* [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink
@ 2025-02-10 13:29 Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 01/11] rtnetlink: Lookup device in target netns when creating link Xiao Liang
` (8 more replies)
0 siblings, 9 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
This patch series includes some netns-related improvements and fixes for
rtnetlink, to make link creation more intuitive:
1) Creating link in another net namespace doesn't conflict with link
names in current one.
2) Refector rtnetlink link creation. Create link in target namespace
directly.
So that
# ip link add netns ns1 link-netns ns2 tun0 type gre ...
will create tun0 in ns1, rather than create it in ns2 and move to ns1.
And don't conflict with another interface named "tun0" in current netns.
Patch 01 avoids link name conflict in different netns.
To achieve 2), there're mainly 3 steps:
- Patch 02 packs newlink() parameters into a struct, including
the original "src_net" along with more netns context. No semantic
changes are introduced.
- Patch 03 ~ 07 converts device drivers to use the explicit netns
extracted from params.
- Patch 08 ~ 09 removes the old netns parameter, and converts
rtnetlink to create device in target netns directly.
Patch 10 ~ 11 adds some tests for link name and link netns.
BTW please note there're some issues found in current code:
- In amt_newlink() drivers/net/amt.c:
amt->net = net;
...
amt->stream_dev = dev_get_by_index(net, ...
Uses net, but amt_lookup_upper_dev() only searches in dev_net.
So the AMT device may not be properly deleted if it's in a different
netns from lower dev.
- In lowpan_newlink() in net/ieee802154/6lowpan/core.c:
wdev = dev_get_by_index(dev_net(ldev), nla_get_u32(tb[IFLA_LINK]));
Looks for IFLA_LINK in dev_net, but in theory the ifindex is defined
in link netns.
And thanks to Kuniyuki for fixing related issues in gtp and pfcp:
https://lore.kernel.org/netdev/20250110014754.33847-1-kuniyu@amazon.com/
---
v9:
- Change the prototype of macvlan_common_newlink().
- Minor fixes of coding style and local variables.
v8:
link: https://lore.kernel.org/all/20250113143719.7948-1-shaw.leon@gmail.com/
- Move dev and ext_ack out from param struct.
- Validate link_net and dev_net are identical for 6lowpan.
v7:
link: https://lore.kernel.org/all/20250104125732.17335-1-shaw.leon@gmail.com/
- Add selftest kconfig.
- Remove a duplicated test of ip6gre.
v6:
link: https://lore.kernel.org/all/20241218130909.2173-1-shaw.leon@gmail.com/
- Split prototype, driver and rtnetlink changes.
- Add more tests for link netns.
- Fix IPv6 tunnel net overwriten in ndo_init().
- Reorder variable declarations.
- Exclude a ip_tunnel-specific patch.
v5:
link: https://lore.kernel.org/all/20241209140151.231257-1-shaw.leon@gmail.com/
- Fix function doc in batman-adv.
- Include peer_net in rtnl newlink parameters.
v4:
link: https://lore.kernel.org/all/20241118143244.1773-1-shaw.leon@gmail.com/
- Pack newlink() parameters to a single struct.
- Use ynl async_msg_queue.empty() in selftest.
v3:
link: https://lore.kernel.org/all/20241113125715.150201-1-shaw.leon@gmail.com/
- Drop "netns_atomic" flag and module parameter. Add netns parameter to
newlink() instead, and convert drivers accordingly.
- Move python NetNSEnter helper to net selftest lib.
v2:
link: https://lore.kernel.org/all/20241107133004.7469-1-shaw.leon@gmail.com/
- Check NLM_F_EXCL to ensure only link creation is affected.
- Add self tests for link name/ifindex conflict and notifications
in different netns.
- Changes in dummy driver and ynl in order to add the test case.
v1:
link: https://lore.kernel.org/all/20241023023146.372653-1-shaw.leon@gmail.com/
Xiao Liang (11):
rtnetlink: Lookup device in target netns when creating link
rtnetlink: Pack newlink() params into struct
net: Use link netns in newlink() of rtnl_link_ops
ieee802154: 6lowpan: Validate link netns in newlink() of rtnl_link_ops
net: ip_tunnel: Use link netns in newlink() of rtnl_link_ops
net: ipv6: Use link netns in newlink() of rtnl_link_ops
net: xfrm: Use link netns in newlink() of rtnl_link_ops
rtnetlink: Remove "net" from newlink params
rtnetlink: Create link directly in target net namespace
selftests: net: Add python context manager for netns entering
selftests: net: Add test cases for link and peer netns
drivers/infiniband/ulp/ipoib/ipoib_netlink.c | 9 +-
drivers/net/amt.c | 11 +-
drivers/net/bareudp.c | 9 +-
drivers/net/bonding/bond_netlink.c | 6 +-
drivers/net/can/dev/netlink.c | 4 +-
drivers/net/can/vxcan.c | 7 +-
.../ethernet/qualcomm/rmnet/rmnet_config.c | 9 +-
drivers/net/geneve.c | 9 +-
drivers/net/gtp.c | 10 +-
drivers/net/ipvlan/ipvlan.h | 3 +-
drivers/net/ipvlan/ipvlan_main.c | 8 +-
drivers/net/ipvlan/ipvtap.c | 6 +-
drivers/net/macsec.c | 9 +-
drivers/net/macvlan.c | 21 +--
drivers/net/macvtap.c | 6 +-
drivers/net/netkit.c | 14 +-
drivers/net/pfcp.c | 9 +-
drivers/net/ppp/ppp_generic.c | 9 +-
drivers/net/team/team_core.c | 6 +-
drivers/net/veth.c | 7 +-
drivers/net/vrf.c | 5 +-
drivers/net/vxlan/vxlan_core.c | 9 +-
drivers/net/wireguard/device.c | 7 +-
drivers/net/wireless/virtual/virt_wifi.c | 8 +-
drivers/net/wwan/wwan_core.c | 16 +-
include/linux/if_macvlan.h | 6 +-
include/net/ip_tunnels.h | 5 +-
include/net/rtnetlink.h | 40 ++++-
net/8021q/vlan_netlink.c | 9 +-
net/batman-adv/soft-interface.c | 9 +-
net/bridge/br_netlink.c | 6 +-
net/caif/chnl_net.c | 5 +-
net/core/rtnetlink.c | 34 +++--
net/hsr/hsr_netlink.c | 12 +-
net/ieee802154/6lowpan/core.c | 7 +-
net/ipv4/ip_gre.c | 22 ++-
net/ipv4/ip_tunnel.c | 10 +-
net/ipv4/ip_vti.c | 9 +-
net/ipv4/ipip.c | 9 +-
net/ipv6/ip6_gre.c | 30 ++--
net/ipv6/ip6_tunnel.c | 19 ++-
net/ipv6/ip6_vti.c | 15 +-
net/ipv6/sit.c | 17 ++-
net/xfrm/xfrm_interface_core.c | 15 +-
tools/testing/selftests/net/Makefile | 1 +
tools/testing/selftests/net/config | 5 +
.../testing/selftests/net/lib/py/__init__.py | 2 +-
tools/testing/selftests/net/lib/py/netns.py | 18 +++
tools/testing/selftests/net/link_netns.py | 141 ++++++++++++++++++
tools/testing/selftests/net/netns-name.sh | 10 ++
50 files changed, 494 insertions(+), 179 deletions(-)
create mode 100755 tools/testing/selftests/net/link_netns.py
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 01/11] rtnetlink: Lookup device in target netns when creating link
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops Xiao Liang
` (7 subsequent siblings)
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
When creating link, lookup for existing device in target net namespace
instead of current one.
For example, two links created by:
# ip link add dummy1 type dummy
# ip link add netns ns1 dummy1 type dummy
should have no conflict since they are in different namespaces.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
---
net/core/rtnetlink.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index cb7fad8d1f95..3b3398283a8f 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3866,20 +3866,26 @@ static int __rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh,
{
struct nlattr ** const tb = tbs->tb;
struct net *net = sock_net(skb->sk);
+ struct net *device_net;
struct net_device *dev;
struct ifinfomsg *ifm;
bool link_specified;
+ /* When creating, lookup for existing device in target net namespace */
+ device_net = (nlh->nlmsg_flags & NLM_F_CREATE) &&
+ (nlh->nlmsg_flags & NLM_F_EXCL) ?
