Unterminated StackProfiler thread in case of failure of benchmark
Evgeny Mandrikov
mandrikov at gmail.com
Mon Jan 12 20:05:12 UTC 2015
Hi Aleksey,
I don't think that possible to bypass stop of profilers after this change
:) but anyway checked my scenarios and as expected it works like a charm.
Thanks for fix!
On Mon, Jan 12, 2015 at 10:43 AM, Aleksey Shipilev <
aleksey.shipilev at oracle.com> wrote:
> On 01/12/2015 12:32 PM, Aleksey Shipilev wrote:
> > On 12/30/2014 05:35 PM, Evgeny Mandrikov wrote:
> >> Seems that there is no notification for profilers in case of failure of
> >> benchmark ( LoopBenchmarkHandler can exit with BenchmarkException before
> >> reaching stopProfilers ) , so that forks won't be terminated if profiler
> >> spawns thread. This is the case for StackProfiler and can be reproduced
> by
> >> addition of ".addProfiler(StackProfiler.class)" into
> FailingForkedBenchTest.
> >
> > Indeed, thanks:
> > https://bugs.openjdk.java.net/browse/CODETOOLS-7901234
>
> And fixed:
> http://hg.openjdk.java.net/code-tools/jmh/rev/5164ebcc4936
>
> The regression test is added to verify, but you may want to verify in
> your scenario as well.
>
> Thanks,
> -Aleksey.
>
>
>
--
Best regards,
Evgeny Mandrikov aka Godin <http://godin.net.ru>
http://twitter.com/_godin_
More information about the jmh-dev
mailing list