Re: NeworkManager reporting incorrect signal quality (always at 100%)

[Ralf Rublack <ralf-rublack gmx de>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Xavier

I think there is a fix in madwifi release 0.9.3 
http://madwifi.org/log/releases/0.9.3

special the changlog http://madwifi.org/changeset/2612

greetings
Ralf

Am Thu, 26 Jul 2007 15:48:02 -0600
schrieb Xavier Callejas <xcalleja ibcinc com>:

> 
> Thank you Dan.
> 
> How can I help to fix this problem?
> 
> openSUSE 10.2 patch por NW+madwifi doesn't fix this problem.
> 
> El Wednesday 25 July 2007 07:47:46 Dan Williams escribió:
> > > 'iwlist' shows the right values.
> >
> > No, it doesn't.  You'll note that the Quality reported by iwlist is
> > completely bogus:
> >
> > Quality=30/70  Signal level=-65 dBm  Noise level=-95 dBm
> >
> > The first # is the difference between signal and noise, which
> > certainly doesn't conform to the WEXT specification.  Second, what
> > does the atheros driver set in the reported quality flags?  What
> > does it call valid?  NM uses the quality reported by the driver as
> > valid before it falls back to a signal/noise ratio.
> >
> > In any case, there are also OpenSUSE patches that special-case
> > madwifi signal quality reporting in NM which could also be
> > malfunctioning.
> >
> > It could be that NM is getting the calculations wrong.  However,
> > madwifi has misreported quality for so long that I'm skeptical.
> >
> > Dan
> 
> 
> 


- -- 

Ralf Rublack
University of Leipzig
Student

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGqTGEXeDgZPr3zMsRAkA0AKCqRrpLuyvSaj+hiqIpLy752igvAwCfU87O
plmQqcsjn2v8YGcjrz9S9JQ=
=k0Lc
-----END PGP SIGNATURE-----