+ tgt_net : net;
+
ifm = nlmsg_data(nlh);
if (ifm->ifi_index > 0) {
link_specified = true;
- dev = __dev_get_by_index(net, ifm->ifi_index);
+ dev = __dev_get_by_index(device_net, ifm->ifi_index);
} else if (ifm->ifi_index < 0) {
NL_SET_ERR_MSG(extack, "ifindex can't be negative");
return -EINVAL;
} else if (tb[IFLA_IFNAME] || tb[IFLA_ALT_IFNAME]) {
link_specified = true;
- dev = rtnl_dev_get(net, tb);
+ dev = rtnl_dev_get(device_net, tb);
} else {
link_specified = false;
dev = NULL;
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 01/11] rtnetlink: Lookup device in target netns when creating link Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-13 6:55 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 04/11] ieee802154: 6lowpan: Validate " Xiao Liang
` (6 subsequent siblings)
8 siblings, 1 reply; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
These netdevice drivers already uses netns parameter in newlink()
callback. Convert them to use rtnl_newlink_link_net() or
rtnl_newlink_peer_net() for clarity and deprecate params->net.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
drivers/infiniband/ulp/ipoib/ipoib_netlink.c | 4 ++--
drivers/net/amt.c | 6 +++---
drivers/net/bareudp.c | 4 ++--
drivers/net/can/vxcan.c | 2 +-
drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c | 4 ++--
drivers/net/geneve.c | 4 ++--
drivers/net/gtp.c | 6 +++---
drivers/net/ipvlan/ipvlan_main.c | 4 ++--
drivers/net/macsec.c | 4 ++--
drivers/net/macvlan.c | 4 ++--
drivers/net/netkit.c | 2 +-
drivers/net/pfcp.c | 6 +++---
drivers/net/ppp/ppp_generic.c | 4 ++--
drivers/net/veth.c | 2 +-
drivers/net/vxlan/vxlan_core.c | 4 ++--
drivers/net/wireguard/device.c | 4 ++--
drivers/net/wireless/virtual/virt_wifi.c | 4 ++--
drivers/net/wwan/wwan_core.c | 2 +-
net/8021q/vlan_netlink.c | 4 ++--
net/hsr/hsr_netlink.c | 8 ++++----
20 files changed, 41 insertions(+), 41 deletions(-)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
index 16cb8ced9f35..53db7c8191e3 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
@@ -101,8 +101,8 @@ static int ipoib_new_child_link(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct net_device *pdev;
struct ipoib_dev_priv *ppriv;
@@ -112,7 +112,7 @@ static int ipoib_new_child_link(struct net_device *dev,
if (!tb[IFLA_LINK])
return -EINVAL;
- pdev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ pdev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (!pdev || pdev->type != ARPHRD_INFINIBAND)
return -ENODEV;
diff --git a/drivers/net/amt.c b/drivers/net/amt.c
index 96b7ec9a2c13..53899b70fae1 100644
--- a/drivers/net/amt.c
+++ b/drivers/net/amt.c
@@ -3165,13 +3165,13 @@ static int amt_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct amt_dev *amt = netdev_priv(dev);
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = params->net;
int err = -EINVAL;
- amt->net = net;
+ amt->net = link_net;
amt->mode = nla_get_u32(data[IFLA_AMT_MODE]);
if (data[IFLA_AMT_MAX_TUNNELS] &&
@@ -3186,7 +3186,7 @@ static int amt_newlink(struct net_device *dev,
amt->hash_buckets = AMT_HSIZE;
amt->nr_tunnels = 0;
get_random_bytes(&amt->hash_seed, sizeof(amt->hash_seed));
- amt->stream_dev = dev_get_by_index(net,
+ amt->stream_dev = dev_get_by_index(link_net,
nla_get_u32(data[IFLA_AMT_LINK]));
if (!amt->stream_dev) {
NL_SET_ERR_MSG_ATTR(extack, tb[IFLA_AMT_LINK],
diff --git a/drivers/net/bareudp.c b/drivers/net/bareudp.c
index fc21dcfb4848..d1473c5f8eef 100644
--- a/drivers/net/bareudp.c
+++ b/drivers/net/bareudp.c
@@ -702,9 +702,9 @@ static int bareudp_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = params->net;
struct bareudp_conf conf;
int err;
@@ -712,7 +712,7 @@ static int bareudp_newlink(struct net_device *dev,
if (err)
return err;
- err = bareudp_configure(net, dev, &conf, extack);
+ err = bareudp_configure(link_net, dev, &conf, extack);
if (err)
return err;
diff --git a/drivers/net/can/vxcan.c b/drivers/net/can/vxcan.c
index 6f8ebb1cfd7b..99a78a757167 100644
--- a/drivers/net/can/vxcan.c
+++ b/drivers/net/can/vxcan.c
@@ -176,8 +176,8 @@ static int vxcan_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *peer_net = rtnl_newlink_peer_net(params);
struct nlattr **data = params->data;
- struct net *peer_net = params->net;
struct nlattr **tb = params->tb;
struct vxcan_priv *priv;
struct net_device *peer;
diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
index 8151e91395e2..ab7e5b6649b2 100644
--- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
+++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
@@ -122,8 +122,8 @@ static int rmnet_newlink(struct net_device *dev,
struct netlink_ext_ack *extack)
{
u32 data_format = RMNET_FLAGS_INGRESS_DEAGGREGATION;
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct net_device *real_dev;
int mode = RMNET_EPMODE_VND;
@@ -137,7 +137,7 @@ static int rmnet_newlink(struct net_device *dev,
return -EINVAL;
}
- real_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ real_dev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (!real_dev) {
NL_SET_ERR_MSG_MOD(extack, "link does not exist");
return -ENODEV;
diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index d373a851930c..c7700deefb00 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -1618,9 +1618,9 @@ static int geneve_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = params->net;
struct geneve_config cfg = {
.df = GENEVE_DF_UNSET,
.use_udp6_rx_checksums = false,
@@ -1634,7 +1634,7 @@ static int geneve_newlink(struct net_device *dev,
if (err)
return err;
- err = geneve_configure(net, dev, extack, &cfg);
+ err = geneve_configure(link_net, dev, extack, &cfg);
if (err)
return err;
diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c
index d6d4326cb908..2cf43e7d0edc 100644
--- a/drivers/net/gtp.c
+++ b/drivers/net/gtp.c
@@ -1466,8 +1466,8 @@ static int gtp_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
unsigned int role = GTP_ROLE_GGSN;
struct gtp_dev *gtp;
struct gtp_net *gn;
@@ -1498,7 +1498,7 @@ static int gtp_newlink(struct net_device *dev,
gtp->restart_count = nla_get_u8_default(data[IFLA_GTP_RESTART_COUNT],
0);
- gtp->net = src_net;
+ gtp->net = link_net;
err = gtp_hashtable_new(gtp, hashsize);
if (err < 0)
@@ -1528,7 +1528,7 @@ static int gtp_newlink(struct net_device *dev,
goto out_encap;
}
- gn = net_generic(src_net, gtp_net_id);
+ gn = net_generic(link_net, gtp_net_id);
list_add(>p->list, &gn->gtp_dev_list);
dev->priv_destructor = gtp_destructor;
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 19ce19ca7e32..b56144ca2fde 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -535,9 +535,9 @@ static int ipvlan_nl_fillinfo(struct sk_buff *skb,
int ipvlan_link_new(struct net_device *dev, struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct ipvl_dev *ipvlan = netdev_priv(dev);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct ipvl_port *port;
struct net_device *phy_dev;
@@ -547,7 +547,7 @@ int ipvlan_link_new(struct net_device *dev, struct rtnl_newlink_params *params,
if (!tb[IFLA_LINK])
return -EINVAL;
- phy_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ phy_dev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (!phy_dev)
return -ENODEV;
diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c
index 1869b0513f57..4de5d63fd577 100644
--- a/drivers/net/macsec.c
+++ b/drivers/net/macsec.c
@@ -4145,10 +4145,10 @@ static int macsec_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct macsec_dev *macsec = macsec_priv(dev);
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = params->net;
rx_handler_func_t *rx_handler;
u8 icv_len = MACSEC_DEFAULT_ICV_LEN;
struct net_device *real_dev;
@@ -4157,7 +4157,7 @@ static int macsec_newlink(struct net_device *dev,
if (!tb[IFLA_LINK])
return -EINVAL;
- real_dev = __dev_get_by_index(net, nla_get_u32(tb[IFLA_LINK]));
+ real_dev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (!real_dev)
return -ENODEV;
if (real_dev->type != ARPHRD_ETHER)
