In preparation for msgr2, make the cluster send us maps with addrvecs
including both LEGACY and MSGR2 addrs instead of a single LEGACY addr.
This means advertising support for SERVER_NAUTILUS and also some older
features: SERVER_MIMIC, MONENC and MONNAMES.
MONNAMES and MONENC are actually pre-argonaut, we just never updated
ceph_monmap_decode() for them. Decoding is unconditional, see commit
23c625ce30 ("libceph: assume argonaut on the server side").
SERVER_MIMIC doesn't bear any meaning for the kernel client.
Since ceph_decode_entity_addrvec() is guarded by encoding version
checks (and in msgr2 case it is guarded implicitly by the fact that
server is speaking msgr2), we assume MSG_ADDR2 for it.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
In preparation for msgr2, factor out finish_auth() so it is suitable
for both existing MAuth message based authentication and upcoming msgr2
authentication exchange.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
In msgr2, initial authentication happens with an exchange of msgr2
control frames -- MAuth message and struct ceph_mon_request_header
aren't used. Make that optional.
Stop reporting cephx protocol as "x". Use "cephx" instead.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
- request service tickets together with auth ticket. Currently we get
auth ticket via CEPHX_GET_AUTH_SESSION_KEY op and then request service
tickets via CEPHX_GET_PRINCIPAL_SESSION_KEY op in a separate message.
Since nautilus, desired service tickets are shared togther with auth
ticket in CEPHX_GET_AUTH_SESSION_KEY reply.
- propagate session key and connection secret, if any. In preparation
for msgr2, update handle_reply() and verify_authorizer_reply() auth
ops to propagate session key and connection secret. Since nautilus,
if secure mode is negotiated, connection secret is shared either in
CEPHX_GET_AUTH_SESSION_KEY reply (for mons) or in a final authorizer
reply (for osds and mdses).
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Make it clear that "need" is a union of "missing" and "have, but up
for renewal" and dout when the ticket goes missing due to expiry or
invalidation by client.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
A pure move, no other changes.
Note that ceph_tcp_recv{msg,page}() and ceph_tcp_send{msg,page}()
helpers are also moved. msgr2 will bring its own, more efficient,
variants based on iov_iter. Switching msgr1 to them was considered
but decided against to avoid subtle regressions.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
In preparation for msgr2, define internal messenger <-> protocol
interface (as opposed to external messenger <-> client interface, which
is struct ceph_connection_operations) consisting of try_read(),
try_write(), revoke(), revoke_incoming(), opened(), reset_session() and
reset_protocol() ops. The semantics are exactly the same as they are
now.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
In preparation for msgr2, rename msgr1 specific states and move the
defines to the header file.
Also drop state transition comments. They don't cover all possible
transitions (e.g. NEGOTIATING -> STANDBY, etc) and currently do more
harm than good.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
unsigned long is a leftover from when con->state used to be a set of
bits managed with set_bit(), clear_bit(), etc. Save a bit of memory.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Our messenger instance addr->port is normally zero -- anything else is
nonsensical because as a client we connect to multiple servers and don't
listen on any port. However, a user can supply an arbitrary addr:port
via ip option and the port is currently preserved. Zero it.
Conversely, make sure our addr->nonce is non-zero. A zero nonce is
special: in combination with a zero port, it is used to blocklist the
entire ip.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Move the logic of grabbing the next message from the queue into its own
function. Like ceph_con_in_msg_alloc(), this is protocol independent.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
ceph_con_in_msg_alloc() is protocol independent, but con->in_hdr (and
struct ceph_msg_header in general) is msgr1 specific. While the struct
is deeply ingrained inside and outside the messenger, con->in_hdr field
can be separated.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Stick with pr_info message because session reset isn't an error most of
the time. When it is (i.e. if the server denies the reconnect attempt),
we get a bunch of other pr_err messages.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
con->peer_global_seq is part of session state. Clear it when
the server tells us to reset, not just in ceph_con_close().
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Move protocol reset bits into ceph_con_reset_protocol(), leaving
just session reset bits.
Note that con->out_skip is now reset on faults. This fixes a crash
in the case of a stateful session getting a fault while in the middle
of revoking a message.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
A fault due to a version mismatch or a feature set mismatch used to be
treated differently from other faults: the connection would get closed
without trying to reconnect and there was a ->bad_proto() connection op
for notifying about that.
