From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Roi Dayan <roid@nvidia.com>,
Vlad Buslov <vladbu@nvidia.com>,
Saeed Mahameed <saeedm@nvidia.com>
Subject: [PATCH 5.15 19/32] net/mlx5: Bridge, Fix devlink deadlock on net namespace deletion
Date: Fri, 4 Feb 2022 10:22:29 +0100 [thread overview]
Message-ID: <20220204091915.888414403@linuxfoundation.org> (raw)
In-Reply-To: <20220204091915.247906930@linuxfoundation.org>
From: Roi Dayan <roid@nvidia.com>
commit 880b517691908fb753019b9b27cd082e7617debd upstream.
When changing mode to switchdev, rep bridge init registered to netdevice
notifier holds the devlink lock and then takes pernet_ops_rwsem.
At that time deleting a netns holds pernet_ops_rwsem and then takes
the devlink lock.
Example sequence is:
$ ip netns add foo
$ devlink dev eswitch set pci/0000:00:08.0 mode switchdev &
$ ip netns del foo
deleting netns trace:
[ 1185.365555] ? devlink_pernet_pre_exit+0x74/0x1c0
[ 1185.368331] ? mutex_lock_io_nested+0x13f0/0x13f0
[ 1185.370984] ? xt_find_table+0x40/0x100
[ 1185.373244] ? __mutex_lock+0x24a/0x15a0
[ 1185.375494] ? net_generic+0xa0/0x1c0
[ 1185.376844] ? wait_for_completion_io+0x280/0x280
[ 1185.377767] ? devlink_pernet_pre_exit+0x74/0x1c0
[ 1185.378686] devlink_pernet_pre_exit+0x74/0x1c0
[ 1185.379579] ? devlink_nl_cmd_get_dumpit+0x3a0/0x3a0
[ 1185.380557] ? xt_find_table+0xda/0x100
[ 1185.381367] cleanup_net+0x372/0x8e0
changing mode to switchdev trace:
[ 1185.411267] down_write+0x13a/0x150
[ 1185.412029] ? down_write_killable+0x180/0x180
[ 1185.413005] register_netdevice_notifier+0x1e/0x210
[ 1185.414000] mlx5e_rep_bridge_init+0x181/0x360 [mlx5_core]
[ 1185.415243] mlx5e_uplink_rep_enable+0x269/0x480 [mlx5_core]
[ 1185.416464] ? mlx5e_uplink_rep_disable+0x210/0x210 [mlx5_core]
[ 1185.417749] mlx5e_attach_netdev+0x232/0x400 [mlx5_core]
[ 1185.418906] mlx5e_netdev_attach_profile+0x15b/0x1e0 [mlx5_core]
[ 1185.420172] mlx5e_netdev_change_profile+0x15a/0x1d0 [mlx5_core]
[ 1185.421459] mlx5e_vport_rep_load+0x557/0x780 [mlx5_core]
[ 1185.422624] ? mlx5e_stats_grp_vport_rep_num_stats+0x10/0x10 [mlx5_core]
[ 1185.424006] mlx5_esw_offloads_rep_load+0xdb/0x190 [mlx5_core]
[ 1185.425277] esw_offloads_enable+0xd74/0x14a0 [mlx5_core]
Fix this by registering rep bridges for per net netdev notifier
instead of global one, which operats on the net namespace without holding
the pernet_ops_rwsem.
