Re: [infinispan-dev] possible bug in TreeCache.put(Fqn fqn, Map<? extends K, ? extends V> data) when passed an empty map

> > Are you referring to not locking a node when adding or removing a child node? This comes from a conversation we had online where I was checking whether there was a real need for TreeCache in this use case.

IIRC, parents used to be locked for writing when adding/removing children. Looking at the Infinispan code, this might not be the case any more: // don't lock parents for child insert/removes! icc.getInvocationContext().setFlags(Flag.SKIP_LOCKING); parentStructure.put(fqn.getLastElement(), fqn); I see we don't have anything in http://community.jboss.org/wiki/Infinispan wrt the tree module. I think it'd be interesting to explain what the lock semantics are in the tree module and how different these are with JBoss Cache. Thoughts?