[patch] wireless network support

[Joe Shaw]

Joe answers your questions!

On Wed, 2004-05-26 at 11:15 +0200, David Zeuthen wrote:
> Yay, I think this is quite nice. One could wonder whether this is
> stretching what the HAL should do; my take on that is that it makes
> good sense to do this; I mean, it's the same thing we do for wired
> networks with the property net.ethernet.link. I'm curious what other
> people think?

At first I wasn't so sure, but once I found that doing scanning of
wireless networks requires root for almost all the drivers, I think it
makes a lot more sense now.

> My only concern is that might cause a storm of D-BUS signal emissions
> and callouts; Kay talked about adding attributes to properties so
> perhaps one day we could say something like
> 
>  hal_property_set_attribute (d, "net.wireless.network0.noise", 
>                              HAL_PROP_ATTR_NO_CALLOUT|
>                              HAL_PROP_ATTR_NO_MASS_EMISSION);
> 
> meaning that D-BUS clients only get updates if they actively watch
> the device, and there will be no callouts in property.d from changes
> to this property.

Yeah.  Right now it's scanning every 10 seconds.  That's too often for
callouts, but in my experience the emission of dbus messages isn't too
overwhelming.

> Of course, this would require API-changes in libhal and in the D-BUS
> network API offered. And we'd have to check that D-BUS can do this,
> cf. the signal emission rules in the system message bus.

I don't know if you need to expose it to libhal.  It seems like
something that only devices inside the daemon would want to bother with.

> I think it would be rather unconstructive to duplicate code for the
> sake of removing dependencies (and iwlib isn't really a policital
> target like some other deps), so can't we just make this a
> compile-time option? If distribution vendors are supporting wireless
> ethernet they sure have iwlib available anyway.

Yes, I agree.

> My only other comment is that the properties should be called
> net.ethernet.wireless instead of just net.wireless, since other
> wireless networking technologies may or may not exist.
> 
> >   net.wireless = true  (bool)
> 
> Can we call this net.ethernet.is_wireless instead to follow what we do
> (or should do, I'll check when updating the spec which is RSN :-]) in
> other parts of the HAL?

Yes.  I'll also add net.ethernet.wireless as a capability as Kristian
suggested.

Joe


_______________________________________________
hal mailing list
hal at freedesktop.org
http://freedesktop.org/mailman/listinfo/hal