[Stackless] 'Stackliss' module

[Bob Ippolito]

Greenlets have two rather large caveats: no cyclic GC, and proper  
finalization is hard.  You can work around these things, but if you  
have a leak it's hard to track down.  Especially because they have  
very little debugger support (in either gdb or pdb).

-bob

On May 2, 2006, at 8:01 AM, Josh Ritter wrote:

>
> Greenlets are hard to discount as they are implemented in a 24k c- 
> extension
> module that works with standard CPython.  This is a pretty nice  
> feature and
> easy to maintain.  I definitely need to at least investigate them.
>
> Anyone have a greenlet horror story?   How about omens and warnings?
> Perhaps a little Stackless evangelism to share?
>
> -Josh Ritter
> Prairie Games, Inc
> http:.//www.prairiegames.com
>
>
>
>
> ----- Original Message -----
> From: "Richard" <richard at ccpgames.com>
> To: <stackless at stackless.com>
> Sent: Tuesday, May 02, 2006 9:31 AM
> Subject: Re: [Stackless] 'Stackliss' module
>
>
>>> From: Josh Ritter on 02 May 2006 14:05
>>>
>>> Is this the latest version of stackliss?  If not, where is it
>>> hosted?  Did
>>> you come up with a way to implement the watchdog functionality?
>>>
>>> http://www.stackless.com/pipermail/stackless/2005-May/001444.html
>>
>> If it is the latest post to the list about it, then it is the latest
>> version.  It is not hosted anywhere, I was trying to get it added to
>> the greenlets package in codespeak.net, but Armin Rigo never added  
>> it.
>>
>>> How does stackliss perform?  A ballpark guesstimate of
>>> abysmal to wonderful
>>> is acceptable :)
>>
>> No idea, sorry :)  I only wrote it because doing so took my fancy at
>> some stage, I think I became disillusioned with greenlets because I
>> thought that there was a reason Stackless did what it did, and if I
>> wanted it done right, I should just use Stackless.
>>
>> And I doubt anyone ever used it.  It has some bugs that I have become
>> aware of, since I have learnt more about how Stackless actually  
>> works.
>> Nothing serious, but.. they should be fixed to bring it into
>> conformity with Stackless.
>>
>> Richard.
>>
>> _______________________________________________
>> Stackless mailing list
>> Stackless at stackless.com
>> http://www.stackless.com/mailman/listinfo/stackless
>
>
> _______________________________________________
> Stackless mailing list
> Stackless at stackless.com
> http://www.stackless.com/mailman/listinfo/stackless


_______________________________________________
Stackless mailing list
Stackless at stackless.com
http://www.stackless.com/mailman/listinfo/stackless