On 2022/11/1 17:34, Chen Zhongjin wrote:
In nf_nat_init(), register_nf_nat_bpf() can fail and return directly without any error handling. Then nf_nat_bysource will leak and registering of &nat_net_ops won't be reverted. This leaves wild ops in subsystem linkedlist and when another module tries to call register_pernet_operations() it triggers page fault: BUG: unable to handle page fault for address: fffffbfff81b964c RIP: 0010:register_pernet_operations+0x1b9/0x5f0 Call Trace: <TASK> register_pernet_subsys+0x29/0x40 ebtables_init+0x58/0x1000 [ebtables] ... Fixes: 820dc0523e05 ("net: netfilter: move bpf_ct_set_nat_info kfunc in nf_nat_bpf.c") Signed-off-by: Chen Zhongjin <chenzhongjin@xxxxxxxxxx> --- net/netfilter/nf_nat_core.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 18319a6e6806..b24b4dfc1ca4 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -1152,7 +1152,12 @@ static int __init nf_nat_init(void) WARN_ON(nf_nat_hook != NULL); RCU_INIT_POINTER(nf_nat_hook, &nat_hook);- return register_nf_nat_bpf();+ ret = register_nf_nat_bpf(); + if (ret < 0) { + kvfree(nf_nat_bysource); + unregister_pernet_subsys(&nat_net_ops); + } + return ret; }
I noticed that follow_master_nat should also be unregistered. Going to send v2, discard this one. Best, Chen
static void __exit nf_nat_cleanup(void)
