RFR(XS): 8004170: G1: Verbose GC output is not getting flushed to log file using JDK 8

John Cuthbertson john.cuthbertson at oracle.com
Tue Dec 4 14:13:17 PST 2012


Hi Kirk,

I thought CMS only used PrintGC and PrintGCDetails - I can't find 
CMSLogLevel mentioned in the source.

If you're referring to the JEP that was recently published - then I 
think the intent is to normalize them.

Cheers,

JohnC

On 12/04/12 09:21, Kirk Pepperdine wrote:
> Would it be worthwhile to explore normalizing the G1LogLevel with CMSLogLevel values?
>
> Regards,
> Kirk
> On 2012-12-03, at 6:52 PM, John Cuthbertson <john.cuthbertson at oracle.com> wrote:
>
>   
>> Hi Everyone,
>>
>> Can I have a couple of volunteers look over the changes for this CR - the webrev can be found at: http://cr.openjdk.java.net/~johnc/8004170/webrev.0/
>>
>> Summary:
>> When we changed the G1 evacuation pause to not use instances of the TraceTime class we forgot to flush the GC log file (which happened during the TraceTime destructor). The fix is to add a flush call to G1CollectedHeap::log_gc_footer(). I also noticed that TraceCPUTime destructor does not flush the specified log file and, since it only seems to be instantiated from GC code, it probably should.
>>
>> Testing:
>> GCOld with -XX:G1LogLevel=fine,finer,finest and -Xloggc. The submitter of the CR has also verified the fix.
>>
>> Thanks,
>>
>> JohnC
>>     
>
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-gc-dev/attachments/20121204/6934b54c/attachment.html 


More information about the hotspot-gc-dev mailing list