fix minor typo in float conversions

R=rsc
APPROVED=rsc
DELTA=2  (0 added, 0 deleted, 2 changed)
OCL=35820
CL=35820

src/cmd/5g/gsubr.c

@@ -791,8 +791,8 @@ gmove(Node *f, Node *t)
 	case CASE(TUINT32, TFLOAT64):
 		fa = AMOVW;
 		a = AMOVWD;
-		ta = AMOVW;
-		goto fltconv;;
+		ta = AMOVD;
+		goto fltconv;
 
 	case CASE(TUINT64, TFLOAT32):
 	case CASE(TUINT64, TFLOAT64):