From: Xiao Liang shaw.leon@gmail.com Date: Mon, 10 Feb 2025 21:29:53 +0800
There are 4 net namespaces involved when creating links:
- source netns - where the netlink socket resides,
- target netns - where to put the device being created,
- link netns - netns associated with the device (backend),
- peer netns - netns of peer device.
Currently, two nets are passed to newlink() callback - "src_net" parameter and "dev_net" (implicitly in net_device). They are set as follows, depending on netlink attributes in the request.
+------------+-------------------+---------+---------+ | peer netns | IFLA_LINK_NETNSID | src_net | dev_net | +------------+-------------------+---------+---------+ | | absent | source | target | | absent +-------------------+---------+---------+ | | present | link | link | +------------+-------------------+---------+---------+ | | absent | peer | target | | present +-------------------+---------+---------+ | | present | peer | link | +------------+-------------------+---------+---------+
When IFLA_LINK_NETNSID is present, the device is created in link netns first and then moved to target netns. This has some side effects, including extra ifindex allocation, ifname validation and link events. These could be avoided if we create it in target netns from the beginning.
On the other hand, the meaning of src_net parameter is ambiguous. It varies depending on how parameters are passed. It is the effective link (or peer netns) by design, but some drivers ignore it and use dev_net instead.
This patch packs existing newlink() parameters, along with the source netns, link netns and peer netns, into a struct. The old "src_net" is renamed to "net" to avoid confusion with real source netns, and will be deprecated later. The use of src_net are converted to params->net trivially.
To make the semantics more clear, two helper functions - rtnl_newlink_link_net() and rtnl_newlink_peer_net() - are provided for netns fallback logic. Peer netns falls back to link netns, and link netns falls back to source netns.
In following patches, to prepare for creating link in target netns directly:
- For device drivers that are aware of the old "src_net", the use of it are replace with one of the two helper functions.
- And for those that takes dev_net() as link netns, we try params->link_net and then dev_net(), in order to maintain compatibility with the old behavior.
Signed-off-by: Xiao Liang shaw.leon@gmail.com
Reviewed-by: Kuniyuki Iwashima kuniyu@amazon.com
I left small comments below.
[...]
diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h index 523025106a64..0f7281e3e448 100644 --- a/include/linux/if_macvlan.h +++ b/include/linux/if_macvlan.h @@ -59,8 +59,10 @@ static inline void macvlan_count_rx(const struct macvlan_dev *vlan,
extern void macvlan_common_setup(struct net_device *dev);
-extern int macvlan_common_newlink(struct net *src_net, struct net_device *dev,
struct nlattr *tb[], struct nlattr *data[],+struct rtnl_newlink_params;
You can just include <net/rtnetlink.h> and remove it from .c files, then this forward declaration will be unnecessary.
+extern int macvlan_common_newlink(struct net_device *dev,
struct rtnl_newlink_params *params, struct netlink_ext_ack *extack);extern void macvlan_dellink(struct net_device *dev, struct list_head *head);
[...]
diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index bc0069a8b6ea..00c086ca0c11 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -69,6 +69,42 @@ static inline int rtnl_msg_family(const struct nlmsghdr *nlh) return AF_UNSPEC; }
+/**
- struct rtnl_newlink_params - parameters of rtnl_link_ops::newlink()
The '\t' after '*' should be single '\s'.
Same for lines below.
- @net: Netns of interest
- @src_net: Source netns of rtnetlink socket
- @link_net: Link netns by IFLA_LINK_NETNSID, NULL if not specified
- @peer_net: Peer netns
- @tb: IFLA_* attributes
- @data: IFLA_INFO_DATA attributes
- */
+struct rtnl_newlink_params {
[...]
+/* Get effective link netns from newlink params. Generally, this is link_net
- and falls back to src_net. But for compatibility, a driver may * choose to
- use dev_net(dev) instead.
- */
+static inline struct net *rtnl_newlink_link_net(struct rtnl_newlink_params *p) +{
- return p->link_net ? : p->src_net;
+}
+/* Get peer netns from newlink params. Fallback to link netns if peer netns is
- not specified explicitly.
- */
+static inline struct net *rtnl_newlink_peer_net(struct rtnl_newlink_params *p) +{
- return p->peer_net ? : rtnl_newlink_link_net(p);
+}
These helpers should belong to patch 2 ?