From: Gilad Sever <gilad9366@xxxxxxxxx>
[ 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@xxxxxxxxx>
Signed-off-by: Daniel Borkmann <daniel@xxxxxxxxxxxxx>
Reviewed-by: Shmulik Ladkani <shmulik.ladkani@xxxxxxxxx>
Reviewed-by: Eyal Birger <eyal.birger@xxxxxxxxx>
Acked-by: Stanislav Fomichev <sdf@xxxxxxxxxx>
Cc: David Ahern <dsahern@xxxxxxxxxx>
Link: https://lore.kernel.org/bpf/20230621104211.301902-4-gilad9366@xxxxxxxxx
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
include/linux/netdevice.h | 9 +++++
net/core/filter.c | 69 ++++++++++++++++++++++-----------------
2 files changed, 48 insertions(+), 30 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c0a4589ab706f..b5df2e59a51d3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -5196,6 +5196,15 @@ static inline bool netif_is_l3_slave(const struct net_device *dev)
return dev->priv_flags & IFF_L3MDEV_SLAVE;
}
+static inline int dev_sdif(const struct net_device *dev)
+{
+#ifdef CONFIG_NET_L3_MASTER_DEV
+ if (netif_is_l3_slave(dev))
+ return dev->ifindex;
+#endif
+ return 0;
+}
+
static inline bool netif_is_bridge_master(const struct net_device *dev)
{
return dev->priv_flags & IFF_EBRIDGE;
diff --git a/net/core/filter.c b/net/core/filter.c
index 60c1ad379a0b6..18eb8049c795c 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6174,12 +6174,11 @@ static struct sock *sk_lookup(struct net *net, struct bpf_sock_tuple *tuple,
static struct sock *
__bpf_skc_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
struct net *caller_net, u32 ifindex, u8 proto, u64 netns_id,
- u64 flags)
+ u64 flags, int sdif)
{
struct sock *sk = NULL;
struct net *net;
u8 family;
- int sdif;
if (len == sizeof(tuple->ipv4))
family = AF_INET;
@@ -6191,10 +6190,12 @@ __bpf_skc_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
if (unlikely(flags || !((s32)netns_id < 0 || netns_id <= S32_MAX)))
goto out;
- if (family == AF_INET)
- sdif = inet_sdif(skb);
- else
- sdif = inet6_sdif(skb);
+ if (sdif < 0) {
+ if (family == AF_INET)
+ sdif = inet_sdif(skb);
+ else
+ sdif = inet6_sdif(skb);
+ }
if ((s32)netns_id < 0) {
net = caller_net;
@@ -6214,10 +6215,11 @@ __bpf_skc_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
static struct sock *
__bpf_sk_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
struct net *caller_net, u32 ifindex, u8 proto, u64 netns_id,
- u64 flags)
+ u64 flags, int sdif)
{
struct sock *sk = __bpf_skc_lookup(skb, tuple, len, caller_net,
- ifindex, proto, netns_id, flags);
+ ifindex, proto, netns_id, flags,
+ sdif);
if (sk) {
struct sock *sk2 = sk_to_full_sk(sk);
@@ -6257,7 +6259,7 @@ bpf_skc_lookup(struct sk_buff *skb, struct bpf_sock_tuple *tuple, u32 len,
}
return __bpf_skc_lookup(skb, tuple, len, caller_net, ifindex, proto,
- netns_id, flags);
+ netns_id, flags, -1);
}
static struct sock *
@@ -6349,12 +6351,13 @@ static const struct bpf_func_proto bpf_sk_lookup_udp_proto = {
BPF_CALL_5(bpf_tc_skc_lookup_tcp, struct sk_buff *, skb,
struct bpf_sock_tuple *, tuple, u32, len, u64, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(skb->dev);
- int ifindex = skb->dev->ifindex;
+ struct net_device *dev = skb->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_skc_lookup(skb, tuple, len, caller_net,
ifindex, IPPROTO_TCP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_tc_skc_lookup_tcp_proto = {
@@ -6372,12 +6375,13 @@ static const struct bpf_func_proto bpf_tc_skc_lookup_tcp_proto = {
BPF_CALL_5(bpf_tc_sk_lookup_tcp, struct sk_buff *, skb,
