I have a few machines on a KVM switch box. If I switch away from my gentoo desktop, then switch back to it, the wheel on the mouse behaves erratically. Scrolling up causes any number of other mouse actions to be interpreted instead (random left/right clicks usually). All other mouse behaviour is normal. This was a problem in all 2.6 series kernels up to and including 2.6.6 and was fixed in 2.6.7 (IIRC). It continued to be fine in 2.6.8 but has returned in 2.6.9. I have the following parameters on my grub config; kernel /boot/bzImage2 root=/dev/hda4 psmouse_noext=1 psmouse_proto=imps removing the mouse parameters has no effect on the described behaviour. The following is in xorg.conf; Section "InputDevice" Identifier "Mouse1" Driver "mouse" Option "Protocol" "imps/2" Option "Device" "/dev/mouse" Option "ZAxisMapping" "4 5" EndSection Once the issue manifests itself, a full restart is the only cure. ie it's nothing to do with XOrg since restarting it doesn't fix it. Reproducible: Always Steps to Reproduce: 1. boot 2.6.9 gentoo-dev-sources kernel 2. use KVM to switch to another machine, then switch back 3. try to use wheel on mouse to scroll up in any application Actual Results: random series of interpreted left/right clicks with varied results. Expected Results: window scrolls up
Upstream issue, perhaps you could try the patch posted at the end here: http://bugme.osdl.org/show_bug.cgi?id=2082 and give feedback.
the patch works to an extent. After switching back to the 2.6.9 machine with the KVM, the mouse wheel initially fails to respond at all in the up direction (previously this generated random click events), then after a few seconds it begins to work as the reset occurs (visible in dmesg).. psmouse.c: Explorer Mouse at isa0060/serio1/input0 - driver resynched. psmouse.c: Explorer Mouse at isa0060/serio1/input0 - driver resynched. psmouse.c: Explorer Mouse at isa0060/serio1/input0 - driver resynched. psmouse.c: Explorer Mouse at isa0060/serio1/input0 - driver resynched. psmouse.c: Explorer Mouse on isa0060/serio1 reports too many errors, issuing reconnect request I can certainly live with this. Will this patch be added to the gentoo-dev-sources kernel or do I need to keep it around and re-apply after any kernel source upgrade? Regards,
I'm not keen to apply this until it is merged into the 2.6.10-rc tree. It won't be unless the author gets some feedback. Could you please sign up for the kernel bugzilla and post your feedback at the end of that bug? (When/if you do come to a conclusive patch with Dmitry, please reopen this bug and i'll add it to gentoo-dev-sources once its sent to Linus)
done.
