Turmvilla scenario keyboard input stuck in a subsystem

[Norman Feske]

Hi Pirmin,

> I'm playing around with the Turmvilla scenario as Norman provides it.

great to hear that you are experimenting with the scenario!

> Sometimes the keyboard input will be stuck in one of the started  
> subsystems. Mostly this is the case with the linux subsystem (I only  
> recognize this when I shut down linux and want to close the nit_fb),  
> but I also had the case where it was stuck in the noux subsystem.
> 
> Is there a way to unstuck the keyboard, or a way how I could debug  
> what happens?

I also noticed this sporadic behavior, which mystifies me quite a bit.
Sometimes there is either a superficial press event (for which no
release event is generated) or a release event goes missing. I notice
this when using a USB HID keyboard. The problem in this situation is
that nitpicker assumes that a key is kept pressing and behaves
differently. For example, unless the number of pressed keys is zero
(tracked by nitpicker's key_cnt variable), nitpicker won't switch the
keyboard focus when clicking with the mouse.

As a workaround - just to regain control over the situation - my
turmvilla branch has the commit "nitpicker: hack to reset key_cnt via
F9". Each time when releasing F9, the key_cnt is hard reseted to zero.
However, this is of course just a messy workaround until we find the
actual problem.

Just out of curiosity, are you using a USB or PS/2 keyboard?

Cheers
Norman

-- 
Dr.-Ing. Norman Feske
Genode Labs

http://www.genode-labs.com · http://genode.org

Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden
Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth