In some production workloads we noticed that connections could
sometimes close extremely prematurely with ETIMEDOUT after
transmitting only 1 TLP and RTO retransmission (when we would normally
expect roughly tcp_retries2 = TCP_RETR2 = 15 RTOs before a connection
closes with ETIMEDOUT).
From tracing we determined that these workloads can suffer from a
scenario where in fast recovery, after some retransmits, a DSACK undo
can happen at a point where the scoreboard is totally clear (we have
retrans_out == sacked_out == lost_out == 0). In such cases, calling
tcp_try_keep_open() means that we do not execute any code path that
clears tp->retrans_stamp to 0. That means that tp->retrans_stamp can
remain erroneously set to the start time of the undone fast recovery,
even after the fast recovery is undone. If minutes or hours elapse,
and then a TLP/RTO/RTO sequence occurs, then the start_ts value in
retransmits_timed_out() (which is from tp->retrans_stamp) will be
erroneously ancient (left over from the fast recovery undone via
DSACKs). Thus this ancient tp->retrans_stamp value can cause the
connection to die very prematurely with ETIMEDOUT via
tcp_write_err().
The fix: we change DSACK undo in fast recovery (TCP_CA_Recovery) to
call tcp_try_to_open() instead of tcp_try_keep_open(). This ensures
that if no retransmits are in flight at the time of DSACK undo in fast
recovery then we properly zero retrans_stamp. Note that calling
tcp_try_to_open() is more consistent with other loss recovery
behavior, since normal fast recovery (CA_Recovery) and RTO recovery
(CA_Loss) both normally end when tp->snd_una meets or exceeds
tp->high_seq and then in tcp_fastretrans_alert() the "default" switch
case executes tcp_try_to_open(). Also note that by inspection this
change to call tcp_try_to_open() implies at least one other nice bug
fix, where now an ECE-marked DSACK that causes an undo will properly
invoke tcp_enter_cwr() rather than ignoring the ECE mark.
Fixes: c7d9d6a185 ("tcp: undo on DSACK during recovery")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Testing determined that the recent commit 9e046bb111 ("tcp: clear
tp->retrans_stamp in tcp_rcv_fastopen_synack()") has a race, and does
not always ensure retrans_stamp is 0 after a TFO payload retransmit.
If transmit completion for the SYN+data skb happens after the client
TCP stack receives the SYNACK (which sometimes happens), then
retrans_stamp can erroneously remain non-zero for the lifetime of the
connection, causing a premature ETIMEDOUT later.
Testing and tracing showed that the buggy scenario is the following
somewhat tricky sequence:
+ Client attempts a TFO handshake. tcp_send_syn_data() sends SYN + TFO
cookie + data in a single packet in the syn_data skb. It hands the
syn_data skb to tcp_transmit_skb(), which makes a clone. Crucially,
it then reuses the same original (non-clone) syn_data skb,
transforming it by advancing the seq by one byte and removing the
FIN bit, and enques the resulting payload-only skb in the
sk->tcp_rtx_queue.
+ Client sets retrans_stamp to the start time of the three-way
handshake.
+ Cookie mismatches or server has TFO disabled, and server only ACKs
SYN.
+ tcp_ack() sees SYN is acked, tcp_clean_rtx_queue() clears
retrans_stamp.
+ Since the client SYN was acked but not the payload, the TFO failure
code path in tcp_rcv_fastopen_synack() tries to retransmit the
payload skb. However, in some cases the transmit completion for the
clone of the syn_data (which had SYN + TFO cookie + data) hasn't
happened. In those cases, skb_still_in_host_queue() returns true
for the retransmitted TFO payload, because the clone of the syn_data
skb has not had its tx completetion.
+ Because skb_still_in_host_queue() finds skb_fclone_busy() is true,
it sets the TSQ_THROTTLED bit and the retransmit does not happen in
the tcp_rcv_fastopen_synack() call chain.
+ The tcp_rcv_fastopen_synack() code next implicitly assumes the
retransmit process is finished, and sets retrans_stamp to 0 to clear
it, but this is later overwritten (see below).
+ Later, upon tx completion, tcp_tsq_write() calls
tcp_xmit_retransmit_queue(), which puts the retransmit in flight and
sets retrans_stamp to a non-zero value.
+ The client receives an ACK for the retransmitted TFO payload data.
+ Since we're in CA_Open and there are no dupacks/SACKs/DSACKs/ECN to
make tcp_ack_is_dubious() true and make us call
tcp_fastretrans_alert() and reach a code path that clears
retrans_stamp, retrans_stamp stays nonzero.
+ Later, if there is a TLP, RTO, RTO sequence, then the connection
will suffer an early ETIMEDOUT due to the erroneously ancient
retrans_stamp.
The fix: this commit refactors the code to have
tcp_rcv_fastopen_synack() retransmit by reusing the relevant parts of
tcp_simple_retransmit() that enter CA_Loss (without changing cwnd) and
call tcp_xmit_retransmit_queue(). We have tcp_simple_retransmit() and
tcp_rcv_fastopen_synack() share code in this way because in both cases
we get a packet indicating non-congestion loss (MTU reduction or TFO
failure) and thus in both cases we want to retransmit as many packets
as cwnd allows, without reducing cwnd. And given that retransmits will
set retrans_stamp to a non-zero value (and may do so in a later
calling context due to TSQ), we also want to enter CA_Loss so that we
track when all retransmitted packets are ACked and clear retrans_stamp
when that happens (to ensure later recurring RTOs are using the
correct retrans_stamp and don't declare ETIMEDOUT prematurely).
Fixes: 9e046bb111 ("tcp: clear tp->retrans_stamp in tcp_rcv_fastopen_synack()")
Fixes: a7abf3cd76 ("tcp: consider using standard rtx logic in tcp_rcv_fastopen_synack()")
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Link: https://patch.msgid.link/20240624144323.2371403-1-ncardwell.sw@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When bonding is configured in BOND_MODE_BROADCAST mode, if two identical
SYN packets are received at the same time and processed on different CPUs,
it can potentially create the same sk (sock) but two different reqsk
(request_sock) in tcp_conn_request().
