harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nathan Beyer" <nbe...@kc.rr.com>
Subject RE: [classlib] java.util.concurrency
Date Wed, 31 May 2006 01:58:27 GMT
Here's an early analysis using the code at Doug's site.

Assumptions:
* Use the code tagged JSR166_PFD [1], as this seems to be the Java 5 spec
version. The HEAD has some new code that depends on Java 6 APIs, like Deque.
* Only use and look at the code that contains a header declaring a public
domain license.

Results:
* java.util.concurrent - Only two or three classes don't compile. The two
missing dependencies are System.nanoTime and java.util.PriorityQueue, which
we'd have to write. Note that the java.util.AbstractQueue is in this
repository as public domain, so that should help.
* java.util.concurrent.atomic - None of these classes compile, as they are
dependent on a VM specific atomic locking mechanism that Harmony would have
to provide.
* java.util.concurrent.locks - Only two classes don't compile, as they are
also dependent on some VM specific locking mechanisms.

Thoughts:
* System.nanoTime() could be stubbed out and possibly temporarily
implemented as a pass through to System.currentTimeMillis().
* Once a java.util.PriorityQueue implementation is completed, a "concurrent"
module could be seeded with the java.util.concurrent package and this would
provide a good start.
* Layout a VM Java API for the necessary locking methods, which could be
part of luni-kernel or a separate concurrent-kernel. Someone with a bit more
VMI hacking would probably have some better thought here, but I thought I
viable patch might be to provide a simple Java implementation of this
locking that uses traditional Java locking. Obviously this would slightly
defeat the point of the atomic locks, but it could provide a functional test
basis.

Please reply with any comments.

-Nathan

[1]
http://gee.cs.oswego.edu/cgi-bin/viewcvs.cgi/jsr166/?only_with_tag=JSR166_PF
D

> -----Original Message-----
> From: Geir Magnusson Jr [mailto:geir@pobox.com]
> Sent: Tuesday, May 30, 2006 6:49 PM
> To: harmony-dev@incubator.apache.org
> Subject: Re: [classlib] java.util.concurrency
> 
> 
> 
> Nathan Beyer wrote:
> > I was browsing Doug's Concurrency interest page [1] and noticed that the
> > code is available in a CVS repository and all of the j.u.c class are
> > copyrighted with a Creative Commons public domain license [2].
> 
> Yes, that's been Doug's claim and my motivation for using the code from
> there.  We'd have to examine it from the POV of the ACQ for any issues
> like that, but I wouldn't imagine we'd have any large problems.
> 
> > The TCK
> > source seems to be there as well. Assuming this is an acceptable license
> by
> > Apache terms, it would seem we could use this code.
> 
> Assuming we can take the code, we could incorporate that TCK code into
> our normal test suite.
> 
> >
> > Additionally, there is a pre-built JAR [3], which could possible be used
> for
> > development purposes if dropped into the bootclasspath.
> 
> Give it a whirl :)  let us know :)
> 
> geir
> 
> >
> > Note, some of the code seems to include maintenance updates that are
> above
> > the Java 5 RI.
> >
> > -Nathan
> >
> > [1] http://gee.cs.oswego.edu/dl/concurrency-interest/
> > [2] http://creativecommons.org/licenses/publicdomain
> > [3] http://gee.cs.oswego.edu/dl/jsr166/dist/jsr166.jar
> >
> >> -----Original Message-----
> >> From: Geir Magnusson Jr [mailto:geir@pobox.com]
> >> Sent: Tuesday, May 30, 2006 9:29 AM
> >> To: harmony-dev@incubator.apache.org
> >> Cc: Doug Lea
> >> Subject: [classlib] java.util.concurrency
> >>
> >> I'm cc-ing Doug on this because I'm certain he doesn't read our list
> >> anymore.
> >>
> >> Doug, can you opine regarding the ability for this project to use your
> >> code for our implementation of j.u.c, keeping in mind our rules
> ensuring
> >> that we don't use any copyrighted material or other IP for which we
> >> don't have permission?
> >>
> >> geir
> >>
> >> ---------------------------------------------------------------------
> >> Terms of use : http://incubator.apache.org/harmony/mailing.html
> >> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> >> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >
> >
> > ---------------------------------------------------------------------
> > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >
> >
> 
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org


---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org


Mime
View raw message