[Nut-upsdev] recent and planned changes
Arnaud Quette
aquette.dev at gmail.com
Mon Sep 19 11:56:40 UTC 2005
2005/9/17, Charles Lepple <clepple at gmail.com>:
>
> On 9/16/05, Peter Selinger <selinger at mathstat.dal.ca> wrote:
> > Arnaud wrote:
> > > I've not had time to dig this yet (some Debian and MGE things to deal
> with),
> > > but there would be some points missing:
> > > - Index. This is a param to differentiate 2 devices with exactly the
> same
> > > info (ie same VID/ PID, no serial, ...). Then you could say "the 2nd
> device
> > > with VID xxxx and PID yyyy"
> >
> > I don't think this will work. There is no fixed meaning to "2nd
> > device". The devices could be enumerated by libusb in any order.
this is simply a dumb method to tell the 2nd device found by newhidups,
whatever the enum order is, that complies to the given params (or simply the
2nd UPS found is none params).
> > - an interesting notion that was introduced by Dave Brownell (linux USB
> > > hacker) is the "physical path" (stable ids, which don't change unless
> usb
> > > topology morphs). For more info, search for HIDIOCGPHYS in hidups.
> > > I find this interesting, but I'm not sure it has been ported...
> >
> > As per Charles' post, this is probably not possible with the current
> > backend. The best I could probably do is offer a "bus" option. As far
> > as I can tell (in Linux), while device numbers are assigned randomly,
> > busses are physical, e.g. 001, 002, etc correspond to particular fixed
> > physical USB ports on my computer. (However, I don't know how "fixed"
> > they really are - perhaps this depends on the order in which they are
> > discovered during booting?) So if someone really owns several
> > indistinguishable UPS devices, they might be able to plug them into
> > different physical ports and use the bus number to match the devices.
>
> If the user has hubs with serial numbers, and the virtual root hubs
> (the host controllers) have serial numbers (in Linux, this includes
> the PCI ID), then there is a chance of a portable solution here.
>
> On the other hand, my (totally unsolicited) recommendation for anyone
> with a need for several UPSes on the same machine is that they spend a
> couple extra dollars and get ones that report their serial number.
this was simply an old sitting idea that we should keep in mind, and maybe
lobby with libusb and kernels guys (*bsd, ...). In the long run, that would
be a nice approach, but I agree that we can't rely on it while it's not
supported everywhere or at least in libusb...
End note: I'll be a bit busy this week with some Debian and MGE stuffs.
so my answers might lag a bit...
Arnaud
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.alioth.debian.org/pipermail/nut-upsdev/attachments/20050919/7f37b6f1/attachment.html
More information about the Nut-upsdev
mailing list