rt2x00 driver

[Dan Williams]

On Mon, 2007-04-09 at 15:58 +0200, KH KH wrote:
> 2007/4/9, Rafael Espíndola <rafael.espindola at gmail.com>:
> > I am having problems with the rt2x00 driver present in the kernel
> > 3045. I have an usb device (Belink f5d7050, fcc id: k7s-f5d7050b).
> >
> > When I plug the device the driver loads correctly:
> >
> > usb 5-1: new high speed USB device using ehci_hcd and address 7
> > usb 5-1: configuration #1 chosen from 1 choice
> > Loading module: rt2x00lib - CVS (N/A) by http://rt2x00.serialmonkey.com.
> > Loading module: rt73usb - CVS (N/A) by http://rt2x00.serialmonkey.com.
> > usbcore: registered new interface driver rt73usb
> > Loading module: rt2500usb - CVS (N/A) by http://rt2x00.serialmonkey.com.
> > usbcore: registered new interface driver rt2500usb
> > wmaster0: Selected rate control algorithm 'simple'
> Ralink usb devices can use rt2500usb or rt73usb (you "may" need to
> uses rt71w-firmware currently in review with the second case!). The
> first step should be to blacklist one or the other doing: echo
> "blacklist rt2500usb" >> /etc/modprobe.d/blacklist
> > iwlist works despite the warning:
> >
> > iwlist wlan0 scan
> Same result with my rt2500pci, the problem is association is failling.
> Sometime you can have better result using rausb0 or ra0 aliases...
> (this has been rapported sometime - not my own experience...)
> > Warning: Driver for device wlan0 has been compiled with version 22
> > of Wireless Extension, while this program supports up to version 20.
> > Some things may be broken...
> Same warning! But since this may prevent to associate correctly with
> Ralink's devices we should consider this as a bug!
> > I have the latest wireless-tools installed. Should I recompile it?
> The Wireless-tools from devel is tha same version as Fedora Core 6,
> this is a problem because this site:
> http://www.hpl.hp.com/personal/Jean_Tourrilhes/Linux/Tools.html#latest
> do not mention WE-22 but beta WE-21, and i down't know where can i
> find tools that supports WE-22 (any link or references?)
> > The problem starts when running wpa_suppilcant:
> >
> > /usr/sbin/wpa_supplicant -B -i wlan0 -D wext -c /et plicant/wpa_supplicant.conf
> > ioctl[SIOCSIWMODE]: Device or resource busy
> > Could not configure driver to use managed mode
> > ioctl[SIOCSIWAUTH]: Operation not supported
> > WEXT auth param 4 value 0x0 - ioctl[SIOCSIWAUTH]: Operation not support
> >
> > Does someone knows what is wrong? What version of rt2x00 is included
> > in the kernel? Do think I should compile a new version? Should I add
> > the dscape patch to wpa_supplicant? According to the rt2x00 site newer
> > versions should work with the wext driver...
> This implies that wpa-supplicant may be updated, i'm only working on
> hostapd for now and i  didn't received answears for this bug,
> requesting it: https://bugzilla.redhat.com/230857

Updated.  We shouldn't be adding new drivers to wpa_supplicant, but
fixing the kernel drivers to correctly implement WEXT.

mac80211 doesn't work well enough with WEXT yet, but work is ongoing in
that area.  mac80211 is still an experimental stack and in no way
'stable'.  You may experience problems while using it with tools other
than 'iwconfig', or even with 'iwconfig'.

Dan

> This was before the 80211 stack named change (from d80211 to mac80211 ).
> But since i didn't recevied any answear may we consider this package
> as orphaned ?!
> > Thanks,
> > Rafael
> 
> Rafael if you could test hostapd it would help, but for now the
> package isn't ready for inclusion in my view (mostly because of
> headers not in kernel)...
> I will try to work on wpa_supplicant update unless someone take it...
> 
> Also i've read some link from ivd the main dev for rt2x00 who said
> that rt2x00 do not seems to be ready for 2.6.21. But solving all
> theses issues may help to have working wireless mac80211 devices "out
> of the box" (dreams)
> 
> Nicolas (kwizart)
> http://kwizart.free.fr/blog
> 
> Some references :
> https://bugzilla.redhat.com/230731
> https://bugzilla.redhat.com/230449
> https://bugzilla.redhat.com/230164
> 
> I was working on legacies version for some time now but they are going
> more bad each day...The paradox is that thoses devices are privilegied
> by FSF because of legal conformity with GPL...
>