RFR: 8140489: Introduce shorthand for average_time_ms in G1CollectorPolicy

Erik Helin erik.helin at oracle.com
Wed Oct 28 10:23:14 UTC 2015


On 2015-10-27, Derek White wrote:
> This looks fine.

Thanks!

On 2015-10-27, Derek White wrote:
> It would be really nice to factor out all of the "update_stats" work from
> record_collection_pause_end(), so we can see the actually policy decisions.
> But maybe you are headed in that direction.

I agree, and yes, I'm looking at more ways to improve the code in
record_collection_pause_end.

Thanks,
Erik

>  - Derek
> 
>   On 10/27/15 11:41 AM, Erik Helin wrote:
> >Hi all,
> >
> >this small patch introduces the little helper method
> >G1CollectorPolicy::average_tims_ms instead of calling
> >g1_phases()->average_time_ms, making the code a bit shorter and easier
> >to read.
> >
> >Enhancement:
> >https://bugs.openjdk.java.net/browse/JDK-8140489
> >
> >Webrev:
> >http://cr.openjdk.java.net/~ehelin/8140489/webrev.00/
> >
> >Testing:
> >JPRT
> >
> >Thanks,
> >Erik
> 



More information about the hotspot-gc-dev mailing list