Merge branch 'tun-allow-to-attach-eBPF-filter'

Jason Wang says:

====================
tun: allow to attach eBPF filter

This series tries to implement eBPF socket filter for tun. This could
be used for implementing efficient virtio-net receive filter for
vhost-net.

Changes from V2:
- fix typo
- remove unnecessary double check
====================

Signed-off-by: David S. Miller <[email protected]>

Author: davem330

drivers/net/tun.c

@@ -196,7 +196,7 @@ struct tun_flow_entry {
 
 #define TUN_NUM_FLOW_ENTRIES 1024
 
-struct tun_steering_prog {
+struct tun_prog {
 	struct rcu_head rcu;
 	struct bpf_prog *prog;
 };
@@ -238,7 +238,13 @@ struct tun_struct {
 	u32 rx_batched;
 	struct tun_pcpu_stats __percpu *pcpu_stats;
 	struct bpf_prog __rcu *xdp_prog;
-	struct tun_steering_prog __rcu *steering_prog;
+	struct tun_prog __rcu *steering_prog;
+	struct tun_prog __rcu *filter_prog;
+};
+
+struct veth {
+	__be16 h_vlan_proto;
+	__be16 h_vlan_TCI;
 };
 
 bool tun_is_xdp_buff(void *ptr)
@@ -590,7 +596,7 @@ static u16 tun_automq_select_queue(struct tun_struct *tun, struct sk_buff *skb)
 
 static u16 tun_ebpf_select_queue(struct tun_struct *tun, struct sk_buff *skb)
 {
-	struct tun_steering_prog *prog;
+	struct tun_prog *prog;
 	u16 ret = 0;
 
 	prog = rcu_dereference(tun->steering_prog);
@@ -1036,12 +1042,25 @@ static void tun_automq_xmit(struct tun_struct *tun, struct sk_buff *skb)
 #endif
 }
 
+static unsigned int run_ebpf_filter(struct tun_struct *tun,
+				    struct sk_buff *skb,
+				    int len)
+{
+	struct tun_prog *prog = rcu_dereference(tun->filter_prog);
+
+	if (prog)
+		len = bpf_prog_run_clear_cb(prog->prog, skb);
+
+	return len;
+}
+
 /* Net device start xmit */
 static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct tun_struct *tun = netdev_priv(dev);
 	int txq = skb->queue_mapping;
 	struct tun_file *tfile;
+	int len = skb->len;
 
 	rcu_read_lock();
 	tfile = rcu_dereference(tun->tfiles[txq]);
@@ -1067,6 +1086,15 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 	    sk_filter(tfile->socket.sk, skb))
 		goto drop;
 
+	len = run_ebpf_filter(tun, skb, len);
+
+	/* Trim extra bytes since we may insert vlan proto & TCI
+	 * in tun_put_user().
+	 */
+	len -= skb_vlan_tag_present(skb) ? sizeof(struct veth) : 0;
+	if (len <= 0 || pskb_trim(skb, len))
+		goto drop;
+
 	if (unlikely(skb_orphan_frags_rx(skb, GFP_ATOMIC)))
 		goto drop;
 
@@ -2054,10 +2082,7 @@ static ssize_t tun_put_user(struct tun_struct *tun,
 
 	if (vlan_hlen) {
 		int ret;
-		struct {
-			__be16 h_vlan_proto;
-			__be16 h_vlan_TCI;
-		} veth;
+		struct veth veth;
 
 		veth.h_vlan_proto = skb->vlan_proto;
 		veth.h_vlan_TCI = htons(skb_vlan_tag_get(skb));
@@ -2184,19 +2209,18 @@ static ssize_t tun_chr_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	return ret;
 }
 
-static void tun_steering_prog_free(struct rcu_head *rcu)
+static void tun_prog_free(struct rcu_head *rcu)
 {
-	struct tun_steering_prog *prog = container_of(rcu,
-					 struct tun_steering_prog, rcu);
+	struct tun_prog *prog = container_of(rcu, struct tun_prog, rcu);
 
 	bpf_prog_destroy(prog->prog);
 	kfree(prog);
 }
 
-static int __tun_set_steering_ebpf(struct tun_struct *tun,
-				   struct bpf_prog *prog)
+static int __tun_set_ebpf(struct tun_struct *tun, struct tun_prog **prog_p,
+			  struct bpf_prog *prog)
 {
-	struct tun_steering_prog *old, *new = NULL;
+	struct tun_prog *old, *new = NULL;
 
 	if (prog) {
 		new = kmalloc(sizeof(*new), GFP_KERNEL);
@@ -2206,13 +2230,13 @@ static int __tun_set_steering_ebpf(struct tun_struct *tun,
 	}
 
 	spin_lock_bh(&tun->lock);
-	old = rcu_dereference_protected(tun->steering_prog,
+	old = rcu_dereference_protected(*prog_p,
 					lockdep_is_held(&tun->lock));
-	rcu_assign_pointer(tun->steering_prog, new);
+	rcu_assign_pointer(*prog_p, new);
 	spin_unlock_bh(&tun->lock);
 
 	if (old)
-		call_rcu(&old->rcu, tun_steering_prog_free);
+		call_rcu(&old->rcu, tun_prog_free);
 
 	return 0;
 }
@@ -2225,7 +2249,8 @@ static void tun_free_netdev(struct net_device *dev)
 	free_percpu(tun->pcpu_stats);
 	tun_flow_uninit(tun);
 	security_tun_dev_free_security(tun->security);
-	__tun_set_steering_ebpf(tun, NULL);
+	__tun_set_ebpf(tun, &tun->steering_prog, NULL);
+	__tun_set_ebpf(tun, &tun->filter_prog, NULL);
 }
 
 static void tun_setup(struct net_device *dev)
@@ -2720,7 +2745,8 @@ static int tun_set_queue(struct file *file, struct ifreq *ifr)
 	return ret;
 }
 
-static int tun_set_steering_ebpf(struct tun_struct *tun, void __user *data)
+static int tun_set_ebpf(struct tun_struct *tun, struct tun_prog **prog_p,
+			void __user *data)
 {
 	struct bpf_prog *prog;
 	int fd;
@@ -2736,7 +2762,7 @@ static int tun_set_steering_ebpf(struct tun_struct *tun, void __user *data)
 			return PTR_ERR(prog);
 	}
 
-	return __tun_set_steering_ebpf(tun, prog);
+	return __tun_set_ebpf(tun, prog_p, prog);
 }
 
 static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
@@ -3016,7 +3042,11 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
 		break;
 
 	case TUNSETSTEERINGEBPF:
-		ret = tun_set_steering_ebpf(tun, argp);
+		ret = tun_set_ebpf(tun, &tun->steering_prog, argp);
+		break;
+
+	case TUNSETFILTEREBPF:
+		ret = tun_set_ebpf(tun, &tun->filter_prog, argp);
 		break;
 
 	default:

include/uapi/linux/if_tun.h

@@ -58,6 +58,7 @@
 #define TUNSETVNETBE _IOW('T', 222, int)
 #define TUNGETVNETBE _IOR('T', 223, int)
 #define TUNSETSTEERINGEBPF _IOR('T', 224, int)
+#define TUNSETFILTEREBPF _IOR('T', 225, int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001