lkml.org 
[lkml]   [1998]   [Jan]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: devfs
Date
From
Leonard N. Zubkoff wrote:
> Richard Gooch wrote:
>
> h0c0i0l0p2 current scheme: no changes required :-)
>
> h0c0t0u0p2 new scheme: requires people to change compatibility
> symlinks and /etc/fstab (if you've gone that far)
>

I *strongly* prefer the latter scheme.

> I think the
> "new scheme" is a good compromise between slavish compatibility with systems
> like Solaris and the unique needs of Linux.

Gotta comment on this... Is there anyone reading this who believes that
the c?t?d?s? SCSI device naming scheme originated with Sun and HP? The
"one true UNIX(tm)" as inherited/purchased/acquired/stolen/borrowed from
AT&T has had this naming scheme since at least SVr?. If those who are
promulgating a gratuitously different naming scheme are doing so out of
a desire to spite Solaris, they are misguided or at least ignorant of
UN*X history. Bottom line: IMNSHO the current argument over the naming
of devices is rehashing old territory except for the valid point that
existing naming schemes didn't anticipate the notions of multi-channel
controllers and RAID devices. Anything we come up with now will more than
likely be found insufficient 15+ years later. (How old *is* System V
anyway?) I look forward to the changes that will make that "problem" a
reality :-).

> Use of "i" rather than "t" would
> be a gratuitous incompatibility, whereas the need for separate host number and
> channel number is fully justified.

Agreed on both counts.

> (placement of devices within the /dev hierarchy)
>
> /dev/sd_h0c0t0u0p2 OR:
> /dev/disc/sd_h0c0t0u0p2
>
> I'd like to see /dev/disk/sd_h?c?t?u?p? personally (i.e., a single directory
> for all disk devices). The spelling "disc" is nonstandard. I think a shallow
> directory tree is a good compromise. It will be easier to find devices if
> there is a shallow hierarchy based on function: disk, tape, cdrom, scsi (for
> generic), etc.

I'm with Leonard, except that I find the "sd_" prefix ugly in the extreme,
and possibly redundant. Are there plans to extend this naming scheme to
handle non-SCSI devices? If not, a prefix serves no purpose unless
that's how we plan to distinguish between raw and block devices in a
common subdirectory. Historically, this has been done by putting raw
devices in one directory (/dev/rdsk) and block in another (/dev/dsk).
If we go with /dev/disk, I could see /dev/disk/h?c?t?u?p? and
/dev/disk/rh?c?t?u?p? assuming there's a place in the evolution of
Linux for raw disk devices :-).

Anyway, more grist for the mill...

--
Bob Tracy | If you have any trouble sounding condescending,
Firewall Security Corp. | find a Unix user to show you how it's done.
rct@frus.com | -- Scott Adams: DNRC Newsletter 3.0

\
 
 \ /
  Last update: 2005-03-22 13:41    [W:0.247 / U:0.164 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site