[6062] in linux-scsi channel archive
No subject found in mail header
daemon@ATHENA.MIT.EDU (owner-linux-scsi@vger.rutgers.edu)
Thu Mar 11 01:34:58 1999
In-Reply-To: <Pine.LNX.4.04.9903102319590.5273-100000@andante.jic.com>
Date: Thu, 11 Mar 1999 15:39:01
From: owner-linux-scsi@vger.rutgers.edu
To: unlisted-recipients: ;(no To-header on input)
-0000=?us-ascii?Q?_(/etc/localtimep=97=12=08|=F7=FF=BF=3Dm =08?=
Reply-To: peterw@dascom.com
From: Peter Waltenberg <peterw@dascom.com>
To: Eric Youngdale <eric@andante.jic.com>
Subject: Re: SCSI generic interface addition
Cc: Douglas Gilbert <dgilbert@interlog.com>
Cc: Douglas Gilbert <dgilbert@interlog.com>, Bernard Hatt <bmh@arkady.demon.co.uk>, linux-scsi@vger.rutgers.edu, Richard Gooch <rgooch@atnf.csiro.au>
X-Orcpt: rfc822;linux-scsi@vger.rutgers.edu
Sender: owner-linux-scsi@vger.rutgers.edu
Precedence: bulk
X-Loop: majordomo@vger.rutgers.edu
>
>> Eric Youngdale writes:
>> > My main objection to this was that it never seemed obvious
>> > that this is something that really had to be in the kernel. There
>> > has got to be a user-space solution to this issue that wouldn't
>> > bloat the kernel.
>>
>> It's a pity so much FUD has been spread about devfs. I have raised
>> points that show that there are certain things you simply can't do
>> without devfs, and I have *NEVER* heard from the anti-devfs crowd how
>> you could do *all* these things in user space.
>
> It doesn't have to be an all or nothing proposition. OK, you do
> the things in the kernel that can't be done in user space, and do the rest
> in user space. Just because you can't do the whole thing in user space
> doesn't mean the whole thing absolutely must be in the kernel.
True, but devfs is optional. You don't see people complaining about supporting
UFS or NTFS in the kernel. You could do a lot of that in user space as well.
*cough* just like NFS ;)
It just happens that in this case you can do it neatly and elegantly in the kernel, and probably
with less code than hooks to user-space would take.
I'd also point out that keeping the user-space and kernel components in sync is a major
PITA for almost everything that takes that hybrid approach. NFS (tried to keep up with
knfsd lately ?), SMB/Samba, NCPFS , pcmcia, even PPP.
Since it doesn't contribute anything to runtime code bloat if you don't enable it I also
fail to see any rational reason for rejecting it other than the "addiction" one. I.e. we won't
be able to remove it once it goes in. However, if it's that useful .....
Peter
----------------------------------
E-Mail: Peter Waltenberg <peterw@dascom.com>
Date: 11-Mar-99
Time: 15:30:21
This message was sent by XFMail
----------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
