netdev - Re: [PATCH net] vlan: consolidate VLAN parsing code and limit max parsing depth

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 7 Jul 2020 20:01:49 +0900
From:   Toshiaki Makita <toshiaki.makita1@...il.com>
To:     Toke Høiland-Jørgensen <toke@...hat.com>
Cc:     davem@...emloft.net, netdev@...r.kernel.org,
        cake@...ts.bufferbloat.net, Davide Caratti <dcaratti@...hat.com>,
        Jiri Pirko <jiri@...nulli.us>,
        Jamal Hadi Salim <jhs@...atatu.com>,
        Cong Wang <xiyou.wangcong@...il.com>,
        Daniel Borkmann <daniel@...earbox.net>
Subject: Re: [PATCH net] vlan: consolidate VLAN parsing code and limit max
 parsing depth

On 2020/07/07 19:57, Toke Høiland-Jørgensen wrote:
> Toshiaki Makita <toshiaki.makita1@...il.com> writes:
> 
>> On 2020/07/06 21:29, Toke Høiland-Jørgensen wrote:
>>> Toshiaki pointed out that we now have two very similar functions to extract
>>> the L3 protocol number in the presence of VLAN tags. And Daniel pointed out
>>> that the unbounded parsing loop makes it possible for maliciously crafted
>>> packets to loop through potentially hundreds of tags.
>>>
>>> Fix both of these issues by consolidating the two parsing functions and
>>> limiting the VLAN tag parsing to an arbitrarily-chosen, but hopefully
>>> conservative, max depth of 32 tags. As part of this, switch over
>>> __vlan_get_protocol() to use skb_header_pointer() instead of
>>> pskb_may_pull(), to avoid the possible side effects of the latter and keep
>>> the skb pointer 'const' through all the parsing functions.
>>>
>>> Reported-by: Toshiaki Makita <toshiaki.makita1@...il.com>
>>> Reported-by: Daniel Borkmann <daniel@...earbox.net>
>>> Fixes: d7bf2ebebc2b ("sched: consistently handle layer3 header accesses in the presence of VLANs")
>>> Signed-off-by: Toke Høiland-Jørgensen <toke@...hat.com>
>>> ---
>> ...
>>> @@ -623,13 +597,12 @@ static inline __be16 __vlan_get_protocol(struct sk_buff *skb, __be16 type,
>>>    			vlan_depth = ETH_HLEN;
>>>    		}
>>>    		do {
>>> -			struct vlan_hdr *vh;
>>> +			struct vlan_hdr vhdr, *vh;
>>>    
>>> -			if (unlikely(!pskb_may_pull(skb,
>>> -						    vlan_depth + VLAN_HLEN)))
>>> +			vh = skb_header_pointer(skb, vlan_depth, sizeof(vhdr), &vhdr);
>>
>> Some drivers which use vlan_get_protocol to get IP protocol for checksum offload discards
>> packets when it cannot get the protocol.
>> I guess for such users this function should try to get protocol even if it is not in skb header?
>> I'm not sure such a case can happen, but since you care about this, you know real cases where
>> vlan tag can be in skb frags?
> 
> skb_header_pointer() will still succeed in reading the data, it'll just
> do so by copying it into the buffer on the stack (vhdr) instead of
> moving the SKB data itself around...

True, probably I need some more coffee...
Thanks.

Toshiaki Makita