[patch net-next 0/7] net: introduce alternative names for network interfaces

[patch net-next 0/7] net: introduce alternative names for network interfaces

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

In the past, there was repeatedly discussed the IFNAMSIZ (16) limit for
netdevice name length. Now when we have PF and VF representors
with port names like "pfXvfY", it became quite common to hit this limit:
0123456789012345
enp131s0f1npf0vf6
enp131s0f1npf0vf22

Udev cannot rename these interfaces out-of-the-box and user needs to
create custom rules to handle them.

Also, udev has multiple schemes of netdev names. From udev code:
 * Type of names:
 *   b<number>                             - BCMA bus core number
 *   c<bus_id>                             - bus id of a grouped CCW or CCW device,
 *                                           with all leading zeros stripped [s390]
 *   o<index>[n<phys_port_name>|d<dev_port>]
 *                                         - on-board device index number
 *   s<slot>[f<function>][n<phys_port_name>|d<dev_port>]
 *                                         - hotplug slot index number
 *   x<MAC>                                - MAC address
 *   [P<domain>]p<bus>s<slot>[f<function>][n<phys_port_name>|d<dev_port>]
 *                                         - PCI geographical location
 *   [P<domain>]p<bus>s<slot>[f<function>][u<port>][..][c<config>][i<interface>]
 *                                         - USB port number chain
 *   v<slot>                               - VIO slot number (IBM PowerVM)
 *   a<vendor><model>i<instance>           - Platform bus ACPI instance id
 *   i<addr>n<phys_port_name>              - Netdevsim bus address and port name

One device can be often renamed by multiple patterns at the
same time (e.g. pci address/mac).

This patchset introduces alternative names for network interfaces.
Main goal is to:
1) Overcome the IFNAMSIZ limitation (altname limitation is 128 bytes)
2) Allow to have multiple names at the same time (multiple udev patterns)
3) Allow to use alternative names as handle for commands

The patchset introduces two new commands to add/delete list of properties.
Currently only alternative names are implemented but the ifrastructure
could be easily extended later on. This is very similar to the list of vlan
and tunnels being added/deleted to/from bridge ports.

See following examples.

$ ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff

-> Add alternative names for dummy0:
    
$ ip link prop add dummy0 altname someothername
$ ip link prop add dummy0 altname someotherveryveryveryverylongname 
$ ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname someothername
    altname someotherveryveryveryverylongname
$ ip link show someotherveryveryveryverylongname
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname someothername
    altname someotherveryveryveryverylongname

-> Add bridge brx, add it's alternative name and use alternative names to
   do enslavement.
    
$ ip link add name brx type bridge
$ ip link prop add brx altname mypersonalsuperspecialbridge
$ ip link set someotherveryveryveryverylongname master mypersonalsuperspecialbridge
$ ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop master brx state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname someothername
    altname someotherveryveryveryverylongname
3: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname mypersonalsuperspecialbridge

-> Add ipv4 address to the bridge using alternative name:

$ ip addr add 192.168.0.1/24 dev mypersonalsuperspecialbridge
$ ip addr show mypersonalsuperspecialbridge
3: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname mypersonalsuperspecialbridge
    inet 192.168.0.1/24 scope global brx
       valid_lft forever preferred_lft forever

-> Delete one of dummy0 alternative names:

$ ip link prop del dummy0 altname someotherveryveryveryverylongname
$ ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop master brx state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname someothername
3: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname mypersonalsuperspecialbridge

-> Add multiple alternative names at once

$ ip link prop add dummy0 altname a altname b altname c altname d
$ ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop master brx state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname someothername
    altname a
    altname b
    altname c
    altname d
3: brx: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:67:a9:67:46:86 brd ff:ff:ff:ff:ff:ff
    altname mypersonalsuperspecialbridge

Jiri Pirko (7):
  net: procfs: use index hashlist instead of name hashlist
  net: introduce name_node struct to be used in hashlist
  net: rtnetlink: add linkprop commands to add and delete alternative
    ifnames
  net: rtnetlink: put alternative names to getlink message
  net: rtnetlink: unify the code in __rtnl_newlink get dev with the rest
  net: rtnetlink: introduce helper to get net_device instance by ifname
  net: rtnetlink: add possibility to use alternative names as message
    handle

 include/linux/netdevice.h      |  14 ++-
 include/uapi/linux/if.h        |   1 +
 include/uapi/linux/if_link.h   |   2 +
 include/uapi/linux/rtnetlink.h |   7 ++
 net/core/dev.c                 | 153 +++++++++++++++++++++---
 net/core/net-procfs.c          |   4 +-
 net/core/rtnetlink.c           | 206 +++++++++++++++++++++++++++++----
 security/selinux/nlmsgtab.c    |   4 +-
 8 files changed, 348 insertions(+), 43 deletions(-)

-- 
2.21.0

[patch net-next 1/7] net: procfs: use index hashlist instead of name hashlist

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Name hashlist is going to be used for more than just dev->name, so use
rather index hashlist for iteration over net_device instances.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 net/core/net-procfs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/core/net-procfs.c b/net/core/net-procfs.c
index 36347933ec3a..6bbd06f7dc7d 100644
--- a/net/core/net-procfs.c
+++ b/net/core/net-procfs.c
@@ -20,8 +20,8 @@ static inline struct net_device *dev_from_same_bucket(struct seq_file *seq, loff
 	struct hlist_head *h;
 	unsigned int count = 0, offset = get_offset(*pos);
 
-	h = &net->dev_name_head[get_bucket(*pos)];
-	hlist_for_each_entry_rcu(dev, h, name_hlist) {
+	h = &net->dev_index_head[get_bucket(*pos)];
+	hlist_for_each_entry_rcu(dev, h, index_hlist) {
 		if (++count == offset)
 			return dev;
 	}
-- 
2.21.0

[patch net-next 2/7] net: introduce name_node struct to be used in hashlist

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Introduce name_node structure to hold name of device and put it into
hashlist instead of putting there struct net_device directly. Add a
necessary infrastructure to manipulate the hashlist. This prepares
the code to use the same hashlist for alternative names introduced
later in this set.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- make char *name const in struct netdev_name_node and in
  netdev_name_node_alloc()
- use INIT_HLIST_NODE in netdev_name_node_alloc() and
  change kzalloc to kmalloc.
- added patch description
---
 include/linux/netdevice.h | 10 +++-
 net/core/dev.c            | 97 +++++++++++++++++++++++++++++++--------
 2 files changed, 87 insertions(+), 20 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9eda1c31d1f7..e92bc5467256 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -925,6 +925,12 @@ struct dev_ifalias {
 struct devlink;
 struct tlsdev_ops;
 
