From owner-freebsd-hackers  Sun Jan  7 22:55:41 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id WAA17266
          for hackers-outgoing; Sun, 7 Jan 1996 22:55:41 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA17261
          for <hackers@FreeBSD.ORG>; Sun, 7 Jan 1996 22:55:34 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id RAA13341; Mon, 8 Jan 1996 17:50:47 +1100
Date: Mon, 8 Jan 1996 17:50:47 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199601080650.RAA13341@godzilla.zeta.org.au>
To: bjj@sequent.com, wilko@yedi.iaf.nl
Subject: Re: Number of superblocks on big disks.
Cc: hackers@FreeBSD.ORG, mrcpu@cdsnet.net, roberto@keltia.freenix.fr
Sender: owner-hackers@FreeBSD.ORG
Precedence: bulk

>takes longer, though.  A quick look at the source to FreeBSD newfs
>suggests that contrary to the manual page, we default to `-i 4096'*,
>wheras SunOS is defaulting to `-i 2048', which would spend twice as much
>time writing inodes.

>* fsize defaults to pp->p_fsize, which defaults to 1k, and density defaults
>  to NFPI (4) * fsize => 4096.

I fixed the code to match the man page in my version, but 4.4lite2 fixed
the man page to match the code.  The latter seems best - if the frag
size is large then you probably don't plan to have many small files,
so you probably want less inodes.

Bruce