rtnl_link_register() may fail and can impact registering the device.
The fix checks the return value and pushes the error upstream.
Signed-off-by: Aditya Pakki <[email protected]>
---
net/batman-adv/main.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
index 69c0d85bceb3..e0007f242823 100644
--- a/net/batman-adv/main.c
+++ b/net/batman-adv/main.c
@@ -109,7 +109,10 @@ static int __init batadv_init(void)
batadv_debugfs_init();
register_netdevice_notifier(&batadv_hard_if_notifier);
- rtnl_link_register(&batadv_link_ops);
+ ret = rtnl_link_register(&batadv_link_ops);
+ if (ret < 0)
+ return ret;
+
batadv_netlink_register();
pr_info("B.A.T.M.A.N. advanced %s (compatibility version %i) loaded\n",
--
2.17.1
On Monday, 24 December 2018 18.49.26 CET Aditya Pakki wrote:
[...]
> diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c
> index 69c0d85bceb3..e0007f242823 100644
> --- a/net/batman-adv/main.c
> +++ b/net/batman-adv/main.c
> @@ -109,7 +109,10 @@ static int __init batadv_init(void)
> batadv_debugfs_init();
>
> register_netdevice_notifier(&batadv_hard_if_notifier);
> - rtnl_link_register(&batadv_link_ops);
> + ret = rtnl_link_register(&batadv_link_ops);
> + if (ret < 0)
> + return ret;
> +
> batadv_netlink_register();
Uhm, but shouldn't the already initialized parts be unregistered/uninitialized
in this case?
Rejecting this patch because this could cause a new bug.
Also "Fixes: " line is missing
Kind regards,
Sven
On Monday, 24 December 2018 18.49.26 CET Aditya Pakki wrote:
> rtnl_link_register() may fail and can impact registering the device.
> The fix checks the return value and pushes the error upstream.
Regarding the commit message - what is "batman-adv/main"? Shouldn't this be
"batman-adv: "?
And the device can still be registered without batadv_link_ops - just not
using rtnl.
Kind regards,
Sven