This changed a long time ago, see commits 6384bb8b8e ("libceph: kill
bad_proto ceph connection op") and 0fa6ebc600 ("libceph: fix protocol
feature mismatch failure path"). Nowadays these aren't any different
from other faults (i.e. we try to reconnect even though the mismatch
won't resolve until the server is replaced). reset_connection() calls
there are rather confusing because reset_connection() resets a session
together an individual instance of the protocol. This is cleaned up
in the next patch.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
The current setting allows the backoff to climb up to 5 minutes. This
is too high -- it becomes hard to tell whether the client is stuck on
something or just in backoff.
In userspace, ms_max_backoff is defaulted to 15 seconds. Let's do the
same.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Alexei Starovoitov says:
====================
pull-request: bpf 2020-12-10
The following pull-request contains BPF updates for your *net* tree.
We've added 21 non-merge commits during the last 12 day(s) which contain
a total of 21 files changed, 163 insertions(+), 88 deletions(-).
The main changes are:
1) Fix propagation of 32-bit signed bounds from 64-bit bounds, from Alexei.
2) Fix ring_buffer__poll() return value, from Andrii.
3) Fix race in lwt_bpf, from Cong.
4) Fix test_offload, from Toke.
5) Various xsk fixes.
Please consider pulling these changes from:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git
Thanks a lot!
Also thanks to reporters, reviewers and testers of commits in this pull-request:
Cong Wang, Hulk Robot, Jakub Kicinski, Jean-Philippe Brucker, John
Fastabend, Magnus Karlsson, Maxim Mikityanskiy, Yonghong Song
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
1) Switch to RCU in x_tables to fix possible NULL pointer dereference,
from Subash Abhinov Kasiviswanathan.
2) Fix netlink dump of dynset timeouts later than 23 days.
3) Add comment for the indirect serialization of the nft commit mutex
with rtnl_mutex.
4) Remove bogus check for confirmed conntrack when matching on the
conntrack ID, from Brett Mastbergen.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When cwnd is not a multiple of the TSO skb size of N*MSS, we can get
into persistent scenarios where we have the following sequence:
(1) ACK for full-sized skb of N*MSS arrives
-> tcp_write_xmit() transmit full-sized skb with N*MSS
-> move pacing release time forward
-> exit tcp_write_xmit() because pacing time is in the future
(2) TSQ callback or TCP internal pacing timer fires
-> try to transmit next skb, but TSO deferral finds remainder of
available cwnd is not big enough to trigger an immediate send
now, so we defer sending until the next ACK.
(3) repeat...
So we can get into a case where we never mark ourselves as
cwnd-limited for many seconds at a time, even with
bulk/infinite-backlog senders, because:
o In case (1) above, every time in tcp_write_xmit() we have enough
cwnd to send a full-sized skb, we are not fully using the cwnd
(because cwnd is not a multiple of the TSO skb size). So every time we
send data, we are not cwnd limited, and so in the cwnd-limited
tracking code in tcp_cwnd_validate() we mark ourselves as not
cwnd-limited.
o In case (2) above, every time in tcp_write_xmit() that we try to
transmit the "remainder" of the cwnd but defer, we set the local
variable is_cwnd_limited to true, but we do not send any packets, so
sent_pkts is zero, so we don't call the cwnd-limited logic to update
tp->is_cwnd_limited.
Fixes: ca8a226343 ("tcp: make cwnd-limited checks measurement-based, and gentler")
Reported-by: Ingemar Johansson <ingemar.s.johansson@ericsson.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20201209035759.1225145-1-ncardwell.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The offending commit introduces a cleanup callback that is invoked
when the driver module is removed to clean up the tunnel device
flow block. But it returns on the first iteration of the for loop.
The remaining indirect flow blocks will never be freed.
Fixes: 1fac52da59 ("net: flow_offload: consolidate indirect flow_block infrastructure")
CC: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
For DCTCP, we have to retain the ECT bits set by the congestion control
algorithm on the socket when reflecting syn TOS in syn-ack, in order to
make ECN work properly.
Fixes: ac8f1710c1 ("tcp: reflect tos value received in SYN to the socket")
Reported-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: Wei Wang <weiwan@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Syzbot reported a stack overflow in bitmap_from_arr32() called from
ethnl_parse_bitset() when bitset from netlink message is longer than
target bitmap length. While ethnl_compact_sanity_checks() makes sure that
trailing part is all zeros (i.e. the request does not try to touch bits
kernel does not recognize), we also need to cap change_bits to nbits so
that we don't try to write past the prepared bitmaps.