+struct netdev_name_node {
+	struct hlist_node hlist;
+	struct net_device *dev;
+	const char *name;
+};
+
 /*
  * This structure defines the management hooks for network devices.
  * The following hooks can be defined; unless noted otherwise, they are
@@ -1564,7 +1570,7 @@ enum netdev_priv_flags {
  *		(i.e. as seen by users in the "Space.c" file).  It is the name
  *		of the interface.
  *
- *	@name_hlist: 	Device name hash chain, please keep it close to name[]
+ *	@name_node:	Name hashlist node
  *	@ifalias:	SNMP alias
  *	@mem_end:	Shared memory end
  *	@mem_start:	Shared memory start
@@ -1774,7 +1780,7 @@ enum netdev_priv_flags {
 
 struct net_device {
 	char			name[IFNAMSIZ];
-	struct hlist_node	name_hlist;
+	struct netdev_name_node	*name_node;
 	struct dev_ifalias	__rcu *ifalias;
 	/*
 	 *	I/O specific fields
diff --git a/net/core/dev.c b/net/core/dev.c
index bf3ed413abaf..b9088dee0701 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -228,6 +228,67 @@ static inline void rps_unlock(struct softnet_data *sd)
 #endif
 }
 
+static struct netdev_name_node *netdev_name_node_alloc(struct net_device *dev,
+						       const char *name)
+{
+	struct netdev_name_node *name_node;
+
+	name_node = kmalloc(sizeof(*name_node), GFP_KERNEL);
+	if (!name_node)
+		return NULL;
+	INIT_HLIST_NODE(&name_node->hlist);
+	name_node->dev = dev;
+	name_node->name = name;
+	return name_node;
+}
+
+static struct netdev_name_node *
+netdev_name_node_head_alloc(struct net_device *dev)
+{
+	return netdev_name_node_alloc(dev, dev->name);
+}
+
+static void netdev_name_node_free(struct netdev_name_node *name_node)
+{
+	kfree(name_node);
+}
+
+static void netdev_name_node_add(struct net *net,
+				 struct netdev_name_node *name_node)
+{
+	hlist_add_head_rcu(&name_node->hlist,
+			   dev_name_hash(net, name_node->name));
+}
+
+static void netdev_name_node_del(struct netdev_name_node *name_node)
+{
+	hlist_del_rcu(&name_node->hlist);
+}
+
+static struct netdev_name_node *netdev_name_node_lookup(struct net *net,
+							const char *name)
+{
+	struct hlist_head *head = dev_name_hash(net, name);
+	struct netdev_name_node *name_node;
+
+	hlist_for_each_entry(name_node, head, hlist)
+		if (!strcmp(name_node->name, name))
+			return name_node;
+	return NULL;
+}
+
+static struct netdev_name_node *netdev_name_node_lookup_rcu(struct net *net,
+							    const char *name)
+{
+	struct hlist_head *head = dev_name_hash(net, name);
+	struct netdev_name_node *name_node;
+
+	hlist_for_each_entry_rcu(name_node, head, hlist)
+		if (!strcmp(name_node->name, name))
+			return name_node;
+	return NULL;
+}
+
 /* Device list insertion */
 static void list_netdevice(struct net_device *dev)
 {
@@ -237,7 +298,7 @@ static void list_netdevice(struct net_device *dev)
 
 	write_lock_bh(&dev_base_lock);
 	list_add_tail_rcu(&dev->dev_list, &net->dev_base_head);
-	hlist_add_head_rcu(&dev->name_hlist, dev_name_hash(net, dev->name));
+	netdev_name_node_add(net, dev->name_node);
 	hlist_add_head_rcu(&dev->index_hlist,
 			   dev_index_hash(net, dev->ifindex));
 	write_unlock_bh(&dev_base_lock);
@@ -255,7 +316,7 @@ static void unlist_netdevice(struct net_device *dev)
 	/* Unlink dev from the device chain */
 	write_lock_bh(&dev_base_lock);
 	list_del_rcu(&dev->dev_list);
-	hlist_del_rcu(&dev->name_hlist);
+	netdev_name_node_del(dev->name_node);
 	hlist_del_rcu(&dev->index_hlist);
 	write_unlock_bh(&dev_base_lock);
 
@@ -733,14 +794,10 @@ EXPORT_SYMBOL_GPL(dev_fill_metadata_dst);
 
 struct net_device *__dev_get_by_name(struct net *net, const char *name)
 {
-	struct net_device *dev;
-	struct hlist_head *head = dev_name_hash(net, name);
+	struct netdev_name_node *node_name;
 
-	hlist_for_each_entry(dev, head, name_hlist)
-		if (!strncmp(dev->name, name, IFNAMSIZ))
-			return dev;
-
-	return NULL;
+	node_name = netdev_name_node_lookup(net, name);
+	return node_name ? node_name->dev : NULL;
 }
 EXPORT_SYMBOL(__dev_get_by_name);
 
@@ -758,14 +815,10 @@ EXPORT_SYMBOL(__dev_get_by_name);
 
 struct net_device *dev_get_by_name_rcu(struct net *net, const char *name)
 {
-	struct net_device *dev;
-	struct hlist_head *head = dev_name_hash(net, name);
-
-	hlist_for_each_entry_rcu(dev, head, name_hlist)
-		if (!strncmp(dev->name, name, IFNAMSIZ))
-			return dev;
+	struct netdev_name_node *node_name;
 
-	return NULL;
+	node_name = netdev_name_node_lookup_rcu(net, name);
+	return node_name ? node_name->dev : NULL;
 }
 EXPORT_SYMBOL(dev_get_by_name_rcu);
 
@@ -1232,13 +1285,13 @@ int dev_change_name(struct net_device *dev, const char *newname)
 	netdev_adjacent_rename_links(dev, oldname);
 
 	write_lock_bh(&dev_base_lock);
-	hlist_del_rcu(&dev->name_hlist);
+	netdev_name_node_del(dev->name_node);
 	write_unlock_bh(&dev_base_lock);
 
 	synchronize_rcu();
 
 	write_lock_bh(&dev_base_lock);
-	hlist_add_head_rcu(&dev->name_hlist, dev_name_hash(net, dev->name));
+	netdev_name_node_add(net, dev->name_node);
 	write_unlock_bh(&dev_base_lock);
 
 	ret = call_netdevice_notifiers(NETDEV_CHANGENAME, dev);
@@ -8264,6 +8317,8 @@ static void rollback_registered_many(struct list_head *head)
 		dev_uc_flush(dev);
 		dev_mc_flush(dev);
 
+		netdev_name_node_free(dev->name_node);
+
 		if (dev->netdev_ops->ndo_uninit)
 			dev->netdev_ops->ndo_uninit(dev);
 
@@ -8706,6 +8761,10 @@ int register_netdevice(struct net_device *dev)
 	if (ret < 0)
 		goto out;
 
