RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Claes Redestad
Hi,

JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was
recently
integrated into jdk10-hs, causing some trivially fixed test failures:

Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
Bug: https://bugs.openjdk.java.net/browse/JDK-8180479

Thanks!

/Claes

Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Tobias Hartmann-2
Hi Claes,

looks good to me, thanks for fixing!

Best regards,
Tobias

On 17.05.2017 12:16, Claes Redestad wrote:

> Hi,
>
> JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was recently
> integrated into jdk10-hs, causing some trivially fixed test failures:
>
> Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
> Bug: https://bugs.openjdk.java.net/browse/JDK-8180479
>
> Thanks!
>
> /Claes
>
Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Claes Redestad
Hi Tobias,

On 2017-05-17 12:26, Tobias Hartmann wrote:
> Hi Claes,
>
> looks good to me, thanks for fixing!

thanks for reviewing!

/Claes

>
> Best regards,
> Tobias
>
> On 17.05.2017 12:16, Claes Redestad wrote:
>> Hi,
>>
>> JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was recently
>> integrated into jdk10-hs, causing some trivially fixed test failures:
>>
>> Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8180479
>>
>> Thanks!
>>
>> /Claes
>>

Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Vladimir Kozlov
In reply to this post by Claes Redestad
Hi Claes,

Do you have bug to fix C1 and C2 intrinsics too after methods renaming?:

Thanks,
Vladimir

On 5/17/17 3:16 AM, Claes Redestad wrote:

> Hi,
>
> JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was recently
> integrated into jdk10-hs, causing some trivially fixed test failures:
>
> Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
> Bug: https://bugs.openjdk.java.net/browse/JDK-8180479
>
> Thanks!
>
> /Claes
>
Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Claes Redestad
Hi Vladimir, I don't know if that's planned - Paul, Ron?

Thanks!

/Claes

On 2017-05-17 17:20, Vladimir Kozlov wrote:

> Hi Claes,
>
> Do you have bug to fix C1 and C2 intrinsics too after methods renaming?:
>
> Thanks,
> Vladimir
>
> On 5/17/17 3:16 AM, Claes Redestad wrote:
>> Hi,
>>
>> JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was
>> recently
>> integrated into jdk10-hs, causing some trivially fixed test failures:
>>
>> Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
>> Bug: https://bugs.openjdk.java.net/browse/JDK-8180479
>>
>> Thanks!
>>
>> /Claes
>>

Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Vladimir Kozlov
Sorry for noise - I should have looked more
Original rename changes fixed intrinsics - everything is fine

Thanks
Vladimir

> On May 17, 2017, at 9:35 AM, Claes Redestad <[hidden email]> wrote:
>
> Hi Vladimir, I don't know if that's planned - Paul, Ron?
>
> Thanks!
>
> /Claes
>
>> On 2017-05-17 17:20, Vladimir Kozlov wrote:
>> Hi Claes,
>>
>> Do you have bug to fix C1 and C2 intrinsics too after methods renaming?:
>>
>> Thanks,
>> Vladimir
>>
>>> On 5/17/17 3:16 AM, Claes Redestad wrote:
>>> Hi,
>>>
>>> JDK-8159995 changed method names in jdk.internal.misc.Unsafe and was recently
>>> integrated into jdk10-hs, causing some trivially fixed test failures:
>>>
>>> Webrev: http://cr.openjdk.java.net/~redestad/8180479/hotspot.00/
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8180479
>>>
>>> Thanks!
>>>
>>> /Claes
>>>
>

Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Paul Sandoz

> On 17 May 2017, at 09:59, Vladimir Kozlov <[hidden email]> wrote:
>
> Sorry for noise - I should have looked more
> Original rename changes fixed intrinsics - everything is fine
>

Thanks. We avoided changing the name of C2 nodes, which could be a follow up fix.

I now realize i pushed the original fix to jdk10/jdk10 rather than jdk10/hs, sorry about that. I did run everything through JPRT both independently and for the push, so i am unsure why we missed updating those tests.

Paul.


> Thanks
> Vladimir
>
>> On May 17, 2017, at 9:35 AM, Claes Redestad <[hidden email]> wrote:
>>
>> Hi Vladimir, I don't know if that's planned - Paul, Ron?
>>
>> Thanks!
>>
>> /Claes
>>


signature.asc (858 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: RFR: 8180479: [TESTBUG] Some hotspot tests broken after internal Unsafe name changes

Tobias Hartmann-2
Hi Paul,

On 17.05.2017 19:21, Paul Sandoz wrote:
> I now realize i pushed the original fix to jdk10/jdk10 rather than jdk10/hs, sorry about that. I did run everything through JPRT both independently and for the push, so i am unsure why we missed updating those tests.

We do not execute all the HotSpot (compiler) tests on JPRT due to runtime constraints. For example, the two failing tests are not executed.

To avoid such failures in the future, I would suggest to run all hotspot tests on RBT before pushing.

Thanks,
Tobias

>>> On May 17, 2017, at 9:35 AM, Claes Redestad <[hidden email]> wrote:
>>>
>>> Hi Vladimir, I don't know if that's planned - Paul, Ron?
>>>
>>> Thanks!
>>>
>>> /Claes
>>>
>