[PATCH 3/5] ath9k_htc: Fix module unloading issue

[PATCH 3/5] ath9k_htc: Fix module unloading issue

[Sujith]

The maximum number of packets in a single buffer in
stream mode is 10. The driver currently uses 8 - which
caused stack corruption, in the absence of any kind
of OOB checking.

Fixing this to the correct value of 10 fixes the module
unload issue.

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

diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index 4528df4..69bef1d 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -324,7 +324,7 @@ static struct ath9k_htc_hif hif_usb = {
 static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev,
 				    struct sk_buff *skb)
 {
-	struct sk_buff *nskb, *skb_pool[8];
+	struct sk_buff *nskb, *skb_pool[MAX_PKT_NUM_IN_TRANSFER];
 	int index = 0, i = 0, chk_idx, len = skb->len;
 	int rx_remain_len = 0, rx_pkt_len = 0;
 	u16 pkt_len, pkt_tag, pool_index = 0;
diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.h b/drivers/net/wireless/ath/ath9k/hif_usb.h
index ed9708c..179cea4 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.h
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.h
@@ -34,6 +34,7 @@
 
 #define MAX_RX_URB_NUM  8
 #define MAX_RX_BUF_SIZE 16384
+#define MAX_PKT_NUM_IN_TRANSFER 10
 
 #define MAX_REG_OUT_URB_NUM  1
 #define MAX_REG_OUT_BUF_NUM  8
-- 
1.7.0.3

Re: [PATCH 3/5] ath9k_htc: Fix module unloading issue

[Ming Lei]

2010/4/6 Sujith <Sujith.Manoharan@atheros.com>:
> The maximum number of packets in a single buffer in
> stream mode is 10. The driver currently uses 8 - which
> caused stack corruption, in the absence of any kind
> of OOB checking.
>
> Fixing this to the correct value of 10 fixes the module
> unload issue.

Great,  the patch fixes the issue of destruction of urb->use_count,
which caused to sleep forever of usb_kill_urb().

For this kind of dark issue, it is better to add check for 'pool_index'.

Tested-by: Ming Lei <tom.leiming@gmail.com>

>
> Signed-off-by: Sujith <Sujith.Manoharan@atheros.com>
> ---
>  drivers/net/wireless/ath/ath9k/hif_usb.c |    2 +-
>  drivers/net/wireless/ath/ath9k/hif_usb.h |    1 +
>  2 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
> index 4528df4..69bef1d 100644
> --- a/drivers/net/wireless/ath/ath9k/hif_usb.c
> +++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
> @@ -324,7 +324,7 @@ static struct ath9k_htc_hif hif_usb = {
>  static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev,
>                                    struct sk_buff *skb)
>  {
> -       struct sk_buff *nskb, *skb_pool[8];
> +       struct sk_buff *nskb, *skb_pool[MAX_PKT_NUM_IN_TRANSFER];
>        int index = 0, i = 0, chk_idx, len = skb->len;
>        int rx_remain_len = 0, rx_pkt_len = 0;
>        u16 pkt_len, pkt_tag, pool_index = 0;
> diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.h b/drivers/net/wireless/ath/ath9k/hif_usb.h
> index ed9708c..179cea4 100644
> --- a/drivers/net/wireless/ath/ath9k/hif_usb.h
> +++ b/drivers/net/wireless/ath/ath9k/hif_usb.h
> @@ -34,6 +34,7 @@
>
>  #define MAX_RX_URB_NUM  8
>  #define MAX_RX_BUF_SIZE 16384
> +#define MAX_PKT_NUM_IN_TRANSFER 10
>
>  #define MAX_REG_OUT_URB_NUM  1
>  #define MAX_REG_OUT_BUF_NUM  8
> --
> 1.7.0.3
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>



-- 
Lei Ming

Re: [PATCH 3/5] ath9k_htc: Fix module unloading issue

[Sujith]

Ming Lei wrote:
> 2010/4/6 Sujith <Sujith.Manoharan@atheros.com>:
> > The maximum number of packets in a single buffer in
> > stream mode is 10. The driver currently uses 8 - which
> > caused stack corruption, in the absence of any kind
> > of OOB checking.
> >
> > Fixing this to the correct value of 10 fixes the module
> > unload issue.
> 
> Great,  the patch fixes the issue of destruction of urb->use_count,
> which caused to sleep forever of usb_kill_urb().
> 
> For this kind of dark issue, it is better to add check for 'pool_index'.

Right, it will be done in a separate patch.

> Tested-by: Ming Lei <tom.leiming@gmail.com>

Thanks !

Sujith