+	dev->name_node = netdev_name_node_head_alloc(dev);
+	if (!dev->name_node)
+		goto out;
+
 	/* Init, if this function is available */
 	if (dev->netdev_ops->ndo_init) {
 		ret = dev->netdev_ops->ndo_init(dev);
@@ -8827,6 +8886,8 @@ int register_netdevice(struct net_device *dev)
 	return ret;
 
 err_uninit:
+	if (dev->name_node)
+		netdev_name_node_free(dev->name_node);
 	if (dev->netdev_ops->ndo_uninit)
 		dev->netdev_ops->ndo_uninit(dev);
 	if (dev->priv_destructor)
-- 
2.21.0

[patch net-next 3/7] net: rtnetlink: add linkprop commands to add and delete alternative ifnames

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Add two commands to add and delete list of link properties. Implement
the first property type along - alternative ifnames.
Each net device can have multiple alternative names.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- make netdev_name_node_alt_create/destroy() char *name arg const
- converted the patch to add/del list of link properties
---
 include/linux/netdevice.h      |   4 ++
 include/uapi/linux/if.h        |   1 +
 include/uapi/linux/if_link.h   |   2 +
 include/uapi/linux/rtnetlink.h |   7 +++
 net/core/dev.c                 |  58 ++++++++++++++++++-
 net/core/rtnetlink.c           | 103 +++++++++++++++++++++++++++++++++
 security/selinux/nlmsgtab.c    |   4 +-
 7 files changed, 177 insertions(+), 2 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e92bc5467256..48cc71aae466 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -927,10 +927,14 @@ struct tlsdev_ops;
 
 struct netdev_name_node {
 	struct hlist_node hlist;
+	struct list_head list;
 	struct net_device *dev;
 	const char *name;
 };
 
+int netdev_name_node_alt_create(struct net_device *dev, const char *name);
+int netdev_name_node_alt_destroy(struct net_device *dev, const char *name);
+
 /*
  * This structure defines the management hooks for network devices.
  * The following hooks can be defined; unless noted otherwise, they are
diff --git a/include/uapi/linux/if.h b/include/uapi/linux/if.h
index 7fea0fd7d6f5..4bf33344aab1 100644
--- a/include/uapi/linux/if.h
+++ b/include/uapi/linux/if.h
@@ -33,6 +33,7 @@
 #define	IFNAMSIZ	16
 #endif /* __UAPI_DEF_IF_IFNAMSIZ */
 #define	IFALIASZ	256
+#define	ALTIFNAMSIZ	128
 #include <linux/hdlc/ioctl.h>
 
 /* For glibc compatibility. An empty enum does not compile. */
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 4a8c02cafa9a..8aec8769d944 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -167,6 +167,8 @@ enum {
 	IFLA_NEW_IFINDEX,
 	IFLA_MIN_MTU,
 	IFLA_MAX_MTU,
+	IFLA_PROP_LIST,
+	IFLA_ALT_IFNAME, /* Alternative ifname */
 	__IFLA_MAX
 };
 
diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index ce2a623abb75..1418a8362bb7 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -164,6 +164,13 @@ enum {
 	RTM_GETNEXTHOP,
 #define RTM_GETNEXTHOP	RTM_GETNEXTHOP
 
+	RTM_NEWLINKPROP = 108,
+#define RTM_NEWLINKPROP	RTM_NEWLINKPROP
+	RTM_DELLINKPROP,
+#define RTM_DELLINKPROP	RTM_DELLINKPROP
+	RTM_GETLINKPROP,
+#define RTM_GETLINKPROP	RTM_GETLINKPROP
+
 	__RTM_MAX,
 #define RTM_MAX		(((__RTM_MAX + 3) & ~3) - 1)
 };
diff --git a/net/core/dev.c b/net/core/dev.c
index b9088dee0701..36e139793e8e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -245,7 +245,13 @@ static struct netdev_name_node *netdev_name_node_alloc(struct net_device *dev,
 static struct netdev_name_node *
 netdev_name_node_head_alloc(struct net_device *dev)
 {
-	return netdev_name_node_alloc(dev, dev->name);
+	struct netdev_name_node *name_node;
+
+	name_node = netdev_name_node_alloc(dev, dev->name);
+	if (!name_node)
+		return NULL;
+	INIT_LIST_HEAD(&name_node->list);
+	return name_node;
 }
 
 static void netdev_name_node_free(struct netdev_name_node *name_node)
@@ -289,6 +295,55 @@ static struct netdev_name_node *netdev_name_node_lookup_rcu(struct net *net,
 	return NULL;
 }
 
+int netdev_name_node_alt_create(struct net_device *dev, const char *name)
+{
+	struct netdev_name_node *name_node;
+	struct net *net = dev_net(dev);
+
+	name_node = netdev_name_node_lookup(net, name);
+	if (name_node)
+		return -EEXIST;
+	name_node = netdev_name_node_alloc(dev, name);
+	if (!name_node)
+		return -ENOMEM;
+	netdev_name_node_add(net, name_node);
+	/* The node that holds dev->name acts as a head of per-device list. */
+	list_add_tail(&name_node->list, &dev->name_node->list);
+
+	return 0;
+}
+EXPORT_SYMBOL(netdev_name_node_alt_create);
+
+static void __netdev_name_node_alt_destroy(struct netdev_name_node *name_node)
+{
+	list_del(&name_node->list);
+	netdev_name_node_del(name_node);
+	kfree(name_node->name);
+	netdev_name_node_free(name_node);
+}
+
+int netdev_name_node_alt_destroy(struct net_device *dev, const char *name)
+{
+	struct netdev_name_node *name_node;
+	struct net *net = dev_net(dev);
+
+	name_node = netdev_name_node_lookup(net, name);
+	if (!name_node)
+		return -ENOENT;
+	__netdev_name_node_alt_destroy(name_node);
+
+	return 0;
+}
+EXPORT_SYMBOL(netdev_name_node_alt_destroy);
+
+static void netdev_name_node_alt_flush(struct net_device *dev)
+{
+	struct netdev_name_node *name_node, *tmp;
+
+	list_for_each_entry_safe(name_node, tmp, &dev->name_node->list, list)
+		__netdev_name_node_alt_destroy(name_node);
+}
+
 /* Device list insertion */
 static void list_netdevice(struct net_device *dev)
 {
@@ -8317,6 +8372,7 @@ static void rollback_registered_many(struct list_head *head)
 		dev_uc_flush(dev);
 		dev_mc_flush(dev);
 
+		netdev_name_node_alt_flush(dev);
 		netdev_name_node_free(dev->name_node);
 
 		if (dev->netdev_ops->ndo_uninit)
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 1ee6460f8275..e13646993d82 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1750,6 +1750,9 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = {
 	[IFLA_CARRIER_DOWN_COUNT] = { .type = NLA_U32 },
 	[IFLA_MIN_MTU]		= { .type = NLA_U32 },
 	[IFLA_MAX_MTU]		= { .type = NLA_U32 },
+	[IFLA_PROP_LIST]	= { .type = NLA_NESTED },
+	[IFLA_ALT_IFNAME]	= { .type = NLA_STRING,
+				    .len = ALTIFNAMSIZ - 1 },
 };
 
 static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
@@ -3373,6 +3376,103 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	return err;
 }
 
