On Tue, 08 Apr 2003 01:37:17 -0500 seebs@plethora.net (Peter Seebach) wrote:

> In message <20030408154901.3c9bf8b3.ggm@apnic.net>, George Michaelson writes:
> >if I switch ps/2 mice via a dock/undock, new ps/2 behaviours like being
> >scrollable don't seem to be picked up.
> 
> This is quite possibly because many laptops have hardware that will filter
> out the scrollwheel protocol.  The PS/2 protocol for scroll mice is an utter
> monstrosity, and really *can't* reprobe on the fly; there's no way to tell.

If it can't re-probe, then being able to re-tell it is good! I'd wear the risk
of being found out as a liar to be able to try and tell the truth to good
effect.

likewise for the scsi. if some timing thing has stalled it, but umass0 is
attached, being able to tell the kernel to wire a scsi0 would be enough because
scsictl has a 're-probe the bus' function which would bind the sd0 (I think)

gee.. this is like Vax 780s and adding and removing packs on a live host. That
was a bitch to do too sometimes, always crashed for me in 4.2BSD. Never did
understand why.

-George