From owner-freebsd-current Wed Sep 16 00:26:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA21300 for freebsd-current-outgoing; Wed, 16 Sep 1998 00:26:43 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from word.smith.net.au (castles205.castles.com [208.214.165.205]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA21241; Wed, 16 Sep 1998 00:26:29 -0700 (PDT) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost [127.0.0.1]) by word.smith.net.au (8.9.1/8.8.8) with ESMTP id AAA00684; Wed, 16 Sep 1998 00:24:21 -0700 (PDT) (envelope-from mike@word.smith.net.au) Message-Id: <199809160724.AAA00684@word.smith.net.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Poul-Henning Kamp cc: Mike Smith , joelh@gnu.org, Terry Lambert , tom@uniserve.com, gpalmer@FreeBSD.ORG, irc@cooltime.simplenet.com, freebsd-current@FreeBSD.ORG Subject: Re: Download of FreeBSD 3.0-SNAP In-reply-to: Your message of "Wed, 16 Sep 1998 09:04:31 +0200." <3217.905929471@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 16 Sep 1998 00:24:19 -0700 From: Mike Smith Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > In message <199809152258.PAA01641@dingo.cdrom.com>, Mike Smith writes: > > >> Without knowing the true geometry of the disk (which I assume EIDE > >> doesn't allow), we can't optimize for the cylinder patterns anyway, so > >> I suppose it's a moot issue. I suppose that by manually specifying > >> the geometry at format time then it could be slightly optimized for > >> those hdd's that come with true geometry. Was that taken out because > >> the computation was outweighing the seek time benefits, or what? > > > >Disk geometry is nonlinear, and the calculations involved in optimising > >for it are complex and not really compatible with the optimisations for > >this sort of thing already part of the filesystem. > > ... Which have been disabled since 2.0.5 or something, we're now running > on a pseudo geometry which fails horribly on huge disks. It's better than what we'd have otherwise. > According to Kirk, 16 to 32 is the right number of cylinder groups, > all over 50 is waste. we can end up with 200+ cgs on modern disks :-( Yay, even less locality of reference. You were paying attention to the thread on inode allocation policies that went past just recently? > >There's no such thing as "true geometry" anymore. > > Well, there is, but we'd like to avoid having to deal with it. Gawd, and the pedants out from the voodvork came. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message