Arguably, the following is a kludge that should not exist.
The X server should be taught to do "the right thing" with input
devices that come and go.
Per my mail earlier today titled: "X support for USB devices..."
The key question to enable proper support in X is:
"Is there some device we can select on that will return if the configuration
of the USB bus changes?"
This is the fundamental enabler needed out of the operating system.
The rest of this problem, unless things are fouled up at the driver level
in ways they should not be, is SMOP: a simple matter of programming,
in particular, in the X OS/device dependent layer. Don't treat the X server
as something that can't be changed: we already know it needs to, and it
is easier to work on the X server than device drivers, any day.
We have to do work here anyway, so kluding the driver isn't necessary.
- Jim
> Sender: [EMAIL PROTECTED]
> From: Paul Ashton <[EMAIL PROTECTED]>
> Date: Mon, 01 May 2000 20:56:23 +0100
> To: [EMAIL PROTECTED]
> Cc: [EMAIL PROTECTED]
> Subject: Re: [linux-usb] FW: Hot-plugging USB mice?
> -----
> > Is there any way to tell the USB subsystem (the backport patch from
> > 2.3.99pre6, applied to 2.2.15pre20) to leave the mouse driver connected,
> > even when the mouse has been unplugged?
>
> There used to be:
>
> + * 1999/08/19 - Fixed behaviour on mouse
> + * disconnect and suspend/resume. Added module parameter "force=1"
> + * to allow opening of the mouse driver before mouse has been plugged
> + * in (enables consistent XF86Config settings).
>
> But that got removed some time ago. Since we can do this with the
> keyboard I think we should also be able to do it with the mouse.
> Ideally any USB device should be able to have some "spoofing" mode
> that allows this.
>
--
Jim Gettys
Technology and Corporate Development
Compaq Computer Corporation
[EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
