summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--drivers/net/ethernet/intel/igb/igb_ptp.c5
-rw-r--r--drivers/net/gtp.c12
-rw-r--r--include/linux/netfilter.h1
-rw-r--r--net/bridge/br_netfilter_hooks.c96
-rw-r--r--net/bridge/netfilter/nf_conntrack_bridge.c30
-rw-r--r--net/core/rtnetlink.c11
-rw-r--r--net/hsr/hsr_forward.c2
-rw-r--r--net/netfilter/nf_conntrack_core.c1
-rw-r--r--net/netfilter/nft_compat.c20
-rw-r--r--net/tls/tls_sw.c40
-rw-r--r--tools/net/ynl/lib/ynl.c1
-rw-r--r--tools/testing/selftests/netfilter/Makefile3
-rw-r--r--tools/testing/selftests/netfilter/bridge_netfilter.sh188
13 files changed, 383 insertions, 27 deletions
diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index 319c544b9f04..f94570556120 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -957,7 +957,7 @@ static void igb_ptp_tx_hwtstamp(struct igb_adapter *adapter)
igb_ptp_systim_to_hwtstamp(adapter, &shhwtstamps, regval);
/* adjust timestamp for the TX latency based on link speed */
- if (adapter->hw.mac.type == e1000_i210) {
+ if (hw->mac.type == e1000_i210 || hw->mac.type == e1000_i211) {
switch (adapter->link_speed) {
case SPEED_10:
adjust = IGB_I210_TX_LATENCY_10;
@@ -1003,6 +1003,7 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va,
ktime_t *timestamp)
{
struct igb_adapter *adapter = q_vector->adapter;
+ struct e1000_hw *hw = &adapter->hw;
struct skb_shared_hwtstamps ts;
__le64 *regval = (__le64 *)va;
int adjust = 0;
@@ -1022,7 +1023,7 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va,
igb_ptp_systim_to_hwtstamp(adapter, &ts, le64_to_cpu(regval[1]));
/* adjust timestamp for the RX latency based on link speed */
- if (adapter->hw.mac.type == e1000_i210) {
+ if (hw->mac.type == e1000_i210 || hw->mac.type == e1000_i211) {
switch (adapter->link_speed) {
case SPEED_10:
adjust = IGB_I210_RX_LATENCY_10;
diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c
index 2129ae42c703..2b5357d94ff5 100644
--- a/drivers/net/gtp.c
+++ b/drivers/net/gtp.c
@@ -1903,26 +1903,26 @@ static int __init gtp_init(void)
get_random_bytes(&gtp_h_initval, sizeof(gtp_h_initval));
- err = rtnl_link_register(&gtp_link_ops);
+ err = register_pernet_subsys(&gtp_net_ops);
if (err < 0)
goto error_out;
- err = register_pernet_subsys(&gtp_net_ops);
+ err = rtnl_link_register(&gtp_link_ops);
if (err < 0)
- goto unreg_rtnl_link;
+ goto unreg_pernet_subsys;
err = genl_register_family(&gtp_genl_family);
if (err < 0)
- goto unreg_pernet_subsys;
+ goto unreg_rtnl_link;
pr_info("GTP module loaded (pdp ctx size %zd bytes)\n",
sizeof(struct pdp_ctx));
return 0;
-unreg_pernet_subsys:
- unregister_pernet_subsys(&gtp_net_ops);
unreg_rtnl_link:
rtnl_link_unregister(&gtp_link_ops);
+unreg_pernet_subsys:
+ unregister_pernet_subsys(&gtp_net_ops);
error_out:
pr_err("error loading GTP module loaded\n");
return err;
diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 80900d910992..ce660d51549b 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -474,6 +474,7 @@ struct nf_ct_hook {
const struct sk_buff *);
void (*attach)(struct sk_buff *nskb, const struct sk_buff *skb);
void (*set_closing)(struct nf_conntrack *nfct);
+ int (*confirm)(struct sk_buff *skb);
};
extern const struct nf_ct_hook __rcu *nf_ct_hook;
diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index ed1720890757..35e10c5a766d 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -43,6 +43,10 @@
#include <linux/sysctl.h>
#endif
+#if IS_ENABLED(CONFIG_NF_CONNTRACK)
+#include <net/netfilter/nf_conntrack_core.h>
+#endif
+
static unsigned int brnf_net_id __read_mostly;
struct brnf_net {
@@ -553,6 +557,90 @@ static unsigned int br_nf_pre_routing(void *priv,
return NF_STOLEN;
}
+#if IS_ENABLED(CONFIG_NF_CONNTRACK)
+/* conntracks' nf_confirm logic cannot handle cloned skbs referencing
+ * the same nf_conn entry, which will happen for multicast (broadcast)
+ * Frames on bridges.