These two different reqsk will respond with two SYNACK packets, and since
the generation of the seq (ISN) incorporates a timestamp, the final two
SYNACK packets will have different seq values.
The consequence is that when the Client receives and replies with an ACK
to the earlier SYNACK packet, we will reset(RST) it.
========================================================================
This behavior is consistently reproducible in my local setup,
which comprises:
| NETA1 ------ NETB1 |
PC_A --- bond --- | | --- bond --- PC_B
| NETA2 ------ NETB2 |
- PC_A is the Server and has two network cards, NETA1 and NETA2. I have
bonded these two cards using BOND_MODE_BROADCAST mode and configured
them to be handled by different CPU.
- PC_B is the Client, also equipped with two network cards, NETB1 and
NETB2, which are also bonded and configured in BOND_MODE_BROADCAST mode.
If the client attempts a TCP connection to the server, it might encounter
a failure. Capturing packets from the server side reveals:
10.10.10.10.45182 > localhost: Flags [S], seq 320236027,
10.10.10.10.45182 > localhost: Flags [S], seq 320236027,
localhost > 10.10.10.10.45182: Flags [S.], seq 2967855116,
localhost > 10.10.10.10.45182: Flags [S.], seq 2967855123, <==
10.10.10.10.45182 > localhost: Flags [.], ack 4294967290,
10.10.10.10.45182 > localhost: Flags [.], ack 4294967290,
localhost > 10.10.10.10.45182: Flags [R], seq 2967855117, <==
localhost > 10.10.10.10.45182: Flags [R], seq 2967855117,
Two SYNACKs with different seq numbers are sent by localhost,
resulting in an anomaly.
========================================================================
The attempted solution is as follows:
Add a return value to inet_csk_reqsk_queue_hash_add() to confirm if the
ehash insertion is successful (Up to now, the reason for unsuccessful
insertion is that a reqsk for the same connection has already been
inserted). If the insertion fails, release the reqsk.
Due to the refcnt, Kuniyuki suggests also adding a return value check
for the DCCP module; if ehash insertion fails, indicating a successful
insertion of the same connection, simply release the reqsk as well.
Simultaneously, In the reqsk_queue_hash_req(), the start of the
req->rsk_timer is adjusted to be after successful insertion.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: luoxuanqiang <luoxuanqiang@kylinos.cn>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240621013929.1386815-1-luoxuanqiang@kylinos.cn
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Some applications were reporting ETIMEDOUT errors on apparently
good looking flows, according to packet dumps.
We were able to root cause the issue to an accidental setting
of tp->retrans_stamp in the following scenario:
- client sends TFO SYN with data.
- server has TFO disabled, ACKs only SYN but not payload.
- client receives SYNACK covering only SYN.
- tcp_ack() eats SYN and sets tp->retrans_stamp to 0.
- tcp_rcv_fastopen_synack() calls tcp_xmit_retransmit_queue()
to retransmit TFO payload w/o SYN, sets tp->retrans_stamp to "now",
but we are not in any loss recovery state.
- TFO payload is ACKed.
- we are not in any loss recovery state, and don't see any dupacks,
so we don't get to any code path that clears tp->retrans_stamp.
- tp->retrans_stamp stays non-zero for the lifetime of the connection.
- after first RTO, tcp_clamp_rto_to_user_timeout() clamps second RTO
to 1 jiffy due to bogus tp->retrans_stamp.
- on clamped RTO with non-zero icsk_retransmits, retransmits_timed_out()
sets start_ts from tp->retrans_stamp from TFO payload retransmit
hours/days ago, and computes bogus long elapsed time for loss recovery,
and suffers ETIMEDOUT early.
Fixes: a7abf3cd76 ("tcp: consider using standard rtx logic in tcp_rcv_fastopen_synack()")
CC: stable@vger.kernel.org
Co-developed-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Co-developed-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240614130615.396837-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As the comment in this function says, the code currently just clears the
CIPSO part with IPOPT_NOP, rather than removing it completely and
trimming the packet. The other cipso_v4_*_delattr() functions, however,
do the proper removal and also calipso_skbuff_delattr() makes an effort
to remove the CALIPSO options instead of replacing them with padding.
Some routers treat IPv4 packets with anything (even NOPs) in the option
header as a special case and take them through a slower processing path.
Consequently, hardening guides such as STIG recommend to configure such
routers to drop packets with non-empty IP option headers [1][2]. Thus,
users might expect NetLabel to produce packets with minimal padding (or
at least with no padding when no actual options are present).
Implement the proper option removal to address this and to be closer to
what the peer functions do.
[1] https://www.stigviewer.com/stig/juniper_router_rtr/2019-09-27/finding/V-90937
[2] https://www.stigviewer.com/stig/cisco_ios_xe_router_rtr/2021-03-26/finding/V-217001
Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As evident from the definition of ip_options_get(), the IP option
IPOPT_END is used to pad the IP option data array, not IPOPT_NOP. Yet
the loop that walks the IP options to determine the total IP options
length in cipso_v4_delopt() doesn't take IPOPT_END into account.
Fix it by recognizing the IPOPT_END value as the end of actual options.
Fixes: 014ab19a69 ("selinux: Set socket NetLabel based on connection endpoint")
Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Due to timer wheel implementation, a timer will usually fire
after its schedule.
For instance, for HZ=1000, a timeout between 512ms and 4s
has a granularity of 64ms.
For this range of values, the extra delay could be up to 63ms.
For TCP, this means that tp->rcv_tstamp may be after
inet_csk(sk)->icsk_timeout whenever the timer interrupt
finally triggers, if one packet came during the extra delay.
We need to make sure tcp_rtx_probe0_timed_out() handles this case.
Fixes: e89688e3e9 ("net: tcp: fix unexcepted socket die when snd_wnd is 0")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Menglong Dong <imagedong@tencent.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Link: https://lore.kernel.org/r/20240607125652.1472540-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jaroslav reports Dell's OMSA Systems Management Data Engine
expects NLM_DONE in a separate recvmsg(), both for rtnl_dump_ifinfo()
and inet_dump_ifaddr(). We already added a similar fix previously in
commit 460b0d33cf ("inet: bring NLM_DONE out to a separate recv() again")
Instead of modifying all the dump handlers, and making them look
different than modern for_each_netdev_dump()-based dump handlers -
put the workaround in rtnetlink code. This will also help us move
the custom rtnl-locking from af_netlink in the future (in net-next).
