Inconsistent "Fix Version: 15" for JDK-8247367

[Jesper Wilhelmsson]

Hi Aleksey.

Yes, there were indeed a couple of issues that went in after the 16+0 tag (which was the tip for forking the JDK 15 update repo) and before we got hg updater set to map jdk/jdk to JDK 16.

I'll go through these and make sure JBS has the correct fixVersion recorded and notify owners that their fix didn't make it to JDK 15.

For JDK-8247367, this is a P4 enhancement so don't think it should be backported to JDK 15. I'd recommend just closing the backport.

For JDK-8247358 which was in the same state I cleaned it up a little. Again this is a P4 enhancement so backporting to JDK 15 would require approval at this point.

Thanks!
/Jesper


> On 15 Jun 2020, at 17:30, Daniel D. Daugherty <daniel.daugherty at oracle.com> wrote:
> 
> Hi Alexsey,
> 
> This fix was pushed at 2020.06.11 @ 1216 ET/1616 UTC so it did not make
> the cutoff for JDK15. However, it looks like the hgupdater switch from
> JDK15 -> JDK16 for jdk/jdk pushes happened after that point.
> 
> I have changed this bug:
> 
>     JDK-8247367 Shenandoah: pacer should wait on lock instead of exponential backoff
> 
> to have a 'Fix Version/s' value of '16' instead of '15'.
> 
> I have re-opened the backport bug:
> 
>     JDK-8247468 Shenandoah: pacer should wait on lock instead of exponential backoff
> 
> changed the 'Affects Version/s' value to '15'. If you don't intend to
> backport the fix for JDK-8247367 to JDK15, then we can close the backport
> as "Won't Fix".
> 
> The next push to jdk/jdk (for JDK16) was this one:
> 
>     JDK-8207936 TestZipFile failed with java.lang.AssertionError exception
> 
> and it appears to be correct.
> 
> The last push to jdk/jdk (for JDK15) was this one:
> 
>     JDK-8246382 assert in MetaspaceShared::map_archives
> 
> and it also appears to be correct.
> 
> It looks like just your bug (JDK-8247367) got caught in the worm hole. :-)
> 
> Dan
> 
> 
> On 6/15/20 10:05 AM, Aleksey Shipilev wrote:
>> Hi,
>> 
>> I am a bit confused about hgupdater closing some issues as "Fix Versions: 15". For example, see
>> these issue, it is "Fixed in 15":
>>   https://bugs.openjdk.java.net/browse/JDK-8247367
>> 
>> The changeset is indeed in jdk/jdk:
>>   http://hg.openjdk.java.net/jdk/jdk/log?rev=8247367
>> 
>> ...but not in jdk/jdk15:
>>   http://hg.openjdk.java.net/jdk/jdk15/log?rev=8247367
>> 
>> Is there some kind of version switch race that this push got into?
>> E.g. hgupdater switched to 16 for jdk/jdk some time after jdk/jdk -> jdk/jdk15 split?
>> 
>> There might be more issues like this.
>> 
>