All of lore.kernel.org
 help / color / mirror / Atom feed
From: Shawn Starr <shawn.starr@rogers.com>
To: LKML <linux-kernel@vger.kernel.org>
Subject: Re: Crazy Idea: Replacing /dev using sysfs over time
Date: Mon, 26 Sep 2005 20:02:11 -0400	[thread overview]
Message-ID: <200509262002.11834.shawn.starr@rogers.com> (raw)
In-Reply-To: <200509261928.20701.shawn.starr@rogers.com>


Or instead of even needing major/minor we just have:


/sys/class/sound
       `- - audio0
                |
                | - raw
        `-- dsp0
                |
                | - raw

Then instead, let udev know that audio0 and dsp0 belong to one sound card 
device or have it report this in sysfs:

/dev/class/sound
       `--sound0     
               | 
               | -- dev
               | -- device -> ../../../devices/pci0000:00
               `-- audio0
                     |- device  -> ../../../devices/pci0000:00/0000:00:1f.5
               `--dsp0
                     | -device -> ../../../devices/pci0000:00/0000:00:1f.5
               `--mixer0 
        `--sound1
               |
               | -dev   
               | -- device -> ../../../devices/pci0000:00:1a
               `--audio0
                       |- device 
--> ../../../devices/pci0000:00:1a/0000:00:1c.6

And so forth.


Then map sound0 devices in /dev/dsp0 /dev/mixer0 /dev/audio0  with udev

*NOTE: I am not avocating devfs, but more of keeping sysfs as the primary 
structure for devices.

On September 26, 2005 19:28, Shawn Starr wrote: > I wonder if in the future, 
we can just eliminate /dev altogether (or map it
> via sysfs until older apps move away from /dev). It just seems we could
> represent major,minor in a sysfs node:
>
>         /sys/class/block/
>         `-- sda
>
>             |-- sda1
>             |
>                     | - major
>                     | - minor
>                     | - raw
>             |
>             |-- sda2
>             |
>                     | - major
>                     | - minor
>                     | - raw
>
>             `-- sda3
>
> and so forth, or under a different branch elsewhere.
>
> Does it make sense? Logical? Illogical? Do we really need /dev other than
> for historical/legacy purposes?
>
> Shawn.

  parent reply	other threads:[~2005-09-27  0:02 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-09-26 23:28 Crazy Idea: Replacing /dev using sysfs over time Shawn Starr
2005-09-26 23:55 ` Diego Calleja
2005-09-27  0:02 ` Shawn Starr [this message]
2005-09-27 10:25 ` Greg KH
2005-09-27 12:52 ` linux-os (Dick Johnson)

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=200509262002.11834.shawn.starr@rogers.com \
    --to=shawn.starr@rogers.com \
    --cc=linux-kernel@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.