linux/net/openvswitch/vport-internal_dev.c

// SPDX-License-Identifier: GPL-2.0-only
/*
 * Copyright (c) 2007-2012 Nicira, Inc.
 */

#include <linux/if_vlan.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/ethtool.h>
#include <linux/skbuff.h>

#include <net/dst.h>
#include <net/xfrm.h>
#include <net/rtnetlink.h>

#include "datapath.h"
#include "vport-internal_dev.h"
#include "vport-netdev.h"

struct internal_dev {
	struct vport *vport;
};

static struct vport_ops ovs_internal_vport_ops;

static struct internal_dev *internal_dev_priv(struct net_device *netdev)
{
	return netdev_priv(netdev);
}

/* Called with rcu_read_lock_bh. */
static netdev_tx_t
internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
{
	int len, err;

	/* store len value because skb can be freed inside ovs_vport_receive() */
	len = skb->len;

	rcu_read_lock();
	err = ovs_vport_receive(internal_dev_priv(netdev)->vport, skb, NULL);
	rcu_read_unlock();

	if (likely(!err))
		dev_sw_netstats_tx_add(netdev, 1, len);
	else
		netdev->stats.tx_errors++;

	return NETDEV_TX_OK;
}

static int internal_dev_open(struct net_device *netdev)
{
	netif_start_queue(netdev);
	return 0;
}

static int internal_dev_stop(struct net_device *netdev)
{
	netif_stop_queue(netdev);
	return 0;
}

static void internal_dev_getinfo(struct net_device *netdev,
				 struct ethtool_drvinfo *info)
{
	strscpy(info->driver, "openvswitch", sizeof(info->driver));
}

static const struct ethtool_ops internal_dev_ethtool_ops = {
	.get_drvinfo	= internal_dev_getinfo,
	.get_link	= ethtool_op_get_link,
};

static void internal_dev_destructor(struct net_device *dev)
{
	struct vport *vport = ovs_internal_dev_get_vport(dev);

	ovs_vport_free(vport);
}

static const struct net_device_ops internal_dev_netdev_ops = {
	.ndo_open = internal_dev_open,
	.ndo_stop = internal_dev_stop,
	.ndo_start_xmit = internal_dev_xmit,
	.ndo_set_mac_address = eth_mac_addr,
	.ndo_get_stats64 = dev_get_tstats64,
};

static struct rtnl_link_ops internal_dev_link_ops __read_mostly = {
	.kind = "openvswitch",
};

static void do_setup(struct net_device *netdev)
{
	ether_setup(netdev);

	netdev->max_mtu = ETH_MAX_MTU;

	netdev->netdev_ops = &internal_dev_netdev_ops;

	netdev->priv_flags &= ~IFF_TX_SKB_SHARING;
	netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_OPENVSWITCH |
			      IFF_NO_QUEUE;
	netdev->needs_free_netdev = true;
	netdev->priv_destructor = NULL;
	netdev->ethtool_ops = &internal_dev_ethtool_ops;
	netdev->rtnl_link_ops = &internal_dev_link_ops;

	netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST |
			   NETIF_F_HIGHDMA | NETIF_F_HW_CSUM |
			   NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL;

	netdev->vlan_features = netdev->features;
	netdev->hw_enc_features = netdev->features;
	netdev->features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
	netdev->hw_features = netdev->features & ~NETIF_F_LLTX;

	eth_hw_addr_random(netdev);
}

static struct vport *internal_dev_create(const struct vport_parms *parms)
{
	struct vport *vport;
	struct internal_dev *internal_dev;
	struct net_device *dev;
	int err;

	vport = ovs_vport_alloc(0, &ovs_internal_vport_ops, parms);
	if (IS_ERR(vport)) {
		err = PTR_ERR(vport);
		goto error;
	}

	dev = alloc_netdev(sizeof(struct internal_dev),
			   parms->name, NET_NAME_USER, do_setup);
	vport->dev = dev;
	if (!vport->dev) {
		err = -ENOMEM;
		goto error_free_vport;
	}
	vport->dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
	if (!vport->dev->tstats) {
		err = -ENOMEM;
		goto error_free_netdev;
	}

	dev_net_set(vport->dev, ovs_dp_get_net(vport->dp));
	dev->ifindex = parms->desired_ifindex;
	internal_dev = internal_dev_priv(vport->dev);
	internal_dev->vport = vport;

	/* Restrict bridge port to current netns. */
	if (vport->port_no == OVSP_LOCAL)
		vport->dev->features |= NETIF_F_NETNS_LOCAL;

	rtnl_lock();
	err = register_netdevice(vport->dev);
	if (err)
		goto error_unlock;
	vport->dev->priv_destructor = internal_dev_destructor;

	dev_set_promiscuity(vport->dev, 1);
	rtnl_unlock();
	netif_start_queue(vport->dev);

	return vport;

error_unlock:
	rtnl_unlock();
	free_percpu(dev->tstats);
error_free_netdev:
	free_netdev(dev);
error_free_vport:
	ovs_vport_free(vport);
error:
	return ERR_PTR(err);
}

static void internal_dev_destroy(struct vport *vport)
{
	netif_stop_queue(vport->dev);
	rtnl_lock();
	dev_set_promiscuity(vport->dev, -1);

	/* unregister_netdevice() waits for an RCU grace period. */
	unregister_netdevice(vport->dev);
	free_percpu(vport->dev->tstats);
	rtnl_unlock();
}

static netdev_tx_t internal_dev_recv(struct sk_buff *skb)
{
	struct net_device *netdev = skb->dev;

	if (unlikely(!(netdev->flags & IFF_UP))) {
		kfree_skb(skb);
		netdev->stats.rx_dropped++;
		return NETDEV_TX_OK;
	}

	skb_dst_drop(skb);
	nf_reset_ct(skb);
	secpath_reset(skb);

	skb->pkt_type = PACKET_HOST;
	skb->protocol = eth_type_trans(skb, netdev);
	skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
	dev_sw_netstats_rx_add(netdev, skb->len);

	netif_rx(skb);
	return NETDEV_TX_OK;
}

static struct vport_ops ovs_internal_vport_ops = {
	.type		= OVS_VPORT_TYPE_INTERNAL,
	.create		= internal_dev_create,
	.destroy	= internal_dev_destroy,
	.send		= internal_dev_recv,
};

int ovs_is_internal_dev(const struct net_device *netdev)
{
	return netdev->netdev_ops == &internal_dev_netdev_ops;
}

struct vport *ovs_internal_dev_get_vport(struct net_device *netdev)
{
	if (!ovs_is_internal_dev(netdev))
		return NULL;

	return internal_dev_priv(netdev)->vport;
}

int ovs_internal_dev_rtnl_link_register(void)
{
	int err;

	err = rtnl_link_register(&internal_dev_link_ops);
	if (err < 0)
		return err;

	err = ovs_vport_ops_register(&ovs_internal_vport_ops);
	if (err < 0)
		rtnl_link_unregister(&internal_dev_link_ops);

	return err;
}

void ovs_internal_dev_rtnl_link_unregister(void)
{
	ovs_vport_ops_unregister(&ovs_internal_vport_ops);
	rtnl_link_unregister(&internal_dev_link_ops);
}