Date: Mon, 5 Feb 1996 20:29:47 +0200 (SAT) From: Robert Nordier <rnordier@iafrica.com> To: luigi@labinfo.iet.unipi.it (Luigi Rizzo) Cc: hackers@freebsd.org Subject: Re: Some thoughts on FAT filesystems Message-ID: <199602051829.UAA01047@eac.iafrica.com> In-Reply-To: <199602051647.RAA21430@labinfo.iet.unipi.it> from "Luigi Rizzo" at Feb 5, 96 05:47:48 pm
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 5 Feb 1996, Luigi Rizzo wrote: > I'll try to rephrase it: some utility program creates a special file, > called "C:\FASTFAT" with the following features: > > + same size as a single FAT; > + its clusters are spread in the disk, one at the beginning of each CG > + its size does not change with time. > . . . . . Yes, that's clear. :-) It's a shame, but I guess ingenuity spent on the FAT fs may well just be ingenuity wasted.... The FAT3 approach seems too prone to get out of sync, anytime anyone does anything in DOS. It needs setting up; and it's just too _exposed_. Because it's non-standard, users will complain bitterly when it bombs out; because it is visible, it'll be an endless minor source of controversy.... If people really want the extra performance, I'd suggest the option of doing as Windows NT does and writing the FAT asynchronously. At least people are used to that, or used to the partial equivalent of running smartdrv without a write-through cache. And at least that would be a Microsoft-approved mess. :-) -- Robert Nordier
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199602051829.UAA01047>