[10u] RFA: JDK-8203305 and JDK-8203669 (RTM fixes for POWER9)

[Gustavo Romero]

On 06/08/2018 03:13 PM, Vladimir Kozlov wrote:
> I pushed both changesets to
> 
> http://hg.openjdk.java.net/jdk-updates/jdk10u

Thanks a lot, Vladimir!


Regards,
Gustavo
  
> Regards,
> Vladimir
> 
> On 6/8/18 7:02 AM, Gustavo Romero wrote:
>> Hello Vladimir,
>>
>> Do know somebody who is a Committer on the Updates project that could sponsor these two changes for me?
>>
>> It's already approved to be pushed.
>>
>> I don't know anybody listed as "JDK Updates Project – Committer" to ask such a favor...
>>
>>
>> Best regards,
>> Gustavo
>>
>> On 06/08/2018 10:32 AM, Rob McKenna wrote:
>>> A sponsor must push to the jdk-updates/jdk10u forest. There are no
>>> automatic syncs between the jdk project & the jdk-updates project.
>>>
>>>      -Rob
>>>
>>> On 07/06/18 23:18, Gustavo Romero wrote:
>>>> Hi Rob,
>>>>
>>>> Thanks for approving the backports.
>>>>
>>>> Is that necessary any sponsor to push them or the integration will happen
>>>> automatically from now on?
>>>>
>>>>
>>>> Regards,
>>>> Gustavo
>>>>
>>>> On 06/05/2018 04:18 PM, Gustavo Romero wrote:
>>>>> Hi Rob,
>>>>>
>>>>> On 06/04/2018 09:33 PM, Rob McKenna wrote:
>>>>>> Hi Gustavo,
>>>>>>
>>>>>> As per rules 1 & 2, please add a fix request comment to each bug along
>>>>>> with the jdk10u-fix-request label.
>>>>>>
>>>>>> (note, this is the new process - approval mails are no longer necessary)
>>>>>
>>>>> oh I see. Thank you. It's done. I was only unsure about how I could turn
>>>>> conspicuous the fact that it's a patchset and so getting 8203305 implies
>>>>> getting 8203669. I did a crossref: so on request for 8203305 I mention
>>>>> 8203669, and vice versa. Please, let me if it's not ok.
>>>>>
>>>>>
>>>>> Best regards,
>>>>> Gustavo
>>>>>
>>>>>>      -Rob
>>>>>>
>>>>>> On 04/06/18 17:05, Gustavo Romero wrote:
>>>>>>> Hi,
>>>>>>>
>>>>>>> Some distro releases, notably Ubuntu 18.04 LTS "Bionic Beaver", are already
>>>>>>> shipping OpenJDK 10. Ubuntu 18.04 is able to run on the new POWER9
>>>>>>> machines. On some of these new POWER9 machines the JVM won't be able to
>>>>>>> detect a new POWER9 TM mode (called "TM with No Suspend mode") and will
>>>>>>> halt on checking if the JVM is capable of enabling the RTM feature (which
>>>>>>> depends upon TM).
>>>>>>>
>>>>>>> So I would like to request the approval to backport the following two
>>>>>>> changes to 10u that address this issue on POWER9 DD2.1:
>>>>>>>
>>>>>>> Title        : RFR(XS): 8203305: PPC64: Improve TM detection for enabling RTM on Linux / POWER9 [1]
>>>>>>> Bug          : https://bugs.openjdk.java.net/browse/JDK-8203305
>>>>>>> Review thread: http://mail.openjdk.java.net/pipermail/hotspot-dev/2018-May/032313.html
>>>>>>> Patch        : http://hg.openjdk.java.net/jdk/jdk/raw-rev/e2d9b38630c3
>>>>>>>
>>>>>>> Title        : RFR(S) : 8203669: PPC64: Fix jtreg RTM tests after "8203305: Improve TM detection for enabling RTM on Linux / POWER9" [2]
>>>>>>> Bug          : https://bugs.openjdk.java.net/browse/JDK-8203669
>>>>>>> Review thread: http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2018-May/029091.html
>>>>>>> Patch        : http://hg.openjdk.java.net/jdk/jdk/raw-rev/ede65c4fb6da
>>>>>>>
>>>>>>> Both changes apply cleanly to 10u except for small offsets.
>>>>>>>
>>>>>>> Change [1] is PPC-only. Change [2] touches shared code but it just changes
>>>>>>> a jtreg test in regard to PPC and fixes a few typos. Although changes [1]
>>>>>>> and [2] are marked as "Enhancement" in the JBS, I understand that they
>>>>>>> actually fit the "CPU fix merges" as found on Rule 3, in the Request
>>>>>>> for Approval rules [3].
>>>>>>>
>>>>>>> Thank you.
>>>>>>>
>>>>>>>
>>>>>>> Best regards,
>>>>>>> Gustavo
>>>>>>>
>>>>>>> [3] http://openjdk.java.net/projects/jdk-updates/approval.html
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>