136 files changed, 12193 insertions, 5783 deletions

diff --git a/generic/tclTomMath.h b/generic/tclTomMath.h
index 34ac1bb..14cf5b6 100644
--- a/generic/tclTomMath.h
+++ b/generic/tclTomMath.h
@@ -28,9 +28,9 @@ extern "C" {
 
 /* detect 64-bit mode if possible */
 #if defined(NEVER)  /* 128-bit ints fail in too many places */
-   #if !(defined(MP_32BIT) || defined(MP_16BIT) || defined(MP_8BIT))
-      #define MP_64BIT
-   #endif
+#   if !(defined(MP_32BIT) || defined(MP_16BIT) || defined(MP_8BIT))
+#      define MP_64BIT
+#   endif
 #endif
 
 /* some default configurations.
@@ -43,94 +43,94 @@ extern "C" {
  */
 #ifdef MP_8BIT
 #ifndef MP_DIGIT_DECLARED
-   typedef uint8_t              mp_digit;
+typedef uint8_t              mp_digit;
 #define MP_DIGIT_DECLARED
 #endif
 #ifndef MP_WORD_DECLARED
-   typedef uint16_t             mp_word;
+typedef uint16_t             mp_word;
 #define MP_WORD_DECLARED
 #endif
-#define MP_SIZEOF_MP_DIGIT      1
-#ifdef DIGIT_BIT
-#error You must not define DIGIT_BIT when using MP_8BIT
-#endif
+#   define MP_SIZEOF_MP_DIGIT 1
+#   ifdef DIGIT_BIT
+#      error You must not define DIGIT_BIT when using MP_8BIT
+#   endif
 #elif defined(MP_16BIT)
 #ifndef MP_DIGIT_DECLARED
-   typedef uint16_t             mp_digit;
+typedef uint16_t             mp_digit;
 #define MP_DIGIT_DECLARED
 #endif
 #ifndef MP_WORD_DECLARED
-   typedef uint32_t             mp_word;
+typedef uint32_t             mp_word;
 #define MP_WORD_DECLARED
 #endif
-#define MP_SIZEOF_MP_DIGIT      2
-#ifdef DIGIT_BIT
-#error You must not define DIGIT_BIT when using MP_16BIT
-#endif
+#   define MP_SIZEOF_MP_DIGIT 2
+#   ifdef DIGIT_BIT
+#      error You must not define DIGIT_BIT when using MP_16BIT
+#   endif
 #elif defined(MP_64BIT)
-   /* for GCC only on supported platforms */
+/* for GCC only on supported platforms */
 #ifndef MP_DIGIT_DECLARED
-   typedef uint64_t mp_digit;
+typedef uint64_t mp_digit;
 #define MP_DIGIT_DECLARED
 #endif
-#if defined(_WIN32)
+#   if defined(_WIN32)
 #ifndef MP_WORD_DECLARED
-   typedef unsigned __int128    mp_word;
+typedef unsigned __int128    mp_word;
 #define MP_WORD_DECLARED
 #endif
-#elif defined(__GNUC__)
-   typedef unsigned long        mp_word __attribute__ ((mode(TI)));
-#else
-   /* it seems you have a problem
-    * but we assume you can somewhere define your own uint128_t */
+#   elif defined(__GNUC__)
+typedef unsigned long        mp_word __attribute__((mode(TI)));
+#   else
+/* it seems you have a problem
+ * but we assume you can somewhere define your own uint128_t */
 #ifndef MP_WORD_DECLARED
-   typedef uint128_t            mp_word;
+typedef uint128_t            mp_word;
 #define MP_WORD_DECLARED
 #endif
-#endif
+#   endif
 
-   #define DIGIT_BIT            60
+#   define DIGIT_BIT 60
 #else
-   /* this is the default case, 28-bit digits */
+/* this is the default case, 28-bit digits */
 
-   /* this is to make porting into LibTomCrypt easier :-) */
+/* this is to make porting into LibTomCrypt easier :-) */
 #ifndef MP_DIGIT_DECLARED
-   typedef uint32_t             mp_digit;
+typedef uint32_t             mp_digit;
 #define MP_DIGIT_DECLARED
 #endif
 #ifndef MP_WORD_DECLARED
-   typedef uint64_t             mp_word;
+typedef uint64_t             mp_word;
 #define MP_WORD_DECLARED
 #endif
 
-#ifdef MP_31BIT
-   /* this is an extension that uses 31-bit digits */
-   #define DIGIT_BIT            31
-#else
-   /* default case is 28-bit digits, defines MP_28BIT as a handy macro to test */
-   #define DIGIT_BIT            28
-   #define MP_28BIT
-#endif
+#   ifdef MP_31BIT
+/* this is an extension that uses 31-bit digits */
+#      define DIGIT_BIT 31
+#   else
+/* default case is 28-bit digits, defines MP_28BIT as a handy macro to test */
+#      define DIGIT_BIT 28
+#      define MP_28BIT
+#   endif
 #endif
 
 /* otherwise the bits per digit is calculated automatically from the size of a mp_digit */
 #ifndef DIGIT_BIT
-   #define DIGIT_BIT     (((CHAR_BIT * MP_SIZEOF_MP_DIGIT) - 1))  /* bits per digit */
-   typedef uint_least32_t mp_min_u32;
+#   define DIGIT_BIT (((CHAR_BIT * MP_SIZEOF_MP_DIGIT) - 1))  /* bits per digit */
+typedef uint_least32_t mp_min_u32;
 #else
-   typedef mp_digit mp_min_u32;
+typedef mp_digit mp_min_u32;
 #endif
 
 /* use arc4random on platforms that support it */
 #if defined(__FreeBSD__) || defined(__OpenBSD__) || defined(__NetBSD__) || defined(__DragonFly__)
-    #define MP_GEN_RANDOM()    arc4random()
-    #define MP_GEN_RANDOM_MAX  0xffffffff
+#   define MP_GEN_RANDOM()    arc4random()
+#   define MP_GEN_RANDOM_MAX  0xffffffff
 #endif
 
 /* use rand() as fall-back if there's no better rand function */
 #ifndef MP_GEN_RANDOM
-    #define MP_GEN_RANDOM()    rand()
-    #define MP_GEN_RANDOM_MAX  RAND_MAX
+#   define MP_GEN_RANDOM()    rand()
+#   define MP_GEN_RANDOM_MAX  RAND_MAX
 #endif
 
 #define MP_DIGIT_BIT     DIGIT_BIT
@@ -163,9 +163,9 @@ typedef int           mp_err;
 /* you'll have to tune these... */
 #if defined(BUILD_tcl) || !defined(_WIN32)
 MODULE_SCOPE int KARATSUBA_MUL_CUTOFF,
-           KARATSUBA_SQR_CUTOFF,
-           TOOM_MUL_CUTOFF,
-           TOOM_SQR_CUTOFF;
+       KARATSUBA_SQR_CUTOFF,
+       TOOM_MUL_CUTOFF,
+       TOOM_SQR_CUTOFF;
 #endif
 
 /* define this to use lower memory usage routines (exptmods mostly) */
@@ -173,11 +173,11 @@ MODULE_SCOPE int KARATSUBA_MUL_CUTOFF,
 
