On Nov 22, 2005, at 1:40, Matt Fowles wrote:
Call Frame Access
Chip began to pontificate about how one should access call frames.
Chip
suggested using a PMC, but Leo thought that would be too slow.
No, not really. It'll be slower, yes. But my argument was: whenever you
start introspecting a call frame, by almost whatever means, this will
keep the call frame alive[1] (see Continuation or Closure). That is:
timely destruction doesn't work for example and the introspection
feature is adding another level of complexity that isn't needed per se,
because 2 other solutions are already there (or at least implemented
mostly).
leo
[1] a call frame PMC could be stored elsewhere and reused later,
refering to then dead contents. Autrijus mentioned that this will need
weak references to work properly.