+ *
+ * Example:
+ * macvlan0
+ * br0
+ * ethX ethY
+ *
+ * ethX (or Y) receives multicast or broadcast packet containing
+ * an IP packet, not yet in conntrack table.
+ *
+ * 1. skb passes through bridge and fake-ip (br_netfilter)Prerouting.
+ * -> skb->_nfct now references a unconfirmed entry
+ * 2. skb is broad/mcast packet. bridge now passes clones out on each bridge
+ * interface.
+ * 3. skb gets passed up the stack.
+ * 4. In macvlan case, macvlan driver retains clone(s) of the mcast skb
+ * and schedules a work queue to send them out on the lower devices.
+ *
+ * The clone skb->_nfct is not a copy, it is the same entry as the
+ * original skb. The macvlan rx handler then returns RX_HANDLER_PASS.
+ * 5. Normal conntrack hooks (in NF_INET_LOCAL_IN) confirm the orig skb.
+ *
+ * The Macvlan broadcast worker and normal confirm path will race.
+ *
+ * This race will not happen if step 2 already confirmed a clone. In that
+ * case later steps perform skb_clone() with skb->_nfct already confirmed (in
+ * hash table). This works fine.
+ *
+ * But such confirmation won't happen when eb/ip/nftables rules dropped the
+ * packets before they reached the nf_confirm step in postrouting.
+ *
+ * Work around this problem by explicit confirmation of the entry at
+ * LOCAL_IN time, before upper layer has a chance to clone the unconfirmed
+ * entry.
+ *
+ */
+static unsigned int br_nf_local_in(void *priv,
+ struct sk_buff *skb,
+ const struct nf_hook_state *state)
+{
+ struct nf_conntrack *nfct = skb_nfct(skb);
+ const struct nf_ct_hook *ct_hook;
+ struct nf_conn *ct;
+ int ret;
+
+ if (!nfct || skb->pkt_type == PACKET_HOST)
+ return NF_ACCEPT;
+
+ ct = container_of(nfct, struct nf_conn, ct_general);
+ if (likely(nf_ct_is_confirmed(ct)))
+ return NF_ACCEPT;
+
+ WARN_ON_ONCE(skb_shared(skb));
+ WARN_ON_ONCE(refcount_read(&nfct->use) != 1);
+
+ /* We can't call nf_confirm here, it would create a dependency
+ * on nf_conntrack module.
+ */
+ ct_hook = rcu_dereference(nf_ct_hook);
+ if (!ct_hook) {
+ skb->_nfct = 0ul;
+ nf_conntrack_put(nfct);
+ return NF_ACCEPT;
+ }
+
+ nf_bridge_pull_encap_header(skb);
+ ret = ct_hook->confirm(skb);
+ switch (ret & NF_VERDICT_MASK) {
+ case NF_STOLEN:
+ return NF_STOLEN;
+ default:
+ nf_bridge_push_encap_header(skb);
+ break;
+ }
+
+ ct = container_of(nfct, struct nf_conn, ct_general);
+ WARN_ON_ONCE(!nf_ct_is_confirmed(ct));
+
+ return ret;
+}
+#endif
/* PF_BRIDGE/FORWARD *************************************************/
static int br_nf_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb)
@@ -964,6 +1052,14 @@ static const struct nf_hook_ops br_nf_ops[] = {
.hooknum = NF_BR_PRE_ROUTING,
.priority = NF_BR_PRI_BRNF,
},
+#if IS_ENABLED(CONFIG_NF_CONNTRACK)
+ {
+ .hook = br_nf_local_in,
+ .pf = NFPROTO_BRIDGE,
+ .hooknum = NF_BR_LOCAL_IN,
+ .priority = NF_BR_PRI_LAST,
+ },
+#endif
{
.hook = br_nf_forward,
.pf = NFPROTO_BRIDGE,
diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c
index abb090f94ed2..6f877e31709b 100644
--- a/net/bridge/netfilter/nf_conntrack_bridge.c
+++ b/net/bridge/netfilter/nf_conntrack_bridge.c
@@ -291,6 +291,30 @@ static unsigned int nf_ct_bridge_pre(void *priv, struct sk_buff *skb,
return nf_conntrack_in(skb, &bridge_state);
}
+static unsigned int nf_ct_bridge_in(void *priv, struct sk_buff *skb,
+ const struct nf_hook_state *state)
+{
+ enum ip_conntrack_info ctinfo;
+ struct nf_conn *ct;
+
+ if (skb->pkt_type == PACKET_HOST)
+ return NF_ACCEPT;
+
+ /* nf_conntrack_confirm() cannot handle concurrent clones,
+ * this happens for broad/multicast frames with e.g. macvlan on top
+ * of the bridge device.
