Re: [BUG] net: core: netif_receive_skb_list() crash on non-standard ptypes forwarding
From: Edward Cree
Date: Thu Mar 28 2019 - 08:29:44 EST
On 28/03/2019 11:56, Alexander Lobakin wrote:
> Edward Cree wrote 28.03.2019 14:37:
>> On 28/03/2019 10:03, Alexander Lobakin wrote:
>>> Here's a working possible fix for that, not sure if it can break anything
>>> though.
>>>
>>> diff --git a/net/core/dev.c b/net/core/dev.c
>>> index 2b67f2aa59dd..fdcff29df915 100644
>>> --- a/net/core/dev.c
>>> +++ b/net/core/dev.c
>>> @@ -5014,8 +5014,10 @@ static inline void __netif_receive_skb_list_ptype(struct list_head *head,
>>> ÂÂÂÂif (pt_prev->list_func != NULL)
>>> ÂÂÂÂÂÂÂ pt_prev->list_func(head, pt_prev, orig_dev);
>>> ÂÂÂÂelse
>>> -ÂÂÂÂÂÂÂ list_for_each_entry_safe(skb, next, head, list)
>>> +ÂÂÂÂÂÂÂ list_for_each_entry_safe(skb, next, head, list) {
>>> +ÂÂÂÂÂÂÂÂÂÂÂ skb_list_del_init(skb);
>>> ÂÂÂÂÂÂÂÂÂÂÂ pt_prev->func(skb, skb->dev, pt_prev, orig_dev);
>>> +ÂÂÂÂÂÂÂ }
>>> }
>>>
>>> static void __netif_receive_skb_list_core(struct list_head *head, bool pfmemalloc)
>>>
>>> Maybe you could look into this and find another/better solution (or I could
>>> submit this one if that's pretty enough).
>> That looks like the correct fix to me, go ahead and have a
>> Reviewed-by: Edward Cree <ecree@xxxxxxxxxxxxxx>
>
> Cool!
> Do we need a "stable" mark in this case to propose it to LTS backports (4.20+, I suppose)?
Probably add a
Fixes: 88eb1944e18c ("net: core: propagate SKB lists through packet_type lookup")
(then Dave will take care of sending it to -stable.)
-Ed