Yet another strike against the current AA implementation

[Andrei Alexandrescu]

Georg Wrede wrote:
>>> I can't believe you brought it up as anything else but jesting (maybe 
>>> you are and I'm not getting it).
> 
> Having looked into things, it turns out I'm the one that now suspects 
> you of jesting.

What I referred to was inferring a thread's actual stack requirements 
from one trace of its run. That's just untenable.

>> Maybe I'll have to dig up prior art, or something. Clearly I'm not 
>> qualified to properly defend the validity of this idea.
> 
> There is prior art, by the truckload. A /very/ short slide show here: 
> http://zope.stackless.com/StacklessEuroPy.ppt
> 
> I'd kill to get that in D.

Interesting, just you can't claim you were talking about that all along. 
I mean, come on! You say one thing that was untenable, and now you come 
up with a different one that is. And you were right all along?

> And about your comments on stack size, seems regular Python has an 
> in-built limit on recursion, at 1000 deep. That should be diametrically 
> opposite your stance on stack size.

Python has apparently set a maximum to have a guarantee it won't crash 
when given a certain minimum stack size. That's nice, but I'm not quite 
seeing how that's opposite to what I said in this discussion.


Andrei