Fixes: 88db6d1e4f ("ethtool: add ethnl_parse_bitset() helper")
Reported-by: syzbot+9d39fa49d4df294aab93@syzkaller.appspotmail.com
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Link: https://lore.kernel.org/r/3487ee3a98e14cd526f55b6caaa959d2dcbcad9f.1607465316.git.mkubecek@suse.cz
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The isotp socket can be widely configured in its behaviour regarding addressing
types, fill-ups, receive pattern tests and link layer length. Usually all
these settings need to be fixed before bind() and can not be changed
afterwards.
This patch adds a check to enforce the common usage pattern.
Fixes: e057dd3fc2 ("can: add ISO 15765-2:2016 transport protocol")
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Tested-by: Thomas Wagner <thwa1@web.de>
Link: https://lore.kernel.org/r/20201203140604.25488-2-socketcan@hartkopp.net
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Link: https://lore.kernel.org/r/20201204133508.742120-3-mkl@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since commit 7f0a838254 ("bpf, xdp: Maintain info on attached XDP BPF
programs in net_device"), the XDP program attachment info is now maintained
in the core code. This interacts badly with the xdp_attachment_flags_ok()
check that prevents unloading an XDP program with different load flags than
it was loaded with. In practice, two kinds of failures are seen:
- An XDP program loaded without specifying a mode (and which then ends up
in driver mode) cannot be unloaded if the program mode is specified on
unload.
- The dev_xdp_uninstall() hook always calls the driver callback with the
mode set to the type of the program but an empty flags argument, which
means the flags_ok() check prevents the program from being removed,
leading to bpf prog reference leaks.
The original reason this check was added was to avoid ambiguity when
multiple programs were loaded. With the way the checks are done in the core
now, this is quite simple to enforce in the core code, so let's add a check
there and get rid of the xdp_attachment_flags_ok() callback entirely.
Fixes: 7f0a838254 ("bpf, xdp: Maintain info on attached XDP BPF programs in net_device")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/bpf/160752225751.110217.10267659521308669050.stgit@toke.dk
Since commit 656c8e9cc1 ("netfilter: conntrack: Use consistent ct id
hash calculation") the ct id will not change from initialization to
confirmation. Removing the confirmation check allows for things like
adding an element to a 'typeof ct id' set in prerouting upon reception
of the first packet of a new connection, and then being able to
reference that set consistently both before and after the connection
is confirmed.
Fixes: 656c8e9cc1 ("netfilter: conntrack: Use consistent ct id hash calculation")
Signed-off-by: Brett Mastbergen <brett.mastbergen@gmail.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Before commit a337531b94 ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB")
small tcp_rmem[1] values were overridden by tcp_fixup_rcvbuf() to accommodate various MSS.
This is no longer the case, and Hazem Mohamed Abuelfotoh reported
that DRS would not work for MTU 9000 endpoints receiving regular (1500 bytes) frames.
Root cause is that tcp_init_buffer_space() uses tp->rcv_wnd for upper limit
of rcvq_space.space computation, while it can select later a smaller
value for tp->rcv_ssthresh and tp->window_clamp.
ss -temoi on receiver would show :
skmem:(r0,rb131072,t0,tb46080,f0,w0,o0,bl0,d0) rcv_space:62496 rcv_ssthresh:56596
This means that TCP can not increase its window in tcp_grow_window(),
and that DRS can never kick.
Fix this by making sure that rcvq_space.space is not bigger than number of bytes
that can be held in TCP receive queue.
People unable/unwilling to change their kernel can work around this issue by
selecting a bigger tcp_rmem[1] value as in :
echo "4096 196608 6291456" >/proc/sys/net/ipv4/tcp_rmem
Based on an initial report and patch from Hazem Mohamed Abuelfotoh
https://lore.kernel.org/netdev/20201204180622.14285-1-abuehaze@amazon.com/
Fixes: a337531b94 ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB")
Fixes: 041a14d267 ("tcp: start receiver buffer autotuning sooner")
Reported-by: Hazem Mohamed Abuelfotoh <abuehaze@amazon.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
`tipc_node_apply_property` does a null check on a `tipc_link_entry`
pointer but also accesses the same pointer out of the null check block.
