Request for review (XXS): 7067973: test/java/lang/management/MemoryMXBean/CollectionUsageThreshold.java hanging intermittently

[Bengt Rutisson]

Yet another update.

I realized that the original bug (7067973) is not a G1 specific bug. I 
thought it was since is has a test-fail-g1 keyword, but the log output 
in the CR indicates that it was not a G1 failure. The log says "Tenured 
Gen" instead of "G1 Old Gen". So, I am leaving 7067973 as it is and will 
track my proposed change in a separate CR:

7173460: G1: 
java/lang/management/MemoryMXBean/CollectionUsageThreshold.java failes 
with G1

I avoided changing the subject line of this email threads since I 
thought it would just confuse everyone.

Bengt

On 2012-06-01 09:54, Bengt Rutisson wrote:
>
> Hi again,
>
> A quick update on this. I ran the following tests using UTE on a Linux 
> x64 host just to make sure that my change doesn't break any other part 
> of the management API.
>
> java/lang/management
> com/sun/management
> sun/management
> vm.tmtools.testlist
> nsk.monitoring.testlist
>
> I did see some intermittent failures of  
> java/lang/management/ThreadMXBean/ResetPeakThreadCount.java, but I 
> think that is:
>
> 6766097: TEST_BUG: ResetPeakThreadCount fails because it does not join 
> threads
>
> Other than that all tests pass. This includes 
> java/lang/management/MemoryMXBean/CollectionUsageThreshold.java that 
> used to fail.
>
> Bengt
>
> On 2012-05-31 22:04, Bengt Rutisson wrote:
>>
>> Hi all,
>>
>> Can I have a couple of reviews for this really small change?
>> http://cr.openjdk.java.net/~brutisso/7067973/webrev.00/
>>
>> Background:
>> The CollectionUsageThreshold test fails with G1. The test lowers the 
>> notification threshold for the G1 old gen memory pool and expects to 
>> get a notification after a full GC.
>>
>> The problem in G1 is that the decision to send the notification is 
>> done in TraceMemoryManagerStats::~TraceMemoryManagerStats(). This 
>> eventually does pool->get_memory_usage() to get the memory usage 
>> after a collection. The problem is that we update this information in 
>> G1MonitoringSupport::update_sizes() which is called in 
>> G1CollectedHeap::do_collection() _after_ the TraceMemoryManagerStats 
>> scope had been exited.
>>
>> Extending the scope to cover the call to 
>> G1MonitoringSupport::update_sizes() solves the issue.
>>
>> Testing:
>> Before this change the CollectionUsageThreshold failed every time I 
>> ran it. After this change it passes every time I ran it.
>>
>> Thanks,
>> Bengt
>