Re: [PATCH] rtl8187: fix use after free on failure path in rtl8187_init_urbs()

From: Hin-Tak Leung
Date: Mon Sep 02 2013 - 02:41:29 EST


------------------------------
On Mon, Sep 2, 2013 05:06 BST Alexey Khoroshilov wrote:

>On 01.09.2013 10:51, Hin-Tak Leung wrote:
>> ------------------------------
>> On Sat, Aug 31, 2013 22:18 BST Alexey Khoroshilov wrote:
>>
>> In case of __dev_alloc_skb() failure rtl8187_init_urbs()
>> calls usb_free_urb(entry) where 'entry' can points to urb
>> allocated at the previous iteration. That means refcnt will be
>> decremented incorrectly and the urb can be used after memory
>> deallocation.
>>
>> The patch fixes the issue and implements error handling of init_urbs
>> in rtl8187_start().
>>
>> Found by Linux Driver Verification project (linuxtesting.org).
>>
>> Signed-off-by: Alexey Khoroshilov <khoroshilov@xxxxxxxxx>
>> ---
>> drivers/net/wireless/rtl818x/rtl8187/dev.c | 15 ++++++++++-----
>> 1 file changed, 10 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/net/wireless/rtl818x/rtl8187/dev.c b/drivers/net/wireless/rtl818x/rtl8187/dev.c
>> index f49220e..e83d53c 100644
>> --- a/drivers/net/wireless/rtl818x/rtl8187/dev.c
>> +++ b/drivers/net/wireless/rtl818x/rtl8187/dev.c
>> @@ -438,17 +438,16 @@ static int rtl8187_init_urbs(struct ieee80211_hw *dev)
>>          skb_queue_tail(&priv->rx_queue, skb);
>>          usb_anchor_urb(entry, &priv->anchored);
>>          ret = usb_submit_urb(entry, GFP_KERNEL);
>> +        usb_free_urb(entry);
>>          if (ret) {
>>              skb_unlink(skb, &priv->rx_queue);
>>              usb_unanchor_urb(entry);
>>              goto err;
>>          }
>> -        usb_free_urb(entry);
>>      }
>>      return ret;
>>
>> err:
>> -    usb_free_urb(entry);
>>      kfree_skb(skb);
>>      usb_kill_anchored_urbs(&priv->anchored);
>>      return ret;
>> This part looks wrong - you free_urb(entry) then unanchor_urb(entry).
>I do not see any problems here.
>usb_free_urb() just decrements refcnt of the urb.
>While usb_anchor_urb() and usb_unanchor_urb() increment and decrement it
>as well.
>So actual memory deallocation will happen in usb_unanchor_urb().

If the routines work as you say, they probably are misnamed, and/or prototyped wrongly?
Also, you are making assumptions about how they are implemented, and relying
on the implementation details to be fixed for eternity.

I am just saying,

XXX_free(some_entity);
if(condtion)
do_stuff(some_entity);

looks wrong, and if that's intentional, those routines really shouldn't be named as such.

Hin-Tak



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/