a0)**2 */
-    if ((res = mp_add(&a2, &a1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sqr(&tmp1, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    /* now solve the matrix
-
-       0  0  0  0  1
-       1  2  4  8  16
-       1  1  1  1  1
-       16 8  4  2  1
-       1  0  0  0  0
-
-       using 12 subtractions, 4 shifts, 2 small divisions and 1 small multiplication.
-     */
-
-    /* r1 - r4 */
-    if ((res = mp_sub(&w1, &w4, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r0 */
-    if ((res = mp_sub(&w3, &w0, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1/2 */
-    if ((res = mp_div_2(&w1, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3/2 */
-    if ((res = mp_div_2(&w3, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r2 - r0 - r4 */
-    if ((res = mp_sub(&w2, &w0, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w4, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - r2 */
-    if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r2 */
-    if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - 8r0 */
-    if ((res = mp_mul_2d(&w0, 3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w1, &tmp1, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - 8r4 */
-    if ((res = mp_mul_2d(&w4, 3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w3, &tmp1, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* 3r2 - r1 - r3 */
-    if ((res = mp_mul_d(&w2, 3, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w1, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_sub(&w2, &w3, &w2)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1 - r2 */
-    if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3 - r2 */
-    if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r1/3 */
-    if ((res = mp_div_3(&w1, &w1, NULL)) != MP_OKAY) {
-       goto ERR;
-    }
-    /* r3/3 */
-    if ((res = mp_div_3(&w3, &w3, NULL)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    /* at this point shift W[n] by B*n */
-    if ((res = mp_lshd(&w1, 1*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w2, 2*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w3, 3*B)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_lshd(&w4, 4*B)) != MP_OKAY) {
-       goto ERR;
-    }
-
-    if ((res = mp_add(&w0, &w1, b)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&w2, &w3, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&w4, &tmp1, &tmp1)) != MP_OKAY) {
-       goto ERR;
-    }
-    if ((res = mp_add(&tmp1, b, b)) != MP_OKAY) {
-       goto ERR;
-    }
+   mp_int w0, w1, w2, w3, w4, tmp1, a0, a1, a2;
+   int res, B;
+
+   /* init temps */
+   if ((res = mp_init_multi(&w0, &w1, &w2, &w3, &w4, &a0, &a1, &a2, &tmp1, NULL)) != MP_OKAY) {
+      return res;
+   }
+
+   /* B */
+   B = a->used / 3;
+
+   /* a = a2 * B**2 + a1 * B + a0 */
+   if ((res = mp_mod_2d(a, DIGIT_BIT * B, &a0)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   if ((res = mp_copy(a, &a1)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&a1, B);
+   if ((res = mp_mod_2d(&a1, DIGIT_BIT * B, &a1)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   if ((res = mp_copy(a, &a2)) != MP_OKAY) {
+      goto ERR;
+   }
+   mp_rshd(&a2, B*2);
+
+   /* w0 = a0*a0 */
+   if ((res = mp_sqr(&a0, &w0)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   /* w4 = a2 * a2 */
+   if ((res = mp_sqr(&a2, &w4)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   /* w1 = (a2 + 2(a1 + 2a0))**2 */
+   if ((res = mp_mul_2(&a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a2, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   if ((res = mp_sqr(&tmp1, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   /* w3 = (a0 + 2(a1 + 2a2))**2 */
+   if ((res = mp_mul_2(&a2, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_mul_2(&tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   if ((res = mp_sqr(&tmp1, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+
+
+   /* w2 = (a2 + a1 + a0)**2 */
+   if ((res = mp_add(&a2, &a1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, &a0, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sqr(&tmp1, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   /* now solve the matrix
+
+      0  0  0  0  1
+      1  2  4  8  16
+      1  1  1  1  1
+      16 8  4  2  1
+      1  0  0  0  0
+
+      using 12 subtractions, 4 shifts, 2 small divisions and 1 small multiplication.
+    */
+
+   /* r1 - r4 */
+   if ((res = mp_sub(&w1, &w4, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r0 */
+   if ((res = mp_sub(&w3, &w0, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1/2 */
+   if ((res = mp_div_2(&w1, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3/2 */
+   if ((res = mp_div_2(&w3, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r2 - r0 - r4 */
+   if ((res = mp_sub(&w2, &w0, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w4, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - r2 */
+   if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r2 */
+   if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - 8r0 */
+   if ((res = mp_mul_2d(&w0, 3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w1, &tmp1, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - 8r4 */
+   if ((res = mp_mul_2d(&w4, 3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w3, &tmp1, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* 3r2 - r1 - r3 */
+   if ((res = mp_mul_d(&w2, 3, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w1, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_sub(&w2, &w3, &w2)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1 - r2 */
+   if ((res = mp_sub(&w1, &w2, &w1)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3 - r2 */
+   if ((res = mp_sub(&w3, &w2, &w3)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r1/3 */
+   if ((res = mp_div_3(&w1, &w1, NULL)) != MP_OKAY) {
+      goto ERR;
+   }
+   /* r3/3 */
+   if ((res = mp_div_3(&w3, &w3, NULL)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   /* at this point shift W[n] by B*n */
+   if ((res = mp_lshd(&w1, 1*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w2, 2*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w3, 3*B)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_lshd(&w4, 4*B)) != MP_OKAY) {
+      goto ERR;
+   }
+
+   if ((res = mp_add(&w0, &w1, b)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&w2, &w3, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&w4, &tmp1, &tmp1)) != MP_OKAY) {
+      goto ERR;
+   }
+   if ((res = mp_add(&tmp1, b, b)) != MP_OKAY) {
+      goto ERR;
+   }
 
 ERR:
-    mp_clear_multi(&w0, &w1, &w2, &w3, &w4, &a0, &a1, &a2, &tmp1, NULL);
-    return res;
+   mp_clear_multi(&w0, &w1, &w2, &w3, &w4, &a0, &a1, &a2, &tmp1, NULL);
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_toradix.c b/libtommath/bn_mp_toradix.c
index 3337765..5016219 100644
--- a/libtommath/bn_mp_toradix.c
+++ b/libtommath/bn_mp_toradix.c
@@ -16,56 +16,56 @@
  */
 
 /* stores a bignum as a ASCII string in a given radix (2..64) */
-int mp_toradix (mp_int * a, char *str, int radix)
+int mp_toradix(mp_int *a, char *str, int radix)
 {
-  int     res, digs;
-  mp_int  t;
-  mp_digit d;
-  char   *_s = str;
+   int     res, digs;
+   mp_int  t;
+   mp_digit d;
+   char   *_s = str;
 
-  /* check range of the radix */
-  if ((radix < 2) || (radix > 64)) {
-    return MP_VAL;
-  }
+   /* check range of the radix */
+   if ((radix < 2) || (radix > 64)) {
+      return MP_VAL;
+   }
 
-  /* quick out if its zero */
-  if (mp_iszero(a) == MP_YES) {
-     *str++ = '0';
-     *str = '\0';
-     return MP_OKAY;
-  }
+   /* quick out if its zero */
+   if (mp_iszero(a) == MP_YES) {
+      *str++ = '0';
+      *str = '\0';
+      return MP_OKAY;
+   }
 
-  if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+      return res;
+   }
 
-  /* if it is negative output a - */
-  if (t.sign == MP_NEG) {
-    ++_s;
-    *str++ = '-';
-    t.sign = MP_ZPOS;
-  }
+   /* if it is negative output a - */
+   if (t.sign == MP_NEG) {
+      ++_s;
+      *str++ = '-';
+      t.sign = MP_ZPOS;
+   }
 
-  digs = 0;
-  while (mp_iszero (&t) == MP_NO) {
-    if ((res = mp_div_d (&t, (mp_digit) radix, &t, &d)) != MP_OKAY) {
-      mp_clear (&t);
-      return res;
-    }
-    *str++ = mp_s_rmap[d];
-    ++digs;
-  }
+   digs = 0;
+   while (mp_iszero(&t) == MP_NO) {
+      if ((res = mp_div_d(&t, (mp_digit)radix, &t, &d)) != MP_OKAY) {
+         mp_clear(&t);
+         return res;
+      }
+      *str++ = mp_s_rmap[d];
+      ++digs;
+   }
 
-  /* reverse the digits of the string.  In this case _s points
-   * to the first digit [exluding the sign] of the number]
-   */
-  bn_reverse ((unsigned char *)_s, digs);
+   /* reverse the digits of the string.  In this case _s points
+    * to the first digit [exluding the sign] of the number]
+    */
+   bn_reverse((unsigned char *)_s, digs);
 
-  /* append a NULL so the string is properly terminated */
-  *str = '\0';
+   /* append a NULL so the string is properly terminated */
+   *str = '\0';
 
-  mp_clear (&t);
-  return MP_OKAY;
+   mp_clear(&t);
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_toradix_n.c b/libtommath/bn_mp_toradix_n.c
index ae24ada..287d3f8 100644
--- a/libtommath/bn_mp_toradix_n.c
+++ b/libtommath/bn_mp_toradix_n.c
@@ -15,70 +15,70 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* stores a bignum as a ASCII string in a given radix (2..64) 
+/* stores a bignum as a ASCII string in a given radix (2..64)
  *
- * Stores upto maxlen-1 chars and always a NULL byte 
+ * Stores upto maxlen-1 chars and always a NULL byte
  */
-int mp_toradix_n(mp_int * a, char *str, int radix, int maxlen)
+int mp_toradix_n(mp_int *a, char *str, int radix, int maxlen)
 {
-  int     res, digs;
-  mp_int  t;
-  mp_digit d;
-  char   *_s = str;
+   int     res, digs;
+   mp_int  t;
+   mp_digit d;
+   char   *_s = str;
 
-  /* check range of the maxlen, radix */
-  if ((maxlen < 2) || (radix < 2) || (radix > 64)) {
-    return MP_VAL;
-  }
+   /* check range of the maxlen, radix */
+   if ((maxlen < 2) || (radix < 2) || (radix > 64)) {
+      return MP_VAL;
+   }
 
-  /* quick out if its zero */
-  if (mp_iszero(a) == MP_YES) {
-     *str++ = '0';
-     *str = '\0';
-     return MP_OKAY;
-  }
+   /* quick out if its zero */
+   if (mp_iszero(a) == MP_YES) {
+      *str++ = '0';
+      *str = '\0';
+      return MP_OKAY;
+   }
 
-  if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-    return res;
-  }
+   if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+      return res;
+   }
 
-  /* if it is negative output a - */
-  if (t.sign == MP_NEG) {
-    /* we have to reverse our digits later... but not the - sign!! */
-    ++_s;
+   /* if it is negative output a - */
+   if (t.sign == MP_NEG) {
+      /* we have to reverse our digits later... but not the - sign!! */
+      ++_s;
 
-    /* store the flag and mark the number as positive */
-    *str++ = '-';
-    t.sign = MP_ZPOS;
- 
-    /* subtract a char */
-    --maxlen;
-  }
+      /* store the flag and mark the number as positive */
+      *str++ = '-';
+      t.sign = MP_ZPOS;
 
-  digs = 0;
-  while (mp_iszero (&t) == MP_NO) {
-    if (--maxlen < 1) {
-       /* no more room */
-       break;
-    }
-    if ((res = mp_div_d (&t, (mp_digit) radix, &t, &d)) != MP_OKAY) {
-      mp_clear (&t);
-      return res;
-    }
-    *str++ = mp_s_rmap[d];
-    ++digs;
-  }
+      /* subtract a char */
+      --maxlen;
+   }
+
+   digs = 0;
+   while (mp_iszero(&t) == MP_NO) {
+      if (--maxlen < 1) {
+         /* no more room */
+         break;
+      }
+      if ((res = mp_div_d(&t, (mp_digit)radix, &t, &d)) != MP_OKAY) {
+         mp_clear(&t);
+         return res;
+      }
+      *str++ = mp_s_rmap[d];
+      ++digs;
+   }
 
-  /* reverse the digits of the string.  In this case _s points
-   * to the first digit [exluding the sign] of the number
-   */
-  bn_reverse ((unsigned char *)_s, digs);
+   /* reverse the digits of the string.  In this case _s points
+    * to the first digit [exluding the sign] of the number
+    */
+   bn_reverse((unsigned char *)_s, digs);
 
-  /* append a NULL so the string is properly terminated */
-  *str = '\0';
+   /* append a NULL so the string is properly terminated */
+   *str = '\0';
 
-  mp_clear (&t);
-  return MP_OKAY;
+   mp_clear(&t);
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_unsigned_bin_size.c b/libtommath/bn_mp_unsigned_bin_size.c
index f46d0ba..174cdc1 100644
--- a/libtommath/bn_mp_unsigned_bin_size.c
+++ b/libtommath/bn_mp_unsigned_bin_size.c
@@ -16,10 +16,10 @@
  */
 
 /* get the size for an unsigned equivalent */
-int mp_unsigned_bin_size (mp_int * a)
+int mp_unsigned_bin_size(mp_int *a)
 {
-  int     size = mp_count_bits (a);
-  return (size / 8) + (((size & 7) != 0) ? 1 : 0);
+   int     size = mp_count_bits(a);
+   return (size / 8) + (((size & 7) != 0) ? 1 : 0);
 }
 #endif
 
diff --git a/libtommath/bn_mp_xor.c b/libtommath/bn_mp_xor.c
index f51fc8e..4224a7f 100644
--- a/libtommath/bn_mp_xor.c
+++ b/libtommath/bn_mp_xor.c
@@ -16,33 +16,32 @@
  */
 
 /* XOR two ints together */
-int
-mp_xor (mp_int * a, mp_int * b, mp_int * c)
+int mp_xor(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     res, ix, px;
-  mp_int  t, *x;
+   int     res, ix, px;
+   mp_int  t, *x;
 
-  if (a->used > b->used) {
-    if ((res = mp_init_copy (&t, a)) != MP_OKAY) {
-      return res;
-    }
-    px = b->used;
-    x = b;
-  } else {
-    if ((res = mp_init_copy (&t, b)) != MP_OKAY) {
-      return res;
-    }
-    px = a->used;
-    x = a;
-  }
+   if (a->used > b->used) {
+      if ((res = mp_init_copy(&t, a)) != MP_OKAY) {
+         return res;
+      }
+      px = b->used;
+      x = b;
+   } else {
+      if ((res = mp_init_copy(&t, b)) != MP_OKAY) {
+         return res;
+      }
+      px = a->used;
+      x = a;
+   }
 
-  for (ix = 0; ix < px; ix++) {
-     t.dp[ix] ^= x->dp[ix];
-  }
-  mp_clamp (&t);
-  mp_exch (c, &t);
-  mp_clear (&t);
-  return MP_OKAY;
+   for (ix = 0; ix < px; ix++) {
+      t.dp[ix] ^= x->dp[ix];
+   }
+   mp_clamp(&t);
+   mp_exch(c, &t);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_zero.c b/libtommath/bn_mp_zero.c
index a7d59e4..08aac2a 100644
--- a/libtommath/bn_mp_zero.c
+++ b/libtommath/bn_mp_zero.c
@@ -16,18 +16,18 @@
  */
 
