Account Options

  1. Sign in
The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Google Groups Home
« Groups Home
comp.lang.lisp
+ new post
About this group
Subscribe to this group
This is a Usenet group - learn more
Message from discussion off-topic: open challenge to Christian Lynbeck
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 
From:
To:
Cc:
Followup To:
Add Cc | Add Followup-to | Edit Subject
Subject:
Validation:
For verification purposes please type the characters you see in the picture below or the numbers you hear by clicking the accessibility icon. Listen and type the numbers you hear
 
rif  
View profile  
 More options Mar 8 2004, 12:52 pm
Newsgroups: comp.lang.lisp
From: rif <r...@mit.edu>
Date: 08 Mar 2004 12:52:37 -0500
Local: Mon, Mar 8 2004 12:52 pm
Subject: Re: off-topic: open challenge to Christian Lynbeck
Print | View thread | Show original | Report this message | Find messages by this author

Raymond Toy <r...@earthlink.net> writes:
> rif wrote:
> >  generation.  It's worthwhile to know the specifics of your
> > implementation (for instance, it seems that a substantial amount of
> > the last-mile speed up for this benchmark on CMUCL came from figuring
> > out ftruncate could be fast inside a macro, but not inside an inline
> > function).

> I don't believe that is true.  I think if you take Brian's macro and
> made it an inline function, you will find that ftruncate will still be
> inlined in all places where that function is used.

> Well, at least that was true for me on the particular version of cmucl
> that I used.

> Ray

Well, I'm not sure it's about ftruncate per se.  Maybe it's about
maximum inlining depth, or something else.  If I take Brian's CL
version, and I replace his macro with an inline function, I get a
float to pointer coercion on the return value of the inline function,
and the program slows way down.  This is using the 2003-09-10 build of
CMUCL.

rif


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2013 Google