home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!ukma!darwin.sura.net!Sirius.dfn.de!math.fu-berlin.de!unidus.rz.uni-duesseldorf.de!convex.rz.uni-duesseldorf.de!wolfgang
- From: wolfgang@convex.rz.uni-duesseldorf.de (Wolfgang Mueller)
- Subject: Re: Jumptable Performance (Was: Re: shared libs - can everyone be happy with this?)
- Message-ID: <wolfgang.714233807@convex.rz.uni-duesseldorf.de>
- Sender: news@unidus.rz.uni-duesseldorf.de
- Organization: Heinrich-Heine-Universitaet Duesseldorf
- References: <1992Aug17.144719.1961@crd.ge.com> <1992Aug17.151311.29507@ods.com> <NOP.92Aug17135014@theory.Mankato.MSUS.EDU> <1992Aug18.080437.3944@fys.ruu.nl> <1992Aug18.140858.3484@crd.ge.com> <1992Aug18.154149.26416@fys.ruu.nl> <1992Aug19.125541.865@crd.ge.com>
- Date: Wed, 19 Aug 1992 14:16:47 GMT
- Lines: 12
-
- In <1992Aug19.125541.865@crd.ge.com> davidsen@ariel.crd.GE.COM (william E Davidsen) writes:
- > Like most things which seem too good to be true, I'm suspicious. Does
- >anyone have an explanation why adding size and instructions to every
- >library call would make the program use less user CPU (or appear to)?
- >Having used jump tables before I have to be suspicious.
-
- Might it be that with the first jump already invalidating the lookahead cache
- the second jump does not do any more harm, but instead leaves the memory
- interface in a condition more suitable to the following register savings ?
-
- Just a guess,
- Wolfgang R. Mueller <dvs@ze8.rz.uni.duesseldorf.de> .
-