 /* default precision */
 #ifndef MP_PREC
-   #ifndef MP_LOW_MEM
-      #define MP_PREC                 32     /* default digits of precision */
-   #else
-      #define MP_PREC                 8      /* default digits of precision */
-   #endif
+#   ifndef MP_LOW_MEM
+#      define MP_PREC 32        /* default digits of precision */
+#   else
+#      define MP_PREC 8         /* default digits of precision */
+#   endif
 #endif
 
 /* size of comba arrays, should be at least 2 * 2**(BITS_PER_WORD - BITS_PER_DIGIT*2) */
@@ -189,17 +189,17 @@ MODULE_SCOPE int KARATSUBA_MUL_CUTOFF,
 typedef struct mp_int mp_int;
 #endif
 struct mp_int {
-    int used, alloc, sign;
-    mp_digit *dp;
+   int used, alloc, sign;
+   mp_digit *dp;
 };
 
 /* callback for mp_prime_random, should fill dst with random bytes and return how many read [upto len] */
 typedef int ltm_prime_callback(unsigned char *dst, int len, void *dat);
 
 
-#define USED(m)    ((m)->used)
-#define DIGIT(m,k) ((m)->dp[(k)])
-#define SIGN(m)    ((m)->sign)
+#define USED(m)     ((m)->used)
+#define DIGIT(m, k) ((m)->dp[(k)])
+#define SIGN(m)     ((m)->sign)
 
 /* error code to char* string */
 const char *mp_error_to_string(int code);
@@ -278,27 +278,27 @@ int mp_set_long_long(mp_int *a, Tcl_WideUInt b);
 
 /* get a 32-bit value */
 /*
-unsigned long mp_get_int(const mp_int * a);
+unsigned long mp_get_int(const mp_int *a);
 */
 
 /* get a platform dependent unsigned long value */
 /*
-unsigned long mp_get_long(const mp_int * a);
+unsigned long mp_get_long(const mp_int *a);
 */
 
 /* get a platform dependent unsigned long long value */
 /*
-Tcl_WideUInt mp_get_long_long(const mp_int * a);
+Tcl_WideUInt mp_get_long_long(const mp_int *a);
 */
 
 /* initialize and set a digit */
 /*
-int mp_init_set (mp_int * a, mp_digit b);
+int mp_init_set(mp_int *a, mp_digit b);
 */
 
 /* initialize and set 32-bit value */
 /*
-int mp_init_set_int (mp_int * a, unsigned long b);
+int mp_init_set_int(mp_int *a, unsigned long b);
 */
 
 /* copy, b = a */
@@ -318,12 +318,12 @@ void mp_clamp(mp_int *a);
 
 /* import binary data */
 /*
-int mp_import(mp_int* rop, size_t count, int order, size_t size, int endian, size_t nails, const void* op);
+int mp_import(mp_int *rop, size_t count, int order, size_t size, int endian, size_t nails, const void *op);
 */
 
 /* export binary data */
 /*
-int mp_export(void* rop, size_t* countp, int order, size_t size, int endian, size_t nails, mp_int* op);
+int mp_export(void *rop, size_t *countp, int order, size_t size, int endian, size_t nails, mp_int *op);
 */
 
 /* ---> digit manipulation <--- */
@@ -457,7 +457,7 @@ int mp_cmp_d(const mp_int *a, mp_digit b);
 
 /* c = a + b */
 /*
-int mp_add_d(const mp_int *a, const mp_digit b, mp_int *c);
+int mp_add_d(const mp_int *a, mp_digit b, mp_int *c);
 */
 
 /* c = a - b */
@@ -482,10 +482,10 @@ int mp_div_3(const mp_int *a, mp_int *c, mp_digit *d);
 
 /* c = a**b */
 /*
-int mp_expt_d(const mp_int *a, const mp_digit b, mp_int *c);
+int mp_expt_d(const mp_int *a, mp_digit b, mp_int *c);
 */
 /*
-int mp_expt_d_ex (const mp_int * a, const mp_digit b, mp_int * c, int fast);
+int mp_expt_d_ex(const mp_int *a, mp_digit b, mp_int *c, int fast);
 */
 
 /* c = a mod b, 0 <= c < b  */
@@ -543,12 +543,12 @@ int mp_lcm(mp_int *a, mp_int *b, mp_int *c);
 int mp_n_root(mp_int *a, mp_digit b, mp_int *c);
 */
 /*
-int mp_n_root_ex (mp_int * a, mp_digit b, mp_int * c, int fast);
+int mp_n_root_ex(mp_int *a, mp_digit b, mp_int *c, int fast);
 */
 
 /* special sqrt algo */
 /*
-int mp_sqrt(mp_int *arg, mp_int *ret);
+int mp_sqrt(const mp_int *arg, mp_int *ret);
 */
 
 /* special sqrt (mod prime) */
@@ -651,9 +651,9 @@ int mp_exptmod(mp_int *a, mp_int *b, mp_int *c, mp_int *d);
 
 /* number of primes */
 #ifdef MP_8BIT
-#  define PRIME_SIZE      31
+#  define PRIME_SIZE 31
 #else
-#  define PRIME_SIZE      256
+#  define PRIME_SIZE 256
 #endif
 
 /* table of first PRIME_SIZE primes */
@@ -741,7 +741,7 @@ int mp_count_bits(const mp_int *a);
 */
 
 /*
-int mp_unsigned_bin_size(mp_int *a);
+int mp_unsigned_bin_size(const mp_int *a);
 */
 /*
 int mp_read_unsigned_bin(mp_int *a, const unsigned char *b, int c);
@@ -750,7 +750,7 @@ int mp_read_unsigned_bin(mp_int *a, const unsigned char *b, int c);
 int mp_to_unsigned_bin(const mp_int *a, unsigned char *b);
 */
 /*
-int mp_to_unsigned_bin_n (const mp_int * a, unsigned char *b, unsigned long *outlen);
+int mp_to_unsigned_bin_n(const mp_int *a, unsigned char *b, unsigned long *outlen);
 */
 
 /*
@@ -763,7 +763,7 @@ int mp_read_signed_bin(mp_int *a, const unsigned char *b, int c);
 int mp_to_signed_bin(const mp_int *a,  unsigned char *b);
 */
 /*
-int mp_to_signed_bin_n (const mp_int * a, unsigned char *b, unsigned long *outlen);
+int mp_to_signed_bin_n(const mp_int *a, unsigned char *b, unsigned long *outlen);
 */
 
 /*
@@ -773,7 +773,7 @@ int mp_read_radix(mp_int *a, const char *str, int radix);
 int mp_toradix(const mp_int *a, char *str, int radix);
 */
 /*
-int mp_toradix_n(const mp_int * a, char *str, int radix, int maxlen);
+int mp_toradix_n(const mp_int *a, char *str, int radix, int maxlen);
 */
 /*
 int mp_radix_size(const mp_int *a, int radix, int *size);
@@ -801,7 +801,7 @@ int mp_fwrite(const mp_int *a, int radix, FILE *stream);
 #define mp_tohex(M, S)     mp_toradix((M), (S), 16)
 