Note that this change is not touching rtnl_dump_all(). rtnl_dump_all()
is different kettle of fish and a potential problem. We now mix families
in a single recvmsg(), but NLM_DONE is not coalesced.
Tested:
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_addr.yaml \
--dump getaddr --json '{"ifa-family": 2}'
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_route.yaml \
--dump getroute --json '{"rtm-family": 2}'
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_link.yaml \
--dump getlink
Fixes: 3e41af9076 ("rtnetlink: use xarray iterator to implement rtnl_dump_ifinfo()")
Fixes: cdb2f80f1c ("inet: use xa_array iterator to implement inet_dump_ifaddr()")
Reported-by: Jaroslav Pulchart <jaroslav.pulchart@gooddata.com>
Link: https://lore.kernel.org/all/CAK8fFZ7MKoFSEzMBDAOjoUt+vTZRRQgLDNXEOfdCCXSoXXKE0g@mail.gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
According to RFC 1213, we should also take CLOSE-WAIT sockets into
consideration:
"tcpCurrEstab OBJECT-TYPE
...
The number of TCP connections for which the current state
is either ESTABLISHED or CLOSE- WAIT."
After this, CurrEstab counter will display the total number of
ESTABLISHED and CLOSE-WAIT sockets.
The logic of counting
When we increment the counter?
a) if we change the state to ESTABLISHED.
b) if we change the state from SYN-RECEIVED to CLOSE-WAIT.
When we decrement the counter?
a) if the socket leaves ESTABLISHED and will never go into CLOSE-WAIT,
say, on the client side, changing from ESTABLISHED to FIN-WAIT-1.
b) if the socket leaves CLOSE-WAIT, say, on the server side, changing
from CLOSE-WAIT to LAST-ACK.
Please note: there are two chances that old state of socket can be changed
to CLOSE-WAIT in tcp_fin(). One is SYN-RECV, the other is ESTABLISHED.
So we have to take care of the former case.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For TLS offload we mark packets with skb->decrypted to make sure
they don't escape the host without getting encrypted first.
The crypto state lives in the socket, so it may get detached
by a call to skb_orphan(). As a safety check - the egress path
drops all packets with skb->decrypted and no "crypto-safe" socket.
The skb marking was added to sendpage only (and not sendmsg),
because tls_device injected data into the TCP stack using sendpage.
This special case was missed when sendpage got folded into sendmsg.
Fixes: c5c37af6ec ("tcp: Convert do_tcp_sendpages() to use MSG_SPLICE_PAGES")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240530232607.82686-1-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
TCP_CLOSE may or may not have current/rnext keys and should not be
considered "established". The fast-path for TCP_CLOSE is
SKB_DROP_REASON_TCP_CLOSE. This is what tcp_rcv_state_process() does
anyways. Add an early drop path to not spend any time verifying
segment signatures for sockets in TCP_CLOSE state.
Cc: stable@vger.kernel.org # v6.7
Fixes: 0a3a809089 ("net/tcp: Verify inbound TCP-AO signed segments")
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Link: https://lore.kernel.org/r/20240529-tcp_ao-sk_state-v1-1-d69b5d323c52@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=cexP
-----END PGP SIGNATURE-----
Merge tag 'nf-24-05-29' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
Patch #1 syzbot reports that nf_reinject() could be called without
rcu_read_lock() when flushing pending packets at nfnetlink
queue removal, from Eric Dumazet.
Patch #2 flushes ipset list:set when canceling garbage collection to
reference to other lists to fix a race, from Jozsef Kadlecsik.
Patch #3 restores q-in-q matching with nft_payload by reverting
f6ae9f120d ("netfilter: nft_payload: add C-VLAN support").
Patch #4 fixes vlan mangling in skbuff when vlan offload is present
in skbuff, without this patch nft_payload corrupts packets
in this case.
Patch #5 fixes possible nul-deref in tproxy no IP address is found in
netdevice, reported by syzbot and patch from Florian Westphal.
Patch #6 removes a superfluous restriction which prevents loose fib
lookups from input and forward hooks, from Eric Garver.
My assessment is that patches #1, #2 and #5 address possible kernel
crash, anything else in this batch fixes broken features.
netfilter pull request 24-05-29
* tag 'nf-24-05-29' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nft_fib: allow from forward/input without iif selector
netfilter: tproxy: bail out if IP has been disabled on the device
netfilter: nft_payload: skbuff vlan metadata mangle support
netfilter: nft_payload: restore vlan q-in-q match support
netfilter: ipset: Add list flush to cancel_gc
netfilter: nfnetlink_queue: acquire rcu_read_lock() in instance_destroy_rcu()
====================
Link: https://lore.kernel.org/r/20240528225519.1155786-1-pablo@netfilter.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
A recent change to inet_dump_ifaddr had the function incorrectly iterate
over net rather than tgt_net, resulting in the data coming for the
incorrect network namespace.
Fixes: cdb2f80f1c ("inet: use xa_array iterator to implement inet_dump_ifaddr()")
Reported-by: Stéphane Graber <stgraber@stgraber.org>
Closes: https://github.com/lxc/incus/issues/892
Bisected-by: Stéphane Graber <stgraber@stgraber.org>
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Tested-by: Stéphane Graber <stgraber@stgraber.org>
Acked-by: Christian Brauner <brauner@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240528203030.10839-1-aleksandr.mikhalitsyn@canonical.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
__dst_negative_advice() does not enforce proper RCU rules when
sk->dst_cache must be cleared, leading to possible UAF.
RCU rules are that we must first clear sk->sk_dst_cache,
then call dst_release(old_dst).
Note that sk_dst_reset(sk) is implementing this protocol correctly,
while __dst_negative_advice() uses the wrong order.
Given that ip6_negative_advice() has special logic
against RTF_CACHE, this means each of the three ->negative_advice()
existing methods must perform the sk_dst_reset() themselves.