+static int rtnl_alt_ifname(int cmd, struct net_device *dev, struct nlattr *attr,
+			   bool *changed, struct netlink_ext_ack *extack)
+{
+	char *alt_ifname;
+	int err;
+
+	err = nla_validate(attr, attr->nla_len, IFLA_MAX, ifla_policy, extack);
+	if (err)
+		return err;
+
+	alt_ifname = nla_data(attr);
+	if (cmd == RTM_NEWLINKPROP) {
+		alt_ifname = kstrdup(alt_ifname, GFP_KERNEL);
+		if (!alt_ifname)
+			return -ENOMEM;
+		err = netdev_name_node_alt_create(dev, alt_ifname);
+		if (err) {
+			kfree(alt_ifname);
+			return err;
+		}
+	} else if (cmd == RTM_DELLINKPROP) {
+		err = netdev_name_node_alt_destroy(dev, alt_ifname);
+		if (err)
+			return err;
+	} else {
+		WARN_ON(1);
+		return 0;
+	}
+
+	*changed = true;
+	return 0;
+}
+
+static int rtnl_linkprop(int cmd, struct sk_buff *skb, struct nlmsghdr *nlh,
+			 struct netlink_ext_ack *extack)
+{
+	struct net *net = sock_net(skb->sk);
+	struct nlattr *tb[IFLA_MAX + 1];
+	struct net_device *dev;
+	struct ifinfomsg *ifm;
+	bool changed = false;
+	struct nlattr *attr;
+	int err, rem;
+
+	err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy, extack);
+	if (err)
+		return err;
+
+	err = rtnl_ensure_unique_netns(tb, extack, true);
+	if (err)
+		return err;
+
+	ifm = nlmsg_data(nlh);
+	if (ifm->ifi_index > 0) {
+		dev = __dev_get_by_index(net, ifm->ifi_index);
+	} else if (tb[IFLA_IFNAME]) {
+		char ifname[IFNAMSIZ];
+
+		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
+		dev = __dev_get_by_name(net, ifname);
+	} else {
+		return -EINVAL;
+	}
+
+	if (!dev)
+		return -ENODEV;
+
+	if (!tb[IFLA_PROP_LIST])
+		return 0;
+
+	nla_for_each_nested(attr, tb[IFLA_PROP_LIST], rem) {
+		switch (nla_type(attr)) {
+		case IFLA_ALT_IFNAME:
+			err = rtnl_alt_ifname(cmd, dev, attr, &changed, extack);
+			if (err)
+				return err;
+			break;
+		}
+	}
+
+	if (changed)
+		netdev_state_change(dev);
+	return 0;
+}
+
+static int rtnl_newlinkprop(struct sk_buff *skb, struct nlmsghdr *nlh,
+			    struct netlink_ext_ack *extack)
+{
+	return rtnl_linkprop(RTM_NEWLINKPROP, skb, nlh, extack);
+}
+
+static int rtnl_dellinkprop(struct sk_buff *skb, struct nlmsghdr *nlh,
+			    struct netlink_ext_ack *extack)
+{
+	return rtnl_linkprop(RTM_DELLINKPROP, skb, nlh, extack);
+}
+
 static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
 	struct net *net = sock_net(skb->sk);
@@ -5331,6 +5431,9 @@ void __init rtnetlink_init(void)
 	rtnl_register(PF_UNSPEC, RTM_GETROUTE, NULL, rtnl_dump_all, 0);
 	rtnl_register(PF_UNSPEC, RTM_GETNETCONF, NULL, rtnl_dump_all, 0);
 
+	rtnl_register(PF_UNSPEC, RTM_NEWLINKPROP, rtnl_newlinkprop, NULL, 0);
+	rtnl_register(PF_UNSPEC, RTM_DELLINKPROP, rtnl_dellinkprop, NULL, 0);
+
 	rtnl_register(PF_BRIDGE, RTM_NEWNEIGH, rtnl_fdb_add, NULL, 0);
 	rtnl_register(PF_BRIDGE, RTM_DELNEIGH, rtnl_fdb_del, NULL, 0);
 	rtnl_register(PF_BRIDGE, RTM_GETNEIGH, rtnl_fdb_get, rtnl_fdb_dump, 0);
diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c
index 58345ba0528e..c97fdae8f71b 100644
--- a/security/selinux/nlmsgtab.c
+++ b/security/selinux/nlmsgtab.c
@@ -83,6 +83,8 @@ static const struct nlmsg_perm nlmsg_route_perms[] =
 	{ RTM_NEWNEXTHOP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
 	{ RTM_DELNEXTHOP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
 	{ RTM_GETNEXTHOP,	NETLINK_ROUTE_SOCKET__NLMSG_READ  },
+	{ RTM_NEWLINKPROP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
+	{ RTM_DELLINKPROP,	NETLINK_ROUTE_SOCKET__NLMSG_WRITE },
 };
 
 static const struct nlmsg_perm nlmsg_tcpdiag_perms[] =
@@ -166,7 +168,7 @@ int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm)
 		 * structures at the top of this file with the new mappings
 		 * before updating the BUILD_BUG_ON() macro!
 		 */
-		BUILD_BUG_ON(RTM_MAX != (RTM_NEWNEXTHOP + 3));
+		BUILD_BUG_ON(RTM_MAX != (RTM_NEWLINKPROP + 3));
 		err = nlmsg_perm(nlmsg_type, perm, nlmsg_route_perms,
 				 sizeof(nlmsg_route_perms));
 		break;
-- 
2.21.0

[patch net-next 4/7] net: rtnetlink: put alternative names to getlink message

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Extend exiting getlink info message with list of properties. Now the
only ones are alternative names.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- move size initialization after early return in
  rtnl_alt_ifname_list_size()/rtnl_prop_list_size()
- converted to generic property list
---
 net/core/rtnetlink.c | 53 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 53 insertions(+)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e13646993d82..c38917371b84 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -980,6 +980,19 @@ static size_t rtnl_xdp_size(void)
 	return xdp_size;
 }
 
+static size_t rtnl_prop_list_size(const struct net_device *dev)
+{
+	struct netdev_name_node *name_node;
+	size_t size;
+
+	if (list_empty(&dev->name_node->list))
+		return 0;
+	size = nla_total_size(0);
+	list_for_each_entry(name_node, &dev->name_node->list, list)
+		size += nla_total_size(ALTIFNAMSIZ);
+	return size;
+}
+
 static noinline size_t if_nlmsg_size(const struct net_device *dev,
 				     u32 ext_filter_mask)
 {
@@ -1027,6 +1040,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev,
 	       + nla_total_size(4)  /* IFLA_CARRIER_DOWN_COUNT */
 	       + nla_total_size(4)  /* IFLA_MIN_MTU */
 	       + nla_total_size(4)  /* IFLA_MAX_MTU */
+	       + rtnl_prop_list_size(dev)
 	       + 0;
 }
 
