Re: Patch to fix Reference/natReference.

[Tom Tromey <tromey at redhat dot com>]

>>>>> "David" == David Daney <ddaney@avtrex.com> writes:

David> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12013

BTW, there are a couple other buglets in Reference.
Just in case you're interested, these are the ones I know of.

First, the Sun docs say that unless a PhantomReference has clear()
called, its referent will remain phantom-referenceable.  I don't know
how this situation could be detected, but we don't enforce it.  In our
implementation, once we've processed all the PhantomReferences, the
referent will be collected.

Second, String.intern() and references don't play well together.
There's a PR for this one.  It's tempting to just remove our current
intern code and use the Classpath implementation instead.  Another
approach would be to have String.intern communicate with natReference
about intern()d strings.

Tom