netdev - Re: [PATCH net-next] net: add real socket cookies

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: <1426113035.11398.90.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Wed, 11 Mar 2015 15:30:35 -0700
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	David Miller <davem@...emloft.net>
Cc:	netdev@...r.kernel.org, salo@...gle.com
Subject: Re: [PATCH net-next] net: add real socket cookies

On Wed, 2015-03-11 at 17:53 -0400, David Miller wrote:
> From: Eric Dumazet <eric.dumazet@...il.com>
> Date: Tue, 10 Mar 2015 15:45:33 -0700
> 
> > diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
> > index 14d02ea905b6..a097ae38b639 100644
> > --- a/net/ipv4/inet_connection_sock.c
> > +++ b/net/ipv4/inet_connection_sock.c
> > @@ -678,7 +678,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk,
> >  		newsk->sk_write_space = sk_stream_write_space;
> >  
> >  		newsk->sk_mark = inet_rsk(req)->ir_mark;
> > -
> > +		atomic64_cmpxchg(&newsk->sk_cookie, 0,
> > +				 atomic64_read(&inet_rsk(req)->ir_cookie));
> >  		newicsk->icsk_retransmits = 0;
> >  		newicsk->icsk_backoff	  = 0;
> >  		newicsk->icsk_probes_out  = 0;
> 
> I think you have to be more careful here.
> 
> sk_clone_lock() is not going to zero out sk_cookie for you, it only
> does so if the priority has __GFP_ZERO in it and some callsites use
> just plain GFP_ATOMIC.
> 
> Therefore, you can't just assume sk_cookie is zero.
> 
> Just use atomic64_set() here, you have exclusive access to this piece
> of memory at this point in time, and you'll save an unnecessary atomic
> operation as well.

Note that sk_clone_lock() really does the clear for us :

net/core/sock.c:1541:		atomic64_set(&newsk->sk_cookie, 0);

atomic_cmpxchg() is needed here in case a concurrent dumper already
caught the new socket.

I thought of adding a 'u64 cookie' parameter to sk_clone(), but it
probably can be done in a followup patch, not sure it is worth the
complexity.

This would avoid the atomic64_cmpxchg(), but we also simply can avoid
this atomic64_cmpxchg() if inet_rsk(req)->ir_cookie is 0, as it is
probably 0 anyway.

I can send a v2 with :

u64 cookie = atomic64_read(&inet_rsk(req)->ir_cookie);
if (cookie)
	atomic64_cmpxchg(&newsk->sk_cookie, 0, cooke);

Thanks !





--
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