linux/net/core
Toke Høiland-Jørgensen ad1e03b2b3 core: Don't skip generic XDP program execution for cloned SKBs
The current generic XDP handler skips execution of XDP programs entirely if
an SKB is marked as cloned. This leads to some surprising behaviour, as
packets can end up being cloned in various ways, which will make an XDP
program not see all the traffic on an interface.

This was discovered by a simple test case where an XDP program that always
returns XDP_DROP is installed on a veth device. When combining this with
the Scapy packet sniffer (which uses an AF_PACKET) socket on the sending
side, SKBs reliably end up in the cloned state, causing them to be passed
through to the receiving interface instead of being dropped. A minimal
reproducer script for this is included below.

This patch fixed the issue by simply triggering the existing linearisation
code for cloned SKBs instead of skipping the XDP program execution. This
behaviour is in line with the behaviour of the native XDP implementation
for the veth driver, which will reallocate and copy the SKB data if the SKB
is marked as shared.

Reproducer Python script (requires BCC and Scapy):

from scapy.all import TCP, IP, Ether, sendp, sniff, AsyncSniffer, Raw, UDP
from bcc import BPF
import time, sys, subprocess, shlex

SKB_MODE = (1 << 1)
DRV_MODE = (1 << 2)
PYTHON=sys.executable

def client():
    time.sleep(2)
    # Sniffing on the sender causes skb_cloned() to be set
    s = AsyncSniffer()
    s.start()

    for p in range(10):
        sendp(Ether(dst="aa:aa:aa:aa:aa:aa", src="cc:cc:cc:cc:cc:cc")/IP()/UDP()/Raw("Test"),
              verbose=False)
        time.sleep(0.1)

    s.stop()
    return 0

def server(mode):
    prog = BPF(text="int dummy_drop(struct xdp_md *ctx) {return XDP_DROP;}")
    func = prog.load_func("dummy_drop", BPF.XDP)
    prog.attach_xdp("a_to_b", func, mode)

    time.sleep(1)

    s = sniff(iface="a_to_b", count=10, timeout=15)
    if len(s):
        print(f"Got {len(s)} packets - should have gotten 0")
        return 1
    else:
        print("Got no packets - as expected")
        return 0

if len(sys.argv) < 2:
    print(f"Usage: {sys.argv[0]} <skb|drv>")
    sys.exit(1)

if sys.argv[1] == "client":
    sys.exit(client())
elif sys.argv[1] == "server":
    mode = SKB_MODE if sys.argv[2] == 'skb' else DRV_MODE
    sys.exit(server(mode))
else:
    try:
        mode = sys.argv[1]
        if mode not in ('skb', 'drv'):
            print(f"Usage: {sys.argv[0]} <skb|drv>")
            sys.exit(1)
        print(f"Running in {mode} mode")

        for cmd in [
                'ip netns add netns_a',
                'ip netns add netns_b',
                'ip -n netns_a link add a_to_b type veth peer name b_to_a netns netns_b',
                # Disable ipv6 to make sure there's no address autoconf traffic
                'ip netns exec netns_a sysctl -qw net.ipv6.conf.a_to_b.disable_ipv6=1',
                'ip netns exec netns_b sysctl -qw net.ipv6.conf.b_to_a.disable_ipv6=1',
                'ip -n netns_a link set dev a_to_b address aa:aa:aa:aa:aa:aa',
                'ip -n netns_b link set dev b_to_a address cc:cc:cc:cc:cc:cc',
                'ip -n netns_a link set dev a_to_b up',
                'ip -n netns_b link set dev b_to_a up']:
            subprocess.check_call(shlex.split(cmd))

        server = subprocess.Popen(shlex.split(f"ip netns exec netns_a {PYTHON} {sys.argv[0]} server {mode}"))
        client = subprocess.Popen(shlex.split(f"ip netns exec netns_b {PYTHON} {sys.argv[0]} client"))

        client.wait()
        server.wait()
        sys.exit(server.returncode)

    finally:
        subprocess.run(shlex.split("ip netns delete netns_a"))
        subprocess.run(shlex.split("ip netns delete netns_b"))

