RFR (XXS): 8132709: [TESTBUG] gc/g1/TestHumongousShrinkHeap.java might fail on embedded

Dmitry Fazunenko dmitry.fazunenko at oracle.com
Mon Sep 7 10:49:23 UTC 2015


Hi Thomas,


On 07.09.2015 12:44, Thomas Schatzl wrote:
> Hi,
>
> On Fri, 2015-09-04 at 20:30 +0300, Dmitry Fazunenko wrote:
>> Hello,
>>
>> May I have reviews for a small test fix:
>> https://bugs.openjdk.java.net/browse/JDK-8132709
>>
>> webrev:
>> http://cr.openjdk.java.net/~dfazunen/8132709/webrev/
>>
>> Comments: the test allocates/deallocates a number of humongous objects
>> to fragment the heap,
>> and then checks that fragmentation doesn't prevent heap shrinking. The
>> test didn't expect that
>> the amount of free memory could be not enough to allocate several
>> humongous objects (embedded platforms).
>> The fix adds such check.
>    looks good. Shouldn't the bug id updated in the test too? If so, I do
> not need a re-review for this.

I think in tests it's worth mentioning only major CRs like RFE# for a 
feature under test.
All minor CRs causing updates could be retrieved from the repository 
history when needed.

Thanks for the prompt review!
Dima


>
> Thanks,
>    Thomas
>
>




More information about the hotspot-gc-dev mailing list