[11u] RFR: JDK-8223697 jfr tool can't format duration values greater than 1 minute

[Severin Gehwolf]

On Thu, 2019-10-24 at 18:22 +0000, Hohensee, Paul wrote:
> This is an instance of a fanout problem that will become more common
> as tip's code base diverges from that of 11u.

True. FWIW, we have that problem in JDK 8u already (diverged code
base).

> Maintainers, shall we make it a policy to do predecessor backports
> such that we can get clean 'master' backports?

I wouldn't go as far as making it policy, but rather investigate this
on a case-by-case basis. Perhaps it should indeed be the first attempt
to investigate what it would take for a backport to apply cleanly.
However, we need to assess risk involved getting dependencies
backported too.

In this particular case, I believe, the risk of backporting the dep as
well is acceptable since it's small enough, contained to JFR tool code,
no API change and adds some value on its own. Hence, I'd opt to
downport the dep (JDK-8215771) too.

This would be very different if the dependency would be a massive
patch, possible not even API compatible.

That's my $0.02 anyway.

Thanks,
Severin
> 
> On 10/24/19, 7:36 AM, "jdk-updates-dev on behalf of Jie Kang" <jdk-updates-dev-bounces at openjdk.java.net on behalf of jkang at redhat.com> wrote:
> 
>     On Thu, Oct 24, 2019 at 3:49 AM Severin Gehwolf <sgehwolf at redhat.com> wrote:
>     >
>     > Hi Jie,
>     >
>     > On Wed, 2019-10-23 at 16:15 -0400, Jie Kang wrote:
>     > > Hi,
>     > >
>     > > Please review this backport for [1]. The fix did not apply cleanly due
>     > > to line differences from missing backport of [2]. The additions are
>     > > the same but at different lines than in jdk/jdk.
>     > >
>     > > This fix is quite small and useful for the jfr tool in jdk 11. The
>     > > tier one and jfr tests ran successfully on my machine.
>     >
>     > Is there a good reason NOT to backport JDK-8215771 as well? Then this
>     > patch would apply cleanly AFAUI and JDK-8215771 seems useful in itself.
>     > JDK-8215771 seems small enough. Thoughts?
>     
>     Hi Severin,
>     
>     I considered JDK-8215771 to be of the 'feature enhancement' type,
>     while this (JDK-8223697) is of 'bug fix' type. This was my thinking
>     for the backport request of this but not of JDK-8215771. However, I
>     realize I can request a fix for JDK-8215771 to see if it is acceptable
>     to the 11u maintainers. It will indeed make this a clean backport and
>     bring in useful changes. I will proceed as such; thanks for bringing
>     this up!
>     
>     
>     Regards,
>     Jie Kang
>     
>     >
>     > Thanks,
>     > Severin
>     >
>     > > Webrev:
>     > > http://cr.openjdk.java.net/~jkang/JDK-8223697/webrev.01/
>     > > Bug:
>     > > https://bugs.openjdk.java.net/browse/JDK-8223697
>     > >
>     > > [1] https://bugs.openjdk.java.net/browse/JDK-8223697
>     > > [2] https://bugs.openjdk.java.net/browse/JDK-8215771
>     > >
>     > >
>     > > Regards,
>     > > Jie Kang
>     > >
>     >
>     
>