Re: why umsdos?

Anthony Barbachan (barbacha@Hinako.AMBusiness.com)
Sun, 8 Nov 1998 15:55:26 -0500


-----Original Message-----
From: Matthew Wilcox <Matthew.Wilcox@genedata.com>
To: Horvath Akos <maxx@yuc.sch.bme.hu>
Cc: linux-kernel@vger.rutgers.edu <linux-kernel@vger.rutgers.edu>;
humbubba@raptor.cqi.com <humbubba@raptor.cqi.com>
Date: Sunday, November 08, 1998 2:42 PM
Subject: Re: why umsdos?

>On Sun, Nov 08, 1998 at 04:17:40PM +0100, Horvath Akos wrote:
>> On Sun, 8 Nov 1998, Matthew Wilcox wrote:
>> > I don't like umsdos personally. Have you considered using a file on
the
>> > msdos filesystem with an ext2 filesystem on it instead? I have a
machine
>> > with its root filesystem mounted on /dev/loop0. A little playing with
>> > initrd makes this quite easy. This means you have no need for a untfs
>> > (and in my case, no need for a uadfs).
>>
>> Five reasons to not to use this:
>> - You can not resize an ext2fs image - repartitioning is better, because
>> FIPS, presizer or Partition Magic can resize msdos partitions
>
>PartitionMagic can now resize ext2 partitions. The code to do this was
>written by tytso under contract. There will (after a certain amount of
>time) be a standalone ext2resizer. You cannot resize files containing
>ext2 partitions now, I understand that. However you don't necessarily
>need to. If someone's `just trying it out', they will almost certainly
>not need to resize their partition. If they really do, they can be
>introduced to the joys of multiple partitions and a single filesystem
>hierarchy.
>
>> - You can not read az ext2fs image from dos
>
>There are several tools that do this. There's even a VxD for Windows 95.
>
>> - It will be slow (ok, umsdos is slow too)
>
>Then this is not an argument.
>
>> - A newbie will never be able to set up a looproot distribution
>
>I don't understand what point you're trying to make there. There should
>not be any user-visible difference between setting up a looproot and an
>umsdosroot distribution.
>
>> - Looproot is problematic and untested (for example it does not work in
>> vfat but works on an msdos partition)
>
>Why does it not work on vfat? All it does is use the filesystem bmap()
>method. Since vfat's bmap method is in fact the fat bmap method,
>I don't think this is possible. Have you tried it?
>
>> I think, a general ums-like pseudo-ext2-fs over not only umsdos, but all
>> non unix-compatible fs were very good and useful for the linux community.
>
>I disagree. It would be a lot of work for very little gain. Different
>filesystems differ so much that it would be almost impossible, and
>even if you did, it would always be a kludge of limited usefulness.
>Compared to the loop device which we already have with its multiple uses,
>I really don't see the point.
>

How about UMSSMB? Or UMS on netware? UMSNTFS? UMS(any other non-unix
filesystem)? We should not assume that linux users will be using, can use,
or want to use a unix compatable filesystem.

>--
>Matthew Wilcox <willy@bofh.ai>
>"I decry the current tendency to seek patents on algorithms. There are
>better ways to earn a living than to prevent other people from making use
of
>one's contributions to computer science." -- Donald E. Knuth, TAoCP vol 3
>
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.rutgers.edu
>Please read the FAQ at http://www.tux.org/lkml/
>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/