 #ifdef __cplusplus
-   }
+}
 #endif
 
 #endif
diff --git a/libtommath/astylerc b/libtommath/astylerc
new file mode 100644
index 0000000..5d63f7a
--- /dev/null
+++ b/libtommath/astylerc
@@ -0,0 +1,27 @@
+# Artistic Style, see http://astyle.sourceforge.net/
+# full documentation, see: http://astyle.sourceforge.net/astyle.html
+#
+# usage:
+#       astyle --options=astylerc *.[ch]
+
+## Bracket Style Options
+style=kr
+
+## Tab Options
+indent=spaces=3
+
+## Bracket Modify Options
+
+## Indentation Options
+min-conditional-indent=0
+
+## Padding Options
+pad-header
+unpad-paren
+align-pointer=name
+
+## Formatting Options
+break-after-logical
+max-code-length=120
+convert-tabs
+mode=c
diff --git a/libtommath/bn_error.c b/libtommath/bn_error.c
index 0d77411..a51d712 100644
--- a/libtommath/bn_error.c
+++ b/libtommath/bn_error.c
@@ -16,12 +16,12 @@
  */
 
 static const struct {
-     int code;
-     const char *msg;
+   int code;
+   const char *msg;
 } msgs[] = {
-     { MP_OKAY, "Successful" },
-     { MP_MEM,  "Out of heap" },
-     { MP_VAL,  "Value out of range" }
+   { MP_OKAY, "Successful" },
+   { MP_MEM,  "Out of heap" },
+   { MP_VAL,  "Value out of range" }
 };
 
 /* return a char * string for a given code */
@@ -31,9 +31,9 @@ const char *mp_error_to_string(int code)
 
    /* scan the lookup table for the given message */
    for (x = 0; x < (int)(sizeof(msgs) / sizeof(msgs[0])); x++) {
-       if (msgs[x].code == code) {
-          return msgs[x].msg;
-       }
+      if (msgs[x].code == code) {
+         return msgs[x].msg;
+      }
    }
 
    /* generic reply for invalid code */
diff --git a/libtommath/bn_fast_mp_invmod.c b/libtommath/bn_fast_mp_invmod.c
index 12f42de..7771136 100644
--- a/libtommath/bn_fast_mp_invmod.c
+++ b/libtommath/bn_fast_mp_invmod.c
@@ -15,131 +15,132 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* computes the modular inverse via binary extended euclidean algorithm, 
- * that is c = 1/a mod b 
+/* computes the modular inverse via binary extended euclidean algorithm,
+ * that is c = 1/a mod b
  *
- * Based on slow invmod except this is optimized for the case where b is 
+ * Based on slow invmod except this is optimized for the case where b is
  * odd as per HAC Note 14.64 on pp. 610
  */
-int fast_mp_invmod (mp_int * a, mp_int * b, mp_int * c)
+int fast_mp_invmod(mp_int *a, mp_int *b, mp_int *c)
 {
-  mp_int  x, y, u, v, B, D;
-  int     res, neg;
-
-  /* 2. [modified] b must be odd   */
-  if (mp_iseven (b) == MP_YES) {
-    return MP_VAL;
-  }
-
-  /* init all our temps */
-  if ((res = mp_init_multi(&x, &y, &u, &v, &B, &D, NULL)) != MP_OKAY) {
-     return res;
-  }
-
-  /* x == modulus, y == value to invert */
-  if ((res = mp_copy (b, &x)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-
-  /* we need y = |a| */
-  if ((res = mp_mod (a, b, &y)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-
-  /* 3. u=x, v=y, A=1, B=0, C=0,D=1 */
-  if ((res = mp_copy (&x, &u)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-  if ((res = mp_copy (&y, &v)) != MP_OKAY) {
-    goto LBL_ERR;
-  }
-  mp_set (&D, 1);
+   mp_int  x, y, u, v, B, D;
+   int     res, neg;
 
-top:
-  /* 4.  while u is even do */
-  while (mp_iseven (&u) == MP_YES) {
-    /* 4.1 u = u/2 */
-    if ((res = mp_div_2 (&u, &u)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    /* 4.2 if B is odd then */
-    if (mp_isodd (&B) == MP_YES) {
-      if ((res = mp_sub (&B, &x, &B)) != MP_OKAY) {
-        goto LBL_ERR;
-      }
-    }
-    /* B = B/2 */
-    if ((res = mp_div_2 (&B, &B)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
+   /* 2. [modified] b must be odd   */
+   if (mp_iseven(b) == MP_YES) {
+      return MP_VAL;
+   }
 
-  /* 5.  while v is even do */
-  while (mp_iseven (&v) == MP_YES) {
-    /* 5.1 v = v/2 */
-    if ((res = mp_div_2 (&v, &v)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-    /* 5.2 if D is odd then */
-    if (mp_isodd (&D) == MP_YES) {
-      /* D = (D-x)/2 */
-      if ((res = mp_sub (&D, &x, &D)) != MP_OKAY) {
-        goto LBL_ERR;
-      }
-    }
-    /* D = D/2 */
-    if ((res = mp_div_2 (&D, &D)) != MP_OKAY) {
+   /* init all our temps */
+   if ((res = mp_init_multi(&x, &y, &u, &v, &B, &D, NULL)) != MP_OKAY) {
+      return res;
+   }
+
+   /* x == modulus, y == value to invert */
+   if ((res = mp_copy(b, &x)) != MP_OKAY) {
       goto LBL_ERR;
-    }
-  }
+   }
 
-  /* 6.  if u >= v then */
-  if (mp_cmp (&u, &v) != MP_LT) {
-    /* u = u - v, B = B - D */
-    if ((res = mp_sub (&u, &v, &u)) != MP_OKAY) {
+   /* we need y = |a| */
+   if ((res = mp_mod(a, b, &y)) != MP_OKAY) {
       goto LBL_ERR;
-    }
+   }
 
-    if ((res = mp_sub (&B, &D, &B)) != MP_OKAY) {
+   /* 3. u=x, v=y, A=1, B=0, C=0,D=1 */
+   if ((res = mp_copy(&x, &u)) != MP_OKAY) {
       goto LBL_ERR;
-    }
-  } else {
-    /* v - v - u, D = D - B */
-    if ((res = mp_sub (&v, &u, &v)) != MP_OKAY) {
+   }
+   if ((res = mp_copy(&y, &v)) != MP_OKAY) {
       goto LBL_ERR;
-    }
+   }
+   mp_set(&D, 1);
 