+ */
+ ct = nf_ct_get(skb, &ctinfo);
+ if (!ct || nf_ct_is_confirmed(ct) || nf_ct_is_template(ct))
+ return NF_ACCEPT;
+
+ /* let inet prerouting call conntrack again */
+ skb->_nfct = 0;
+ nf_ct_put(ct);
+
+ return NF_ACCEPT;
+}
+
static void nf_ct_bridge_frag_save(struct sk_buff *skb,
struct nf_bridge_frag_data *data)
{
@@ -386,6 +410,12 @@ static struct nf_hook_ops nf_ct_bridge_hook_ops[] __read_mostly = {
.priority = NF_IP_PRI_CONNTRACK,
},
{
+ .hook = nf_ct_bridge_in,
+ .pf = NFPROTO_BRIDGE,
+ .hooknum = NF_BR_LOCAL_IN,
+ .priority = NF_IP_PRI_CONNTRACK_CONFIRM,
+ },
+ {
.hook = nf_ct_bridge_post,
.pf = NFPROTO_BRIDGE,
.hooknum = NF_BR_POST_ROUTING,
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 9c4f427f3a50..ae86f751efc3 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -5169,10 +5169,9 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
struct net *net = sock_net(skb->sk);
struct ifinfomsg *ifm;
struct net_device *dev;
- struct nlattr *br_spec, *attr = NULL;
+ struct nlattr *br_spec, *attr, *br_flags_attr = NULL;
int rem, err = -EOPNOTSUPP;
u16 flags = 0;
- bool have_flags = false;
if (nlmsg_len(nlh) < sizeof(*ifm))
return -EINVAL;
@@ -5190,11 +5189,11 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC);
if (br_spec) {
nla_for_each_nested(attr, br_spec, rem) {
- if (nla_type(attr) == IFLA_BRIDGE_FLAGS && !have_flags) {
+ if (nla_type(attr) == IFLA_BRIDGE_FLAGS && !br_flags_attr) {
if (nla_len(attr) < sizeof(flags))
return -EINVAL;
- have_flags = true;
+ br_flags_attr = attr;
flags = nla_get_u16(attr);
}
@@ -5238,8 +5237,8 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
}
}
- if (have_flags)
- memcpy(nla_data(attr), &flags, sizeof(flags));
+ if (br_flags_attr)
+ memcpy(nla_data(br_flags_attr), &flags, sizeof(flags));
out:
return err;
}
diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c
index 2afe28712a7a..5d68cb181695 100644
--- a/net/hsr/hsr_forward.c
+++ b/net/hsr/hsr_forward.c
@@ -83,7 +83,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb)
return false;
/* Get next tlv */
- total_length += sizeof(struct hsr_sup_tlv) + hsr_sup_tag->tlv.HSR_TLV_length;
+ total_length += hsr_sup_tag->tlv.HSR_TLV_length;
if (!pskb_may_pull(skb, total_length))
return false;
skb_pull(skb, total_length);
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 2e5f3864d353..5b876fa7f9af 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -2756,6 +2756,7 @@ static const struct nf_ct_hook nf_conntrack_hook = {
.get_tuple_skb = nf_conntrack_get_tuple_skb,
.attach = nf_conntrack_attach,
.set_closing = nf_conntrack_set_closing,
+ .confirm = __nf_conntrack_confirm,
};
void nf_conntrack_init_end(void)
diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index 1f9474fefe84..d3d11dede545 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -359,10 +359,20 @@ static int nft_target_validate(const struct nft_ctx *ctx,
if (ctx->family != NFPROTO_IPV4 &&
ctx->family != NFPROTO_IPV6 &&
+ ctx->family != NFPROTO_INET &&
ctx->family != NFPROTO_BRIDGE &&
ctx->family != NFPROTO_ARP)
return -EOPNOTSUPP;
+ ret = nft_chain_validate_hooks(ctx->chain,
+ (1 << NF_INET_PRE_ROUTING) |
+ (1 << NF_INET_LOCAL_IN) |
+ (1 << NF_INET_FORWARD) |
+ (1 << NF_INET_LOCAL_OUT) |
+ (1 << NF_INET_POST_ROUTING));
+ if (ret)
+ return ret;
+
