Egbert wrote: >Mark Vojkovich writes: > > > > XFree86's mouse protocol handling doesn't seem to be able to > > handle interruptions/inconsistencies in the protocol stream that > > kvm switches cause. As far as I can tell, XFree86 has always had > > this problem. Other than switching VTs to force a the server > > to reset its state machine, I don't know of another solution. > > > >The problem are not so much intermediate inconsitencies in the protocol >but rather the fact that the mouse gets powered down and up again >by some kvms loosing all its configuration data. >One may be able to work around this problem by choosing the auto protocol. >This attempts to redetect the mouse when inconsitencies in the data stream >persist. >However I cannot tell for sure. I never had a kvm for testing when I >implemented the auto protocol. > >Egbert. i found this same problem months ago & finally got around to digging up the XF86config file. changing protocol to "auto" does help -- i notice a slight pause on the screen when i switch back to RH9, but full cursor control is restored within about a second after that. i don't know whether it matters, but i have a Belkin F1DS104T. the only other workable solution i found was (as root): service gpm restart but this one isn't terribly convenient. -- jmw _______________________________________________ XFree86 mailing list XFree86@xxxxxxxxxxx http://XFree86.Org/mailman/listinfo/xfree86