@@ -1584,6 +1598,42 @@ static int rtnl_fill_link_af(struct sk_buff *skb,
 	return 0;
 }
 
+static int rtnl_fill_alt_ifnames(struct sk_buff *skb,
+				 const struct net_device *dev)
+{
+	struct netdev_name_node *name_node;
+	int count = 0;
+
+	list_for_each_entry(name_node, &dev->name_node->list, list) {
+		if (nla_put_string(skb, IFLA_ALT_IFNAME, name_node->name))
+			return -EMSGSIZE;
+		count++;
+	}
+	return count;
+}
+
+static int rtnl_fill_prop_list(struct sk_buff *skb,
+			       const struct net_device *dev)
+{
+	struct nlattr *prop_list;
+	int ret;
+
+	prop_list = nla_nest_start(skb, IFLA_PROP_LIST);
+	if (!prop_list)
+		return -EMSGSIZE;
+
+	ret = rtnl_fill_alt_ifnames(skb, dev);
+	if (ret <= 0)
+		goto nest_cancel;
+
+	nla_nest_end(skb, prop_list);
+	return 0;
+
+nest_cancel:
+	nla_nest_cancel(skb, prop_list);
+	return ret;
+}
+
 static int rtnl_fill_ifinfo(struct sk_buff *skb,
 			    struct net_device *dev, struct net *src_net,
 			    int type, u32 pid, u32 seq, u32 change,
@@ -1697,6 +1747,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb,
 		goto nla_put_failure_rcu;
 	rcu_read_unlock();
 
+	if (rtnl_fill_prop_list(skb, dev))
+		goto nla_put_failure;
+
 	nlmsg_end(skb, nlh);
 	return 0;
 
-- 
2.21.0

[patch net-next 5/7] net: rtnetlink: unify the code in __rtnl_newlink get dev with the rest

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

__rtnl_newlink() code flow is a bit different around tb[IFLA_IFNAME]
processing comparing to the other places. Change that to be unified with
the rest.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 net/core/rtnetlink.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index c38917371b84..a0017737442f 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3080,12 +3080,10 @@ static int __rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
-	else {
-		if (ifname[0])
-			dev = __dev_get_by_name(net, ifname);
-		else
-			dev = NULL;
-	}
+	else if (tb[IFLA_IFNAME])
+		dev = __dev_get_by_name(net, ifname);
+	else
+		dev = NULL;
 
 	if (dev) {
 		master_dev = netdev_master_upper_dev_get(dev);
-- 
2.21.0

[patch net-next 6/7] net: rtnetlink: introduce helper to get net_device instance by ifname

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Introduce helper function rtnl_get_dev() that gets net_device structure
instance pointer according to passed ifname or ifname attribute.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 net/core/rtnetlink.c | 45 ++++++++++++++++++++++++--------------------
 1 file changed, 25 insertions(+), 20 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index a0017737442f..77d4719e5be0 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -2778,6 +2778,23 @@ static int do_setlink(const struct sk_buff *skb,
 	return err;
 }
 
+static struct net_device *rtnl_dev_get(struct net *net,
+				       struct nlattr *ifname_attr,
+				       char *ifname)
+{
+	char buffer[IFNAMSIZ];
+
+	if (!ifname) {
+		ifname = buffer;
+		if (ifname_attr)
+			nla_strlcpy(ifname, ifname_attr, IFNAMSIZ);
+		else
+			return NULL;
+	}
+
+	return __dev_get_by_name(net, ifname);
+}
+
 static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 			struct netlink_ext_ack *extack)
 {
@@ -2807,7 +2824,7 @@ static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
 	else if (tb[IFLA_IFNAME])
-		dev = __dev_get_by_name(net, ifname);
+		dev = rtnl_dev_get(net, NULL, ifname);
 	else
 		goto errout;
 
@@ -2880,7 +2897,6 @@ static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	struct net *tgt_net = net;
 	struct net_device *dev = NULL;
 	struct ifinfomsg *ifm;
-	char ifname[IFNAMSIZ];
 	struct nlattr *tb[IFLA_MAX+1];
 	int err;
 	int netnsid = -1;
@@ -2894,9 +2910,6 @@ static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (err < 0)
 		return err;
 
-	if (tb[IFLA_IFNAME])
-		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
-
 	if (tb[IFLA_TARGET_NETNSID]) {
 		netnsid = nla_get_s32(tb[IFLA_TARGET_NETNSID]);
 		tgt_net = rtnl_get_net_ns_capable(NETLINK_CB(skb).sk, netnsid);
@@ -2909,7 +2922,7 @@ static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(tgt_net, ifm->ifi_index);
 	else if (tb[IFLA_IFNAME])
-		dev = __dev_get_by_name(tgt_net, ifname);
+		dev = rtnl_dev_get(net, tb[IFLA_IFNAME], NULL);
 	else if (tb[IFLA_GROUP])
 		err = rtnl_group_dellink(tgt_net, nla_get_u32(tb[IFLA_GROUP]));
 	else
@@ -3081,7 +3094,7 @@ static int __rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
 	else if (tb[IFLA_IFNAME])
-		dev = __dev_get_by_name(net, ifname);
+		dev = rtnl_dev_get(net, NULL, ifname);
 	else
 		dev = NULL;
 
@@ -3363,7 +3376,6 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	struct net *net = sock_net(skb->sk);
 	struct net *tgt_net = net;
 	struct ifinfomsg *ifm;
-	char ifname[IFNAMSIZ];
 	struct nlattr *tb[IFLA_MAX+1];
 	struct net_device *dev = NULL;
 	struct sk_buff *nskb;
@@ -3386,9 +3398,6 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 			return PTR_ERR(tgt_net);
 	}
 
-	if (tb[IFLA_IFNAME])
-		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
-
 	if (tb[IFLA_EXT_MASK])
 		ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
 
@@ -3397,7 +3406,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(tgt_net, ifm->ifi_index);
 	else if (tb[IFLA_IFNAME])
-		dev = __dev_get_by_name(tgt_net, ifname);
+		dev = rtnl_dev_get(tgt_net, tb[IFLA_IFNAME], NULL);
 	else
 		goto out;
 
