netdev - Re: [PATCH net-next 18/32] netfilter: egress: avoid a lockdep splat

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 for Android: free password hash cracker in your pocket

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

Message-ID: <872e2596-4d7c-3b91-341c-0db3a3d7fd57@gmail.com>
Date:   Sun, 27 Feb 2022 18:13:33 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Pablo Neira Ayuso <pablo@...filter.org>,
        netfilter-devel@...r.kernel.org
Cc:     davem@...emloft.net, netdev@...r.kernel.org, kuba@...nel.org
Subject: Re: [PATCH net-next 18/32] netfilter: egress: avoid a lockdep splat


On 1/9/22 15:16, Pablo Neira Ayuso wrote:
> From: Florian Westphal <fw@...len.de>
>
> include/linux/netfilter_netdev.h:97 suspicious rcu_dereference_check() usage!
> 2 locks held by sd-resolve/1100:
>   0: ..(rcu_read_lock_bh){1:3}, at: ip_finish_output2
>   1: ..(rcu_read_lock_bh){1:3}, at: __dev_queue_xmit
>   __dev_queue_xmit+0 ..
>
> The helper has two callers, one uses rcu_read_lock, the other
> rcu_read_lock_bh().  Annotate the dereference to reflect this.
>
> Fixes: 42df6e1d221dd ("netfilter: Introduce egress hook")
> Signed-off-by: Florian Westphal <fw@...len.de>
> Signed-off-by: Pablo Neira Ayuso <pablo@...filter.org>
> ---
>   include/linux/netfilter_netdev.h | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/linux/netfilter_netdev.h b/include/linux/netfilter_netdev.h
> index b71b57a83bb4..b4dd96e4dc8d 100644
> --- a/include/linux/netfilter_netdev.h
> +++ b/include/linux/netfilter_netdev.h
> @@ -94,7 +94,7 @@ static inline struct sk_buff *nf_hook_egress(struct sk_buff *skb, int *rc,
>   		return skb;
>   #endif
>   
> -	e = rcu_dereference(dev->nf_hooks_egress);
> +	e = rcu_dereference_check(dev->nf_hooks_egress, rcu_read_lock_bh_held());
>   	if (!e)
>   		return skb;
>   


It seems other rcu_dereference() uses will also trigger lockdep splat.


nft_do_chain()

...

if (genbit)

     blob = rcu_dereference(chain->blob_gen_1);

else

    blob = rcu_dereference(chain->blob_gen_0);

I wonder how many other places will need a fix ?