------- Additional Comments From bonzini at gcc dot gnu dot org 2005-05-07
09:22 -------
With this version of AWTEvent.java, this is an ice-on-valid:
package java.awt;
import java.util.EventObject;
public abstract class AWTEvent extends EventObject {
AWTEvent() { super(null); }
}
I have prepared a patch which is now under testing, and it is really a "how
could it have worked" situation. When creating ctxp_for_generation we add at
the head of the list, so passes 2 and 3 are invoked in LIFO order. By the time
we try to resolve AWTEvent in B.java, java_complete_class (pass 2) has not been
done on AWTEvent yet. It seems obvious to me that passes 2 and 3 should be
executed in FIFO order.
In addition, find_in_imports should look at ctxp->import_list even if enclosing
is not NULL, if enclosing's TYPE_IMPORT_LIST is NULL (likewise for
find_in_on_demand_imports).
Patch under testing.
--
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |bonzini at gcc dot gnu dot
|dot org |org
Status|UNCONFIRMED |ASSIGNED
Ever Confirmed| |1
Keywords| |ice-on-valid-code
Last reconfirmed|0000-00-00 00:00:00 |2005-05-07 09:22:59
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21436
