RFR(s): 8023541 Race condition in rmid initialization
tristan.yan at oracle.com
Wed Jan 29 07:23:13 UTC 2014
Should variable initialized be volatile here? Otherwise looks good.
On Jan 29, 2014, at 2:51 PM, Stuart Marks <stuart.marks at oracle.com> wrote:
> Hi all,
> Please review this fix to a race condition in rmid initialization. Briefly, rmid subclasses the RMI registry implementation and provides special handling for its own stub. Unfortunately the registry is exported in the super() call, making remote calls possible before rmid's stub initialization is complete. The fix is to ensure that all remote calls wait for initialization before proceeding.
More information about the core-libs-dev