Bug in G1 logging?

Thomas Schatzl thomas.schatzl at oracle.com
Mon Jun 15 08:50:11 UTC 2015


Hi,

On Mon, 2015-06-15 at 10:41 +0200, Thomas Viessmann wrote:
> Hi all,
> 
> There is probably a bug in G1 logging in 8u40. The total of [Other: 
> 214.0 ms]
> does not match the sum  of its sub-categories which amount to just 9.2 ms.
> Hence at least one of these is supposedly wrong. Is this a known bug or 
> should I file a new one?

Can you provide at least command line parameters to see if some (not
mentioned) logging/verification/diagnostic option causes this?

> 
>    [Code Root Fixup: 0.1 ms]
>     [Code Root Purge: 0.0 ms]
>     [Clear CT: 2.0 ms]
>     [Other: 214.0 ms]<<<<<<<<<<<<<<<<<<
>        [Choose CSet: 0.0 ms]
>        [Ref Proc: 4.9 ms]
>        [Ref Enq: 0.3 ms]
>        [Redirty Cards: 2.1 ms]
>           [Parallel Redirty:  1.8  1.8  1.7  1.7  1.7  1.7 1.7  1.7  
> 1.7  1.7  1.6  1.6  1.6
>            Min: 1.6, Avg: 1.7, Max: 1.8, Diff: 0.2, Sum: 22.1]
>           [Redirtied Cards:  27319  24512  23040  24635 18996  28913  
> 20628  19163  17881  17664  15360  18417  16639
>            Min: 15360, Avg: 21012.8, Max: 28913, Diff: 13553, Sum: 273167]
>        [Humongous Reclaim: 0.0 ms]
>           [Humongous Total: 11]
>           [Humongous Candidate: 0]
>           [Humongous Reclaimed: 0]
>        [Free CSet: 1.9 ms]
>           [Young Free CSet: 1.9 ms]
>           [Non-Young Free CSet: 0.0 ms]
>     [Eden: 3400.0M(3232.0M)->0.0B(848.0M) Survivors: 176.0M->376.0M 
> Heap: 12.6G(24.0G)->9673.3M(24.0G)]

Thanks,
  Thomas




More information about the hotspot-gc-dev mailing list