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index f903b414eaeb..4e9d54be887c 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1444,9 +1444,9 @@ int macvlan_common_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct macvlan_dev *vlan = netdev_priv(dev);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct net_device *lowerdev;
struct macvlan_port *port;
@@ -1457,7 +1457,7 @@ int macvlan_common_newlink(struct net_device *dev,
if (!tb[IFLA_LINK])
return -EINVAL;
- lowerdev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ lowerdev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (lowerdev == NULL)
return -ENODEV;
diff --git a/drivers/net/netkit.c b/drivers/net/netkit.c
index 640a2dbbbd28..751347392570 100644
--- a/drivers/net/netkit.c
+++ b/drivers/net/netkit.c
@@ -331,13 +331,13 @@ static int netkit_new_link(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *peer_net = rtnl_newlink_peer_net(params);
enum netkit_scrub scrub_prim = NETKIT_SCRUB_DEFAULT;
enum netkit_scrub scrub_peer = NETKIT_SCRUB_DEFAULT;
struct nlattr *peer_tb[IFLA_MAX + 1], **tbp, *attr;
enum netkit_action policy_prim = NETKIT_PASS;
enum netkit_action policy_peer = NETKIT_PASS;
struct nlattr **data = params->data;
- struct net *peer_net = params->net;
enum netkit_mode mode = NETKIT_L3;
unsigned char ifname_assign_type;
struct nlattr **tb = params->tb;
diff --git a/drivers/net/pfcp.c b/drivers/net/pfcp.c
index 7b0575940e1d..f873a92d2445 100644
--- a/drivers/net/pfcp.c
+++ b/drivers/net/pfcp.c
@@ -188,12 +188,12 @@ static int pfcp_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct pfcp_dev *pfcp = netdev_priv(dev);
- struct net *net = params->net;
struct pfcp_net *pn;
int err;
- pfcp->net = net;
+ pfcp->net = link_net;
err = pfcp_add_sock(pfcp);
if (err) {
@@ -207,7 +207,7 @@ static int pfcp_newlink(struct net_device *dev,
goto exit_del_pfcp_sock;
}
- pn = net_generic(net, pfcp_net_id);
+ pn = net_generic(link_net, pfcp_net_id);
list_add(&pfcp->list, &pn->pfcp_dev_list);
netdev_dbg(dev, "registered new PFCP interface\n");
diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index b3340f8a6149..6220866258fc 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -1307,8 +1307,8 @@ static int ppp_nl_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct ppp_config conf = {
.unit = -1,
@@ -1346,7 +1346,7 @@ static int ppp_nl_newlink(struct net_device *dev,
if (!tb[IFLA_IFNAME] || !nla_len(tb[IFLA_IFNAME]) || !*(char *)nla_data(tb[IFLA_IFNAME]))
conf.ifname_is_set = false;
- err = ppp_dev_configure(src_net, dev, &conf);
+ err = ppp_dev_configure(link_net, dev, &conf);
out_unlock:
mutex_unlock(&ppp_mutex);
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 7dfda89f072f..ba3ae2d8092f 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -1769,8 +1769,8 @@ static int veth_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *peer_net = rtnl_newlink_peer_net(params);
struct nlattr **data = params->data;
- struct net *peer_net = params->net;
struct nlattr **tb = params->tb;
int err;
struct net_device *peer;
diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c
index 69ee76172cd2..d342ba899a69 100644
--- a/drivers/net/vxlan/vxlan_core.c
+++ b/drivers/net/vxlan/vxlan_core.c
@@ -4401,8 +4401,8 @@ static int vxlan_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct vxlan_config conf;
int err;
@@ -4411,7 +4411,7 @@ static int vxlan_newlink(struct net_device *dev,
if (err)
return err;
- return __vxlan_dev_create(src_net, dev, &conf, extack);
+ return __vxlan_dev_create(link_net, dev, &conf, extack);
}
static int vxlan_changelink(struct net_device *dev, struct nlattr *tb[],
diff --git a/drivers/net/wireguard/device.c b/drivers/net/wireguard/device.c
index 404cf05bd72b..c496d35b266d 100644
--- a/drivers/net/wireguard/device.c
+++ b/drivers/net/wireguard/device.c
@@ -311,11 +311,11 @@ static int wg_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct wg_device *wg = netdev_priv(dev);
- struct net *src_net = params->net;
int ret = -ENOMEM;
- rcu_assign_pointer(wg->creating_net, src_net);
+ rcu_assign_pointer(wg->creating_net, link_net);
init_rwsem(&wg->static_identity.lock);
mutex_init(&wg->socket_update_lock);
mutex_init(&wg->device_update_lock);
diff --git a/drivers/net/wireless/virtual/virt_wifi.c b/drivers/net/wireless/virtual/virt_wifi.c
index 26905b2b3ba3..f9d11a023313 100644
--- a/drivers/net/wireless/virtual/virt_wifi.c
+++ b/drivers/net/wireless/virtual/virt_wifi.c
@@ -524,7 +524,7 @@ static int virt_wifi_newlink(struct net_device *dev,
struct netlink_ext_ack *extack)
{
struct virt_wifi_netdev_priv *priv = netdev_priv(dev);
- struct net *src_net = params->net;
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **tb = params->tb;
int err;
@@ -534,7 +534,7 @@ static int virt_wifi_newlink(struct net_device *dev,
netif_carrier_off(dev);
priv->upperdev = dev;
- priv->lowerdev = __dev_get_by_index(src_net,
+ priv->lowerdev = __dev_get_by_index(link_net,
nla_get_u32(tb[IFLA_LINK]));
if (!priv->lowerdev)
diff --git a/drivers/net/wwan/wwan_core.c b/drivers/net/wwan/wwan_core.c
index a05c49b4e7f8..63a47d420bc5 100644
--- a/drivers/net/wwan/wwan_core.c
+++ b/drivers/net/wwan/wwan_core.c
@@ -1065,7 +1065,7 @@ static void wwan_create_default_link(struct wwan_device *wwandev,
struct nlattr *tb[IFLA_MAX + 1], *linkinfo[IFLA_INFO_MAX + 1];
struct nlattr *data[IFLA_WWAN_MAX + 1];
struct rtnl_newlink_params params = {
- .net = &init_net,
+ .src_net = &init_net,
.tb = tb,
.data = data,
};
diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c
index 91df0f96e32a..a000b1ef0520 100644
--- a/net/8021q/vlan_netlink.c
+++ b/net/8021q/vlan_netlink.c
@@ -139,9 +139,9 @@ static int vlan_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct vlan_dev_priv *vlan = vlan_dev_priv(dev);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
struct net_device *real_dev;
unsigned int max_mtu;
@@ -158,7 +158,7 @@ static int vlan_newlink(struct net_device *dev,
return -EINVAL;
}
- real_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ real_dev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK]));
if (!real_dev) {
NL_SET_ERR_MSG_MOD(extack, "link does not exist");
return -ENODEV;
diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c
index 39add538ba99..b120470246cc 100644
--- a/net/hsr/hsr_netlink.c
+++ b/net/hsr/hsr_netlink.c
@@ -33,8 +33,8 @@ static int hsr_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *link_net = rtnl_newlink_link_net(params);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
enum hsr_version proto_version;
unsigned char multicast_spec;
u8 proto = HSR_PROTOCOL_HSR;
@@ -48,7 +48,7 @@ static int hsr_newlink(struct net_device *dev,
NL_SET_ERR_MSG_MOD(extack, "Slave1 device not specified");
return -EINVAL;
}
- link[0] = __dev_get_by_index(src_net,
+ link[0] = __dev_get_by_index(link_net,
nla_get_u32(data[IFLA_HSR_SLAVE1]));
if (!link[0]) {
NL_SET_ERR_MSG_MOD(extack, "Slave1 does not exist");
@@ -58,7 +58,7 @@ static int hsr_newlink(struct net_device *dev,
NL_SET_ERR_MSG_MOD(extack, "Slave2 device not specified");
return -EINVAL;
}
- link[1] = __dev_get_by_index(src_net,
+ link[1] = __dev_get_by_index(link_net,
nla_get_u32(data[IFLA_HSR_SLAVE2]));
if (!link[1]) {
NL_SET_ERR_MSG_MOD(extack, "Slave2 does not exist");
@@ -71,7 +71,7 @@ static int hsr_newlink(struct net_device *dev,
}
if (data[IFLA_HSR_INTERLINK])
- interlink = __dev_get_by_index(src_net,
+ interlink = __dev_get_by_index(link_net,
nla_get_u32(data[IFLA_HSR_INTERLINK]));
if (interlink && interlink == link[0]) {
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 04/11] ieee802154: 6lowpan: Validate link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 01/11] rtnetlink: Lookup device in target netns when creating link Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 05/11] net: ip_tunnel: Use " Xiao Liang
` (5 subsequent siblings)
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
Device denoted by IFLA_LINK is in link_net (IFLA_LINK_NETNSID) or
source netns by design, but 6lowpan uses dev_net.