Fixes: d445516966 ("net: xdp: support xdp generic on virtual devices")
Reported-by: Stepan Horacek <shoracek@redhat.com>
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-02-11 17:01:29 -08:00
..
bpf_sk_storage.c bpf: Improve bucket_log calculation logic 2020-02-07 23:01:41 +01:00
datagram.c net: add queue argument to __skb_wait_for_more_packets and __skb_{,try_}recv_datagram 2019-12-09 09:59:07 +01:00
datagram.h net/core: Allow the compiler to verify declaration and definition consistency 2019-03-27 13:49:44 -07:00
dev_addr_lists.c net: remove unnecessary variables and callback 2019-10-24 14:53:49 -07:00
dev_ioctl.c net: Introduce peer to peer one step PTP time stamping. 2019-12-25 19:51:34 -08:00
dev.c core: Don't skip generic XDP program execution for cloned SKBs 2020-02-11 17:01:29 -08:00
devlink.c devlink: report 0 after hitting end in region read 2020-02-05 14:23:20 +01:00
drop_monitor.c drop_monitor: Do not cancel uninitialized work item 2020-02-07 18:48:36 +01:00
dst_cache.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
dst.c net: print proper warning on dst underflow 2019-09-26 09:05:56 +02:00
failover.c failover: allow name change on IFF_UP slave interfaces 2019-04-10 22:12:26 -07:00
fib_notifier.c net: fib_notifier: propagate extack down to the notifier block callback 2019-10-04 11:10:56 -07:00
fib_rules.c net: fib_notifier: propagate extack down to the notifier block callback 2019-10-04 11:10:56 -07:00
filter.c treewide: remove redundant IS_ERR() before error code check 2020-02-04 03:05:27 +00:00
flow_dissector.c flow_dissector: Fix to use new variables for port ranges in bpf hook 2020-01-27 11:25:07 +01:00
flow_offload.c net: core: rename indirect block ingress cb function 2019-12-06 20:45:09 -08:00
gen_estimator.c net_sched: gen_estimator: extend packet counter to 64bit 2019-11-06 21:51:36 -08:00
gen_stats.c net_sched: add TCA_STATS_PKT64 attribute 2019-11-05 18:20:55 -08:00
gro_cells.c gro_cells: make sure device is up in gro_cells_receive() 2019-03-10 11:07:14 -07:00
hwbm.c net: hwbm: Make the hwbm_pool lock a mutex 2019-06-09 19:40:10 -07:00
link_watch.c net: link_watch: prevent starvation when processing linkwatch wq 2019-07-01 19:02:47 -07:00
lwt_bpf.c net: ipv6_stub: use ip6_dst_lookup_flow instead of ip6_dst_lookup 2019-12-04 12:27:13 -08:00
lwtunnel.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
Makefile ethtool: move to its own directory 2019-12-12 17:07:05 -08:00
neighbour.c neigh_stat_seq_next() should increase position index 2020-01-24 11:42:18 +01:00
net_namespace.c netns: Constify exported functions 2020-01-17 13:25:24 +01:00
net-procfs.c net: procfs: use index hashlist instead of name hashlist 2019-10-01 14:47:19 -07:00
net-sysfs.c net-sysfs: Call dev_hold always in rx_queue_add_kobject 2019-12-17 22:57:11 -08:00
net-sysfs.h
net-traces.c page_pool: add tracepoints for page_pool with details need by XDP 2019-06-19 11:23:13 -04:00
netclassid_cgroup.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
netevent.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
netpoll.c net: fix skb use after free in netpoll 2019-08-27 20:52:02 -07:00
netprio_cgroup.c netprio: use css ID instead of cgroup ID 2019-11-12 08:18:03 -08:00
page_pool.c page_pool: help compiler remove code in case CONFIG_NUMA=n 2020-01-02 15:37:52 -08:00
pktgen.c proc: convert everything to "struct proc_ops" 2020-02-04 03:05:26 +00:00
ptp_classifier.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 295 2019-06-05 17:36:38 +02:00
request_sock.c tcp: add rcu protection around tp->fastopen_rsk 2019-10-13 10:13:08 -07:00
rtnetlink.c net/core: Do not clear VF index for node/port GUIDs query 2020-01-30 15:20:26 +01:00
scm.c y2038: socket: remove timespec reference in timestamping 2019-11-15 14:38:29 +01:00
secure_seq.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
skbuff.c net: Support GRO/GSO fraglist chaining. 2020-01-27 11:00:21 +01:00
skmsg.c net, sk_msg: Don't check if sock is locked when tearing down psock 2020-01-22 20:30:20 +01:00
sock_diag.c sock: make cookie generation global instead of per netns 2019-08-09 13:14:46 -07:00
sock_map.c bpf, sockhash: Synchronize_rcu before free'ing map 2020-02-07 22:36:26 +01:00
sock_reuseport.c soreuseport: Cleanup duplicate initialization of more_reuse->max_socks. 2020-01-27 11:01:16 +01:00
sock.c xsk, net: Make sock_def_readable() have external linkage 2020-01-22 00:08:52 +01:00
stream.c tcp: make sure EPOLLOUT wont be missed 2019-08-19 13:07:43 -07:00
sysctl_net_core.c net, sysctl: Fix compiler warning when only cBPF is present 2019-12-19 17:17:51 +01:00
timestamping.c net: Introduce a new MII time stamping interface. 2019-12-25 19:51:33 -08:00
tso.c net: Use skb accessors in network core 2019-07-22 20:47:56 -07:00
utils.c net: Fix skb->csum update in inet_proto_csum_replace16(). 2020-01-24 20:54:30 +01:00
xdp.c treewide: Use sizeof_field() macro 2019-12-09 10:36:44 -08:00