struct bpf_sock_tuple *, tuple, u32, len, u64, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(skb->dev);
- int ifindex = skb->dev->ifindex;
+ struct net_device *dev = skb->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_sk_lookup(skb, tuple, len, caller_net,
ifindex, IPPROTO_TCP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_tc_sk_lookup_tcp_proto = {
@@ -6395,12 +6399,13 @@ static const struct bpf_func_proto bpf_tc_sk_lookup_tcp_proto = {
BPF_CALL_5(bpf_tc_sk_lookup_udp, struct sk_buff *, skb,
struct bpf_sock_tuple *, tuple, u32, len, u64, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(skb->dev);
- int ifindex = skb->dev->ifindex;
+ struct net_device *dev = skb->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_sk_lookup(skb, tuple, len, caller_net,
ifindex, IPPROTO_UDP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_tc_sk_lookup_udp_proto = {
@@ -6432,12 +6437,13 @@ static const struct bpf_func_proto bpf_sk_release_proto = {
BPF_CALL_5(bpf_xdp_sk_lookup_udp, struct xdp_buff *, ctx,
struct bpf_sock_tuple *, tuple, u32, len, u32, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(ctx->rxq->dev);
- int ifindex = ctx->rxq->dev->ifindex;
+ struct net_device *dev = ctx->rxq->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_sk_lookup(NULL, tuple, len, caller_net,
ifindex, IPPROTO_UDP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_xdp_sk_lookup_udp_proto = {
@@ -6455,12 +6461,13 @@ static const struct bpf_func_proto bpf_xdp_sk_lookup_udp_proto = {
BPF_CALL_5(bpf_xdp_skc_lookup_tcp, struct xdp_buff *, ctx,
struct bpf_sock_tuple *, tuple, u32, len, u32, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(ctx->rxq->dev);
- int ifindex = ctx->rxq->dev->ifindex;
+ struct net_device *dev = ctx->rxq->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_skc_lookup(NULL, tuple, len, caller_net,
ifindex, IPPROTO_TCP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_xdp_skc_lookup_tcp_proto = {
@@ -6478,12 +6485,13 @@ static const struct bpf_func_proto bpf_xdp_skc_lookup_tcp_proto = {
BPF_CALL_5(bpf_xdp_sk_lookup_tcp, struct xdp_buff *, ctx,
struct bpf_sock_tuple *, tuple, u32, len, u32, netns_id, u64, flags)
{
- struct net *caller_net = dev_net(ctx->rxq->dev);
- int ifindex = ctx->rxq->dev->ifindex;
+ struct net_device *dev = ctx->rxq->dev;
+ int ifindex = dev->ifindex, sdif = dev_sdif(dev);
+ struct net *caller_net = dev_net(dev);
return (unsigned long)__bpf_sk_lookup(NULL, tuple, len, caller_net,
ifindex, IPPROTO_TCP, netns_id,
- flags);
+ flags, sdif);
}
static const struct bpf_func_proto bpf_xdp_sk_lookup_tcp_proto = {
@@ -6503,7 +6511,8 @@ BPF_CALL_5(bpf_sock_addr_skc_lookup_tcp, struct bpf_sock_addr_kern *, ctx,
{
return (unsigned long)__bpf_skc_lookup(NULL, tuple, len,
sock_net(ctx->sk), 0,
- IPPROTO_TCP, netns_id, flags);
+ IPPROTO_TCP, netns_id, flags,
+ -1);
}
static const struct bpf_func_proto bpf_sock_addr_skc_lookup_tcp_proto = {
@@ -6522,7 +6531,7 @@ BPF_CALL_5(bpf_sock_addr_sk_lookup_tcp, struct bpf_sock_addr_kern *, ctx,
{
return (unsigned long)__bpf_sk_lookup(NULL, tuple, len,
sock_net(ctx->sk), 0, IPPROTO_TCP,
- netns_id, flags);
+ netns_id, flags, -1);
}
static const struct bpf_func_proto bpf_sock_addr_sk_lookup_tcp_proto = {
@@ -6541,7 +6550,7 @@ BPF_CALL_5(bpf_sock_addr_sk_lookup_udp, struct bpf_sock_addr_kern *, ctx,
{
return (unsigned long)__bpf_sk_lookup(NULL, tuple, len,
sock_net(ctx->sk), 0, IPPROTO_UDP,
- netns_id, flags);
+ netns_id, flags, -1);
}
static const struct bpf_func_proto bpf_sock_addr_sk_lookup_udp_proto = {
--
2.39.2