Note dev->netns_local is set to true and currently link_net is
implemented via a netns change. These together effectively reject
IFLA_LINK_NETNSID.
This patch adds a validation to ensure link_net is either NULL or
identical to dev_net. Thus it would be fine to continue using dev_net
when rtnetlink core begins to create devices directly in target netns.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
net/ieee802154/6lowpan/core.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c
index 704bf9e3f097..ee318d46817d 100644
--- a/net/ieee802154/6lowpan/core.c
+++ b/net/ieee802154/6lowpan/core.c
@@ -143,6 +143,8 @@ static int lowpan_newlink(struct net_device *ldev,
if (!tb[IFLA_LINK])
return -EINVAL;
+ if (params->link_net && !net_eq(params->link_net, dev_net(ldev)))
+ return -EINVAL;
/* find and hold wpan device */
wdev = dev_get_by_index(dev_net(ldev), nla_get_u32(tb[IFLA_LINK]));
if (!wdev)
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 05/11] net: ip_tunnel: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (2 preceding siblings ...)
2025-02-10 13:29 ` [PATCH net-next v9 04/11] ieee802154: 6lowpan: Validate " Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-13 6:20 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 06/11] net: ipv6: " Xiao Liang
` (4 subsequent siblings)
8 siblings, 1 reply; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
When link_net is set, use it as link netns instead of dev_net(). This
prepares for rtnetlink core to create device in target netns directly,
in which case the two namespaces may be different.
Convert common ip_tunnel_newlink() to accept an extra link netns
argument. Don't overwrite ip_tunnel.net in ip_tunnel_init().
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
include/net/ip_tunnels.h | 5 +++--
net/ipv4/ip_gre.c | 8 +++++---
net/ipv4/ip_tunnel.c | 10 ++++++----
net/ipv4/ip_vti.c | 3 ++-
net/ipv4/ipip.c | 3 ++-
5 files changed, 18 insertions(+), 11 deletions(-)
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 1aa31bdb2b31..ae1f2dda4533 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -406,8 +406,9 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
bool log_ecn_error);
int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[],
struct ip_tunnel_parm_kern *p, __u32 fwmark);
-int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
- struct ip_tunnel_parm_kern *p, __u32 fwmark);
+int ip_tunnel_newlink(struct net *net, struct net_device *dev,
+ struct nlattr *tb[], struct ip_tunnel_parm_kern *p,
+ __u32 fwmark);
void ip_tunnel_setup(struct net_device *dev, unsigned int net_id);
bool ip_tunnel_netlink_encap_parms(struct nlattr *data[],
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 1fe9b13d351c..26d15f907551 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -1413,7 +1413,8 @@ static int ipgre_newlink(struct net_device *dev,
err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark);
if (err < 0)
return err;
- return ip_tunnel_newlink(dev, tb, &p, fwmark);
+ return ip_tunnel_newlink(params->link_net ? : dev_net(dev), dev, tb, &p,
+ fwmark);
}
static int erspan_newlink(struct net_device *dev,
@@ -1433,7 +1434,8 @@ static int erspan_newlink(struct net_device *dev,
err = erspan_netlink_parms(dev, data, tb, &p, &fwmark);
if (err)
return err;
- return ip_tunnel_newlink(dev, tb, &p, fwmark);
+ return ip_tunnel_newlink(params->link_net ? : dev_net(dev), dev, tb, &p,
+ fwmark);
}
static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
@@ -1701,7 +1703,7 @@ static struct rtnl_link_ops erspan_link_ops __read_mostly = {
struct net_device *gretap_fb_dev_create(struct net *net, const char *name,
u8 name_assign_type)
{
- struct rtnl_newlink_params params = { .net = net };
+ struct rtnl_newlink_params params = { .src_net = net };
struct nlattr *tb[IFLA_MAX + 1];
struct net_device *dev;
LIST_HEAD(list_kill);
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index 09b73acf037a..618a50d5c0c2 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -1213,11 +1213,11 @@ void ip_tunnel_delete_nets(struct list_head *net_list, unsigned int id,
}
EXPORT_SYMBOL_GPL(ip_tunnel_delete_nets);
-int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
- struct ip_tunnel_parm_kern *p, __u32 fwmark)
+int ip_tunnel_newlink(struct net *net, struct net_device *dev,
+ struct nlattr *tb[], struct ip_tunnel_parm_kern *p,
+ __u32 fwmark)
{
struct ip_tunnel *nt;
- struct net *net = dev_net(dev);
struct ip_tunnel_net *itn;
int mtu;
int err;
@@ -1326,7 +1326,9 @@ int ip_tunnel_init(struct net_device *dev)
}
tunnel->dev = dev;
- tunnel->net = dev_net(dev);
+ if (!tunnel->net)
+ tunnel->net = dev_net(dev);
+
strscpy(tunnel->parms.name, dev->name);
iph->version = 4;
iph->ihl = 5;
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index b901bee03e6d..159b4473290e 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -585,7 +585,8 @@ static int vti_newlink(struct net_device *dev,
__u32 fwmark = 0;
vti_netlink_parms(data, &parms, &fwmark);
- return ip_tunnel_newlink(dev, tb, &parms, fwmark);
+ return ip_tunnel_newlink(params->link_net ? : dev_net(dev), dev, tb,
+ &parms, fwmark);
}
static int vti_changelink(struct net_device *dev, struct nlattr *tb[],
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index a8b844bcfc64..bab0bf90c908 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -455,7 +455,8 @@ static int ipip_newlink(struct net_device *dev,
}
ipip_netlink_parms(data, &p, &t->collect_md, &fwmark);
- return ip_tunnel_newlink(dev, tb, &p, fwmark);
+ return ip_tunnel_newlink(params->link_net ? : dev_net(dev), dev, tb, &p,
+ fwmark);
}
static int ipip_changelink(struct net_device *dev, struct nlattr *tb[],
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (3 preceding siblings ...)
2025-02-10 13:29 ` [PATCH net-next v9 05/11] net: ip_tunnel: Use " Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-13 7:05 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 07/11] net: xfrm: " Xiao Liang
` (3 subsequent siblings)
8 siblings, 1 reply; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
When link_net is set, use it as link netns instead of dev_net(). This
prepares for rtnetlink core to create device in target netns directly,
in which case the two namespaces may be different.
