From: Ben Hutchings <bhutchings@solarflare.com>
To: Greg KH <greg@kroah.com>
Cc: Matt Domsch <Matt_Domsch@dell.com>,
Kay Sievers <kay.sievers@vrfy.org>, dann frazier <dannf@hp.com>,
linux-hotplug@vger.kernel.org, Narendra_K@dell.com,
netdev@vger.kernel.org, Jordan_Hargrave@dell.com,
Charles_Rose@dell.com
Subject: Re: [PATCH] udev: create empty regular files to represent net
Date: Thu, 29 Oct 2009 16:49:35 +0000 [thread overview]
Message-ID: <1256834975.2827.63.camel@achroite> (raw)
In-Reply-To: <20091029142554.GA16869@kroah.com>
On Thu, 2009-10-29 at 07:25 -0700, Greg KH wrote:
> On Thu, Oct 29, 2009 at 08:11:25AM -0500, Matt Domsch wrote:
> > Netdev team - are you in agreement that having multiple names to
> > address the same netdevice is a worthwhile thing to add, to allow a
> > variety of naming schemes to exist simultaneously? If not, this whole
> > discussion will be moot, and my basic problem, that the ethX naming
> > convention is nondeterministic, but we need determinism, remains
> > unresolved.
>
> I'm still totally confused as to why you think this. What is wrong with
> what we do today, which is name network devices in a deterministic
> manner by their MAC in userspace? That name goes into the kernel, and
> everyone uses the same name and is happy.
>
> If you don't like naming by MAC, then pick some other deterministic
> naming scheme that works for your hardware and write udev rules for it.
>
> You could easily name them in a way that could keep the lowest number
> (eth0) for the lowest PCI id if you so desired and your BIOS guaranteed
> it.
>
> This way the kernel has only one name, and so does userspace, and
> everyone is happy.
I thought there was a general trend in udev development to provide
default rules that work for almost everyone, so few users/administrators
need to override or add to them. Compare disks and net devices:
1. Stable kernel device id
Disks: block device number
Net devices: ifindex
2. Unique identifier (across reboot)
Disks: label or UUID (each with limitations)
Net devices: (MAC address, subtype)
3. Name assignment mechanism
Disks: kernel suggests a name; udev can assign any number
Net devices: kernel assigns a single name; udev can override it
4. Default name assignment policy
Disks: names disk by device path (id), label and UUID
Net devices: assigns arbitrary stable names per (MAC address, subtype)
5. Naming by users
Disks: user can identify by any method without having to choose on a
system-wide basis
Net devices: user must identify by single name; policy can be overridden
on a system-wide basis
I fully understand the technical reasons for differences 3-5, but why
should users have to put up with it?
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
WARNING: multiple messages have this Message-ID (diff)
From: Ben Hutchings <bhutchings@solarflare.com>
To: Greg KH <greg@kroah.com>
Cc: Matt Domsch <Matt_Domsch@dell.com>,
Kay Sievers <kay.sievers@vrfy.org>, dann frazier <dannf@hp.com>,
linux-hotplug@vger.kernel.org, Narendra_K@dell.com,
netdev@vger.kernel.org, Jordan_Hargrave@dell.com,
Charles_Rose@dell.com
Subject: Re: [PATCH] udev: create empty regular files to represent net interfaces
Date: Thu, 29 Oct 2009 16:49:35 +0000 [thread overview]
Message-ID: <1256834975.2827.63.camel@achroite> (raw)
In-Reply-To: <20091029142554.GA16869@kroah.com>
On Thu, 2009-10-29 at 07:25 -0700, Greg KH wrote:
> On Thu, Oct 29, 2009 at 08:11:25AM -0500, Matt Domsch wrote:
> > Netdev team - are you in agreement that having multiple names to
> > address the same netdevice is a worthwhile thing to add, to allow a
> > variety of naming schemes to exist simultaneously? If not, this whole
> > discussion will be moot, and my basic problem, that the ethX naming
> > convention is nondeterministic, but we need determinism, remains
> > unresolved.
>
> I'm still totally confused as to why you think this. What is wrong with
> what we do today, which is name network devices in a deterministic
> manner by their MAC in userspace? That name goes into the kernel, and
> everyone uses the same name and is happy.
>
> If you don't like naming by MAC, then pick some other deterministic
> naming scheme that works for your hardware and write udev rules for it.
>
> You could easily name them in a way that could keep the lowest number
> (eth0) for the lowest PCI id if you so desired and your BIOS guaranteed
> it.
