It is stated that ath9k_htc_rx_msg() either frees the provided skb or
passes its management to another callback function. However, the skb is
not freed in case there is no another callback function, and Syzkaller was
able to cause a memory leak. Also minor comment fix.

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Fixes: fb9987d0f748 ("ath9k_htc: Support for AR9271 chipset.")
Reported-by: syzbot+e008dccab31bd3647609@syzkaller.appspotmail.com
Reported-by: syzbot+6692c72009680f7c4eb2@syzkaller.appspotmail.com
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
---
v1->v2: added Reported-by tag
v2->v3: use 'goto invalid' instead of freeing skb in place
v3->v4: fix lost comment

 drivers/net/wireless/ath/ath9k/htc_hst.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath9k/htc_hst.c b/drivers/net/wireless/ath/ath9k/htc_hst.c
index ca05b07a45e6..fe62ff668f75 100644
--- a/drivers/net/wireless/ath/ath9k/htc_hst.c
+++ b/drivers/net/wireless/ath/ath9k/htc_hst.c
@@ -391,7 +391,7 @@ static void ath9k_htc_fw_panic_report(struct htc_target *htc_handle,
  * HTC Messages are handled directly here and the obtained SKB
  * is freed.
  *
- * Service messages (Data, WMI) passed to the corresponding
+ * Service messages (Data, WMI) are passed to the corresponding
  * endpoint RX handlers, which have to free the SKB.
  */
 void ath9k_htc_rx_msg(struct htc_target *htc_handle,
@@ -478,6 +478,8 @@ void ath9k_htc_rx_msg(struct htc_target *htc_handle,
 		if (endpoint->ep_callbacks.rx)
 			endpoint->ep_callbacks.rx(endpoint->ep_callbacks.priv,
 						  skb, epid);
+		else
+			goto invalid;
 	}
 }
 
-- 
2.34.1

Fedor Pchelkin <pchelkin@ispras.ru> wrote:

> It is stated that ath9k_htc_rx_msg() either frees the provided skb or
> passes its management to another callback function. However, the skb is
> not freed in case there is no another callback function, and Syzkaller was
> able to cause a memory leak. Also minor comment fix.
> 
> Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
> 
> Fixes: fb9987d0f748 ("ath9k_htc: Support for AR9271 chipset.")
> Reported-by: syzbot+e008dccab31bd3647609@syzkaller.appspotmail.com
> Reported-by: syzbot+6692c72009680f7c4eb2@syzkaller.appspotmail.com
> Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
> Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
> Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
> Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>

Patch applied to ath-next branch of ath.git, thanks.

9b25e3985477 wifi: ath9k: htc_hst: free skb in ath9k_htc_rx_msg() if there is no callback function

-- 
https://patchwork.kernel.org/project/linux-wireless/patch/20230104123546.51427-1-pchelkin@ispras.ru/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

Fedor Pchelkin <pchelkin@ispras.ru> writes:

> It is stated that ath9k_htc_rx_msg() either frees the provided skb or
> passes its management to another callback function. However, the skb is
> not freed in case there is no another callback function, and Syzkaller was
> able to cause a memory leak. Also minor comment fix.
>
> Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
>
> Fixes: fb9987d0f748 ("ath9k_htc: Support for AR9271 chipset.")
> Reported-by: syzbot+e008dccab31bd3647609@syzkaller.appspotmail.com
> Reported-by: syzbot+6692c72009680f7c4eb2@syzkaller.appspotmail.com
> Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
> Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>

Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>