[PATCH RFC net-next v4 05/12] vhost/vsock: add netns support

[Bobby Eshleman <bobbyeshleman@gmail.com>]

From: Bobby Eshleman <bobbyeshleman@meta.com>

Add the ability to isolate vsock flows using namespaces.

The namespace for a VM is inherited from the PID that opened the
vhost-vsock device.

Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
---
 drivers/vhost/vsock.c | 48 ++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 38 insertions(+), 10 deletions(-)

diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
index 802153e23073..863419533a3f 100644
--- a/drivers/vhost/vsock.c
+++ b/drivers/vhost/vsock.c
@@ -46,6 +46,8 @@ static DEFINE_READ_MOSTLY_HASHTABLE(vhost_vsock_hash, 8);
 struct vhost_vsock {
 	struct vhost_dev dev;
 	struct vhost_virtqueue vqs[2];
+	struct net *net;
+	netns_tracker ns_tracker;
 
 	/* Link to global vhost_vsock_hash, writes use vhost_vsock_mutex */
 	struct hlist_node hash;
@@ -59,6 +61,22 @@ struct vhost_vsock {
 	bool seqpacket_allow;
 };
 
+static void vhost_vsock_net_set(struct vhost_vsock *vsock, struct net *net)
+{
+	if (net_eq(net, vsock_global_net()))
+		vsock->net = vsock_global_net();
+	else
+		vsock->net = get_net_track(net, &vsock->ns_tracker, GFP_KERNEL);
+}
+
+static void vhost_vsock_net_put(struct vhost_vsock *vsock)
+{
+	if (net_eq(vsock->net, vsock_global_net()))
+		return;
+
+	put_net_track(vsock->net, &vsock->ns_tracker);
+}
+
 static u32 vhost_transport_get_local_cid(void)
 {
 	return VHOST_VSOCK_DEFAULT_HOST_CID;
@@ -67,7 +85,7 @@ static u32 vhost_transport_get_local_cid(void)
 /* Callers that dereference the return value must hold vhost_vsock_mutex or the
  * RCU read lock.
  */
-static struct vhost_vsock *vhost_vsock_get(u32 guest_cid)
+static struct vhost_vsock *vhost_vsock_get(u32 guest_cid, struct net *net)
 {
 	struct vhost_vsock *vsock;
 
@@ -78,9 +96,8 @@ static struct vhost_vsock *vhost_vsock_get(u32 guest_cid)
 		if (other_cid == 0)
 			continue;
 
-		if (other_cid == guest_cid)
+		if (other_cid == guest_cid && vsock_net_check_mode(net, vsock->net))
 			return vsock;
-
 	}
 
 	return NULL;
@@ -272,13 +289,14 @@ static int
 vhost_transport_send_pkt(struct sk_buff *skb)
 {
 	struct virtio_vsock_hdr *hdr = virtio_vsock_hdr(skb);
+	struct net *net = virtio_vsock_skb_net(skb);
 	struct vhost_vsock *vsock;
 	int len = skb->len;
 
 	rcu_read_lock();
 
 	/* Find the vhost_vsock according to guest context id  */
-	vsock = vhost_vsock_get(le64_to_cpu(hdr->dst_cid));
+	vsock = vhost_vsock_get(le64_to_cpu(hdr->dst_cid), net);
 	if (!vsock) {
 		rcu_read_unlock();
 		kfree_skb(skb);
@@ -305,7 +323,7 @@ vhost_transport_cancel_pkt(struct vsock_sock *vsk)
 	rcu_read_lock();
 
 	/* Find the vhost_vsock according to guest context id  */
-	vsock = vhost_vsock_get(vsk->remote_addr.svm_cid);
+	vsock = vhost_vsock_get(vsk->remote_addr.svm_cid, sock_net(sk_vsock(vsk)));
 	if (!vsock)
 		goto out;
 
@@ -403,7 +421,7 @@ static bool vhost_transport_msgzerocopy_allow(void)
 	return true;
 }
 
-static bool vhost_transport_seqpacket_allow(u32 remote_cid);
+static bool vhost_transport_seqpacket_allow(struct vsock_sock *vsk, u32 remote_cid);
 
 static struct virtio_transport vhost_transport = {
 	.transport = {
@@ -459,13 +477,14 @@ static struct virtio_transport vhost_transport = {
 	.send_pkt = vhost_transport_send_pkt,
 };
 
-static bool vhost_transport_seqpacket_allow(u32 remote_cid)
+static bool vhost_transport_seqpacket_allow(struct vsock_sock *vsk, u32 remote_cid)
 {
+	struct net *net = sock_net(sk_vsock(vsk));
 	struct vhost_vsock *vsock;
 	bool seqpacket_allow = false;
 
 	rcu_read_lock();
-	vsock = vhost_vsock_get(remote_cid);
+	vsock = vhost_vsock_get(remote_cid, net);
 
 	if (vsock)
 		seqpacket_allow = vsock->seqpacket_allow;
@@ -525,6 +544,7 @@ static void vhost_vsock_handle_tx_kick(struct vhost_work *work)
 			continue;
 		}
 
+		virtio_vsock_skb_set_net(skb, vsock->net);
 		total_len += sizeof(*hdr) + skb->len;
 
 		/* Deliver to monitoring devices all received packets */
@@ -651,10 +671,16 @@ static void vhost_vsock_free(struct vhost_vsock *vsock)
 
 static int vhost_vsock_dev_open(struct inode *inode, struct file *file)
 {
+
 	struct vhost_virtqueue **vqs;
 	struct vhost_vsock *vsock;
+	struct net *net;
 	int ret;
 
+	net = get_net_ns_by_pid(current->pid);
+	if (IS_ERR(net))
+		return PTR_ERR(net);
+
 	/* This struct is large and allocation could fail, fall back to vmalloc
 	 * if there is no other way.
 	 */
@@ -668,6 +694,7 @@ static int vhost_vsock_dev_open(struct inode *inode, struct file *file)
 		goto out;
 	}
 
+	vhost_vsock_net_set(vsock, net);
 	vsock->guest_cid = 0; /* no CID assigned yet */
 	vsock->seqpacket_allow = false;
 
@@ -707,7 +734,7 @@ static void vhost_vsock_reset_orphans(struct sock *sk)
 	 */
 
 	/* If the peer is still valid, no need to reset connection */
-	if (vhost_vsock_get(vsk->remote_addr.svm_cid))
+	if (vhost_vsock_get(vsk->remote_addr.svm_cid, sock_net(sk)))
 		return;
 
 	/* If the close timeout is pending, let it expire.  This avoids races
@@ -752,6 +779,7 @@ static int vhost_vsock_dev_release(struct inode *inode, struct file *file)
 	virtio_vsock_skb_queue_purge(&vsock->send_pkt_queue);
 
 	vhost_dev_cleanup(&vsock->dev);
+	vhost_vsock_net_put(vsock);
 	kfree(vsock->dev.vqs);
 	vhost_vsock_free(vsock);
 	return 0;
@@ -778,7 +806,7 @@ static int vhost_vsock_set_cid(struct vhost_vsock *vsock, u64 guest_cid)
 
 	/* Refuse if CID is already in use */
 	mutex_lock(&vhost_vsock_mutex);
-	other = vhost_vsock_get(guest_cid);
+	other = vhost_vsock_get(guest_cid, vsock->net);
 	if (other && other != vsock) {
 		mutex_unlock(&vhost_vsock_mutex);
 		return -EADDRINUSE;

-- 
2.47.3