Comment # 16 on bug 1200983 from Takashi Iwai

(In reply to Michael Hirmke from comment #14)
> (In reply to Takashi Iwai from comment #13)
> > (In reply to Michael Hirmke from comment #12)
> > > (In reply to Stefan Dirsch from comment #10)
> > > > Thanks for testing. Indeed this makes it a kernel issue. You could also test
> > > > the mouse on the Linux console via 'evtest /dev/input/eventXX', but this is
> > > > a bit complicated to figure out the right device. So I only wanted to
> > > > mention keyboard here.
> > > 
> > > Keyboard and mouse both have problems one the console, too.
> > > The keyboard vanishes after a few seconds and returns, as soon as a key is
> > > pressed, but the first keypress is not shown. The mouse never works on the
> > > console - no events are shown.
> > 
> > What do you mean "vanishes" here? It means that the /dev/input/event* device
> > is gone?
> 
> no, there is no mouse pointer and the keyboard light is off.
> 
> > 
> > The point to check here is whether the device file persists and can
> > read/write from a user-space (no matter whether it's processed or not).  Or
> > is the (USB?) audio device itself disappears and reappears automatically.
> 
> Because the keyboard also has problems, this isn't easy to debug.
> What exactly would you like me to test?

If you have two USB keyboards, you can check it if a problem happens only a
single one or on both.  Also, if you can access remotely e.g. via ssh, it can
be checked without interrupting the input device status, too.