>
> This way the kernel has only one name, and so does userspace, and
> everyone is happy.
I thought there was a general trend in udev development to provide
default rules that work for almost everyone, so few users/administrators
need to override or add to them. Compare disks and net devices:
1. Stable kernel device id
Disks: block device number
Net devices: ifindex
2. Unique identifier (across reboot)
Disks: label or UUID (each with limitations)
Net devices: (MAC address, subtype)
3. Name assignment mechanism
Disks: kernel suggests a name; udev can assign any number
Net devices: kernel assigns a single name; udev can override it
4. Default name assignment policy
Disks: names disk by device path (id), label and UUID
Net devices: assigns arbitrary stable names per (MAC address, subtype)
5. Naming by users
Disks: user can identify by any method without having to choose on a
system-wide basis
Net devices: user must identify by single name; policy can be overridden
on a system-wide basis
I fully understand the technical reasons for differences 3-5, but why
should users have to put up with it?
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
--
To unsubscribe from this list: send the line "unsubscribe linux-hotplug" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2009-10-29 16:49 UTC|newest]
Thread overview: 210+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <EDA0A4495861324DA2618B4C45DCB3EE5894ED@blrx3m08.blr.amer.dell.com>
2009-10-09 14:00 ` PATCH: Network Device Naming mechanism and policy Narendra K
2009-10-09 14:00 ` Narendra K
2009-10-09 14:51 ` Matt Domsch
2009-10-09 14:51 ` Matt Domsch
2009-10-09 16:23 ` Bryan Kadzban
2009-10-09 16:23 ` Bryan Kadzban
2009-10-09 16:56 ` Marco d'Itri
2009-10-09 16:56 ` Marco d'Itri
2009-10-12 10:41 ` Scott James Remnant
2009-10-12 10:41 ` Scott James Remnant
2009-10-12 11:31 ` Ben Hutchings
2009-10-12 11:31 ` Ben Hutchings
2009-10-12 17:37 ` Bill Nottingham
2009-10-12 17:37 ` Bill Nottingham
2009-10-13 18:06 ` Dan Williams
2009-10-13 18:06 ` Dan Williams
2009-10-13 18:53 ` Ben Hutchings
2009-10-13 18:53 ` Ben Hutchings
2009-10-13 19:53 ` John W. Linville
2009-10-13 19:53 ` John W. Linville
2009-10-09 16:36 ` Greg KH
2009-10-09 16:36 ` Greg KH
2009-10-09 17:17 ` Matt Domsch
2009-10-09 17:17 ` Matt Domsch
2009-10-09 17:22 ` Greg KH
2009-10-09 17:22 ` Greg KH
2009-10-09 21:09 ` Matt Domsch
2009-10-09 21:09 ` Matt Domsch
2009-10-10 2:44 ` Stephen Hemminger
2009-10-10 2:44 ` Stephen Hemminger
2009-10-10 4:40 ` Matt Domsch
2009-10-10 4:40 ` Matt Domsch
2009-10-10 5:23 ` Greg KH
2009-10-10 5:23 ` Greg KH
2009-10-10 8:17 ` Sujit K M
2009-10-10 8:29 ` Sujit K M
2009-10-10 16:27 ` Greg KH
2009-10-10 16:27 ` Greg KH
2009-10-10 19:00 ` Ben Hutchings
2009-10-10 19:00 ` Ben Hutchings
2009-10-10 21:10 ` Greg KH
2009-10-10 21:10 ` Greg KH
2009-10-10 12:47 ` Matt Domsch
2009-10-10 12:47 ` Matt Domsch
2009-10-10 16:25 ` Greg KH
2009-10-10 16:25 ` Greg KH
2009-10-10 17:34 ` Bryan Kadzban
2009-10-10 17:34 ` Bryan Kadzban
2009-10-10 21:13 ` Greg KH
2009-10-10 21:13 ` Greg KH
2009-10-12 6:21 ` Bryan Kadzban
2009-10-12 6:21 ` Bryan Kadzban
2009-10-12 16:19 ` Bryan Kadzban
2009-10-12 16:19 ` Bryan Kadzban
2009-10-11 16:40 ` David Zeuthen
2009-10-11 16:40 ` David Zeuthen
2009-10-11 18:47 ` Greg KH
2009-10-11 18:47 ` Greg KH
2009-10-10 18:11 ` Bill Fink
2009-10-10 18:11 ` Bill Fink
2009-10-10 18:35 ` Kay Sievers
2009-10-10 18:35 ` Kay Sievers
2009-10-11 21:10 ` Rob Townley
2009-10-11 21:10 ` Rob Townley
2009-10-11 23:04 ` Matt Domsch
2009-10-11 23:04 ` Matt Domsch
