[Patch bpf] udp: validate checksum in udp_read_sock()

[Patch bpf] udp: validate checksum in udp_read_sock()

[Cong Wang]

From: Cong Wang <cong.wang@bytedance.com>

It turns out the skb's in sock receive queue could have
bad checksums, as both ->poll() and ->recvmsg() validate
checksums. We have to do the same for ->read_sock() path
too before they are redirected in sockmap.

Fixes: d7f571188ecf ("udp: Implement ->read_sock() for sockmap")
Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Reported-by: John Fastabend <john.fastabend@gmail.com>
Cc: Lorenz Bauer <lmb@cloudflare.com>
Cc: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
---
 net/ipv4/udp.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 8536b2a7210b..0ae8ab5e05b4 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1808,6 +1808,17 @@ int udp_read_sock(struct sock *sk, read_descriptor_t *desc,
 		skb = skb_recv_udp(sk, 0, 1, &err);
 		if (!skb)
 			return err;
+
+		if (udp_lib_checksum_complete(skb)) {
+			__UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS,
+					IS_UDPLITE(sk));
+			__UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS,
+					IS_UDPLITE(sk));
+			atomic_inc(&sk->sk_drops);
+			kfree_skb(skb);
+			continue;
+		}
+
 		used = recv_actor(desc, skb, 0, skb->len);
 		if (used <= 0) {
 			if (!copied)
-- 
2.30.2

RE: [Patch bpf] udp: validate checksum in udp_read_sock()

[John Fastabend]

Cong Wang wrote:
> From: Cong Wang <cong.wang@bytedance.com>
> 
> It turns out the skb's in sock receive queue could have
> bad checksums, as both ->poll() and ->recvmsg() validate
> checksums. We have to do the same for ->read_sock() path
> too before they are redirected in sockmap.
> 
> Fixes: d7f571188ecf ("udp: Implement ->read_sock() for sockmap")
> Reported-by: Daniel Borkmann <daniel@iogearbox.net>
> Reported-by: John Fastabend <john.fastabend@gmail.com>
> Cc: Lorenz Bauer <lmb@cloudflare.com>
> Cc: Jakub Sitnicki <jakub@cloudflare.com>
> Signed-off-by: Cong Wang <cong.wang@bytedance.com>
> ---
>  net/ipv4/udp.c | 11 +++++++++++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> index 8536b2a7210b..0ae8ab5e05b4 100644
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
> @@ -1808,6 +1808,17 @@ int udp_read_sock(struct sock *sk, read_descriptor_t *desc,
>  		skb = skb_recv_udp(sk, 0, 1, &err);
>  		if (!skb)
>  			return err;
> +
> +		if (udp_lib_checksum_complete(skb)) {
> +			__UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS,
> +					IS_UDPLITE(sk));
> +			__UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS,
> +					IS_UDPLITE(sk));
> +			atomic_inc(&sk->sk_drops);
> +			kfree_skb(skb);

We could use sock_drop() here? Otherwise looks good thanks.

> +			continue;
> +		}
> +
>  		used = recv_actor(desc, skb, 0, skb->len);
>  		if (used <= 0) {
>  			if (!copied)
> -- 
> 2.30.2
> 

Re: [Patch bpf] udp: validate checksum in udp_read_sock()

[Cong Wang]

