Date: Mon, 22 Jul 1996 20:58:02 +0200 (SAT) From: Robert Nordier <rnordier@iafrica.com> To: pst@shockwave.com (Paul Traina) Cc: jkh@time.cdrom.com, hackers@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: anyone working on upgrading the msdosfs to NetBSD levels? Message-ID: <199607221858.UAA00383@eac.iafrica.com> In-Reply-To: <199607220534.WAA03617@precipice.shockwave.com> from "Paul Traina" at Jul 21, 96 10:34:02 pm
next in thread | previous in thread | raw e-mail | index | archive | help
Paul Traina wrote: > > Do we -need- a complete rewrite? It looks like the current NetBSD version > is actually in pretty damn good shape. Unfortunately, detangling changes > is not going to be bloody easy (<insert obligatory flame about gratuitous > source code changes>). > > Paul I ran some fairly extensive tests on the FreeBSD msdosfs several months ago. At that stage, it seemed to me to have fundamental problems that did not appear to be addressed (or easily addressable), even taking into account NetBSD diffs. Unfortunately - as you mention - the NetBSD code contains many changes that are more stylistic than functional. Over the last few weeks, I've been working on a few msdosfs patches myself. However, the most serious FreeBSD msdosfs problem (where it causes corruption to other partitions) apparently needs a 64-cluster FIPS-ed FAT filesystem with an EIDE controller translating to 64 heads, 63 sectors to be readily reproducible. (At least, I haven't reproduced it on any other 64-cluster filesystems.) Personally, I'd be glad if someone did succumb to the urge to bring NetBSD changes across to the msdosfs. That way, I could forget msdosfs patches, and concentrate on getting the new vfatfs finished. The vfatfs would then also have to be demonstrably better than an improved msdosfs to be worth adopting: a good incentive, and probably a good thing for FreeBSD generally. -- Robert Nordier
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199607221858.UAA00383>