From: Saeed Mahameed <saeed@kernel•org>
To: "David S. Miller" <davem@davemloft•net>,
Jakub Kicinski <kuba@kernel•org>
Cc: netdev@vger•kernel.org, Roi Dayan <roid@nvidia•com>,
Vlad Buslov <vladbu@nvidia•com>,
Saeed Mahameed <saeedm@nvidia•com>
Subject: [net 08/18] net/mlx5: Bridge, Fix devlink deadlock on net namespace deletion
Date: Tue, 1 Feb 2022 21:03:54 -0800 [thread overview]
Message-ID: <20220202050404.100122-9-saeed@kernel.org> (raw)
In-Reply-To: <20220202050404.100122-1-saeed@kernel.org>
From: Roi Dayan <roid@nvidia•com>
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>
---
drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c
index d5cb27667005..48dc121b2cb4 100644
--- 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_priv *priv)
}
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 mlx5e_priv *priv)
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);
--
2.34.1
next prev parent reply other threads:[~2022-02-02 5:06 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-02 5:03 [pull request][net 00/18] mlx5 fixes 2022-02-01 Saeed Mahameed
2022-02-02 5:03 ` [net 01/18] net/mlx5: Bridge, take rtnl lock in init error handler Saeed Mahameed
2022-02-02 14:30 ` patchwork-bot+netdevbpf
2022-02-02 5:03 ` [net 02/18] net/mlx5: Bridge, ensure dev_name is null-terminated Saeed Mahameed
2022-02-02 5:03 ` [net 03/18] net/mlx5e: TC, Reject rules with drop and modify hdr action Saeed Mahameed
2022-02-02 5:03 ` [net 04/18] net/mlx5e: Fix module EEPROM query Saeed Mahameed
2022-02-02 5:03 ` [net 05/18] net/mlx5: Use del_timer_sync in fw reset flow of halting poll Saeed Mahameed
2022-02-02 5:03 ` [net 06/18] net/mlx5e: TC, Reject rules with forward and drop actions Saeed Mahameed
2022-02-02 5:03 ` [net 07/18] net/mlx5: Fix offloading with ESWITCH_IPV4_TTL_MODIFY_ENABLE Saeed Mahameed
2022-02-02 5:03 ` Saeed Mahameed [this message]
2022-02-02 5:03 ` [net 09/18] net/mlx5e: Fix wrong calculation of header index in HW_GRO Saeed Mahameed
2022-02-02 5:03 ` [net 10/18] net/mlx5e: Fix broken SKB allocation in HW-GRO Saeed Mahameed
2022-02-02 5:03 ` [net 11/18] net/mlx5e: Fix handling of wrong devices during bond netevent Saeed Mahameed
2022-02-02 5:03 ` [net 12/18] net/mlx5: E-Switch, Fix uninitialized variable modact Saeed Mahameed
2022-02-02 5:03 ` [net 13/18] net/mlx5e: Don't treat small ceil values as unlimited in HTB offload Saeed Mahameed
2022-02-02 5:04 ` [net 14/18] net/mlx5e: IPsec: Fix crypto offload for non TCP/UDP encapsulated traffic Saeed Mahameed
2022-02-02 5:04 ` [net 15/18] net/mlx5e: IPsec: Fix tunnel mode crypto offload for non TCP/UDP traffic Saeed Mahameed
2022-02-02 5:04 ` [net 16/18] net/mlx5e: Avoid implicit modify hdr for decap drop rule Saeed Mahameed
2022-02-02 5:04 ` [net 17/18] net/mlx5e: Use struct_group() for memcpy() region Saeed Mahameed
2022-02-02 5:04 ` [net 18/18] net/mlx5e: Avoid field-overflowing memcpy() Saeed Mahameed
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=20220202050404.100122-9-saeed@kernel.org \
--to=saeed@kernel$(echo .)org \
--cc=davem@davemloft$(echo .)net \
--cc=kuba@kernel$(echo .)org \
--cc=netdev@vger$(echo .)kernel.org \
--cc=roid@nvidia$(echo .)com \
--cc=saeedm@nvidia$(echo .)com \
--cc=vladbu@nvidia$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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