RFR(S): 8213527: [testbug] Fix tests using UseCompressedOops for 32-bit.

[David Holmes]

Hi Goetz,

On 8/11/2018 9:28 PM, Lindenmaier, Goetz wrote:
> Ah, thanks to pointing me to that change.
> 
> Should I remove the VMProps case from this change? I guess the other
> fix is needed anyways.

Yes please. Consider the fix to PrintMetaspaceDcmd Reviewed.

Thanks,
David

> Best regards,
>    Goetz.
> 
>> -----Original Message-----
>> From: David Holmes <david.holmes at oracle.com>
>> Sent: Donnerstag, 8. November 2018 12:07
>> To: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>; hotspot-runtime-
>> dev at openjdk.java.net
>> Subject: Re: RFR(S): 8213527: [testbug] Fix tests using UseCompressedOops for
>> 32-bit.
>>
>> Hi Goetz,
>>
>> Boris already has this issue out for review under:
>>
>> 8213410: UseCompressedOops @requirement check fails fails on 32-bit system
>>
>> We're currently waiting on feedback from jtreg folk on exactly how
>> @requires is expected to work to see which way to fix it cleanly.
>>
>> I would not want to special case UseCompressedOops in VMProps.java
>>
>> Thanks,
>> David
>>
>> On 8/11/2018 9:00 PM, Lindenmaier, Goetz wrote:
>>> Hi,
>>>
>>> A few tests that use the UseCompressedOops flag are failing on 32-bit
>> platforms.
>>>
>>> Adapt the test script of PrintMetaspaceDcmd to only test the relevant case on
>> 32-bit and to omit the unsupported flags.
>>>
>>> Adapt VMProps.vmOptFinalFlag to return 'false' on 32-bit for
>> UseCompressedOops. The flag is not defined at all in 32-bit builds, thus currently
>> 'null' is returned by vmOptFinalFlag.
>>>
>>> http://cr.openjdk.java.net/~goetz/wr18/8213527-32bit_tests/01/
>>>
>>> Best regards,
>>>     Goetz.
>>>