Re: [PATCH net] net: Defer netdev KOBJ_ADD uevent until the device is published
From: Jakub Kicinski <kuba@kernel.org>
Date: 2026-08-04 21:09:59
Also in:
lkml
On Tue, 4 Aug 2026 18:51:01 +0300 Dragos Tatulea wrote:
netdev_register_kobject() calls device_add(), which emits KOBJ_ADD and
wakes udev, but register_netdevice() only makes the device findable by
name later, in list_netdevice(). A udev worker that reacts to the uevent
can therefore run against a device that no lookup can find yet.
This used to be harmless because the ethtool ioctl took the rtnl_lock
when looking the device up, and register_netdevice() runs under rtnl, so
the worker simply blocked until registration finished. The commit in the
fixes tag moved the lookup out from under rtnl for ops-locked drivers.
Now there is a short window in register_netdevice() between
netdev_register_kobject() until list_netdevice() when the device is not
findable by name.
This was reproduced with the mlx5 driver on a kernel with KASAN enabled
during devlink reload: systemd-udevd's net_driver builtin gets -ENODEV
from ETHTOOL_GDRVINFO, which was preventing interface renaming.
Suppress the uevent in netdev_register_kobject() and emit it from
register_netdevice() next to rtmsg_ifinfo(). This is the last point in
register_netdevice() where no error can happen, so only fully registered
devices are announced: the registration error paths never reach it, and
the device_del() that unwinds them stays silent as well, leaving
userspace with neither an add nor a remove.
Fixes: f994752b1127 ("net: ethtool: optionally skip rtnl_lock on IOCTL path")
Assisted-by: Claude:claude-opus-5
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Shahar Shitrit <redacted>LGTM, one nit..
quoted hunk ↗ jump to hunk
net/core/dev.c | 1 + net/core/net-sysfs.c | 16 ++++++++++++++++ net/core/net-sysfs.h | 1 + 3 files changed, 18 insertions(+)diff --git a/net/core/dev.c b/net/core/dev.c index c1c1be1a6962..32cc092d5d11 100644 --- a/net/core/dev.c +++ b/net/core/dev.c@@ -11499,6 +11499,7 @@ int register_netdevice(struct net_device *dev) * Prevent userspace races by waiting until the network * device is fully setup before sending notifications. */ + netdev_uevent_add(dev); if (!(dev->rtnl_link_ops && dev->rtnl_link_initializing)) rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U, GFP_KERNEL, 0, NULL);diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 0e71c9ed41e8..9a403259a3d0 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c@@ -2334,6 +2334,11 @@ int netdev_register_kobject(struct net_device *ndev) *groups++ = &wireless_group; #endif /* CONFIG_SYSFS */ + /* Hold back the KOBJ_ADD uevent until the device is discoverable + * by name.
nit: maybe just say "until the device is listed" ? Or there's some less than obvious relevance of the name lookup vs ifindex lookup?
quoted hunk ↗ jump to hunk
+ */ + dev_set_uevent_suppress(dev, 1); + error = device_add(dev); if (error) return error;@@ -2349,6 +2354,17 @@ int netdev_register_kobject(struct net_device *ndev) return error; } +/* Announce a fully registered device to userspace. This pairs with the uevent + * suppression from netdev_register_kobject();
nit: s/;/./ ?
+ */
+void netdev_uevent_add(struct net_device *ndev)
+{
+ struct device *dev = &ndev->dev;
+
+ dev_set_uevent_suppress(dev, 0);
+ kobject_uevent(&dev->kobj, KOBJ_ADD);
+}Please wait for Clashiko to cross-review before posting v2. -- pw-bot: cr