2009-10-12 3:00 ` Greg KH
2009-10-12 3:00 ` Greg KH
2009-10-12 18:35 ` Rob Townley
2009-10-12 18:35 ` Rob Townley
2009-10-12 18:44 ` Matt Domsch
2009-10-12 18:44 ` Matt Domsch
2009-10-12 17:45 ` Bill Nottingham
2009-10-12 17:45 ` Bill Nottingham
2009-10-12 17:55 ` Greg KH
2009-10-12 17:55 ` Greg KH
2009-10-12 18:07 ` Bill Nottingham
2009-10-12 18:07 ` Bill Nottingham
2009-10-12 18:15 ` Greg KH
2009-10-12 18:15 ` Greg KH
2009-10-10 18:32 ` Stephen Hemminger
2009-10-10 18:32 ` Stephen Hemminger
2009-10-10 21:06 ` Greg KH
2009-10-10 21:06 ` Greg KH
2009-10-13 18:02 ` Dan Williams
2009-10-13 18:02 ` Dan Williams
2009-10-13 18:53 ` Narendra_K
2009-10-13 18:56 ` Narendra_K
2009-10-12 7:30 ` Kurt Van Dijck
2009-10-12 7:30 ` Kurt Van Dijck
2009-10-11 0:37 ` Marco d'Itri
2009-10-11 0:37 ` Marco d'Itri
2009-10-13 15:08 ` dann frazier
2009-10-13 15:08 ` dann frazier
2009-10-13 17:13 ` Narendra_K
2009-10-13 17:25 ` Narendra_K
2009-10-13 17:36 ` dann frazier
2009-10-13 17:36 ` dann frazier
2009-10-16 0:32 ` dann frazier
2009-10-16 0:32 ` dann frazier
2009-10-16 14:02 ` Narendra_K
2009-10-16 14:14 ` Narendra_K
2009-10-16 15:20 ` dann frazier
2009-10-16 15:20 ` dann frazier
2009-10-16 15:33 ` Ben Hutchings
2009-10-16 15:33 ` Ben Hutchings
2009-10-16 15:41 ` dann frazier
2009-10-16 15:41 ` dann frazier
2009-10-16 21:40 ` dann frazier
2009-10-16 21:40 ` dann frazier
2009-10-19 11:30 ` Narendra_K
2009-10-19 11:42 ` Narendra_K
2009-10-19 16:14 ` Bryan Kadzban
2009-10-19 16:14 ` Bryan Kadzban
2009-11-04 14:23 ` Narendra_K
2009-11-04 14:35 ` Narendra_K
2009-11-06 8:49 ` Marco d'Itri
2009-11-06 8:49 ` Marco d'Itri
2009-11-06 22:06 ` Matt Domsch
2009-11-06 22:06 ` Matt Domsch
2009-11-06 22:35 ` Marco d'Itri
2009-11-06 22:35 ` Marco d'Itri
2009-11-06 23:17 ` dann frazier
2009-11-06 23:17 ` dann frazier
2009-11-09 14:41 ` Narendra_K
2009-11-09 14:53 ` Narendra_K
2009-11-10 17:23 ` Stephen Hemminger
2009-11-10 17:23 ` Stephen Hemminger
2009-11-11 6:31 ` Narendra_K
2009-11-11 6:43 ` Narendra_K
2009-11-06 22:05 ` Domsch, Matt
2009-11-06 22:05 ` Domsch, Matt
2009-10-22 6:36 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-22 6:36 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-27 20:55 ` Matt Domsch
2009-10-27 20:55 ` Matt Domsch
2009-10-28 8:23 ` [PATCH] udev: create empty regular files to represent net Kay Sievers
2009-10-28 8:23 ` [PATCH] udev: create empty regular files to represent net interfaces Kay Sievers
2009-10-28 13:03 ` Matt Domsch
2009-10-28 13:03 ` Matt Domsch
2009-10-28 15:09 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-28 15:09 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-28 16:09 ` Jordan_Hargrave
2009-10-28 16:09 ` Jordan_Hargrave
2009-10-28 16:09 ` Jordan_Hargrave
2009-10-28 16:09 ` Jordan_Hargrave
2009-10-28 16:11 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-28 16:11 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-28 13:06 ` [PATCH] udev: create empty regular files to represent net Ben Hutchings
2009-10-28 13:06 ` [PATCH] udev: create empty regular files to represent net interfaces Ben Hutchings
2009-10-28 19:15 ` Narendra_K
2009-10-28 19:27 ` Narendra_K
2009-10-29 13:11 ` Matt Domsch
2009-10-29 13:11 ` Matt Domsch
2009-10-29 14:25 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-29 14:25 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-29 16:44 ` Narendra_K
2009-10-29 16:56 ` Narendra_K
2009-10-29 16:52 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-29 16:52 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-29 17:22 ` [PATCH] udev: create empty regular files to represent netinterfaces Narendra_K