 /* set to zero */
-void mp_zero (mp_int * a)
+void mp_zero(mp_int *a)
 {
-  int       n;
-  mp_digit *tmp;
+   int       n;
+   mp_digit *tmp;
 
-  a->sign = MP_ZPOS;
-  a->used = 0;
+   a->sign = MP_ZPOS;
+   a->used = 0;
 
-  tmp = a->dp;
-  for (n = 0; n < a->alloc; n++) {
-     *tmp++ = 0;
-  }
+   tmp = a->dp;
+   for (n = 0; n < a->alloc; n++) {
+      *tmp++ = 0;
+   }
 }
 #endif
 
diff --git a/libtommath/bn_prime_tab.c b/libtommath/bn_prime_tab.c
index 5252130..c8fadcd 100644
--- a/libtommath/bn_prime_tab.c
+++ b/libtommath/bn_prime_tab.c
@@ -15,43 +15,43 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 const mp_digit ltm_prime_tab[] = {
-  0x0002, 0x0003, 0x0005, 0x0007, 0x000B, 0x000D, 0x0011, 0x0013,
-  0x0017, 0x001D, 0x001F, 0x0025, 0x0029, 0x002B, 0x002F, 0x0035,
-  0x003B, 0x003D, 0x0043, 0x0047, 0x0049, 0x004F, 0x0053, 0x0059,
-  0x0061, 0x0065, 0x0067, 0x006B, 0x006D, 0x0071, 0x007F,
+   0x0002, 0x0003, 0x0005, 0x0007, 0x000B, 0x000D, 0x0011, 0x0013,
+   0x0017, 0x001D, 0x001F, 0x0025, 0x0029, 0x002B, 0x002F, 0x0035,
+   0x003B, 0x003D, 0x0043, 0x0047, 0x0049, 0x004F, 0x0053, 0x0059,
+   0x0061, 0x0065, 0x0067, 0x006B, 0x006D, 0x0071, 0x007F,
 #ifndef MP_8BIT
-  0x0083,
-  0x0089, 0x008B, 0x0095, 0x0097, 0x009D, 0x00A3, 0x00A7, 0x00AD,
-  0x00B3, 0x00B5, 0x00BF, 0x00C1, 0x00C5, 0x00C7, 0x00D3, 0x00DF,
-  0x00E3, 0x00E5, 0x00E9, 0x00EF, 0x00F1, 0x00FB, 0x0101, 0x0107,
-  0x010D, 0x010F, 0x0115, 0x0119, 0x011B, 0x0125, 0x0133, 0x0137,
+   0x0083,
+   0x0089, 0x008B, 0x0095, 0x0097, 0x009D, 0x00A3, 0x00A7, 0x00AD,
+   0x00B3, 0x00B5, 0x00BF, 0x00C1, 0x00C5, 0x00C7, 0x00D3, 0x00DF,
+   0x00E3, 0x00E5, 0x00E9, 0x00EF, 0x00F1, 0x00FB, 0x0101, 0x0107,
+   0x010D, 0x010F, 0x0115, 0x0119, 0x011B, 0x0125, 0x0133, 0x0137,
 
-  0x0139, 0x013D, 0x014B, 0x0151, 0x015B, 0x015D, 0x0161, 0x0167,
-  0x016F, 0x0175, 0x017B, 0x017F, 0x0185, 0x018D, 0x0191, 0x0199,
-  0x01A3, 0x01A5, 0x01AF, 0x01B1, 0x01B7, 0x01BB, 0x01C1, 0x01C9,
-  0x01CD, 0x01CF, 0x01D3, 0x01DF, 0x01E7, 0x01EB, 0x01F3, 0x01F7,
-  0x01FD, 0x0209, 0x020B, 0x021D, 0x0223, 0x022D, 0x0233, 0x0239,
-  0x023B, 0x0241, 0x024B, 0x0251, 0x0257, 0x0259, 0x025F, 0x0265,
-  0x0269, 0x026B, 0x0277, 0x0281, 0x0283, 0x0287, 0x028D, 0x0293,
-  0x0295, 0x02A1, 0x02A5, 0x02AB, 0x02B3, 0x02BD, 0x02C5, 0x02CF,
+   0x0139, 0x013D, 0x014B, 0x0151, 0x015B, 0x015D, 0x0161, 0x0167,
+   0x016F, 0x0175, 0x017B, 0x017F, 0x0185, 0x018D, 0x0191, 0x0199,
+   0x01A3, 0x01A5, 0x01AF, 0x01B1, 0x01B7, 0x01BB, 0x01C1, 0x01C9,
+   0x01CD, 0x01CF, 0x01D3, 0x01DF, 0x01E7, 0x01EB, 0x01F3, 0x01F7,
+   0x01FD, 0x0209, 0x020B, 0x021D, 0x0223, 0x022D, 0x0233, 0x0239,
+   0x023B, 0x0241, 0x024B, 0x0251, 0x0257, 0x0259, 0x025F, 0x0265,
+   0x0269, 0x026B, 0x0277, 0x0281, 0x0283, 0x0287, 0x028D, 0x0293,
+   0x0295, 0x02A1, 0x02A5, 0x02AB, 0x02B3, 0x02BD, 0x02C5, 0x02CF,
 
-  0x02D7, 0x02DD, 0x02E3, 0x02E7, 0x02EF, 0x02F5, 0x02F9, 0x0301,
-  0x0305, 0x0313, 0x031D, 0x0329, 0x032B, 0x0335, 0x0337, 0x033B,
-  0x033D, 0x0347, 0x0355, 0x0359, 0x035B, 0x035F, 0x036D, 0x0371,
-  0x0373, 0x0377, 0x038B, 0x038F, 0x0397, 0x03A1, 0x03A9, 0x03AD,
-  0x03B3, 0x03B9, 0x03C7, 0x03CB, 0x03D1, 0x03D7, 0x03DF, 0x03E5,
-  0x03F1, 0x03F5, 0x03FB, 0x03FD, 0x0407, 0x0409, 0x040F, 0x0419,
-  0x041B, 0x0425, 0x0427, 0x042D, 0x043F, 0x0443, 0x0445, 0x0449,
-  0x044F, 0x0455, 0x045D, 0x0463, 0x0469, 0x047F, 0x0481, 0x048B,
+   0x02D7, 0x02DD, 0x02E3, 0x02E7, 0x02EF, 0x02F5, 0x02F9, 0x0301,
+   0x0305, 0x0313, 0x031D, 0x0329, 0x032B, 0x0335, 0x0337, 0x033B,
+   0x033D, 0x0347, 0x0355, 0x0359, 0x035B, 0x035F, 0x036D, 0x0371,
+   0x0373, 0x0377, 0x038B, 0x038F, 0x0397, 0x03A1, 0x03A9, 0x03AD,
+   0x03B3, 0x03B9, 0x03C7, 0x03CB, 0x03D1, 0x03D7, 0x03DF, 0x03E5,
+   0x03F1, 0x03F5, 0x03FB, 0x03FD, 0x0407, 0x0409, 0x040F, 0x0419,
+   0x041B, 0x0425, 0x0427, 0x042D, 0x043F, 0x0443, 0x0445, 0x0449,
+   0x044F, 0x0455, 0x045D, 0x0463, 0x0469, 0x047F, 0x0481, 0x048B,
 
-  0x0493, 0x049D, 0x04A3, 0x04A9, 0x04B1, 0x04BD, 0x04C1, 0x04C7,
-  0x04CD, 0x04CF, 0x04D5, 0x04E1, 0x04EB, 0x04FD, 0x04FF, 0x0503,
-  0x0509, 0x050B, 0x0511, 0x0515, 0x0517, 0x051B, 0x0527, 0x0529,
-  0x052F, 0x0551, 0x0557, 0x055D, 0x0565, 0x0577, 0x0581, 0x058F,
-  0x0593, 0x0595, 0x0599, 0x059F, 0x05A7, 0x05AB, 0x05AD, 0x05B3,
-  0x05BF, 0x05C9, 0x05CB, 0x05CF, 0x05D1, 0x05D5, 0x05DB, 0x05E7,
-  0x05F3, 0x05FB, 0x0607, 0x060D, 0x0611, 0x0617, 0x061F, 0x0623,
-  0x062B, 0x062F, 0x063D, 0x0641, 0x0647, 0x0649, 0x064D, 0x0653
+   0x0493, 0x049D, 0x04A3, 0x04A9, 0x04B1, 0x04BD, 0x04C1, 0x04C7,
+   0x04CD, 0x04CF, 0x04D5, 0x04E1, 0x04EB, 0x04FD, 0x04FF, 0x0503,
+   0x0509, 0x050B, 0x0511, 0x0515, 0x0517, 0x051B, 0x0527, 0x0529,
+   0x052F, 0x0551, 0x0557, 0x055D, 0x0565, 0x0577, 0x0581, 0x058F,
+   0x0593, 0x0595, 0x0599, 0x059F, 0x05A7, 0x05AB, 0x05AD, 0x05B3,
+   0x05BF, 0x05C9, 0x05CB, 0x05CF, 0x05D1, 0x05D5, 0x05DB, 0x05E7,
+   0x05F3, 0x05FB, 0x0607, 0x060D, 0x0611, 0x0617, 0x061F, 0x0623,
+   0x062B, 0x062F, 0x063D, 0x0641, 0x0647, 0x0649, 0x064D, 0x0653
 #endif
 };
 #endif
diff --git a/libtommath/bn_reverse.c b/libtommath/bn_reverse.c
index dc87a4e..71e3d03 100644
--- a/libtommath/bn_reverse.c
+++ b/libtommath/bn_reverse.c
@@ -16,21 +16,20 @@
  */
 
 /* reverse an array, used for radix code */
-void
-bn_reverse (unsigned char *s, int len)
+void bn_reverse(unsigned char *s, int len)
 {
-  int     ix, iy;
-  unsigned char t;
+   int     ix, iy;
+   unsigned char t;
 
-  ix = 0;
-  iy = len - 1;
-  while (ix < iy) {
-    t     = s[ix];
-    s[ix] = s[iy];
-    s[iy] = t;
-    ++ix;
-    --iy;
-  }
+   ix = 0;
+   iy = len - 1;
+   while (ix < iy) {
+      t     = s[ix];
+      s[ix] = s[iy];
+      s[iy] = t;
+      ++ix;
+      --iy;
+   }
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_add.c b/libtommath/bn_s_mp_add.c
index 7a100e8..6ba65da 100644
--- a/libtommath/bn_s_mp_add.c
+++ b/libtommath/bn_s_mp_add.c
@@ -16,91 +16,90 @@
  */
 
 /* low level addition, based on HAC pp.594, Algorithm 14.7 */
-int
-s_mp_add (mp_int * a, mp_int * b, mp_int * c)
+int s_mp_add(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int *x;
-  int     olduse, res, min, max;
-
-  /* find sizes, we let |a| <= |b| which means we have to sort
-   * them.  "x" will point to the input with the most digits
-   */
-  if (a->used > b->used) {
-    min = b->used;
-    max = a->used;
-    x = a;
-  } else {
-    min = a->used;
-    max = b->used;
-    x = b;
-  }
-
-  /* init result */
-  if (c->alloc < (max + 1)) {
-    if ((res = mp_grow (c, max + 1)) != MP_OKAY) {
-      return res;
-    }
-  }
-
-  /* get old used digit count and set new one */
-  olduse = c->used;
-  c->used = max + 1;
-
-  {
-    mp_digit u, *tmpa, *tmpb, *tmpc;
-    int i;
-
-    /* alias for digit pointers */
-
-    /* first input */
-    tmpa = a->dp;
-
-    /* second input */
-    tmpb = b->dp;
-
-    /* destination */
-    tmpc = c->dp;
-
-    /* zero the carry */
-    u = 0;
-    for (i = 0; i < min; i++) {
-      /* Compute the sum at one digit, T[i] = A[i] + B[i] + U */
-      *tmpc = *tmpa++ + *tmpb++ + u;
-
-      /* U = carry bit of T[i] */
-      u = *tmpc >> ((mp_digit)DIGIT_BIT);
-
-      /* take away carry bit from T[i] */
-      *tmpc++ &= MP_MASK;
-    }
-
-    /* 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++) {
-        /* T[i] = X[i] + U */
-        *tmpc = x->dp[i] + u;
-
-        /* U = carry bit of T[i] */
-        u = *tmpc >> ((mp_digit)DIGIT_BIT);
-
-        /* take away carry bit from T[i] */
-        *tmpc++ &= MP_MASK;
+   mp_int *x;
+   int     olduse, res, min, max;
+
+   /* find sizes, we let |a| <= |b| which means we have to sort
+    * them.  "x" will point to the input with the most digits
+    */
+   if (a->used > b->used) {
+      min = b->used;
+      max = a->used;
+      x = a;
+   } else {
+      min = a->used;
+      max = b->used;
+      x = b;
+   }
+
+   /* init result */
+   if (c->alloc < (max + 1)) {
+      if ((res = mp_grow(c, max + 1)) != MP_OKAY) {
+         return res;
       }
-    }
+   }
 
-    /* add carry */
-    *tmpc++ = u;
+   /* get old used digit count and set new one */
+   olduse = c->used;
+   c->used = max + 1;
 
-    /* clear digits above oldused */
-    for (i = c->used; i < olduse; i++) {
-      *tmpc++ = 0;
-    }
-  }
+   {
+      mp_digit u, *tmpa, *tmpb, *tmpc;
+      int i;
 
-  mp_clamp (c);
-  return MP_OKAY;
+      /* alias for digit pointers */
+
+      /* first input */
+      tmpa = a->dp;
+
+      /* second input */
+      tmpb = b->dp;
+
+      /* destination */
+      tmpc = c->dp;
+
+      /* zero the carry */
+      u = 0;
+      for (i = 0; i < min; i++) {
+         /* Compute the sum at one digit, T[i] = A[i] + B[i] + U */
+         *tmpc = *tmpa++ + *tmpb++ + u;
+
+         /* U = carry bit of T[i] */
+         u = *tmpc >> ((mp_digit)DIGIT_BIT);
+
+         /* take away carry bit from T[i] */
+         *tmpc++ &= MP_MASK;
+      }
+
+      /* 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++) {
+            /* T[i] = X[i] + U */
+            *tmpc = x->dp[i] + u;
+
+            /* U = carry bit of T[i] */
+            u = *tmpc >> ((mp_digit)DIGIT_BIT);
+
+            /* take away carry bit from T[i] */
+            *tmpc++ &= MP_MASK;
+         }
+      }
+
+      /* add carry */
+      *tmpc++ = u;
+
+      /* clear digits above oldused */
+      for (i = c->used; i < olduse; i++) {
+         *tmpc++ = 0;
+      }
+   }
+
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_exptmod.c b/libtommath/bn_s_mp_exptmod.c
index ab820d4..f8e6b3b 100644
--- a/libtommath/bn_s_mp_exptmod.c
+++ b/libtommath/bn_s_mp_exptmod.c
@@ -15,235 +15,237 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 #ifdef MP_LOW_MEM
-   #define TAB_SIZE 32
+#   define TAB_SIZE 32
 #else
-   #define TAB_SIZE 256
+#   define TAB_SIZE 256
 #endif
 
