New finalization support code is in the Trunk now (was: Re: [squeak-dev] Re: Multiple finalizers per single object)

Igor Stasenko siguctua at gmail.com
Sun Sep 26 02:42:08 UTC 2010


Great news! Thanks Levente.
I will test if everything ok on my VM(s), since i have Squeak&Cog with
new finalization both :)

On 26 September 2010 04:29, Levente Uzonyi <leves at elte.hu> wrote:
> On Fri, 24 Sep 2010, Igor Stasenko wrote:
>
>> 2010/9/24 Levente Uzonyi <leves at elte.hu>:
>
> snip
>
>>>
>>> Today or tomorrow I'll check the code more in-depth and prepare mczs from
>>> your changesets (if noone else does it before me).
>>>
>> Please, do :)
>
> It's in the Trunk now. I did the following changes:
> - there was a bug in the evaluated code in WeakFinalizationRegistry class >>
> #migrateOldRegistries, the first argument of forgetClass:logged: was a
> Symbol instead of a Class, so WeakFinalizationRegistry was still around.
> - WeakRegistry >> #keys is now protected
> - I merged the old comment of WeakRegistry with the new one
> = I merged phase-0 with phase-1
> - I omitted phase-3, because I added the migration code to phase-2
> - I omitted phase-4 because those tests are already in the Trunk
>
> What I didn't do so far, is that I didn't build a VM with the new
> finalization support, so someone should check if it works as expected.
>
>
> Levente
>
> P.S.: Seems like the MC postscript is evaluated before the class
> initializers which is a bit surprising. This is why I couldn't merge phase-2
> into phase-1.
>
>



-- 
Best regards,
Igor Stasenko AKA sig.



More information about the Squeak-dev mailing list