if (nft_is_base_chain(ctx->chain)) {
const struct nft_base_chain *basechain =
nft_base_chain(ctx->chain);
@@ -610,10 +620,20 @@ static int nft_match_validate(const struct nft_ctx *ctx,
if (ctx->family != NFPROTO_IPV4 &&
ctx->family != NFPROTO_IPV6 &&
+ ctx->family != NFPROTO_INET &&
ctx->family != NFPROTO_BRIDGE &&
ctx->family != NFPROTO_ARP)
return -EOPNOTSUPP;
+ ret = nft_chain_validate_hooks(ctx->chain,
+ (1 << NF_INET_PRE_ROUTING) |
+ (1 << NF_INET_LOCAL_IN) |
+ (1 << NF_INET_FORWARD) |
+ (1 << NF_INET_LOCAL_OUT) |
+ (1 << NF_INET_POST_ROUTING));
+ if (ret)
+ return ret;
+
if (nft_is_base_chain(ctx->chain)) {
const struct nft_base_chain *basechain =
nft_base_chain(ctx->chain);
diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index de96959336c4..211f57164cb6 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -52,6 +52,7 @@ struct tls_decrypt_arg {
struct_group(inargs,
bool zc;
bool async;
+ bool async_done;
u8 tail;
);
@@ -274,22 +275,30 @@ static int tls_do_decryption(struct sock *sk,
DEBUG_NET_WARN_ON_ONCE(atomic_read(&ctx->decrypt_pending) < 1);
atomic_inc(&ctx->decrypt_pending);
} else {
+ DECLARE_CRYPTO_WAIT(wait);
+
aead_request_set_callback(aead_req,
CRYPTO_TFM_REQ_MAY_BACKLOG,
- crypto_req_done, &ctx->async_wait);
+ crypto_req_done, &wait);
+ ret = crypto_aead_decrypt(aead_req);
+ if (ret == -EINPROGRESS || ret == -EBUSY)
+ ret = crypto_wait_req(ret, &wait);
+ return ret;
}
ret = crypto_aead_decrypt(aead_req);
+ if (ret == -EINPROGRESS)
+ return 0;
+
if (ret == -EBUSY) {
ret = tls_decrypt_async_wait(ctx);
- ret = ret ?: -EINPROGRESS;
+ darg->async_done = true;
+ /* all completions have run, we're not doing async anymore */
+ darg->async = false;
+ return ret;
}
- if (ret == -EINPROGRESS) {
- if (darg->async)
- return 0;
- ret = crypto_wait_req(ret, &ctx->async_wait);
- }
+ atomic_dec(&ctx->decrypt_pending);
darg->async = false;
return ret;
@@ -1588,8 +1597,11 @@ static int tls_decrypt_sg(struct sock *sk, struct iov_iter *out_iov,
/* Prepare and submit AEAD request */
err = tls_do_decryption(sk, sgin, sgout, dctx->iv,
data_len + prot->tail_size, aead_req, darg);
- if (err)
+ if (err) {
+ if (darg->async_done)
+ goto exit_free_skb;
goto exit_free_pages;
+ }
darg->skb = clear_skb ?: tls_strp_msg(ctx);
clear_skb = NULL;
@@ -1601,6 +1613,9 @@ static int tls_decrypt_sg(struct sock *sk, struct iov_iter *out_iov,
return err;
}
+ if (unlikely(darg->async_done))
+ return 0;
+
if (prot->tail_size)
darg->tail = dctx->tail;
@@ -1948,6 +1963,7 @@ int tls_sw_recvmsg(struct sock *sk,
struct strp_msg *rxm;
struct tls_msg *tlm;
ssize_t copied = 0;
+ ssize_t peeked = 0;
bool async = false;
int target, err;
bool is_kvec = iov_iter_is_kvec(&msg->msg_iter);
@@ -2095,8 +2111,10 @@ put_on_rx_list:
if (err < 0)
goto put_on_rx_list_err;
- if (is_peek)
+ if (is_peek) {
+ peeked += chunk;
goto put_on_rx_list;
+ }
if (partially_consumed) {
rxm->offset += chunk;
@@ -2135,8 +2153,8 @@ recv_end:
/* Drain records from the rx_list & copy if required */
if (is_peek || is_kvec)
- err = process_rx_list(ctx, msg, &control, copied,
- decrypted, is_peek, NULL);
+ err = process_rx_list(ctx, msg, &control, copied + peeked,
+ decrypted - peeked, is_peek, NULL);
else
err = process_rx_list(ctx, msg, &control, 0,
async_copy_bytes, is_peek, NULL);
