Re: Blocking in WeakIdentityMap

Patrick Wright wrote:
> We found a discussion in the mailing lists related to WIM from last
> summer but it was inconclusive.
> http://archives.java.sun.com/cgi-bin/wa?A2=ind0607&L=jini-users&D=0&P=10153
>
> Thoughts?

...
>
> "http-8080-Processor12" daemon prio=10 tid=0xacf75400 nid=0x5605
> waiting for monitor entry [0xaca83000..0xaca84030]
>   java.lang.Thread.State: BLOCKED (on object monitor)
>        at net.jini.security.policy.DynamicPolicyProvider.getDomainPermissions(DynamicPolicyProvider.java:259)
>        - waiting to lock <0xae6ec030> (a com.sun.jini.collection.WeakIdentityMap)
>        at net.jini.security.policy.DynamicPolicyProvider.implies(DynamicPolicyProvider.java:170)
>        at com.sun.jini.start.AggregatePolicyProvider.implies(AggregatePolicyProvider.java:201)
>        at java.security.ProtectionDomain.implies(ProtectionDomain.java:213)
>        at java.lang.Thread.run(Thread.java:619)

Hi Patrick!  This trace is a different issue than the issue that I was
investigating at that time.  The issue shown in this trace is another contention
point for mulitple threads running in a secure Java/Jini environment.  I've been
meaning to go look into this issue as well, because I do see this as a point of
contention in my use of Jini as well.

I don't have much to say about what the real issue is, other than the fact that
a large lock is in place with the use of the WeakIdentityMap as a sync for all
of the domains' permissions.

Gregg Wonderly

--------------------------------------------------------------------------
Getting Started:     http://www.jini.org/wiki/Category:Getting_Started
Community Web Site:  http://jini.org
jini-users Archive:  http://archives.java.sun.com/archives/jini-users.html
Unsubscribing:       email "signoff JINI-USERS"  to [email protected]