This triggers a warning on Coverity Static Analyzer because we're
implying that `e->link` can BE null.
Move "Update MTU for node link entry" line into if block to make sure
that we're not in a state that `e->link` is null.
Signed-off-by: Cengiz Can <cengiz@kernel.wtf>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add an explicit comment in the code to describe the indirect
serialization of the holders of the commit_mutex with the rtnl_mutex.
Commit 90d2723c6d ("netfilter: nf_tables: do not hold reference on
netdevice from preparation phase") already describes this, but a comment
in this case is better for reference.
Reported-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use nf_msecs_to_jiffies64 and nf_jiffies64_to_msecs as provided by
8e1102d5a1 ("netfilter: nf_tables: support timeouts larger than 23
days"), otherwise ruleset listing breaks.
Fixes: a8b1e36d0d ("netfilter: nft_dynset: fix element timeout for HZ != 1000")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When running concurrent iptables rules replacement with data, the per CPU
sequence count is checked after the assignment of the new information.
The sequence count is used to synchronize with the packet path without the
use of any explicit locking. If there are any packets in the packet path using
the table information, the sequence count is incremented to an odd value and
is incremented to an even after the packet process completion.
The new table value assignment is followed by a write memory barrier so every
CPU should see the latest value. If the packet path has started with the old
table information, the sequence counter will be odd and the iptables
replacement will wait till the sequence count is even prior to freeing the
old table info.
However, this assumes that the new table information assignment and the memory
barrier is actually executed prior to the counter check in the replacement
thread. If CPU decides to execute the assignment later as there is no user of
the table information prior to the sequence check, the packet path in another
CPU may use the old table information. The replacement thread would then free
the table information under it leading to a use after free in the packet
processing context-
Unable to handle kernel NULL pointer dereference at virtual
address 000000000000008e
pc : ip6t_do_table+0x5d0/0x89c
lr : ip6t_do_table+0x5b8/0x89c
ip6t_do_table+0x5d0/0x89c
ip6table_filter_hook+0x24/0x30
nf_hook_slow+0x84/0x120
ip6_input+0x74/0xe0
ip6_rcv_finish+0x7c/0x128
ipv6_rcv+0xac/0xe4
__netif_receive_skb+0x84/0x17c
process_backlog+0x15c/0x1b8
napi_poll+0x88/0x284
net_rx_action+0xbc/0x23c
__do_softirq+0x20c/0x48c
This could be fixed by forcing instruction order after the new table
information assignment or by switching to RCU for the synchronization.
Fixes: 80055dab5d ("netfilter: x_tables: make xt_replace_table wait until old rules are not used anymore")
Reported-by: Sean Tranchetti <stranche@codeaurora.org>
Reported-by: kernel test robot <lkp@intel.com>
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Steffen Klassert says:
====================
pull request (net): ipsec 2020-12-07
1) Sysbot reported fixes for the new 64/32 bit compat layer.
From Dmitry Safonov.
2) Fix a memory leak in xfrm_user_policy that was introduced
by adding the 64/32 bit compat layer. From Yu Kuai.
* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec:
net: xfrm: fix memory leak in xfrm_user_policy()
xfrm/compat: Don't allocate memory with __GFP_ZERO
xfrm/compat: memset(0) 64-bit padding at right place
xfrm/compat: Translate by copying XFRMA_UNSPEC attribute
====================
Link: https://lore.kernel.org/r/20201207093937.2874932-1-steffen.klassert@secunet.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When do cat /proc/net/netstat, the output isn't append with a new line, it looks like this:
[root@localhost ~]# cat /proc/net/netstat
...
MPTcpExt: 0 0 0 0 0 0 0 0 0 0 0 0 0[root@localhost ~]#
This is because in mptcp_seq_show(), if mptcp isn't in use, net->mib.mptcp_statistics is NULL,
so it just puts all 0 after "MPTcpExt:", and return, forgot the '\n'.
After this patch:
[root@localhost ~]# cat /proc/net/netstat
...
MPTcpExt: 0 0 0 0 0 0 0 0 0 0 0 0 0
[root@localhost ~]#
Fixes: fc518953bc ("mptcp: add and use MIB counter infrastructure")
Signed-off-by: Jianguo Wu <wujianguo@chinatelecom.cn>
Acked-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/142e2fd9-58d9-bb13-fb75-951cccc2331e@163.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When enabling multicast snooping, bridge module deadlocks on multicast_lock
if 1) IPv6 is enabled, and 2) there is an existing querier on the same L2
network.
