[RFC][plugin]: fix fixme

[Dr Andrew John Hughes]

On 10:46 Wed 30 Mar     , Deepak Bhole wrote:

snip...

> Why not synchronize the whole function? The whole body is synchronized
> on a static variable anyway. I think the multi-threaded map access is a
> genuine potential problem. Making the reference and unreference
> functions synchronized and keeping everything else as is should be
> enough to address that.
> 

They aren't the same thing.  Making the methods synchronised is equivalent
to wrapping the body in synchronized (this) and wouldn't protect against
access to the maps from another instance.

I don't see the immediate reason for three maps, but I don't know the rest
of this code that well.  Can you explain how 'multi-threaded map access is
a genuine potential problem'?  nextID is not going to take a long amount of
time so keeping objects locked over that period doesn't seem an issue.

> Cheers,
> Deepak
> 

-- 
Andrew :)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Support Free Java!
Contribute to GNU Classpath and IcedTea
http://www.gnu.org/software/classpath
http://icedtea.classpath.org
PGP Key: F5862A37 (https://keys.indymedia.org/)
Fingerprint = EA30 D855 D50F 90CD F54D  0698 0713 C3ED F586 2A37