What's the name of the problem that relates to optimizing closures on a stack-based system?

Question

I remember hearing about a general optimization problem that relates to function closures, stating that in general it's difficult to optimize the creation of a closure using only stack-based memory management. Do any of you remember the name of this optimization problem, possibly with an example or link to relevant page?

Answer 1

It sounds like you're thinking of the upward funarg problem.

Answer 2

Perhaps you're thinking of escape analysis.

It concerns the distinction between what the Lisp community calls its two kinds of extent: dynamic extent and indefinite extent. Objects of the former can be stack-allocated, while the latter cannot, as their lifetime likely exceeds the scope of their allocation.