-    if ((res = mp_sub (&D, &B, &D)) != MP_OKAY) {
-      goto LBL_ERR;
-    }
-  }
-
-  /* if not zero goto step 4 */
-  if (mp_iszero (&u) == MP_NO) {
-    goto top;
-  }
-
-  /* now a = C, b = D, gcd == g*v */
-
-  /* if v != 1 then there is no inverse */
-  if (mp_cmp_d (&v, 1) != MP_EQ) {
-    res = MP_VAL;
-    goto LBL_ERR;
-  }
-
-  /* b is now the inverse */
-  neg = a->sign;
-  while (D.sign == MP_NEG) {
-    if ((res = mp_add (&D, b, &D)) != MP_OKAY) {
+top:
+   /* 4.  while u is even do */
+   while (mp_iseven(&u) == MP_YES) {
+      /* 4.1 u = u/2 */
+      if ((res = mp_div_2(&u, &u)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+      /* 4.2 if B is odd then */
+      if (mp_isodd(&B) == MP_YES) {
+         if ((res = mp_sub(&B, &x, &B)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* B = B/2 */
+      if ((res = mp_div_2(&B, &B)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* 5.  while v is even do */
+   while (mp_iseven(&v) == MP_YES) {
+      /* 5.1 v = v/2 */
+      if ((res = mp_div_2(&v, &v)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+      /* 5.2 if D is odd then */
+      if (mp_isodd(&D) == MP_YES) {
+         /* D = (D-x)/2 */
+         if ((res = mp_sub(&D, &x, &D)) != MP_OKAY) {
+            goto LBL_ERR;
+         }
+      }
+      /* D = D/2 */
+      if ((res = mp_div_2(&D, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* 6.  if u >= v then */
+   if (mp_cmp(&u, &v) != MP_LT) {
+      /* u = u - v, B = B - D */
+      if ((res = mp_sub(&u, &v, &u)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+
+      if ((res = mp_sub(&B, &D, &B)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   } else {
+      /* v - v - u, D = D - B */
+      if ((res = mp_sub(&v, &u, &v)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+
+      if ((res = mp_sub(&D, &B, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+
+   /* if not zero goto step 4 */
+   if (mp_iszero(&u) == MP_NO) {
+      goto top;
+   }
+
+   /* now a = C, b = D, gcd == g*v */
+
+   /* if v != 1 then there is no inverse */
+   if (mp_cmp_d(&v, 1) != MP_EQ) {
+      res = MP_VAL;
       goto LBL_ERR;
-    }
-  }
-  mp_exch (&D, c);
-  c->sign = neg;
-  res = MP_OKAY;
-
-LBL_ERR:mp_clear_multi (&x, &y, &u, &v, &B, &D, NULL);
-  return res;
+   }
+
+   /* b is now the inverse */
+   neg = a->sign;
+   while (D.sign == MP_NEG) {
+      if ((res = mp_add(&D, b, &D)) != MP_OKAY) {
+         goto LBL_ERR;
+      }
+   }
+   mp_exch(&D, c);
+   c->sign = neg;
+   res = MP_OKAY;
+
+LBL_ERR:
+   mp_clear_multi(&x, &y, &u, &v, &B, &D, NULL);
+   return res;
 }
 #endif
 
diff --git a/libtommath/bn_fast_mp_montgomery_reduce.c b/libtommath/bn_fast_mp_montgomery_reduce.c
index 16d5ff7..f2c38bf 100644
--- a/libtommath/bn_fast_mp_montgomery_reduce.c
+++ b/libtommath/bn_fast_mp_montgomery_reduce.c
@@ -23,147 +23,147 @@
  *
  * Based on Algorithm 14.32 on pp.601 of HAC.
 */
-int fast_mp_montgomery_reduce (mp_int * x, mp_int * n, mp_digit rho)
+int fast_mp_montgomery_reduce(mp_int *x, mp_int *n, mp_digit rho)
 {
-  int     ix, res, olduse;
-  mp_word W[MP_WARRAY];
-
-  /* get old used count */
-  olduse = x->used;
-
-  /* grow a as required */
-  if (x->alloc < (n->used + 1)) {
-    if ((res = mp_grow (x, n->used + 1)) != MP_OKAY) {
-      return res;
-    }
-  }
-
-  /* first we have to get the digits of the input into
-   * an array of double precision words W[...]
-   */
-  {
-    mp_word *_W;
-    mp_digit *tmpx;
-
-    /* alias for the W[] array */
-    _W   = W;
-
-    /* alias for the digits of  x*/
-    tmpx = x->dp;
-
-    /* copy the digits of a into W[0..a->used-1] */
-    for (ix = 0; ix < x->used; ix++) {
-      *_W++ = *tmpx++;
-    }
-
-    /* zero the high words of W[a->used..m->used*2] */
-    for (; ix < ((n->used * 2) + 1); ix++) {
-      *_W++ = 0;
-    }
-  }
-
-  /* now we proceed to zero successive digits
-   * from the least significant upwards
-   */
-  for (ix = 0; ix < n->used; ix++) {
-    /* mu = ai * m' mod b
-     *
-     * We avoid a double precision multiplication (which isn't required)
-     * by casting the value down to a mp_digit.  Note this requires
-     * that W[ix-1] have  the carry cleared (see after the inner loop)
-     */
-    mp_digit mu;
-    mu = (mp_digit) (((W[ix] & MP_MASK) * rho) & MP_MASK);
-
-    /* a = a + mu * m * b**i
-     *
-     * This is computed in place and on the fly.  The multiplication
-     * by b**i is handled by offseting which columns the results
-     * are added to.
-     *
-     * Note the comba method normally doesn't handle carries in the
-     * inner loop In this case we fix the carry from the previous
-     * column since the Montgomery reduction requires digits of the
-     * result (so far) [see above] to work.  This is
-     * handled by fixing up one carry after the inner loop.  The
-     * carry fixups are done in order so after these loops the
-     * first m->used words of W[] have the carries fixed
-     */
-    {
-      int iy;
-      mp_digit *tmpn;
+   int     ix, res, olduse;
+   mp_word W[MP_WARRAY];
+
+   /* get old used count */
+   olduse = x->used;
+
+   /* grow a as required */
+   if (x->alloc < (n->used + 1)) {
+      if ((res = mp_grow(x, n->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* first we have to get the digits of the input into
+    * an array of double precision words W[...]
+    */
+   {
       mp_word *_W;
+      mp_digit *tmpx;
+
+      /* alias for the W[] array */
+      _W   = W;
 
-      /* alias for the digits of the modulus */
-      tmpn = n->dp;
+      /* alias for the digits of  x*/
+      tmpx = x->dp;
 
-      /* Alias for the columns set by an offset of ix */
-      _W = W + ix;
+      /* copy the digits of a into W[0..a->used-1] */
+      for (ix = 0; ix < x->used; ix++) {
+         *_W++ = *tmpx++;
+      }
 
