RFR(M): 8155634: Clean out old logging and dead code from SurvRateGroup

[Jon Masamitsu]

On 05/02/2016 05:40 AM, Mikael Gerdin wrote:
> Hi,
>
> Please review this change to clean out unused code from SurvRateGroup. 
> The G1YoungSurvRateNumRegionsSummary flag is only accessible in 
> notproduct builds and it is my opinion that the ability to print the 
> predicted aggregate survivor rates for eden regions at JVM shutdown is 
> not a particularly useful feature.
>
> Testing: Local testing with gc test suite.
> Bug: https://bugs.openjdk.java.net/browse/JDK-8155634
> Webrev: http://cr.openjdk.java.net/~mgerdin/8155634/webrev.0/

Mikael,

Changes look good.

I don't have much of an opinion on whether the survivor rate summary is
useful but I note that

http://cr.openjdk.java.net/~mgerdin/8155634/webrev.0/src/share/vm/gc/g1/g1CollectedHeap.cpp.udiff.html

  void G1CollectedHeap::print_tracing_info() const {
    g1_rem_set()->print_summary_info();
    concurrent_mark()->print_summary_info();
- g1_policy()->print_yg_surv_rate_info();
  }

the remembered set and concurrent marking summaries are being kept.
Can you explain why survivor rate is the exception?  All seems to
me averages over such an extended period that they all deserve the
same fate.

Thanks.

Jon


>
> /Mikael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/hotspot-gc-dev/attachments/20160502/108c09b9/attachment.htm>