[8u-dev] RFR+RFA (M): JDK-8176100: [REDO][REDO] G1 Needs pre barrier on dereference of weak JNI handles

[Hohensee, Paul]

Hi Andrew,

The commit is http://hg.openjdk.java.net/jdk8u/jdk8u-dev/hotspot/rev/32bc598624bd. Both problems are in jni.cpp.

It wasn't intentional: I overlooked it moving our internal patch to jdk8u-dev. It's there because a compiler for another internally-supported platform complained. We weren't aware of 8062808. If you like, I'll create a backport issue for 8062808 and point it to the 8176100 backport patch, or I could revert the 8062808-specific code and and do 8062808 separately.

I'll file an issue to fix the copyright date.

Thanks,

Paul

On 5/14/19, 6:34 AM, "Andrew John Hughes" <gnu.andrew at redhat.com> wrote:

    
    
    On 07/05/2019 21:45, Hohensee, Paul wrote:
    > Thank you, Andrey. Pushed in my name since the author, Mikael Gerdin, is
    > inactive.
    > 
    
    I noticed when integrating this downstream that JDK-8064786 [0] was also
    silently included in this change. Was this intentional?
    
    8062808, which prompts this change, is on my list to backport and we
    have the fix locally in RPMs. But there should be no need for 8064786 if
    the -Wreturn-type warning is not turned on by 8062808. Were you seeing
    such build failures with jdk8u-dev?
    
    Incidentally, there also seemed to be one line which rolled back a
    copyright line. If the copyright line is already later than the proposed
    change, it should be left as is.
    
    [0] https://bugs.openjdk.java.net/browse/JDK-8064786
    
    Thanks,
    -- 
    Andrew :)
    
    Senior Free Java Software Engineer
    Red Hat, Inc. (http://www.redhat.com)
    
    PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
    Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222
    https://keybase.io/gnu_andrew