I've uploaded the new versions of the packages to The Inbox, and moved the 
old ones to The Treated Inbox.
The main difference is that the Semaphore class is not modified, and its 
instances are handled like any other object. There's also a new #cleanUp: 
method which will allow freeing up some memory. And the caches are 
shrinked whenever the external objects are wiped.
Since Semaphores don't know their indexes anymore, the new part of the API 
has also changed.
The following snippet will load the new versions to an up-to-date Trunk 
image:

Installer squeakInbox
 	package: 'System-ul.687';
 	package: 'Sound-ul.40';
 	package: 'Kernel-ul.883';
 	package: 'Network-ul.154';
 	package: 'Files-ul.140';
 	install


Levente

On Sat, 25 Oct 2014, David T. Lewis wrote:

> On Sat, Oct 25, 2014 at 10:02:22PM +0200, Levente Uzonyi wrote:
>> On Sat, 25 Oct 2014, David T. Lewis wrote:
>>
>>> On Fri, Oct 24, 2014 at 09:52:11PM +0200, Levente Uzonyi wrote:
>>>> On Fri, 24 Oct 2014, Bert Freudenberg wrote:
>>>>
>>>>> Another thought: is it really necessary to cache the index in the
>>>>> semaphore? Wouldn't scanning the externalObjectsTable for self be
>>>>> efficient enough? It's pretty small, after all.
>>>>
>>>> That's exactly what the current implementation does (scans the table), and
>>>> this is what I want to avoid, because it doesn't scale. The table is small
>>>> when you start your image, but it can easily grow large - especially on a
>>>> server (3 semaphores per socket) - and it never gets shrinked. So once it
>>>> grows, it'll stay slow "forever".
>>>>
>>>> And no, it's not absolutely necessary to store the index in the
>>>> semaphores. The new implementation uses an IdentityDictionary to store the
>>>> indexes of non-semaphores. The same method could work for semaphores too.
>>>> But it's convenient to store the index in the semaphore, otherwise the
>>>> method which creates and registers a new semaphore has to return two
>>>> values - the index and the semaphore - which is cumbersome.
>>>> It's also faster to ask for a new Semaphore from the
>>>> ExternalSemaphoreTable, because that way it knows that the
>>>> Semaphore is not registered.
>>>>
>>>
>>> I don't see the indexesByObjects dictionary actually being updated, except
>>> in the case of setting a new externalObjectsArray in the
>>> ExternalSemaphoreTable.
>>> I think I am overlooking something?
>>
>> It's updated in #safelyRegisterExternalObject: and
>> #safelyUnregisterExternalObject:, but - as the class comment states - it's
>> only used for non-Semaphores.
>>
>> Based on the feedback there's too much resistance to add an instance
>> variable to Semaphore, so I'm about to remove that part from the
>> implementation.
>
> I don't think the instance variable is horrible, but I do think it would
> feel cleaner if you can do without it, so +1 if you are able to do that.
>
> Thanks
> Dave
>
>
>