Greetings!


Recently I purchased a brand new laptop and sadly the hardware (like chip set, ...) is hardly documented at all so I can only make guesses. Naturally I wanted to install Linux on it and ran into a very weird problem; I have some trouble explaining it as English isn't my mother language.

I was installing Gentoo from a Live CD and was typing commands. Suddenly the characters on the screen started acting strangely *lol*. What happened is that there was some kind of lag between the actual keystroke and the system recognizing it. But it didn't take a specific time for a character to appear but rather 2-3 more keystrokes.

For example if I type 'reboot' (without the quotes) I see 'rebo' on the screen. After I press ENTER, the second 'o' appears and nothing happens, as if I really hadn't entered the command yet. Only after two more keystrokes (for the 't' and the ENTER), the command gets executed. But these two keystrokes are also lagged and will appear if I enter the next command.

This happens not only to typed text but also to control key combinations. Like when I want to change to another terminal using 'ALT + F2', I have to press 2 or 3 more keys for the terminal to actually change.

I cannot see any relation between any (system-)events and this behavior except that it happens much more frequently when the system is under heavy load (compiling etc.).

After a clean boot everything works fine. Then after some seemingly random time this happens and won't get better until after a reboot.

Also it happens with both the Live CD kernel and a self-compiled one.

There is one more thing which might be of importance: Gentoo provides compile-time-optimized LiveCDs for several CPU architectures. I have a Celeron 2.6 GHz and got the Live CD for Pentium4. Also I configured my kernel to be optimized for Pentium4. Might this be the wrong architecture type and the cause for my problem?


I am out of ideas, can anyone please help me find out what to do? And sorry about the confusing explanation.

I am right now compiling a kernel for a different target architecture to test if this is the problem. I will post the result. Any hints are - of course - still welcome. =)