-int s_mp_exptmod (mp_int * G, mp_int * X, mp_int * P, mp_int * Y, int redmode)
+int s_mp_exptmod(mp_int *G, mp_int *X, mp_int *P, mp_int *Y, int redmode)
 {
-  mp_int  M[TAB_SIZE], res, mu;
-  mp_digit buf;
-  int     err, bitbuf, bitcpy, bitcnt, mode, digidx, x, y, winsize;
-  int (*redux)(mp_int*,mp_int*,mp_int*);
-
-  /* find window size */
-  x = mp_count_bits (X);
-  if (x <= 7) {
-    winsize = 2;
-  } else if (x <= 36) {
-    winsize = 3;
-  } else if (x <= 140) {
-    winsize = 4;
-  } else if (x <= 450) {
-    winsize = 5;
-  } else if (x <= 1303) {
-    winsize = 6;
-  } else if (x <= 3529) {
-    winsize = 7;
-  } else {
-    winsize = 8;
-  }
+   mp_int  M[TAB_SIZE], res, mu;
+   mp_digit buf;
+   int     err, bitbuf, bitcpy, bitcnt, mode, digidx, x, y, winsize;
+   int (*redux)(mp_int *,mp_int *,mp_int *);
+
+   /* find window size */
+   x = mp_count_bits(X);
+   if (x <= 7) {
+      winsize = 2;
+   } else if (x <= 36) {
+      winsize = 3;
+   } else if (x <= 140) {
+      winsize = 4;
+   } else if (x <= 450) {
+      winsize = 5;
+   } else if (x <= 1303) {
+      winsize = 6;
+   } else if (x <= 3529) {
+      winsize = 7;
+   } else {
+      winsize = 8;
+   }
 
 #ifdef MP_LOW_MEM
-    if (winsize > 5) {
-       winsize = 5;
-    }
+   if (winsize > 5) {
+      winsize = 5;
+   }
 #endif
 
-  /* init M array */
-  /* init first cell */
-  if ((err = mp_init(&M[1])) != MP_OKAY) {
-     return err; 
-  }
-
-  /* now init the second half of the array */
-  for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
-    if ((err = mp_init(&M[x])) != MP_OKAY) {
-      for (y = 1<<(winsize-1); y < x; y++) {
-        mp_clear (&M[y]);
-      }
-      mp_clear(&M[1]);
+   /* init M array */
+   /* init first cell */
+   if ((err = mp_init(&M[1])) != MP_OKAY) {
       return err;
-    }
-  }
-
-  /* create mu, used for Barrett reduction */
-  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;
-     }
-     redux = mp_reduce;
-  } else {
-     if ((err = mp_reduce_2k_setup_l (P, &mu)) != MP_OKAY) {
-        goto LBL_MU;
-     }
-     redux = mp_reduce_2k_l;
-  }    
-
-  /* create M table
-   *
-   * The M table contains powers of the base, 
-   * e.g. M[x] = G**x mod P
-   *
-   * 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 
-   */
-  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)], 
-                       &M[1 << (winsize - 1)])) != MP_OKAY) {
-      goto LBL_MU;
-    }
+   }
+
+   /* now init the second half of the array */
+   for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
+      if ((err = mp_init(&M[x])) != MP_OKAY) {
+         for (y = 1<<(winsize-1); y < x; y++) {
+            mp_clear(&M[y]);
+         }
+         mp_clear(&M[1]);
+         return err;
+      }
+   }
 
-    /* reduce modulo P */
-    if ((err = redux (&M[1 << (winsize - 1)], P, &mu)) != MP_OKAY) {
-      goto LBL_MU;
-    }
-  }
-
-  /* create upper table, that is M[x] = M[x-1] * M[1] (mod P)
-   * for x = (2**(winsize - 1) + 1) to (2**winsize - 1)
-   */
-  for (x = (1 << (winsize - 1)) + 1; x < (1 << winsize); x++) {
-    if ((err = mp_mul (&M[x - 1], &M[1], &M[x])) != MP_OKAY) {
+   /* create mu, used for Barrett reduction */
+   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;
+      }
+      redux = mp_reduce;
+   } else {
+      if ((err = mp_reduce_2k_setup_l(P, &mu)) != MP_OKAY) {
+         goto LBL_MU;
+      }
+      redux = mp_reduce_2k_l;
+   }
+
+   /* create M table
+    *
+    * The M table contains powers of the base,
+    * e.g. M[x] = G**x mod P
+    *
+    * 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;
-    }
-    if ((err = redux (&M[x], P, &mu)) != MP_OKAY) {
+   }
+
+   /* 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;
-    }
-  }
-
-  /* setup result */
-  if ((err = mp_init (&res)) != MP_OKAY) {
-    goto LBL_MU;
-  }
-  mp_set (&res, 1);
-
-  /* set initial mode and bit cnt */
-  mode   = 0;
-  bitcnt = 1;
-  buf    = 0;
-  digidx = X->used - 1;
-  bitcpy = 0;
-  bitbuf = 0;
-
-  for (;;) {
-    /* grab next digit as required */
-    if (--bitcnt == 0) {
-      /* if digidx == -1 we are out of digits */
-      if (digidx == -1) {
-        break;
+   }
+
+   for (x = 0; x < (winsize - 1); x++) {
+      /* square it */
+      if ((err = mp_sqr(&M[1 << (winsize - 1)],
+                        &M[1 << (winsize - 1)])) != MP_OKAY) {
+         goto LBL_MU;
       }
-      /* read next digit and reset the bitcnt */
-      buf    = X->dp[digidx--];
-      bitcnt = (int) DIGIT_BIT;
-    }
-
-    /* grab the next msb from the exponent */
-    y     = (buf >> (mp_digit)(DIGIT_BIT - 1)) & 1;
-    buf <<= (mp_digit)1;
-
-    /* if the bit is zero and mode == 0 then we ignore it
-     * These represent the leading zero bits before the first 1 bit
-     * in the exponent.  Technically this opt is not required but it
-     * does lower the # of trivial squaring/reductions used
-     */
-    if ((mode == 0) && (y == 0)) {
-      continue;
-    }
-
-    /* if the bit is zero and mode == 1 then we square */
-    if ((mode == 1) && (y == 0)) {
-      if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-        goto LBL_RES;
+
+      /* reduce modulo P */
+      if ((err = redux(&M[1 << (winsize - 1)], P, &mu)) != MP_OKAY) {
+         goto LBL_MU;
       }
-      if ((err = redux (&res, P, &mu)) != MP_OKAY) {
-        goto LBL_RES;
+   }
+
+   /* create upper table, that is M[x] = M[x-1] * M[1] (mod P)
+    * for x = (2**(winsize - 1) + 1) to (2**winsize - 1)
+    */
+   for (x = (1 << (winsize - 1)) + 1; x < (1 << winsize); x++) {
+      if ((err = mp_mul(&M[x - 1], &M[1], &M[x])) != MP_OKAY) {
+         goto LBL_MU;
       }
-      continue;
-    }
-
-    /* else we add it to the window */
-    bitbuf |= (y << (winsize - ++bitcpy));
-    mode    = 2;
-
-    if (bitcpy == winsize) {
-      /* ok window is filled so square as required and multiply  */
-      /* square first */
-      for (x = 0; x < winsize; x++) {
-        if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-          goto LBL_RES;
-        }
-        if ((err = redux (&res, P, &mu)) != MP_OKAY) {
-          goto LBL_RES;
-        }
+      if ((err = redux(&M[x], P, &mu)) != MP_OKAY) {
+         goto LBL_MU;
       }
+   }
 
-      /* then multiply */
-      if ((err = mp_mul (&res, &M[bitbuf], &res)) != MP_OKAY) {
-        goto LBL_RES;
-      }
-      if ((err = redux (&res, P, &mu)) != MP_OKAY) {
-        goto LBL_RES;
+   /* setup result */
+   if ((err = mp_init(&res)) != MP_OKAY) {
+      goto LBL_MU;
+   }
+   mp_set(&res, 1);
+
+   /* set initial mode and bit cnt */
+   mode   = 0;
+   bitcnt = 1;
+   buf    = 0;
+   digidx = X->used - 1;
+   bitcpy = 0;
+   bitbuf = 0;
+
+   for (;;) {
+      /* grab next digit as required */
+      if (--bitcnt == 0) {
+         /* if digidx == -1 we are out of digits */
+         if (digidx == -1) {
+            break;
+         }
+         /* read next digit and reset the bitcnt */
+         buf    = X->dp[digidx--];
+         bitcnt = (int)DIGIT_BIT;
       }
 
-      /* empty window and reset */
-      bitcpy = 0;
-      bitbuf = 0;
-      mode   = 1;
-    }
-  }
-
-  /* if bits remain then square/multiply */
-  if ((mode == 2) && (bitcpy > 0)) {
-    /* square then multiply if the bit is set */
-    for (x = 0; x < bitcpy; x++) {
-      if ((err = mp_sqr (&res, &res)) != MP_OKAY) {
-        goto LBL_RES;
-      }
-      if ((err = redux (&res, P, &mu)) != MP_OKAY) {
-        goto LBL_RES;
+      /* grab the next msb from the exponent */
+      y     = (buf >> (mp_digit)(DIGIT_BIT - 1)) & 1;
+      buf <<= (mp_digit)1;
+
+      /* if the bit is zero and mode == 0 then we ignore it
+       * These represent the leading zero bits before the first 1 bit
+       * in the exponent.  Technically this opt is not required but it
+       * does lower the # of trivial squaring/reductions used
+       */
+      if ((mode == 0) && (y == 0)) {
+         continue;
       }
 
-      bitbuf <<= 1;
-      if ((bitbuf & (1 << winsize)) != 0) {
-        /* then multiply */
-        if ((err = mp_mul (&res, &M[1], &res)) != MP_OKAY) {
-          goto LBL_RES;
-        }
-        if ((err = redux (&res, P, &mu)) != MP_OKAY) {
-          goto LBL_RES;
-        }
+      /* if the bit is zero and mode == 1 then we square */
+      if ((mode == 1) && (y == 0)) {
+         if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, &mu)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         continue;
       }
-    }
-  }
 
-  mp_exch (&res, Y);
-  err = MP_OKAY;
-LBL_RES:mp_clear (&res);
-LBL_MU:mp_clear (&mu);
+      /* else we add it to the window */
+      bitbuf |= (y << (winsize - ++bitcpy));
+      mode    = 2;
+
+      if (bitcpy == winsize) {
+         /* ok window is filled so square as required and multiply  */
+         /* square first */
+         for (x = 0; x < winsize; x++) {
+            if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+            if ((err = redux(&res, P, &mu)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+         }
+
+         /* then multiply */
+         if ((err = mp_mul(&res, &M[bitbuf], &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, &mu)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+
+         /* empty window and reset */
+         bitcpy = 0;
+         bitbuf = 0;
+         mode   = 1;
+      }
+   }
+
+   /* if bits remain then square/multiply */
+   if ((mode == 2) && (bitcpy > 0)) {
+      /* square then multiply if the bit is set */
+      for (x = 0; x < bitcpy; x++) {
+         if ((err = mp_sqr(&res, &res)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+         if ((err = redux(&res, P, &mu)) != MP_OKAY) {
+            goto LBL_RES;
+         }
+
+         bitbuf <<= 1;
+         if ((bitbuf & (1 << winsize)) != 0) {
+            /* then multiply */
+            if ((err = mp_mul(&res, &M[1], &res)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+            if ((err = redux(&res, P, &mu)) != MP_OKAY) {
+               goto LBL_RES;
+            }
+         }
+      }
+   }
+
+   mp_exch(&res, Y);
+   err = MP_OKAY;
+LBL_RES:
+   mp_clear(&res);
+LBL_MU:
+   mp_clear(&mu);
 LBL_M:
-  mp_clear(&M[1]);
-  for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
-    mp_clear (&M[x]);
-  }
-  return err;
+   mp_clear(&M[1]);
+   for (x = 1<<(winsize-1); x < (1 << winsize); x++) {
+      mp_clear(&M[x]);
+   }
+   return err;
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_mul_digs.c b/libtommath/bn_s_mp_mul_digs.c
index 8f1bf97..c72c2a8 100644
--- a/libtommath/bn_s_mp_mul_digs.c
+++ b/libtommath/bn_s_mp_mul_digs.c
@@ -16,72 +16,72 @@
  */
 
 /* multiplies |a| * |b| and only computes upto 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.
  */
-int s_mp_mul_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
+int s_mp_mul_digs(mp_int *a, mp_int *b, mp_int *c, int digs)
 {
-  mp_int  t;
-  int     res, pa, pb, ix, iy;
-  mp_digit u;
-  mp_word r;
-  mp_digit tmpx, *tmpt, *tmpy;
+   mp_int  t;
+   int     res, pa, pb, ix, iy;
+   mp_digit u;
+   mp_word r;
+   mp_digit tmpx, *tmpt, *tmpy;
 
-  /* can we use the fast multiplier? */
-  if (((digs) < MP_WARRAY) &&
-      (MIN (a->used, b->used) < 
-          (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
-    return fast_s_mp_mul_digs (a, b, c, digs);
-  }
+   /* can we use the fast multiplier? */
+   if (((digs) < MP_WARRAY) &&
+       (MIN(a->used, b->used) <
+        (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
+      return fast_s_mp_mul_digs(a, b, c, digs);
+   }
 
-  if ((res = mp_init_size (&t, digs)) != MP_OKAY) {
-    return res;
-  }
-  t.used = digs;
+   if ((res = mp_init_size(&t, digs)) != MP_OKAY) {
+      return res;
+   }
+   t.used = digs;
 
-  /* compute the digits of the product directly */
-  pa = a->used;
-  for (ix = 0; ix < pa; ix++) {
-    /* set the carry to zero */
-    u = 0;
+   /* compute the digits of the product directly */
+   pa = a->used;
+   for (ix = 0; ix < pa; ix++) {
+      /* set the carry to zero */
+      u = 0;
 
-    /* limit ourselves to making digs digits of output */
-    pb = MIN (b->used, digs - ix);
+      /* limit ourselves to making digs digits of output */
+      pb = MIN(b->used, digs - ix);
 
-    /* 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;
+      /* setup some aliases */
+      /* copy of the digit from a used within the nested loop */
+      tmpx = a->dp[ix];
 
-    /* compute the columns of the output and propagate the carry */
-    for (iy = 0; iy < pb; iy++) {
-      /* compute the column as a mp_word */
-      r       = (mp_word)*tmpt +
-                ((mp_word)tmpx * (mp_word)*tmpy++) +
-                (mp_word)u;
+      /* an alias for the destination shifted ix places */
+      tmpt = t.dp + ix;
 
-      /* the new column is the lower part of the result */
-      *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
+      /* an alias for the digits of b */
+      tmpy = b->dp;
 
