Downport 8217338: [Containers] Improve systemd slice memory limit support

[Lindenmaier, Goetz]

Hi Severin, everybody,

I wanted to follow your plan and let you do this downport. 
But it was in one of my queues in between of other changes
that were approved and which I can push.
Now I accidentally pushed this change along with the others, 
sorry!

I ran the change through our testing and it's all green. 
But I did not request downport, so it was not approved (yet).
Also I thought it needs some discussion as Andrew Haley
had questioned it in a comment:
https://bugs.openjdk.java.net/browse/JDK-8217338
It brings some changes in internal classes in java.base.

What should I do? Should I back it out, or should 
I request downport and only back it out in case downport
is rejected?

Me personally am in favour of bringing the cgroup improvements
to 11. I think it's an important thing for Java to support.

Best regards
  Goetz

PS: the current process of bringing changes to 11 that were
downported by Oracle is extremely cumbersome.  I am doing
lots of completely pointless webrevs/reviews, and I have to 
coordinate with a row of other resources: reviewer, approval, 
testing. In the end I have 15-30 open changes for downport, 
some not reviewed, some not approved, some in the testing, 
some I need to fix as they did not pass testing. This leads to errors.
And only from the testing I ever got useful feedback ... 
It would be much better if I could just push all changes that
applied without substantial conflicts after they passed testing.




> -----Original Message-----
> From: Severin Gehwolf <sgehwolf at redhat.com>
> Sent: Dienstag, 7. Januar 2020 15:12
> To: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>
> Cc: jdk-updates-dev at openjdk.java.net
> Subject: Re: Downport 8217338: [Containers] Improve systemd slice memory
> limit support
> 
> Hi Goetz!
> 
> On Fri, 2020-01-03 at 14:38 +0000, Lindenmaier, Goetz wrote:
> > Hi Severin,
> >
> > I'm working on downporting the 11.0.7-oracle changes.
> > https://bugs.openjdk.java.net/browse/JDK-8232207: Linux
> os::available_memory re-reads cgroup configuration on every invocation
> > depends on 8217338.  You had once intended to downport
> > https://bugs.openjdk.java.net/browse/JDK-8217338.
> > The downport was then abandoned.
> >
> > I am wondering whether you want to pick this up again?
> 
> Sure, I can pick it up again. I'll try to get it done in the next week
> or so.
> 
> > 8217338 is productive in 13 for a while now, and it applies
> > clean.
> > I'm not sure about the Java changes, do they change the API?
> 
> Not that I'm aware of. Cgroup related code is Linux only and in an
> jdk.internal package.
> 
> > The hotspot changes help to apply 8232207, though.
> >
> > I would like to wait with 8232207 until this is decided.
> 
> Thanks for reaching out.
> 
> Backporting 8232207 is a good idea.
> 
> Thanks,
> Severin