On 12/21/2015 10:33 PM, Paul Eggert wrote:
> Daniel Colascione wrote:
>> Most programs you use don't bother trying to detect or recover from sack
>> overflow. Why should they? C stack overflow is a programming error.
> 
> Emacs is different because it's programmable, and it must respond
> robustly to programming errors.

It's programmable in Lisp. Lisp stack overflows shouldn't kill Emacs.
I'm suggesting that we shouldn't care about *C* stack overflows.

>> The current stack overflow scheme can't recover reliably
>> because it doesn't fix maintain invariants.
> 
> It works well enough now for people to get work done. Introducing
> modules should not break this.

Modules can just dereference NULL or call abort. Writing a module is
more dangerous than writing a bit of elisp.