diff --git a/tools/net/ynl/lib/ynl.c b/tools/net/ynl/lib/ynl.c
index 6e6d474c8366..45e49671ae87 100644
--- a/tools/net/ynl/lib/ynl.c
+++ b/tools/net/ynl/lib/ynl.c
@@ -523,6 +523,7 @@ ynl_get_family_info_mcast(struct ynl_sock *ys, const struct nlattr *mcasts)
ys->mcast_groups[i].name[GENL_NAMSIZ - 1] = 0;
}
}
+ i++;
}
return 0;
diff --git a/tools/testing/selftests/netfilter/Makefile b/tools/testing/selftests/netfilter/Makefile
index db27153eb4a0..936c3085bb83 100644
--- a/tools/testing/selftests/netfilter/Makefile
+++ b/tools/testing/selftests/netfilter/Makefile
@@ -7,7 +7,8 @@ TEST_PROGS := nft_trans_stress.sh nft_fib.sh nft_nat.sh bridge_brouter.sh \
nft_queue.sh nft_meta.sh nf_nat_edemux.sh \
ipip-conntrack-mtu.sh conntrack_tcp_unreplied.sh \
conntrack_vrf.sh nft_synproxy.sh rpath.sh nft_audit.sh \
- conntrack_sctp_collision.sh xt_string.sh
+ conntrack_sctp_collision.sh xt_string.sh \
+ bridge_netfilter.sh
HOSTPKG_CONFIG := pkg-config
diff --git a/tools/testing/selftests/netfilter/bridge_netfilter.sh b/tools/testing/selftests/netfilter/bridge_netfilter.sh
new file mode 100644
index 000000000000..659b3ab02c8b
--- /dev/null
+++ b/tools/testing/selftests/netfilter/bridge_netfilter.sh
@@ -0,0 +1,188 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Test bridge netfilter + conntrack, a combination that doesn't really work,
+# with multicast/broadcast packets racing for hash table insertion.
+
+# eth0 br0 eth0
+# setup is: ns1 <->,ns0 <-> ns3
+# ns2 <-' `'-> ns4
+
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+ret=0
+
+sfx=$(mktemp -u "XXXXXXXX")
+ns0="ns0-$sfx"
+ns1="ns1-$sfx"
+ns2="ns2-$sfx"
+ns3="ns3-$sfx"
+ns4="ns4-$sfx"
+
+ebtables -V > /dev/null 2>&1
+if [ $? -ne 0 ];then
+ echo "SKIP: Could not run test without ebtables"
+ exit $ksft_skip
+fi
+
+ip -Version > /dev/null 2>&1
+if [ $? -ne 0 ];then
+ echo "SKIP: Could not run test without ip tool"
+ exit $ksft_skip
+fi
+
+for i in $(seq 0 4); do
+ eval ip netns add \$ns$i
+done
+
+cleanup() {
+ for i in $(seq 0 4); do eval ip netns del \$ns$i;done
+}
+
+trap cleanup EXIT
+
+do_ping()
+{
+ fromns="$1"
+ dstip="$2"
+
+ ip netns exec $fromns ping -c 1 -q $dstip > /dev/null
+ if [ $? -ne 0 ]; then
+ echo "ERROR: ping from $fromns to $dstip"
+ ip netns exec ${ns0} nft list ruleset
+ ret=1
+ fi
+}
+
+bcast_ping()
+{
+ fromns="$1"
+ dstip="$2"
+
+ for i in $(seq 1 1000); do
+ ip netns exec $fromns ping -q -f -b -c 1 -q $dstip > /dev/null 2>&1
+ if [ $? -ne 0 ]; then
+ echo "ERROR: ping -b from $fromns to $dstip"
+ ip netns exec ${ns0} nft list ruleset
+ fi
+ done
+}
+
+ip link add veth1 netns ${ns0} type veth peer name eth0 netns ${ns1}
+if [ $? -ne 0 ]; then
+ echo "SKIP: Can't create veth device"
+ exit $ksft_skip
+fi
+
+ip link add veth2 netns ${ns0} type veth peer name eth0 netns $ns2
+ip link add veth3 netns ${ns0} type veth peer name eth0 netns $ns3
+ip link add veth4 netns ${ns0} type veth peer name eth0 netns $ns4
+
+ip -net ${ns0} link set lo up
+
+for i in $(seq 1 4); do
+ ip -net ${ns0} link set veth$i up
+done
+
+ip -net ${ns0} link add br0 type bridge stp_state 0 forward_delay 0 nf_call_iptables 1 nf_call_ip6tables 1 nf_call_arptables 1
+if [ $? -ne 0 ]; then
+ echo "SKIP: Can't create bridge br0"
+ exit $ksft_skip
+fi
+
+# make veth0,1,2 part of bridge.
