summaryrefslogtreecommitdiff
path: root/drivers/net/tun.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/net/tun.c')
-rw-r--r--drivers/net/tun.c147
1 files changed, 114 insertions, 33 deletions
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index b52258c327d2..28583aa0c17d 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -78,6 +78,7 @@
#include <linux/mutex.h>
#include <linux/uaccess.h>
+#include <linux/proc_fs.h>
/* Uncomment to enable debugging */
/* #define TUN_DEBUG 1 */
@@ -181,7 +182,6 @@ struct tun_file {
struct tun_struct *detached;
struct ptr_ring tx_ring;
struct xdp_rxq_info xdp_rxq;
- int xdp_pending_pkts;
};
struct tun_flow_entry {
@@ -656,7 +656,7 @@ static struct tun_struct *tun_enable_queue(struct tun_file *tfile)
return tun;
}
-static void tun_ptr_free(void *ptr)
+void tun_ptr_free(void *ptr)
{
if (!ptr)
return;
@@ -668,6 +668,7 @@ static void tun_ptr_free(void *ptr)
__skb_array_destroy_skb(ptr);
}
}
+EXPORT_SYMBOL_GPL(tun_ptr_free);
static void tun_queue_purge(struct tun_file *tfile)
{
@@ -742,8 +743,15 @@ static void __tun_detach(struct tun_file *tfile, bool clean)
static void tun_detach(struct tun_file *tfile, bool clean)
{
+ struct tun_struct *tun;
+ struct net_device *dev;
+
rtnl_lock();
+ tun = rtnl_dereference(tfile->tun);
+ dev = tun ? tun->dev : NULL;
__tun_detach(tfile, clean);
+ if (dev)
+ netdev_state_change(dev);
rtnl_unlock();
}
@@ -1613,7 +1621,6 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
unsigned int delta = 0;
char *buf;
size_t copied;
- bool xdp_xmit = false;
int err, pad = TUN_RX_PAD;
rcu_read_lock();
@@ -1643,6 +1650,7 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
else
*skb_xdp = 0;
+ preempt_disable();
rcu_read_lock();
xdp_prog = rcu_dereference(tun->xdp_prog);
if (xdp_prog && !*skb_xdp) {
@@ -1662,15 +1670,22 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
case XDP_REDIRECT:
get_page(alloc_frag->page);
alloc_frag->offset += buflen;
- ++tfile->xdp_pending_pkts;
err = xdp_do_redirect(tun->dev, &xdp, xdp_prog);
+ xdp_do_flush_map();
if (err)
goto err_redirect;
rcu_read_unlock();
+ preempt_enable();
return NULL;
case XDP_TX:
- xdp_xmit = true;
- /* fall through */
+ get_page(alloc_frag->page);
+ alloc_frag->offset += buflen;
+ if (tun_xdp_xmit(tun->dev, &xdp))
+ goto err_redirect;
+ tun_xdp_flush(tun->dev);
+ rcu_read_unlock();
+ preempt_enable();
+ return NULL;
case XDP_PASS:
delta = orig_data - xdp.data;
break;
@@ -1688,6 +1703,7 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
skb = build_skb(buf, buflen);
if (!skb) {
rcu_read_unlock();
+ preempt_enable();
return ERR_PTR(-ENOMEM);
}
@@ -1696,14 +1712,8 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
get_page(alloc_frag->page);
alloc_frag->offset += buflen;
- if (xdp_xmit) {
- skb->dev = tun->dev;
- generic_xdp_tx(skb, xdp_prog);
- rcu_read_unlock();
- return NULL;
- }
-
rcu_read_unlock();
+ preempt_enable();
return skb;
@@ -1711,6 +1721,7 @@ err_redirect:
put_page(alloc_frag->page);
err_xdp:
rcu_read_unlock();
+ preempt_enable();
this_cpu_inc(tun->pcpu_stats->rx_dropped);
return NULL;
}
@@ -1984,11 +1995,6 @@ static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from)