Note the check against NULL dst is centralized in
__dst_negative_advice(), there is no need to duplicate
it in various callbacks.
Many thanks to Clement Lecigne for tracking this issue.
This old bug became visible after the blamed commit, using UDP sockets.
Fixes: a87cb3e48e ("net: Facility to report route quality of connected sockets")
Reported-by: Clement Lecigne <clecigne@google.com>
Diagnosed-by: Clement Lecigne <clecigne@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <tom@herbertland.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240528114353.1794151-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
syzbot reports:
general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN PTI
KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f]
[..]
RIP: 0010:nf_tproxy_laddr4+0xb7/0x340 net/ipv4/netfilter/nf_tproxy_ipv4.c:62
Call Trace:
nft_tproxy_eval_v4 net/netfilter/nft_tproxy.c:56 [inline]
nft_tproxy_eval+0xa9a/0x1a00 net/netfilter/nft_tproxy.c:168
__in_dev_get_rcu() can return NULL, so check for this.
Reported-and-tested-by: syzbot+b94a6818504ea90d7661@syzkaller.appspotmail.com
Fixes: cc6eb43385 ("tproxy: use the interface primary IP address as a default value for --on-ip")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Jason commit made checks against ACK sequence less strict
and can be exploited by attackers to establish spoofed flows
with less probes.
Innocent users might use tcp_rmem[1] == 1,000,000,000,
or something more reasonable.
An attacker can use a regular TCP connection to learn the server
initial tp->rcv_wnd, and use it to optimize the attack.
If we make sure that only the announced window (smaller than 65535)
is used for ACK validation, we force an attacker to use
65537 packets to complete the 3WHS (assuming server ISN is unknown)
Fixes: 378979e94e ("tcp: remove 64 KByte limit for initial tp->rcv_wnd value")
Link: https://datatracker.ietf.org/meeting/119/materials/slides-119-tcpm-ghost-acks-00
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Link: https://lore.kernel.org/r/20240523130528.60376-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Syzkaller was able to trigger
kernel BUG at net/core/gro.c:424 !
RIP: 0010:gro_pull_from_frag0 net/core/gro.c:424 [inline]
RIP: 0010:gro_try_pull_from_frag0 net/core/gro.c:446 [inline]
RIP: 0010:dev_gro_receive+0x242f/0x24b0 net/core/gro.c:571
Due to using an incorrect NAPI_GRO_CB(skb)->network_offset.
The referenced commit sets this offset to 0 in skb_gro_reset_offset.
That matches the expected case in dev_gro_receive:
pp = INDIRECT_CALL_INET(ptype->callbacks.gro_receive,
ipv6_gro_receive, inet_gro_receive,
&gro_list->list, skb);
But syzkaller injected an skb with protocol ETH_P_TEB into an ip6gre
device (by writing the IP6GRE encapsulated version to a TAP device).
The result was a first call to eth_gro_receive, and thus an extra
ETH_HLEN in network_offset that should not be there. First issue hit
is when computing offset from network header in ipv6_gro_pull_exthdrs.
Initialize both offsets in the network layer gro_receive.
This pairs with all reads in gro_receive, which use
skb_gro_receive_network_offset().
Fixes: 186b1ea73a ("net: gro: use cb instead of skb->network_header")
Reported-by: syzkaller <syzkaller@googlegroups.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
CC: Richard Gobert <richardbgobert@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240523141434.1752483-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cited commit started returning an error when user space requests to dump
the interface's IPv4 addresses and IPv4 is disabled on the interface.
Restore the previous behavior and do not return an error.
Before cited commit:
# ip address show dev dummy1
10: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether e2:40:68:98:d0:18 brd ff:ff:ff:ff:ff:ff
inet6 fe80::e040:68ff:fe98:d018/64 scope link proto kernel_ll
valid_lft forever preferred_lft forever
# ip link set dev dummy1 mtu 67
# ip address show dev dummy1
10: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 67 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether e2:40:68:98:d0:18 brd ff:ff:ff:ff:ff:ff
After cited commit:
# ip address show dev dummy1
10: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether 32:2d:69:f2:9c:99 brd ff:ff:ff:ff:ff:ff
inet6 fe80::302d:69ff:fef2:9c99/64 scope link proto kernel_ll
valid_lft forever preferred_lft forever
# ip link set dev dummy1 mtu 67
# ip address show dev dummy1
RTNETLINK answers: No such device
Dump terminated
With this patch:
# ip address show dev dummy1
10: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether de:17:56:bb:57:c0 brd ff:ff:ff:ff:ff:ff
inet6 fe80::dc17:56ff:febb:57c0/64 scope link proto kernel_ll
valid_lft forever preferred_lft forever
# ip link set dev dummy1 mtu 67
# ip address show dev dummy1
10: dummy1: <BROADCAST,NOARP,UP,LOWER_UP> mtu 67 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether de:17:56:bb:57:c0 brd ff:ff:ff:ff:ff:ff
I fixed the exact same issue for IPv6 in commit c04f7dfe6e ("ipv6: Fix
address dump when IPv6 is disabled on an interface"), but noted [1] that
I am not doing the change for IPv4 because I am not aware of a way to
disable IPv4 on an interface other than unregistering it. I clearly
missed the above case.
[1] https://lore.kernel.org/netdev/20240321173042.2151756-1-idosch@nvidia.com/
Fixes: cdb2f80f1c ("inet: use xa_array iterator to implement inet_dump_ifaddr()")
Reported-by: Carolina Jubran <cjubran@nvidia.com>
Reported-by: Yamen Safadi <ysafadi@nvidia.com>
Tested-by: Carolina Jubran <cjubran@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240523110257.334315-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
regression you have been notified of in the past weeks.
The TCP window fix will require some follow-up, already queued.
Current release - regressions:
- af_unix: fix garbage collection of embryos
Previous releases - regressions:
- af_unix: fix race between GC and receive path
- ipv6: sr: fix missing sk_buff release in seg6_input_core
- tcp: remove 64 KByte limit for initial tp->rcv_wnd value
- eth: r8169: fix rx hangup
- eth: lan966x: remove ptp traps in case the ptp is not enabled.
- eth: ixgbe: fix link breakage vs cisco switches.
