netdev - Re: [PATCH net-next 3/3] bareudp: remove unnecessary udp_encap_enable() in bareudp_socket

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 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]

Date:   Tue, 10 Mar 2020 10:17:55 +0900
From:   Taehee Yoo <ap420073@...il.com>
To:     Martin Varghese <martinvarghesenokia@...il.com>
Cc:     David Miller <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 3/3] bareudp: remove unnecessary
 udp_encap_enable() in bareudp_socket_create()

On Mon, 9 Mar 2020 at 15:03, Martin Varghese
<martinvarghesenokia@...il.com> wrote:
>

Hi Martin,
Thanek you for the review!

> On Sun, Mar 08, 2020 at 01:19:30AM +0000, Taehee Yoo wrote:
> > In the current code, udp_encap_enable() is called in
> > bareudp_socket_create().
> > But, setup_udp_tunnel_sock() internally calls udp_encap_enable().
> > So, udp_encap_enable() is unnecessary.
> >
> > Signed-off-by: Taehee Yoo <ap420073@...il.com>
> > ---
> >  drivers/net/bareudp.c | 3 ---
> >  1 file changed, 3 deletions(-)
> >
> > diff --git a/drivers/net/bareudp.c b/drivers/net/bareudp.c
> > index c9d0d68467f7..71a2f480f70e 100644
> > --- a/drivers/net/bareudp.c
> > +++ b/drivers/net/bareudp.c
> > @@ -250,9 +250,6 @@ static int bareudp_socket_create(struct bareudp_dev *bareudp, __be16 port)
> >       tunnel_cfg.encap_destroy = NULL;
> >       setup_udp_tunnel_sock(bareudp->net, sock, &tunnel_cfg);
> >
> > -     if (sock->sk->sk_family == AF_INET6)
> > -             udp_encap_enable();
> > -
> udp_encap_enable is not called for V6 sockets so we need to have the above lines of code

This patch is already merged into net-next.
So, you could send a revert patch.

In addition, I'm not familiar with the socket API.
So I'm a little bit curious about why you didn't create
separated ipv4 and ipv6 sockets?
Vxlan, geneve, etc create separated ipv4, ipv6 UDP socket.
In the bareudp modules, it creates a single socket and it tries to call
both udp_encap_enable() and udpv6_encap_enable.
Is there any special reason or both two ways are actually the same things?

Thank you so much!
Taehee Yoo