2009-10-29 17:34 ` Narendra_K
2009-10-29 17:50 ` [PATCH] udev: create empty regular files to represent dann frazier
2009-10-29 17:50 ` [PATCH] udev: create empty regular files to represent netinterfaces dann frazier
2009-10-29 16:49 ` Ben Hutchings [this message]
2009-10-29 16:49 ` [PATCH] udev: create empty regular files to represent net interfaces Ben Hutchings
2009-10-29 16:55 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-29 16:55 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-29 17:12 ` [PATCH] udev: create empty regular files to represent net Ben Hutchings
2009-10-29 17:12 ` [PATCH] udev: create empty regular files to represent net interfaces Ben Hutchings
2009-10-29 17:20 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-29 17:20 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-29 17:46 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-29 17:46 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-30 3:30 ` [PATCH] udev: create empty regular files to represent net Marco d'Itri
2009-10-30 3:30 ` [PATCH] udev: create empty regular files to represent net interfaces Marco d'Itri
2009-10-30 5:38 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-30 5:38 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-30 6:22 ` [PATCH] udev: create empty regular files to represent net Marco d'Itri
2009-10-30 6:22 ` [PATCH] udev: create empty regular files to represent net interfaces Marco d'Itri
2009-10-30 15:00 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-30 15:00 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-30 15:13 ` Narendra_K
2009-10-30 15:25 ` Narendra_K
2009-10-30 16:08 ` [PATCH] udev: create empty regular files to represent net dann frazier
2009-10-30 16:08 ` [PATCH] udev: create empty regular files to represent net interfaces dann frazier
2009-10-30 16:53 ` [PATCH] udev: create empty regular files to represent netinterfaces Narendra_K
2009-10-30 16:54 ` Narendra_K
2009-10-30 17:05 ` [PATCH] udev: create empty regular files to represent dann frazier
2009-10-30 17:05 ` [PATCH] udev: create empty regular files to represent netinterfaces dann frazier
2009-10-30 17:10 ` [PATCH] udev: create empty regular files to represent net interfaces Matt Domsch
2009-10-30 17:10 ` Matt Domsch
2009-10-30 17:13 ` [PATCH] udev: create empty regular files to represent net Greg KH
2009-10-30 17:13 ` [PATCH] udev: create empty regular files to represent net interfaces Greg KH
2009-10-30 7:45 ` Hannes Reinecke
2009-10-30 7:45 ` Hannes Reinecke
2009-10-30 16:22 ` Bryan Kadzban
2009-10-30 16:22 ` Bryan Kadzban
2009-10-30 16:34 ` [PATCH] udev: create empty regular files to represent net Stephen Hemminger
2009-10-30 16:34 ` [PATCH] udev: create empty regular files to represent net interfaces Stephen Hemminger
2009-10-13 19:51 ` PATCH: Network Device Naming mechanism and policy Greg KH
2009-10-13 19:51 ` Greg KH
2009-10-13 20:00 ` Jordan_Hargrave
2009-10-13 20:00 ` Jordan_Hargrave
2009-10-13 20:19 ` Greg KH
2009-10-13 20:19 ` Greg KH
2009-10-13 22:05 ` Matt Domsch
2009-10-13 22:05 ` Matt Domsch
2009-10-13 22:08 ` dann frazier
2009-10-13 22:08 ` dann frazier
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1256834975.2827.63.camel@achroite \
--to=bhutchings@solarflare.com \
--cc=Charles_Rose@dell.com \
--cc=Jordan_Hargrave@dell.com \
--cc=Matt_Domsch@dell.com \
--cc=Narendra_K@dell.com \
--cc=dannf@hp.com \
--cc=greg@kroah.com \
--cc=kay.sievers@vrfy.org \
--cc=linux-hotplug@vger.kernel.org \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.