ISPN-1881 Refactor CommandAwareRpcDispatcher to treat calls to single re...

[maniksurtani]

...cipients in a more optimised manner

• Also make SuccessfulResponse a singleton for null payloads

See https://issues.jboss.org/browse/ISPN-1881

Also pls merge t_1881_5 for branch 5.1.x

[Sanne]

you shouldn't remove if (trace) as checking the static constant will always be better than actually invoking the tracef method

[maniksurtani]

Whoops - I'm not sure why I removed that. :/

On 27 February 2012 11:11, Sanne Grinovero <
reply@reply.github.com

wrote:

@@ -410,7 +462,7 @@ public synchronized void futureDone(Future
objectFuture) {
if (sc.processed) {
// This can happen - it is a race condition in JGroups'
NotifyingFuture.setListener() where a listener
// could be notified twice.

•        if (trace) log.tracef("Not processing callback; already
  

  processed callback for sender %s", sc.address);

•        log.tracef("Not processing callback; already processed
  

  callback for sender %s", sc.address);

you shouldn't remove if (trace) as checking the static constant will
always be better than actually invoking the tracef method

---

Reply to this email directly or view it on GitHub:
https://github.com/infinispan/infinispan/pull/974/files#r490659

Manik

Sent from my iPad 4 beta 3 with a telekinetic keyboard