+for i in $(seq 1 3); do
+ ip -net ${ns0} link set veth$i master br0
+done
+
+# add a macvlan on top of the bridge.
+MACVLAN_ADDR=ba:f3:13:37:42:23
+ip -net ${ns0} link add link br0 name macvlan0 type macvlan mode private
+ip -net ${ns0} link set macvlan0 address ${MACVLAN_ADDR}
+ip -net ${ns0} link set macvlan0 up
+ip -net ${ns0} addr add 10.23.0.1/24 dev macvlan0
+
+# add a macvlan on top of veth4.
+MACVLAN_ADDR=ba:f3:13:37:42:24
+ip -net ${ns0} link add link veth4 name macvlan4 type macvlan mode vepa
+ip -net ${ns0} link set macvlan4 address ${MACVLAN_ADDR}
+ip -net ${ns0} link set macvlan4 up
+
+# make the macvlan part of the bridge.
+# veth4 is not a bridge port, only the macvlan on top of it.
+ip -net ${ns0} link set macvlan4 master br0
+
+ip -net ${ns0} link set br0 up
+ip -net ${ns0} addr add 10.0.0.1/24 dev br0
+ip netns exec ${ns0} sysctl -q net.bridge.bridge-nf-call-iptables=1
+ret=$?
+if [ $ret -ne 0 ] ; then
+ echo "SKIP: bridge netfilter not available"
+ ret=$ksft_skip
+fi
+
+# for testing, so namespaces will reply to ping -b probes.
+ip netns exec ${ns0} sysctl -q net.ipv4.icmp_echo_ignore_broadcasts=0
+
+# enable conntrack in ns0 and drop broadcast packets in forward to
+# avoid them from getting confirmed in the postrouting hook before
+# the cloned skb is passed up the stack.
+ip netns exec ${ns0} nft -f - <<EOF
+table ip filter {
+ chain input {
+ type filter hook input priority 1; policy accept
+ iifname br0 counter
+ ct state new accept
+ }
+}
+
+table bridge filter {
+ chain forward {
+ type filter hook forward priority 0; policy accept
+ meta pkttype broadcast ip protocol icmp counter drop
+ }
+}
+EOF
+
+# place 1, 2 & 3 in same subnet, connected via ns0:br0.
+# ns4 is placed in same subnet as well, but its not
+# part of the bridge: the corresponding veth4 is not
+# part of the bridge, only its macvlan interface.
+for i in $(seq 1 4); do
+ eval ip -net \$ns$i link set lo up
+ eval ip -net \$ns$i link set eth0 up
+done
+for i in $(seq 1 2); do
+ eval ip -net \$ns$i addr add 10.0.0.1$i/24 dev eth0
+done
+
+ip -net ${ns3} addr add 10.23.0.13/24 dev eth0
+ip -net ${ns4} addr add 10.23.0.14/24 dev eth0
+
+# test basic connectivity
+do_ping ${ns1} 10.0.0.12
+do_ping ${ns3} 10.23.0.1
+do_ping ${ns4} 10.23.0.1
+
+if [ $ret -eq 0 ];then
+ echo "PASS: netns connectivity: ns1 can reach ns2, ns3 and ns4 can reach ns0"
+fi
+
+bcast_ping ${ns1} 10.0.0.255
+
+# This should deliver broadcast to macvlan0, which is on top of ns0:br0.
+bcast_ping ${ns3} 10.23.0.255
+
+# same, this time via veth4:macvlan4.
+bcast_ping ${ns4} 10.23.0.255
+
+read t < /proc/sys/kernel/tainted
+
+if [ $t -eq 0 ];then
+ echo PASS: kernel not tainted
+else
+ echo ERROR: kernel is tainted
+ ret=1
+fi
+
+exit $ret
generated by cgit (git 2.34.1) at 2024-06-09 06:29:29 +0000