Lei Chen writes: > James Carlson wrote: > >> In fact, this daemon is used to handle wireless USB device connection > >> context, while the devfsadmd manages reconfiguration and device dynamic > >> events. The new daemon needs to interact with WUSB host to setup a > >> wireless USB device's connection context. The interactions between > >> daemon and host are through ioctl. > >> > > > > Is this different from regular USB? How so? I suspect I'm missing > > some background. > > > Yes, wireless USB is a new technology which defines a protocol for > providing short-range, high-bandwidth wireless radio communication. The > wireless nature presents the first significant difference for a wusb > device from a wired one. That is, by some means defined by WUSB, > wireless host and device must first establish a secure relationship > before they can connect to each other. They have to set up a connection > context. The wikipage http://en.wikipedia.org/wiki/Wireless_USB has a > brief introduction and links.
That doesn't really answer my question, as far as I can tell. Isn't that "secure relationship" part and parcel of the whole device attachment process? In other words, I'm asking what makes this issue architecturally different from (say) enumerating USB devices attached to a hub. We don't have a separate daemon that does that process, so why is it that this new USB interconnect mechanism is fundamentally different? At a guess, it has something to do with the security relationship, but I'm having a difficult time seeing how that's a fundamental change in nature. As a way to illustrate the point: suppose we had a new kind of SCSI drive that included encryption. Suppose that drive required new commands when first connected in order to establish the key in use (the "security relationship" between the drive and host). Would we need or want a new daemon to handle that? Or would it more naturally be an extension of the existing devices framework? What I'm concerned about here is that having a separate daemon to manage a group of devices will likely lead to even more confusion in the whole device allocation and permissions area. (Have you considered whether your new daemon can run in a non-global zone or how non-global zone users get access to wireless USB nodes? Or, for that matter, how the device allocation will work with Trusted Extensions?) -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 35 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677
