Update to net/gnome-netstatus to support new wlan system in -CURRENT

[Coleman Kane]

On Wed, 2008-04-23 at 14:13 -0400, Joe Marcus Clarke wrote:
> On Wed, 2008-04-23 at 14:04 -0400, Coleman Kane wrote:
> > On Wed, 2008-04-23 at 10:46 -0700, Sam Leffler wrote:
> > > Coleman Kane wrote:
> > > > On Wed, 2008-04-23 at 10:35 -0700, Sam Leffler wrote:
> > > >   
> > > >> Coleman Kane wrote:
> > > >>     
> > > >>> On Wed, 2008-04-23 at 13:18 -0400, Coleman Kane wrote:
> > > >>>   
> > > >>>       
> > > >>>> Hi,
> > > >>>>
> > > >>>> I put together some changes to the net/gnome-netstatus applet to allow
> > > >>>> it to detect and work with the new wlan interface system that was just
> > > >>>> introduced in CURRENT.
> > > >>>>
> > > >>>> This new code doesn't identify non-wlanN interfaces as wifi anymore.
> > > >>>>
> > > >>>> I think it may need some help in getting signal-strength detection
> > > >>>> properly using the if_ndis driver. Mine keeps telling me that the signal
> > > >>>> strength is always 100% no matter where I walk in my apt.
> > > >>>>
> > > >>>>     
> > > >>>>         
> > > >>> Yes, I do mean to actually attach the patch too.
> > > >>>
> > > >>>   
> > > >>>       
> > > >> As I told you previously keying off the ifnet name is a mistake.  What 
> > > >> you can do is check the media status and for any 802.3 interface look 
> > > >> for a backpointer in the MIB to a parent device that must be marked as 
> > > >> an 802.11 device.  Right now this can be defeated if ifnet is renamed 
> > > >> such that you cannot identify the wlanX unity # to find the MIB entry 
> > > >> under net.wlan.X.%parent.
> > > >>
> > > >> Separately you appear to have added "an", "rum", and "ray".  I'm not 
> > > >> sure what this code does but again using ifnet names is a mistake.  In 
> > > >> 7.x you can do a similar search of the net.wlan.X MIB space using the 
> > > >> %parent backpointer to identify an 802.11 device.  wlandebug.c has crude 
> > > >> code you can crib (for 7.x).
> > > >>
> > > >> I don't recall if 6.x has the backpointer in the MIB namespace to use.
> > > >>
> > > >>     Sam
> > > >>
> > > >>     
> > > >
> > > > The original code was written this way, so I was just mimicking it. Your
> > > > idea sounds a lot more stable. The addition of the "rum" and "ray"
> > > > interfaces is done by the original patch, as the upstream GNOME code
> > > > hadn't been written to expect either of those yet. The "an" device isn't
> > > > added, it's part of the surrounding text of the patch. You're looking at
> > > > a patch-of-a-patch...
> > > >
> > > > Anyhow, I'm not the original author but these changes probably make it
> > > > better now than it was before (and my netstatus applet works now). I'll
> > > > see if I can find some time to re-implement this properly over the
> > > > weekend. We might be able to get more iface info in there as well.
> > > >
> > > >   
> > > Ok, thanks.  FWIW there is a large body of 802.11-related code that 
> > > should be lifted form various parts of the src tree into a lib80211 or 
> > > similar.  For example there is now an XML database of regulatory domain 
> > > information that ifconfig uses.  UI apps could easily want to use this 
> > > info along w/ many other bits that are mostly used by ifconfig at the 
> > > moment.
> > > 
> > > I am happy to work w/ anyone that's interested in this.
> > > 
> > >     Sam
> > > 
> > 
> > Sam,
> > 
> > Maybe this would be a better application of my time...
> > 
> > I'll start looking into it this weekend...dump some info onto
> > wiki.freebsd.org....etc...
> > 
> > Then port gnome-netstatus to that in some manner (while preserving prior
> > compatibility).
> > 
> > I still recommend applying my changes to the port in the meantime, as it
> > will ensure that current users aren't twiddling their thumbs on a
> > now-broken applet.
> 
> Thanks for doing this.  g-n-s was on my TODO list after seeing the vap
> changes in -CURRENT.  At the time the patch was originally created (back
> in the 5.X days I believe), I only had an aironet device, and keying off
> device name was easiest.  It would definitely be better to modernize
> this code.
> 
> You have permission to commit this patch in the time being.
> 
> Joe

Thanks, done.

-- 
Coleman Kane
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 195 bytes
Desc: This is a digitally signed message part
Url : http://lists.freebsd.org/pipermail/freebsd-gnome/attachments/20080423/c845487e/attachment.pgp