Re: [linux-usb] Re: USB device allocation
Kernel List (kernel@gladden-fields.demon.co.uk)
Sun, 10 Oct 1999 14:09:34 +0100
On Sat, Oct 09, 1999 at 07:46:42PM -0400, Theodore Y. Ts'o wrote:
> The difference here is that what we're providing is a unified scheme by
> which the kernel *notifies* a user-space process that some new device
> has appeared on the scene. It's a one-way communication, with no
> requirement that kernel code wait for the user-space process to do
> whatever it needs to do. The user-space daemon can ignore the event; it
> can create a node in /dev; it can initiate a GUI dialog box asking the
> user what how some new device should be configured; it can play "God
> Save the Queen" in three-part harmony using the multimedia speakers.
> It's completely up to user-space daemon, so it's therefore much more
> flexible than a kernel-only devfs scheme.
>
ok, I understand this, and I have always been put off installing
devfs because I use alsa and kerneli patches and I dont want
to wait weeks while one catches up with the others and devfs is
too big for me to alter myself.
I like the idea of a kmod style system.
New device plugs into system, driver wakes up and calls kmod equiv. This
looks in devices.conf file for the minor major assigned to this device
and create a device node in appropriate place. Sort of this layout I
have stuck in my head.
0 0 /dev/adevice1 /ftp/dev/adevice1
1 2-5 /dev/mouse
2 7-15 /dev/vdu
Where the major/minor is searched and the devicefiles following
are created. If it is a range and some are already taken then
0,1,2,3,++ are added to the end of the name.
Drivers built into kernel would never call the kmod equiv so
their names would have to entered manually. Of course Im sure
they could be built at boot time from /proc/devices or similar.
This would require the expanded dev_t that is being discussed.
Feel free to shoot me down as I am not a harcore kernel hacker.
Graeme
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/