Set correct netns in priv before registering device, and avoid
overwriting it in ndo_init() path.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
net/ipv6/ip6_gre.c | 20 ++++++++++----------
net/ipv6/ip6_tunnel.c | 13 ++++++++-----
net/ipv6/ip6_vti.c | 10 ++++++----
net/ipv6/sit.c | 11 +++++++----
4 files changed, 31 insertions(+), 23 deletions(-)
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index 863852abe8ea..108600dc716f 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
tunnel = netdev_priv(dev);
tunnel->dev = dev;
- tunnel->net = dev_net(dev);
+ if (!tunnel->net)
+ tunnel->net = dev_net(dev);
strcpy(tunnel->parms.name, dev->name);
ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
@@ -1882,7 +1883,8 @@ static int ip6erspan_tap_init(struct net_device *dev)
tunnel = netdev_priv(dev);
tunnel->dev = dev;
- tunnel->net = dev_net(dev);
+ if (!tunnel->net)
+ tunnel->net = dev_net(dev);
strcpy(tunnel->parms.name, dev->name);
ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
@@ -1971,7 +1973,7 @@ static bool ip6gre_netlink_encap_parms(struct nlattr *data[],
return ret;
}
-static int ip6gre_newlink_common(struct net *src_net, struct net_device *dev,
+static int ip6gre_newlink_common(struct net *link_net, struct net_device *dev,
struct nlattr *tb[], struct nlattr *data[],
struct netlink_ext_ack *extack)
{
@@ -1992,7 +1994,7 @@ static int ip6gre_newlink_common(struct net *src_net, struct net_device *dev,
eth_hw_addr_random(dev);
nt->dev = dev;
- nt->net = dev_net(dev);
+ nt->net = link_net;
err = register_netdevice(dev);
if (err)
@@ -2009,11 +2011,10 @@ static int ip6gre_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *net = params->link_net ? : dev_net(dev);
struct ip6_tnl *nt = netdev_priv(dev);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
- struct net *net = dev_net(dev);
struct ip6gre_net *ign;
int err;
@@ -2028,7 +2029,7 @@ static int ip6gre_newlink(struct net_device *dev,
return -EEXIST;
}
- err = ip6gre_newlink_common(src_net, dev, tb, data, extack);
+ err = ip6gre_newlink_common(net, dev, tb, data, extack);
if (!err) {
ip6gre_tnl_link_config(nt, !tb[IFLA_MTU]);
ip6gre_tunnel_link_md(ign, nt);
@@ -2248,11 +2249,10 @@ static int ip6erspan_newlink(struct net_device *dev,
struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *net = params->link_net ? : dev_net(dev);
struct ip6_tnl *nt = netdev_priv(dev);
struct nlattr **data = params->data;
- struct net *src_net = params->net;
struct nlattr **tb = params->tb;
- struct net *net = dev_net(dev);
struct ip6gre_net *ign;
int err;
@@ -2268,7 +2268,7 @@ static int ip6erspan_newlink(struct net_device *dev,
return -EEXIST;
}
- err = ip6gre_newlink_common(src_net, dev, tb, data, extack);
+ err = ip6gre_newlink_common(net, dev, tb, data, extack);
if (!err) {
ip6erspan_tnl_link_config(nt, !tb[IFLA_MTU]);
ip6erspan_tunnel_link_md(ign, nt);
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 54b843d20870..2438dc627e02 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -253,8 +253,7 @@ static void ip6_dev_free(struct net_device *dev)
static int ip6_tnl_create2(struct net_device *dev)
{
struct ip6_tnl *t = netdev_priv(dev);
- struct net *net = dev_net(dev);
- struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
+ struct ip6_tnl_net *ip6n = net_generic(t->net, ip6_tnl_net_id);
int err;
dev->rtnl_link_ops = &ip6_link_ops;
@@ -1878,7 +1877,8 @@ ip6_tnl_dev_init_gen(struct net_device *dev)
int t_hlen;
t->dev = dev;
- t->net = dev_net(dev);
+ if (!t->net)
+ t->net = dev_net(dev);
ret = dst_cache_init(&t->dst_cache, GFP_KERNEL);
if (ret)
@@ -2008,13 +2008,16 @@ static int ip6_tnl_newlink(struct net_device *dev,
{
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = dev_net(dev);
- struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id);
struct ip_tunnel_encap ipencap;
+ struct ip6_tnl_net *ip6n;
struct ip6_tnl *nt, *t;
+ struct net *net;
int err;
+ net = params->link_net ? : dev_net(dev);
+ ip6n = net_generic(net, ip6_tnl_net_id);
nt = netdev_priv(dev);
+ nt->net = net;
if (ip_tunnel_netlink_encap_parms(data, &ipencap)) {
err = ip6_tnl_encap_setup(nt, &ipencap);
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c
index 993f85aeb882..4aa1e7821951 100644
--- a/net/ipv6/ip6_vti.c
+++ b/net/ipv6/ip6_vti.c
@@ -177,8 +177,7 @@ vti6_tnl_unlink(struct vti6_net *ip6n, struct ip6_tnl *t)
static int vti6_tnl_create2(struct net_device *dev)
{
struct ip6_tnl *t = netdev_priv(dev);
- struct net *net = dev_net(dev);
- struct vti6_net *ip6n = net_generic(net, vti6_net_id);
+ struct vti6_net *ip6n = net_generic(t->net, vti6_net_id);
int err;
dev->rtnl_link_ops = &vti6_link_ops;
@@ -925,7 +924,8 @@ static inline int vti6_dev_init_gen(struct net_device *dev)
struct ip6_tnl *t = netdev_priv(dev);
t->dev = dev;
- t->net = dev_net(dev);
+ if (!t->net)
+ t->net = dev_net(dev);
netdev_hold(dev, &t->dev_tracker, GFP_KERNEL);
netdev_lockdep_set_classes(dev);
return 0;
@@ -1002,13 +1002,15 @@ static int vti6_newlink(struct net_device *dev,
struct netlink_ext_ack *extack)
{
struct nlattr **data = params->data;
- struct net *net = dev_net(dev);
struct ip6_tnl *nt;
+ struct net *net;
+ net = params->link_net ? : dev_net(dev);
nt = netdev_priv(dev);
vti6_netlink_parms(data, &nt->parms);
nt->parms.proto = IPPROTO_IPV6;
+ nt->net = net;
if (vti6_locate(net, &nt->parms, 0))
return -EEXIST;
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index e2bd52cabdee..e870271ed04a 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -201,8 +201,7 @@ static void ipip6_tunnel_clone_6rd(struct net_device *dev, struct sit_net *sitn)
static int ipip6_tunnel_create(struct net_device *dev)
{
struct ip_tunnel *t = netdev_priv(dev);
- struct net *net = dev_net(dev);
- struct sit_net *sitn = net_generic(net, sit_net_id);
+ struct sit_net *sitn = net_generic(t->net, sit_net_id);
int err;
__dev_addr_set(dev, &t->parms.iph.saddr, 4);
@@ -270,6 +269,7 @@ static struct ip_tunnel *ipip6_tunnel_locate(struct net *net,
nt = netdev_priv(dev);
nt->parms = *parms;
+ nt->net = net;
if (ipip6_tunnel_create(dev) < 0)
goto failed_free;
@@ -1449,7 +1449,8 @@ static int ipip6_tunnel_init(struct net_device *dev)
int err;
tunnel->dev = dev;
- tunnel->net = dev_net(dev);
+ if (!tunnel->net)
+ tunnel->net = dev_net(dev);
strcpy(tunnel->parms.name, dev->name);
ipip6_tunnel_bind_dev(dev);
@@ -1556,15 +1557,17 @@ static int ipip6_newlink(struct net_device *dev,
{
struct nlattr **data = params->data;
struct nlattr **tb = params->tb;
- struct net *net = dev_net(dev);
struct ip_tunnel *nt;
struct ip_tunnel_encap ipencap;
#ifdef CONFIG_IPV6_SIT_6RD
struct ip_tunnel_6rd ip6rd;
#endif
+ struct net *net;
int err;
+ net = params->link_net ? : dev_net(dev);
nt = netdev_priv(dev);
+ nt->net = net;
if (ip_tunnel_netlink_encap_parms(data, &ipencap)) {
err = ip_tunnel_encap_setup(nt, &ipencap);
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 07/11] net: xfrm: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (4 preceding siblings ...)
2025-02-10 13:29 ` [PATCH net-next v9 06/11] net: ipv6: " Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 08/11] rtnetlink: Remove "net" from newlink params Xiao Liang
` (2 subsequent siblings)
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
When link_net is set, use it as link netns instead of dev_net(). This
prepares for rtnetlink core to create device in target netns directly,
in which case the two namespaces may be different.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
net/xfrm/xfrm_interface_core.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/net/xfrm/xfrm_interface_core.c b/net/xfrm/xfrm_interface_core.c
index 5659a6cadd51..622445f041d3 100644
--- a/net/xfrm/xfrm_interface_core.c
+++ b/net/xfrm/xfrm_interface_core.c
@@ -242,10 +242,9 @@ static void xfrmi_dev_free(struct net_device *dev)
gro_cells_destroy(&xi->gro_cells);
}
-static int xfrmi_create(struct net_device *dev)
+static int xfrmi_create(struct net *net, struct net_device *dev)
{
struct xfrm_if *xi = netdev_priv(dev);
- struct net *net = dev_net(dev);
struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
int err;
@@ -819,11 +818,12 @@ static int xfrmi_newlink(struct net_device *dev,
struct netlink_ext_ack *extack)
{
struct nlattr **data = params->data;
- struct net *net = dev_net(dev);
struct xfrm_if_parms p = {};
struct xfrm_if *xi;
+ struct net *net;
int err;
+ net = params->link_net ? : dev_net(dev);
xfrmi_netlink_parms(data, &p);
if (p.collect_md) {
struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
@@ -852,7 +852,7 @@ static int xfrmi_newlink(struct net_device *dev,
xi->net = net;
xi->dev = dev;
- err = xfrmi_create(dev);
+ err = xfrmi_create(net, dev);
return err;
}
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 08/11] rtnetlink: Remove "net" from newlink params
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (5 preceding siblings ...)
