Subject: Re: Work-in-progress "wedges" implementation
To: Eric Haszlakiewicz <>
From: Jason Thorpe <>
List: tech-kern
Date: 09/24/2004 17:31:54
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII; format=flowed

On Sep 23, 2004, at 9:29 AM, Eric Haszlakiewicz wrote:

> On Wed, Sep 22, 2004 at 04:35:37PM -0700, Jason Thorpe wrote:
>> Oh.  Well, eventually a whole-disk e.g. /dev/sd0 will be there.  The
>> wedges are also named /dev/dk0, /dev/dk1, etc.  The /dev/dk/WegeName
>> names are really intended to just be aliases.  If the wedge doesn't
>> have a name, you can still access via the /dev/dkN path.
> 	is that /dev/sd0 as a directory, to make it possible to access the
> wedges on a particular device?
> e.g. /dev/sd0/whole-disk, /dev/sd0/wedge0, /dev/sd0/wedge1, etc..?

Well, you can't have both a file called "sd0" and a directory called 
"sd0" in the same place...

> I saw someone mention using a full dmesg path (/dev/foo/umass1/sd2/...)
> but since device names are unique (can't have sd2 at umass1 and sd2 at 
> umass4)
> that doesn't seem necessary.

Full dmesg path appeals to me in a lot of ways... but, then again, 
we're digressing into the devfs discussion that I'm trying to avoid for 
the moment :-)

         -- Jason R. Thorpe <>

content-type: application/pgp-signature; x-mac-type=70674453;
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
content-transfer-encoding: 7bit

Version: GnuPG v1.2.3 (Darwin)