[concurrency-interest] AtomicReference and java.lang.ref packagedesign thoughts

DT dt at flyingtroika.com
Mon Sep 1 16:30:30 EDT 2014

thanks, so AtomicReference does not have anything common with a strong 
reference , is it correct? in the sense how strong reference is implemented.

On 9/1/2014 1:29 AM, David Holmes wrote:
> Hi DT,
> AtomicReference is an atomic class that operates on object references as
> opposed to one of the primitive types.
> The java.lang.ref package concerns types that manage references to other
> objects for the purpose of interacting with the garbage collector.
> AtomicReference has nothing to do with interacting with the garbage
> collector.
> Cheers,
> David Holmes
>> -----Original Message-----
>> From: concurrency-interest-bounces at cs.oswego.edu
>> [mailto:concurrency-interest-bounces at cs.oswego.edu]On Behalf Of DT
>> Sent: Monday, 1 September 2014 2:48 PM
>> To: concurrency-interest
>> Subject: [concurrency-interest] AtomicReference and java.lang.ref
>> packagedesign thoughts
>> Are any plans to move AtomicReference to the java.lang.ref package? Just
>> interesting to understand what was the idea behind the AtomicReference
>> implementation. AtomicReference is used for concurrency reason and its
>> some sort of a strong reference. So it makes sense to put it under
>> concurrency package but at the same time it represents a reference
>> object and the construct for Atomicreference is the same as any other
>> references - AtomicReference aRef_ = new AtomicReference() , or
>> SoftReference sRef_ = new SoftReference();
>> The ref package has multiple reference implementations such as
>> Reference, PhantomReference, SoftReference, WeakReference  - and all of
>> them are managed by ReferenceQueue in the direct or abstract way, though
>> AtomicReference does not belong to  ref package and is not extended from
>> Reference class. My understanding is that its not meant to be used in
>> any way with a ReferenceQueue for any type of weak cache implementations
>> either. So should it be under the ref package or under concurrency
>> package if we look at this from high level?
>> Seems to me there is some misconception about references which makes it
>> hard to understand in terms of the memory and gc behaviors.
>> Thanks,
>> DT
>> _______________________________________________
>> Concurrency-interest mailing list
>> Concurrency-interest at cs.oswego.edu
>> http://cs.oswego.edu/mailman/listinfo/concurrency-interest

More information about the Concurrency-interest mailing list