2025-02-10 13:29 ` [PATCH net-next v9 07/11] net: xfrm: " Xiao Liang
@ 2025-02-10 13:29 ` Xiao Liang
2025-02-10 13:30 ` [PATCH net-next v9 09/11] rtnetlink: Create link directly in target net namespace Xiao Liang
2025-02-10 13:30 ` [PATCH net-next v9 10/11] selftests: net: Add python context manager for netns entering Xiao Liang
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:29 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
Now that devices have been converted to use the specific netns instead
of ambiguous "net", let's remove it from newlink parameters.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
include/net/rtnetlink.h | 2 --
net/core/rtnetlink.c | 6 ------
2 files changed, 8 deletions(-)
diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h
index 00c086ca0c11..dd51240431b8 100644
--- a/include/net/rtnetlink.h
+++ b/include/net/rtnetlink.h
@@ -72,7 +72,6 @@ static inline int rtnl_msg_family(const struct nlmsghdr *nlh)
/**
* struct rtnl_newlink_params - parameters of rtnl_link_ops::newlink()
*
- * @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
@@ -80,7 +79,6 @@ static inline int rtnl_msg_family(const struct nlmsghdr *nlh)
* @data: IFLA_INFO_DATA attributes
*/
struct rtnl_newlink_params {
- struct net *net;
struct net *src_net;
struct net *link_net;
struct net *peer_net;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 7e98f3cd102b..e9af0775fa6b 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3797,12 +3797,6 @@ static int rtnl_newlink_create(struct sk_buff *skb, struct ifinfomsg *ifm,
dev->ifindex = ifm->ifi_index;
- params.net = params.src_net;
- if (link_net)
- params.net = link_net;
- if (peer_net)
- params.net = peer_net;
-
if (ops->newlink)
err = ops->newlink(dev, ¶ms, extack);
else
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 09/11] rtnetlink: Create link directly in target net namespace
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (6 preceding siblings ...)
2025-02-10 13:29 ` [PATCH net-next v9 08/11] rtnetlink: Remove "net" from newlink params Xiao Liang
@ 2025-02-10 13:30 ` Xiao Liang
2025-02-10 13:30 ` [PATCH net-next v9 10/11] selftests: net: Add python context manager for netns entering Xiao Liang
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:30 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
Make rtnl_newlink_create() create device in target namespace directly.
Avoid extra netns change when link netns is provided.
Device drivers has been converted to be aware of link netns, that is not
assuming device netns is and link netns is the same when ops->newlink()
is called.
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
net/core/rtnetlink.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e9af0775fa6b..a11b2c1f0985 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3788,8 +3788,8 @@ static int rtnl_newlink_create(struct sk_buff *skb, struct ifinfomsg *ifm,
name_assign_type = NET_NAME_ENUM;
}
- dev = rtnl_create_link(link_net ? : tgt_net, ifname,
- name_assign_type, ops, tb, extack);
+ dev = rtnl_create_link(tgt_net, ifname, name_assign_type, ops, tb,
+ extack);
if (IS_ERR(dev)) {
err = PTR_ERR(dev);
goto out;
@@ -3809,11 +3809,6 @@ static int rtnl_newlink_create(struct sk_buff *skb, struct ifinfomsg *ifm,
err = rtnl_configure_link(dev, ifm, portid, nlh);
if (err < 0)
goto out_unregister;
- if (link_net) {
- err = dev_change_net_namespace(dev, tgt_net, ifname);
- if (err < 0)
- goto out_unregister;
- }
if (tb[IFLA_MASTER]) {
err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]), extack);
if (err)
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH net-next v9 10/11] selftests: net: Add python context manager for netns entering
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
` (7 preceding siblings ...)
2025-02-10 13:30 ` [PATCH net-next v9 09/11] rtnetlink: Create link directly in target net namespace Xiao Liang
@ 2025-02-10 13:30 ` Xiao Liang
8 siblings, 0 replies; 17+ messages in thread
From: Xiao Liang @ 2025-02-10 13:30 UTC (permalink / raw)
To: netdev, linux-kselftest, Kuniyuki Iwashima, Jakub Kicinski
Cc: David S. Miller, David Ahern, Eric Dumazet, Paolo Abeni,
Andrew Lunn, Simon Horman, Shuah Khan, Donald Hunter,
Alexander Aring, Stefan Schmidt, Miquel Raynal, Steffen Klassert,
Herbert Xu, linux-rdma, linux-can, osmocom-net-gprs, bpf,
linux-ppp, wireguard, linux-wireless, b.a.t.m.a.n, bridge,
linux-wpan, linux-kernel
Change netns of current thread and switch back on context exit.
For example:
with NetNSEnter("ns1"):
ip("link add dummy0 type dummy")
The command be executed in netns "ns1".
Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
---
tools/testing/selftests/net/lib/py/__init__.py | 2 +-
tools/testing/selftests/net/lib/py/netns.py | 18 ++++++++++++++++++
2 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/lib/py/__init__.py b/tools/testing/selftests/net/lib/py/__init__.py
index 54d8f5eba810..e2d6c7b63019 100644
--- a/tools/testing/selftests/net/lib/py/__init__.py
+++ b/tools/testing/selftests/net/lib/py/__init__.py
@@ -2,7 +2,7 @@
from .consts import KSRC
from .ksft import *
-from .netns import NetNS
+from .netns import NetNS, NetNSEnter
from .nsim import *
from .utils import *
from .ynl import NlError, YnlFamily, EthtoolFamily, NetdevFamily, RtnlFamily
diff --git a/tools/testing/selftests/net/lib/py/netns.py b/tools/testing/selftests/net/lib/py/netns.py
index ecff85f9074f..8e9317044eef 100644
--- a/tools/testing/selftests/net/lib/py/netns.py
+++ b/tools/testing/selftests/net/lib/py/netns.py
@@ -1,9 +1,12 @@
# SPDX-License-Identifier: GPL-2.0
from .utils import ip
+import ctypes
import random
import string
+libc = ctypes.cdll.LoadLibrary('libc.so.6')
+
class NetNS:
def __init__(self, name=None):
@@ -29,3 +32,18 @@ class NetNS:
def __repr__(self):
return f"NetNS({self.name})"
+
+
+class NetNSEnter:
+ def __init__(self, ns_name):
+ self.ns_path = f"/run/netns/{ns_name}"
+
+ def __enter__(self):
+ self.saved = open("/proc/thread-self/ns/net")
+ with open(self.ns_path) as ns_file:
+ libc.setns(ns_file.fileno(), 0)
+ return self
+
+ def __exit__(self, exc_type, exc_value, traceback):
+ libc.setns(self.saved.fileno(), 0)
+ self.saved.close()
--
2.48.1
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 05/11] net: ip_tunnel: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 ` [PATCH net-next v9 05/11] net: ip_tunnel: Use " Xiao Liang
@ 2025-02-13 6:20 ` Kuniyuki Iwashima
0 siblings, 0 replies; 17+ messages in thread
From: Kuniyuki Iwashima @ 2025-02-13 6:20 UTC (permalink / raw)
To: shaw.leon
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba, kuniyu,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
From: Xiao Liang <shaw.leon@gmail.com>
Date: Mon, 10 Feb 2025 21:29:56 +0800
> When link_net is set, use it as link netns instead of dev_net(). This
> prepares for rtnetlink core to create device in target netns directly,
> in which case the two namespaces may be different.
>
> Convert common ip_tunnel_newlink() to accept an extra link netns
> argument. Don't overwrite ip_tunnel.net in ip_tunnel_init().
Why... ? see a comment below.
[...]
> diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
> index 1fe9b13d351c..26d15f907551 100644
> --- a/net/ipv4/ip_gre.c
> +++ b/net/ipv4/ip_gre.c
> @@ -1413,7 +1413,8 @@ static int ipgre_newlink(struct net_device *dev,
> err = ipgre_netlink_parms(dev, data, tb, &p, &fwmark);
> if (err < 0)
> return err;
> - return ip_tunnel_newlink(dev, tb, &p, fwmark);
> + return ip_tunnel_newlink(params->link_net ? : dev_net(dev), dev, tb, &p,
This is duplicate at all call sites, let's move it into
ip_tunnel_newlink() by passing params.
