Re: [PATCHv2] mac80211: dynamic wep

[Michael Wu <flamingice@sourmilk.net>]

[-- Attachment #1: Type: text/plain, Size: 2276 bytes --]

On Friday 10 August 2007 16:31, Volker Braun wrote:
> This patch fixes my problems with "dynamic wep" (widely used in
> universities), and I can now successfully associate and transfer
> data using mac80211+iwl4965. Main changes:
> 1) Allow privacy mismatch until associated
> 2) Decrypt unicast frames with the per-STA key, not making any
>    assumptions about it being key index 0.
>
> Signed-off-by: Volker Braun <volker.braun@physik.hu-berlin.de>
> ---
Hm, you somehow got a copy of this to me without actually putting me on the CC 
list.. you should probably use CC instead.

> -0400 @@ -357,7 +357,7 @@ struct ieee80211_cts {
>  #define WLAN_CAPABILITY_IBSS		(1<<1)
>  #define WLAN_CAPABILITY_CF_POLLABLE	(1<<2)
>  #define WLAN_CAPABILITY_CF_POLL_REQUEST	(1<<3)
> -#define WLAN_CAPABILITY_PRIVACY		(1<<4)
> +#define WLAN_CAPABILITY_PRIVACY		(1<<4) /* Force WEP on data packets */
Anyone interested in what this bit is can read the spec. Also, your comment is 
inaccurate, as this bit is not specific to WEP.

> +	if (rx->fc & IEEE80211_FCTL_PROTECTED && /* WEP */
Inaccurate and unnecessary comment.

> -			if (keyidx >= 0 && keyidx < NUM_DEFAULT_KEYS &&
> -			    (!rx->sta || !rx->sta->key || keyidx > 0))
> -				rx->key = rx->sdata->keys[keyidx];
So it looks like the keyidx > 0 check here is the source of the RX problems. I 
think we can store the keyidx of the individual key to compare against 
instead of assuming all individual keys have a keyidx of 0.

> -0400 @@ -1131,10 +1131,11 @@ static int ieee80211_privacy_mismatch(st
>  	bss = ieee80211_rx_bss_get(dev, ifsta->bssid);
>  	if (!bss)
>  		return 0;
> -
> -	if (ieee80211_sta_wep_configured(dev) !=
> -	    !!(bss->capability & WLAN_CAPABILITY_PRIVACY))
> -		res = 1;
> +
> +	if (ifsta->associated && ieee80211_sta_wep_configured(dev) !=
> +	    !!(bss->capability & WLAN_CAPABILITY_PRIVACY)) {
> +		res = 1;  /* associated and WEP encryption mismatch */
> +	}
Another unneeded comment. Don't add braces when it's not needed.

This part seems fairly evil. I suspect a better solution here is to allow 
unencrypted frames when the interface is dormant (netif_dormant()) and then 
we might be able to get rid of the specific check for WEP.

-Michael Wu

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]