The deadlock was caused by the following sequence: While holding the lock,
br_multicast_open calls br_multicast_join_snoopers, which eventually causes
IP stack to (attempt to) send out a Listener Report (in igmp6_join_group).
Since the destination Ethernet address is a multicast address, br_dev_xmit
feeds the packet back to the bridge via br_multicast_rcv, which in turn
calls br_multicast_add_group, which then deadlocks on multicast_lock.
The fix is to move the call br_multicast_join_snoopers outside of the
critical section. This works since br_multicast_join_snoopers only deals
with IP and does not modify any multicast data structures of the bridge,
so there's no need to hold the lock.
Steps to reproduce:
1. sysctl net.ipv6.conf.all.force_mld_version=1
2. have another querier
3. ip link set dev bridge type bridge mcast_snooping 0 && \
ip link set dev bridge type bridge mcast_snooping 1 < deadlock >
A typical call trace looks like the following:
[ 936.251495] _raw_spin_lock+0x5c/0x68
[ 936.255221] br_multicast_add_group+0x40/0x170 [bridge]
[ 936.260491] br_multicast_rcv+0x7ac/0xe30 [bridge]
[ 936.265322] br_dev_xmit+0x140/0x368 [bridge]
[ 936.269689] dev_hard_start_xmit+0x94/0x158
[ 936.273876] __dev_queue_xmit+0x5ac/0x7f8
[ 936.277890] dev_queue_xmit+0x10/0x18
[ 936.281563] neigh_resolve_output+0xec/0x198
[ 936.285845] ip6_finish_output2+0x240/0x710
[ 936.290039] __ip6_finish_output+0x130/0x170
[ 936.294318] ip6_output+0x6c/0x1c8
[ 936.297731] NF_HOOK.constprop.0+0xd8/0xe8
[ 936.301834] igmp6_send+0x358/0x558
[ 936.305326] igmp6_join_group.part.0+0x30/0xf0
[ 936.309774] igmp6_group_added+0xfc/0x110
[ 936.313787] __ipv6_dev_mc_inc+0x1a4/0x290
[ 936.317885] ipv6_dev_mc_inc+0x10/0x18
[ 936.321677] br_multicast_open+0xbc/0x110 [bridge]
[ 936.326506] br_multicast_toggle+0xec/0x140 [bridge]
Fixes: 4effd28c12 ("bridge: join all-snoopers multicast address")
Signed-off-by: Joseph Huang <Joseph.Huang@garmin.com>
Acked-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Link: https://lore.kernel.org/r/20201204235628.50653-1-Joseph.Huang@garmin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
migrate_disable() is just a wrapper for preempt_disable() in
non-RT kernel. It is safe to replace it, and RT kernel will
benefit.
Note that it is introduced since Feb 2020.
Suggested-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20201205075946.497763-2-xiyou.wangcong@gmail.com
The per-cpu bpf_redirect_info is shared among all skb_do_redirect()
and BPF redirect helpers. Callers on RX path are all in BH context,
disabling preemption is not sufficient to prevent BH interruption.
In production, we observed strange packet drops because of the race
condition between LWT xmit and TC ingress, and we verified this issue
is fixed after we disable BH.
Although this bug was technically introduced from the beginning, that
is commit 3a0af8fd61 ("bpf: BPF for lightweight tunnel infrastructure"),
at that time call_rcu() had to be call_rcu_bh() to match the RCU context.
So this patch may not work well before RCU flavor consolidation has been
completed around v5.0.
Update the comments above the code too, as call_rcu() is now BH friendly.
Signed-off-by: Dongdong Wang <wangdongdong.6@bytedance.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Cong Wang <cong.wang@bytedance.com>
Link: https://lore.kernel.org/bpf/20201205075946.497763-1-xiyou.wangcong@gmail.com
Guillaume noticed that: for segments udp_queue_rcv_one_skb() returns the
proto, and it should pass "ret" unmodified to ip_protocol_deliver_rcu().
Otherwize, with a negtive value passed, it will underflow inet_protos.
This can be reproduced with IPIP FOU:
# ip fou add port 5555 ipproto 4
# ethtool -K eth1 rx-gro-list on
Fixes: cf329aa42b ("udp: cope with UDP GRO packet misdirection")
Reported-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
