[11u]: RFR JDK-8227368: EnumSet.class serialization broken in JDK 9+

[Peter Levart]

Hi Christoph,

The CSR was involved for the fix in JDK 13, because for JDK 13 the 
published specification did change (and the JCK tests too probably), but 
Stuart modified the original patch so that it does not change the 
published specification and passes the JDK 11 JCK test(s). So it is 
believed that JSR process is not needed for this modified patch for JDK 
11u. Of course, it is up to you and Andrew Haley to decide. It is true 
that this is a behavioral change (reverting unintended past behavioral 
change). The EnumSet.class object serialized with JDK 11u4 would fail to 
be deserialized with JDK 11u5 (if this patch gets into 11u5) and vice versa.

So what do you think? Is CSR still required for this patch?

If it is required, I'll open a JDK11 specific CSR. The contents of the 
JDK 11 specific CSR would probably differ since the patch differs too, 
right? Do I link the CSR to the original issue too or just to the 
backport issues?

Regards, Peter

On 8/8/19 5:14 PM, Langer, Christoph wrote:
> Hi Peter,
>
> the fix has gotten jdk11u approval.
>
> However, it's an item that involves a CSR. So, before you can push to jdk11u, you'll need to open a JDK 11 specific CSR and have it reviewed. The contents should be copy&paste from the original CSR.
>
> I suggest you link the CSR to both, the 11-pool and the 11.0.6-oracle backport item.
>
> Thanks
> Christoph
>
>> -----Original Message-----
>> From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net> On
>> Behalf Of Peter Levart
>> Sent: Donnerstag, 8. August 2019 13:24
>> To: Stuart Marks <stuart.marks at oracle.com>
>> Cc: jdk-updates-dev at openjdk.java.net
>> Subject: Re: [11u]: RFR JDK-8227368: EnumSet.class serialization broken in
>> JDK 9+
>>
>> Hi Stuart,
>>
>> I see the JDK-8227368 has got an approval label (jdk11u-fix-yes). Should
>> the change you have prepared for this backport [1] be pushed to
>> jdk-updates/jdk11u-dev now?
>>
>> Regards, Peter
>>
>> On 7/18/19 1:11 AM, Stuart Marks wrote:
>>>
>>> On 7/17/19 3:48 PM, Stuart Marks wrote:
>>>> I've set the fix-version of the backport record JDK-8227650 to 11.0.5
>>>> ...
>>> I changed this to 11-pool on advice from Joe Darcy, as this will make
>>> it more likely that that the automatic hg-updater will pick up this
>>> backport issue when the fix gets pushed to some 11-update release.
>>>
>>> s'marks