bpf: Fix bpf socket lookup from tc/xdp to respect socket VRF bindings
authorGilad Sever <gilad9366@gmail.com>
Wed, 21 Jun 2023 10:42:10 +0000 (13:42 +0300)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 19 Jul 2023 14:21:11 +0000 (16:21 +0200)
commitcfb310623413d5b13cfb723ed7e22b22fd5bf6a7
tree7efc607d58e55653b4f7d741bbcb8367fc8fbbe7
parentc7415c521abdc1d174d8b03dca62ae847ba1c49c
bpf: Fix bpf socket lookup from tc/xdp to respect socket VRF bindings

[ Upstream commit 9a5cb79762e0eda17ca15c2a6eaca4622383c21c ]

When calling bpf_sk_lookup_tcp(), bpf_sk_lookup_udp() or
bpf_skc_lookup_tcp() from tc/xdp ingress, VRF socket bindings aren't
respoected, i.e. unbound sockets are returned, and bound sockets aren't
found.

VRF binding is determined by the sdif argument to sk_lookup(), however
when called from tc the IP SKB control block isn't initialized and thus
inet{,6}_sdif() always returns 0.

Fix by calculating sdif for the tc/xdp flows by observing the device's
l3 enslaved state.

The cg/sk_skb hooking points which are expected to support
inet{,6}_sdif() pass sdif=-1 which makes __bpf_skc_lookup() use the
existing logic.

Fixes: 6acc9b432e67 ("bpf: Add helper to retrieve socket in BPF")
Signed-off-by: Gilad Sever <gilad9366@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Shmulik Ladkani <shmulik.ladkani@gmail.com>
Reviewed-by: Eyal Birger <eyal.birger@gmail.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Cc: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/bpf/20230621104211.301902-4-gilad9366@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
include/linux/netdevice.h
net/core/filter.c