-      /* inner loop */
-      for (iy = 0; iy < n->used; iy++) {
-          *_W++ += ((mp_word)mu) * ((mp_word)*tmpn++);
+      /* zero the high words of W[a->used..m->used*2] */
+      for (; ix < ((n->used * 2) + 1); ix++) {
+         *_W++ = 0;
       }
-    }
-
-    /* now fix carry for next digit, W[ix+1] */
-    W[ix + 1] += W[ix] >> ((mp_word) DIGIT_BIT);
-  }
-
-  /* now we have to propagate the carries and
-   * shift the words downward [all those least
-   * significant digits we zeroed].
-   */
-  {
-    mp_digit *tmpx;
-    mp_word *_W, *_W1;
-
-    /* nox fix rest of carries */
-
-    /* alias for current word */
-    _W1 = W + ix;
-
-    /* alias for next word, where the carry goes */
-    _W = W + ++ix;
-
-    for (; ix <= ((n->used * 2) + 1); ix++) {
-      *_W++ += *_W1++ >> ((mp_word) DIGIT_BIT);
-    }
-
-    /* copy out, A = A/b**n
-     *
-     * The result is A/b**n but instead of converting from an
-     * array of mp_word to mp_digit than calling mp_rshd
-     * we just copy them in the right order
-     */
-
-    /* alias for destination word */
-    tmpx = x->dp;
-
-    /* alias for shifted double precision result */
-    _W = W + n->used;
-
-    for (ix = 0; ix < (n->used + 1); ix++) {
-      *tmpx++ = (mp_digit)(*_W++ & ((mp_word) MP_MASK));
-    }
-
-    /* zero oldused digits, if the input a was larger than
-     * m->used+1 we'll have to clear the digits
-     */
-    for (; ix < olduse; ix++) {
-      *tmpx++ = 0;
-    }
-  }
-
-  /* set the max used and clamp */
-  x->used = n->used + 1;
-  mp_clamp (x);
-
-  /* if A >= m then A = A - m */
-  if (mp_cmp_mag (x, n) != MP_LT) {
-    return s_mp_sub (x, n, x);
-  }
-  return MP_OKAY;
+   }
+
+   /* now we proceed to zero successive digits
+    * from the least significant upwards
+    */
+   for (ix = 0; ix < n->used; ix++) {
+      /* mu = ai * m' mod b
+       *
+       * We avoid a double precision multiplication (which isn't required)
+       * by casting the value down to a mp_digit.  Note this requires
+       * that W[ix-1] have  the carry cleared (see after the inner loop)
+       */
+      mp_digit mu;
+      mu = (mp_digit)(((W[ix] & MP_MASK) * rho) & MP_MASK);
+
+      /* a = a + mu * m * b**i
+       *
+       * This is computed in place and on the fly.  The multiplication
+       * by b**i is handled by offseting which columns the results
+       * are added to.
+       *
+       * Note the comba method normally doesn't handle carries in the
+       * inner loop In this case we fix the carry from the previous
+       * column since the Montgomery reduction requires digits of the
+       * result (so far) [see above] to work.  This is
+       * handled by fixing up one carry after the inner loop.  The
+       * carry fixups are done in order so after these loops the
+       * first m->used words of W[] have the carries fixed
+       */
+      {
+         int iy;
+         mp_digit *tmpn;
+         mp_word *_W;
+
+         /* alias for the digits of the modulus */
+         tmpn = n->dp;
+
+         /* Alias for the columns set by an offset of ix */
+         _W = W + ix;
+
+         /* inner loop */
+         for (iy = 0; iy < n->used; iy++) {
+            *_W++ += ((mp_word)mu) * ((mp_word)*tmpn++);
+         }
+      }
+
+      /* now fix carry for next digit, W[ix+1] */
+      W[ix + 1] += W[ix] >> ((mp_word) DIGIT_BIT);
+   }
+
+   /* now we have to propagate the carries and
+    * shift the words downward [all those least
+    * significant digits we zeroed].
+    */
+   {
+      mp_digit *tmpx;
+      mp_word *_W, *_W1;
+
+      /* nox fix rest of carries */
+
+      /* alias for current word */
+      _W1 = W + ix;
+
+      /* alias for next word, where the carry goes */
+      _W = W + ++ix;
+
+      for (; ix <= ((n->used * 2) + 1); ix++) {
+         *_W++ += *_W1++ >> ((mp_word) DIGIT_BIT);
+      }
+
+      /* copy out, A = A/b**n
+       *
+       * The result is A/b**n but instead of converting from an
+       * array of mp_word to mp_digit than calling mp_rshd
+       * we just copy them in the right order
+       */
+
+      /* alias for destination word */
+      tmpx = x->dp;
+
+      /* alias for shifted double precision result */
+      _W = W + n->used;
+
+      for (ix = 0; ix < (n->used + 1); ix++) {
+         *tmpx++ = (mp_digit)(*_W++ & ((mp_word) MP_MASK));
+      }
+
+      /* zero oldused digits, if the input a was larger than
+       * m->used+1 we'll have to clear the digits
+       */
+      for (; ix < olduse; ix++) {
+         *tmpx++ = 0;
+      }
+   }
+
+   /* set the max used and clamp */
+   x->used = n->used + 1;
+   mp_clamp(x);
+
+   /* if A >= m then A = A - m */
+   if (mp_cmp_mag(x, n) != MP_LT) {
+      return s_mp_sub(x, n, x);
+   }
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_mul_digs.c b/libtommath/bn_fast_s_mp_mul_digs.c
index 641b574..558d151 100644
--- a/libtommath/bn_fast_s_mp_mul_digs.c
+++ b/libtommath/bn_fast_s_mp_mul_digs.c
@@ -17,39 +17,39 @@
 
 /* Fast (comba) multiplier
  *
- * This is the fast column-array [comba] multiplier.  It is 
- * designed to compute the columns of the product first 
- * then handle the carries afterwards.  This has the effect 
+ * This is the fast column-array [comba] multiplier.  It is
+ * designed to compute the columns of the product first
+ * then handle the carries afterwards.  This has the effect
  * of making the nested loops that compute the columns very
  * simple and schedulable on super-scalar processors.
  *
- * This has been modified to produce a variable number of 
- * digits of output so if say only a half-product is required 
- * you don't have to compute the upper half (a feature 
+ * This has been modified to produce a variable number of
+ * digits of output so if say only a half-product is required
+ * you don't have to compute the upper half (a feature
  * required for fast Barrett reduction).
  *
  * Based on Algorithm 14.12 on pp.595 of HAC.
  *
  */
-int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs)
+int fast_s_mp_mul_digs(const mp_int *a, const mp_int *b, mp_int *c, int digs)
 {
-  int     olduse, res, pa, ix, iz;
-  mp_digit W[MP_WARRAY];
-  mp_word  _W;
+   int     olduse, res, pa, ix, iz;
+   mp_digit W[MP_WARRAY];
+   mp_word  _W;
 
-  /* grow the destination as required */
-  if (c->alloc < digs) {
-    if ((res = mp_grow (c, digs)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* grow the destination as required */
+   if (c->alloc < digs) {
+      if ((res = mp_grow(c, digs)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* number of output digits to produce */
-  pa = MIN(digs, a->used + b->used);
+   /* number of output digits to produce */
+   pa = MIN(digs, a->used + b->used);
 
-  /* clear the carry */
-  _W = 0;
-  for (ix = 0; ix < pa; ix++) { 
+   /* clear the carry */
+   _W = 0;
+   for (ix = 0; ix < pa; ix++) {
       int      tx, ty;
       int      iy;
       mp_digit *tmpx, *tmpy;
@@ -62,7 +62,7 @@ int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs
       tmpx = a->dp + tx;
       tmpy = b->dp + ty;
 
-      /* this is the number of times the loop will iterrate, essentially 
+      /* this is the number of times the loop will iterrate, essentially
          while (tx++ < a->used && ty-- >= 0) { ... }
        */
       iy = MIN(a->used-tx, ty+1);
@@ -78,27 +78,27 @@ int fast_s_mp_mul_digs (const mp_int * a, const mp_int * b, mp_int * c, int digs
 
       /* make next carry */
       _W = _W >> ((mp_word)DIGIT_BIT);
-  }
+   }
 
-  /* setup dest */
-  olduse  = c->used;
-  c->used = pa;
+   /* setup dest */
+   olduse  = c->used;
+   c->used = pa;
 
-  {
-    mp_digit *tmpc;
-    tmpc = c->dp;
-    for (ix = 0; ix < (pa + 1); ix++) {
-      /* now extract the previous digit [below the carry] */
-      *tmpc++ = W[ix];
-    }
+   {
+      mp_digit *tmpc;
+      tmpc = c->dp;
+      for (ix = 0; ix < (pa + 1); ix++) {
+         /* now extract the previous digit [below the carry] */
+         *tmpc++ = W[ix];
+      }
 
