### diff bn_mp_karatsuba_sqr.c @ 282:91fbc376f010libtommath-origlibtommath-0.35

Import of libtommath 0.35 From ltm-0.35.tar.bz2 SHA1 of 3f193dbae9351e92d02530994fa18236f7fde01c
author Matt Johnston Wed, 08 Mar 2006 13:16:18 +0000 97db060d0ef5
line wrap: on
line diff
```--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/bn_mp_karatsuba_sqr.c	Wed Mar 08 13:16:18 2006 +0000
@@ -0,0 +1,117 @@
+#include <tommath.h>
+#ifdef BN_MP_KARATSUBA_SQR_C
+/* LibTomMath, multiple-precision integer library -- Tom St Denis
+ *
+ * LibTomMath is a library that provides multiple-precision
+ * integer arithmetic as well as number theoretic functionality.
+ *
+ * The library was designed directly after the MPI library by
+ * Michael Fromberger but has been written from scratch with
+ * additional optimizations in place.
+ *
+ * The library is free for all purposes without any express
+ * guarantee it works.
+ *
+ * Tom St Denis, [email protected], http://math.libtomcrypt.org
+ */
+
+/* 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
+ * tuned to perform recursive squarings.
+ */
+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;
+
+  {
+    register int x;
+    register 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 (mp_sub (&x1, &x0, &t1) != MP_OKAY)
+    goto X1X1;           /* t1 = x1 - x0 */
+  if (mp_sqr (&t1, &t1) != MP_OKAY)
+    goto X1X1;           /* t1 = (x1 - x0) * (x1 - x0) */
+
+  if (s_mp_add (&x0x0, &x1x1, &t2) != MP_OKAY)
+    goto X1X1;           /* t2 = x0x0 + x1x1 */
+  if (mp_sub (&t2, &t1, &t1) != MP_OKAY)
+    goto X1X1;           /* t1 = x0x0 + x1x1 - (x1-x0)*(x1-x0) */
+
+  /* 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;
+}
+#endif```