On Thursday 14 September 2006 10:54, Attilio Fiandrotti wrote: > >> Some info about the background: the bug > >> was found in DFB 0.9.25.1, and we're running with the linux_input > >> input module disabled in directfbrc. > > > > This is no longer true: > > rootskel-gtk (0.09) unstable; urgency=low > > > > * Update for compatibility with GTK 2.8 library. > > - Change default font size to 9pt. > > - Enable linux_input again as otherwise the mouse does not work > > in vmware. > > - Keep gtk config file gdk-pixbuf.loaders for now because > > file in gtk udeb is empty; remove gtk.immodules as it is no > > longer needed. > > > > It does seem however that input support for mice and touchpads is in > > general somewhat unreliable. > > so, IIUC, you re-enable loading of module linux_input , right ?
Correct. IMO the regular mouse no longer working in vmware is a bigger problem than touchpads not working. > Please note that my touchpad only works with linux_input disabled, like > many other touchpads. I thought at the time that removing that module was the wrong solution: if a module blocks other devices, then the module needs to be fixed (or the support for other devices improved); removing it can only lead to other devices not working. > Indeed, mouse detection under DFB is ometimes an headache, and i don't > know if this is related to DFB only or to missing detection of correct > kernel modules... For i386 we have the following modules available: $ cat input-modules hid ? usbhid ? usbkbd $ cat mouse-modules evdev psmouse (NB: amd64 additionally has keybdev optionally included in input-modules.) Feel free to: - check that these modules are actually loaded - check that the correct devices are present in /dev/input - experiment with other modules I'd suggest comparing what you have in the installer with what you have in an installed system running the same kernel. Checking /proc/usb/devices may help too (after mounting usbfs on /proc/bus/usb).
pgpHm3DQHMqaI.pgp
Description: PGP signature