- eth: ice: prevent ethtool from corrupting the channels.
Previous releases - always broken:
- openvswitch: set the skbuff pkt_type for proper pmtud support.
- tcp: Fix shift-out-of-bounds in dctcp_update_alpha().
Misc:
- a bunch of selftests stabilization patches.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=zd2T
-----END PGP SIGNATURE-----
Merge tag 'net-6.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Paolo Abeni:
"Quite smaller than usual. Notably it includes the fix for the unix
regression from the past weeks. The TCP window fix will require some
follow-up, already queued.
Current release - regressions:
- af_unix: fix garbage collection of embryos
Previous releases - regressions:
- af_unix: fix race between GC and receive path
- ipv6: sr: fix missing sk_buff release in seg6_input_core
- tcp: remove 64 KByte limit for initial tp->rcv_wnd value
- eth: r8169: fix rx hangup
- eth: lan966x: remove ptp traps in case the ptp is not enabled
- eth: ixgbe: fix link breakage vs cisco switches
- eth: ice: prevent ethtool from corrupting the channels
Previous releases - always broken:
- openvswitch: set the skbuff pkt_type for proper pmtud support
- tcp: Fix shift-out-of-bounds in dctcp_update_alpha()
Misc:
- a bunch of selftests stabilization patches"
* tag 'net-6.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (25 commits)
r8169: Fix possible ring buffer corruption on fragmented Tx packets.
idpf: Interpret .set_channels() input differently
ice: Interpret .set_channels() input differently
nfc: nci: Fix handling of zero-length payload packets in nci_rx_work()
net: relax socket state check at accept time.
tcp: remove 64 KByte limit for initial tp->rcv_wnd value
net: ti: icssg_prueth: Fix NULL pointer dereference in prueth_probe()
tls: fix missing memory barrier in tls_init
net: fec: avoid lock evasion when reading pps_enable
Revert "ixgbe: Manual AN-37 for troublesome link partners for X550 SFI"
testing: net-drv: use stats64 for testing
net: mana: Fix the extra HZ in mana_hwc_send_request
net: lan966x: Remove ptp traps in case the ptp is not enabled.
openvswitch: Set the skbuff pkt_type for proper pmtud support.
selftest: af_unix: Make SCM_RIGHTS into OOB data.
af_unix: Fix garbage collection of embryos carrying OOB with SCM_RIGHTS
tcp: Fix shift-out-of-bounds in dctcp_update_alpha().
selftests/net: use tc rule to filter the na packet
ipv6: sr: fix memleak in seg6_hmac_init_algo
af_unix: Update unix_sk(sk)->oob_skb under sk_receive_queue lock.
...
Recently, we had some servers upgraded to the latest kernel and noticed
the indicator from the user side showed worse results than before. It is
caused by the limitation of tp->rcv_wnd.
In 2018 commit a337531b94 ("tcp: up initial rmem to 128KB and SYN rwin
to around 64KB") limited the initial value of tp->rcv_wnd to 65535, most
CDN teams would not benefit from this change because they cannot have a
large window to receive a big packet, which will be slowed down especially
in long RTT. Small rcv_wnd means slow transfer speed, to some extent. It's
the side effect for the latency/time-sensitive users.
To avoid future confusion, current change doesn't affect the initial
receive window on the wire in a SYN or SYN+ACK packet which are set within
65535 bytes according to RFC 7323 also due to the limit in
__tcp_transmit_skb():
th->window = htons(min(tp->rcv_wnd, 65535U));
In one word, __tcp_transmit_skb() already ensures that constraint is
respected, no matter how large tp->rcv_wnd is. The change doesn't violate
RFC.
Let me provide one example if with or without the patch:
Before:
client --- SYN: rwindow=65535 ---> server
client <--- SYN+ACK: rwindow=65535 ---- server
client --- ACK: rwindow=65536 ---> server
Note: for the last ACK, the calculation is 512 << 7.
After:
client --- SYN: rwindow=65535 ---> server
client <--- SYN+ACK: rwindow=65535 ---- server
client --- ACK: rwindow=175232 ---> server
Note: I use the following command to make it work:
ip route change default via [ip] dev eth0 metric 100 initrwnd 120
For the last ACK, the calculation is 1369 << 7.
When we apply such a patch, having a large rcv_wnd if the user tweak this
knob can help transfer data more rapidly and save some rtts.
Fixes: a337531b94 ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB")
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20240521134220.12510-1-kerneljasonxing@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=S6Xt
-----END PGP SIGNATURE-----
Merge tag 'net-accept-more-20240515' of git://git.kernel.dk/linux
Pull more io_uring updates from Jens Axboe:
"This adds support for IORING_CQE_F_SOCK_NONEMPTY for io_uring accept
requests.
This is very similar to previous work that enabled the same hint for
doing receives on sockets. By far the majority of the work here is
refactoring to enable the networking side to pass back whether or not
the socket had more pending requests after accepting the current one,
the last patch just wires it up for io_uring.
Not only does this enable applications to know whether there are more
connections to accept right now, it also enables smarter logic for
io_uring multishot accept on whether to retry immediately or wait for
a poll trigger"
* tag 'net-accept-more-20240515' of git://git.kernel.dk/linux:
io_uring/net: wire up IORING_CQE_F_SOCK_NONEMPTY for accept
net: pass back whether socket was empty post accept
net: have do_accept() take a struct proto_accept_arg argument
net: change proto and proto_ops accept type
We're going to send an RST due to invalid syn packet which is already
checked whether 1) it is in sequence, 2) it is a retransmitted skb.
As RFC 793 says, if the state of socket is not CLOSED/LISTEN/SYN-SENT,
then we should send an RST when receiving bad syn packet:
"fourth, check the SYN bit,...If the SYN is in the window it is an
error, send a reset"
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Link: https://lore.kernel.org/r/20240510122502.27850-6-kerneljasonxing@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There are two possible cases where TCP layer can send an RST. Since they
happen in the same place, I think using one independent reason is enough
to identify this special situation.
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Link: https://lore.kernel.org/r/20240510122502.27850-5-kerneljasonxing@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This adds an 'is_empty' argument to struct proto_accept_arg, which can
be used to pass back information on whether or not the given socket has
more connections to accept post the one just accepted.