-      /* get the carry word from the result */
-      u       = (mp_digit) (r >> ((mp_word) DIGIT_BIT));
-    }
-    /* set carry if it is placed below digs */
-    if ((ix + iy) < digs) {
-      *tmpt = u;
-    }
-  }
+      /* compute the columns of the output and propagate the carry */
+      for (iy = 0; iy < pb; iy++) {
+         /* compute the column as a mp_word */
+         r       = (mp_word)*tmpt +
+                   ((mp_word)tmpx * (mp_word)*tmpy++) +
+                   (mp_word)u;
 
-  mp_clamp (&t);
-  mp_exch (&t, c);
+         /* the new column is the lower part of the result */
+         *tmpt++ = (mp_digit)(r & ((mp_word) MP_MASK));
 
-  mp_clear (&t);
-  return MP_OKAY;
+         /* get the carry word from the result */
+         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+      }
+      /* set carry if it is placed below digs */
+      if ((ix + iy) < digs) {
+         *tmpt = u;
+      }
+   }
+
+   mp_clamp(&t);
+   mp_exch(&t, c);
+
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_mul_high_digs.c b/libtommath/bn_s_mp_mul_high_digs.c
index 031f17b..0e12f3b 100644
--- a/libtommath/bn_s_mp_mul_high_digs.c
+++ b/libtommath/bn_s_mp_mul_high_digs.c
@@ -18,61 +18,60 @@
 /* multiplies |a| * |b| and does not compute the lower digs digits
  * [meant to get the higher part of the product]
  */
-int
-s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
+int s_mp_mul_high_digs(mp_int *a, mp_int *b, mp_int *c, int digs)
 {
-  mp_int  t;
-  int     res, pa, pb, ix, iy;
-  mp_digit u;
-  mp_word r;
-  mp_digit tmpx, *tmpt, *tmpy;
+   mp_int  t;
+   int     res, pa, pb, ix, iy;
+   mp_digit u;
+   mp_word r;
+   mp_digit tmpx, *tmpt, *tmpy;
 
-  /* can we use the fast multiplier? */
+   /* can we use the fast multiplier? */
 #ifdef BN_FAST_S_MP_MUL_HIGH_DIGS_C
-  if (((a->used + b->used + 1) < MP_WARRAY)
-      && (MIN (a->used, b->used) < (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
-    return fast_s_mp_mul_high_digs (a, b, c, digs);
-  }
+   if (((a->used + b->used + 1) < MP_WARRAY)
+       && (MIN(a->used, b->used) < (1 << ((CHAR_BIT * sizeof(mp_word)) - (2 * DIGIT_BIT))))) {
+      return fast_s_mp_mul_high_digs(a, b, c, digs);
+   }
 #endif
 
-  if ((res = mp_init_size (&t, a->used + b->used + 1)) != MP_OKAY) {
-    return res;
-  }
-  t.used = a->used + b->used + 1;
+   if ((res = mp_init_size(&t, a->used + b->used + 1)) != MP_OKAY) {
+      return res;
+   }
+   t.used = a->used + b->used + 1;
 
-  pa = a->used;
-  pb = b->used;
-  for (ix = 0; ix < pa; ix++) {
-    /* clear the carry */
-    u = 0;
+   pa = a->used;
+   pb = b->used;
+   for (ix = 0; ix < pa; ix++) {
+      /* clear the carry */
+      u = 0;
 
-    /* left hand side of A[ix] * B[iy] */
-    tmpx = a->dp[ix];
+      /* left hand side of A[ix] * B[iy] */
+      tmpx = a->dp[ix];
 
-    /* alias to the address of where the digits will be stored */
-    tmpt = &(t.dp[digs]);
+      /* alias to the address of where the digits will be stored */
+      tmpt = &(t.dp[digs]);
 
-    /* alias for where to read the right hand side from */
-    tmpy = b->dp + (digs - ix);
+      /* alias for where to read the right hand side from */
+      tmpy = b->dp + (digs - ix);
 
-    for (iy = digs - ix; iy < pb; iy++) {
-      /* calculate the double precision result */
-      r       = (mp_word)*tmpt +
-                ((mp_word)tmpx * (mp_word)*tmpy++) +
-                (mp_word)u;
+      for (iy = digs - ix; iy < pb; iy++) {
+         /* calculate the double precision result */
+         r       = (mp_word)*tmpt +
+                   ((mp_word)tmpx * (mp_word)*tmpy++) +
+                   (mp_word)u;
 
-      /* get the lower part */
-      *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
+         /* get the lower part */
+         *tmpt++ = (mp_digit)(r & ((mp_word) MP_MASK));
 
-      /* carry the carry */
-      u       = (mp_digit) (r >> ((mp_word) DIGIT_BIT));
-    }
-    *tmpt = u;
-  }
-  mp_clamp (&t);
-  mp_exch (&t, c);
-  mp_clear (&t);
-  return MP_OKAY;
+         /* carry the carry */
+         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+      }
+      *tmpt = u;
+   }
+   mp_clamp(&t);
+   mp_exch(&t, c);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_sqr.c b/libtommath/bn_s_mp_sqr.c
index ac0e157..2910f24 100644
--- a/libtommath/bn_s_mp_sqr.c
+++ b/libtommath/bn_s_mp_sqr.c
@@ -16,66 +16,66 @@
  */
 
 /* low level squaring, b = a*a, HAC pp.596-597, Algorithm 14.16 */
-int s_mp_sqr (mp_int * a, mp_int * b)
+int s_mp_sqr(mp_int *a, mp_int *b)
 {
-  mp_int  t;
-  int     res, ix, iy, pa;
-  mp_word r;
-  mp_digit u, tmpx, *tmpt;
+   mp_int  t;
+   int     res, ix, iy, pa;
+   mp_word r;
+   mp_digit u, tmpx, *tmpt;
 
-  pa = a->used;
-  if ((res = mp_init_size (&t, (2 * pa) + 1)) != MP_OKAY) {
-    return res;
-  }
+   pa = a->used;
+   if ((res = mp_init_size(&t, (2 * pa) + 1)) != MP_OKAY) {
+      return res;
+   }
 
-  /* default used is maximum possible size */
-  t.used = (2 * pa) + 1;
+   /* default used is maximum possible size */
+   t.used = (2 * pa) + 1;
 
-  for (ix = 0; ix < pa; ix++) {
-    /* first calculate the digit at 2*ix */
-    /* calculate double precision result */
-    r = (mp_word)t.dp[2*ix] +
-        ((mp_word)a->dp[ix] * (mp_word)a->dp[ix]);
+   for (ix = 0; ix < pa; ix++) {
+      /* first calculate the digit at 2*ix */
+      /* calculate double precision result */
+      r = (mp_word)t.dp[2*ix] +
+          ((mp_word)a->dp[ix] * (mp_word)a->dp[ix]);
 
-    /* store lower part in result */
-    t.dp[ix+ix] = (mp_digit) (r & ((mp_word) MP_MASK));
+      /* store lower part in result */
+      t.dp[ix+ix] = (mp_digit)(r & ((mp_word)MP_MASK));
 
-    /* get the carry */
-    u           = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+      /* get the carry */
+      u           = (mp_digit)(r >> ((mp_word)DIGIT_BIT));
 
-    /* left hand side of A[ix] * A[iy] */
-    tmpx        = a->dp[ix];
+      /* left hand side of A[ix] * A[iy] */
+      tmpx        = a->dp[ix];
 
-    /* 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 where to store the results */
+      tmpt        = t.dp + ((2 * ix) + 1);
 
-      /* now calculate the double precision result, note we use
-       * addition instead of *2 since it's easier to optimize
-       */
-      r       = ((mp_word) *tmpt) + r + r + ((mp_word) u);
+      for (iy = ix + 1; iy < pa; iy++) {
+         /* first calculate the product */
+         r       = ((mp_word)tmpx) * ((mp_word)a->dp[iy]);
 
-      /* store lower part */
-      *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
+         /* now calculate the double precision result, note we use
+          * addition instead of *2 since it's easier to optimize
+          */
+         r       = ((mp_word) *tmpt) + r + r + ((mp_word) u);
 
-      /* get carry */
-      u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
-    }
-    /* propagate upwards */
-    while (u != ((mp_digit) 0)) {
-      r       = ((mp_word) *tmpt) + ((mp_word) u);
-      *tmpt++ = (mp_digit) (r & ((mp_word) MP_MASK));
-      u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
-    }
-  }
+         /* store lower part */
+         *tmpt++ = (mp_digit)(r & ((mp_word) MP_MASK));
 
-  mp_clamp (&t);
-  mp_exch (&t, b);
-  mp_clear (&t);
-  return MP_OKAY;
+         /* get carry */
+         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+      }
+      /* propagate upwards */
+      while (u != ((mp_digit) 0)) {
+         r       = ((mp_word) *tmpt) + ((mp_word) u);
+         *tmpt++ = (mp_digit)(r & ((mp_word) MP_MASK));
+         u       = (mp_digit)(r >> ((mp_word) DIGIT_BIT));
+      }
+   }
+
+   mp_clamp(&t);
+   mp_exch(&t, b);
+   mp_clear(&t);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_s_mp_sub.c b/libtommath/bn_s_mp_sub.c
index 8091f4a..f61aa83 100644
--- a/libtommath/bn_s_mp_sub.c
+++ b/libtommath/bn_s_mp_sub.c
@@ -16,70 +16,69 @@
  */
 
 /* low level subtraction (assumes |a| > |b|), HAC pp.595 Algorithm 14.9 */
-int
-s_mp_sub (mp_int * a, mp_int * b, mp_int * c)
+int s_mp_sub(mp_int *a, mp_int *b, mp_int *c)
 {
-  int     olduse, res, min, max;
+   int     olduse, res, min, max;
 
-  /* find sizes */
-  min = b->used;
-  max = a->used;
+   /* find sizes */
+   min = b->used;
+   max = a->used;
 
-  /* init result */
-  if (c->alloc < max) {
-    if ((res = mp_grow (c, max)) != MP_OKAY) {
-      return res;
-    }
-  }
-  olduse = c->used;
-  c->used = max;
+   /* init result */
+   if (c->alloc < max) {
+      if ((res = mp_grow(c, max)) != MP_OKAY) {
+         return res;
+      }
+   }
+   olduse = c->used;
+   c->used = max;
 
-  {
-    mp_digit u, *tmpa, *tmpb, *tmpc;
-    int i;
+   {
+      mp_digit u, *tmpa, *tmpb, *tmpc;
+      int i;
 
-    /* alias for digit pointers */
-    tmpa = a->dp;
-    tmpb = b->dp;
-    tmpc = c->dp;
+      /* alias for digit pointers */
+      tmpa = a->dp;
+      tmpb = b->dp;
+      tmpc = c->dp;
 
-    /* set carry to zero */
-    u = 0;
-    for (i = 0; i < min; i++) {
-      /* T[i] = A[i] - B[i] - U */
-      *tmpc = (*tmpa++ - *tmpb++) - u;
+      /* set carry to zero */
+      u = 0;
+      for (i = 0; i < min; i++) {
+         /* T[i] = A[i] - B[i] - U */
+         *tmpc = (*tmpa++ - *tmpb++) - u;
 
-      /* U = carry bit of T[i]
-       * Note this saves performing an AND operation since
-       * if a carry does occur it will propagate all the way to the
-       * MSB.  As a result a single shift is enough to get the carry
-       */
-      u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
+         /* U = carry bit of T[i]
+          * Note this saves performing an AND operation since
+          * if a carry does occur it will propagate all the way to the
+          * MSB.  As a result a single shift is enough to get the carry
+          */
+         u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
 
-      /* Clear carry from T[i] */
-      *tmpc++ &= MP_MASK;
-    }
+         /* Clear carry from T[i] */
+         *tmpc++ &= MP_MASK;
+      }
 
-    /* now copy higher words if any, e.g. if A has more digits than B  */
-    for (; i < max; i++) {
-      /* T[i] = A[i] - U */
-      *tmpc = *tmpa++ - u;
+      /* now copy higher words if any, e.g. if A has more digits than B  */
+      for (; i < max; i++) {
+         /* T[i] = A[i] - U */
+         *tmpc = *tmpa++ - u;
 
-      /* U = carry bit of T[i] */
-      u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
+         /* U = carry bit of T[i] */
+         u = *tmpc >> ((mp_digit)((CHAR_BIT * sizeof(mp_digit)) - 1));
 
-      /* Clear carry from T[i] */
-      *tmpc++ &= MP_MASK;
-    }
+         /* Clear carry from T[i] */
+         *tmpc++ &= MP_MASK;
+      }
 
-    /* clear digits above used (since we may not have grown result above) */
-    for (i = c->used; i < olduse; i++) {
-      *tmpc++ = 0;
-    }
-  }
+      /* clear digits above used (since we may not have grown result above) */
+      for (i = c->used; i < olduse; i++) {
+         *tmpc++ = 0;
+      }
+   }
 
-  mp_clamp (c);
-  return MP_OKAY;
+   mp_clamp(c);
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bncore.c b/libtommath/bncore.c
index e80ec99..cfd19f0 100644
--- a/libtommath/bncore.c
+++ b/libtommath/bncore.c
@@ -21,14 +21,14 @@
 -------------------------------------------------------------
  Intel P4 Northwood     /GCC v3.4.1   /        88/       128/LTM 0.32 ;-)
  AMD Athlon64           /GCC v3.4.4   /        80/       120/LTM 0.35
- 
+
 */
 
 int     KARATSUBA_MUL_CUTOFF = 80,      /* Min. number of digits before Karatsuba multiplication is used. */
         KARATSUBA_SQR_CUTOFF = 120,     /* Min. number of digits before Karatsuba squaring is used. */
-        
+
         TOOM_MUL_CUTOFF      = 350,      /* no optimal values of these are known yet so set em high */
-        TOOM_SQR_CUTOFF      = 400; 
+        TOOM_SQR_CUTOFF      = 400;
 #endif
 
 /* ref:         $Format:%D$ */
diff --git a/libtommath/makefile b/libtommath/makefile
index fdd2435..64d8fcd 100644
--- a/libtommath/makefile
+++ b/libtommath/makefile
@@ -149,3 +149,6 @@ new_file:
 
 perlcritic:
 	perlcritic *.pl
+
+astyle:
+	astyle --options=astylerc $(OBJECTS:.o=.c)
diff --git a/libtommath/tommath.h b/libtommath/tommath.h
index 80bae69..2aaea5b 100644
--- a/libtommath/tommath.h
+++ b/libtommath/tommath.h
@@ -33,9 +33,9 @@ extern "C" {
     defined(__sparcv9) || defined(__sparc_v9__) || defined(__sparc64__) || \
     defined(__ia64) || defined(__ia64__) || defined(__itanium__) || defined(_M_IA64) || \
     defined(__LP64__) || defined(_LP64) || defined(__64BIT__)
-   #if !(defined(MP_32BIT) || defined(MP_16BIT) || defined(MP_8BIT))
-      #define MP_64BIT
-   #endif
+#   if !(defined(MP_32BIT) || defined(MP_16BIT) || defined(MP_8BIT))
+#      define MP_64BIT
+#   endif
 #endif
 
