netdev - Re: [PATCH v2 net-next 4/9] tcp: Restrict SO

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: <6822b191d0399_104f1029490@willemb.c.googlers.com.notmuch>
Date: Mon, 12 May 2025 22:42:25 -0400
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: Kuniyuki Iwashima <kuniyu@...zon.com>, 
 willemdebruijn.kernel@...il.com
Cc: brauner@...nel.org, 
 davem@...emloft.net, 
 edumazet@...gle.com, 
 horms@...nel.org, 
 kuba@...nel.org, 
 kuni1840@...il.com, 
 kuniyu@...zon.com, 
 netdev@...r.kernel.org, 
 pabeni@...hat.com, 
 willemb@...gle.com
Subject: Re: [PATCH v2 net-next 4/9] tcp: Restrict SO_TXREHASH to TCP socket.

Kuniyuki Iwashima wrote:
> From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
> Date: Mon, 12 May 2025 15:18:12 -0400
> > Kuniyuki Iwashima wrote:
> > > sk->sk_txrehash is only used for TCP.
> > > 
> > > Let's restrict SO_TXREHASH to TCP to reflect this.
> > > 
> > > Later, we will make sk_txrehash a part of the union for other
> > > protocol families, so we set 0 explicitly in getsockopt().
> > > 
> > > Signed-off-by: Kuniyuki Iwashima <kuniyu@...zon.com>
> > > ---
> > >  net/core/sock.c | 9 +++++++--
> > >  1 file changed, 7 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/net/core/sock.c b/net/core/sock.c
> > > index b64df2463300..5c84a608ddd7 100644
> > > --- a/net/core/sock.c
> > > +++ b/net/core/sock.c
> > > @@ -1276,6 +1276,8 @@ int sk_setsockopt(struct sock *sk, int level, int optname,
> > >  		return 0;
> > >  		}
> > >  	case SO_TXREHASH:
> > > +		if (!sk_is_tcp(sk))
> > > +			return -EOPNOTSUPP;
> > >  		if (val < -1 || val > 1)
> > >  			return -EINVAL;
> > >  		if ((u8)val == SOCK_TXREHASH_DEFAULT)
> > > @@ -2102,8 +2104,11 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
> > >  		break;
> > >  
> > >  	case SO_TXREHASH:
> > > -		/* Paired with WRITE_ONCE() in sk_setsockopt() */
> > > -		v.val = READ_ONCE(sk->sk_txrehash);
> > > +		if (sk_is_tcp(sk))
> > > +			/* Paired with WRITE_ONCE() in sk_setsockopt() */
> > > +			v.val = READ_ONCE(sk->sk_txrehash);
> > > +		else
> > > +			v.val = 0;
> > 
> > Here and in the following getsockopt calls: should the call fail with
> > EOPNOTSUPP rather than return a value that is legal where the option
> > is supported (in TCP).
> 
> I was wondering which is better but didn't have preference, so will
> return -EOPNOTSUPP in v3.

It's a reminder that this is breaking an existing API.

It is unlikely to affect any real users in this case, as SO_TXREHASH
never was function for Unix sockets. But for this and subsequent such
changes we have to be aware that it is in principle a user visible
change.