Hm, from my understanding, the problem is this:

* a bunch of drivers don't advertise that they do crypto stuff natively;
* net80211 however supports software encryption for other things;
* and what we're doing is somehow passing those caps all the way down
to the supplicant;
* .. that believes that it's not supported.

Now, I _think_ that maybe we should be teaching net80211 to fill in
the capabilities for what it supports before kicking them up to the
supplicant layer, since it knows if there's hardware or softwrae
support for things.

So, ok. i'll talk with jouni/rui and see if we can come up with a better fix.

Thanks!


-a


On 15 March 2014 16:26, sean bruno <sbr...@ignoranthack.me> wrote:
> Will do. Rui generated this, but it may be masking something else.
>
> http://people.freebsd.org/~rpaulo/wpa.patch
>
>
> -------- Original message --------
> From: Adrian Chadd
> Date:03/16/2014 8:18 AM (GMT+09:00)
> To: Allan Jude
> Cc: freebsd-wireless@freebsd.org,Sean Bruno
> Subject: Re: iwn(4) and WEP
>
> Please commit the wpa patch!
>
>
> -a
>
>
> On 15 March 2014 02:43, Allan Jude <free...@allanjude.com> wrote:
>> On 2014-03-15 13:46, Adrian Chadd wrote:
>>
>> Can you put this up on http or attach as .txt so I can read it in this
>> here
>> phone mail app?
>>
>> Adrian
>>
>> On Mar 14, 2014 9:41 PM, "Allan Jude" <free...@allanjude.com> wrote:
>>>
>>> On 03/14/2014 22:52, Sean Bruno wrote:
>>> > It looks like the wpa_supplicant 2.0 import now checks the
>>> > cryptocapability mask when trying to validate WEP usage.
>>> >
>>> > iwn(4) does not populate the mask with any values so wpa_supplicant
>>> > sees
>>> > a value of 0.
>>> >
>>> > I added the capabilites and can associate, but dhclient cannot get an
>>> > IP
>>> > from the AP.  I assume then, that once again, I don't know what I'm
>>> > doing.
>>> >
>>> > More or less, this is what I tried:
>>> >
>>> > Index: if_iwn.c
>>> > ===================================================================
>>> > --- if_iwn.c  (revision 263173)
>>> > +++ if_iwn.c  (working copy)
>>> > @@ -564,6 +564,8 @@
>>> >               | IEEE80211_C_PMGT              /* Station-side power
>>> > mgmt
>>> > */
>>> >               ;
>>> >
>>> > +        ic->ic_cryptocaps = IEEE80211_CRYPTO_WEP;
>>> > +
>>> >       /* Read MAC address, channels, etc from EEPROM. */
>>> >       if ((error = iwn_read_eeprom(sc, macaddr)) != 0) {
>>> >               device_printf(dev, "could not read EEPROM, error %d\n",
>>> >
>>> >
>>> > sean
>>> >
>>> > p.s. Allan Jude came up with a hack for wpa_supplicant that seems to
>>> > work, so I've bcc'd him here to get a patch out of him.
>>>
>>> Attached is my (horrible) patch that no one should use. But it is
>>> available for those of us at the conference who just need something that
>>> will work for the rest of the weekend.
>>>
>>>
>>>
>>> _______________________________________________
>>> freebsd-wireless@freebsd.org mailing list
>>> http://lists.freebsd.org/mailman/listinfo/freebsd-wireless
>>> To unsubscribe, send any mail to
>>> "freebsd-wireless-unsubscr...@freebsd.org"
>>
>>
>> Here is the web url for my patch (which no one should use):
>> http://www.allanjude.com/bsd/wpa_supplicant.wep.patch
>>
>> Here is the patch rpaulo sent us on IRC:
>> http://people.freebsd.org/~rpaulo/wpa.patch
>>
>>
_______________________________________________
freebsd-wireless@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-wireless
To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"

Reply via email to