@@ -3480,16 +3489,12 @@ static int rtnl_linkprop(int cmd, struct sk_buff *skb, struct nlmsghdr *nlh,
 		return err;
 
 	ifm = nlmsg_data(nlh);
-	if (ifm->ifi_index > 0) {
+	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
-	} else if (tb[IFLA_IFNAME]) {
-		char ifname[IFNAMSIZ];
-
-		nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
-		dev = __dev_get_by_name(net, ifname);
-	} else {
+	else if (tb[IFLA_IFNAME])
+		dev = rtnl_dev_get(net, tb[IFLA_IFNAME], NULL);
+	else
 		return -EINVAL;
-	}
 
 	if (!dev)
 		return -ENODEV;
-- 
2.21.0

[patch net-next 7/7] net: rtnetlink: add possibility to use alternative names as message handle

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Extend the basic rtnetlink commands to use alternative interface names
as a handle instead of ifindex and ifname.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- rebased on top of changes in the previous patches
---
 net/core/rtnetlink.c | 29 ++++++++++++++++++-----------
 1 file changed, 18 insertions(+), 11 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 77d4719e5be0..49fa910b58af 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -2780,14 +2780,17 @@ static int do_setlink(const struct sk_buff *skb,
 
 static struct net_device *rtnl_dev_get(struct net *net,
 				       struct nlattr *ifname_attr,
+				       struct nlattr *altifname_attr,
 				       char *ifname)
 {
-	char buffer[IFNAMSIZ];
+	char buffer[ALTIFNAMSIZ];
 
 	if (!ifname) {
 		ifname = buffer;
 		if (ifname_attr)
 			nla_strlcpy(ifname, ifname_attr, IFNAMSIZ);
+		else if (altifname_attr)
+			nla_strlcpy(ifname, altifname_attr, ALTIFNAMSIZ);
 		else
 			return NULL;
 	}
@@ -2823,8 +2826,8 @@ static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
-	else if (tb[IFLA_IFNAME])
-		dev = rtnl_dev_get(net, NULL, ifname);
+	else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME])
+		dev = rtnl_dev_get(net, NULL, tb[IFLA_ALT_IFNAME], ifname);
 	else
 		goto errout;
 
@@ -2921,8 +2924,9 @@ static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(tgt_net, ifm->ifi_index);
-	else if (tb[IFLA_IFNAME])
-		dev = rtnl_dev_get(net, tb[IFLA_IFNAME], NULL);
+	else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME])
+		dev = rtnl_dev_get(net, tb[IFLA_IFNAME],
+				   tb[IFLA_ALT_IFNAME], NULL);
 	else if (tb[IFLA_GROUP])
 		err = rtnl_group_dellink(tgt_net, nla_get_u32(tb[IFLA_GROUP]));
 	else
@@ -3093,8 +3097,8 @@ static int __rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
-	else if (tb[IFLA_IFNAME])
-		dev = rtnl_dev_get(net, NULL, ifname);
+	else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME])
+		dev = rtnl_dev_get(net, NULL, tb[IFLA_ALT_IFNAME], ifname);
 	else
 		dev = NULL;
 
@@ -3358,6 +3362,7 @@ static int rtnl_valid_getlink_req(struct sk_buff *skb,
 
 		switch (i) {
 		case IFLA_IFNAME:
+		case IFLA_ALT_IFNAME:
 		case IFLA_EXT_MASK:
 		case IFLA_TARGET_NETNSID:
 			break;
@@ -3405,8 +3410,9 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(tgt_net, ifm->ifi_index);
-	else if (tb[IFLA_IFNAME])
-		dev = rtnl_dev_get(tgt_net, tb[IFLA_IFNAME], NULL);
+	else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME])
+		dev = rtnl_dev_get(tgt_net, tb[IFLA_IFNAME],
+				   tb[IFLA_ALT_IFNAME], NULL);
 	else
 		goto out;
 
@@ -3491,8 +3497,9 @@ static int rtnl_linkprop(int cmd, struct sk_buff *skb, struct nlmsghdr *nlh,
 	ifm = nlmsg_data(nlh);
 	if (ifm->ifi_index > 0)
 		dev = __dev_get_by_index(net, ifm->ifi_index);
-	else if (tb[IFLA_IFNAME])
-		dev = rtnl_dev_get(net, tb[IFLA_IFNAME], NULL);
+	else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME])
+		dev = rtnl_dev_get(net, tb[IFLA_IFNAME],
+				   tb[IFLA_ALT_IFNAME], NULL);
 	else
 		return -EINVAL;
 
-- 
2.21.0

[patch net-next 1/2] ip: add support for alternative name addition/deletion/list

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Implement addition/deletion of lists of properties, currently
alternative ifnames. Also extent the ip link show command to list them.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- convert to generic property lists
- added patch description
- use ll_name_to_index() and ifindex as handle
- fixed parsing with nested flag on
- adjusted output
- added arg name to check_altifname header prototype
- added help and man parts
---
 include/uapi/linux/if_link.h   |  2 +
 include/uapi/linux/rtnetlink.h |  7 +++
 include/utils.h                |  1 +
 ip/ipaddress.c                 | 20 ++++++++-
 ip/iplink.c                    | 82 +++++++++++++++++++++++++++++++++-
 lib/utils.c                    | 19 +++++---
 man/man8/ip-link.8.in          | 11 +++++
 7 files changed, 134 insertions(+), 8 deletions(-)

diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index d36919fb4024..1c49f436424d 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -167,6 +167,8 @@ enum {
 	IFLA_NEW_IFINDEX,
 	IFLA_MIN_MTU,
 	IFLA_MAX_MTU,
+	IFLA_PROP_LIST,
+	IFLA_ALT_IFNAME, /* Alternative ifname */
 	__IFLA_MAX
 };
 
diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index 358e83ee134d..4b93791cdbd4 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -164,6 +164,13 @@ enum {
 	RTM_GETNEXTHOP,
 #define RTM_GETNEXTHOP	RTM_GETNEXTHOP
 
+	RTM_NEWLINKPROP = 108,
+#define RTM_NEWLINKPROP	RTM_NEWLINKPROP
+	RTM_DELLINKPROP,
+#define RTM_DELLINKPROP	RTM_DELLINKPROP
+	RTM_GETLINKPROP,
+#define RTM_GETLINKPROP	RTM_GETLINKPROP
+
 	__RTM_MAX,
 #define RTM_MAX		(((__RTM_MAX + 3) & ~3) - 1)
 };
diff --git a/include/utils.h b/include/utils.h
index 794d36053634..001491a1eb40 100644
--- a/include/utils.h
+++ b/include/utils.h
@@ -196,6 +196,7 @@ void duparg(const char *, const char *) __attribute__((noreturn));
 void duparg2(const char *, const char *) __attribute__((noreturn));
 int nodev(const char *dev);
 int check_ifname(const char *);
+int check_altifname(const char *name);
 int get_ifname(char *, const char *);
 const char *get_ifname_rta(int ifindex, const struct rtattr *rta);
 bool matches(const char *prefix, const char *string);
diff --git a/ip/ipaddress.c b/ip/ipaddress.c
index bc8f5ba13c33..2881b2e356ff 100644
--- a/ip/ipaddress.c
+++ b/ip/ipaddress.c
@@ -879,7 +879,7 @@ int print_linkinfo(struct nlmsghdr *n, void *arg)
 	if (filter.up && !(ifi->ifi_flags&IFF_UP))
 		return -1;
 