result = tun_get_user(tun, tfile, NULL, from,
file->f_flags & O_NONBLOCK, false);
- if (tfile->xdp_pending_pkts) {
- tfile->xdp_pending_pkts = 0;
- xdp_do_flush_map();
- }
-
tun_put(tun);
return result;
}
@@ -2286,11 +2292,67 @@ static int tun_validate(struct nlattr *tb[], struct nlattr *data[],
return -EINVAL;
}
+static size_t tun_get_size(const struct net_device *dev)
+{
+ BUILD_BUG_ON(sizeof(u32) != sizeof(uid_t));
+ BUILD_BUG_ON(sizeof(u32) != sizeof(gid_t));
+
+ return nla_total_size(sizeof(uid_t)) + /* OWNER */
+ nla_total_size(sizeof(gid_t)) + /* GROUP */
+ nla_total_size(sizeof(u8)) + /* TYPE */
+ nla_total_size(sizeof(u8)) + /* PI */
+ nla_total_size(sizeof(u8)) + /* VNET_HDR */
+ nla_total_size(sizeof(u8)) + /* PERSIST */
+ nla_total_size(sizeof(u8)) + /* MULTI_QUEUE */
+ nla_total_size(sizeof(u32)) + /* NUM_QUEUES */
+ nla_total_size(sizeof(u32)) + /* NUM_DISABLED_QUEUES */
+ 0;
+}
+
+static int tun_fill_info(struct sk_buff *skb, const struct net_device *dev)
+{
+ struct tun_struct *tun = netdev_priv(dev);
+
+ if (nla_put_u8(skb, IFLA_TUN_TYPE, tun->flags & TUN_TYPE_MASK))
+ goto nla_put_failure;
+ if (uid_valid(tun->owner) &&
+ nla_put_u32(skb, IFLA_TUN_OWNER,
+ from_kuid_munged(current_user_ns(), tun->owner)))
+ goto nla_put_failure;
+ if (gid_valid(tun->group) &&
+ nla_put_u32(skb, IFLA_TUN_GROUP,
+ from_kgid_munged(current_user_ns(), tun->group)))
+ goto nla_put_failure;
+ if (nla_put_u8(skb, IFLA_TUN_PI, !(tun->flags & IFF_NO_PI)))
+ goto nla_put_failure;
+ if (nla_put_u8(skb, IFLA_TUN_VNET_HDR, !!(tun->flags & IFF_VNET_HDR)))
+ goto nla_put_failure;
+ if (nla_put_u8(skb, IFLA_TUN_PERSIST, !!(tun->flags & IFF_PERSIST)))
+ goto nla_put_failure;
+ if (nla_put_u8(skb, IFLA_TUN_MULTI_QUEUE,
+ !!(tun->flags & IFF_MULTI_QUEUE)))
+ goto nla_put_failure;
+ if (tun->flags & IFF_MULTI_QUEUE) {
+ if (nla_put_u32(skb, IFLA_TUN_NUM_QUEUES, tun->numqueues))
+ goto nla_put_failure;
+ if (nla_put_u32(skb, IFLA_TUN_NUM_DISABLED_QUEUES,
+ tun->numdisabled))
+ goto nla_put_failure;
+ }
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
static struct rtnl_link_ops tun_link_ops __read_mostly = {
.kind = DRV_NAME,
.priv_size = sizeof(struct tun_struct),
.setup = tun_setup,
.validate = tun_validate,
+ .get_size = tun_get_size,
+ .fill_info = tun_fill_info,
};
static void tun_sock_write_space(struct sock *sk)
@@ -2325,13 +2387,6 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
ret = tun_get_user(tun, tfile, m->msg_control, &m->msg_iter,
m->msg_flags & MSG_DONTWAIT,
m->msg_flags & MSG_MORE);
-
- if (tfile->xdp_pending_pkts >= NAPI_POLL_WEIGHT ||
- !(m->msg_flags & MSG_MORE)) {
- tfile->xdp_pending_pkts = 0;
- xdp_do_flush_map();
- }
-
tun_put(tun);
return ret;
}
@@ -2514,10 +2569,15 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
/* One or more queue has already been attached, no need
* to initialize the device again.
*/
+ netdev_state_change(dev);
return 0;
}
- }
- else {
+
+ tun->flags = (tun->flags & ~TUN_FEATURES) |
+ (ifr->ifr_flags & TUN_FEATURES);
+
+ netdev_state_change(dev);
+ } else {
char *name;
unsigned long flags = 0;
int queues = ifr->ifr_flags & IFF_MULTI_QUEUE ?
