[1581] in java-interest
Finalization and GC
daemon@ATHENA.MIT.EDU (Christian Fabre)
Fri Sep 8 12:03:07 1995
To: java-interest@java.Eng.Sun.COM
Date: Fri, 08 Sep 1995 15:00:45 +0200
From: Christian Fabre <fabre@gr.osf.org>
Hi everybody,
[Sorry if this issue has been raised before.]
In "The Java Language Specification" dated of May 1995, on page 15
there is a footnote that reads:
If a finalized object references another finalized object, the
objects are finalized in the reverse order of their creation.
For the run-time, the easiest order of dealocation of objects seems to
depend very much on GC algorithm. Forcing a given order might put too
much burden on some GC algorithms. Also, what is the "order" of
creation on a SMP machine?
Maybe the specs be more neutral about that and say that the order of
dealocation is "platform dependent".
Christian.
=====
Christian Fabre
OSF Research Institute Net: fabre@gr.osf.org
2 avenue de Vignate Tel: +33 76.63.48.90
38610 Gieres - France Fax: +33 76.51.05.32
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
