netdev - Re: [PATCH] net: skb content must be visible for lockless skb

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 PHC
Open Source and information security mailing list archives

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

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

Date:   Mon, 1 Aug 2022 21:45:13 +0300
From:   Kirill Tkhai <tkhai@...ru>
To:     Eric Dumazet <edumazet@...gle.com>
Cc:     Paolo Abeni <pabeni@...hat.com>,
        "David S. Miller" <davem@...emloft.net>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>
Subject: Re: [PATCH] net: skb content must be visible for lockless skb_peek()
 and its variations

On 01.08.2022 10:39, Eric Dumazet wrote:
> On Mon, Aug 1, 2022 at 9:00 AM Kirill Tkhai <tkhai@...ru> wrote:
>>
>> On 01.08.2022 09:52, Paolo Abeni wrote:
>>> On Sun, 2022-07-31 at 23:39 +0300, Kirill Tkhai wrote:
>>>> From: Kirill Tkhai <tkhai@...ru>
>>>>
>>>> Currently, there are no barriers, and skb->xxx update may become invisible on cpu2.
>>>> In the below example var2 may point to intial_val0 instead of expected var1:
>>>>
>>>> [cpu1]                                       [cpu2]
>>>> skb->xxx = initial_val0;
>>>> ...
>>>> skb->xxx = var1;                     skb = READ_ONCE(prev_skb->next);
>>>> <no barrier>                         <no barrier>
>>>> WRITE_ONCE(prev_skb->next, skb);     var2 = skb->xxx;
>>>>
>>>> This patch adds barriers and fixes the problem. Note, that __skb_peek() is not patched,
>>>> since it's a lowlevel function, and a caller has to understand the things it does (and
>>>> also __skb_peek() is used under queue lock in some places).
>>>>
>>>> Signed-off-by: Kirill Tkhai <tkhai@...ru>
>>>> ---
>>>> Hi, David, Eric and other developers,
>>>>
>>>> picking unix sockets code I found this problem,
>>>
>>> Could you please report exactly how/where the problem maifests (e.g.
>>> the involved call paths/time sequence)?
>>
>> I didn't get why call paths in the patch description are not enough for you. Please, explain
>> what you want.
>>
>>>> and for me it looks like it exists. If there
>>>> are arguments that everything is OK and it's expected, please, explain.
>>>
>>> I don't see why such barriers are needed for the locked peek/tail
>>> variants, as the spin_lock pair implies a full memory barrier.
>>
>> This is for lockless skb_peek() calls and the patch is called in that way :). For locked skb_peek()
>> this is not needed. I'm not sure we need separate skb_peek() and skb_peek_lockless(). Do we?
> 
> We prefer explicit _lockless variants to document the precise points
> they are needed.
> 
> A new helper (and its initial usage) will clearly point to the problem
> you saw in af_unix.

The problem is:

unix_stream_sendmsg()	unix_stream_read_generic()
  skb->len = size;	  skb = skb_peek();
  skb_queue_tail(skb);	  unix_skb_len(skb); <- here we read wrong len

 
> BTW, smp_mb__after_spinlock() in your patch does not really make sense to me.
> Please add in your changelog the precise issue you are seeing.

What is about queue part, do you recommend to have a separate helper for that?
skb_queue_tail_for_lockless()?