Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Ilya Maximets <i.maximets@samsung.com>
To:     netdev@vger.kernel.org
Cc:     linux-kernel@vger.kernel.org, bpf@vger.kernel.org,
        xdp-newbies@vger.kernel.org,
        "David S. Miller" <davem@davemloft.net>,
        =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= <bjorn.topel@intel.com>,
        Magnus Karlsson <magnus.karlsson@intel.com>,
        Jonathan Lemon <jonathan.lemon@gmail.com>,
        Jakub Kicinski <jakub.kicinski@netronome.com>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Ilya Maximets <i.maximets@samsung.com>
Subject: [PATCH bpf v2] xdp: fix race on generic receive path
Date:   Wed,  3 Jul 2019 15:09:16 +0300
Message-Id: <20190703120916.19973-1-i.maximets@samsung.com>
Content-Type: text/plain; charset="utf-8"
CMS-TYPE: 201P
References: <CGME20190703120922eucas1p2d97e3b994425ecdd2dadd13744ac2a77@eucas1p2.samsung.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Unlike driver mode, generic xdp receive could be triggered
by different threads on different CPU cores at the same time
leading to the fill and rx queue breakage. For example, this
could happen while sending packets from two processes to the
first interface of veth pair while the second part of it is
open with AF_XDP socket.

Need to take a lock for each generic receive to avoid race.

Fixes: c497176cb2e4 ("xsk: add Rx receive functions and poll support")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
---

Version 2:
    * spin_lock_irqsave --> spin_lock_bh.

 include/net/xdp_sock.h |  2 ++
 net/xdp/xsk.c          | 31 ++++++++++++++++++++++---------
 2 files changed, 24 insertions(+), 9 deletions(-)

diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index d074b6d60f8a..ac3c047d058c 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -67,6 +67,8 @@ struct xdp_sock {
 	 * in the SKB destructor callback.
 	 */
 	spinlock_t tx_completion_lock;
+	/* Protects generic receive. */
+	spinlock_t rx_lock;
 	u64 rx_dropped;
 };
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index a14e8864e4fa..5e0637db92ea 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -123,13 +123,17 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	u64 addr;
 	int err;
 
-	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index)
-		return -EINVAL;
+	spin_lock_bh(&xs->rx_lock);
+
+	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index) {
+		err = -EINVAL;
+		goto out_unlock;
+	}
 
 	if (!xskq_peek_addr(xs->umem->fq, &addr) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
-		xs->rx_dropped++;
-		return -ENOSPC;
+		err = -ENOSPC;
+		goto out_drop;
 	}
 
 	addr += xs->umem->headroom;
@@ -138,13 +142,21 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	memcpy(buffer, xdp->data_meta, len + metalen);
 	addr += metalen;
 	err = xskq_produce_batch_desc(xs->rx, addr, len);
-	if (!err) {
-		xskq_discard_addr(xs->umem->fq);
-		xsk_flush(xs);
-		return 0;
-	}
+	if (err)
+		goto out_drop;
+
+	xskq_discard_addr(xs->umem->fq);
+	xskq_produce_flush_desc(xs->rx);
 
+	spin_unlock_bh(&xs->rx_lock);
+
+	xs->sk.sk_data_ready(&xs->sk);
+	return 0;
+
+out_drop:
 	xs->rx_dropped++;
+out_unlock:
+	spin_unlock_bh(&xs->rx_lock);
 	return err;
 }
 
@@ -765,6 +777,7 @@ static int xsk_create(struct net *net, struct socket *sock, int protocol,
 
 	xs = xdp_sk(sk);
 	mutex_init(&xs->mutex);
+	spin_lock_init(&xs->rx_lock);
 	spin_lock_init(&xs->tx_completion_lock);
 
 	mutex_lock(&net->xdp.lock);
-- 
2.17.1