To utilize this information, the caller should initialize the 'is_empty'
field to, eg, -1 and then check for 0/1 after the accept. If the field
has been set, the caller knows whether there are more pending connections
or not. If the field remains -1 after the accept call, the protocol
doesn't support passing back this information.
This patch wires it up for ipv4/6 TCP.
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Rather than pass in flags, error pointer, and whether this is a kernel
invocation or not, add a struct proto_accept_arg struct as the argument.
This then holds all of these arguments, and prepares accept for being
able to pass back more information.
No functional changes in this patch.
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTFp0I1jqZrAX+hPRXbK58LschIgwUCZkGcZAAKCRDbK58LschI
g6o6APwLsqhrM2w71VUN5ciCxu4H5VDtZp6wkdqtVbxxU4qNxQEApKgYgKt8ZLF3
Kily5c7m+S4ZXhMX21rb8JhSAz0dfQk=
=5Dk7
-----END PGP SIGNATURE-----
Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says:
====================
pull-request: bpf-next 2024-05-13
We've added 119 non-merge commits during the last 14 day(s) which contain
a total of 134 files changed, 9462 insertions(+), 4742 deletions(-).
The main changes are:
1) Add BPF JIT support for 32-bit ARCv2 processors, from Shahab Vahedi.
2) Add BPF range computation improvements to the verifier in particular
around XOR and OR operators, refactoring of checks for range computation
and relaxing MUL range computation so that src_reg can also be an unknown
scalar, from Cupertino Miranda.
3) Add support to attach kprobe BPF programs through kprobe_multi link in
a session mode, meaning, a BPF program is attached to both function entry
and return, the entry program can decide if the return program gets
executed and the entry program can share u64 cookie value with return
program. Session mode is a common use-case for tetragon and bpftrace,
from Jiri Olsa.
4) Fix a potential overflow in libbpf's ring__consume_n() and improve libbpf
as well as BPF selftest's struct_ops handling, from Andrii Nakryiko.
5) Improvements to BPF selftests in context of BPF gcc backend,
from Jose E. Marchesi & David Faust.
6) Migrate remaining BPF selftest tests from test_sock_addr.c to prog_test-
-style in order to retire the old test, run it in BPF CI and additionally
expand test coverage, from Jordan Rife.
7) Big batch for BPF selftest refactoring in order to remove duplicate code
around common network helpers, from Geliang Tang.
8) Another batch of improvements to BPF selftests to retire obsolete
bpf_tcp_helpers.h as everything is available vmlinux.h,
from Martin KaFai Lau.
9) Fix BPF map tear-down to not walk the map twice on free when both timer
and wq is used, from Benjamin Tissoires.
10) Fix BPF verifier assumptions about socket->sk that it can be non-NULL,
from Alexei Starovoitov.
11) Change BTF build scripts to using --btf_features for pahole v1.26+,
from Alan Maguire.
12) Small improvements to BPF reusing struct_size() and krealloc_array(),
from Andy Shevchenko.
13) Fix s390 JIT to emit a barrier for BPF_FETCH instructions,
from Ilya Leoshkevich.
14) Extend TCP ->cong_control() callback in order to feed in ack and
flag parameters and allow write-access to tp->snd_cwnd_stamp
from BPF program, from Miao Xu.
15) Add support for internal-only per-CPU instructions to inline
bpf_get_smp_processor_id() helper call for arm64 and riscv64 BPF JITs,
from Puranjay Mohan.
16) Follow-up to remove the redundant ethtool.h from tooling infrastructure,
from Tushar Vyavahare.
17) Extend libbpf to support "module:<function>" syntax for tracing
programs, from Viktor Malik.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (119 commits)
bpf: make list_for_each_entry portable
bpf: ignore expected GCC warning in test_global_func10.c
bpf: disable strict aliasing in test_global_func9.c
selftests/bpf: Free strdup memory in xdp_hw_metadata
selftests/bpf: Fix a few tests for GCC related warnings.
bpf: avoid gcc overflow warning in test_xdp_vlan.c
tools: remove redundant ethtool.h from tooling infra
selftests/bpf: Expand ATTACH_REJECT tests
selftests/bpf: Expand getsockname and getpeername tests
sefltests/bpf: Expand sockaddr hook deny tests
selftests/bpf: Expand sockaddr program return value tests
selftests/bpf: Retire test_sock_addr.(c|sh)
selftests/bpf: Remove redundant sendmsg test cases
selftests/bpf: Migrate ATTACH_REJECT test cases
selftests/bpf: Migrate expected_attach_type tests
selftests/bpf: Migrate wildcard destination rewrite test
selftests/bpf: Migrate sendmsg6 v4 mapped address tests
selftests/bpf: Migrate sendmsg deny test cases
selftests/bpf: Migrate WILDCARD_IP test
selftests/bpf: Handle SYSCALL_EPERM and SYSCALL_ENOTSUPP test cases
...
====================
Link: https://lore.kernel.org/r/20240513134114.17575-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A way for an application to know if an MPTCP connection fell back to TCP
is to use getsockopt(MPTCP_INFO) and look for errors. The issue with
this technique is that the same errors -- EOPNOTSUPP (IPv4) and
ENOPROTOOPT (IPv6) -- are returned if there was a fallback, *or* if the
kernel doesn't support this socket option. The userspace then has to
look at the kernel version to understand what the errors mean.
It is not clean, and it doesn't take into account older kernels where
the socket option has been backported. A cleaner way would be to expose
this info to the TCP socket level. In case of MPTCP socket where no
fallback happened, the socket options for the TCP level will be handled
in MPTCP code, in mptcp_getsockopt_sol_tcp(). If not, that will be in
TCP code, in do_tcp_getsockopt(). So MPTCP simply has to set the value
1, while TCP has to set 0.
If the socket option is not supported, one of these two errors will be
reported:
- EOPNOTSUPP (95 - Operation not supported) for MPTCP sockets
- ENOPROTOOPT (92 - Protocol not available) for TCP sockets, e.g. on the
socket received after an 'accept()', when the client didn't request to
use MPTCP: this socket will be a TCP one, even if the listen socket
was an MPTCP one.
