From 4ec9031cbeb73a66979560bbb6d355329be762de Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 8 Apr 2025 12:59:50 -0700 Subject: netdev: add "ops compat locking" helpers Add helpers to "lock a netdev in a backward-compatible way", which for ops-locked netdevs will mean take the instance lock. For drivers which haven't opted into the ops locking we'll take rtnl_lock. The scoped foreach is dropping and re-taking the lock for each device, even if prev and next are both under rtnl_lock. I hope that's fine since we expect that netdev nl to be mostly supported by modern drivers, and modern drivers should also opt into the instance locking. Note that these helpers are mostly needed for queue related state, because drivers modify queue config in their ops in a non-atomic way. Or differently put, queue changes don't have a clear-cut API like NAPI configuration. Any state that can should just use the instance lock directly, not the "compat" hacks. Reviewed-by: Joe Damato Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250408195956.412733-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- include/net/netdev_lock.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'include/net/netdev_lock.h') diff --git a/include/net/netdev_lock.h b/include/net/netdev_lock.h index c316b551df8d..5706835a660c 100644 --- a/include/net/netdev_lock.h +++ b/include/net/netdev_lock.h @@ -64,6 +64,22 @@ netdev_ops_assert_locked_or_invisible(const struct net_device *dev) netdev_ops_assert_locked(dev); } +static inline void netdev_lock_ops_compat(struct net_device *dev) +{ + if (netdev_need_ops_lock(dev)) + netdev_lock(dev); + else + rtnl_lock(); +} + +static inline void netdev_unlock_ops_compat(struct net_device *dev) +{ + if (netdev_need_ops_lock(dev)) + netdev_unlock(dev); + else + rtnl_unlock(); +} + static inline int netdev_lock_cmp_fn(const struct lockdep_map *a, const struct lockdep_map *b) { -- cgit v1.2.3 From 22cbc1ee268b7ec0000848708944daa61c6e4909 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 15 Apr 2025 20:04:47 -0700 Subject: netdev: fix the locking for netdev notifications Kuniyuki reports that the assert for netdev lock fires when there are netdev event listeners (otherwise we skip the netlink event generation). Correct the locking when coming from the notifier. The NETDEV_XDP_FEAT_CHANGE notifier is already fully locked, it's the documentation that's incorrect. Fixes: 99e44f39a8f7 ("netdev: depend on netdev->lock for xdp features") Reported-by: syzkaller Reported-by: Kuniyuki Iwashima Link: https://lore.kernel.org/20250410171019.62128-1-kuniyu@amazon.com Acked-by: Stanislav Fomichev Link: https://patch.msgid.link/20250416030447.1077551-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/networking/netdevices.rst | 4 +++- include/linux/netdevice.h | 2 +- include/net/netdev_lock.h | 16 ++++++++++++++++ net/core/lock_debug.c | 4 +++- net/core/netdev-genl.c | 4 ++++ 5 files changed, 27 insertions(+), 3 deletions(-) (limited to 'include/net/netdev_lock.h') diff --git a/Documentation/networking/netdevices.rst b/Documentation/networking/netdevices.rst index 77fe1f7b93be..7ebb6c36482d 100644 --- a/Documentation/networking/netdevices.rst +++ b/Documentation/networking/netdevices.rst @@ -387,12 +387,14 @@ For device drivers that implement shaping or queue management APIs, some of the notifiers (``enum netdev_cmd``) are running under the netdev instance lock. +The following netdev notifiers are always run under the instance lock: +* ``NETDEV_XDP_FEAT_CHANGE`` + For devices with locked ops, currently only the following notifiers are running under the lock: * ``NETDEV_CHANGE`` * ``NETDEV_REGISTER`` * ``NETDEV_UP`` -* ``NETDEV_XDP_FEAT_CHANGE`` The following notifiers are running without the lock: * ``NETDEV_UNREGISTER`` diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e6036b82ef4c..0321fd952f70 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2520,7 +2520,7 @@ struct net_device { * @net_shaper_hierarchy, @reg_state, @threaded * * Double protects: - * @up, @moving_ns, @nd_net, @xdp_flags + * @up, @moving_ns, @nd_net, @xdp_features * * Double ops protects: * @real_num_rx_queues, @real_num_tx_queues diff --git a/include/net/netdev_lock.h b/include/net/netdev_lock.h index 5706835a660c..2a753813f849 100644 --- a/include/net/netdev_lock.h +++ b/include/net/netdev_lock.h @@ -48,6 +48,22 @@ static inline void netdev_unlock_ops(struct net_device *dev) netdev_unlock(dev); } +static inline void netdev_lock_ops_to_full(struct net_device *dev) +{ + if (netdev_need_ops_lock(dev)) + netdev_assert_locked(dev); + else + netdev_lock(dev); +} + +static inline void netdev_unlock_full_to_ops(struct net_device *dev) +{ + if (netdev_need_ops_lock(dev)) + netdev_assert_locked(dev); + else + netdev_unlock(dev); +} + static inline void netdev_ops_assert_locked(const struct net_device *dev) { if (netdev_need_ops_lock(dev)) diff --git a/net/core/lock_debug.c b/net/core/lock_debug.c index 6fade574bc2a..9e9fb25314b9 100644 --- a/net/core/lock_debug.c +++ b/net/core/lock_debug.c @@ -18,10 +18,12 @@ int netdev_debug_event(struct notifier_block *nb, unsigned long event, /* Keep enum and don't add default to trigger -Werror=switch */ switch (cmd) { + case NETDEV_XDP_FEAT_CHANGE: + netdev_assert_locked(dev); + fallthrough; case NETDEV_CHANGE: case NETDEV_REGISTER: case NETDEV_UP: - case NETDEV_XDP_FEAT_CHANGE: netdev_ops_assert_locked(dev); fallthrough; case NETDEV_DOWN: diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index b64c614a00c4..2c104947d224 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -963,10 +963,14 @@ static int netdev_genl_netdevice_event(struct notifier_block *nb, switch (event) { case NETDEV_REGISTER: + netdev_lock_ops_to_full(netdev); netdev_genl_dev_notify(netdev, NETDEV_CMD_DEV_ADD_NTF); + netdev_unlock_full_to_ops(netdev); break; case NETDEV_UNREGISTER: + netdev_lock(netdev); netdev_genl_dev_notify(netdev, NETDEV_CMD_DEV_DEL_NTF); + netdev_unlock(netdev); break; case NETDEV_XDP_FEAT_CHANGE: netdev_genl_dev_notify(netdev, NETDEV_CMD_DEV_CHANGE_NTF); -- cgit v1.2.3 From 4c2bd7913f52b1e5c978edf56cdef39c30a1f603 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Sat, 17 May 2025 13:08:10 -0700 Subject: net: let lockdep compare instance locks AFAIU always returning -1 from lockdep's compare function basically disables checking of dependencies between given locks. Try to be a little more precise about what guarantees that instance locks won't deadlock. Right now we only nest them under protection of rtnl_lock. Mostly in unregister_netdevice_many() and dev_close_many(). Acked-by: Stanislav Fomichev Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250517200810.466531-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- include/net/netdev_lock.h | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'include/net/netdev_lock.h') diff --git a/include/net/netdev_lock.h b/include/net/netdev_lock.h index 2a753813f849..c345afecd4c5 100644 --- a/include/net/netdev_lock.h +++ b/include/net/netdev_lock.h @@ -99,16 +99,15 @@ static inline void netdev_unlock_ops_compat(struct net_device *dev) static inline int netdev_lock_cmp_fn(const struct lockdep_map *a, const struct lockdep_map *b) { - /* Only lower devices currently grab the instance lock, so no - * real ordering issues can occur. In the near future, only - * hardware devices will grab instance lock which also does not - * involve any ordering. Suppress lockdep ordering warnings - * until (if) we start grabbing instance lock on pure SW - * devices (bond/team/veth/etc). - */ if (a == b) return 0; - return -1; + + /* Allow locking multiple devices only under rtnl_lock, + * the exact order doesn't matter. + * Note that upper devices don't lock their ops, so nesting + * mostly happens in batched device removal for now. + */ + return lockdep_rtnl_is_held() ? -1 : 1; } #define netdev_lockdep_set_classes(dev) \ -- cgit v1.2.3