RFR(XS): 8066589: Make importing sa-jdi.jar optional on its existance

[Dmitry Samersoff]

Volker,

I'm OK with your fix.

-Dmitry

On 2014-12-09 20:42, Volker Simonis wrote:
> That's fine, but we currently can not build on AIX because of this
> issue and I don't want to stay without builds until the mentioned bugs
> are fixed (especially when the fix for the issue is so simple and
> non-intrusive for other platforms).
> 
> Regards,
> Volker
> 
> On Tue, Dec 9, 2014 at 6:23 PM, Dmitry Samersoff
> <dmitry.samersoff at oracle.com> wrote:
>> Volker,
>>
>> Please notice, there is work-in-progress to cease out any dependencies
>> to jdk.hotspot.agent from other modules.
>>
>>
>> Please, see:
>>
>> https://bugs.openjdk.java.net/browse/JDK-8059038
>> https://bugs.openjdk.java.net/browse/JDK-8059035
>>
>> -Dmitry
>>
>>
>> On 2014-12-09 20:12, Volker Simonis wrote:
>>> On Fri, Dec 5, 2014 at 10:11 AM, Erik Joelsson <erik.joelsson at oracle.com> wrote:
>>>> Hello Volker,
>>>>
>>>> Are these the only conditions for when sa-jdi.jar is not built? If so, then
>>>> I suppose this is fine.
>>>
>>> Yes. But with my proposed solution any new platform may easily add
>>> itself to the list of platforms which don't have the SA-agent.
>>>
>>>>
>>>> The old Import.gmk would only copy sa-jdi.jar if it existed, and I think we
>>>> can keep that behavior, so just an existence check on sa-jdi.jar is good
>>>> enough in Import.gmk. In Gensrc-jdk.jdi.gmk, checking if
>>>> $(SUPPORT_OUTPUTDIR)/gensrc/jdk.hotspot.agent/_the.sa.services exists should
>>>> be fine with me. We lose a bit of error checking in the build doing it that
>>>> way as we won't fail if that file is missing for other reasons.
>>>>
>>>
>>> I don't quite understand. If a platform doesn't support the SA-agent
>>> there's no need for any error checking. This fix doesn't change the
>>> behaviour on any other platform except for aix-ppc64 and ZERO. Any
>>> other platforms which don't support the SA just add themselves to the
>>> lisst in the if-statement without affecting the other platforms.
>>>
>>>
>>>
>>>> Note that this hacking of the service provider files is a temporary hack
>>>> until service providers are properly handled in the modular world, so no
>>>> need for fancy solutions.
>>>
>>> OK fine. I've added one more tiny fix which was needed to build on
>>> AIX. It's in an if-def AIX anyway, so it won't impact other platforms.
>>> It just fixes the location of the static version of libjli:
>>>
>>> http://cr.openjdk.java.net/~simonis/webrevs/8066589.v2/
>>>
>>> OK to push now?
>>>
>>> Thanks,
>>> Volker
>>>
>>>>
>>>> /Erik
>>>>
>>>>
>>>> On 2014-12-04 18:49, Volker Simonis wrote:
>>>>>
>>>>> Hi,
>>>>>
>>>>> could you please review this tiny change which fixes the build on
>>>>> platforms which don't built the SA agent after the recent modualrity
>>>>> integrations:
>>>>>
>>>>> http://cr.openjdk.java.net/~simonis/webrevs/8066589
>>>>> https://bugs.openjdk.java.net/browse/JDK-8066589
>>>>>
>>>>> I've tested that the fix works on AIX but I havn't had a chance to build
>>>>> Zero.
>>>>>
>>>>> @Xerxes: maybe you could check if my suggested fix also solves your
>>>>> build problems. I'm also no sure if the "ifneq ($(JVM_VARIANT_ZERO),
>>>>> true)" clause also catches the ZEROSHARK case (altough I think it
>>>>> should). If not we would need yet another "ifneq
>>>>> ($(JVM_VARIANT_ZEROSHARK), true)"
>>>>>
>>>>> Thanks,
>>>>> Volker
>>>>
>>>>
>>
>>
>> --
>> Dmitry Samersoff
>> Oracle Java development team, Saint Petersburg, Russia
>> * I would love to change the world, but they won't give me the sources.


-- 
Dmitry Samersoff
Oracle Java development team, Saint Petersburg, Russia
* I would love to change the world, but they won't give me the sources.