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpc++ = 0;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpc++ = 0;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_mul_high_digs.c b/libtommath/bn_fast_s_mp_mul_high_digs.c
index 08f0355..588d80b 100644
--- a/libtommath/bn_fast_s_mp_mul_high_digs.c
+++ b/libtommath/bn_fast_s_mp_mul_high_digs.c
@@ -24,24 +24,24 @@
  *
  * Based on Algorithm 14.12 on pp.595 of HAC.
  */
-int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
+int fast_s_mp_mul_high_digs(mp_int *a, mp_int *b, mp_int *c, int digs)
 {
-  int     olduse, res, pa, ix, iz;
-  mp_digit W[MP_WARRAY];
-  mp_word  _W;
+   int     olduse, res, pa, ix, iz;
+   mp_digit W[MP_WARRAY];
+   mp_word  _W;
 
-  /* grow the destination as required */
-  pa = a->used + b->used;
-  if (c->alloc < pa) {
-    if ((res = mp_grow (c, pa)) != MP_OKAY) {
-      return res;
-    }
-  }
+   /* grow the destination as required */
+   pa = a->used + b->used;
+   if (c->alloc < pa) {
+      if ((res = mp_grow(c, pa)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* number of output digits to produce */
-  pa = a->used + b->used;
-  _W = 0;
-  for (ix = digs; ix < pa; ix++) { 
+   /* number of output digits to produce */
+   pa = a->used + b->used;
+   _W = 0;
+   for (ix = digs; ix < pa; ix++) {
       int      tx, ty, iy;
       mp_digit *tmpx, *tmpy;
 
@@ -53,7 +53,7 @@ int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
       tmpx = a->dp + tx;
       tmpy = b->dp + ty;
 
-      /* this is the number of times the loop will iterrate, essentially its 
+      /* this is the number of times the loop will iterrate, essentially its
          while (tx++ < a->used && ty-- >= 0) { ... }
        */
       iy = MIN(a->used-tx, ty+1);
@@ -68,28 +68,28 @@ int fast_s_mp_mul_high_digs (mp_int * a, mp_int * b, mp_int * c, int digs)
 
       /* make next carry */
       _W = _W >> ((mp_word)DIGIT_BIT);
-  }
-  
-  /* setup dest */
-  olduse  = c->used;
-  c->used = pa;
+   }
+
+   /* setup dest */
+   olduse  = c->used;
+   c->used = pa;
 
-  {
-    mp_digit *tmpc;
+   {
+      mp_digit *tmpc;
 
-    tmpc = c->dp + digs;
-    for (ix = digs; ix < pa; ix++) {
-      /* now extract the previous digit [below the carry] */
-      *tmpc++ = W[ix];
-    }
+      tmpc = c->dp + digs;
+      for (ix = digs; ix < pa; ix++) {
+         /* now extract the previous digit [below the carry] */
+         *tmpc++ = W[ix];
+      }
 
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpc++ = 0;
-    }
-  }
-  mp_clamp (c);
-  return MP_OKAY;
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpc++ = 0;
+      }
+   }
+   mp_clamp(c);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_fast_s_mp_sqr.c b/libtommath/bn_fast_s_mp_sqr.c
index 282d303..161f785 100644
--- a/libtommath/bn_fast_s_mp_sqr.c
+++ b/libtommath/bn_fast_s_mp_sqr.c
@@ -16,32 +16,32 @@
  */
 
 /* the jist of squaring...
- * you do like mult except the offset of the tmpx [one that 
- * starts closer to zero] can't equal the offset of tmpy.  
+ * you do like mult except the offset of the tmpx [one that
+ * starts closer to zero] can't equal the offset of tmpy.
  * So basically you set up iy like before then you min it with
- * (ty-tx) so that it never happens.  You double all those 
+ * (ty-tx) so that it never happens.  You double all those
  * you add in the inner loop
 
 After that loop you do the squares and add them in.
 */
 
-int fast_s_mp_sqr (const mp_int * a, mp_int * b)
+int fast_s_mp_sqr(const mp_int *a, mp_int *b)
 {
-  int       olduse, res, pa, ix, iz;
-  mp_digit   W[MP_WARRAY], *tmpx;
-  mp_word   W1;
-
-  /* grow the destination as required */
-  pa = a->used + a->used;
-  if (b->alloc < pa) {
-    if ((res = mp_grow (b, pa)) != MP_OKAY) {
-      return res;
-    }
-  }
-
-  /* number of output digits to produce */
-  W1 = 0;
-  for (ix = 0; ix < pa; ix++) { 
+   int       olduse, res, pa, ix, iz;
+   mp_digit   W[MP_WARRAY], *tmpx;
+   mp_word   W1;
+
+   /* grow the destination as required */
+   pa = a->used + a->used;
+   if (b->alloc < pa) {
+      if ((res = mp_grow(b, pa)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* number of output digits to produce */
+   W1 = 0;
+   for (ix = 0; ix < pa; ix++) {
       int      tx, ty, iy;
       mp_word  _W;
       mp_digit *tmpy;
@@ -62,7 +62,7 @@ int fast_s_mp_sqr (const mp_int * a, mp_int * b)
        */
       iy = MIN(a->used-tx, ty+1);
 
-      /* now for squaring tx can never equal ty 
+      /* now for squaring tx can never equal ty
        * we halve the distance since they approach at a rate of 2x
        * and we have to round because odd cases need to be executed
        */
@@ -86,26 +86,26 @@ int fast_s_mp_sqr (const mp_int * a, mp_int * b)
 
       /* make next carry */
       W1 = _W >> ((mp_word)DIGIT_BIT);
-  }
-
-  /* setup dest */
-  olduse  = b->used;
-  b->used = a->used+a->used;
-
-  {
-    mp_digit *tmpb;
-    tmpb = b->dp;
-    for (ix = 0; ix < pa; ix++) {
-      *tmpb++ = W[ix] & MP_MASK;
-    }
-
-    /* clear unused digits [that existed in the old copy of c] */
-    for (; ix < olduse; ix++) {
-      *tmpb++ = 0;
-    }
-  }
-  mp_clamp (b);
-  return MP_OKAY;
+   }
+
+   /* setup dest */
+   olduse  = b->used;
+   b->used = a->used+a->used;
+
+   {
+      mp_digit *tmpb;
+      tmpb = b->dp;
+      for (ix = 0; ix < pa; ix++) {
+         *tmpb++ = W[ix] & MP_MASK;
+      }
+
+      /* clear unused digits [that existed in the old copy of c] */
+      for (; ix < olduse; ix++) {
+         *tmpb++ = 0;
+      }
+   }
+   mp_clamp(b);
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_2expt.c b/libtommath/bn_mp_2expt.c
index 989bb9f..bdc868f 100644
--- a/libtommath/bn_mp_2expt.c
+++ b/libtommath/bn_mp_2expt.c
@@ -15,31 +15,31 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* computes a = 2**b 
+/* computes a = 2**b
  *
  * Simple algorithm which zeroes the int, grows it then just sets one bit
  * as required.
  */
 int
-mp_2expt (mp_int * a, int b)
+mp_2expt(mp_int *a, int b)
 {
-  int     res;
+   int     res;
 
-  /* zero a as per default */
-  mp_zero (a);
+   /* zero a as per default */
+   mp_zero(a);
 
-  /* grow a to accomodate the single bit */
-  if ((res = mp_grow (a, (b / DIGIT_BIT) + 1)) != MP_OKAY) {
-    return res;
-  }
+   /* grow a to accomodate the single bit */
+   if ((res = mp_grow(a, (b / DIGIT_BIT) + 1)) != MP_OKAY) {
+      return res;
+   }
 
