netfilter: nf_tables: don't prevent event handler from device cleanup on netns exit
authorFlorian Westphal <fw@strlen.de>
Thu, 2 Aug 2018 19:44:41 +0000 (21:44 +0200)
committerPablo Neira Ayuso <pablo@netfilter.org>
Thu, 16 Aug 2018 17:37:03 +0000 (19:37 +0200)
commit6a48de0144767f2c6880540c0a4ac6741e3c440b
treef3111dac7df6aa291d8d8b60dcb50f5a508c9d70
parentd209df3e7f7002d9099fdb0f6df0f972b4386a63
netfilter: nf_tables: don't prevent event handler from device cleanup on netns exit

When a netnsamespace exits, the nf_tables pernet_ops will remove all rules.
However, there is one caveat:

Base chains that register ingress hooks will cause use-after-free:
device is already gone at that point.

The device event handlers prevent this from happening:
netns exit synthesizes unregister events for all devices.

However, an improper fix for a race condition made the notifiers a no-op
in case they get called from netns exit path, so revert that part.

This is safe now as the previous patch fixed nf_tables pernet ops
and device notifier initialisation ordering.

Fixes: 0a2cf5ee432c2 ("netfilter: nf_tables: close race between netns exit and rmmod")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
net/netfilter/nf_tables_api.c
net/netfilter/nft_chain_filter.c