With this new option, the kernel can return a clear answer to both "Is
this kernel new enough to tell me the fallback status?" and "If it is
new enough, is it currently a TCP or MPTCP socket?" questions, while not
breaking the previous method.
Acked-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240509-upstream-net-next-20240509-mptcp-tcp_is_mptcp-v1-1-f846df999202@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
{inet,ipv6}_gro_receive functions perform flush checks (ttl, flags,
iph->id, ...) against all packets in a loop. These flush checks are used in
all merging UDP and TCP flows.
These checks need to be done only once and only against the found p skb,
since they only affect flush and not same_flow.
This patch leverages correct network header offsets from the cb for both
outer and inner network headers - allowing these checks to be done only
once, in tcp_gro_receive and udp_gro_receive_segment. As a result,
NAPI_GRO_CB(p)->flush is not used at all. In addition, flush_id checks are
more declarative and contained in inet_gro_flush, thus removing the need
for flush_id in napi_gro_cb.
This results in less parsing code for non-loop flush tests for TCP and UDP
flows.
To make sure results are not within noise range - I've made netfilter drop
all TCP packets, and measured CPU performance in GRO (in this case GRO is
responsible for about 50% of the CPU utilization).
perf top while replaying 64 parallel IP/TCP streams merging in GRO:
(gro_receive_network_flush is compiled inline to tcp_gro_receive)
net-next:
6.94% [kernel] [k] inet_gro_receive
3.02% [kernel] [k] tcp_gro_receive
patch applied:
4.27% [kernel] [k] tcp_gro_receive
4.22% [kernel] [k] inet_gro_receive
perf top while replaying 64 parallel IP/IP/TCP streams merging in GRO (same
results for any encapsulation, in this case inet_gro_receive is top
offender in net-next)
net-next:
10.09% [kernel] [k] inet_gro_receive
2.08% [kernel] [k] tcp_gro_receive
patch applied:
6.97% [kernel] [k] inet_gro_receive
3.68% [kernel] [k] tcp_gro_receive
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240509190819.2985-3-richardbgobert@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch converts references of skb->network_header to napi_gro_cb's
network_offset and inner_network_offset.
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240509190819.2985-2-richardbgobert@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
I missed that (struct ifaddrmsg)->ifa_flags was only 8bits,
while (struct in_ifaddr)->ifa_flags is 32bits.
Use a temporary 32bit variable as I did in set_ifa_lifetime()
and check_lifetime().
Fixes: 3ddc2231c8 ("inet: annotate data-races around ifa->ifa_flags")
Reported-by: Yu Watanabe <watanabe.yu@gmail.com>
Dianosed-by: Yu Watanabe <watanabe.yu@gmail.com>
Closes: https://github.com/systemd/systemd/pull/32666#issuecomment-2103977928
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240510072932.2678952-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=RD9p
-----END PGP SIGNATURE-----
Merge tag 'nf-next-24-05-12' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next
Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for net-next:
Patch #1 skips transaction if object type provides no .update interface.
Patch #2 skips NETDEV_CHANGENAME which is unused.
Patch #3 enables conntrack to handle Multicast Router Advertisements and
Multicast Router Solicitations from the Multicast Router Discovery
protocol (RFC4286) as untracked opposed to invalid packets.
From Linus Luessing.
Patch #4 updates DCCP conntracker to mark invalid as invalid, instead of
dropping them, from Jason Xing.
Patch #5 uses NF_DROP instead of -NF_DROP since NF_DROP is 0,
also from Jason.
Patch #6 removes reference in netfilter's sysctl documentation on pickup
entries which were already removed by Florian Westphal.
Patch #7 removes check for IPS_OFFLOAD flag to disable early drop which
allows to evict entries from the conntrack table,
also from Florian.
Patches #8 to #16 updates nf_tables pipapo set backend to allocate
the datastructure copy on-demand from preparation phase,
to better deal with OOM situations where .commit step is too late
to fail. Series from Florian Westphal.
Patch #17 adds a selftest with packetdrill to cover conntrack TCP state
transitions, also from Florian.
Patch #18 use GFP_KERNEL to clone elements from control plane to avoid
quick atomic reserves exhaustion with large sets, reporter refers
to million entries magnitude.
* tag 'nf-next-24-05-12' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
netfilter: nf_tables: allow clone callbacks to sleep
selftests: netfilter: add packetdrill based conntrack tests
netfilter: nft_set_pipapo: remove dirty flag
netfilter: nft_set_pipapo: move cloning of match info to insert/removal path
netfilter: nft_set_pipapo: prepare pipapo_get helper for on-demand clone
netfilter: nft_set_pipapo: merge deactivate helper into caller
netfilter: nft_set_pipapo: prepare walk function for on-demand clone
netfilter: nft_set_pipapo: prepare destroy function for on-demand clone
netfilter: nft_set_pipapo: make pipapo_clone helper return NULL
netfilter: nft_set_pipapo: move prove_locking helper around
netfilter: conntrack: remove flowtable early-drop test
netfilter: conntrack: documentation: remove reference to non-existent sysctl
netfilter: use NF_DROP instead of -NF_DROP
netfilter: conntrack: dccp: try not to drop skb in conntrack
netfilter: conntrack: fix ct-state for ICMPv6 Multicast Router Discovery
netfilter: nf_tables: remove NETDEV_CHANGENAME from netdev chain event handler
netfilter: nf_tables: skip transaction if update object is not implemented
====================
Link: https://lore.kernel.org/r/20240512161436.168973-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
DCCP is going away soon, and had no twsk_unique() method.
We can directly call tcp_twsk_unique() for TCP sockets.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240507164140.940547-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Cross-merge networking fixes after downstream PR.
No conflicts.
Adjacent changes:
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c
35d92abfba ("net: hns3: fix kernel crash when devlink reload during initialization")
2a1a1a7b5f ("net: hns3: add command queue trace for hns3")
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Introduce a tracepoint for icmp_send, which can help users to get more
detail information conveniently when icmp abnormal events happen.