-  /* set the used count of where the bit will go */
-  a->used = (b / DIGIT_BIT) + 1;
+   /* set the used count of where the bit will go */
+   a->used = (b / DIGIT_BIT) + 1;
 
-  /* put the single bit in its place */
-  a->dp[b / DIGIT_BIT] = ((mp_digit)1) << (b % DIGIT_BIT);
+   /* put the single bit in its place */
+   a->dp[b / DIGIT_BIT] = ((mp_digit)1) << (b % DIGIT_BIT);
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_abs.c b/libtommath/bn_mp_abs.c
index e7c5e25..343a102 100644
--- a/libtommath/bn_mp_abs.c
+++ b/libtommath/bn_mp_abs.c
@@ -15,26 +15,26 @@
  * Tom St Denis, tstdenis82@gmail.com, http://libtom.org
  */
 
-/* b = |a| 
+/* b = |a|
  *
  * Simple function copies the input and fixes the sign to positive
  */
 int
-mp_abs (mp_int * a, mp_int * b)
+mp_abs(mp_int *a, mp_int *b)
 {
-  int     res;
+   int     res;
 
-  /* copy a to b */
-  if (a != b) {
-     if ((res = mp_copy (a, b)) != MP_OKAY) {
-       return res;
-     }
-  }
+   /* copy a to b */
+   if (a != b) {
+      if ((res = mp_copy(a, b)) != MP_OKAY) {
+         return res;
+      }
+   }
 
-  /* force the sign of b to positive */
-  b->sign = MP_ZPOS;
+   /* force the sign of b to positive */
+   b->sign = MP_ZPOS;
 
-  return MP_OKAY;
+   return MP_OKAY;
 }
 #endif
 
diff --git a/libtommath/bn_mp_add.c b/libtommath/bn_mp_add.c
index a68c02e..d31d5a0 100644
--- a/libtommath/bn_mp_add.c
+++ b/libtommath/bn_mp_add.c
@@ -16,34 +16,34 @@
  */
 
 /* high level addition (handles signs) */
-int mp_add (const mp_int * a, const mp_int * b, mp_int * c)
+int mp_add(const mp_int *a, const mp_int *b, mp_int *c)
 {
-  int     sa, sb, res;
+   int     sa, sb, res;
 
-  /* get sign of both inputs */
-  sa = a->sign;
-  sb = b->sign;
+   /* get sign of both inputs */
+   sa = a->sign;
+   sb = b->sign;
 
-  /* handle two cases, not four */
-  if (sa == sb) {
-    /* both positive or both negative */
-    /* add their magnitudes, copy the sign */
-    c->sign = sa;
-    res = s_mp_add (a, b, c);
-  } else {
-    /* one positive, the other negative */
-    /* subtract the one with the greater magnitude from */
-    /* the one of the lesser magnitude.  The result gets */
-    /* the sign of the one with the greater magnitude. */
-    if (mp_cmp_mag (a, b) == MP_LT) {
-      c->sign = sb;
-      res = s_mp_sub (b, a, c);
-    } else {
+   /* handle two cases, not four */
+   if (sa == sb) {
+      /* both positive or both negative */
+      /* add their magnitudes, copy the sign */
       c->sign = sa;
-      res = s_mp_sub (a, b, c);
-    }
-  }
-  return res;
+      res = s_mp_add(a, b, c);
+   } else {
+      /* one positive, the other negative */
+      /* subtract the one with the greater magnitude from */
+      /* the one of the lesser magnitude.  The result gets */
+      /* the sign of the one with the greater magnitude. */
+      if (mp_cmp_mag(a, b) == MP_LT) {
+         c->sign = sb;
+         res = s_mp_sub(b, a, c);
+      } else {
+         c->sign = sa;
+         res = s_mp_sub(a, b, c);
+      }
+   }
+   return res;
 }
 
 #endif
diff --git a/libtommath/bn_mp_add_d.c b/libtommath/bn_mp_add_d.c
index c2f2e0e..5270d27 100644
--- a/libtommath/bn_mp_add_d.c
+++ b/libtommath/bn_mp_add_d.c
@@ -17,93 +17,93 @@
 
 /* single digit addition */
 int
-mp_add_d (const mp_int * a, mp_digit b, mp_int * c)
+mp_add_d(const mp_int *a, mp_digit b, mp_int *c)
 {
-  int     res, ix, oldused;
-  mp_digit *tmpa, *tmpc, mu;
-
-  /* grow c as required */
-  if (c->alloc < (a->used + 1)) {
-     if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
-        return res;
-     }
-  }
-
-  /* if a is negative and |a| >= b, call c = |a| - b */
-  if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) {
-     mp_int a_ = *a;
-     /* temporarily fix sign of a */
-     a_.sign = MP_ZPOS;
-
-     /* c = |a| - b */
-     res = mp_sub_d(&a_, b, c);
-
-     /* fix sign  */
-     c->sign = MP_NEG;
-
-     /* clamp */
-     mp_clamp(c);
-
-     return res;
-  }
-
-  /* old number of used digits in c */
-  oldused = c->used;
-
-  /* source alias */
-  tmpa    = a->dp;
-
-  /* destination alias */
-  tmpc    = c->dp;
-
-  /* if a is positive */
-  if (a->sign == MP_ZPOS) {
-     /* add digit, after this we're propagating
-      * the carry.
-      */
-     *tmpc   = *tmpa++ + b;
-     mu      = *tmpc >> DIGIT_BIT;
-     *tmpc++ &= MP_MASK;
-
-     /* now handle rest of the digits */
-     for (ix = 1; ix < a->used; ix++) {
-        *tmpc   = *tmpa++ + mu;
-        mu      = *tmpc >> DIGIT_BIT;
-        *tmpc++ &= MP_MASK;
-     }
-     /* set final carry */
-     ix++;
-     *tmpc++  = mu;
-
-     /* setup size */
-     c->used = a->used + 1;
-  } else {
-     /* a was negative and |a| < b */
-     c->used  = 1;
-
-     /* the result is a single digit */
-     if (a->used == 1) {
-        *tmpc++  =  b - a->dp[0];
-     } else {
-        *tmpc++  =  b;
-     }
-
-     /* setup count so the clearing of oldused
-      * can fall through correctly
-      */
-     ix       = 1;
-  }
-
-  /* sign always positive */
-  c->sign = MP_ZPOS;
-
-  /* now zero to oldused */
-  while (ix++ < oldused) {
-     *tmpc++ = 0;
-  }
-  mp_clamp(c);
-
-  return MP_OKAY;
+   int     res, ix, oldused;
+   mp_digit *tmpa, *tmpc, mu;
+
+   /* grow c as required */
+   if (c->alloc < (a->used + 1)) {
+      if ((res = mp_grow(c, a->used + 1)) != MP_OKAY) {
+         return res;
+      }
+   }
+
+   /* if a is negative and |a| >= b, call c = |a| - b */
+   if ((a->sign == MP_NEG) && ((a->used > 1) || (a->dp[0] >= b))) {
+      mp_int a_ = *a;
+      /* temporarily fix sign of a */
+      a_.sign = MP_ZPOS;
+
+      /* c = |a| - b */
+      res = mp_sub_d(&a_, b, c);
+
+      /* fix sign  */
+      c->sign = MP_NEG;
+
+      /* clamp */
+      mp_clamp(c);
+
+      return res;
+   }
+
+   /* old number of used digits in c */
+   oldused = c->used;
+
+   /* source alias */
+   tmpa    = a->dp;
+
+   /* destination alias */
+   tmpc    = c->dp;
+
+   /* if a is positive */
+   if (a->sign == MP_ZPOS) {
+      /* add digit, after this we're propagating
+       * the carry.
+       */
+      *tmpc   = *tmpa++ + b;
+      mu      = *tmpc >> DIGIT_BIT;
+      *tmpc++ &= MP_MASK;
+
+      /* now handle rest of the digits */
+      for (ix = 1; ix < a->used; ix++) {
+         *tmpc   = *tmpa++ + mu;
+         mu      = *tmpc >> DIGIT_BIT;
+         *tmpc++ &= MP_MASK;
+      }
+      /* set final carry */
+      ix++;
+      *tmpc++  = mu;
+
+      /* setup size */
+      c->used = a->used + 1;
+   } else {
+      /* a was negative and |a| < b */
+      c->used  = 1;
+
+      /* the result is a single digit */
+      if (a->used == 1) {
+         *tmpc++  =  b - a->dp[0];
+      } else {
+         *tmpc++  =  b;
+      }
+
+      /* setup count so the clearing of oldused
+       * can fall through correctly
+       */
+      ix       = 1;
+   }
+
+   /* sign always positive */
+   c->sign = MP_ZPOS;
+
+   /* now zero to oldused */
+   while (ix++ < oldused) {
+      *tmpc++ = 0;
+   }
+   mp_clamp(c);
+
+   return MP_OKAY;
 }
 
