Re: XFree + KDE + USB mouse = APM misbehavior

[Thomas Winischhofer <tw@webit.com>]

On Wednesday 05 December 2001 19:07, Michel Dänzer wrote:
> On Wed, 2001-12-05 at 11:00, Thomas Winischhofer wrote:
> > I use the following setup:
> >
> > Laptop computer
> > Xfree 4.1.0-9
> > KDE 2.2.2 (from yesterday)
> >
> > X is configured to use the built-in touchpad as primary mouse
> > input device, and an USB mouse as secondary. Works perfect, except
> > for the following issue:
> >
> > Under KDE 2.2.2, the box goes into suspend mode if
> >
> > - I don't use the keyboard and the touchpad but just the USB mouse
> > AND
> >
> > - I only MOVE the USB mouse during the suspend time-out (an do no
> > mouseclicks).
> >
> > It seems that either X or KDE don't take secondary mouse movements
> > into account for resetting the suspend time-out...?
> >
> > Can anyone confirm this?
> > Is this a KDE or X bug?
>
> Neither AFAICT. I don't see either having anything to do with APM
> suspend. I'd rather suspect apmd or the kernel APM code.

I can now answer my own question:

It's a design flaw within kde's laptop daemon. This thingy 
checks for keyboard and mouse events through parsing /proc/interrupts, 
searching for the keywords "[K|k]eyboard" and "[M|m]ouse" and checking 
whether or not interrupts have occurred.

Thus, no wonder the usb mouse doesn't count so to speak; in 
/proc/interrupts the usb mouse does not appear as "mouse" but as 
"usb-ohci" (or whatever driver you use) - and might as such share irq 
with other devices (such as sound in my case). :(

I will file a bug because this is obviously not a good way - 
considering some people using USB input devices only.

Thomas