ath9k: Fix bug in validating received data length for edma
authorVasanthakumar Thiagarajan <vasanth@atheros.com>
Thu, 20 May 2010 21:34:48 +0000 (14:34 -0700)
committerJohn W. Linville <linville@tuxdriver.com>
Wed, 2 Jun 2010 20:13:32 +0000 (16:13 -0400)
The rx status length should also be taken into account while
validating the length of a received frame.

Signed-off-by: Vasanthakumar Thiagarajan <vasanth@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
drivers/net/wireless/ath/ath9k/recv.c

index 978b4d9..1618f8f 100644 (file)
@@ -830,6 +830,7 @@ static bool ath9k_rx_accept(struct ath_common *common,
 {
        struct ath_hw *ah = common->ah;
        __le16 fc;
+       u8 rx_status_len = ah->caps.rx_status_len;
 
        fc = hdr->frame_control;
 
@@ -840,7 +841,7 @@ static bool ath9k_rx_accept(struct ath_common *common,
          * we can take a hint that hardware corrupted it, so ignore
          * those frames.
          */
-       if (rx_stats->rs_datalen > common->rx_bufsize)
+       if (rx_stats->rs_datalen > (common->rx_bufsize - rx_status_len))
                return false;
 
        /*