RFR(M): 8044186 Introduce a reproducible random generator
david.r.chase at oracle.com
Fri Sep 19 21:21:45 UTC 2014
On 2014-09-19, at 5:03 PM, Igor Ignatyev <igor.ignatyev at oracle.com> wrote:
> don't you need an additional check RANDOM_GENERATOR == null after line #355 to prevent double creating of Random which leads to the problems w/ reproducing on MT environments?
Alternately just drop the test inside the synchronized block, because how often do we call that anyway?
There’s one minor consolation, there’d be double-printing of initial values in the log file if that race happened.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 801 bytes
Desc: Message signed with OpenPGP using GPGMail
More information about the hotspot-compiler-dev