RFR (XS) 8220290: gc_epilogue(full = false) is called on Full GC path after JDK-8215221

Thomas Schatzl thomas.schatzl at oracle.com
Fri Mar 8 10:58:25 UTC 2019


Hi,

On Thu, 2019-03-07 at 23:19 +0100, Thomas Schatzl wrote:
> Hi,
> 
> On Thu, 2019-03-07 at 23:08 +0100, Aleksey Shipilev wrote:
> > On 3/7/19 9:17 PM, Thomas Schatzl wrote:
> > > Hi,
> > > 
> > > On Thu, 2019-03-07 at 19:05 +0100, Aleksey Shipilev wrote:
> > > > Bug:
> > > >  https://bugs.openjdk.java.net/browse/JDK-8220290
> > > > 
> > > > Fix:
> > > >  http://cr.openjdk.java.net/~shade/8220290/webrev.01/
> > > > 
> > > > This regression is caused by recent refactoring in JDK-8215221.
> > > > The comments in the bug should describe what happened.
> > > > 
> > > > Testing: Linux x86_64 fastdebug: 20 rounds of
> > > > gc/arguments/TestSurvivorRatioFlag with CMS, 5 rounds
> > > > of hotspot/gc with {Serial, Parallel, CMS}, hotspot tier1, jdk-
> > > > submit (in progress)
> > > 
> > >   looks good.
> > 
> > Thanks!
> > 
> > > Already started hs-tier1-5 run some time ago, but there are
> > > problems with the CI right now...
> > 
> > Okay, I'll go catch some sleep, and maybe wait for that testing to
> > be completed?
> 
> That's probably best. The CI having problems also mean that we won't
> get annoyed by automated failure notices too in the meantime :)
> 
> I will give a heads up tomorrow as soon as the testing completed.

Testing almost complete, no issues. Ship it.

Thomas




More information about the hotspot-gc-dev mailing list