| Message ID | 20210504144000.18155-1-fw@strlen.de |
|---|---|
| State | Accepted |
| Delegated to: | Pablo Neira |
| Headers | show |
| Series | [nf] netfilter: conntrack: unregister ipv4 sockopts on error unwind | expand |
diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 89e5bac384d7..dc9ca12b0489 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -664,7 +664,7 @@ int nf_conntrack_proto_init(void) #if IS_ENABLED(CONFIG_IPV6) cleanup_sockopt: - nf_unregister_sockopt(&so_getorigdst6); + nf_unregister_sockopt(&so_getorigdst); #endif return ret; }
When ipv6 sockopt register fails, the ipv4 one needs to be removed. Fixes: a0ae2562c6c ("netfilter: conntrack: remove l3proto abstraction") Signed-off-by: Florian Westphal <fw@strlen.de> --- The should be no confict/overlap with an existing sockopt range, so I don't think the registration can fail in the first place. However, its broken as-is. An alternate solution (in -next), might be to change return type to void and have nf_register_sockopt WARN() instead. net/netfilter/nf_conntrack_proto.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)