RFR(S): 8167299: -XX:+PrintRelocations crashes the VM

[Volker Simonis]

PING - can I please have a sponsor for this change?

Thanks,
Volker


On Thu, Oct 6, 2016 at 7:15 PM, Vladimir Kozlov
<vladimir.kozlov at oracle.com> wrote:
> Looks good.
>
> Thanks,
> Vladimir
>
>
> On 10/6/16 9:32 AM, Volker Simonis wrote:
>>
>> Hi,
>>
>> can I please have a review and sponsor for this small fix which
>> re-enables the usage of -XX:+PrintRelocations:
>>
>> https://bugs.openjdk.java.net/browse/JDK-8167299
>> http://cr.openjdk.java.net/~simonis/webrevs/2016/8167299/
>>
>> If we're running with "PrintNMethods" (which is as debug option) a
>> call to Relocation::code() in Relocation::method_value() may return
>> NULL. We must check this and handle it appropriately.
>>
>> Also, if running with "-XX:+WizardMode",
>> RelocIterator::print_current() calls oop->print_value_on() without
>> checking if oop is actually a valid oop. This can be easily fixed by
>> calling oop->is_oop() before calling oop->print_value_on().
>>
>> Thank you and best regards,
>> Volker
>>
>