REPOST: Re: affects on extended modules

[Curtis Jensen]

Kragen Sitaker <kragen at pobox.com> wrote in message news:<83pu5zrvbl.fsf at panacea.canonical.org>...
> Curtis Jensen <cjensen at bioeng.ucsd.edu> writes:
> > We have created a python interface to some core libraries of our own
> > making.  We also have a C interface to these same libraries.  However,
> > the the python interface seems to affect the speed of the extended
> > libraries.  ie.  some library routines have their own benchmark code,
> > and the time of exection from the start of the library routine to the
> > end of the library routine (not including any python code execution),
> > takes longer than it's C counterpart.
> 
> In the Python version, the code is in a Python extension module,
> right?  A .so or .dll file?  Is it also in the C counterpart?  (If
> that's not it, can you provide more details on how you compiled and
> linked the two?)
> 
> In general, referring to dynamically loaded things through symbols ---
> even from within the same file --- tends to be slower than referring
> to things that aren't dynamically loaded.
> 
> What architecture are you on?  If you're on the x86, maybe Numeric is
> being stupid and allocating things that aren't maximally aligned.  But
> you'd probably notice a pretty drastic difference in that case.
> 
> ... or maybe Numeric is being stupid and allocating things in a way
> that causes cache-line contention.
> 
> Hope this helps.


I have been unable to find any references to dynamically loaded
objects being slower than statically linked objects.  I understand
that there is an initial cost when the library is loaded that wouldn't
be there in the static case, though, I don't know why it would be
slower after that.  Can you please provide some references for this,
or explain why this would be?  Thanks.

--
Curtis Jensen

========= WAS CANCELLED BY =======:
Path: news.sol.net!spool1-milwwi.newsops.execpc.com!newsfeeds.sol.net!news-out.visi.com!hermes.visi.com!newspeer.monmouth.com!news-xfer.nuri.net!news-hub.kaist.ac.kr!feeder.kornet.net!news1.kornet.net!ua4canc3ll3r
From: cjensen at bioeng.ucsd.edu (Curtis Jensen)
Newsgroups: comp.lang.python
Subject: cmsg cancel <1be8d8b.0112271702.69a599a at posting.google.com>
Control: cancel <1be8d8b.0112271702.69a599a at posting.google.com>
Date: Mon, 31 Dec 2001 03:54:32 GMT
Organization: A poorly-installed InterNetNews site
Lines: 2
Message-ID: <cancel.1be8d8b.0112271702.69a599a at posting.google.com>
NNTP-Posting-Host: 211.57.49.2
X-Trace: news2.kornet.net 1009775464 27193 211.57.49.2 (31 Dec 2001 05:11:04 GMT)
X-Complaints-To: usenet at news2.kornet.net
NNTP-Posting-Date: Mon, 31 Dec 2001 05:11:04 +0000 (UTC)
X-No-Archive: yes
X-Unac4ncel: yes
X-Commentary: I love NewsAgent 1.10 and the Sandblaster Cancel Engine Build 74 (19 March 1999)

This message was cancelled from within Mozilla.