Re: [PATCH v3] mac80211: remove wep dependency

["John W. Linville" <linville@tuxdriver.com>]

On Tue, Jul 20, 2010 at 12:59:48PM +0200, Johannes Berg wrote:
> On Wed, 2010-07-07 at 15:07 -0400, John W. Linville wrote:
> > The current mac80211 code assumes that WEP is always available.  If WEP
> > fails to initialize, ieee80211_register_hw will always fail.
> > 
> > In some cases (e.g. FIPS certification), the cryptography used by WEP is
> > unavailable.  However, in such cases there is no good reason why CCMP
> > encryption (or even no link level encryption) cannot be used.  So, this
> > patch removes mac80211's assumption that WEP (and TKIP) will always be
> > available for use.
> > 
> > Signed-off-by: John W. Linville <linville@tuxdriver.com>
> > ---
> > v3 -> actually post changed patch...
> > v2 -> make it safe to call ieee80211_wep_free even if ieee80211_wep_init
> > 	had failed
> 
> I think this is missing a change like this:
> 
> --- a/net/mac80211/cfg.c
> +++ b/net/mac80211/cfg.c
> @@ -1228,6 +1228,10 @@ static int ieee80211_scan(struct wiphy *wiphy,
>  static int ieee80211_auth(struct wiphy *wiphy, struct net_device *dev,
>  			  struct cfg80211_auth_request *req)
>  {
> +	if (req->auth_type == NL80211_AUTHTYPE_SHARED_KEY &&
> +	    IS_ERR(sdata->local->wep_tx_tfm))
> +		return -EINVAL;
> +
>  	return ieee80211_mgd_auth(IEEE80211_DEV_TO_SUB_IF(dev), req);
>  }
>  
> 
> since otherwise you could end up trying to encrypt the auth frame with
> the uninitialised WEP TFM?

Yes, I think that makes sense.  The check in ieee80211_wep_encrypt_data
will make it safe either way, but that will trigger a warning
in ieee80211_send_auth and leave us unable to return an error to
the ieee80211_auth caller.  Also, it will send the auth request
unencrypted...

Care to send that as a "real" patch? :-)

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.