 /* some default configurations.
@@ -47,68 +47,68 @@ extern "C" {
  * [any size beyond that is ok provided it doesn't overflow the data type]
  */
 #ifdef MP_8BIT
-   typedef uint8_t              mp_digit;
-   typedef uint16_t             mp_word;
-#define MP_SIZEOF_MP_DIGIT      1
-#ifdef DIGIT_BIT
-#error You must not define DIGIT_BIT when using MP_8BIT
-#endif
+typedef uint8_t              mp_digit;
+typedef uint16_t             mp_word;
+#   define MP_SIZEOF_MP_DIGIT 1
+#   ifdef DIGIT_BIT
+#      error You must not define DIGIT_BIT when using MP_8BIT
+#   endif
 #elif defined(MP_16BIT)
-   typedef uint16_t             mp_digit;
-   typedef uint32_t             mp_word;
-#define MP_SIZEOF_MP_DIGIT      2
-#ifdef DIGIT_BIT
-#error You must not define DIGIT_BIT when using MP_16BIT
-#endif
+typedef uint16_t             mp_digit;
+typedef uint32_t             mp_word;
+#   define MP_SIZEOF_MP_DIGIT 2
+#   ifdef DIGIT_BIT
+#      error You must not define DIGIT_BIT when using MP_16BIT
+#   endif
 #elif defined(MP_64BIT)
-   /* for GCC only on supported platforms */
-   typedef uint64_t mp_digit;
-#if defined(_WIN32)
-   typedef unsigned __int128    mp_word;
-#elif defined(__GNUC__)
-   typedef unsigned long        mp_word __attribute__ ((mode(TI)));
-#else
-   /* it seems you have a problem
-    * but we assume you can somewhere define your own uint128_t */
-   typedef uint128_t            mp_word;
-#endif
-
-   #define DIGIT_BIT            60
+/* for GCC only on supported platforms */
+typedef uint64_t mp_digit;
+#   if defined(_WIN32)
+typedef unsigned __int128    mp_word;
+#   elif defined(__GNUC__)
+typedef unsigned long        mp_word __attribute__((mode(TI)));
+#   else
+/* it seems you have a problem
+ * but we assume you can somewhere define your own uint128_t */
+typedef uint128_t            mp_word;
+#   endif
+
+#   define DIGIT_BIT 60
 #else
-   /* this is the default case, 28-bit digits */
-
-   /* this is to make porting into LibTomCrypt easier :-) */
-   typedef uint32_t             mp_digit;
-   typedef uint64_t             mp_word;
-
-#ifdef MP_31BIT
-   /* this is an extension that uses 31-bit digits */
-   #define DIGIT_BIT            31
-#else
-   /* default case is 28-bit digits, defines MP_28BIT as a handy macro to test */
-   #define DIGIT_BIT            28
-   #define MP_28BIT
-#endif
+/* this is the default case, 28-bit digits */
+
+/* this is to make porting into LibTomCrypt easier :-) */
+typedef uint32_t             mp_digit;
+typedef uint64_t             mp_word;
+
+#   ifdef MP_31BIT
+/* this is an extension that uses 31-bit digits */
+#      define DIGIT_BIT 31
+#   else
+/* default case is 28-bit digits, defines MP_28BIT as a handy macro to test */
+#      define DIGIT_BIT 28
+#      define MP_28BIT
+#   endif
 #endif
 
 /* otherwise the bits per digit is calculated automatically from the size of a mp_digit */
 #ifndef DIGIT_BIT
-   #define DIGIT_BIT     (((CHAR_BIT * MP_SIZEOF_MP_DIGIT) - 1))  /* bits per digit */
-   typedef uint_least32_t mp_min_u32;
+#   define DIGIT_BIT (((CHAR_BIT * MP_SIZEOF_MP_DIGIT) - 1))  /* bits per digit */
+typedef uint_least32_t mp_min_u32;
 #else
-   typedef mp_digit mp_min_u32;
+typedef mp_digit mp_min_u32;
 #endif
 
 /* use arc4random on platforms that support it */
 #if defined(__FreeBSD__) || defined(__OpenBSD__) || defined(__NetBSD__) || defined(__DragonFly__)
-    #define MP_GEN_RANDOM()    arc4random()
-    #define MP_GEN_RANDOM_MAX  0xffffffff
+#   define MP_GEN_RANDOM()    arc4random()
+#   define MP_GEN_RANDOM_MAX  0xffffffff
 #endif
 
 /* use rand() as fall-back if there's no better rand function */
 #ifndef MP_GEN_RANDOM
-    #define MP_GEN_RANDOM()    rand()
-    #define MP_GEN_RANDOM_MAX  RAND_MAX
+#   define MP_GEN_RANDOM()    rand()
+#   define MP_GEN_RANDOM_MAX  RAND_MAX
 #endif
 
 #define MP_DIGIT_BIT     DIGIT_BIT
@@ -140,20 +140,20 @@ typedef int           mp_err;
 
 /* you'll have to tune these... */
 extern int KARATSUBA_MUL_CUTOFF,
-           KARATSUBA_SQR_CUTOFF,
-           TOOM_MUL_CUTOFF,
-           TOOM_SQR_CUTOFF;
+       KARATSUBA_SQR_CUTOFF,
+       TOOM_MUL_CUTOFF,
+       TOOM_SQR_CUTOFF;
 
 /* define this to use lower memory usage routines (exptmods mostly) */
 /* #define MP_LOW_MEM */
 
 /* default precision */
 #ifndef MP_PREC
-   #ifndef MP_LOW_MEM
-      #define MP_PREC                 32     /* default digits of precision */
-   #else
-      #define MP_PREC                 8      /* default digits of precision */
-   #endif
+#   ifndef MP_LOW_MEM
+#      define MP_PREC 32        /* default digits of precision */
+#   else
+#      define MP_PREC 8         /* default digits of precision */
+#   endif
 #endif
 
 /* size of comba arrays, should be at least 2 * 2**(BITS_PER_WORD - BITS_PER_DIGIT*2) */
@@ -161,17 +161,17 @@ extern int KARATSUBA_MUL_CUTOFF,
 
 /* the infamous mp_int structure */
 typedef struct  {
-    int used, alloc, sign;
-    mp_digit *dp;
+   int used, alloc, sign;
+   mp_digit *dp;
 } mp_int;
 
 /* callback for mp_prime_random, should fill dst with random bytes and return how many read [upto len] */
 typedef int ltm_prime_callback(unsigned char *dst, int len, void *dat);
 
 
-#define USED(m)    ((m)->used)
-#define DIGIT(m,k) ((m)->dp[(k)])
-#define SIGN(m)    ((m)->sign)
+#define USED(m)     ((m)->used)
+#define DIGIT(m, k) ((m)->dp[(k)])
+#define SIGN(m)     ((m)->sign)
 
 /* error code to char* string */
 const char *mp_error_to_string(int code);
@@ -223,19 +223,19 @@ int mp_set_long(mp_int *a, unsigned long b);
 int mp_set_long_long(mp_int *a, unsigned long long b);
 
 /* get a 32-bit value */
-unsigned long mp_get_int(mp_int * a);
+unsigned long mp_get_int(mp_int *a);
 
 /* get a platform dependent unsigned long value */
-unsigned long mp_get_long(mp_int * a);
+unsigned long mp_get_long(mp_int *a);
 
 /* get a platform dependent unsigned long long value */
-unsigned long long mp_get_long_long(mp_int * a);
+unsigned long long mp_get_long_long(mp_int *a);
 
 /* initialize and set a digit */
-int mp_init_set (mp_int * a, mp_digit b);
+int mp_init_set(mp_int *a, mp_digit b);
 
 /* initialize and set 32-bit value */
-int mp_init_set_int (mp_int * a, unsigned long b);
+int mp_init_set_int(mp_int *a, unsigned long b);
 
 /* copy, b = a */
 int mp_copy(mp_int *a, mp_int *b);
@@ -247,10 +247,10 @@ int mp_init_copy(mp_int *a, mp_int *b);
 void mp_clamp(mp_int *a);
 
 /* import binary data */
-int mp_import(mp_int* rop, size_t count, int order, size_t size, int endian, size_t nails, const void* op);
+int mp_import(mp_int *rop, size_t count, int order, size_t size, int endian, size_t nails, const void *op);
 
 /* export binary data */
-int mp_export(void* rop, size_t* countp, int order, size_t size, int endian, size_t nails, mp_int* op);
+int mp_export(void *rop, size_t *countp, int order, size_t size, int endian, size_t nails, mp_int *op);
 
 /* ---> digit manipulation <--- */
 
@@ -350,7 +350,7 @@ int mp_div_3(mp_int *a, mp_int *c, mp_digit *d);
 
 /* c = a**b */
 int mp_expt_d(mp_int *a, mp_digit b, mp_int *c);
-int mp_expt_d_ex (mp_int * a, mp_digit b, mp_int * c, int fast);
+int mp_expt_d_ex(mp_int *a, mp_digit b, mp_int *c, int fast);
 
 /* c = a mod b, 0 <= c < b  */
 int mp_mod_d(mp_int *a, mp_digit b, mp_digit *c);
@@ -386,7 +386,7 @@ int mp_lcm(mp_int *a, mp_int *b, mp_int *c);
  * returns error if a < 0 and b is even
  */
 int mp_n_root(mp_int *a, mp_digit b, mp_int *c);
-int mp_n_root_ex (mp_int * a, mp_digit b, mp_int * c, int fast);
+int mp_n_root_ex(mp_int *a, mp_digit b, mp_int *c, int fast);
 
 /* special sqrt algo */
 int mp_sqrt(mp_int *arg, mp_int *ret);
@@ -455,9 +455,9 @@ int mp_exptmod(mp_int *a, mp_int *b, mp_int *c, mp_int *d);
 
 /* number of primes */
 #ifdef MP_8BIT
-   #define PRIME_SIZE      31
+#  define PRIME_SIZE 31
 #else
-   #define PRIME_SIZE      256
+#  define PRIME_SIZE 256
 #endif
 
 /* table of first PRIME_SIZE primes */
@@ -529,16 +529,16 @@ int mp_count_bits(mp_int *a);
 int mp_unsigned_bin_size(mp_int *a);
 int mp_read_unsigned_bin(mp_int *a, const unsigned char *b, int c);
 int mp_to_unsigned_bin(mp_int *a, unsigned char *b);
-int mp_to_unsigned_bin_n (mp_int * a, unsigned char *b, unsigned long *outlen);
+int mp_to_unsigned_bin_n(mp_int *a, unsigned char *b, unsigned long *outlen);
 
 int mp_signed_bin_size(mp_int *a);
 int mp_read_signed_bin(mp_int *a, const unsigned char *b, int c);
 int mp_to_signed_bin(mp_int *a,  unsigned char *b);
-int mp_to_signed_bin_n (mp_int * a, unsigned char *b, unsigned long *outlen);
+int mp_to_signed_bin_n(mp_int *a, unsigned char *b, unsigned long *outlen);
 
 int mp_read_radix(mp_int *a, const char *str, int radix);
 int mp_toradix(mp_int *a, char *str, int radix);
-int mp_toradix_n(mp_int * a, char *str, int radix, int maxlen);
+int mp_toradix_n(mp_int *a, char *str, int radix, int maxlen);
 int mp_radix_size(mp_int *a, int radix, int *size);
 
 #ifndef LTM_NO_FILE
@@ -559,7 +559,7 @@ int mp_fwrite(mp_int *a, int radix, FILE *stream);
 #define mp_tohex(M, S)     mp_toradix((M), (S), 16)
 
 #ifdef __cplusplus
-   }
+}
 #endif
 
 #endif
diff --git a/libtommath/tommath_class.h b/libtommath/tommath_class.h
index e860613..f700d66 100644
--- a/libtommath/tommath_class.h
+++ b/libtommath/tommath_class.h
@@ -1,220 +1,220 @@
 #if !(defined(LTM1) && defined(LTM2) && defined(LTM3))
 #if defined(LTM2)
-#define LTM3
+#   define LTM3
 #endif
 #if defined(LTM1)
-#define LTM2
+#   define LTM2
 #endif
 #define LTM1
 
