netdev - Re: [PATCH] net/l2tp: Fix reference count leak in l2tp_udp_recv

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: <20210909091005.GB7098@katalix.com>
Date:   Thu, 9 Sep 2021 10:10:05 +0100
From:   Tom Parkin <tparkin@...alix.com>
To:     Xiyu Yang <xiyuyang19@...an.edu.cn>
Cc:     "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Cong Wang <cong.wang@...edance.com>,
        Randy Dunlap <rdunlap@...radead.org>,
        Xin Xiong <xiongx18@...an.edu.cn>,
        "Gong, Sishuai" <sishuai@...due.edu>,
        Matthias Schiffer <mschiffer@...verse-factory.net>,
        Bhaskar Chowdhury <unixbhaskar@...il.com>,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        yuanxzhang@...an.edu.cn, Xin Tan <tanxin.ctf@...il.com>
Subject: Re: [PATCH] net/l2tp: Fix reference count leak in l2tp_udp_recv_core

On  Thu, Sep 09, 2021 at 10:01:56 +0100, Tom Parkin wrote:
> On  Thu, Sep 09, 2021 at 12:32:00 +0800, Xiyu Yang wrote:
> > The reference count leak issue may take place in an error handling
> > path. If both conditions of tunnel->version == L2TP_HDR_VER_3 and the
> > return value of l2tp_v3_ensure_opt_in_linear is nonzero, the function
> > would directly jump to label invalid, without decrementing the reference
> > count of the l2tp_session object session increased earlier by
> > l2tp_tunnel_get_session(). This may result in refcount leaks.
> 
> I agree with your analysis.  Thanks for catching this!
> 

Also: I forgot to mention I think this fixes:

4522a70db7aa ("l2tp: fix reading optional fields of L2TPv3")

> > 
> > Fix this issue by decrease the reference count before jumping to the
> > label invalid.
> > 
> > Signed-off-by: Xiyu Yang <xiyuyang19@...an.edu.cn>
> > Signed-off-by: Xin Xiong <xiongx18@...an.edu.cn>
> > Signed-off-by: Xin Tan <tanxin.ctf@...il.com>
> > ---
> >  net/l2tp/l2tp_core.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
> > index 53486b162f01..93271a2632b8 100644
> > --- a/net/l2tp/l2tp_core.c
> > +++ b/net/l2tp/l2tp_core.c
> > @@ -869,8 +869,10 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
> >  	}
> >  
> >  	if (tunnel->version == L2TP_HDR_VER_3 &&
> > -	    l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr))
> > +	    l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr)) {
> > +		l2tp_session_dec_refcount(session);
> >  		goto invalid;
> > +	}
> 
> The error paths in l2tp_udp_recv_core are a bit convoluted because of
> the check (!session || !session->recv_skb) which may or may not need
> to drop a session reference if triggered.
> 
> I think it could be simplified since session->recv_skb is always set
> for all the current session types in the tree, but doing that is probably
> a little patch series on its own.
> 
> >  
> >  	l2tp_recv_common(session, skb, ptr, optr, hdrflags, length);
> >  	l2tp_session_dec_refcount(session);
> > -- 
> > 2.7.4
> > 
> 
> -- 
> Tom Parkin
> Katalix Systems Ltd
> https://katalix.com
> Catalysts for your Embedded Linux software development



-- 
Tom Parkin
Katalix Systems Ltd
https://katalix.com
Catalysts for your Embedded Linux software development

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)