1. Giving an usecase example:
=============================
When an application experiences packet loss due to an unreachable UDP
destination port, the kernel will send an exception message through the
icmp_send function. By adding a trace point for icmp_send, developers or
system administrators can obtain detailed information about the UDP
packet loss, including the type, code, source address, destination address,
source port, and destination port. This facilitates the trouble-shooting
of UDP packet loss issues especially for those network-service
applications.
2. Operation Instructions:
==========================
Switch to the tracing directory.
cd /sys/kernel/tracing
Filter for destination port unreachable.
echo "type==3 && code==3" > events/icmp/icmp_send/filter
Enable trace event.
echo 1 > events/icmp/icmp_send/enable
3. Result View:
================
udp_client_erro-11370 [002] ...s.12 124.728002:
icmp_send: icmp_send: type=3, code=3.
From 127.0.0.1:41895 to 127.0.0.1:6666 ulen=23
skbaddr=00000000589b167a
Signed-off-by: Peilin He <he.peilin@zte.com.cn>
Signed-off-by: xu xin <xu.xin16@zte.com.cn>
Reviewed-by: Yunkai Zhang <zhang.yunkai@zte.com.cn>
Cc: Yang Yang <yang.yang29@zte.com.cn>
Cc: Liu Chun <liu.chun2@zte.com.cn>
Cc: Xuexin Jiang <jiang.xuexin@zte.com.cn>
Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Simon reported that ndo_change_mtu() methods were never
updated to use WRITE_ONCE(dev->mtu, new_mtu) as hinted
in commit 501a90c945 ("inet: protect against too small
mtu values.")
We read dev->mtu without holding RTNL in many places,
with READ_ONCE() annotations.
It is time to take care of ndo_change_mtu() methods
to use corresponding WRITE_ONCE()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Simon Horman <horms@kernel.org>
Closes: https://lore.kernel.org/netdev/20240505144608.GB67882@kernel.org/
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240506102812.3025432-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We want to be able to run rtnl_fill_ifinfo() under RCU protection
instead of RTNL in the future.
All rtnl_link_ops->get_link_net() methods already using dev_net()
are ready. I added READ_ONCE() annotations on others.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=QiGZ
-----END PGP SIGNATURE-----
Merge tag 'ipsec-next-2024-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next
Steffen Klassert says:
====================
pull request (net-next): ipsec-next 2024-05-03
1) Remove Obsolete UDP_ENCAP_ESPINUDP_NON_IKE Support.
This was defined by an early version of an IETF draft
that did not make it to a standard.
2) Introduce direction attribute for xfrm states.
xfrm states have a direction, a stsate can be used
either for input or output packet processing.
Add a direction to xfrm states to make it clear
for what a xfrm state is used.
* tag 'ipsec-next-2024-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next:
xfrm: Restrict SA direction attribute to specific netlink message types
xfrm: Add dir validation to "in" data path lookup
xfrm: Add dir validation to "out" data path lookup
xfrm: Add Direction to the SA in or out
udpencap: Remove Obsolete UDP_ENCAP_ESPINUDP_NON_IKE Support
====================
Link: https://lore.kernel.org/r/20240503082732.2835810-1-steffen.klassert@secunet.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
At the beginning in 2009 one patch [1] introduced collecting drop
counter in nf_conntrack_in() by returning -NF_DROP. Later, another
patch [2] changed the return value of tcp_packet() which now is
renamed to nf_conntrack_tcp_packet() from -NF_DROP to NF_DROP. As
we can see, that -NF_DROP should be corrected.
Similarly, there are other two points where the -NF_DROP is used.
Well, as NF_DROP is equal to 0, inverting NF_DROP makes no sense
as patch [2] said many years ago.
[1]
commit 7d1e04598e ("netfilter: nf_conntrack: account packets drop by tcp_packet()")
[2]
commit ec8d540969 ("netfilter: conntrack: fix dropping packet after l4proto->packet()")
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When forwarding TCP after GRO, software segmentation is very expensive,
especially when the checksum needs to be recalculated.
One case where that's currently unavoidable is when routing packets over
PPPoE. Performance improves significantly when using fraglist GRO
implemented in the same way as for UDP.
When NETIF_F_GRO_FRAGLIST is enabled, perform a lookup for an established
socket in the same netns as the receiving device. While this may not
cover all relevant use cases in multi-netns configurations, it should be
good enough for most configurations that need this.
Here's a measurement of running 2 TCP streams through a MediaTek MT7622
device (2-core Cortex-A53), which runs NAT with flow offload enabled from
one ethernet port to PPPoE on another ethernet port + cake qdisc set to
1Gbps.
rx-gro-list off: 630 Mbit/s, CPU 35% idle
rx-gro-list on: 770 Mbit/s, CPU 40% idle
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Pull the code out of tcp_gro_receive in order to access the tcp header
from tcp4/6_gro_receive.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This pulls the flow port matching out of tcp_gro_receive, so that it can be
reused for the next change, which adds the TCP fraglist GRO heuristic.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This implements fraglist GRO similar to how it's handled in UDP, however
no functional changes are added yet. The next change adds a heuristic for
using fraglist GRO instead of regular GRO.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Preparation for adding TCP fraglist GRO support. It expects packets to be
combined in a similar way as UDP fraglist GSO packets.
For IPv4 packets, NAT is handled in the same way as UDP fraglist GSO.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This helper function will be used for TCP fraglist GRO support
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
-----BEGIN PGP SIGNATURE-----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=aeD1
-----END PGP SIGNATURE-----
Merge tag 'ipsec-2024-05-02' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec
Steffen Klassert says:
====================
pull request (net): ipsec 2024-05-02
1) Fix an error pointer dereference in xfrm_in_fwd_icmp.
From Antony Antony.
2) Preserve vlan tags for ESP transport mode software GRO.
From Paul Davey.
3) Fix a spelling mistake in an uapi xfrm.h comment.
From Anotny Antony.
* tag 'ipsec-2024-05-02' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec:
xfrm: Correct spelling mistake in xfrm.h comment
xfrm: Preserve vlan tags for transport mode software GRO
xfrm: fix possible derferencing in error path
====================
Link: https://lore.kernel.org/r/20240502084838.2269355-1-steffen.klassert@secunet.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
