Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!convex!central!newstop!exodus!appserv!slovax.Eng.Sun.COM!lm
From: lm@slovax.Eng.Sun.COM (Larry McVoy)
Newsgroups: comp.unix.wizards
Subject: Re: Shared libraries (was Re: Window system bashing (was Re: X11 bashing))
Message-ID: <572@appserv.Eng.Sun.COM>
Date: 8 May 91 05:45:06 GMT
References: <BZS.91May4202942@world.std.com>
Sender: news@appserv.Eng.Sun.COM
Organization: Sun Microsystems, Mt. View, CA.
Lines: 20

bzs@world.std.com (Barry Shein) writes:
> >Dynamically linked shared libraries (a la Sun) will incur a performance hit,
> >but it is completely misleading to talk about the hit in percentage terms.
> 
> Have you actually made any attempt to measure this effect? Or are you
> just drawing conclusions from a few facts you have lying around?

I have (made an attempt to measure it).  I measured:

	main() { exit(0); }

statically and dynamically on an SS1 (20 MHZ sparc).  I found that
dynamically linked version cost about 20milliseconds more than
statically linked version.

I do not know (not having measured it and not really groking how
dynamic linking works) if each libc reference costs 20milliseconds on
first reference.   I hope not.
---
Larry McVoy, Sun Microsystems     (415) 336-7627       ...!sun!lm or lm@sun.com