Re: [8u] RFR for backport of 8156028: G1YoungGenSizer _adaptive_size not correct when setting NewSize and MaxNewSize to the same value

Liang Mao maoliang.ml at alibaba-inc.com
Wed Nov 27 07:33:40 UTC 2019


Hi Paul,

The webrev is updated. Since I'm not a committer, could you please help to push?
http://cr.openjdk.java.net/~ddong/8156028/webrev.00/

Thanks,
Liang


------------------------------------------------------------------
From:Hohensee, Paul <hohensee at amazon.com>
Send Time:2019 Nov. 26 (Tue.) 22:20
To:"MAO, Liang" <maoliang.ml at alibaba-inc.com>; jdk8u-dev <jdk8u-dev at openjdk.java.net>
Subject:Re: [8u] RFR for backport of 8156028: G1YoungGenSizer _adaptive_size not correct when setting NewSize and MaxNewSize to the same value

This looks fine, except that it would be good to add a newline after the declaration of _adaptive_size to match the original patch.

Thanks,
Paul

 On 11/26/19, 12:20 AM, "jdk8u-dev on behalf of Liang Mao" <jdk8u-dev-bounces at openjdk.java.net on behalf of maoliang.ml at alibaba-inc.com> wrote:

    Hi,
    Original bug:
https://bugs.openjdk.java.net/browse/JDK-8156028
http://hg.openjdk.java.net/jdk/jdk/rev/ba8be1a71dec

    Original patch does not apply cleanly to 8u, because class G1YoungGenSizer has been
    moved to seperated file.

    Since G1 is already a stable feature in JDK8U, the obvious bug that the young gen size
    cannot be adaptive with different NewSize and MaxNewSize makes users confused. The
    alternative way to use experimental option *G1NewSizePercent* is not intuitive and formal.

    8u webrev:
http://cr.openjdk.java.net/~ddong/8156028/webrev.00/
    Testing: x86_64 build, affected tests, tier1
    Thanks,
    Liang






More information about the jdk8u-dev mailing list