Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 24 Nov 2002 02:05:15 -0800
From:      David Schultz <dschultz@uclink.Berkeley.EDU>
To:        Cliff Sarginson <cls@raggedclown.net>
Cc:        freebsd-questions@FreeBSD.ORG
Subject:   Re: Confirmation: ext2fs requires kernel rebuild?
Message-ID:  <20021124100514.GB3172@HAL9000.homeunix.com>
In-Reply-To: <20021121232340.GA70539@raggedclown.net>
References:  <p05200f1ab9ff72b26ae1@[66.92.104.201]> <20021119100625.GC679@HAL9000.homeunix.com> <p05200f24ba0073e6b323@[66.92.104.201]> <20021121035034.GA2591@HAL9000.homeunix.com> <p05200f09ba020d2729e6@[66.92.104.201]> <20021121072556.GB1590@raggedclown.net> <20021121230912.GE6062@HAL9000.homeunix.com> <20021121232340.GA70539@raggedclown.net>

next in thread | previous in thread | raw e-mail | index | archive | help
Thus spake Cliff Sarginson <cls@raggedclown.net>:
> On Thu, Nov 21, 2002 at 03:09:12PM -0800, David Schultz wrote:
> > Thus spake Cliff Sarginson <cls@raggedclown.net>:
> > FreeBSD support for ext2fs is a specific instance of the more
> > general problem that features that very few people care about tend
> > not to get maintained.  FreeBSD already has a filesystem that is
> > more complete and faster than ext2fs,[1] so the only people using
> > both UFS and ext2fs are also running Linux.  If you mount an
> > ext2fs filesystem r/w, consider yourself a beta tester.
> > 
> Oh I am not comparing their relative merits. But the thing is there are
> people who for one reason or another run heterogenous networks. I mean
> it is not a crime to run both Linux and FreeBSD (or is it ?). My feeling
> is that it is a shame that is all. 
> 
> As for your scathing comment about the evolution of EXT2, I have no comment
> since it sounds like opinion rather than fact. 

The point of my post was to point out why nobody is interested in
maintaining FreeBSD's ext2fs support.  But ext2fs really is a
simplified version of FFS!  Basically, they took out support for
fragments and vastly simplified the allocation policies.  Some of
the things they took out aren't even used anymore in FFS, like the
code to take rotational offsets on the disk into consideration.
On the other hand, they left out Kirk's realloc algorithm, which
has been shown to significantly reduce fragmentation as
filesystems age.

(FYI, some people have expressed interest in porting ReiserFS to
FreeBSD.  Hans Reiser is willing to negotiate special exceptions
to the license, as long as Microsoft can't build the next version
of Windows on top of FreeBSD and be able to use his filesystem for
free.)

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021124100514.GB3172>