> + fwmark);
> }
>
> static int erspan_newlink(struct net_device *dev,
>
>
> diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
> index 09b73acf037a..618a50d5c0c2 100644
> --- a/net/ipv4/ip_tunnel.c
> +++ b/net/ipv4/ip_tunnel.c
> @@ -1213,11 +1213,11 @@ void ip_tunnel_delete_nets(struct list_head *net_list, unsigned int id,
> }
> EXPORT_SYMBOL_GPL(ip_tunnel_delete_nets);
>
> -int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[],
> - struct ip_tunnel_parm_kern *p, __u32 fwmark)
> +int ip_tunnel_newlink(struct net *net, struct net_device *dev,
> + struct nlattr *tb[], struct ip_tunnel_parm_kern *p,
> + __u32 fwmark)
> {
> struct ip_tunnel *nt;
> - struct net *net = dev_net(dev);
> struct ip_tunnel_net *itn;
> int mtu;
> int err;
> @@ -1326,7 +1326,9 @@ int ip_tunnel_init(struct net_device *dev)
> }
>
> tunnel->dev = dev;
> - tunnel->net = dev_net(dev);
> + if (!tunnel->net)
> + tunnel->net = dev_net(dev);
Isn't tunnel->net always non-NULL ?
ip_tunnel_newlink
-> netdev_priv(dev)->net = net
-> register_netdevice(dev)
-> dev->netdev_ops->ndo_init(dev)
-> ip_tunnel_init(dev)
-> netdev_priv(dev)->net = dev_net(dev)
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 ` [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops Xiao Liang
@ 2025-02-13 6:55 ` Kuniyuki Iwashima
0 siblings, 0 replies; 17+ messages in thread
From: Kuniyuki Iwashima @ 2025-02-13 6:55 UTC (permalink / raw)
To: shaw.leon
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba, kuniyu,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
From: Xiao Liang <shaw.leon@gmail.com>
Date: Mon, 10 Feb 2025 21:29:54 +0800
> These netdevice drivers already uses netns parameter in newlink()
> callback. Convert them to use rtnl_newlink_link_net() or
> rtnl_newlink_peer_net() for clarity and deprecate params->net.
>
> Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
> diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
> index 8151e91395e2..ab7e5b6649b2 100644
> --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
> +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_config.c
> @@ -122,8 +122,8 @@ static int rmnet_newlink(struct net_device *dev,
> struct netlink_ext_ack *extack)
> {
> u32 data_format = RMNET_FLAGS_INGRESS_DEAGGREGATION;
> + struct net *link_net = rtnl_newlink_link_net(params);
nit: reverse xmas tree
> struct nlattr **data = params->data;
> - struct net *src_net = params->net;
> struct nlattr **tb = params->tb;
> struct net_device *real_dev;
> int mode = RMNET_EPMODE_VND;
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
2025-02-10 13:29 ` [PATCH net-next v9 06/11] net: ipv6: " Xiao Liang
@ 2025-02-13 7:05 ` Kuniyuki Iwashima
[not found] ` <CABAhCOT8sCV4RgBWwfYjCw2xoZbdiYG8yuWReigx-u5DibTaiA@mail.gmail.com>
0 siblings, 1 reply; 17+ messages in thread
From: Kuniyuki Iwashima @ 2025-02-13 7:05 UTC (permalink / raw)
To: shaw.leon
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba, kuniyu,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
From: Xiao Liang <shaw.leon@gmail.com>
Date: Mon, 10 Feb 2025 21:29:57 +0800
> When link_net is set, use it as link netns instead of dev_net(). This
> prepares for rtnetlink core to create device in target netns directly,
> in which case the two namespaces may be different.
>
> Set correct netns in priv before registering device, and avoid
> overwriting it in ndo_init() path.
>
> Signed-off-by: Xiao Liang <shaw.leon@gmail.com>
> ---
> net/ipv6/ip6_gre.c | 20 ++++++++++----------
> net/ipv6/ip6_tunnel.c | 13 ++++++++-----
> net/ipv6/ip6_vti.c | 10 ++++++----
> net/ipv6/sit.c | 11 +++++++----
> 4 files changed, 31 insertions(+), 23 deletions(-)
>
> diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
> index 863852abe8ea..108600dc716f 100644
> --- a/net/ipv6/ip6_gre.c
> +++ b/net/ipv6/ip6_gre.c
> @@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
> tunnel = netdev_priv(dev);
>
> tunnel->dev = dev;
> - tunnel->net = dev_net(dev);
> + if (!tunnel->net)
> + tunnel->net = dev_net(dev);
Same question as patch 5 for here and other parts.
Do we need this check and assignment ?
ip6gre_newlink_common
-> nt->net = dev_net(dev)
-> register_netdevice
-> ndo_init / ip6gre_tunnel_init()
-> ip6gre_tunnel_init_common
-> tunnel->net = dev_net(dev)
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
[not found] ` <CABAhCOT8sCV4RgBWwfYjCw2xoZbdiYG8yuWReigx-u5DibTaiA@mail.gmail.com>
@ 2025-02-13 9:55 ` Xiao Liang
2025-02-13 11:00 ` Kuniyuki Iwashima
0 siblings, 1 reply; 17+ messages in thread
From: Xiao Liang @ 2025-02-13 9:55 UTC (permalink / raw)
To: Kuniyuki Iwashima
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
On Thu, Feb 13, 2025 at 4:37 PM Xiao Liang <shaw.leon@gmail.com> wrote:
>
> On Thu, Feb 13, 2025 at 3:05 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
> >
> [...]
> > > diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
> > > index 863852abe8ea..108600dc716f 100644
> > > --- a/net/ipv6/ip6_gre.c
> > > +++ b/net/ipv6/ip6_gre.c
> > > @@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
> > > tunnel = netdev_priv(dev);
> > >
> > > tunnel->dev = dev;
> > > - tunnel->net = dev_net(dev);
> > > + if (!tunnel->net)
> > > + tunnel->net = dev_net(dev);
> >
> > Same question as patch 5 for here and other parts.
> > Do we need this check and assignment ?
> >
> > ip6gre_newlink_common
> > -> nt->net = dev_net(dev)
> > -> register_netdevice
> > -> ndo_init / ip6gre_tunnel_init()
> > -> ip6gre_tunnel_init_common
> > -> tunnel->net = dev_net(dev)
>
> Will remove this line.
However, fb tunnel of ip6_tunnel, ip6_vti and sit can have
tunnel->net == NULL here. Take ip6_tunnel for example:
ip6_tnl_init_net()
-> ip6_fb_tnl_dev_init()
-> register_netdev()
-> register_netdevice()
-> ip6_tnl_dev_init()
This code path (including ip6_fb_tnl_dev_init()) doesn't set
tunnel->net. But for ip6_gre, ip6gre_fb_tunnel_init() does.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
2025-02-13 9:55 ` Xiao Liang
@ 2025-02-13 11:00 ` Kuniyuki Iwashima
2025-02-14 9:22 ` Xiao Liang
0 siblings, 1 reply; 17+ messages in thread
From: Kuniyuki Iwashima @ 2025-02-13 11:00 UTC (permalink / raw)
To: shaw.leon
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba, kuniyu,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
From: Xiao Liang <shaw.leon@gmail.com>
Date: Thu, 13 Feb 2025 17:55:32 +0800
> On Thu, Feb 13, 2025 at 4:37 PM Xiao Liang <shaw.leon@gmail.com> wrote:
> >
> > On Thu, Feb 13, 2025 at 3:05 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
> > >
> > [...]
> > > > diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
> > > > index 863852abe8ea..108600dc716f 100644
> > > > --- a/net/ipv6/ip6_gre.c
> > > > +++ b/net/ipv6/ip6_gre.c
> > > > @@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
> > > > tunnel = netdev_priv(dev);
> > > >
> > > > tunnel->dev = dev;
> > > > - tunnel->net = dev_net(dev);
> > > > + if (!tunnel->net)
> > > > + tunnel->net = dev_net(dev);
> > >
> > > Same question as patch 5 for here and other parts.
> > > Do we need this check and assignment ?
> > >
> > > ip6gre_newlink_common
> > > -> nt->net = dev_net(dev)
> > > -> register_netdevice
> > > -> ndo_init / ip6gre_tunnel_init()
> > > -> ip6gre_tunnel_init_common
> > > -> tunnel->net = dev_net(dev)
> >
> > Will remove this line.
>
> However, fb tunnel of ip6_tunnel, ip6_vti and sit can have
> tunnel->net == NULL here. Take ip6_tunnel for example:
>
> ip6_tnl_init_net()
> -> ip6_fb_tnl_dev_init()
> -> register_netdev()
> -> register_netdevice()
> -> ip6_tnl_dev_init()
>
> This code path (including ip6_fb_tnl_dev_init()) doesn't set
> tunnel->net. But for ip6_gre, ip6gre_fb_tunnel_init() does.
Ah, okay. Then, let's set net in a single place, which would
be better than spreading net assignment and adding null check
in ->ndo_init(), and maybe apply the same to IPv4 tunnels ?
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
2025-02-13 11:00 ` Kuniyuki Iwashima
@ 2025-02-14 9:22 ` Xiao Liang
2025-02-14 10:51 ` Kuniyuki Iwashima
0 siblings, 1 reply; 17+ messages in thread
From: Xiao Liang @ 2025-02-14 9:22 UTC (permalink / raw)
To: Kuniyuki Iwashima
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
On Thu, Feb 13, 2025 at 7:00 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
>
> From: Xiao Liang <shaw.leon@gmail.com>
> Date: Thu, 13 Feb 2025 17:55:32 +0800
> > On Thu, Feb 13, 2025 at 4:37 PM Xiao Liang <shaw.leon@gmail.com> wrote:
> > >
> > > On Thu, Feb 13, 2025 at 3:05 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
> > > >
> > > [...]
> > > > > diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
> > > > > index 863852abe8ea..108600dc716f 100644
> > > > > --- a/net/ipv6/ip6_gre.c
> > > > > +++ b/net/ipv6/ip6_gre.c
> > > > > @@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
> > > > > tunnel = netdev_priv(dev);
> > > > >
> > > > > tunnel->dev = dev;
> > > > > - tunnel->net = dev_net(dev);
> > > > > + if (!tunnel->net)
> > > > > + tunnel->net = dev_net(dev);
> > > >
> > > > Same question as patch 5 for here and other parts.
> > > > Do we need this check and assignment ?
> > > >
> > > > ip6gre_newlink_common
> > > > -> nt->net = dev_net(dev)
> > > > -> register_netdevice
> > > > -> ndo_init / ip6gre_tunnel_init()
> > > > -> ip6gre_tunnel_init_common
> > > > -> tunnel->net = dev_net(dev)
> > >
> > > Will remove this line.
> >
> > However, fb tunnel of ip6_tunnel, ip6_vti and sit can have
> > tunnel->net == NULL here. Take ip6_tunnel for example:
> >
> > ip6_tnl_init_net()
> > -> ip6_fb_tnl_dev_init()
> > -> register_netdev()
> > -> register_netdevice()
> > -> ip6_tnl_dev_init()
> >
> > This code path (including ip6_fb_tnl_dev_init()) doesn't set
> > tunnel->net. But for ip6_gre, ip6gre_fb_tunnel_init() does.
>
> Ah, okay. Then, let's set net in a single place, which would
> be better than spreading net assignment and adding null check
> in ->ndo_init(), and maybe apply the same to IPv4 tunnels ?
Tunnels are created in three ways: a) rtnetlink newlink,
b) ioctl SIOCADDTUNNEL and c) during per netns init (fb).
The code paths don't have much in common, and refactoring
to set net in a single place is somewhat beyond the scope
of this series. But for now I think we could put a general rule:
net should be set prior to register_netdevice().
For IPv4 tunnels, tunnel->net of a) is set in ip_tunnel_newlink().
b) and c) are set in __ip_tunnel_create():
ip_tunnel_init_net() -> __ip_tunnel_create()
ip_tunnel_ctl() -> ip_tunnel_create() -> __ip_tunnel_create()
So net has already been initialized when register_netdevice()
is called.
But it varies for IPv6 tunnels. Some set net for a) or c) while
some don't. This patch has "fixed" for a). As for c) we can
adopt the way of ip6_gre - setting net in *_fb_tunnel_init(),
then remove the check in ndo_init().
Is it reasonable?
Thanks.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH net-next v9 06/11] net: ipv6: Use link netns in newlink() of rtnl_link_ops
2025-02-14 9:22 ` Xiao Liang
@ 2025-02-14 10:51 ` Kuniyuki Iwashima
0 siblings, 0 replies; 17+ messages in thread
From: Kuniyuki Iwashima @ 2025-02-14 10:51 UTC (permalink / raw)
To: shaw.leon
Cc: alex.aring, andrew+netdev, b.a.t.m.a.n, bpf, bridge, davem,
donald.hunter, dsahern, edumazet, herbert, horms, kuba, kuniyu,
linux-can, linux-kernel, linux-kselftest, linux-ppp, linux-rdma,
linux-wireless, linux-wpan, miquel.raynal, netdev,
osmocom-net-gprs, pabeni, shuah, stefan, steffen.klassert,
wireguard
From: Xiao Liang <shaw.leon@gmail.com>
Date: Fri, 14 Feb 2025 17:22:28 +0800
> On Thu, Feb 13, 2025 at 7:00 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
> >
> > From: Xiao Liang <shaw.leon@gmail.com>
> > Date: Thu, 13 Feb 2025 17:55:32 +0800
> > > On Thu, Feb 13, 2025 at 4:37 PM Xiao Liang <shaw.leon@gmail.com> wrote:
> > > >
> > > > On Thu, Feb 13, 2025 at 3:05 PM Kuniyuki Iwashima <kuniyu@amazon.com> wrote:
> > > > >
> > > > [...]
> > > > > > diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
> > > > > > index 863852abe8ea..108600dc716f 100644
> > > > > > --- a/net/ipv6/ip6_gre.c
> > > > > > +++ b/net/ipv6/ip6_gre.c
> > > > > > @@ -1498,7 +1498,8 @@ static int ip6gre_tunnel_init_common(struct net_device *dev)
> > > > > > tunnel = netdev_priv(dev);
> > > > > >
> > > > > > tunnel->dev = dev;
> > > > > > - tunnel->net = dev_net(dev);
> > > > > > + if (!tunnel->net)
> > > > > > + tunnel->net = dev_net(dev);
> > > > >
> > > > > Same question as patch 5 for here and other parts.
> > > > > Do we need this check and assignment ?
> > > > >
> > > > > ip6gre_newlink_common
> > > > > -> nt->net = dev_net(dev)
> > > > > -> register_netdevice
> > > > > -> ndo_init / ip6gre_tunnel_init()
> > > > > -> ip6gre_tunnel_init_common
> > > > > -> tunnel->net = dev_net(dev)
> > > >
> > > > Will remove this line.
> > >
> > > However, fb tunnel of ip6_tunnel, ip6_vti and sit can have
> > > tunnel->net == NULL here. Take ip6_tunnel for example:
> > >
> > > ip6_tnl_init_net()
> > > -> ip6_fb_tnl_dev_init()
> > > -> register_netdev()
> > > -> register_netdevice()
> > > -> ip6_tnl_dev_init()
> > >
> > > This code path (including ip6_fb_tnl_dev_init()) doesn't set
> > > tunnel->net. But for ip6_gre, ip6gre_fb_tunnel_init() does.
> >
> > Ah, okay. Then, let's set net in a single place, which would
> > be better than spreading net assignment and adding null check
> > in ->ndo_init(), and maybe apply the same to IPv4 tunnels ?
>
> Tunnels are created in three ways: a) rtnetlink newlink,
> b) ioctl SIOCADDTUNNEL and c) during per netns init (fb).
> The code paths don't have much in common, and refactoring
> to set net in a single place is somewhat beyond the scope
> of this series. But for now I think we could put a general rule:
> net should be set prior to register_netdevice().
>
> For IPv4 tunnels, tunnel->net of a) is set in ip_tunnel_newlink().
> b) and c) are set in __ip_tunnel_create():
> ip_tunnel_init_net() -> __ip_tunnel_create()
> ip_tunnel_ctl() -> ip_tunnel_create() -> __ip_tunnel_create()
> So net has already been initialized when register_netdevice()
> is called.
>
> But it varies for IPv6 tunnels. Some set net for a) or c) while
> some don't. This patch has "fixed" for a). As for c) we can
> adopt the way of ip6_gre - setting net in *_fb_tunnel_init(),
> then remove the check in ndo_init().
>
> Is it reasonable?
Yes, fair enough.
^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2025-05-21 17:28 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-02-10 13:29 [PATCH net-next v9 00/11] net: Improve netns handling in rtnetlink Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 01/11] rtnetlink: Lookup device in target netns when creating link Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 03/11] net: Use link netns in newlink() of rtnl_link_ops Xiao Liang
2025-02-13 6:55 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 04/11] ieee802154: 6lowpan: Validate " Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 05/11] net: ip_tunnel: Use " Xiao Liang
2025-02-13 6:20 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 06/11] net: ipv6: " Xiao Liang
2025-02-13 7:05 ` Kuniyuki Iwashima
[not found] ` <CABAhCOT8sCV4RgBWwfYjCw2xoZbdiYG8yuWReigx-u5DibTaiA@mail.gmail.com>
2025-02-13 9:55 ` Xiao Liang
2025-02-13 11:00 ` Kuniyuki Iwashima
2025-02-14 9:22 ` Xiao Liang
2025-02-14 10:51 ` Kuniyuki Iwashima
2025-02-10 13:29 ` [PATCH net-next v9 07/11] net: xfrm: " Xiao Liang
2025-02-10 13:29 ` [PATCH net-next v9 08/11] rtnetlink: Remove "net" from newlink params Xiao Liang
2025-02-10 13:30 ` [PATCH net-next v9 09/11] rtnetlink: Create link directly in target net namespace Xiao Liang
2025-02-10 13:30 ` [PATCH net-next v9 10/11] selftests: net: Add python context manager for netns entering Xiao Liang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).