 #if defined(LTM_ALL)
-#define BN_ERROR_C
-#define BN_FAST_MP_INVMOD_C
-#define BN_FAST_MP_MONTGOMERY_REDUCE_C
-#define BN_FAST_S_MP_MUL_DIGS_C
-#define BN_FAST_S_MP_MUL_HIGH_DIGS_C
-#define BN_FAST_S_MP_SQR_C
-#define BN_MP_2EXPT_C
-#define BN_MP_ABS_C
-#define BN_MP_ADD_C
-#define BN_MP_ADD_D_C
-#define BN_MP_ADDMOD_C
-#define BN_MP_AND_C
-#define BN_MP_CLAMP_C
-#define BN_MP_CLEAR_C
-#define BN_MP_CLEAR_MULTI_C
-#define BN_MP_CMP_C
-#define BN_MP_CMP_D_C
-#define BN_MP_CMP_MAG_C
-#define BN_MP_CNT_LSB_C
-#define BN_MP_COPY_C
-#define BN_MP_COUNT_BITS_C
-#define BN_MP_DIV_C
-#define BN_MP_DIV_2_C
-#define BN_MP_DIV_2D_C
-#define BN_MP_DIV_3_C
-#define BN_MP_DIV_D_C
-#define BN_MP_DR_IS_MODULUS_C
-#define BN_MP_DR_REDUCE_C
-#define BN_MP_DR_SETUP_C
-#define BN_MP_EXCH_C
-#define BN_MP_EXPORT_C
-#define BN_MP_EXPT_D_C
-#define BN_MP_EXPT_D_EX_C
-#define BN_MP_EXPTMOD_C
-#define BN_MP_EXPTMOD_FAST_C
-#define BN_MP_EXTEUCLID_C
-#define BN_MP_FREAD_C
-#define BN_MP_FWRITE_C
-#define BN_MP_GCD_C
-#define BN_MP_GET_INT_C
-#define BN_MP_GET_LONG_C
-#define BN_MP_GET_LONG_LONG_C
-#define BN_MP_GROW_C
-#define BN_MP_IMPORT_C
-#define BN_MP_INIT_C
-#define BN_MP_INIT_COPY_C
-#define BN_MP_INIT_MULTI_C
-#define BN_MP_INIT_SET_C
-#define BN_MP_INIT_SET_INT_C
-#define BN_MP_INIT_SIZE_C
-#define BN_MP_INVMOD_C
-#define BN_MP_INVMOD_SLOW_C
-#define BN_MP_IS_SQUARE_C
-#define BN_MP_JACOBI_C
-#define BN_MP_KARATSUBA_MUL_C
-#define BN_MP_KARATSUBA_SQR_C
-#define BN_MP_LCM_C
-#define BN_MP_LSHD_C
-#define BN_MP_MOD_C
-#define BN_MP_MOD_2D_C
-#define BN_MP_MOD_D_C
-#define BN_MP_MONTGOMERY_CALC_NORMALIZATION_C
-#define BN_MP_MONTGOMERY_REDUCE_C
-#define BN_MP_MONTGOMERY_SETUP_C
-#define BN_MP_MUL_C
-#define BN_MP_MUL_2_C
-#define BN_MP_MUL_2D_C
-#define BN_MP_MUL_D_C
-#define BN_MP_MULMOD_C
-#define BN_MP_N_ROOT_C
-#define BN_MP_N_ROOT_EX_C
-#define BN_MP_NEG_C
-#define BN_MP_OR_C
-#define BN_MP_PRIME_FERMAT_C
-#define BN_MP_PRIME_IS_DIVISIBLE_C
-#define BN_MP_PRIME_IS_PRIME_C
-#define BN_MP_PRIME_MILLER_RABIN_C
-#define BN_MP_PRIME_NEXT_PRIME_C
-#define BN_MP_PRIME_RABIN_MILLER_TRIALS_C
-#define BN_MP_PRIME_RANDOM_EX_C
-#define BN_MP_RADIX_SIZE_C
-#define BN_MP_RADIX_SMAP_C
-#define BN_MP_RAND_C
-#define BN_MP_READ_RADIX_C
-#define BN_MP_READ_SIGNED_BIN_C
-#define BN_MP_READ_UNSIGNED_BIN_C
-#define BN_MP_REDUCE_C
-#define BN_MP_REDUCE_2K_C
-#define BN_MP_REDUCE_2K_L_C
-#define BN_MP_REDUCE_2K_SETUP_C
-#define BN_MP_REDUCE_2K_SETUP_L_C
-#define BN_MP_REDUCE_IS_2K_C
-#define BN_MP_REDUCE_IS_2K_L_C
-#define BN_MP_REDUCE_SETUP_C
-#define BN_MP_RSHD_C
-#define BN_MP_SET_C
-#define BN_MP_SET_INT_C
-#define BN_MP_SET_LONG_C
-#define BN_MP_SET_LONG_LONG_C
-#define BN_MP_SHRINK_C
-#define BN_MP_SIGNED_BIN_SIZE_C
-#define BN_MP_SQR_C
-#define BN_MP_SQRMOD_C
-#define BN_MP_SQRT_C
-#define BN_MP_SQRTMOD_PRIME_C
-#define BN_MP_SUB_C
-#define BN_MP_SUB_D_C
-#define BN_MP_SUBMOD_C
-#define BN_MP_TO_SIGNED_BIN_C
-#define BN_MP_TO_SIGNED_BIN_N_C
-#define BN_MP_TO_UNSIGNED_BIN_C
-#define BN_MP_TO_UNSIGNED_BIN_N_C
-#define BN_MP_TOOM_MUL_C
-#define BN_MP_TOOM_SQR_C
-#define BN_MP_TORADIX_C
-#define BN_MP_TORADIX_N_C
-#define BN_MP_UNSIGNED_BIN_SIZE_C
-#define BN_MP_XOR_C
-#define BN_MP_ZERO_C
-#define BN_PRIME_TAB_C
-#define BN_REVERSE_C
-#define BN_S_MP_ADD_C
-#define BN_S_MP_EXPTMOD_C
-#define BN_S_MP_MUL_DIGS_C
-#define BN_S_MP_MUL_HIGH_DIGS_C
-#define BN_S_MP_SQR_C
-#define BN_S_MP_SUB_C
-#define BNCORE_C
+#   define BN_ERROR_C
+#   define BN_FAST_MP_INVMOD_C
+#   define BN_FAST_MP_MONTGOMERY_REDUCE_C
+#   define BN_FAST_S_MP_MUL_DIGS_C
+#   define BN_FAST_S_MP_MUL_HIGH_DIGS_C
+#   define BN_FAST_S_MP_SQR_C
+#   define BN_MP_2EXPT_C
+#   define BN_MP_ABS_C
+#   define BN_MP_ADD_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_ADDMOD_C
+#   define BN_MP_AND_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_CMP_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_CNT_LSB_C
+#   define BN_MP_COPY_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_DIV_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_DIV_3_C
+#   define BN_MP_DIV_D_C
+#   define BN_MP_DR_IS_MODULUS_C
+#   define BN_MP_DR_REDUCE_C
+#   define BN_MP_DR_SETUP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_EXPORT_C
+#   define BN_MP_EXPT_D_C
+#   define BN_MP_EXPT_D_EX_C
+#   define BN_MP_EXPTMOD_C
+#   define BN_MP_EXPTMOD_FAST_C
+#   define BN_MP_EXTEUCLID_C
+#   define BN_MP_FREAD_C
+#   define BN_MP_FWRITE_C
+#   define BN_MP_GCD_C
+#   define BN_MP_GET_INT_C
+#   define BN_MP_GET_LONG_C
+#   define BN_MP_GET_LONG_LONG_C
+#   define BN_MP_GROW_C
+#   define BN_MP_IMPORT_C
+#   define BN_MP_INIT_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_INIT_SET_C
+#   define BN_MP_INIT_SET_INT_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_INVMOD_C
+#   define BN_MP_INVMOD_SLOW_C
+#   define BN_MP_IS_SQUARE_C
+#   define BN_MP_JACOBI_C
+#   define BN_MP_KARATSUBA_MUL_C
+#   define BN_MP_KARATSUBA_SQR_C
+#   define BN_MP_LCM_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_MOD_C
+#   define BN_MP_MOD_2D_C
+#   define BN_MP_MOD_D_C
+#   define BN_MP_MONTGOMERY_CALC_NORMALIZATION_C
+#   define BN_MP_MONTGOMERY_REDUCE_C
+#   define BN_MP_MONTGOMERY_SETUP_C
+#   define BN_MP_MUL_C
+#   define BN_MP_MUL_2_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_MULMOD_C
+#   define BN_MP_N_ROOT_C
+#   define BN_MP_N_ROOT_EX_C
+#   define BN_MP_NEG_C
+#   define BN_MP_OR_C
+#   define BN_MP_PRIME_FERMAT_C
+#   define BN_MP_PRIME_IS_DIVISIBLE_C
+#   define BN_MP_PRIME_IS_PRIME_C
+#   define BN_MP_PRIME_MILLER_RABIN_C
+#   define BN_MP_PRIME_NEXT_PRIME_C
+#   define BN_MP_PRIME_RABIN_MILLER_TRIALS_C
+#   define BN_MP_PRIME_RANDOM_EX_C
+#   define BN_MP_RADIX_SIZE_C
+#   define BN_MP_RADIX_SMAP_C
+#   define BN_MP_RAND_C
+#   define BN_MP_READ_RADIX_C
+#   define BN_MP_READ_SIGNED_BIN_C
+#   define BN_MP_READ_UNSIGNED_BIN_C
+#   define BN_MP_REDUCE_C
+#   define BN_MP_REDUCE_2K_C
+#   define BN_MP_REDUCE_2K_L_C
+#   define BN_MP_REDUCE_2K_SETUP_C
+#   define BN_MP_REDUCE_2K_SETUP_L_C
+#   define BN_MP_REDUCE_IS_2K_C
+#   define BN_MP_REDUCE_IS_2K_L_C
+#   define BN_MP_REDUCE_SETUP_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_SET_C
+#   define BN_MP_SET_INT_C
+#   define BN_MP_SET_LONG_C
+#   define BN_MP_SET_LONG_LONG_C
+#   define BN_MP_SHRINK_C
+#   define BN_MP_SIGNED_BIN_SIZE_C
+#   define BN_MP_SQR_C
+#   define BN_MP_SQRMOD_C
+#   define BN_MP_SQRT_C
+#   define BN_MP_SQRTMOD_PRIME_C
+#   define BN_MP_SUB_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_SUBMOD_C
+#   define BN_MP_TO_SIGNED_BIN_C
+#   define BN_MP_TO_SIGNED_BIN_N_C
+#   define BN_MP_TO_UNSIGNED_BIN_C
+#   define BN_MP_TO_UNSIGNED_BIN_N_C
+#   define BN_MP_TOOM_MUL_C
+#   define BN_MP_TOOM_SQR_C
+#   define BN_MP_TORADIX_C
+#   define BN_MP_TORADIX_N_C
+#   define BN_MP_UNSIGNED_BIN_SIZE_C
+#   define BN_MP_XOR_C
+#   define BN_MP_ZERO_C
+#   define BN_PRIME_TAB_C
+#   define BN_REVERSE_C
+#   define BN_S_MP_ADD_C
+#   define BN_S_MP_EXPTMOD_C
+#   define BN_S_MP_MUL_DIGS_C
+#   define BN_S_MP_MUL_HIGH_DIGS_C
+#   define BN_S_MP_SQR_C
+#   define BN_S_MP_SUB_C
+#   define BNCORE_C
 #endif
 
 #if defined(BN_ERROR_C)
-   #define BN_MP_ERROR_TO_STRING_C
+#   define BN_MP_ERROR_TO_STRING_C
 #endif
 
 #if defined(BN_FAST_MP_INVMOD_C)
-   #define BN_MP_ISEVEN_C
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_COPY_C
-   #define BN_MP_MOD_C
-   #define BN_MP_SET_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_ISODD_C
-   #define BN_MP_SUB_C
-   #define BN_MP_CMP_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_CMP_D_C
-   #define BN_MP_ADD_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_ISEVEN_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_COPY_C
+#   define BN_MP_MOD_C
+#   define BN_MP_SET_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_ISODD_C
+#   define BN_MP_SUB_C
+#   define BN_MP_CMP_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_ADD_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_FAST_MP_MONTGOMERY_REDUCE_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_MP_GROW_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_FAST_S_MP_MUL_DIGS_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_FAST_S_MP_MUL_HIGH_DIGS_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_FAST_S_MP_SQR_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_2EXPT_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_GROW_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_GROW_C
 #endif
 
 #if defined(BN_MP_ABS_C)
-   #define BN_MP_COPY_C
+#   define BN_MP_COPY_C
 #endif
 
 #if defined(BN_MP_ADD_C)
-   #define BN_S_MP_ADD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_S_MP_ADD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_ADD_D_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_ADDMOD_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_ADD_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_MOD_C
+#   define BN_MP_INIT_C
+#   define BN_MP_ADD_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_MOD_C
 #endif
 
 #if defined(BN_MP_AND_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_CLAMP_C)
@@ -224,11 +224,11 @@
 #endif
 
 #if defined(BN_MP_CLEAR_MULTI_C)
-   #define BN_MP_CLEAR_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_CMP_C)
-   #define BN_MP_CMP_MAG_C
+#   define BN_MP_CMP_MAG_C
 #endif
 
 #if defined(BN_MP_CMP_D_C)
@@ -238,81 +238,81 @@
 #endif
 
 #if defined(BN_MP_CNT_LSB_C)
-   #define BN_MP_ISZERO_C
+#   define BN_MP_ISZERO_C
 #endif
 
 #if defined(BN_MP_COPY_C)
-   #define BN_MP_GROW_C
+#   define BN_MP_GROW_C
 #endif
 
 #if defined(BN_MP_COUNT_BITS_C)
 #endif
 
 #if defined(BN_MP_DIV_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_COPY_C
-   #define BN_MP_ZERO_C
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_SET_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_ABS_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_CMP_C
-   #define BN_MP_SUB_C
-   #define BN_MP_ADD_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_MULTI_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_INIT_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_COPY_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_SET_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_ABS_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_CMP_C
+#   define BN_MP_SUB_C
+#   define BN_MP_ADD_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_INIT_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_DIV_2_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_DIV_2D_C)
-   #define BN_MP_COPY_C
-   #define BN_MP_ZERO_C
-   #define BN_MP_MOD_2D_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_COPY_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_MOD_2D_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_DIV_3_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_DIV_D_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_COPY_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_DIV_3_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_COPY_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_DIV_3_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_DR_IS_MODULUS_C)
 #endif
 
 #if defined(BN_MP_DR_REDUCE_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_DR_SETUP_C)
@@ -322,96 +322,96 @@
 #endif
 
 #if defined(BN_MP_EXPORT_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_EXPT_D_C)
-   #define BN_MP_EXPT_D_EX_C
+#   define BN_MP_EXPT_D_EX_C
 #endif
 
 #if defined(BN_MP_EXPT_D_EX_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_SET_C
-   #define BN_MP_MUL_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_SQR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_SET_C
+#   define BN_MP_MUL_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_SQR_C
 #endif
 
 #if defined(BN_MP_EXPTMOD_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_INVMOD_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_ABS_C
-   #define BN_MP_CLEAR_MULTI_C
-   #define BN_MP_REDUCE_IS_2K_L_C
-   #define BN_S_MP_EXPTMOD_C
-   #define BN_MP_DR_IS_MODULUS_C
-   #define BN_MP_REDUCE_IS_2K_C
-   #define BN_MP_ISODD_C
-   #define BN_MP_EXPTMOD_FAST_C
+#   define BN_MP_INIT_C
+#   define BN_MP_INVMOD_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_ABS_C
+#   define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_REDUCE_IS_2K_L_C
+#   define BN_S_MP_EXPTMOD_C
+#   define BN_MP_DR_IS_MODULUS_C
+#   define BN_MP_REDUCE_IS_2K_C
+#   define BN_MP_ISODD_C
+#   define BN_MP_EXPTMOD_FAST_C
 #endif
 
 #if defined(BN_MP_EXPTMOD_FAST_C)
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_MONTGOMERY_SETUP_C
-   #define BN_FAST_MP_MONTGOMERY_REDUCE_C
-   #define BN_MP_MONTGOMERY_REDUCE_C
-   #define BN_MP_DR_SETUP_C
-   #define BN_MP_DR_REDUCE_C
-   #define BN_MP_REDUCE_2K_SETUP_C
-   #define BN_MP_REDUCE_2K_C
-   #define BN_MP_MONTGOMERY_CALC_NORMALIZATION_C
-   #define BN_MP_MULMOD_C
-   #define BN_MP_SET_C
-   #define BN_MP_MOD_C
-   #define BN_MP_COPY_C
-   #define BN_MP_SQR_C
-   #define BN_MP_MUL_C
-   #define BN_MP_EXCH_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_MONTGOMERY_SETUP_C
+#   define BN_FAST_MP_MONTGOMERY_REDUCE_C
+#   define BN_MP_MONTGOMERY_REDUCE_C
+#   define BN_MP_DR_SETUP_C
+#   define BN_MP_DR_REDUCE_C
+#   define BN_MP_REDUCE_2K_SETUP_C
+#   define BN_MP_REDUCE_2K_C
+#   define BN_MP_MONTGOMERY_CALC_NORMALIZATION_C
+#   define BN_MP_MULMOD_C
+#   define BN_MP_SET_C
+#   define BN_MP_MOD_C
+#   define BN_MP_COPY_C
+#   define BN_MP_SQR_C
+#   define BN_MP_MUL_C
+#   define BN_MP_EXCH_C
 #endif
 
 #if defined(BN_MP_EXTEUCLID_C)
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_SET_C
-   #define BN_MP_COPY_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_DIV_C
-   #define BN_MP_MUL_C
-   #define BN_MP_SUB_C
-   #define BN_MP_NEG_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_SET_C
+#   define BN_MP_COPY_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_DIV_C
+#   define BN_MP_MUL_C
+#   define BN_MP_SUB_C
+#   define BN_MP_NEG_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_FREAD_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_S_RMAP_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_CMP_D_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_S_RMAP_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_CMP_D_C
 #endif
 
