[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <ad7f5f275bcda1ee058d7bd3020b7d85cd44b9f6.1636390483.git.leonro@nvidia.com>
Date: Mon, 8 Nov 2021 19:05:36 +0200
From: Leon Romanovsky <leon@...nel.org>
To: "David S . Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>
Cc: Leon Romanovsky <leonro@...dia.com>,
Ido Schimmel <idosch@...dia.com>, Jiri Pirko <jiri@...dia.com>,
netdev <netdev@...r.kernel.org>
Subject: [RFC PATCH 14/16] devlink: Require devlink lock during device reload
From: Leon Romanovsky <leonro@...dia.com>
Devlink reload was implemented as a special command which does _SET_
operation, but doesn't take devlink->lock, while recursive devlink
calls that were part of .reload_up()/.reload_down() sequence took it.
This fragile flow was possible due to holding a big devlink lock
(devlink_mutex), which effectively stopped all devlink activities,
even unrelated to reloaded devlink.
So let's make sure that devlink reload behaves as other commands.
Signed-off-by: Leon Romanovsky <leonro@...dia.com>
---
net/core/devlink.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/core/devlink.c b/net/core/devlink.c
index 60af6a3fc130..32d274bfd049 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -8771,7 +8771,6 @@ static const struct genl_small_ops devlink_nl_ops[] = {
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = devlink_nl_cmd_reload,
.flags = GENL_ADMIN_PERM,
- .internal_flags = DEVLINK_NL_FLAG_NO_LOCK,
},
{
.cmd = DEVLINK_CMD_PARAM_GET,
@@ -11462,6 +11461,7 @@ static void __net_exit devlink_pernet_pre_exit(struct net *net)
if (!net_eq(devlink_net(devlink), net))
goto retry;
+ mutex_lock(&devlink->lock);
WARN_ON(!(devlink->features & DEVLINK_F_RELOAD));
err = devlink_reload(devlink, &init_net,
DEVLINK_RELOAD_ACTION_DRIVER_REINIT,
@@ -11469,6 +11469,7 @@ static void __net_exit devlink_pernet_pre_exit(struct net *net)
&actions_performed, NULL);
if (err && err != -EOPNOTSUPP)
pr_warn("Failed to reload devlink instance into init_net\n");
+ mutex_unlock(&devlink->lock);
retry:
devlink_put(devlink);
}
--
2.33.1
Powered by blists - more mailing lists