regcomp.sym: longj field is a boolean

The comments could lead one to thinking one could specify any of the
argument fields that nodes can have.  But in fact, the value is a
boolean, 0 meaning to use the normal offset field of all regnodes; and 1
meaning to use the ARG field that some regnodes have.  If a regnode had
more than just the one argument field, the one that corresponds to that
would be used.

This commit enforces that, and changes regcomp.sym to not use  '2',
which is misleading.

It clarifies the comments about this and what '.' means in the flags
field

1 files changed, 6 insertions, 1 deletions

diff --git a/regen/regcomp.pl b/regen/regcomp.pl
index ffb7d7a13f..cb9861318d 100644
--- a/regen/regcomp.pl
+++ b/regen/regcomp.pl
@@ -54,7 +54,7 @@ use strict;
 # code          Op      what code is associated with this node (???)
 # args          Op      what type of args the node has (which regnode struct)
 # flags         Op      (???)
-# longj         Op      Whether this node is a longjump
+# longj         Op      Boolean as to if this node is a longjump
 # comment       Both    Comment about node, if any
 # pod_comment   Both    Special comments for pod output (preceding lines in def)
 
@@ -89,6 +89,11 @@ sub register_node {
     $node->{id}= 0 + @all;
     push @all, $node;
     $all{ $node->{name} }= $node;
+
+    if ($node->{longj} && $node->{longj} != 1) {
+        die "longj field must be in [01] if present in ", Dumper($node);
+    }
+
 }
 
 # Parse and add an opcode definition to the global state.