Distribution: RH9 Problem Description: I have a USB Logitech optical mouse ( which the mouse wheel doesn't work ) and a ps/2 at keyboard. When I press a key and move the mouse just after pressing the key, the key will repeat itself depending on how much I move the mouse. So, I press the 'b' key ( console or xterm in X ) and then wiggle the mouse back and forth : I then have 10 b's instead of just 1. The most it will repeat is generally 8-12 extra characters. Steps to reproduce: Boot to console -> login. Press 'b' on the keyboard and wiggle mouse directly after pressing. Voila : 10 b's on the screen. Works every time :> I have started to peruse the /drivers/input code (nice code btw :>)and am trying to diagnose this myself. However, a first glance at the usb stuff leads me to believe I will either need a lot of time or some help. Thanks, Simon.
Created attachment 746 [details] dmesg Big thanks to Dave Jones et all for getting the Radeon 9200 to work with AGP3 (8x) :>
Created attachment 747 [details] lsmod
Created attachment 748 [details] proc usb
Created attachment 749 [details] proc devices
Created attachment 750 [details] proc interrupts
Aha! Compiled all the usb stuff in kernel and voila -> it all works :> Mouse wheel works and no more weird problems with key repeats. I couldn't debug the modules because klogd : Aug 28 23:59:09 localhost kernel: klogd 1.4.1, log source = /proc/kmsg started. Aug 28 23:59:09 localhost kernel: Inspecting /boot/System.map-2.6.0-test4 Aug 28 23:59:09 localhost kernel: Loaded 21455 symbols from /boot/System.map-2.6.0-test4. Aug 28 23:59:09 localhost kernel: Symbols match kernel version 2.6.0. Aug 28 23:59:09 localhost kernel: No module symbols loaded - kernel modules not enabled. Adding sybmols for modules is listed as a TODO for module-init-tools-0.9.13 and 0.9.14 won't compile. Am I missing something here? Should I be able to debug modules? Could this be the cause of why it wasn't working? Needless to say all the debugging info for USB came pouring in the console when I compiled USB into the kernel. I've been looking over the driver/usb/input and driver/input code for the past week, so I am perfectly willing to trace the problem I was having with USB compiled as modules if I have in fact found a problem and am not just missing something :>
Most likely the BIOS is playing tricks with us. What if you load the USB modules _before_ the input modules? (You'll need CONFIG_EMBEDDED for this).
What is the current status of this bug? Are you seeing this craziness with 2.6.11-rc*? Thanks, Nish
Looking back I think this problem was specific to RH9. I was never able to trace exactly why it was happening( and I spent a lot of time learning about USB and input devices ) but I haven't had the problem on any distro's since then. It probably stemmed from my manual conversion from 2.4 to 2.5/6. I eventually had to abandon that setup as my boot process was so mangled due to all my testing. I say close the bug.
Consider it close :)