 #if defined(BN_MP_FWRITE_C)
-   #define BN_MP_RADIX_SIZE_C
-   #define BN_MP_TORADIX_C
+#   define BN_MP_RADIX_SIZE_C
+#   define BN_MP_TORADIX_C
 #endif
 
 #if defined(BN_MP_GCD_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_ABS_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_CNT_LSB_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_EXCH_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_ABS_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_CNT_LSB_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_EXCH_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_GET_INT_C)
@@ -427,402 +427,402 @@
 #endif
 
 #if defined(BN_MP_IMPORT_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_INIT_C)
 #endif
 
 #if defined(BN_MP_INIT_COPY_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_COPY_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_COPY_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_INIT_MULTI_C)
-   #define BN_MP_ERR_C
-   #define BN_MP_INIT_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ERR_C
+#   define BN_MP_INIT_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_INIT_SET_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_SET_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SET_C
 #endif
 
 #if defined(BN_MP_INIT_SET_INT_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_SET_INT_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SET_INT_C
 #endif
 
 #if defined(BN_MP_INIT_SIZE_C)
-   #define BN_MP_INIT_C
+#   define BN_MP_INIT_C
 #endif
 
 #if defined(BN_MP_INVMOD_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_ISODD_C
-   #define BN_MP_CMP_D_C
-   #define BN_FAST_MP_INVMOD_C
-   #define BN_MP_INVMOD_SLOW_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_ISODD_C
+#   define BN_MP_CMP_D_C
+#   define BN_FAST_MP_INVMOD_C
+#   define BN_MP_INVMOD_SLOW_C
 #endif
 
 #if defined(BN_MP_INVMOD_SLOW_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_MOD_C
-   #define BN_MP_COPY_C
-   #define BN_MP_ISEVEN_C
-   #define BN_MP_SET_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_ISODD_C
-   #define BN_MP_ADD_C
-   #define BN_MP_SUB_C
-   #define BN_MP_CMP_C
-   #define BN_MP_CMP_D_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_MOD_C
+#   define BN_MP_COPY_C
+#   define BN_MP_ISEVEN_C
+#   define BN_MP_SET_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_ISODD_C
+#   define BN_MP_ADD_C
+#   define BN_MP_SUB_C
+#   define BN_MP_CMP_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_IS_SQUARE_C)
-   #define BN_MP_MOD_D_C
-   #define BN_MP_INIT_SET_INT_C
-   #define BN_MP_MOD_C
-   #define BN_MP_GET_INT_C
-   #define BN_MP_SQRT_C
-   #define BN_MP_SQR_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_MOD_D_C
+#   define BN_MP_INIT_SET_INT_C
+#   define BN_MP_MOD_C
+#   define BN_MP_GET_INT_C
+#   define BN_MP_SQRT_C
+#   define BN_MP_SQR_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_JACOBI_C)
-   #define BN_MP_ISNEG_C
-   #define BN_MP_CMP_D_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_CNT_LSB_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_MOD_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ISNEG_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_CNT_LSB_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_MOD_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_KARATSUBA_MUL_C)
-   #define BN_MP_MUL_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_S_MP_ADD_C
-   #define BN_MP_ADD_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_MUL_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_S_MP_ADD_C
+#   define BN_MP_ADD_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_KARATSUBA_SQR_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_SQR_C
-   #define BN_S_MP_ADD_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_ADD_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_SQR_C
+#   define BN_S_MP_ADD_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_ADD_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_LCM_C)
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_GCD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_DIV_C
-   #define BN_MP_MUL_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_GCD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_DIV_C
+#   define BN_MP_MUL_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_LSHD_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_RSHD_C
+#   define BN_MP_GROW_C
+#   define BN_MP_RSHD_C
 #endif
 
 #if defined(BN_MP_MOD_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_DIV_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_ADD_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_DIV_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_ADD_C
 #endif
 
 #if defined(BN_MP_MOD_2D_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_COPY_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_COPY_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_MOD_D_C)
-   #define BN_MP_DIV_D_C
+#   define BN_MP_DIV_D_C
 #endif
 
 #if defined(BN_MP_MONTGOMERY_CALC_NORMALIZATION_C)
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_2EXPT_C
-   #define BN_MP_SET_C
-   #define BN_MP_MUL_2_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_2EXPT_C
+#   define BN_MP_SET_C
+#   define BN_MP_MUL_2_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_MONTGOMERY_REDUCE_C)
-   #define BN_FAST_MP_MONTGOMERY_REDUCE_C
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_FAST_MP_MONTGOMERY_REDUCE_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_MONTGOMERY_SETUP_C)
 #endif
 
 #if defined(BN_MP_MUL_C)
-   #define BN_MP_TOOM_MUL_C
-   #define BN_MP_KARATSUBA_MUL_C
-   #define BN_FAST_S_MP_MUL_DIGS_C
-   #define BN_S_MP_MUL_C
-   #define BN_S_MP_MUL_DIGS_C
+#   define BN_MP_TOOM_MUL_C
+#   define BN_MP_KARATSUBA_MUL_C
+#   define BN_FAST_S_MP_MUL_DIGS_C
+#   define BN_S_MP_MUL_C
+#   define BN_S_MP_MUL_DIGS_C
 #endif
 
 #if defined(BN_MP_MUL_2_C)
-   #define BN_MP_GROW_C
+#   define BN_MP_GROW_C
 #endif
 
 #if defined(BN_MP_MUL_2D_C)
-   #define BN_MP_COPY_C
-   #define BN_MP_GROW_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_COPY_C
+#   define BN_MP_GROW_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_MUL_D_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_MULMOD_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_MUL_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_MOD_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_MUL_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_MOD_C
 #endif
 
 #if defined(BN_MP_N_ROOT_C)
-   #define BN_MP_N_ROOT_EX_C
+#   define BN_MP_N_ROOT_EX_C
 #endif
 
 #if defined(BN_MP_N_ROOT_EX_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_SET_C
-   #define BN_MP_COPY_C
-   #define BN_MP_EXPT_D_EX_C
-   #define BN_MP_MUL_C
-   #define BN_MP_SUB_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_DIV_C
-   #define BN_MP_CMP_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SET_C
+#   define BN_MP_COPY_C
+#   define BN_MP_EXPT_D_EX_C
+#   define BN_MP_MUL_C
+#   define BN_MP_SUB_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_DIV_C
+#   define BN_MP_CMP_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_NEG_C)
-   #define BN_MP_COPY_C
-   #define BN_MP_ISZERO_C
+#   define BN_MP_COPY_C
+#   define BN_MP_ISZERO_C
 #endif
 
 #if defined(BN_MP_OR_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_PRIME_FERMAT_C)
-   #define BN_MP_CMP_D_C
-   #define BN_MP_INIT_C
-   #define BN_MP_EXPTMOD_C
-   #define BN_MP_CMP_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_INIT_C
+#   define BN_MP_EXPTMOD_C
+#   define BN_MP_CMP_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_PRIME_IS_DIVISIBLE_C)
-   #define BN_MP_MOD_D_C
+#   define BN_MP_MOD_D_C
 #endif
 
 #if defined(BN_MP_PRIME_IS_PRIME_C)
-   #define BN_MP_CMP_D_C
-   #define BN_MP_PRIME_IS_DIVISIBLE_C
-   #define BN_MP_INIT_C
-   #define BN_MP_SET_C
-   #define BN_MP_PRIME_MILLER_RABIN_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_PRIME_IS_DIVISIBLE_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SET_C
+#   define BN_MP_PRIME_MILLER_RABIN_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_PRIME_MILLER_RABIN_C)
-   #define BN_MP_CMP_D_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_CNT_LSB_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_EXPTMOD_C
-   #define BN_MP_CMP_C
-   #define BN_MP_SQRMOD_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_CNT_LSB_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_EXPTMOD_C
+#   define BN_MP_CMP_C
+#   define BN_MP_SQRMOD_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_PRIME_NEXT_PRIME_C)
-   #define BN_MP_CMP_D_C
-   #define BN_MP_SET_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_ISEVEN_C
-   #define BN_MP_MOD_D_C
-   #define BN_MP_INIT_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_PRIME_MILLER_RABIN_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_SET_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_ISEVEN_C
+#   define BN_MP_MOD_D_C
+#   define BN_MP_INIT_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_PRIME_MILLER_RABIN_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_PRIME_RABIN_MILLER_TRIALS_C)
 #endif
 
 #if defined(BN_MP_PRIME_RANDOM_EX_C)
-   #define BN_MP_READ_UNSIGNED_BIN_C
-   #define BN_MP_PRIME_IS_PRIME_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_MUL_2_C
-   #define BN_MP_ADD_D_C
+#   define BN_MP_READ_UNSIGNED_BIN_C
+#   define BN_MP_PRIME_IS_PRIME_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_MUL_2_C
+#   define BN_MP_ADD_D_C
 #endif
 
 #if defined(BN_MP_RADIX_SIZE_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_DIV_D_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_DIV_D_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_RADIX_SMAP_C)
-   #define BN_MP_S_RMAP_C
+#   define BN_MP_S_RMAP_C
 #endif
 
 #if defined(BN_MP_RAND_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_LSHD_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_LSHD_C
 #endif
 
 #if defined(BN_MP_READ_RADIX_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_S_RMAP_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_ISZERO_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_S_RMAP_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_ISZERO_C
 #endif
 
 #if defined(BN_MP_READ_SIGNED_BIN_C)
-   #define BN_MP_READ_UNSIGNED_BIN_C
+#   define BN_MP_READ_UNSIGNED_BIN_C
 #endif
 
 #if defined(BN_MP_READ_UNSIGNED_BIN_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_ZERO_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_REDUCE_C)
-   #define BN_MP_REDUCE_SETUP_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_MUL_C
-   #define BN_S_MP_MUL_HIGH_DIGS_C
-   #define BN_FAST_S_MP_MUL_HIGH_DIGS_C
-   #define BN_MP_MOD_2D_C
-   #define BN_S_MP_MUL_DIGS_C
-   #define BN_MP_SUB_C
-   #define BN_MP_CMP_D_C
-   #define BN_MP_SET_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_ADD_C
-   #define BN_MP_CMP_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_REDUCE_SETUP_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_MUL_C
+#   define BN_S_MP_MUL_HIGH_DIGS_C
+#   define BN_FAST_S_MP_MUL_HIGH_DIGS_C
+#   define BN_MP_MOD_2D_C
+#   define BN_S_MP_MUL_DIGS_C
+#   define BN_MP_SUB_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_SET_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_ADD_C
+#   define BN_MP_CMP_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_REDUCE_2K_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_MUL_D_C
-   #define BN_S_MP_ADD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_MUL_D_C
+#   define BN_S_MP_ADD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_REDUCE_2K_L_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_MUL_C
-   #define BN_S_MP_ADD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_MUL_C
+#   define BN_S_MP_ADD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_REDUCE_2K_SETUP_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_2EXPT_C
-   #define BN_MP_CLEAR_C
-   #define BN_S_MP_SUB_C
+#   define BN_MP_INIT_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_2EXPT_C
+#   define BN_MP_CLEAR_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_REDUCE_2K_SETUP_L_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_2EXPT_C
-   #define BN_MP_COUNT_BITS_C
-   #define BN_S_MP_SUB_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_C
+#   define BN_MP_2EXPT_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_S_MP_SUB_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_REDUCE_IS_2K_C)
-   #define BN_MP_REDUCE_2K_C
-   #define BN_MP_COUNT_BITS_C
+#   define BN_MP_REDUCE_2K_C
+#   define BN_MP_COUNT_BITS_C
 #endif
 
 #if defined(BN_MP_REDUCE_IS_2K_L_C)
 #endif
 
 #if defined(BN_MP_REDUCE_SETUP_C)
-   #define BN_MP_2EXPT_C
-   #define BN_MP_DIV_C
+#   define BN_MP_2EXPT_C
+#   define BN_MP_DIV_C
 #endif
 
 #if defined(BN_MP_RSHD_C)
-   #define BN_MP_ZERO_C
+#   define BN_MP_ZERO_C
 #endif
 
 #if defined(BN_MP_SET_C)
-   #define BN_MP_ZERO_C
+#   define BN_MP_ZERO_C
 #endif
 
 #if defined(BN_MP_SET_INT_C)
-   #define BN_MP_ZERO_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_SET_LONG_C)
@@ -835,155 +835,155 @@
 #endif
 
 #if defined(BN_MP_SIGNED_BIN_SIZE_C)
-   #define BN_MP_UNSIGNED_BIN_SIZE_C
+#   define BN_MP_UNSIGNED_BIN_SIZE_C
 #endif
 
 #if defined(BN_MP_SQR_C)
-   #define BN_MP_TOOM_SQR_C
-   #define BN_MP_KARATSUBA_SQR_C
-   #define BN_FAST_S_MP_SQR_C
-   #define BN_S_MP_SQR_C
+#   define BN_MP_TOOM_SQR_C
+#   define BN_MP_KARATSUBA_SQR_C
+#   define BN_FAST_S_MP_SQR_C
+#   define BN_S_MP_SQR_C
 #endif
 
 #if defined(BN_MP_SQRMOD_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_SQR_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_MOD_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SQR_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_MOD_C
 #endif
 
 #if defined(BN_MP_SQRT_C)
-   #define BN_MP_N_ROOT_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_ZERO_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_DIV_C
-   #define BN_MP_ADD_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_N_ROOT_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_DIV_C
+#   define BN_MP_ADD_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_SQRTMOD_PRIME_C)
-   #define BN_MP_CMP_D_C
-   #define BN_MP_ZERO_C
-   #define BN_MP_JACOBI_C
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_MOD_D_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_EXPTMOD_C
-   #define BN_MP_COPY_C
-   #define BN_MP_SUB_D_C
-   #define BN_MP_ISEVEN_C
-   #define BN_MP_SET_INT_C
-   #define BN_MP_SQRMOD_C
-   #define BN_MP_MULMOD_C
-   #define BN_MP_SET_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_CMP_D_C
+#   define BN_MP_ZERO_C
+#   define BN_MP_JACOBI_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_MOD_D_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_EXPTMOD_C
+#   define BN_MP_COPY_C
+#   define BN_MP_SUB_D_C
+#   define BN_MP_ISEVEN_C
+#   define BN_MP_SET_INT_C
+#   define BN_MP_SQRMOD_C
+#   define BN_MP_MULMOD_C
+#   define BN_MP_SET_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_SUB_C)
-   #define BN_S_MP_ADD_C
-   #define BN_MP_CMP_MAG_C
-   #define BN_S_MP_SUB_C
+#   define BN_S_MP_ADD_C
+#   define BN_MP_CMP_MAG_C
+#   define BN_S_MP_SUB_C
 #endif
 
 #if defined(BN_MP_SUB_D_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_ADD_D_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_ADD_D_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_MP_SUBMOD_C)
-   #define BN_MP_INIT_C
-   #define BN_MP_SUB_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_MOD_C
+#   define BN_MP_INIT_C
+#   define BN_MP_SUB_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_MOD_C
 #endif
 
