[PATCH] Zero register

Boredom + insomnia = ....

This is just a simple zero op for numeric registers. Its minutely
faster than set $REG, 0 and on intel at least, and I don't know how it
maps to JIT.

I won't commit this myself because I'm still unsure the minor speed is
worth it, however I feel compiler freaks might like having it.

-Melvin


Index: core.ops
===================================================================
RCS file: /cvs/public/parrot/core.ops,v
retrieving revision 1.96
diff -u -r1.96 core.ops
--- core.ops    18 Feb 2002 08:26:03 -0000      1.96
+++ core.ops    28 Feb 2002 08:42:45 -0000
@@ -537,6 +537,29 @@
    goto NEXT();
  }

+########################################
+
+=item B<zero>(in INT)
+
+=item B<zero>(in NUM)
+
+    Zero out register $1
+    Faster than set, $1, 0  (we hope)
+
+=cut
+
+inline op zero(in INT) {
+       $1 = 0;
+       goto NEXT();
+}
+
+inline op zero(in NUM) {
+       $1 = 0;
+       goto NEXT();
+}
+
+########################################
+
  =item B<set_keyed>(out PMC, out PMC, in PMC, in PMC)

      $1[$2] = $3[$4];

• [PATCH] Zero register by Melvin Smith