@@ -2594,6 +2654,9 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
~(NETIF_F_HW_VLAN_CTAG_TX |
NETIF_F_HW_VLAN_STAG_TX);
+ tun->flags = (tun->flags & ~TUN_FEATURES) |
+ (ifr->ifr_flags & TUN_FEATURES);
+
INIT_LIST_HEAD(&tun->disabled);
err = tun_attach(tun, file, false, ifr->ifr_flags & IFF_NAPI);
if (err < 0)
@@ -2608,9 +2671,6 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr)
tun_debug(KERN_INFO, tun, "tun_set_iff\n");
- tun->flags = (tun->flags & ~TUN_FEATURES) |
- (ifr->ifr_flags & TUN_FEATURES);
-
/* Make sure persistent devices do not get stuck in
* xoff state.
*/
@@ -2757,6 +2817,9 @@ static int tun_set_queue(struct file *file, struct ifreq *ifr)
} else
ret = -EINVAL;
+ if (ret >= 0)
+ netdev_state_change(tun->dev);
+
unlock:
rtnl_unlock();
return ret;
@@ -2789,6 +2852,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
struct tun_struct *tun;
void __user* argp = (void __user*)arg;
struct ifreq ifr;
+ struct net *net;
kuid_t owner;
kgid_t group;
int sndbuf;
@@ -2796,8 +2860,10 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
unsigned int ifindex;
int le;
int ret;
+ bool do_notify = false;
- if (cmd == TUNSETIFF || cmd == TUNSETQUEUE || _IOC_TYPE(cmd) == SOCK_IOC_TYPE) {
+ if (cmd == TUNSETIFF || cmd == TUNSETQUEUE ||
+ (_IOC_TYPE(cmd) == SOCK_IOC_TYPE && cmd != SIOCGSKNS)) {
if (copy_from_user(&ifr, argp, ifreq_len))
return -EFAULT;
} else {
@@ -2817,6 +2883,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
rtnl_lock();
tun = tun_get(tfile);
+ net = sock_net(&tfile->sk);
if (cmd == TUNSETIFF) {
ret = -EEXIST;
if (tun)
@@ -2824,7 +2891,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
ifr.ifr_name[IFNAMSIZ-1] = '\0';
- ret = tun_set_iff(sock_net(&tfile->sk), file, &ifr);
+ ret = tun_set_iff(net, file, &ifr);
if (ret)
goto unlock;
@@ -2846,6 +2913,14 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
tfile->ifindex = ifindex;
goto unlock;
}
+ if (cmd == SIOCGSKNS) {
+ ret = -EPERM;
+ if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
+ goto unlock;
+
+ ret = open_related_ns(&net->ns, get_net_ns);
+ goto unlock;
+ }
ret = -EBADFD;
if (!tun)
@@ -2882,10 +2957,12 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
if (arg && !(tun->flags & IFF_PERSIST)) {
tun->flags |= IFF_PERSIST;
__module_get(THIS_MODULE);
+ do_notify = true;
}
if (!arg && (tun->flags & IFF_PERSIST)) {
tun->flags &= ~IFF_PERSIST;
module_put(THIS_MODULE);
+ do_notify = true;
}
tun_debug(KERN_INFO, tun, "persist %s\n",
@@ -2900,6 +2977,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
break;
}
tun->owner = owner;
+ do_notify = true;
tun_debug(KERN_INFO, tun, "owner set to %u\n",
from_kuid(&init_user_ns, tun->owner));
break;
@@ -2912,6 +2990,7 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
break;
}
tun->group = group;
+ do_notify = true;
tun_debug(KERN_INFO, tun, "group set to %u\n",
from_kgid(&init_user_ns, tun->group));
break;
@@ -3071,6 +3150,9 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
break;
}
+ if (do_notify)
+ netdev_state_change(tun->dev);
+
unlock:
rtnl_unlock();
if (tun)
@@ -3163,7 +3245,6 @@ static int tun_chr_open(struct inode *inode, struct file * file)
sock_set_flag(&tfile->sk, SOCK_ZEROCOPY);
memset(&tfile->tx_ring, 0, sizeof(tfile->tx_ring));
- tfile->xdp_pending_pkts = 0;
return 0;
}
generated by cgit v1.2.3 (git 2.46.0) at 2025-11-07 08:31:02 +0000