--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/libtommath/bn_fast_s_mp_mul_digs.c	Wed Mar 08 13:23:49 2006 +0000
@@ -0,0 +1,105 @@
+#include <tommath.h>
+#ifdef BN_FAST_S_MP_MUL_DIGS_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
+ */
+
+/* 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 
+ * 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 
+ * 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     olduse, res, pa, ix, iz;
+  mp_digit W[MP_WARRAY];
+  register mp_word  _W;
+
+  /* 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);
+
+  /* clear the carry */
+  _W = 0;
+  for (ix = 0; ix < pa; ix++) { 
+      int      tx, ty;
+      int      iy;
+      mp_digit *tmpx, *tmpy;
+
+      /* get offsets into the two bignums */
+      ty = MIN(b->used-1, ix);
+      tx = ix - ty;
+
+      /* setup temp aliases */
+      tmpx = a->dp + tx;
+      tmpy = b->dp + ty;
+
+      /* this is the number of times the loop will iterrate, essentially 
+         while (tx++ < a->used && ty-- >= 0) { ... }
+       */
+      iy = MIN(a->used-tx, ty+1);
+
+      /* execute loop */
+      for (iz = 0; iz < iy; ++iz) {
+         _W += ((mp_word)*tmpx++)*((mp_word)*tmpy--);
+      }
+
+      /* store term */
+      W[ix] = ((mp_digit)_W) & MP_MASK;
+
+      /* make next carry */
+      _W = _W >> ((mp_word)DIGIT_BIT);
+  }
+
+  /* store final carry */
+  W[ix] = (mp_digit)(_W & MP_MASK);
+
+  /* setup dest */
+  olduse  = c->used;
+  c->used = pa;
+
+  {
+    register 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;
+}
+#endif