Summary of Lightweight finalization for the JDK without any drawbacks

[Jaroslav Tulach]

Hi.
I'd like to thank everyone in participating in our discussion about lightweight 
finalization and shaping and sharping my understanding of the problem. I believe the 
discussion has now slowed down and it is time for another summary. I have added one 
into the issue (while also changing its subject):

https://bugs.openjdk.java.net/browse/JDK-8051843?focusedCommentId=13539596&p
age=com.atlassian.jira.plugin.system.issuetabpanels:comment-
tabpanel#comment-13539596[1] 

I believe our understanding of the problem has been deepen enough to focus on 
creation of the actual public API for lightweight finalization. It will be my task for up-
coming week to start necessary steps to propose such API.

-jt

PS: I'd like to especially thank to Andrew for pointing out the "this" reference problem 
and Doug for commenting it. I've reported 
https://bugs.openjdk.java.net/browse/JDK-8055183[2] to resolve it somehow. I'll be 
watching from distance eager to know the result.


Dne St 13. srpna 2014 13:12:43, Jaroslav Tulach napsal(a):
> Dne So 9. srpna 2014 07:18:54, Doug Lea napsal(a):
> > > Dne Pá 8. srpna 2014 15:02:41, Stanimir Simeonoff napsal(a):
> > > 
> > > 
> > > According to articles provided by Andrew[1], toString() can be on stack,
> > > even if obj
> > > (e.g. this for toString()) is garbage collected. This is a bit
> > > surprising
> > > (an OOP fan
> > > would expect that this is GC root while its instance method is running),
> > > but such
> > > behavior is said to be in accordance to the 1.8 spec.
> > > 
> > > Question for Doug (as a member of jmm-dev list) then is: would not it be
> > > wise to
> > > change the spec to require this to be GC root while its instance method
> > > is
> > > being
> > > executed?
> > 
> > Possible changes are being discussed. It is not clear yet
> > exactly what rules would better match people's expectations
> > while remaining implementable by JVMs, and/or whether they
> > would force performance degradations even
> > for the vast majority of code not using finalizers. It is
> > very possible the end result will just be to clarify rules and
> > provide mechanisms allowing better programmer control.
> > 
> > Any changes would go into effect for JDK9 at the earliest,
> > so if you can find a way to avoid use of finalizers
> > here, it would be a better option.
> 
> I'll do my best to come up with a lightweight alternative to finalization.
> However:
> 
> As far as I understand Andrew's inquiry, the problem is not (that much)
> related to finalizers, rather to question whether following method can ever
> finish or not:
> 
>     private void gcThis() {
>         Reference<?> ref = new WeakReference<>(this);
>         while (ref.get() != null) {
>             System.gc();
>         }
>         System.err.println("ref is gone: " + ref);
>     }
> 
> On Oracle's JDK8 the method has never finished for me. However when I tried
> IBM's JDK (identification can be found below) the method finished every
> time! A bit of consistency would be beneficial in this area, imho.
> 
> -jt
> 
> PS: While I understand why the two JDKs behave the way they behave, I don't
> think this is explainable to regular Java developers. But maybe they don't
> care that much and one day it will be easy to google out solution on
> stackoverflow to prevent "this" being garbage collected by changing the last
> method line to
> 
>         System.err.println("ref is gone: " + ref + " for " + this);
> 
> The version of IBM's Java was:
> 
> ~/bin/ibmjava7$ ./bin/java -version
> java version "1.7.0"
> Java(TM) SE Runtime Environment (build pxa6470_27sr1fp1-20140708_01(SR1
> FP1)) IBM J9 VM (build 2.7, JRE 1.7.0 Linux amd64-64 Compressed References
> 20140707_205525 (JIT enabled, AOT enabled)
> J9VM - R27_Java727_SR1_20140707_1408_B205525
> JIT  - tr.r13.java_20140410_61421.07
> GC   - R27_Java727_SR1_20140707_1408_B205525_CMPRSS
> J9CL - 20140707_205525)
> JCL - 20140707_01 based on Oracle 7u65-b16


--------
[1] 
https://bugs.openjdk.java.net/browse/JDK-8051843?focusedCommentId=13539596&p
age=com.atlassian.jira.plugin.system.issuetabpanels:comment-
tabpanel#comment-13539596
[2] https://bugs.openjdk.java.net/browse/JDK-8055183