Replay svn r88850.v2.6.7

re-format everything through astyle. Taken from libtom/libtommath (pull request [https://github.com/libtom/libtommath/pull/85|85])
author: Barry Warsaw <barry@python.org> 2011-06-04 00:02:47 (GMT)
committer: Barry Warsaw <barry@python.org> 2011-06-04 00:02:47 (GMT)
commit: 16ec24a1920dd4e94e2890032b5c4160621d526d (patch)
tree: b1222fde01e4cb336002e8319ee1467d3b16c86c /PC
parent: 9c53584ebdb45c27d4b9c7cb7235bc5bb77c610f (diff)
download: cpython-2.6.7.zip
cpython-2.6.7.tar.gz
cpython-2.6.7.tar.bz2
0 files changed, 0 insertions, 0 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
author	Barry Warsaw <barry@python.org>	2011-06-04 00:02:47 (GMT)
committer	Barry Warsaw <barry@python.org>	2011-06-04 00:02:47 (GMT)
commit	16ec24a1920dd4e94e2890032b5c4160621d526d (patch)
tree	b1222fde01e4cb336002e8319ee1467d3b16c86c /PC
parent	9c53584ebdb45c27d4b9c7cb7235bc5bb77c610f (diff)
download	cpython-2.6.7.zip cpython-2.6.7.tar.gz cpython-2.6.7.tar.bz2
author	jan.nijtmans <nijtmans@users.sourceforge.net>	2017-09-15 14:44:00 (GMT)
committer	jan.nijtmans <nijtmans@users.sourceforge.net>	2017-09-15 14:44:00 (GMT)
commit	30f7e69b182d1267056ee2628a860891f6555aa3 (patch)
tree	4598f9d64ea232476fcf18bbef6bd413444c0c07
parent	b167e2845e3a2d5b2ffe3a877cb19c3f430aeacf (diff)
download	tcl-30f7e69b182d1267056ee2628a860891f6555aa3.zip tcl-30f7e69b182d1267056ee2628a860891f6555aa3.tar.gz tcl-30f7e69b182d1267056ee2628a860891f6555aa3.tar.bz2