[PATCH 7/7] ath9k_htc: Fix TKIP encryption

[PATCH 7/7] ath9k_htc: Fix TKIP encryption

[Sujith]

The FCS has to be stripped before the packet
is given to mac80211. Also, remove a redundant
assignment of skb length and include the FCS_LEN
when checking padding.

Fixing this issue makes TKIP work.

Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
---
 drivers/net/wireless/ath/ath9k/htc_drv_txrx.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
index befe574..b171f8e 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
@@ -450,7 +450,7 @@ static bool ath9k_rx_prepare(struct ath9k_htc_priv *priv,
 	padpos = ath9k_cmn_padpos(fc);
 
 	padsize = padpos & 3;
-	if (padsize && skb->len >= padpos+padsize) {
+	if (padsize && skb->len >= padpos+padsize+FCS_LEN) {
 		memmove(skb->data + padsize, skb->data, padpos);
 		skb_pull(skb, padsize);
 	}
@@ -537,6 +537,9 @@ static bool ath9k_rx_prepare(struct ath9k_htc_priv *priv,
 	rx_status->antenna = rxbuf->rxstatus.rs_antenna;
 	rx_status->flag |= RX_FLAG_TSFT;
 
+	/* Strip FCS */
+	skb_trim(skb, skb->len - FCS_LEN);
+
 	return true;
 
 rx_next:
@@ -646,7 +649,6 @@ void ath9k_htc_rxep(void *drv_priv, struct sk_buff *skb,
 	spin_lock(&priv->rx.rxbuflock);
 	memcpy(&rxbuf->rxstatus, rxstatus, HTC_RX_FRAME_HEADER_SIZE);
 	skb_pull(skb, HTC_RX_FRAME_HEADER_SIZE);
-	skb->len = rxstatus->rs_datalen;
 	rxbuf->skb = skb;
 	rxbuf->in_process = true;
 	spin_unlock(&priv->rx.rxbuflock);
-- 
1.7.0.3

Re: [PATCH 7/7] ath9k_htc: Fix TKIP encryption

[Christian Lamparter]

On Monday 29 March 2010 12:37:22 Sujith wrote:
> The FCS has to be stripped before the packet
> is given to mac80211. Also, remove a redundant
> assignment of skb length and include the FCS_LEN
> when checking padding.
> 
> Fixing this issue makes TKIP work.
> 
> Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
> ---
you don't need to strip the FCS, mac80211 can do that if necessary.
(Or not, if you want to operate a network sniffer)

---
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_init.c b/drivers/net/wireless/ath/ath9k/htc_drv_init.c
index 10c8760..d5e015e 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_init.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_init.c
@@ -509,6 +509,7 @@ static void ath9k_set_hw_capab(struct ath9k_htc_priv *priv,
 	struct ath_common *common = ath9k_hw_common(priv->ah);
 
 	hw->flags = IEEE80211_HW_SIGNAL_DBM |
+		IEEE80211_HW_RX_INCLUDES_FCS |
 		IEEE80211_HW_AMPDU_AGGREGATION |
 		IEEE80211_HW_SPECTRUM_MGMT |
 		IEEE80211_HW_HAS_RATE_CONTROL;


>  drivers/net/wireless/ath/ath9k/htc_drv_txrx.c |    6 ++++--
>  1 files changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
> index befe574..b171f8e 100644
> --- a/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
> +++ b/drivers/net/wireless/ath/ath9k/htc_drv_txrx.c
> @@ -646,7 +649,6 @@ void ath9k_htc_rxep(void *drv_priv, struct sk_buff *skb,
>  	spin_lock(&priv->rx.rxbuflock);
>  	memcpy(&rxbuf->rxstatus, rxstatus, HTC_RX_FRAME_HEADER_SIZE);
>  	skb_pull(skb, HTC_RX_FRAME_HEADER_SIZE);
> -	skb->len = rxstatus->rs_datalen;
>  	rxbuf->skb = skb;
>  	rxbuf->in_process = true;
>  	spin_unlock(&priv->rx.rxbuflock);
> 
ok, this one looks odd...

BTW, 

Re: [PATCH 7/7] ath9k_htc: Fix TKIP encryption

[Sujith]

Christian Lamparter wrote:
> On Monday 29 March 2010 12:37:22 Sujith wrote:
> > The FCS has to be stripped before the packet
> > is given to mac80211. Also, remove a redundant
> > assignment of skb length and include the FCS_LEN
> > when checking padding.
> > 
> > Fixing this issue makes TKIP work.
> > 
> > Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
> > ---
> you don't need to strip the FCS, mac80211 can do that if necessary.
> (Or not, if you want to operate a network sniffer)
> 

I was aware of the flag - but there was some confusion about
whether the target FW strips the FCS or not. And it looks like it doesn't.
I'll send a modified patch, thanks for the review.

Sujith