11.0.9 misses JDK-8250861

[Gil Tene]

> On Oct 29, 2020, at 1:31 AM, Lindenmaier, Goetz <goetz.lindenmaier at sap.com> wrote:
> 
> Hi,
> 
> So should we release a 11.0.9.0.1 version?

Without opining in any way on whether or not a respin of 11.0.9 is needed or not,
I'd like to chime in on the version number for such a respin *if* one is done. We have
not done any of those in the 11+ version range this far (we did e.g. 8u265 as an urgent
update after 8u262, but the versioning convention there is different).

According to JEP322 [1], the first four elements in the version are assigned meaning:
$FEATURE.$INTERIM.$UPDATE.$PATCH

And "...The fifth and later elements of version numbers are reserved for use by
downstream consumers of the JDK code base."

As such, I think the 11u (and later updates projects) should not use the fifth element,
and that the right version for an 11.0.9 "patch" (if one is done) would be 11.0.9.1

[1] https://openjdk.java.net/jeps/322

> I could post a RFR for the version change.
> I think we should also bump the release date, then.
> But to which date? Maybe Monday, 2.11.?
> 
> Actually, we have detected another problem introduced
> in 11.0.8, Christoph has the details.
> It increases memory footprint severely and came up
> in our productive systems. Unfortunately, no fix
> exists yet in OpenJDK, we are working on it.
> But it might be another candidate for a 11.0.9.0.1.
> 
> Best regards,
>  Goetz.
> 
> 
> 
> 
> -----Original Message-----
> From: jdk-updates-dev <jdk-updates-dev-retn at openjdk.java.net> On Behalf Of Andrew Haley
> Sent: Donnerstag, 29. Oktober 2020 12:10
> To: jdk-updates-dev at openjdk.java.net
> Subject: Re: 11.0.9 misses JDK-8250861
> 
> Hi,
> 
> On 29/10/2020 10:45, Severin Gehwolf wrote:
>> 
>> On Thu, 2020-10-29 at 10:22 +0000, Doerr, Martin wrote:
>>> Hi,
>>> 
>>> we have observed a crash with SIGSEGV in the C2 code on SPARC (but
>>> it's not in platform specific code) in 11.0.9. This code was
>>> introduced with
>>> http://hg.openjdk.java.net/jdk-updates/jdk11u/rev/ed3959f95671
>>> 
>>> Fix is already available:
>>> https://bugs.openjdk.java.net/browse/JDK-8250861
>>> 
>>> I've requested backport for 11.0.10 so I'm planning to push it to
>>> jdk11u-dev. Does anybody need it in jdk11u?
>> 
>> Approved for 11.0.10.
>> 
>> I'll let others chime in on whether or not this warrants a respin for
>> 11.0.9.
> 
> I would have thought so.
> 
> --
> Andrew Haley  (he/him)
> Java Platform Lead Engineer
> Red Hat UK Ltd. <https://www.redhat.com>
> https://keybase.io/andrewhaley
> EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671
>