-	parse_rtattr(tb, IFLA_MAX, IFLA_RTA(ifi), len);
+	parse_rtattr_flags(tb, IFLA_MAX, IFLA_RTA(ifi), len, NLA_F_NESTED);
 
 	name = get_ifname_rta(ifi->ifi_index, tb[IFLA_IFNAME]);
 	if (!name)
@@ -1139,7 +1139,23 @@ int print_linkinfo(struct nlmsghdr *n, void *arg)
 		close_json_array(PRINT_JSON, NULL);
 	}
 
-	print_string(PRINT_FP, NULL, "%s", "\n");
+	if (tb[IFLA_PROP_LIST]) {
+		struct rtattr *i, *proplist = tb[IFLA_PROP_LIST];
+		int rem = RTA_PAYLOAD(proplist);
+
+		open_json_array(PRINT_JSON, "altnames");
+		for (i = RTA_DATA(proplist); RTA_OK(i, rem);
+		     i = RTA_NEXT(i, rem)) {
+			if (i->rta_type != IFLA_ALT_IFNAME)
+				continue;
+			print_string(PRINT_FP, "NULL", "%s    altname ", _SL_);
+			print_string(PRINT_ANY, NULL,
+				     "%s", rta_getattr_str(i));
+		}
+		close_json_array(PRINT_JSON, NULL);
+	}
+
+	print_string(PRINT_FP, NULL, "%s", _SL_);
 	fflush(fp);
 	return 1;
 }
diff --git a/ip/iplink.c b/ip/iplink.c
index 212a088535da..e3f8a28fe94c 100644
--- a/ip/iplink.c
+++ b/ip/iplink.c
@@ -111,7 +111,9 @@ void iplink_usage(void)
 		"\n"
 		"	ip link xstats type TYPE [ ARGS ]\n"
 		"\n"
-		"	ip link afstats [ dev DEVICE ]\n");
+		"	ip link afstats [ dev DEVICE ]\n"
+		"	ip link prop add dev DEVICE [ altname NAME .. ]\n"
+		"	ip link prop del dev DEVICE [ altname NAME .. ]\n");
 
 	if (iplink_have_newlink()) {
 		fprintf(stderr,
@@ -1617,6 +1619,81 @@ static int iplink_afstats(int argc, char **argv)
 	return 0;
 }
 
+static int iplink_prop_mod(int argc, char **argv, struct iplink_req *req)
+{
+	struct rtattr *proplist;
+	char *dev = NULL;
+	char *name;
+
+	proplist = addattr_nest(&req->n, sizeof(*req),
+				IFLA_PROP_LIST | NLA_F_NESTED);
+
+	while (argc > 0) {
+		if (matches(*argv, "altname") == 0) {
+			NEXT_ARG();
+			if (check_altifname(*argv))
+				invarg("not a valid altname", *argv);
+			name = *argv;
+			addattr_l(&req->n, sizeof(*req), IFLA_ALT_IFNAME,
+				  name, strlen(name) + 1);
+		} else if (matches(*argv, "help") == 0) {
+			usage();
+		} else {
+			if (strcmp(*argv, "dev") == 0)
+				NEXT_ARG();
+			if (dev)
+				duparg2("dev", *argv);
+			if (check_altifname(*argv))
+				invarg("\"dev\" not a valid ifname", *argv);
+			dev = *argv;
+		}
+		argv++; argc--;
+	}
+	addattr_nest_end(&req->n, proplist);
+
+	if (!dev) {
+		fprintf(stderr, "Not enough of information: \"dev\" argument is required.\n");
+		exit(-1);
+	}
+
+	req->i.ifi_index = ll_name_to_index(dev);
+	if (!req->i.ifi_index)
+		return nodev(dev);
+
+	if (rtnl_talk(&rth, &req->n, NULL) < 0)
+		return -2;
+
+	return 0;
+}
+
+static int iplink_prop(int argc, char **argv)
+{
+	struct iplink_req req = {
+		.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
+		.n.nlmsg_flags = NLM_F_REQUEST,
+		.i.ifi_family = preferred_family,
+	};
+
+	if (argc <= 0) {
+		usage();
+		exit(-1);
+	}
+
+	if (matches(*argv, "add") == 0) {
+		req.n.nlmsg_flags |= NLM_F_EXCL | NLM_F_CREATE | NLM_F_APPEND;
+		req.n.nlmsg_type = RTM_NEWLINKPROP;
+	} else if (matches(*argv, "del") == 0) {
+		req.n.nlmsg_flags |= RTM_DELLINK;
+		req.n.nlmsg_type = RTM_DELLINKPROP;
+	} else if (matches(*argv, "help") == 0) {
+		usage();
+	} else {
+		fprintf(stderr, "Operator required\n");
+		exit(-1);
+	}
+	return iplink_prop_mod(argc - 1, argv + 1, &req);
+}
+
 static void do_help(int argc, char **argv)
 {
 	struct link_util *lu = NULL;
@@ -1674,6 +1751,9 @@ int do_iplink(int argc, char **argv)
 		return 0;
 	}
 
+	if (matches(*argv, "prop") == 0)
+		return iplink_prop(argc-1, argv+1);
+
 	if (matches(*argv, "help") == 0) {
 		do_help(argc-1, argv+1);
 		return 0;
diff --git a/lib/utils.c b/lib/utils.c
index 95d46ff210aa..bbb3bdcfa80b 100644
--- a/lib/utils.c
+++ b/lib/utils.c
@@ -824,14 +824,10 @@ int nodev(const char *dev)
 	return -1;
 }
 
-int check_ifname(const char *name)
+static int __check_ifname(const char *name)
 {
-	/* These checks mimic kernel checks in dev_valid_name */
 	if (*name == '\0')
 		return -1;
-	if (strlen(name) >= IFNAMSIZ)
-		return -1;
-
 	while (*name) {
 		if (*name == '/' || isspace(*name))
 			return -1;
@@ -840,6 +836,19 @@ int check_ifname(const char *name)
 	return 0;
 }
 
