linux.git/net/ipv4/netfilter, branch v5.15

netfilter: conntrack: fix boot failure with nf_conntrack.enable_hooks=1

2021-09-28T11:04:55+00:00

This is a revert of
7b1957b049 ("netfilter: nf_defrag_ipv4: use net_generic infra")
and a partial revert of
8b0adbe3e3 ("netfilter: nf_defrag_ipv6: use net_generic infra").

If conntrack is builtin and kernel is booted with:
nf_conntrack.enable_hooks=1

.... kernel will fail to boot due to a NULL deref in
nf_defrag_ipv4_enable(): Its called before the ipv4 defrag initcall is
made, so net_generic() returns NULL.

To resolve this, move the user refcount back to struct net so calls
to those functions are possible even before their initcalls have run.

Fixes: 7b1957b04956 ("netfilter: nf_defrag_ipv4: use net_generic infra")
Fixes: 8b0adbe3e38d ("netfilter: nf_defrag_ipv6: use net_generic infra").
Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: iptable_raw: drop bogus net_init annotation

2021-09-21T01:46:56+00:00

This is a leftover from the times when this function was wired up via
pernet_operations.  Now its called when userspace asks for the table.

With CONFIG_NET_NS=n, iptable_raw_table_init memory has been discarded
already and we get a kernel crash.

Other tables are fine, __net_init annotation was removed already.

Fixes: fdacd57c79b7 ("netfilter: x_tables: never register tables by default")
Reported-by: youling 257 
Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: x_tables: handle xt_register_template() returning an error value

2021-08-25T11:06:48+00:00

Commit fdacd57c79b7 ("netfilter: x_tables: never register tables by
default") introduces the function xt_register_template(), and in one case,
a call to that function was missing the error-case handling.

Handle when xt_register_template() returns an error value.

This was identified with the clang-analyzer's Dead-Store analysis.

Signed-off-by: Lukas Bulwahn 
Reviewed-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: x_tables: never register tables by default

2021-08-09T08:22:01+00:00

For historical reasons x_tables still register tables by default in the
initial namespace.
Only newly created net namespaces add the hook on demand.

This means that the init_net always pays hook cost, even if no filtering
rules are added (e.g. only used inside a single netns).

Note that the hooks are added even when 'iptables -L' is called.
This is because there is no way to tell 'iptables -A' and 'iptables -L'
apart at kernel level.

The only solution would be to register the table, but delay hook
registration until the first rule gets added (or policy gets changed).

That however means that counters are not hooked either, so 'iptables -L'
would always show 0-counters even when traffic is flowing which might be
unexpected.

This keeps table and hook registration consistent with what is already done
in non-init netns: first iptables(-save) invocation registers both table
and hooks.

This applies the same solution adopted for ebtables.
All tables register a template that contains the l3 family, the name
and a constructor function that is called when the initial table has to
be added.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: ipt_CLUSTERIP: use clusterip_net to store pernet warning

2021-08-01T10:00:50+00:00

No need to use struct net for this.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: ipt_CLUSTERIP: only add arp mangle hook when required

2021-08-01T10:00:50+00:00

Do not register the arp mangling hooks from pernet init path.

As-is, load of the module is enough for these hooks to become active
in each net namespace.

Use checkentry instead so hook is only added if a CLUSTERIP rule is used.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: nf_tables: add and use nft_sk helper

2021-05-28T23:04:53+00:00

This allows to change storage placement later on without changing readers.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: arptables: use pernet ops struct during unregister

2021-05-03T21:04:01+00:00

Like with iptables and ebtables, hook unregistration has to use the
pernet ops struct, not the template.

This triggered following splat:
  hook not found, pf 3 num 0
  WARNING: CPU: 0 PID: 224 at net/netfilter/core.c:480 __nf_unregister_net_hook+0x1eb/0x610 net/netfilter/core.c:480
[..]
 nf_unregister_net_hook net/netfilter/core.c:502 [inline]
 nf_unregister_net_hooks+0x117/0x160 net/netfilter/core.c:576
 arpt_unregister_table_pre_exit+0x67/0x80 net/ipv4/netfilter/arp_tables.c:1565

Fixes: f9006acc8dfe5 ("netfilter: arp_tables: pass table pointer via nf_hook_ops")
Reported-by: syzbot+dcccba8a1e41a38cb9df@syzkaller.appspotmail.com
Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: allow to turn off xtables compat layer

2021-04-26T16:16:56+00:00

The compat layer needs to parse untrusted input (the ruleset)
to translate it to a 64bit compatible format.

We had a number of bugs in this department in the past, so allow users
to turn this feature off.

Add CONFIG_NETFILTER_XTABLES_COMPAT kconfig knob and make it default to y
to keep existing behaviour.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso

netfilter: arp_tables: pass table pointer via nf_hook_ops

2021-04-26T01:20:46+00:00

Same change as previous patch.  Only difference:
no need to handle NULL template_ops parameter, the only caller
(arptable_filter) always passes non-NULL argument.

This removes all remaining accesses to net->ipv4.arptable_filter.

Signed-off-by: Florian Westphal 
Signed-off-by: Pablo Neira Ayuso