 #if defined(BN_MP_TO_SIGNED_BIN_C)
-   #define BN_MP_TO_UNSIGNED_BIN_C
+#   define BN_MP_TO_UNSIGNED_BIN_C
 #endif
 
 #if defined(BN_MP_TO_SIGNED_BIN_N_C)
-   #define BN_MP_SIGNED_BIN_SIZE_C
-   #define BN_MP_TO_SIGNED_BIN_C
+#   define BN_MP_SIGNED_BIN_SIZE_C
+#   define BN_MP_TO_SIGNED_BIN_C
 #endif
 
 #if defined(BN_MP_TO_UNSIGNED_BIN_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_ISZERO_C
-   #define BN_MP_DIV_2D_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_DIV_2D_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_TO_UNSIGNED_BIN_N_C)
-   #define BN_MP_UNSIGNED_BIN_SIZE_C
-   #define BN_MP_TO_UNSIGNED_BIN_C
+#   define BN_MP_UNSIGNED_BIN_SIZE_C
+#   define BN_MP_TO_UNSIGNED_BIN_C
 #endif
 
 #if defined(BN_MP_TOOM_MUL_C)
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_MOD_2D_C
-   #define BN_MP_COPY_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_MUL_C
-   #define BN_MP_MUL_2_C
-   #define BN_MP_ADD_C
-   #define BN_MP_SUB_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_DIV_3_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_MOD_2D_C
+#   define BN_MP_COPY_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_MUL_C
+#   define BN_MP_MUL_2_C
+#   define BN_MP_ADD_C
+#   define BN_MP_SUB_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_DIV_3_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_TOOM_SQR_C)
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_MOD_2D_C
-   #define BN_MP_COPY_C
-   #define BN_MP_RSHD_C
-   #define BN_MP_SQR_C
-   #define BN_MP_MUL_2_C
-   #define BN_MP_ADD_C
-   #define BN_MP_SUB_C
-   #define BN_MP_DIV_2_C
-   #define BN_MP_MUL_2D_C
-   #define BN_MP_MUL_D_C
-   #define BN_MP_DIV_3_C
-   #define BN_MP_LSHD_C
-   #define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_MOD_2D_C
+#   define BN_MP_COPY_C
+#   define BN_MP_RSHD_C
+#   define BN_MP_SQR_C
+#   define BN_MP_MUL_2_C
+#   define BN_MP_ADD_C
+#   define BN_MP_SUB_C
+#   define BN_MP_DIV_2_C
+#   define BN_MP_MUL_2D_C
+#   define BN_MP_MUL_D_C
+#   define BN_MP_DIV_3_C
+#   define BN_MP_LSHD_C
+#   define BN_MP_CLEAR_MULTI_C
 #endif
 
 #if defined(BN_MP_TORADIX_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_DIV_D_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_S_RMAP_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_DIV_D_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_S_RMAP_C
 #endif
 
 #if defined(BN_MP_TORADIX_N_C)
-   #define BN_MP_ISZERO_C
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_DIV_D_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_S_RMAP_C
+#   define BN_MP_ISZERO_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_DIV_D_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_S_RMAP_C
 #endif
 
 #if defined(BN_MP_UNSIGNED_BIN_SIZE_C)
-   #define BN_MP_COUNT_BITS_C
+#   define BN_MP_COUNT_BITS_C
 #endif
 
 #if defined(BN_MP_XOR_C)
-   #define BN_MP_INIT_COPY_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_COPY_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_MP_ZERO_C)
@@ -996,62 +996,63 @@
 #endif
 
 #if defined(BN_S_MP_ADD_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BN_S_MP_EXPTMOD_C)
-   #define BN_MP_COUNT_BITS_C
-   #define BN_MP_INIT_C
-   #define BN_MP_CLEAR_C
-   #define BN_MP_REDUCE_SETUP_C
-   #define BN_MP_REDUCE_C
-   #define BN_MP_REDUCE_2K_SETUP_L_C
-   #define BN_MP_REDUCE_2K_L_C
-   #define BN_MP_MOD_C
-   #define BN_MP_COPY_C
-   #define BN_MP_SQR_C
-   #define BN_MP_MUL_C
-   #define BN_MP_SET_C
-   #define BN_MP_EXCH_C
+#   define BN_MP_COUNT_BITS_C
+#   define BN_MP_INIT_C
+#   define BN_MP_CLEAR_C
+#   define BN_MP_REDUCE_SETUP_C
+#   define BN_MP_REDUCE_C
+#   define BN_MP_REDUCE_2K_SETUP_L_C
+#   define BN_MP_REDUCE_2K_L_C
+#   define BN_MP_MOD_C
+#   define BN_MP_COPY_C
+#   define BN_MP_SQR_C
+#   define BN_MP_MUL_C
+#   define BN_MP_SET_C
+#   define BN_MP_EXCH_C
 #endif
 
 #if defined(BN_S_MP_MUL_DIGS_C)
-   #define BN_FAST_S_MP_MUL_DIGS_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_FAST_S_MP_MUL_DIGS_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_S_MP_MUL_HIGH_DIGS_C)
-   #define BN_FAST_S_MP_MUL_HIGH_DIGS_C
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_FAST_S_MP_MUL_HIGH_DIGS_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_S_MP_SQR_C)
-   #define BN_MP_INIT_SIZE_C
-   #define BN_MP_CLAMP_C
-   #define BN_MP_EXCH_C
-   #define BN_MP_CLEAR_C
+#   define BN_MP_INIT_SIZE_C
+#   define BN_MP_CLAMP_C
+#   define BN_MP_EXCH_C
+#   define BN_MP_CLEAR_C
 #endif
 
 #if defined(BN_S_MP_SUB_C)
-   #define BN_MP_GROW_C
-   #define BN_MP_CLAMP_C
+#   define BN_MP_GROW_C
+#   define BN_MP_CLAMP_C
 #endif
 
 #if defined(BNCORE_C)
 #endif
 
 #ifdef LTM3
-#define LTM_LAST
+#   define LTM_LAST
 #endif
+
 #include <tommath_superclass.h>
 #include <tommath_class.h>
 #else
-#define LTM_LAST
+#   define LTM_LAST
 #endif
diff --git a/libtommath/tommath_private.h b/libtommath/tommath_private.h
index aeda684..ee39694 100644
--- a/libtommath/tommath_private.h
+++ b/libtommath/tommath_private.h
@@ -19,39 +19,39 @@
 #include <ctype.h>
 
 #ifndef MIN
-  #define MIN(x,y) (((x) < (y)) ? (x) : (y))
+#define MIN(x, y) (((x) < (y)) ? (x) : (y))
 #endif
 
 #ifndef MAX
-  #define MAX(x,y) (((x) > (y)) ? (x) : (y))
+#define MAX(x, y) (((x) > (y)) ? (x) : (y))
 #endif
 
 #ifdef __cplusplus
 extern "C" {
 
 /* C++ compilers don't like assigning void * to mp_digit * */
-#define  OPT_CAST(x)  (x *)
+#define OPT_CAST(x) (x *)
 
 #else
 
 /* C on the other hand doesn't care */
-#define  OPT_CAST(x)
+#define OPT_CAST(x)
 
 #endif
 
 /* define heap macros */
 #ifndef XMALLOC
-   /* default to libc stuff */
-   #define XMALLOC  malloc
-   #define XFREE    free
-   #define XREALLOC realloc
-   #define XCALLOC  calloc
+/* default to libc stuff */
+#   define XMALLOC   malloc
+#   define XFREE     free
+#   define XREALLOC  realloc
+#   define XCALLOC   calloc
 #else
-   /* prototypes for our heap functions */
-   extern void *XMALLOC(size_t n);
-   extern void *XREALLOC(void *p, size_t n);
-   extern void *XCALLOC(size_t n, size_t s);
-   extern void XFREE(void *p);
+/* prototypes for our heap functions */
+extern void *XMALLOC(size_t n);
+extern void *XREALLOC(void *p, size_t n);
+extern void *XCALLOC(size_t n, size_t s);
+extern void XFREE(void *p);
 #endif
 
 /* lowlevel functions, do not call! */
@@ -69,10 +69,10 @@ int mp_toom_mul(mp_int *a, mp_int *b, mp_int *c);
 int mp_karatsuba_sqr(mp_int *a, mp_int *b);
 int mp_toom_sqr(mp_int *a, mp_int *b);
 int fast_mp_invmod(mp_int *a, mp_int *b, mp_int *c);
-int mp_invmod_slow (mp_int * a, mp_int * b, mp_int * c);
+int mp_invmod_slow(mp_int *a, mp_int *b, mp_int *c);
 int fast_mp_montgomery_reduce(mp_int *x, mp_int *n, mp_digit rho);
 int mp_exptmod_fast(mp_int *G, mp_int *X, mp_int *P, mp_int *Y, int redmode);
-int s_mp_exptmod (mp_int * G, mp_int * X, mp_int * P, mp_int * Y, int redmode);
+int s_mp_exptmod(mp_int *G, mp_int *X, mp_int *P, mp_int *Y, int redmode);
 void bn_reverse(unsigned char *s, int len);
 
 extern const char *mp_s_rmap;
@@ -112,7 +112,7 @@ int func_name (mp_int * a, type b)                       \
 }
 
 #ifdef __cplusplus
-   }
+}
 #endif
 
 #endif
diff --git a/libtommath/tommath_superclass.h b/libtommath/tommath_superclass.h
index a2f4d93..da53793 100644
--- a/libtommath/tommath_superclass.h
+++ b/libtommath/tommath_superclass.h
@@ -14,60 +14,60 @@
 
 /* Works for RSA only, mpi.o is 68KiB */
 #ifdef SC_RSA_1
-   #define BN_MP_SHRINK_C
-   #define BN_MP_LCM_C
-   #define BN_MP_PRIME_RANDOM_EX_C
-   #define BN_MP_INVMOD_C
-   #define BN_MP_GCD_C
-   #define BN_MP_MOD_C
-   #define BN_MP_MULMOD_C
-   #define BN_MP_ADDMOD_C
-   #define BN_MP_EXPTMOD_C
-   #define BN_MP_SET_INT_C
-   #define BN_MP_INIT_MULTI_C
-   #define BN_MP_CLEAR_MULTI_C
-   #define BN_MP_UNSIGNED_BIN_SIZE_C
-   #define BN_MP_TO_UNSIGNED_BIN_C
-   #define BN_MP_MOD_D_C
-   #define BN_MP_PRIME_RABIN_MILLER_TRIALS_C
-   #define BN_REVERSE_C
-   #define BN_PRIME_TAB_C
+#   define BN_MP_SHRINK_C
+#   define BN_MP_LCM_C
+#   define BN_MP_PRIME_RANDOM_EX_C
+#   define BN_MP_INVMOD_C
+#   define BN_MP_GCD_C
+#   define BN_MP_MOD_C
+#   define BN_MP_MULMOD_C
+#   define BN_MP_ADDMOD_C
+#   define BN_MP_EXPTMOD_C
+#   define BN_MP_SET_INT_C
+#   define BN_MP_INIT_MULTI_C
+#   define BN_MP_CLEAR_MULTI_C
+#   define BN_MP_UNSIGNED_BIN_SIZE_C
+#   define BN_MP_TO_UNSIGNED_BIN_C
+#   define BN_MP_MOD_D_C
+#   define BN_MP_PRIME_RABIN_MILLER_TRIALS_C
+#   define BN_REVERSE_C
+#   define BN_PRIME_TAB_C
 
-   /* other modifiers */
-   #define BN_MP_DIV_SMALL                    /* Slower division, not critical */
+/* other modifiers */
+#   define BN_MP_DIV_SMALL                    /* Slower division, not critical */
 
-   /* here we are on the last pass so we turn things off.  The functions classes are still there
-    * but we remove them specifically from the build.  This also invokes tweaks in functions
-    * like removing support for even moduli, etc...
-    */
-#ifdef LTM_LAST
-   #undef  BN_MP_TOOM_MUL_C
-   #undef  BN_MP_TOOM_SQR_C
-   #undef  BN_MP_KARATSUBA_MUL_C
-   #undef  BN_MP_KARATSUBA_SQR_C
-   #undef  BN_MP_REDUCE_C
-   #undef  BN_MP_REDUCE_SETUP_C
-   #undef  BN_MP_DR_IS_MODULUS_C
-   #undef  BN_MP_DR_SETUP_C
-   #undef  BN_MP_DR_REDUCE_C
-   #undef  BN_MP_REDUCE_IS_2K_C
-   #undef  BN_MP_REDUCE_2K_SETUP_C
-   #undef  BN_MP_REDUCE_2K_C
-   #undef  BN_S_MP_EXPTMOD_C
-   #undef  BN_MP_DIV_3_C
-   #undef  BN_S_MP_MUL_HIGH_DIGS_C
-   #undef  BN_FAST_S_MP_MUL_HIGH_DIGS_C
-   #undef  BN_FAST_MP_INVMOD_C
+/* here we are on the last pass so we turn things off.  The functions classes are still there
+ * but we remove them specifically from the build.  This also invokes tweaks in functions
+ * like removing support for even moduli, etc...
+ */
+#   ifdef LTM_LAST
+#      undef BN_MP_TOOM_MUL_C
+#      undef BN_MP_TOOM_SQR_C
+#      undef BN_MP_KARATSUBA_MUL_C
+#      undef BN_MP_KARATSUBA_SQR_C
+#      undef BN_MP_REDUCE_C
+#      undef BN_MP_REDUCE_SETUP_C
+#      undef BN_MP_DR_IS_MODULUS_C
+#      undef BN_MP_DR_SETUP_C
+#      undef BN_MP_DR_REDUCE_C
+#      undef BN_MP_REDUCE_IS_2K_C
+#      undef BN_MP_REDUCE_2K_SETUP_C
+#      undef BN_MP_REDUCE_2K_C
+#      undef BN_S_MP_EXPTMOD_C
+#      undef BN_MP_DIV_3_C
+#      undef BN_S_MP_MUL_HIGH_DIGS_C
+#      undef BN_FAST_S_MP_MUL_HIGH_DIGS_C
+#      undef BN_FAST_MP_INVMOD_C
 
-   /* To safely undefine these you have to make sure your RSA key won't exceed the Comba threshold
-    * which is roughly 255 digits [7140 bits for 32-bit machines, 15300 bits for 64-bit machines] 
-    * which means roughly speaking you can handle upto 2536-bit RSA keys with these defined without
-    * trouble.  
-    */
-   #undef  BN_S_MP_MUL_DIGS_C
-   #undef  BN_S_MP_SQR_C
-   #undef  BN_MP_MONTGOMERY_REDUCE_C
-#endif
+/* To safely undefine these you have to make sure your RSA key won't exceed the Comba threshold
+ * which is roughly 255 digits [7140 bits for 32-bit machines, 15300 bits for 64-bit machines]
+ * which means roughly speaking you can handle upto 2536-bit RSA keys with these defined without
+ * trouble.
+ */
+#      undef BN_S_MP_MUL_DIGS_C
+#      undef BN_S_MP_SQR_C
+#      undef BN_MP_MONTGOMERY_REDUCE_C
+#   endif
 
 #endif