 #endif
diff --git a/libtommath/bn_mp_addmod.c b/libtommath/bn_mp_addmod.c
index dc06788..0521974 100644
--- a/libtommath/bn_mp_addmod.c
+++ b/libtommath/bn_mp_addmod.c
@@ -17,22 +17,22 @@
 
 /* d = a + b (mod c) */
 int
-mp_addmod (mp_int * a, mp_int * b, mp_int * c, mp_int * d)
+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 e90ab37..c6e8cf7 100644
--- a/libtommath/bn_mp_and.c
+++ b/libtommath/bn_mp_and.c
@@ -17,39 +17,39 @@
 
 /* AND two ints together */
 int
-mp_and (const mp_int * a, const mp_int * b, mp_int * c)
+mp_and(const mp_int *a, const mp_int *b, mp_int *c)
 {
-  int     res, ix, px;
-  mp_int  t;
-  const mp_int *x;
+   int     res, ix, px;
+   mp_int  t;
+   const mp_int *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..e5c8907 100644
--- a/libtommath/bn_mp_clamp.c
+++ b/libtommath/bn_mp_clamp.c
@@ -15,7 +15,7 @@
  * 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
@@ -23,19 +23,19 @@
  * are no more leading digits
  */
 void
-mp_clamp (mp_int * a)
+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..eada2d7 100644
--- a/libtommath/bn_mp_clear.c
+++ b/libtommath/bn_mp_clear.c
@@ -17,25 +17,25 @@
 
 /* clear one (frees)  */
 void
-mp_clear (mp_int * a)
+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 09cafa0..37f616e 100644
--- a/libtommath/bn_mp_cmp.c
+++ b/libtommath/bn_mp_cmp.c
@@ -17,24 +17,24 @@
 
 /* compare two ints (signed)*/
 int
-mp_cmp (const mp_int * a, const mp_int * b)
+mp_cmp(const mp_int *a, const 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 3768f06..576a073 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(const mp_int * a, mp_digit b)
+int mp_cmp_d(const 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 3b7f1b4..e2c723f 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 (const mp_int * a, const mp_int * b)
+int mp_cmp_mag(const mp_int *a, const 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 10a901f..9a94d3d 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 435729c..0f31138 100644
--- a/libtommath/bn_mp_copy.c
+++ b/libtommath/bn_mp_copy.c
@@ -17,49 +17,49 @@
 
 /* copy, b = a */
 int
-mp_copy (const mp_int * a, mp_int * b)
+mp_copy(const 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 10ab695..f2cd190 100644
--- a/libtommath/bn_mp_count_bits.c
+++ b/libtommath/bn_mp_count_bits.c
@@ -17,26 +17,26 @@
 
 /* returns the number of bits in an int */
 int
-mp_count_bits (const mp_int * a)
+mp_count_bits(const 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 30b36b6..6bad443 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(const mp_int * a, const mp_int * b, mp_int * c, mp_int * d)
+int mp_div(const mp_int *a, const 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 (const mp_int * a, const mp_int * b, mp_int * c, mp_int * d)
+int mp_div(const mp_int *a, const 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 0e981a0..edc8982 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(const mp_int * a, mp_int * b)
+int mp_div_2(const 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 d76de1a..eae3498 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 (const mp_int * a, int b, mp_int * c, mp_int * d)
+int mp_div_2d(const 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 5e75438..8011965 100644
--- a/libtommath/bn_mp_div_3.c
+++ b/libtommath/bn_mp_div_3.c
@@ -17,59 +17,59 @@
 
 /* divide by three (based on routine from MPI and the GMP manual) */
 int
-mp_div_3 (const mp_int * a, mp_int *c, mp_digit * d)
+mp_div_3(const 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 4767565..9997646 100644
--- a/libtommath/bn_mp_div_d.c
+++ b/libtommath/bn_mp_div_d.c
@@ -21,7 +21,7 @@ static int s_is_power_of_two(mp_digit b, int *p)
 
    /* quick out - if (b & (b-1)) isn't zero, b isn't a power of two */
    if ((b == 0) || ((b & (b-1)) != 0)) {
-       return 0;
+      return 0;
    }
    for (x = 1; x < DIGIT_BIT; x++) {
       if (b == (((mp_digit)1)<<x)) {
@@ -33,78 +33,78 @@ static int s_is_power_of_two(mp_digit b, int *p)
 }
 
 /* single digit division (based on routine from MPI) */
-int mp_div_d (const mp_int * a, mp_digit b, mp_int * c, mp_digit * d)
+int mp_div_d(const mp_int *a, mp_digit b, mp_int *c, mp_digit *d)
 {
-  mp_int  q;
-  mp_word w;
-  mp_digit t;
-  int     res, ix;
+   mp_int  q;
+   mp_word w;
+   mp_digit t;
+   int     res, ix;
 
-  /* cannot divide by zero */
-  if (b == 0) {
-     return MP_VAL;
-  }
+   /* 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;
-  }
+   /* 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;
-  }
+   /* 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..e90dc55 100644
--- a/libtommath/bn_mp_dr_reduce.c
+++ b/libtommath/bn_mp_dr_reduce.c
@@ -30,64 +30,64 @@
  * 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)
+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..b49c81a 100644
--- a/libtommath/bn_mp_dr_setup.c
+++ b/libtommath/bn_mp_dr_setup.c
@@ -21,8 +21,8 @@ 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..a95560a 100644
--- a/libtommath/bn_mp_exch.c
+++ b/libtommath/bn_mp_exch.c
@@ -15,17 +15,17 @@
  * 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)
+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..43228a8 100644
--- a/libtommath/bn_mp_export.c
+++ b/libtommath/bn_mp_export.c
@@ -18,67 +18,68 @@
 /* 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