+int check_ifname(const char *name)
+{
+	/* These checks mimic kernel checks in dev_valid_name */
+	if (strlen(name) >= IFNAMSIZ)
+		return -1;
+	return __check_ifname(name);
+}
+
+int check_altifname(const char *name)
+{
+	return __check_ifname(name);
+}
+
 /* buf is assumed to be IFNAMSIZ */
 int get_ifname(char *buf, const char *name)
 {
diff --git a/man/man8/ip-link.8.in b/man/man8/ip-link.8.in
index a8ae72d2b097..7cb4d56726d8 100644
--- a/man/man8/ip-link.8.in
+++ b/man/man8/ip-link.8.in
@@ -244,6 +244,17 @@ ip-link \- network device configuration
 .IR VLAN-QOS " ] ["
 .B proto
 .IR VLAN-PROTO " ] ]"
+.in -8
+
+.ti -8
+.BI "ip link prop add"
+.RB "[ " altname
+.IR NAME " .. ]"
+
+.ti -8
+.BI "ip link prop del"
+.RB "[ " altname
+.IR NAME " .. ]"
 
 .SH "DESCRIPTION"
 .SS ip link add - add virtual link
-- 
2.21.0

[patch net-next 2/2] ip: allow to use alternative names as handle

[Jiri Pirko]

From: Jiri Pirko <jiri@mellanox.com>

Extend ll_name_to_index() to get the index of a netdevice using
alternative interface name. Allow alternative long names to pass checks
in couple of ip link/addr commands.

Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
rfc->v1:
- added patch description
---
 ip/iplink.c  |  5 +++--
 lib/ll_map.c | 41 ++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 43 insertions(+), 3 deletions(-)

diff --git a/ip/iplink.c b/ip/iplink.c
index e3f8a28fe94c..1684e259b538 100644
--- a/ip/iplink.c
+++ b/ip/iplink.c
@@ -931,7 +931,7 @@ int iplink_parse(int argc, char **argv, struct iplink_req *req, char **type)
 				NEXT_ARG();
 			if (dev != name)
 				duparg2("dev", *argv);
-			if (check_ifname(*argv))
+			if (check_altifname(*argv))
 				invarg("\"dev\" not a valid ifname", *argv);
 			dev = *argv;
 		}
@@ -1106,7 +1106,8 @@ int iplink_get(char *name, __u32 filt_mask)
 
 	if (name) {
 		addattr_l(&req.n, sizeof(req),
-			  IFLA_IFNAME, name, strlen(name) + 1);
+			  !check_ifname(name) ? IFLA_IFNAME : IFLA_ALT_IFNAME,
+			  name, strlen(name) + 1);
 	}
 	addattr32(&req.n, sizeof(req), IFLA_EXT_MASK, filt_mask);
 
diff --git a/lib/ll_map.c b/lib/ll_map.c
index e0ed54bf77c9..04dfb0f2320b 100644
--- a/lib/ll_map.c
+++ b/lib/ll_map.c
@@ -70,7 +70,7 @@ static struct ll_cache *ll_get_by_name(const char *name)
 		struct ll_cache *im
 			= container_of(n, struct ll_cache, name_hash);
 
-		if (strncmp(im->name, name, IFNAMSIZ) == 0)
+		if (strcmp(im->name, name) == 0)
 			return im;
 	}
 
@@ -240,6 +240,43 @@ int ll_index_to_flags(unsigned idx)
 	return im ? im->flags : -1;
 }
 
+static int altnametoindex(const char *name)
+{
+	struct {
+		struct nlmsghdr		n;
+		struct ifinfomsg	ifm;
+		char			buf[1024];
+	} req = {
+		.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
+		.n.nlmsg_flags = NLM_F_REQUEST,
+		.n.nlmsg_type = RTM_GETLINK,
+	};
+	struct rtnl_handle rth = {};
+	struct nlmsghdr *answer;
+	struct ifinfomsg *ifm;
+	int rc = 0;
+
+	if (rtnl_open(&rth, 0) < 0)
+		return 0;
+
+	addattr32(&req.n, sizeof(req), IFLA_EXT_MASK,
+		  RTEXT_FILTER_VF | RTEXT_FILTER_SKIP_STATS);
+	addattr_l(&req.n, sizeof(req), IFLA_ALT_IFNAME, name, strlen(name) + 1);
+
+	if (rtnl_talk_suppress_rtnl_errmsg(&rth, &req.n, &answer) < 0)
+		goto out;
+
+	ifm = NLMSG_DATA(answer);
+	rc = ifm->ifi_index;
+
+	free(answer);
+
+	rtnl_close(&rth);
+out:
+	return rc;
+}
+
+
 unsigned ll_name_to_index(const char *name)
 {
 	const struct ll_cache *im;
@@ -257,6 +294,8 @@ unsigned ll_name_to_index(const char *name)
 		idx = if_nametoindex(name);
 	if (idx == 0)
 		idx = ll_idx_a2n(name);
+	if (idx == 0)
+		idx = altnametoindex(name);
 	return idx;
 }
 
-- 
2.21.0

Re: [patch net-next 1/2] ip: add support for alternative name addition/deletion/list

[Jiri Pirko]

In subject of this and the next one should be iproute2-next, of course.

Re: [patch net-next 1/2] ip: add support for alternative name addition/deletion/list

[Stephen Hemminger]

On Mon, 30 Sep 2019 11:59:02 +0200
Jiri Pirko <jiri@resnulli.us> wrote:

> +		open_json_array(PRINT_JSON, "altnames");
> +		for (i = RTA_DATA(proplist); RTA_OK(i, rem);
> +		     i = RTA_NEXT(i, rem)) {
> +			if (i->rta_type != IFLA_ALT_IFNAME)
> +				continue;
> +			print_string(PRINT_FP, "NULL", "%s    altname ", _SL_);

You can pass real NULL versus quoted NULL when doing print to file only

Re: [patch net-next 1/2] ip: add support for alternative name addition/deletion/list

[Jiri Pirko]

Mon, Sep 30, 2019 at 06:27:18PM CEST, stephen@networkplumber.org wrote:
>On Mon, 30 Sep 2019 11:59:02 +0200
>Jiri Pirko <jiri@resnulli.us> wrote:
>
>> +		open_json_array(PRINT_JSON, "altnames");
>> +		for (i = RTA_DATA(proplist); RTA_OK(i, rem);
>> +		     i = RTA_NEXT(i, rem)) {
>> +			if (i->rta_type != IFLA_ALT_IFNAME)
>> +				continue;
>> +			print_string(PRINT_FP, "NULL", "%s    altname ", _SL_);
>
>You can pass real NULL versus quoted NULL when doing print to file only

Okay, will do.

Re: [patch net-next 0/7] net: introduce alternative names for network interfaces

[David Miller]

From: Jiri Pirko <jiri@resnulli.us>
Date: Mon, 30 Sep 2019 11:48:13 +0200

> This patchset introduces alternative names for network interfaces.
> Main goal is to:
> 1) Overcome the IFNAMSIZ limitation (altname limitation is 128 bytes)
> 2) Allow to have multiple names at the same time (multiple udev patterns)
> 3) Allow to use alternative names as handle for commands

Ok, let's see where this goes.

Series applied, thanks Jiri.