[8u40] Request for approval: 8064391: More thread safety problems in core reflection

[Joel Borggrén-Franck]

Hi Martin,

I still like to backport your patch as is, because:

- It is small and "obviously correct"
- Fixing an existing, but hard to repro race
- I was assuming "Here are changes we made locally at Google" implied it
  was in production

If any of this was incorrect, I agree we should wait and reiterate this
some more on core-libs-dev. If my assumptions were correct I think this
should go in as is, potentially followed by more fixes later on.

cheers
/Joel

On 2014-11-13, Martin Buchholz wrote:
> Hold on...  There are some additional thread safety improvements
> ("volatile") that have not yet landed in jdk9, and I think Peter
> Levart has proposed additions to my proposed fixes for 7 and 8 (and
> I've fallen behind keeping up with the Class* changes)
> 
> On Thu, Nov 13, 2014 at 6:05 AM, Joel Borggrén-Franck
> <joel.franck at oracle.com> wrote:
> > Hi,
> >
> > Can I get approval to push a fix for: "8064391: More thread safety problems in core reflection" to 8u.
> >
> > This is a 8u version of a more general fix in 9 http://hg.openjdk.java.net/jdk9/dev/jdk/rev/1e5c2bd424d5
> >
> > The fix is provided by Martin Bucholz : http://cr.openjdk.java.net/~martin/webrevs/openjdk8/Class-thread-safety/
> > Reviewed by me here: http://mail.openjdk.java.net/pipermail/core-libs-dev/2014-November/029644.html
> >
> > JBS issue: https://bugs.openjdk.java.net/browse/JDK-8064391
> >
> > cheers
> > /Joel