Re: [spr29787] Closure ... will be stack allocated
Edi Weitz <[email protected]> Fri, 14 Jan 2005 19:43:48 +0100
| Newsgroups | gmane.lisp.allegro |
|---|---|
| Message-ID | <[email protected]> |
Hi Charley! On Fri, 14 Jan 2005 07:22:29 -0800, "Charles A. Cox" <[email protected]> wrote: > It is an optimization note. In general, closures are > indefinite-extent objects and need to be allocated on the heap. > However, if the compiler can determine, or if a user explicitly > states, that a closure has dynamic-extent, then allocating the > closure space on the stack has better run-time efficiency. Thanks for the info, I suspected something like that but I wanted to be sure. > I believe the reason the note is being displayed is for those > programmers explicitly trying to achieve stack-consed closures. > Without this message, it's not immediately obvious whether the > compiler actually performed the desired optimization. Hmm, I'm pretty sure I didn't explicitely ask for that in my code... Thanks, Edi.