Re: When/why did function calls get cheap?

Peter Finderup Lund <firefly@diku.dk>
21 Feb 2003 00:48:22 -0500

          From comp.compilers
Related articles
When/why did function calls get cheap? peter@javamonkey.com (Peter Seibel) (2003-02-12)
Re: When/why did function calls get cheap? strohm@airmail.net (John R. Strohm) (2003-02-13)
Re: When/why did function calls get cheap? gah@ugcs.caltech.edu (Glen Herrmannsfeldt) (2003-02-13)
Re: When/why did function calls get cheap? bje@redhat.com (Ben Elliston) (2003-02-21)
Re: When/why did function calls get cheap? joachim_d@gmx.de (Joachim Durchholz) (2003-02-21)
Re: When/why did function calls get cheap? marcov@toad.stack.nl (Marco van de Voort) (2003-02-21)
Re: When/why did function calls get cheap? firefly@diku.dk (Peter Finderup Lund) (2003-02-21)
Re: When/why did function calls get cheap? firefly@diku.dk (Peter Finderup Lund) (2003-02-21)
Re: When/why did function calls get cheap? anton@mips.complang.tuwien.ac.at (2003-02-21)
Re: When/why did function calls get cheap? jplevyak@yahoo.com (John Plevyak) (2003-02-21)
Re: When/why did function calls get cheap? {spamtrap}@qeng-ho.org (Arthur Chance) (2003-02-24)
Re: When/why did function calls get cheap? gah@ugcs.caltech.edu (Glen Herrmannsfeldt) (2003-02-24)
Re: When/why did function calls get cheap? alexc@std.com (Alex Colvin) (2003-02-24)
Re: When/why did function calls get cheap? vbdis@aol.com (2003-02-24)
[9 later articles]
| List of all articles for this month |
From: Peter Finderup Lund <firefly@diku.dk>
Newsgroups: comp.compilers
Date: 21 Feb 2003 00:48:22 -0500
Organization: Department of Computer Science, University of Copenhagen
References: 03-02-073 03-02-085
Keywords: architecture, performance
Posted-Date: 21 Feb 2003 00:48:21 EST

On 13 Feb 2003, John R. Strohm wrote:


> Function calls aren't that expensive. They are becoming more
> expensive nowadays, because of the cost of cache misses on processors
> with horrendously long pipelines and superscalar execution and all
> that stuff.


I'm not sure that is correct. Many CPUs have hidden internal return
stacks coupled to their branch prediction so function calls/returns most
often don't require stalls/pipeline flushes. For trace caches, the
calls/returns could possibly be squashed completely out of the stored
trace.


-Peter

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.