This fix assumes handlers returned from defaultFactory are interchangeable and that defaultFactory is thread-safe. If these assumptions are valid, looks good to me.
> On 21 Nov 2018, at 14:31, Chris Hegarty <chris.hega...@oracle.com> wrote: > > >> On 21/11/18 13:04, Seán Coffey wrote: >> Thanks to all for the feedback. It makes sense to reduce the scope of the >> lock where possible. >> I've updated the webrev : >> http://cr.openjdk.java.net/~coffeys/webrev.8213942.v2/webrev/ > > Thanks Sean, this looks good. > > -Chris.
