netdev - Re: [PATCH] net: core: sock: fix information leak to userland

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]

Message-ID: <20101030144951.GA25135@albatros>
Date:	Sat, 30 Oct 2010 18:49:51 +0400
From:	Vasiliy Kulikov <segooon@...il.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	kernel-janitors@...r.kernel.org,
	"David S. Miller" <davem@...emloft.net>,
	"Eric W. Biederman" <ebiederm@...ssion.com>,
	Herbert Xu <herbert@...dor.apana.org.au>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net: core: sock: fix information leak to userland

On Sat, Oct 30, 2010 at 16:35 +0200, Eric Dumazet wrote:
> Le samedi 30 octobre 2010 à 18:26 +0400, Vasiliy Kulikov a écrit :
> > "Address" variable might be not fully initialized in sock->ops->get_name().
> > The only current implementation is get_name(), it leaves some padding
> > fields of sockaddr_tipc uninitialized.  It leads to leaking of contents
> > of kernel stack memory.
> > 
> > Signed-off-by: Vasiliy Kulikov <segooon@...il.com>
> > ---
> >  Compile tested.
> > 
> >  net/core/sock.c |    1 +
> >  1 files changed, 1 insertions(+), 0 deletions(-)
> > 
> > diff --git a/net/core/sock.c b/net/core/sock.c
> > index 3eed542..759dd81 100644
> > --- a/net/core/sock.c
> > +++ b/net/core/sock.c
> > @@ -930,6 +930,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
> >  	{
> >  		char address[128];
> >  
> > +		memset(&address, 0, sizeof(address));
> >  		if (sock->ops->getname(sock, (struct sockaddr *)address, &lv, 2))
> >  			return -ENOTCONN;
> >  		if (lv < len)
> 
> ???
> 
> Please fix the real bug.

What if somebody want to create his own implementation of getname()?
IMO it's much safer to introduce memset() here and relax getname()'s
responsibilities.  Quite many drivers "forget" to initialize outputs
structures.  E.g. new net_device's private field is kzalloc'ed to
simplify driver's code.

-- 
Vasiliy
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html