On Mon, Oct 11, 2021 at 10:06 PM John Fastabend
<john.fastabend@gmail.com> wrote:
>
> Cong Wang wrote:
> > From: Cong Wang <cong.wang@bytedance.com>
> >
> > It turns out the skb's in sock receive queue could have
> > bad checksums, as both ->poll() and ->recvmsg() validate
> > checksums. We have to do the same for ->read_sock() path
> > too before they are redirected in sockmap.
> >
> > Fixes: d7f571188ecf ("udp: Implement ->read_sock() for sockmap")
> > Reported-by: Daniel Borkmann <daniel@iogearbox.net>
> > Reported-by: John Fastabend <john.fastabend@gmail.com>
> > Cc: Lorenz Bauer <lmb@cloudflare.com>
> > Cc: Jakub Sitnicki <jakub@cloudflare.com>
> > Signed-off-by: Cong Wang <cong.wang@bytedance.com>
> > ---
> >  net/ipv4/udp.c | 11 +++++++++++
> >  1 file changed, 11 insertions(+)
> >
> > diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> > index 8536b2a7210b..0ae8ab5e05b4 100644
> > --- a/net/ipv4/udp.c
> > +++ b/net/ipv4/udp.c
> > @@ -1808,6 +1808,17 @@ int udp_read_sock(struct sock *sk, read_descriptor_t *desc,
> >               skb = skb_recv_udp(sk, 0, 1, &err);
> >               if (!skb)
> >                       return err;
> > +
> > +             if (udp_lib_checksum_complete(skb)) {
> > +                     __UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS,
> > +                                     IS_UDPLITE(sk));
> > +                     __UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS,
> > +                                     IS_UDPLITE(sk));
> > +                     atomic_inc(&sk->sk_drops);
> > +                     kfree_skb(skb);
>
> We could use sock_drop() here? Otherwise looks good thanks.

sock_drop() is in include/linux/skmsg.h, I think we need to move it
to sock.h before using it here in net/ipv4/udp.c, right?

And there are other similar patterns which can be replaced with
sock_drop(), so we can do the replacement for all in a separate
patch.

Thanks.

Re: [Patch bpf] udp: validate checksum in udp_read_sock()

[John Fastabend]

Cong Wang wrote:
> On Mon, Oct 11, 2021 at 10:06 PM John Fastabend
> <john.fastabend@gmail.com> wrote:
> >
> > Cong Wang wrote:
> > > From: Cong Wang <cong.wang@bytedance.com>
> > >
> > > It turns out the skb's in sock receive queue could have
> > > bad checksums, as both ->poll() and ->recvmsg() validate
> > > checksums. We have to do the same for ->read_sock() path
> > > too before they are redirected in sockmap.
> > >
> > > Fixes: d7f571188ecf ("udp: Implement ->read_sock() for sockmap")
> > > Reported-by: Daniel Borkmann <daniel@iogearbox.net>
> > > Reported-by: John Fastabend <john.fastabend@gmail.com>
> > > Cc: Lorenz Bauer <lmb@cloudflare.com>
> > > Cc: Jakub Sitnicki <jakub@cloudflare.com>
> > > Signed-off-by: Cong Wang <cong.wang@bytedance.com>
> > > ---
> > >  net/ipv4/udp.c | 11 +++++++++++
> > >  1 file changed, 11 insertions(+)
> > >
> > > diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> > > index 8536b2a7210b..0ae8ab5e05b4 100644
> > > --- a/net/ipv4/udp.c
> > > +++ b/net/ipv4/udp.c
> > > @@ -1808,6 +1808,17 @@ int udp_read_sock(struct sock *sk, read_descriptor_t *desc,
> > >               skb = skb_recv_udp(sk, 0, 1, &err);
> > >               if (!skb)
> > >                       return err;
> > > +
> > > +             if (udp_lib_checksum_complete(skb)) {
> > > +                     __UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS,
> > > +                                     IS_UDPLITE(sk));
> > > +                     __UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS,
> > > +                                     IS_UDPLITE(sk));
> > > +                     atomic_inc(&sk->sk_drops);
> > > +                     kfree_skb(skb);
> >
> > We could use sock_drop() here? Otherwise looks good thanks.
> 
> sock_drop() is in include/linux/skmsg.h, I think we need to move it
> to sock.h before using it here in net/ipv4/udp.c, right?

Yes it would be necessary. Lets not do it here otherwise backports
will be ugly.

Acked-by: John Fastabend <john.fastabend@gmail.com>

> 
> And there are other similar patterns which can be replaced with
> sock_drop(), so we can do the replacement for all in a separate
> patch.
> 
> Thanks.