RFR (S): CR 8006697: ContendedPaddingWidth should be range-checked

David Holmes david.holmes at oracle.com
Mon May 6 22:05:38 PDT 2013


Range check seems reasonable.

Minor nit: you could use INTX_FORMAT in the printf rather than 
"(int)ContendedPaddingWidth".

Thanks,
David

On 3/05/2013 6:40 PM, Aleksey Shipilev wrote:
> Friendly reminder.
>
> -Aleksey.
>
> On 04/29/2013 04:44 PM, Aleksey Shipilev wrote:
>> Hi,
>>
>> Long-lasting fix for this CR is available here:
>>    http://cr.openjdk.java.net/~shade/8006997/webrev.03/
>>
>> We need to test for the sensible padding sizes, as well as the
>> alignment. Hence, these two checks during startup. I had selected [0;
>> 8192] bytes as the sensible range for padding, please speak up if you
>> think this is wrong.
>>
>> Testing:
>>    - full JPRT cycle
>>    - adhoc cmdline testing with Linux x86_64/fastdebug
>>      (the issue is trivial enough so I haven't built the jtreg test)
>>
>> -Aleksey.
>>
>


More information about the hotspot-runtime-dev mailing list