netdev - Re: [PATCH v2] net-tcp: Find dst with sk's xfrm policy not ctl

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] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAM2q-ny=r-U-6n6F+02QON1B8NHJ5TZrrOa7x3CAfkrUtRWnwQ@mail.gmail.com>
Date:   Wed, 6 Jul 2022 03:10:27 +0000
From:   서세욱 <ssewook@...il.com>
To:     Paolo Abeni <pabeni@...hat.com>
Cc:     Sewook Seo <sewookseo@...gle.com>,
        Linux Network Development Mailing List 
        <netdev@...r.kernel.org>, "David S . Miller" <davem@...emloft.net>,
        Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
        David Ahern <dsahern@...nel.org>,
        Jakub Kicinski <kuba@...nel.org>,
        Maciej Żenczykowski <maze@...gle.com>,
        Eric Dumazet <edumazet@...gle.com>,
        Steffen Klassert <steffen.klassert@...unet.com>,
        Sehee Lee <seheele@...gle.com>
Subject: Re: [PATCH v2] net-tcp: Find dst with sk's xfrm policy not ctl_sk

Hi, Paolo.

 If you are targting net, please add a suitable Fixes: tag.
 > I'm targeting net-next, and will update the subject.

It looks like the cloned policy will be overwrited by later resets and
possibly leaked? nobody calls xfrm_sk_free_policy() on the old policy.
> Is it possible that a later reset overwrites sk_ctl's sk_policy? I thought ctl_sk is a percpu variable and it's preempted. Maybe I might miss something, please let me know if my understanding is wrong.

Thanks.


2022년 7월 5일 (화) 오전 8:25, Paolo Abeni <pabeni@...hat.com>님이 작성:
>
> Hello,
>
> On Fri, 2022-07-01 at 15:44 +0000, Sewook Seo wrote:
> > From: sewookseo <sewookseo@...gle.com>
> >
> > If we set XFRM security policy by calling setsockopt with option
> > IPV6_XFRM_POLICY, the policy will be stored in 'sock_policy' in 'sock'
> > struct. However tcp_v6_send_response doesn't look up dst_entry with the
> > actual socket but looks up with tcp control socket. This may cause a
> > problem that a RST packet is sent without ESP encryption & peer's TCP
> > socket can't receive it.
> > This patch will make the function look up dest_entry with actual socket,
> > if the socket has XFRM policy(sock_policy), so that the TCP response
> > packet via this function can be encrypted, & aligned on the encrypted
> > TCP socket.
> >
> > Tested: We encountered this problem when a TCP socket which is encrypted
> > in ESP transport mode encryption, receives challenge ACK at SYN_SENT
> > state. After receiving challenge ACK, TCP needs to send RST to
> > establish the socket at next SYN try. But the RST was not encrypted &
> > peer TCP socket still remains on ESTABLISHED state.
> > So we verified this with test step as below.
> > [Test step]
> > 1. Making a TCP state mismatch between client(IDLE) & server(ESTABLISHED).
> > 2. Client tries a new connection on the same TCP ports(src & dst).
> > 3. Server will return challenge ACK instead of SYN,ACK.
> > 4. Client will send RST to server to clear the SOCKET.
> > 5. Client will retransmit SYN to server on the same TCP ports.
> > [Expected result]
> > The TCP connection should be established.
> >
> > Effort: net-tcp
>
> This looks like a stray "internal" tag?
>
> > Cc: Maciej Żenczykowski <maze@...gle.com>
> > Cc: Eric Dumazet <edumazet@...gle.com>
> > Cc: Steffen Klassert <steffen.klassert@...unet.com>
> > Cc: Sehee Lee <seheele@...gle.com>
> > Signed-off-by: Sewook Seo <sewookseo@...gle.com>
>
> Is this targeting -net -or -net-next? IMHO this could land in either
> trees. If you are targting net, please add a suitable Fixes: tag.
>
>
> > ---
> > Changelog since v1:
> > - Remove unnecessary null check of sk at ip_output.c
> >   Narrow down patch scope: sending RST at SYN_SENT state
> >   Remove unnecessay condition to call xfrm_sk_free_policy()
> >   Verified at KASAN build
> >
> >  net/ipv4/ip_output.c | 7 ++++++-
> >  net/ipv4/tcp_ipv4.c  | 5 +++++
> >  net/ipv6/tcp_ipv6.c  | 7 ++++++-
> >  3 files changed, 17 insertions(+), 2 deletions(-)
> >
> > diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
> > index 00b4bf26fd93..1da430c8fee2 100644
> > --- a/net/ipv4/ip_output.c
> > +++ b/net/ipv4/ip_output.c
> > @@ -1704,7 +1704,12 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb,
> >                          tcp_hdr(skb)->source, tcp_hdr(skb)->dest,
> >                          arg->uid);
> >       security_skb_classify_flow(skb, flowi4_to_flowi_common(&fl4));
> > -     rt = ip_route_output_key(net, &fl4);
> > +#ifdef CONFIG_XFRM
> > +     if (sk->sk_policy[XFRM_POLICY_OUT])
> > +             rt = ip_route_output_flow(net, &fl4, sk);
> > +     else
> > +#endif
> > +             rt = ip_route_output_key(net, &fl4);
> >       if (IS_ERR(rt))
> >               return;
> >
> > diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
> > index fda811a5251f..459669f9e13f 100644
> > --- a/net/ipv4/tcp_ipv4.c
> > +++ b/net/ipv4/tcp_ipv4.c
> > @@ -819,6 +819,10 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
> >               ctl_sk->sk_priority = (sk->sk_state == TCP_TIME_WAIT) ?
> >                                  inet_twsk(sk)->tw_priority : sk->sk_priority;
> >               transmit_time = tcp_transmit_time(sk);
> > +#ifdef CONFIG_XFRM
> > +             if (sk->sk_policy[XFRM_POLICY_OUT] && sk->sk_state == TCP_SYN_SENT)
> > +                     xfrm_sk_clone_policy(ctl_sk, sk);
> > +#endif
>
> It looks like the cloned policy will be overwrited by later resets and
> possibly leaked? nobody calls xfrm_sk_free_policy() on the old policy.
>
> Thanks!
>
> Paolo
>