Re: [PATCH v2] wifi: ath9k: hif_usb: clean up skbs if ath9k_hif_usb_rx_stream() fails

[Toke Høiland-Jørgensen]

Fedor Pchelkin <pchelkin@xxxxxxxxx> writes:

> Syzkaller detected a memory leak of skbs in ath9k_hif_usb_rx_stream().
> While processing skbs in ath9k_hif_usb_rx_stream(), the already allocated
> skbs in skb_pool are not freed if ath9k_hif_usb_rx_stream() fails. If we
> have an incorrect pkt_len or pkt_tag, the skb is dropped and all the
> associated skb_pool buffers should be cleaned, too.
>
> Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
>
> Fixes: 6ce708f54cc8 ("ath9k: Fix out-of-bound memcpy in ath9k_hif_usb_rx_stream")
> Fixes: 44b23b488d44 ("ath9k: hif_usb: Reduce indent 1 column")
> Reported-by: syzbot+e9632e3eb038d93d6bc6@xxxxxxxxxxxxxxxxxxxxxxxxx
> Signed-off-by: Fedor Pchelkin <pchelkin@xxxxxxxxx>
> Signed-off-by: Alexey Khoroshilov <khoroshilov@xxxxxxxxx>
> ---
> v1->v2: added Reported-by tag
>
>  drivers/net/wireless/ath/ath9k/hif_usb.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
> index 1a2e0c7eeb02..d02cec114280 100644
> --- a/drivers/net/wireless/ath/ath9k/hif_usb.c
> +++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
> @@ -586,14 +586,14 @@ static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev,
>  
>  		if (pkt_tag != ATH_USB_RX_STREAM_MODE_TAG) {
>  			RX_STAT_INC(hif_dev, skb_dropped);
> -			return;
> +			goto invalid_pkt;
>  		}
>  
>  		if (pkt_len > 2 * MAX_RX_BUF_SIZE) {
>  			dev_err(&hif_dev->udev->dev,
>  				"ath9k_htc: invalid pkt_len (%x)\n", pkt_len);
>  			RX_STAT_INC(hif_dev, skb_dropped);
> -			return;
> +			goto invalid_pkt;
>  		}
>  
>  		pad_len = 4 - (pkt_len & 0x3);
> @@ -654,6 +654,11 @@ static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev,
>  				 skb_pool[i]->len, USB_WLAN_RX_PIPE);
>  		RX_STAT_INC(hif_dev, skb_completed);
>  	}
> +	return;
> +invalid_pkt:
> +	for (i = 0; i < pool_index; i++)
> +		kfree_skb(skb_pool[i]);
> +	return;

Hmm, so in the other error cases (if SKB allocation fails), we just
'goto err' and call the receive handler for the packets already in
skb_pool. Why can't we do the same here?

Also, I think there's another bug in that function, which this change
will make worse? Specifically, in the start of that function,
hif_dev->remain_skb is moved to skb_pool[0], but not cleared from
hif_dev itself. So if we then hit the invalid check and free it, the
next time the function is called, we'll get the same remain_skb pointer,
which has now been freed.

So I think we'll need to clear out hif_dev->remain_skb after moving it
to skb_pool. Care to add that as well?

-Toke