Speed up wide_int_storage::operator=(wide_int_storage const&)

wide_int_storage shows up high in the profile for the testcase in
PR114855 where the apparent issue is that the conditional jump
on 'precision' after the (inlined) memcpy stalls the pipeline due
to the data dependence and required store-to-load forwarding.  We
can add scheduling freedom by instead testing precision as from the
source which speeds up the function by 30%.  I've applied the
same logic to the copy CTOR.

	* wide-int.h (wide_int_storage::wide_int_storage): Branch
	on source precision to avoid data dependence on memcpy
	destination.
	(wide_int_storage::operator=): Likewise.

diff --git a/gcc/wide-int.h b/gcc/wide-int.h
index 64b8bf2040c1b59102d06684a8bb13a96cc35254..777f017f5ae04c45d2554eab374ff086b6dcea73 100644 (file)
--- a/gcc/wide-int.h
+++ b/gcc/wide-int.h
@@ -1196,7 +1196,7 @@ inline wide_int_storage::wide_int_storage (const T &x)
 inline wide_int_storage::wide_int_storage (const wide_int_storage &x)
 {
   memcpy (this, &x, sizeof (wide_int_storage));
-  if (UNLIKELY (precision > WIDE_INT_MAX_INL_PRECISION))
+  if (UNLIKELY (x.precision > WIDE_INT_MAX_INL_PRECISION))
     {
       u.valp = XNEWVEC (HOST_WIDE_INT, CEIL (precision, HOST_BITS_PER_WIDE_INT));
       memcpy (u.valp, x.u.valp, len * sizeof (HOST_WIDE_INT));
@@ -1219,9 +1219,9 @@ wide_int_storage::operator = (const wide_int_storage &x)
       XDELETEVEC (u.valp);
     }
   memcpy (this, &x, sizeof (wide_int_storage));
-  if (UNLIKELY (precision > WIDE_INT_MAX_INL_PRECISION))
+  if (UNLIKELY (x.precision > WIDE_INT_MAX_INL_PRECISION))
     {
-      u.valp = XNEWVEC (HOST_WIDE_INT, CEIL (precision, HOST_BITS_PER_WIDE_INT));
+      u.valp = XNEWVEC (HOST_WIDE_INT, CEIL (x.precision, HOST_BITS_PER_WIDE_INT));
       memcpy (u.valp, x.u.valp, len * sizeof (HOST_WIDE_INT));
     }
   return *this;