Fixes: 19e9bfa044f3 ("net/mlx5: Bridge, add offload infrastructure")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
@@ -491,7 +491,7 @@ void mlx5e_rep_bridge_init(struct mlx5e_
}
br_offloads->netdev_nb.notifier_call = mlx5_esw_bridge_switchdev_port_event;
- err = register_netdevice_notifier(&br_offloads->netdev_nb);
+ err = register_netdevice_notifier_net(&init_net, &br_offloads->netdev_nb);
if (err) {
esw_warn(mdev, "Failed to register bridge offloads netdevice notifier (err=%d)\n",
err);
@@ -526,7 +526,7 @@ void mlx5e_rep_bridge_cleanup(struct mlx
return;
cancel_delayed_work_sync(&br_offloads->update_work);
- unregister_netdevice_notifier(&br_offloads->netdev_nb);
+ unregister_netdevice_notifier_net(&init_net, &br_offloads->netdev_nb);
unregister_switchdev_blocking_notifier(&br_offloads->nb_blk);
unregister_switchdev_notifier(&br_offloads->nb);
destroy_workqueue(br_offloads->wq);
next prev parent reply other threads:[~2022-02-04 9:24 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-04 9:22 [PATCH 5.15 00/32] 5.15.20-rc1 review Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 01/32] PCI: pciehp: Fix infinite loop in IRQ handler upon power fault Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 02/32] selftests: mptcp: fix ipv6 routing setup Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 03/32] net: ipa: use a bitmap for endpoint replenish_enabled Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 04/32] net: ipa: prevent concurrent replenish Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 05/32] drm/vc4: hdmi: Make sure the device is powered with CEC Greg Kroah-Hartman
2022-02-05 17:12 ` Guenter Roeck
2022-02-05 17:56 ` Greg Kroah-Hartman
2022-02-05 18:41 ` Guenter Roeck
2022-02-06 12:09 ` Greg Kroah-Hartman
2022-02-06 17:32 ` Guenter Roeck
2022-02-04 9:22 ` [PATCH 5.15 06/32] cgroup-v1: Require capabilities to set release_agent Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 07/32] Revert "mm/gup: small refactoring: simplify try_grab_page()" Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 08/32] ovl: dont fail copy up if no fileattr support on upper Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 09/32] lockd: fix server crash on reboot of client holding lock Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 10/32] lockd: fix failure to cleanup client locks Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 11/32] net/mlx5e: IPsec: Fix tunnel mode crypto offload for non TCP/UDP traffic Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 12/32] net/mlx5: Bridge, take rtnl lock in init error handler Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 13/32] net/mlx5: Bridge, ensure dev_name is null-terminated Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 14/32] net/mlx5e: Fix handling of wrong devices during bond netevent Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 15/32] net/mlx5: Use del_timer_sync in fw reset flow of halting poll Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 16/32] net/mlx5e: Fix module EEPROM query Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 17/32] net/mlx5: Fix offloading with ESWITCH_IPV4_TTL_MODIFY_ENABLE Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 18/32] net/mlx5e: Dont treat small ceil values as unlimited in HTB offload Greg Kroah-Hartman
2022-02-04 9:22 ` Greg Kroah-Hartman [this message]
2022-02-04 9:22 ` [PATCH 5.15 20/32] net/mlx5: E-Switch, Fix uninitialized variable modact Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 21/32] ipheth: fix EOVERFLOW in ipheth_rcvbulk_callback Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 22/32] i40e: Fix reset bw limit when DCB enabled with 1 TC Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 23/32] i40e: Fix reset path while removing the driver Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 24/32] net: amd-xgbe: ensure to reset the tx_timer_active flag Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 25/32] net: amd-xgbe: Fix skb data length underflow Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 26/32] fanotify: Fix stale file descriptor in copy_event_to_user() Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 27/32] net: sched: fix use-after-free in tc_new_tfilter() Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 28/32] rtnetlink: make sure to refresh master_dev/m_ops in __rtnl_newlink() Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 29/32] cpuset: Fix the bug that subpart_cpus updated wrongly in update_cpumask() Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 30/32] e1000e: Handshake with CSME starts from ADL platforms Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 31/32] af_packet: fix data-race in packet_setsockopt / packet_setsockopt Greg Kroah-Hartman
2022-02-04 9:22 ` [PATCH 5.15 32/32] tcp: add missing tcp_skb_can_collapse() test in tcp_shift_skb_data() Greg Kroah-Hartman
2022-02-04 12:21 ` [PATCH 5.15 00/32] 5.15.20-rc1 review Bagas Sanjaya
2022-02-04 17:48 ` Florian Fainelli
2022-02-04 20:31 ` Shuah Khan
2022-02-04 21:08 ` Guenter Roeck
2022-02-04 22:42 ` Ron Economos
2022-02-04 23:04 ` Justin Forbes
2022-02-05 0:18 ` Fox Chen
2022-02-05 5:07 ` Slade Watkins
2022-02-05 6:51 ` Naresh Kamboju
2022-02-05 14:32 ` Sudip Mukherjee
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220204091915.888414403@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=roid@nvidia.com \
--cc=saeedm@nvidia.com \
--cc=stable@vger.kernel.org \
--cc=vladbu@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox