[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

weird-o-rama: Type5 keyboard suddenly becomes PC104 keyboard...




I have an Ultra5 running a custom 2.6.8 kernel that I'm using as a LAMP server. I have a PCI USB 2.0 board installed so that I can use my PC104 USB keyboard & mouse and put the Ultra5 on my 4-port USB KVM switch while I set everything up & debug, then I'll yank the USB board and move the machine across the building to the server room, where there's plenty of space for the Sun type 5 keyboard & mouse. The kernel I'm running now ("USB kernel") has USB and full HID compiled-in and sun KB & mouse as modules (not loaded). I also compiled the exact same kernel, but with Sun keyboard & mouse compiled-in and all USB stuff as modules ("Sun kernel"). The whole time using the USB kernel, I had to have the Sun KB attached so that the machine wouldn't boot to a serial console, and to interact with the OpenFirmware prompt.

I just tried booting with the Sun kernel after (using the USB kernel) running 'dpkg-reconfigure console-data' followed by 'dpkg-reconfigure console-common'. Using these two commands, I set the system keyboard map to "sunkeymap", which is apparently correct for a Sun type 5 US keyboard. After rebooting, the system took no input from the USB keyboard and did take input from the Sun keyboard, as expected. However, what I didn't expect was this: the keys on the sun keyboard, even with the "sunkeymap" console mapping, generate exactly the same output as the keys on the USB keyboard when I run it with the Sun keyboard map. It's as though the Sun keyboard's firmware was reprogrammed to generate PC104 scan codes. No, I'm not on drugs. And the Sun keyboard works normally at the OpenFirmware prompt. Anyone have any clue what's causing this? Any idea how I can fix this?

As a very slight aside, my keyboard actually has "MODEL: TYPE 6" molded into the bottom, but it ran perfectly with the type 5 keymap before, and I've only ever read about type 6 keyboards in USB form.

Thanks.

Dan.



Reply to: