[
https://issues.apache.org/jira/browse/OWB-579?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13043422#comment-13043422
]
Kevan Miller commented on OWB-579:
----------------------------------
There were some comments/questions about why Aries should be marking their
methods final. Their motivation seems to have been performance. I asked around
to see if there was any truth in this. I got the following response from Derek
Inglis, who works on IBM's JIT:
{quote}
Marking methods final does give JIT performance improvements for IBM JVM and
Oracle's JVMs. Having said that much of the JIT work over the years has been
to decrease the overhead associated with nonfinal methods, such that the
advantage of final has decreased. In summary, I would encourage using final
on methods, however the advantage for a small number of methods, isn't big
enough to worry about if avoiding final for those methods solves other issues.
{quote}
Personally, I think OWB should ignore synthetic, final methods. I also think
Aries should not create them...
> check for non-proxyiable methods should exclude synthetic methods
> -----------------------------------------------------------------
>
> Key: OWB-579
> URL: https://issues.apache.org/jira/browse/OWB-579
> Project: OpenWebBeans
> Issue Type: Bug
> Components: Core
> Affects Versions: 1.1.1
> Reporter: David Jencks
> Assignee: David Jencks
> Fix For: 1.1.1
>
>
> Aries trunk has some weaving code that adds some final synthetic methods to
> just about every class. AFAICT javassist can still proxy these classes but
> OWB check for proxiablility fails.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
