[11u]: 8262121: [11u] Redo 8244287: JFR: Methods samples have line number 0

[Jaroslav Bachorík]

Hi Christop,

On Mon, Mar 15, 2021 at 4:45 PM Langer, Christoph <christoph.langer at sap.com>
wrote:

> Hi Goetz,
>
> it would have been nice if there was an incremental webrev compared to the
> previous version. But I think I've figured out that the increment was a fix
> in the test. ��
>
> So, it seems that the update has indeed fixed the test failures that we
> observed before, hence it should be ok.
>
> The indentation in line 82 of TestStackFrameLineNumbers looks a bit
> awkward, you should fix that. But no need for a further webrev.
>
> 82                       System.out.println("=== Frame: " + frame);
>
> Question to Jaroslav: It seems like the test is a valuable regression
> test. Shouldn't tit be added to jdk/jdk head as well?
>

Goetz is the original author of the test - I just fixed the corner case
where no samples were generated due to low activity. But yes, it would
definitely make sense to have this test in jdk/jdk head too.

-JB-


>
> Best regards
> Christoph
>
>
> > -----Original Message-----
> > From: jdk-updates-dev <jdk-updates-dev-retn at openjdk.java.net> On
> > Behalf Of Lindenmaier, Goetz
> > Sent: Freitag, 12. März 2021 11:41
> > To: jdk-updates-dev <jdk-updates-dev at openjdk.java.net>
> > Cc: 'Jaroslav Bachorík' <jaroslav.bachorik at datadoghq.com>
> > Subject: RE: [11u]: 8262121: [11u] Redo 8244287: JFR: Methods samples
> have
> > line number 0
> >
> > Hi,
> >
> > We had seen sporadic test failures in the test added with this
> > change.
> > Jaroslav proposed a fix. Thanks!  Since applying that fix I did not see
> > the problem any more.
> >
> > This new webrev contains the fix by Jaroslav:
> > http://cr.openjdk.java.net/~goetz/wr21/8262121-Redo-8244287-
> > JFR_line_no_0-jdk11/02/
> >
> > Best regards,
> >   Goetz.
> >
> >
> >
> > > -----Original Message-----
> > > From: Lindenmaier, Goetz
> > > Sent: Monday, February 22, 2021 2:18 PM
> > > To: jdk-updates-dev <jdk-updates-dev at openjdk.java.net>
> > > Subject: [11u]: 8262121: [11u] Redo 8244287: JFR: Methods samples have
> > > line number 0
> > >
> > > Hi,
> > >
> > > We tried to backport "8244287: JFR: Methods samples have line number 0"
> > > to 11.0.9 [1].  The first downport had to be backed out, though [2].
> > >
> > > I now propose this change for the issue. Please review.
> > > http://cr.openjdk.java.net/~goetz/wr21/8262121-Redo-8244287-
> > > JFR_line_no_0-jdk11/01/
> > > I have verified that we do not see the crashes we saw with the
> > > original change.
> > >
> > > New Issue:
> > > https://bugs.openjdk.java.net/browse/JDK-8262121
> > >
> > > Original Change:
> > > https://bugs.openjdk.java.net/browse/JDK-8244287
> > > Previous downport:
> > >  https://bugs.openjdk.java.net/browse/JDK-8253576
> > > Backout:
> > > https://bugs.openjdk.java.net/browse/JDK-8253813
> > >
> > > I don't understand why the previous backport is
> > > flagged as 11.010. It was pushed on 2020-09-23,
> > > which was in the 11.0.9 time frame. But we had
> > > issues with the automatic tagging in the past, probably
> > > this is the reason.
> > >
> > > I will only push this to 11.0.12.
> > > I would also adapt the backport information so that
> > > JBS shows that this is only fixed in 11.0.12.
> > >
> > > Best regards,
> > >   Goetz
> > >
> > > [1] https://mail.openjdk.java.net/pipermail/jdk-updates-dev/2020-
> > > June/003312.html
> > > [2] http://mail.openjdk.java.net/pipermail/jdk-updates-dev/2020-
> > > September/003886.html
>